MeGUI Feature Request Thread [Archive] - Page 2

Doom9

20th February 2006, 15:21

@dimzon: I don't use ICQ.. never liked that bloatware.

E.g. launch one single-threaded job per core, instead of launching one job and hoping it will use multiple coresMy haircut instantly turned to spike-style when I read that. The interdependencies nightmare is just too frightening. We can have intersecting logfiles, temporary output files and job interdependencies that you don't know about. E.g. it must not be possible to launch an audio dependent video job prior to completion of the audio job. Any interdependent job needs to run in the proper order (but you can break that by moving jobs around). Furthermore, what about progress during simultaneous encoding? What about picking the next ready job to be processed? Now suddenly you add another layer of dependency that is called "don't touch that series because it's being encoded by another "me"".
And then suddenly a job starter wouldn't just have to check for the usual, but check each input and output file for the proper read/write permission to ensure that encoding can proceed.. and then of course it can always happen that the check succeeds, and due to the multithreaded nature, the other job processor starts just in that instance, is a bit faster, goes through the checks as well, everything passes, and then once encoding starts, you have a big explosion because two processes try to write to the same file or one tries to write to a file the other is trying to read from.

Sharktooth

20th February 2006, 15:36

lol... well... it's a bit tricky... :D

Richard Berg

20th February 2006, 18:36

On my multi-cpu machines, I just run 2 copies of MeGUI (from separate folders, of course, so the xml files don't conflict). It's not that hard, and won't be very common right now.

If we want to support the scenario within MeGUI, the best way would be to have multiple completely independent queues. Each would have its own tab in the UI and its own progress window.

Sharktooth

20th February 2006, 18:56

uhm... that way it's ugly.
well, i think this idea should be dropped at this point. we can already default megui to use always 2 threads for x264 even on single core non hyperthreaded cpus (for the reasons explained above).
and we can detect if there are more than 2 execution units and rise the number of threads to 4 for quad core or dual core hyperthreaded cpus...

Morte66

20th February 2006, 19:22

On my multi-cpu machines, I just run 2 copies of MeGUI (from separate folders, of course, so the xml files don't conflict). It's not that hard, and won't be very common right now.

Me too, it gets me from about 4.5 to 6fps on the stuff I've been doing lately. That's what made me put the request in. Oh well never mind, and thanks for looking at it guys.

dimzon

20th February 2006, 19:26

uhm... that way it's ugly.
well, i think this idea should be dropped at this point. we can already default megui to use always 2 threads for x264 even on single core non hyperthreaded cpus (for the reasons explained above).
And suspicions can be checked up:
We (rissain comminity ) take 2 clips coded with identical adjustments, in 1 and in 2 slices and we measure speed of their decoding on 2 processors nero the decoder (which supports SMP).

1 slice- 169 fps (both of the processor are loaded)
2 slice - 167 fps (both of the processor are loaded)

Sharktooth

20th February 2006, 19:55

dimzon

20th February 2006, 20:04

interesting. maybe nero found a way to multithread the decoder using something different than slices. never looked into it...
however qt uses slices for that purpouse (but as we all know QT = coded like crap...).
Elecard h264 encoders perform multithread encoding without slices too
Slices is easyest but not best way for multithreading

Sharktooth

20th February 2006, 20:48

indeed... but since i never used the nero decoder i thought it was using slices (like quicktime).
well, if all the mess is for accelerating quicktime playback we can drop even the 2 slices idea for single/non-multithreaded cpus.

foxyshadis

20th February 2006, 23:41

Well, Doom9, what about integrating Elder as a possibility? It isn't finished, but it already deals with the multiple encodings, segments, stats files, merging, and all that rubbish for you. And it works for xvid and x264. Obviously it wouldn't be bundled by default, since it's perl, but it'd be really nice to see it officially supported. Could just flip a switch and use its brand of massively multithreaded encoding rather than threads. (And since xvid and x264 support mt encoding anyway, once xvid makes it configurable you could move #threads to a global #cores option.)

dimzon

20th February 2006, 23:47

i believe it's easy to port Elder from rerl to C#
i'm interested in it

Richard Berg

21st February 2006, 04:06

Integrating Elder looks even more difficult than trying to sort out the dependency difficulties discussed earlier.

The more I think about it, the more I like having independent queues. The existing OO design should make it reasonably easy.

Sharktooth

21st February 2006, 08:32

IMHO idipendent queues is an ugly solution.

dimzon

21st February 2006, 08:46

IMHO idipendent queues is an ugly solution.
Lets wait for Doom9 refactoring first

Mutant_Fruit

21st February 2006, 09:11

IMHO idipendent queues is an ugly solution.
I agree aswell. I wouldn't like the idea of an app i was running having 4 independant queues because i have a dual-core dual-cpu machine.

How about this to help solve the dependancy issue... If a "job" is created, instead put all the jobs and dependant jobs inside an arraylist (just for an example, there's probably better ways of doing it. So we can go something like...

Arraylist Job1 = new Arraylist();
Job1.Add(job1_pass1);
Job1.Add(job2_pass2);
Job1.Add(job3_pass3);
....

Then...
foreach(Job job in Job1)
{
display information in the queue
}

Then, if someone wants to multithread by running two jobs at once, you just have to check that you aren't running two jobs from the one arraylist at the same time. Also, if someone has 10 "Jobs" in the queue, re-ordering them could be made safer because instead of having 20-30 unlinked jobs, we'd now have 10 Arraylists. When they'd try to move "job7_pass3" upwards, instead we'd move the Job7 arraylist above the Job6 Arraylist, then we'd print out the queue again, thus keeping Job7's jobs intact.

I don't know how it works at the moment, but how does that sound? Stupid or not?

dimzon

21st February 2006, 09:23

Actually one job series can contain:
multiple independed audio jobs
2-3 depended video jobs (multipass)
mux job

in this case you can run multiple audio jobs at same time

talking about such architecture we are close to "Processor controlled via Data-Flow" ideology (used in all modern CPU)

what does we need to organize such functionality:
jobs must contain references to related jobs (i.e. 2-nd pass must refer on 1-st pass, mux must refer to 2-nd pass and all audio etc)
in this case we can run any job wich related jobs is already done

berrinam

21st February 2006, 10:29

IMHO idipendent queues is an ugly solution.
My main gripe with independant queues is that the user would have to be actively thinking about which queue is likely to take the longest and would have to do his/her best to ensure that the encoding time balances up. This is IMHO too much work for the user who is only interested in getting a good end-result video.

How about this to help solve the dependancy issue... If a "job" is created, instead put all the jobs and dependant jobs inside an arraylist (just for an example, there's probably better ways of doing it. So we can go something like...
....
I don't know how it works at the moment, but how does that sound? Stupid or not?
I like this suggestion because it is simple. The issue about audio jobs not relying on each other, which dimzon raised, is not a huge concern, IMO, because the time taken for encoding audio is so small relative to video encoding.

what does we need to organize such functionality:
jobs must contain references to related jobs (i.e. 2-nd pass must refer on 1-st pass, mux must refer to 2-nd pass and all audio etc)
in this case we can run any job wich related jobs is already done
Didn't you suggest this sort of thing before? I agree that this is the optimum solution in customisability and reliability, but probably the most complex code-wise.

dimzon

21st February 2006, 11:53

but probably the most complex code-wise.
I don't think so - proper OO design can make it easy

Doom9

21st February 2006, 16:12

in this case you can run multiple audio jobs at same timeNo, you cannot, because at the time you pass from audio to video encoding, that's when a bunch of postprocessing routines kick in that make sure you end up with the proper video bitrate.. instead you create an undeterministic waiting period from the time the first audio job has been processed, to the time the second has and the queue can continue.
Actually, it's probably not even as hard to run two queues.. as the statusupdate contains a reference to the job that triggered the update, and related jobs are already linked (they need be for out-of-order processing and proper between job updates).
what does we need to organize such functionality:A job is a part of a linked list.. always has been for about a year now.. you can always get to the start of a series of jobs and to the end of it.. it takes no special treatment. It wouldn't even take too long to change existing functions to start/end jobs and the whole markJobxyz shebang to permit two jobs being processed in parallel (admittedly you have to place locks everywhere or it'll end up in disaster, but if you're a creative locksmith, you'll manage). What I find very perxplexing is that not a single person here but me is even thinking about the most basic question of how the heck you keep a user appraised of what's going on with two jobs in parallel, and how you need to explain a user that suddenly there's only one job being processed when there are 3 left in the queue. And on top of that, what's wrong with opening explorer, select your megui folder, press control-c, go to another folder, press control-v and you now can run megui two times completely independent without no worries whatsoever and not a single line of code needs to be written? What's wrong with the KISS approach for the 0.00001% of the users who cannot live with stock functionality?

cc979

1st March 2006, 17:36

just tried the 0.2.3.2097 devbuild works pretty good since the last time i tried, been busy with other things just wondering if there's any plans to show a rendering rate like the same as the video part ?

would be useful for benchmarking and comparing results with others

Doom9

1st March 2006, 20:12

just wondering if there's any plans to show a rendering rate like the same as the video part ?And what would a rendering rate be? I understand rendering in terms of POVRay.. essentially an fps rate.

dimzon

1st March 2006, 21:31

And what would a rendering rate be? I understand rendering in terms of POVRay.. essentially an fps rate.
encoding_time/audio_duration

cc979

2nd March 2006, 04:42

And what would a rendering rate be? I understand rendering in terms of POVRay.. essentially an fps rate.

something like kBps per second, processed/rendered sorry about that

Doom9

2nd March 2006, 08:21

encoding_time/audio_durationBut what does that ratio signify? The only thing I can imagine that would make some sense is indicating a X times realtime indicator. Also, wouldn't it be possible to have a progress indicator even in the first pass? I mean, you should be able to get the track length from avisynth, should you not?

berrinam

2nd March 2006, 09:40

From the AviSynth documentation for the 'Normalize' filter:
The calculation of the peak value is done the first time the audio is requested, so there will be some seconds until AviSynth continues.So the first pass is done all at one point, so MeGUI can't really get any indication of the progress at any point.

dimzon

2nd March 2006, 09:48

The only thing I can imagine that would make some sense is indicating a X times realtime indicator.
Yes, i think it's good to have such information

I mean, you should be able to get the track length from avisynth, should you not?
Yes, I can get he track length from avisynth, no problem here (actually I can get samplerate and samplecount -> i can transform it to duration )

ChronoCross

6th March 2006, 08:21

foxyshadis

6th March 2006, 10:09

I'd like to have crf mode save stats files. It's possible by custom commandline but it wouldn't hurt encoding at all to just save it by default.

The larger issue, 2-pass based on crf, is quite complex because you can't write the second job until you have the bitrate of the first, so no need for that. I go by "if the crf is good enough keep it, otherwise run a second pass to squeeze more quality out". It's slower but I don't always run the second, plus it's a less arbitrary bitrate target.

foxyshadis

6th March 2006, 20:40

ChronoCross

6th March 2006, 21:05

2nd req: Can Exhaustive mode be removed from megui? (Whether or not it's removed from normal x264.) I doubt there's any purpose in leaving it in there in megui, when people will see it and think a slower encoding mode will improve quality. (When it's only infintesimally better than umh.) For backups and normal encoding it has no place at all.

The option should still be left available to those who wish to use it. Even if it is a small difference some people are interested in the absolute best quality.

Richard Berg

6th March 2006, 21:10

We just need to make sure the default profiles based on it. I have a feeling 90% of users stick to Sharktooth's settings, and the ones who don't are on their own :)

foxyshadis

6th March 2006, 21:14

People can copy it into the command line if they need it that badly. I'm not sure why they'd use megui if they need something like esa.

ChronoCross

7th March 2006, 00:58

People can copy it into the command line if they need it that badly. I'm not sure why they'd use megui if they need something like esa.

ease of use. I use megui for that reason. I could do everything by hand but I prefer being able to quickly change my settings.

foxyshadis

7th March 2006, 01:21

In the other thread I suggested renaming it debug, which could dissuade people who think it provides a noticeable jump in quality, like moving up from dia or hex to umh. How about doing that instead?

ChronoCross

7th March 2006, 01:38

In the other thread I suggested renaming it debug, which could dissuade people who think it provides a noticeable jump in quality, like moving up from dia or hex to umh. How about doing that instead?

Yeah but like I said earlier. what is the point. it technically can provide better quality. pengvado said that the others sometimes beat it but not always. it's technically the best one you can use. we should just leave it as is. It's still useable and not really a debug mode.

It was also pointed out that most people just use the profiles as is. Which is the truth. the only people using it are people who know what it does.

berrinam

7th March 2006, 05:49

I agree that it is just a debug mode and it would discourage people from using it if it was removed from MeGUI. I agree that there a quite a few people who think they should enable everything, and then complain about how slow it is.

However, MeGUI is the only GUI at the moment which supports all of x264's features, and I don't think that we should give up on that, just for the sake of better user choices.

I think the only bridge between these two problems is some form of warning the user that it is pointless. However, I think the beginning user will just ignore any such warning, so it comes down to the choice between limiting the program and trusting the users, and I prefer to trust the users, as long as they are given a warning.

berrinam

7th March 2006, 05:52

MAke it so the source analyser uses "idle" system resources. I have it set to analyse the whole clip. when doing so I noticed that it sucks up all my CPU. Can you make it idle like virtualdub or x264 does while encoding? Thanks
What do other people think about the best way to manage this? I personally don't want mine to run on 'idle', because I want it to have priority over background tasks. I could put a choice somewhere, but where? Settings?

berrinam

7th March 2006, 05:52

<Two requests about crf mode>
Added to the list.

ChronoCross

7th March 2006, 06:35

What do other people think about the best way to manage this? I personally don't want mine to run on 'idle', because I want it to have priority over background tasks. I could put a choice somewhere, but where? Settings?

The source analyser settings. That would make sense to have it there. I like having it on idle just so my computer doesn't have a cow checking email while analysing a long source.

berrinam

7th March 2006, 06:36

Yeah, I think that's probably the best place, but still.... is it really necessary?

Just out of interest, how long does it take to analyse a long source?

ChronoCross

7th March 2006, 06:58

Yeah, I think that's probably the best place, but still.... is it really necessary?

Just out of interest, how long does it take to analyse a long source?

29 mins === approx 1 hour per pass.

berrinam

7th March 2006, 07:06

So much? I just timed on a whole DVD, and it took me 2 minutes and 11 seconds.

1. Do you have a slow/hi-res input?
2. Have you changed the settings for source detection, so it analyses more than 1% of the film?

ChronoCross

7th March 2006, 07:10

I have it analysing 100% of the film. that's why it takes so long. but I'm just being thurough.

Doom9

7th March 2006, 11:59

geez.. doing the whole movie is in the same league as using esa me ;) But I agree that everything should be runnable at idle priority.

ChronoCross

7th March 2006, 19:25

geez.. doing the whole movie is in the same league as using esa me ;) But I agree that everything should be runnable at idle priority.

That's why I made the some people are anal retentive about quality testing statement in the x264 and megui discusion threads about esa lol

berrinam

7th March 2006, 20:44

@ChronoCross: If you're willing to subject yourself to such long analysis, would you perhaps be willing to share your results with me? My last commit added diagnostics to Source Detection, so it shares much more information about what it's doing now. I wonder how much more accuracy you get by doing 100% as opposed to 1%.

If you're interested, I simply ask that when you analyse something, analyse it both at 100% and then at 1%, and show me the results. The idea is that they should be the same, but who knows?

ChronoCross

7th March 2006, 20:51

yeah sure. I'll run it now.

fogbav

13th March 2006, 22:07

Automatic Encoding All .TS Files and MPG Files from A Directory ans Subdirs ... Encode Video with fixed PRofile XXX AC3 with fixed Audio Profile YYY and all MP2 Tracks with Audio Profile ZZZ ...

Would be very usefull for all Dbox2 Owners recording streams to a special Folder an get them compressed when they are at work ..8)

Tagging working Folders make it possible to encode from different PCs so 2 or more "Encodings" can happen trough a Network share to the "recording folder"

berrinam

14th March 2006, 06:05

I'm not Doom9, but I'm pretty sure that I know what his response is going to be: no way. It requires a huge amount of organisation and has so many ways that the user could stuff it up, and what are the benefits? It saves a tiny amount of time in user setup (of course, the user could set it up with the one click encoder in MeGUI anyway, which is already very easy) for some few people who havze Dbox2s.

fogbav

14th March 2006, 19:34

Hmm ... sad to here this ... i programmed some lines with kixtart and everything work very well with all command lines put up in some "spagetti code" programming ... 8)

ts -> ProjectX -> Xvid/mp4 (over ini file - configure Codec +passes)-> Audio --> AAC 5.1 and AAC v2 with winamp codec -> Mp4box -> delete org files ...

but my programming know how is not the best ... so I thought it was a good idea to request this to the profis here in this way ...

Maybe a can spend some time with AUTOIT to transfer all files to the "one Click" encoder in megui and get my autoenconding ...

I have a time eating job and 2 kidz so the only way to get space free on my HD is to programm something that is doing the job for me ... and .. everything exists in Megui / StaxRip etc ...

anyway ... thanks

FoGBaV

dimzon

14th March 2006, 19:43

time eating job and 2 kidz so the only way to get space free on my HD is to programm something that is doing the job for me ...
Or teach Your wife how to use MeGUI ;)

ChronoCross

14th March 2006, 20:12

Request: A Button that allows for megui to always be shown in the tray.

berrinam

15th March 2006, 07:09

Request: A Button that allows for megui to always be shown in the tray.
Why?

Also, do you have any news on the Source Detector? Just to jog your memory:

@ChronoCross: If you're willing to subject yourself to such long analysis, would you perhaps be willing to share your results with me? My last commit added diagnostics to Source Detection, so it shares much more information about what it's doing now. I wonder how much more accuracy you get by doing 100% as opposed to 1%.

If you're interested, I simply ask that when you analyse something, analyse it both at 100% and then at 1%, and show me the results. The idea is that they should be the same, but who knows?
yeah sure. I'll run it now.
If you haven't got any results, then never mind for the moment, because I'm planning on slightly modifying the Source Detector again, so that it uses tritical's IsCombedTIVTC, which is more accurate than DeComb's IsCombed.

berrinam

15th March 2006, 07:11

A feature request of my own:

Add extensive commandline implementation for MeGUI. Although it is meant to be a front-end and nothing else should need to run it, commandline support could lead to integration into the shell, meaning you could right-click on a media file and select 'Open With MeGUI' or 'OneClickEncode with MeGUI', etc.

stax76

15th March 2006, 10:37

Add extensive commandline implementation for MeGUI. Although it is meant to be a front-end and nothing else should need to run it, commandline support could lead to integration into the shell, meaning you could right-click on a media file and select 'Open With MeGUI' or 'OneClickEncode with MeGUI', etc.

Maybe you like my CLI architecture, it has a nifty parser as well as a command engine like Visual Studio, the command engine is powered by attributes and reflection and drive both, the menus and the CLI. Adding a new CLI switch is as simple as adding a new method:

<Command("Show Message Box", Nothing)> _
Public Sub ShowMsgBox( _
<Name("Message"), Description("The message may contain macros."), Editor(GetType(MacroStringTypeEditor), GetType(UITypeEditor))> _
ByVal message As String, _
<Name("Title"), Description("The title may contain macros."), DefaultValue("StaxRip"), Editor(GetType(MacroStringTypeEditor), GetType(UITypeEditor))> _
ByVal title As String, _
<Name("Icon"), DefaultValue(GetType(MsgIcon), "Info")> _
ByVal icon As MsgIcon)

Msg.Show(Macro.Solve(message), Macro.Solve(title), icon)
End Sub

After adding the method the command is now available at the menus (http://www.planetdvb.net/staxrip/screenshots/038.png) as well as at the command line:

StaxRip.exe -ShowMessageBox:a_string,"a string in quotes",Info

That means no manual commandline parsing and all types with a TypeConverter will work which include primitive types (IConvertible) and enums and lots of other framework types like Point, Size etc.

As you can see it uses the syntax of the .NET framework tools, the syntax rocks from both, user and development (parsing) perspective. The parser is not case sensitive and supports '/' as prefix like: /showmessagebox:arg,arg2...

berrinam

16th March 2006, 12:03

What do other people think about this: ability to customise scripts used for Source Detection.

Basically, this would mean that instead of having two scripts hard-coded into MeGUI, these two scripts would be accessible via the Source Detector settings. The scripts could be edited by the user and MeGUI would accept them as long as they returned the correct variables within AviSynth.

Pros:
-Customisability -- it could make it easier for the user to change the Comb detection (or something else) to a more accurate method. For instance, my next update will include a change from IsCombed to IsCombedTIVTC, which is much more accurate.

Cons:
-More work for me.
-Many more ways to stuff up.
-Mainstream users probably won't have a clue/won't wont to experiment with these scripts anyway, so for most people it is just a useless feature.

I'm leaning towards not doing it, because it's more work for me and I can't see much in the way of tangible benefits. If someone can give me a good reason to allow it, though, I'll consider going the whole way and giving the user complete control over these scripts.

Sharktooth

16th March 2006, 13:27

uhm... naah.

ChronoCross

16th March 2006, 19:36

same....by being able to customize it, it means you have enough knowledge of IVTC just to do it manaully anyway.

Bathrone

19th March 2006, 08:27

Im just learning MeGUI and AVC encoding so apologies for any user errors :) Also, love the work so far on MeGUI and x264 :)

How about extending the mux capabilities to be able to demux as well? And be able to mux in more streams like an additional audio one after the file has been done.

And with the D2V creation, being able to put start and end points rather than having to go into DGIndex to set it if the user doesnt want to transcode the whole thing.

I wonder too if there is a way to prevent user's changing Sharktooth's profiles by mistake?

ChronoCross

19th March 2006, 09:44

I wonder too if there is a way to prevent user's changing Sharktooth's profiles by mistake?

You can't currently change the profiles. If you make a change tp the profile it's simply made to the settings. if you reselect the same profile whatever you changed will change back to what the profile is.

berrinam

19th March 2006, 10:43

How about extending the mux capabilities to be able to demux as well? And be able to mux in more streams like an additional audio one after the file has been done.You should use a dedicated muxing tool to do this -- YAMB for MP4, or mkvToolnix for Matroska.

And with the D2V creation, being able to put start and end points rather than having to go into DGIndex to set it if the user doesnt want to transcode the whole thing.That's really just video cutting. It's not so important to do it in DGIndex, as long as it is done somewhere before the encoding is done. This cutting will eventually be implemented in Avisynth, but it will probably be a while from now, so you'll just have to be patient until then.

I wonder too if there is a way to prevent user's changing Sharktooth's profiles by mistake?
You can't currently change the profiles.That's obviously not right. How would anyone manage if they couldn't changing them. I think I know what you mean, though. If you select a profile in the main window, then press config, modify it and press ok, then you won't have changed Sharktooth's profile. However, you shouldn't rely on this, because this is in fact a documented bug.

In fact, there was a feature called Safe Profile Alteration, but I believe it has been removed from MeGUI because it was a hassle and not particularly useful to us. It would be a good idea to check the settings to see if it is actually there, but if it isn't, then it's just up to the user not to fiddle with what they don't understand.

dimzon

20th March 2006, 06:48

You should use a dedicated muxing tool to do this -- YAMB for MP4, or mkvToolnix for Matroska.
Maybe we can add support for this into MeGUI? Why not?

berrinam

20th March 2006, 06:56

Maybe we can add support for this into MeGUI?What, specifically, do you mean by this?
Why not?What does it add to MeGUI's purpose, which is to be a frontend for encoding?

In particular, what advantage is there in integrating it into MeGUI as opposed to running it as a separate, already-working tool? It's good having primitive muxing in MeGUI, because it means that entire encodes can be managed automatically. But unless demuxing can be somehow integrated into an automated workflow in MeGUI, then it saves the user basically no work by our putting it in MeGUI, but it takes a lot of work for us to replicate YAMB's code (or mkvmergegui's, which is even more powerful).

dimzon

20th March 2006, 07:39

In particular, what advantage is there in integrating it into MeGUI as opposed to running it as a separate, already-working tool? It's good having primitive muxing in MeGUI, because it means that entire encodes can be managed automatically. But unless demuxing can be somehow integrated into an automated workflow in MeGUI, then it saves the user basically no work by our putting it in MeGUI, but it takes a lot of work for us to replicate YAMB's code (or mkvmergegui's, which is even more powerful).

MkvMerge has it's own GUI so I doesn't propose to implement yet another Matrosska muxer

But YAMB is "third-party" tool. I really think it's better to invite Kurtnoise to join in MeGUI development and move all YAMB functionality into MeGUI... YAMB is not stable until MP4BOX is beta. So Kurtnoise must perform adaptation sometimes for ne MP4BOX version. In other case we must support muxing part of code too... So joining MeGUI and YAMB we will get all code in one place, it's much easy to support

berrinam

20th March 2006, 07:53

In other case we must support muxing part of code too... So joining MeGUI and YAMB we will get all code in one place, it's much easy to support
It's not really going to be such a big hassle to change MP4Box code in MeGUI -- all of the relevant code should be contained in the one generateMP4BoxCommandline function. However, integrating YAMB into MeGUI could be a fair bit more work, and there is still no benefit for the user. Also, it just makes MeGUI even more bloated.

dimzon

20th March 2006, 08:07

However, integrating YAMB into MeGUI could be a fair bit more work, and there is still no benefit for the user. Also, it just makes MeGUI even more bloated.
I doesn't think is too hard to implement it. And I really dislike to use too much applications at once - so I prefer to add additional functionality to MeGUI. And other users doe't like it too. Really you need different applications to encode / mux / split / extract MP4. In opposit way VirtualDub perform everything in one application. This (usability/comfort) is one of reason why users prefer x264 VfW and AVI container. Anycase this work can be done/planned with lowest priority...

berrinam

20th March 2006, 08:37

I can partly understand your reasons for wanting it.... just two final questions:
What exactly do you have in mind for it?
If we are going to put the level of flexibility with editing that VDub allows into MeGUI (which I think is a good idea for SOME application; not necessarily MeGUI), then wouldn't it be better to go the whole way and support matroska cutting as well. Once the GUI is done, the generation of the commandlines are trivial by comparison, as long as the code is well-designed.

Doom9

20th March 2006, 09:42

In fact, there was a feature called Safe Profile Alteration, but I believe it has been removed from MeGUI because it was a hassle and not particularly useful to us.No, it's still there, but I noticed that it hasn't been updated when a codec's option get updated. And profiles are updated just fine.. just switch between two of them in a codec configuration dialog.

but it takes a lot of work for us to replicate YAMB's codeDid you say lot? You probably mean LOOOOOOOOOOOOTS.. it would add another dimension to the "ideal mux path" finding (and the ideal encoder output finding.. that's already two dimensions too many).. and that code is making me insane as it is.

dimzon

20th March 2006, 10:41

it would add another dimension to the "ideal mux path" finding
Why? Maybe Your code have suboptimal architecture? Anycase we must wait until Your commit...

berrinam

21st March 2006, 22:21

A few things to consider:

If MeGUI is meant to target an audience that knows nothing about video encoding (that's what I imagine the One Click Encoder is for), then it might be worthwhile adding some kind of 'quality' scheme, as a %, like other commercial programs do (wmeconder, for example). This would basically be implemented as a codec-specific mapping from a quality % to quantizer or crf-type mode.

Add extra support for Drag'n'Drop in the other windows. So, in a mux window, you could drag all the inputs in.

Also, I wonder how useful the 'and close' feature in the D2V creator and One Click Encoder is. Perhaps it would be better if it were checked by default, because the default behavior ATM is that pressing Queue/Go gives no indication that anything has happened. In fact, MeGUI has several quirks like this that make it confusing for the new user.

kurt

22nd March 2006, 14:33

Feature Request for xvid_encraw: cqm support with squid_80's build should work (maybe in cvs too?) - could this also be included in MeGUI?

and what about adding a custom commandline option in the xvid encoder dialog? (this feature exists for x264 under zones)

thx in advcance
(sorry, if this has been asked before)

Doom9

22nd March 2006, 16:13

Feature Request for xvid_encraw:all of the options of squid's build are already supported in the code on my harddisk.. but it's other parts that are extremely broken so it will take weeks if not month before the whole thing is usable again. Add to that that I'm under constant deadline pressure at work so at home I really don't feel like writing more code.

kurt

22nd March 2006, 16:22

ok, Doom - thx for the clarification and there's no need to rush :)

cc979

22nd March 2006, 21:39

is it possible, on the log screen to have a clear log function because would quicker than reloading it, easier to kept track of things on multible encodes and it would save you from re-opening files too

berrinam

23rd March 2006, 09:32

is it possible, on the log screen to have a clear log functionYep, it's possible.
because would quicker than reloading it, easier to kept track of things on multible encodes and it would save you from re-opening files tooI don't understand what you are saying.

cc979

23rd March 2006, 17:55

cc979

23rd March 2006, 18:49

after reading thru the dev section and seeing all talk macros, magic numbers and things with saved settings and that

could it be possible to save the script from the main screen in some sort of xml file with all the settings, SAR, DAR included so you can encode stuff later and not worry and possibly with option to re-save it as as .avs script

cc979

24th March 2006, 16:33

just curious do the anime presets in Convolution3DYV12.dll work ok, i have added them to the ScriptServer.cs file seems ok

Ogig

26th March 2006, 17:19

(Didn't know whether to post in bug thread or here, hope it's the right place.)

When using the Avisynth Script Generator I noticed that 720 is the maximum width. Is there a reason for this limitation ?

GmorG McRoth

28th March 2006, 13:10

I'm not sure if somone else mentioned it already (If so, I'm with him), but could it be possible to change color of font in status window? currently is unreadable in most of windows themes I like to use. Screen shot (http://gmorg.mcroth.googlepages.com/window.jpg)

berrinam

29th March 2006, 10:23

When using the Avisynth Script Generator I noticed that 720 is the maximum width. Is there a reason for this limitation ?Yep -- your source is only 720 pixels wide, and there's no point in upsizing unless you use a very advanced upsizer like iiP, in which case MeGUI can give you no help.

I'm not sure if somone else mentioned it already (If so, I'm with him), but could it be possible to change color of font in status window? currently is unreadable in most of windows themes I like to use. Screen shot (http://gmorg.mcroth.googlepages.com/window.jpg)
No. The font works well in the standard Windows themes, and there have been many other people posting skinned screenshots with no troubles like that. I don't know, but I'm pretty sure that MeGUI skins properly, so if there's something up, I would say there is something wrong with your font. And it simply requires way too much work to make every font skinnable. Perhaps dimzon can contribute his experience with HTMLayout (http://forum.doom9.org/showthread.php?p=806360#post806360) to helping your problem, but other than that, I would say it is way too much work.

berrinam

29th March 2006, 10:25

i must have been really tired, i meant it would be easier to clear the log screen - than reloading megui with opened avs files - so you could make profile changes quickly on test encodes and at a quick clear log encode and see problems quickerI can't really see the benefits, but I don't imagine it would be too much work, either, so I've added it to the list. I'm also considering a different approach to the log (see my next post no, don't. I changed my mind) which may mean this isn't necessary.
could it be possible to save the script from the main screen in some sort of xml file with all the settings, SAR, DAR included so you can encode stuff later and not worry and possibly with option to re-save it as as .avs scriptWhat is there to store other than SAR/DAR?

dimzon

29th March 2006, 10:38

Perhaps dimzon can contribute his experience with HTMLayout (http://forum.doom9.org/showthread.php?p=806360#post806360) to helping your problem

Actually I want test IUI paradigm and HTMLayout in BeHappy and (if success) propose it to MeGUI ;) Actually current MeGUI look and feel is too complex for novice.

berrinam

29th March 2006, 11:46

Two AviSynth-related ideas:

Cropping can be done by the resizer. This is undoubtedly faster, and I think it also allows for mod1 cropping, which is much better than at the moment. To integrate this into the script creator (which I think we should), the templates would need to be reorganised, because there is no guarantee that they are going to be consecutive. I propose the addition of another tag, <resize-crop>, which will insert the resizer with cropping if relevant.

According to foxyshadis, ColorMatrix() is a good idea on most sources, as DVDs tend to use Rec.701 and MPEG-4 uses Rec.601 (I have no idea about the numbers or what they mean, I'm just taking it on trust). More importantly, DGDecode hints the stream, so that ColorMatrix() will autodetect the right thing to do. So it won't be bad in any cases. I think, then, that ColorMatrix should be checked by default, not unchecked.

I said I was going to propose another idea about the log. I've changed my mind -- I don't like it.

Actually I want test IUI paradigm and HTMLayout in BeHappy and (if success) propose it to MeGUI ;)Yes, sure. I was just saying that this might be a solution sometime in the future.

Actually current MeGUI look and feel is too complex for novice.I totally agree. MeGUI still needs a lot of work, but it is very promising.

@dimzon: Would you be willing to implement this:
Be able to choose the sampling rate for audio
Description: See quake74's post (http://forum.doom9.org/showthread.php?p=764825#post764825)
Status : Nobody is working on it.

dimzon

29th March 2006, 12:23

@dimzon: Would you be willing to implement this:
No problem - but I'm waiting Doom9 refactoring first (to drop out BeSweet support). Actually current Doom9 refactoring is a huge bottleneck for development...

GmorG McRoth

29th March 2006, 14:14

No. The font works well in the standard Windows themes, and there have been many other people posting skinned screenshots with no troubles like that. I don't know, but I'm pretty sure that MeGUI skins properly, so if there's something up, I would say there is something wrong with your font. And it simply requires way too much work to make every font skinnable. Perhaps dimzon can contribute his experience with HTMLayout (http://forum.doom9.org/showthread.php?p=806360#post806360) to helping your problem, but other than that, I would say it is way too much work.

I know I have no right to complain, It is free software. I was just asking for improvement. I have little to no programming knowlage, I was thinking that changing font color is not a big deal, but if you say, its much to rewrite, to change one font, it's indeed bigger chore than it's worth. I attach, for your consideration screenshots of other skins.

windows default:
XP Blue (http://gmorg.mcroth.googlepages.com/XPblue.jpg)
XP Olive (http://gmorg.mcroth.googlepages.com/XPolive.jpg)
XP Silver (http://gmorg.mcroth.googlepages.com/XPsilver.jpg)
Windows Classic (http://gmorg.mcroth.googlepages.com/ClassicWindows.jpg)
Windows Classic High Contrast (http://gmorg.mcroth.googlepages.com/Classicwindowshighcontrast1.jpg) (works best XD)

Custom Thames:
Lamina RC (http://gmorg.mcroth.googlepages.com/LaminaRC.jpg)
ReLuna - Bluetiful (http://gmorg.mcroth.googlepages.com/RElunaBluetiful.jpg)
ReLuna - Em3lent (http://gmorg.mcroth.googlepages.com/window.jpg)
Sustenance - Slate (http://gmorg.mcroth.googlepages.com/Sustenance-Slate.jpg)

I hope I did not taken too much of your time.

foxyshadis

29th March 2006, 14:39

The problem is more that it uses the disabled text color than that it uses grey, making it harder to read than it needs to be. Do they have to be disabled textboxes instead of captions?

Ogig

29th March 2006, 19:44

Yep -- your source is only 720 pixels wide, and there's no point in upsizing unless you use a very advanced upsizer like iiP, in which case MeGUI can give you no help.

What about anamorph encoded content. See http://forum.doom9.org/showthread.php?t=106987 for details.

berrinam

29th March 2006, 21:10

What about anamorph encoded content. See http://forum.doom9.org/showthread.php?t=106987 for details.What about it? Just check 'Retain full resolution and set SAR in encoder' and then you're done. Still no need to upsize.

thuongshoo

1st April 2006, 17:29

I have just tried to use Megui . Speed of it isn't quick .Only 2 frm/s . Can you add "stream data pipeline " ?
http://img117.imageshack.us/img117/9162/pipeline7xs.th.png (http://img117.imageshack.us/my.php?image=pipeline7xs.png)
thanks !

Doom9

1st April 2006, 19:20

megui has very limited influence on speed.. it does nothing complex with the stdout from the encoder. And those settings would affect an encoder.. megui is just a control software, it has no influence on buffers within an encoder software.

Actually current Doom9 refactoring is a huge bottleneck for development...I posted the sources on Thursday so that you can help speed up the process ;)

@GmorG McRoth: If you can show me that MeGUI violates Microsoft suggested design guidelines by using read-only textboxes, this is a problem of whomever designed those skins and any other software using read-only textboxes will run into the exact same problem.. so the skin's bad, not the software. I have never liked skinnable software and anything that doesn't adhere to the platforms normal look and design.. breaking style makes it a lot harder for users to get to know a certain software.

GmorG McRoth

1st April 2006, 20:16

@Doom9 I don't know microsoft design guidelines, also I had no idea that they were "disabled textboxes" untill foxyshadis said so. For me grey text on grey background is design flaw so I tried to bring it to developers atention. As I see it's mostly microsoft foult for making "disabled textboxes" font color setting hiden somewhere in system (or in skin file). Ill try to fix this on my side then.

Thanks for MeGUI by the way it's great software despite this one thing.

thuongshoo

4th April 2006, 12:06

I'm not a programmer . If using x264.exe , can we make function "stream data pipeline" ?
Virtualdub can't use aac encoder and make .mp4 . What's a pity .

berrinam

4th April 2006, 13:49

I'm not a programmer . If using x264.exe , can we make function "stream data pipeline" ?
Virtualdub can't use aac encoder and make .mp4 . What's a pity .
I don't understand you. What do you mean by "stream data pipeline"?

EDIT: I read your earlier post and Googled it. I found this result:

(From http://www.virtualdub.org/docs_faster )
What do the Performance settings do?

Short answer: leave them alone.
...
Stream data pipelining controls the number of audio and video buffers in flight in VirtualDub's pipeline; this is the companion to the AVI output buffering setting. One buffer is used for one video frame, or one video frame's worth of audio data. Again, this value rarely speeds anything up if you're doing a slow operation like recompression, and raising it can have extremely bad consequences -- setting 256 buffers on a 640x480, 24-bit uncompressed RGB file could conceivably cause VirtualDub to allocate 235MB of memory. Increasing it on a direct stream copy operation with a highly compressed input could help, but that tends to already be speedy with the defaults.

So, that issue is not to do with MeGUI, it's x264's problem, and I would also say it isn't a problem -- x264 is just slow, because it is very powerful

jmk

4th April 2006, 14:14

i like that megui now calculates the --sar value for x264 according to what you put in into the "sample aspect ratio" in the config dialogue.

but would it somehow be possible to have a checkbox that disables the automatic calculation and passes the values that i input into those fields to x264 as the --sar values instead? (like it used to be...).

thx,
jmk

btw: i am using v0.2.3.2118

Doom9

4th April 2006, 15:53

according to what you put in into the "sample aspect ratio" in the config dialogue.it hardly does that.. rather it fills those values, but only if you tell it to. Unless you configure megui to signal dar in the avisynth script creator or one click encoder, whatever you configured in the two sar fields in the x264 configuration will be applied (and by default that's 1:1 so nothing will be set).

berrinam

4th April 2006, 22:07

it hardly does that.. rather it fills those values, but only if you tell it to. Unless you configure megui to signal dar in the avisynth script creator or one click encoder, whatever you configured in the two sar fields in the x264 configuration will be applied (and by default that's 1:1 so nothing will be set).
Actually, those fields are treated as DAR, Display Aspect Ratio (the labels might be wrong, though). This means that MeGUI does in fact convert them to sar just before encoding. I made it that way because I made a general decision that DAR is more useful than SAR. I think it is worth allowing either.

thuongshoo

5th April 2006, 02:44

I don't understand you. What do you mean by "stream data pipeline"?

EDIT: I read your earlier post and Googled it. I found this result:

So, that issue is not to do with MeGUI, it's x264's problem, and I would also say it isn't a problem -- x264 is just slow, because it is very powerful
"stream data pipeline" , is a fucntion , .Perhaps you don't understand "function" .:D
yes ,
Stream data pipelining controls the number of audio and video buffers in flight in VirtualDub's pipeline
So can you do the same in Megui ?
I want to have a function which named "Stream data pipeline" , controls the number of audio and video buffers in flight in Megui's pipeline :D
Virtualdub have 3 function concerns with "pipeline" , you can view the picture which I posted .
Virtualdub uses x264vfw while Megui uses x264.exe ,so I don't know whether Megui can boost speed as Virtualdub :D
Again, this value rarely speeds anything up if you're doing a slow operation like recompression, and raising it can have extremely bad consequences
yes , I agree with Averly Lee but it is not bad function completely . If I use 256 buffers (maximum) , output file can be bad but I use less 256 buffers , Virtualdub is ok .Adding , Virtualdub using .Avi as output container , it is said that .Avi is not a good container for x264
The speed actually increase if I increase "AVI buffer" .I use Virtualdub and the speed is often 16 frame/s while is only 2 frame/s for Megui

berrinam

5th April 2006, 08:47

I still don't completely understand what it does, however, I feel confident enough to say, "It has nothing to do with MeGUI". Make a feature request either to the Avisynth devs or the x264 devs for that.

As far as the actual processing goes, MeGUI does nothing. It just sits and watches x264 and Avisynth do all the work, and reports back with how much they've done.

Doom9

5th April 2006, 10:41

thuongshoo

5th April 2006, 12:09

You don't seem to understand how megui works. Megui has no pipelines.. it doesn't read video data.. x264.exe reads video data, encodes and writes video data.. it's all in one tool. All megui does is give parameters to x264.exe and display the status as reported by x264.exe.
yes .Thank for explaining . Before I also guess that Megui is only actually a Gui ,dont do anything , transcoding is job of encoder . Perhaps this is a advantage of x264.exe

thuongshoo

6th April 2006, 02:45

wrote by thuongshoo
Perhaps this is a advantage of x264.exe
I'm sorry !
Perhaps this is a disadvantage of x264.exe :D

Richard Berg

6th April 2006, 03:43

I'm not sure you understand how x264 (or VDub for that matter) work either. x264's backend will always be slower than disk I/O, so adding buffers would just be a waste of memory. It loads Avisynth, which does have PVideoFrames (equivalent to one of VDub's pipelines), but since it's a nonlinear editor the allocation is much much more complex than VDub. There's no way to fit the concept of a static # of pipelines into Avisynth: the number of frames and clips in the system is dynamic, and it also has an aggressive cache. The closest equivalent would be the SetMemoryMax function, but as the name implies, that's a maximum rather than a static #.

All of these programs are open source. If you come up with any workable ideas, submit a patch.

thuongshoo

6th April 2006, 05:19

Megui convert file by x264.exe and Avisynth . It is good .I have a mpeg-2 which can't be transcoded by Virtualdub but Megui can do well . I'm very surprise .
Averly Lee says that Using a large buffer make VD encode while it waits for disk I/O .
Oh ! all thing is theory . If increasing AVI buffer and pipeline , processing speed will increase . Have you try to do this with VD ?
VD uses x264vfw.dll while Megui uses x264.exe
Maybe we can't do this in MEgui because it use Avisynth . I guess that . I don't dare say because I'm a pro programmer :D

denise

6th April 2006, 08:54

difference setting between x264 MeGUI and x264vfw VD ---> difference speed.

Use MP4 box -aviraw video to extract raw h264 from .avi that made by VD . Mux new .264 and .aac in to .mp4 as you desire.

foxyshadis

6th April 2006, 09:22

Heavy pipelining is only useful for huge disk operations, on the order of tens of megabytes read and written a second. Capture and direct stream processing fall into this, heavy final encoding does not. x264 already uses asynch i/o afaik, which provides a big boost, and pipelining doesn't gain much over that.

He discusses it in this entry (http://www.virtualdub.org/blog/pivot/entry.php?id=31):

If your render operation is projected to take 14 hours and its speed is better measured in SPF than FPS, just leave the buffering settings alone. Tweaking them isn't likely to speed anything up.

thuongshoo

6th April 2006, 13:49

wrote by thuongshoo
because I'm a pro programmer
oh ! shock ! shock ! I'm sorry
because I'm not pro programmer :D

To denise : thank you ! I will try
To foxyshadis : I will read it .My english is n ot good so I need time to understand it .

Kayaker

6th April 2006, 14:33

I'd love to have vorbis sound suppont in MeGUI.

Until you resolve the avisynth sound support for Vorbis would it possible to add at least muxing of Vorbis sound with mkvmege ?

Cos mkverge is perfectly capable of doing so.

Thanks a lot for the BEST encoding GUI.

Bathrone

8th April 2006, 03:54

Just a useability suggestion on the UI - buttons could be made inactive until they can be actually used. e.g. Not having the job tab start button enabled until a job exists in the que.

Thanks to all involved in MeGUI, a great app that Im sure is setting the stage for our AVC future.

Kostarum Rex Persia

8th April 2006, 04:46

Hi guys, I have one reasonable suggestion for future MeGUI builds?

Instead supporting ONLY AVS input, devs should introduce direct work with AVI. MPG and other video files, without need for AVS script.

That's because most people doesn't need filters in AviSynth, so direct work with MPG and AVI files would be nice.

http://img438.imageshack.us/img438/6660/suggest9wo.jpg

Any good thoughs, devs. I don't need AviSynth scripts very often, I just want to encode in MP4 or MKV container directly, without AVS scripts.

chipzoller

9th April 2006, 23:20

I don't need AviSynth scripts very often, I just want to encode in MP4 or MKV container directly, without AVS scripts.

I agree this would be nice. Most of the time I don't NEED scripts but have to make them anyway.

ChronoCross

10th April 2006, 00:06

It does have a meinimize to tray already.

View --> minmize to tray

foxyshadis

10th April 2006, 01:49

There's nothing nice about his request, all that needs to change is the drag-drop (and file open) handlers need to be extended to accept and handle directshow types (mkv, mp4, mov, wmv, etc). And, possibly, avisynth input renamed to video input, except then we'd get the exact opposite requests from people asking for an avisynth input.

ChronoCross

10th April 2006, 02:02

Not to mention demultiplexing issues, Remuxing control, container capabilites(in temrs of handling audio and video streams).

There is a whole lot to consider if Megui were to start accepting all types of files.

chipzoller

10th April 2006, 02:12

It does have a meinimize to tray already.

View --> minmize to tray

My bad...never seen it under 'view'

foxyshadis

10th April 2006, 02:18

Well, the video just gets sent to a directshowsource call, so if that fails you just pop up a dialog saying "No system splitter or decoder available for this format! Please install [haali's and ffdshow|realplayer|WMP codecs]."

If you get beyond that into multiple streams, subtitles, chapters, etc, you're in for a world of pain for not much gain. ;_;

berrinam

10th April 2006, 14:10

Ok, done! You can now open any video source that works with DirectShowSource in a number of ways:
Go to the AviSynth script creator and open it there
Drag it onto MeGUI, which will open it in the AviSynth script creator
Use File->Open, which will also open it in the AviSynth script creator

At the moment, the only supported filetypes for DirectShow are *.avi, *.mp4, *.mkv, and *.rmvb. Everything else is accessible through the 'All files' filter. If someone wants to givve a me a better list of file extensions for DSS-loadable files, please go ahead.

Kostarum Rex Persia

10th April 2006, 18:54

Yuo don't inderstand my point berrinam. We, common users, want to open video files directly, without AVS interfering. I hate to make AVS script every time.

ChronoCross

10th April 2006, 19:09

Yuo don't inderstand my point berrinam. We, common users, want to open video files directly, without AVS interfering. I hate to make AVS script every time.

your an idiot. do you have any idea how much would be required to build a directshow decoder directly into megui??? Seriously man....it's best to do your decoding through avisynth. it allows us to deal with things better as the framework is already finished.

dimzon

10th April 2006, 19:20

your an idiot. do you have any idea how much would be required to build a directshow decoder directly into megui??? Seriously man....it's best to do your decoding through avisynth. it allows us to deal with things better as the framework is already finished.
maybe we can create AVS script impicitly without showing any avs-creation GUI

ChronoCross

10th April 2006, 20:09

true. but I think what KRP is asking for is basically virtualdub. Where the video file is handled implicitly where the end of encoding all the stuff is done already.

That would mean that megui would have to be able to differentiate between all tracks(video, audio, subs) Their given track names and other info be able to wrap that in avisynth with no user input, and then figure out how to keep them into the next container with the same attribs.

Considering all the things that can go wrong in this process(audio compatibility, video changes and whatnot) there would have to be a substantial amount of time put in to get every possible situation and prereq for each container both input and output.

cc979

10th April 2006, 20:47

@berrinam cheers for clear log button,
just noticed the resize disable nice touch - but maybe a idea to turn on resize automaticly if crop is used and its not a mod of 8/16(not remember which)

shon3i

10th April 2006, 21:15

maybe we can create AVS script impicitly without showing any avs-creation GUI
For example something like AutoGK

berrinam

10th April 2006, 22:13

@cc979: I dunno.... if I turned resizing off, I wouldn't want it to turn back on when I selected my cropping. You have a point, though.

@all: Yep, I am considering doing that implicit avs creation. There needs to be some way to smoothly integrate the two, so if people can come up with some ideas, that would be great.

The way that I thin the implicit avs creation would work would not be like AutoGK, as shon3i said, but simply a single input line. So, if you load a video file, then MeGUI creates an avs that has a DirectShowSource call (or AviSource, etc), with no resizing or cropping. If you want to configure it yourself, then you (somehow) tell MeGUI to open this with the AviSynth script creator, and if you want AutoGK-style automated script generation, you use the One Click Encoder (this could do with some extension, to support directshowsource files for input -- perhaps using the MediaInfo lib for track information).

cc979

10th April 2006, 23:01

@berrinam ... if autocrop is used and resizer is disabled maybe increase the crop so it is a mod of 16 or re-enable the resize or give a choice

depends on the encode maybe

Romario

11th April 2006, 02:52

your an idiot. do you have any idea how much would be required to build a directshow decoder directly into megui??? Seriously man....it's best to do your decoding through avisynth. it allows us to deal with things better as the framework is already finished.

Beside your bad maners:( , I support KRP request for direct opening of video files in future MeGUI builds. Common users needs that.

ChronoCross

11th April 2006, 03:14

your basically the same person so whatever. Besides ignoring the technical reasons for this MeGUI is NOT Virtualdub. Re-read a few of my posts and you'll understand why implementing it will be not worth the effort.

foxyshadis

11th April 2006, 03:47

If you need a virtualdub for newer formats, please, use avidemux! It was made to be a virtualdub-style editor for them. MeGUI is just an encoder that levarages smart algorithms and the power of avisynth to make your encoding easier, but it will never be an NLE.

Doom9

11th April 2006, 11:08

Here's the only thing that makes sense from a programmer's perspective: relax the filter to other types. If the file selected has the extension .avs, everything works as normal. If the extension is .avi, megui writes a new avs named yourmovie.avi.avs containing:
avisource("yourmovie.avi")
and loads that avs. The avs is written in the same directory as the avi.
If the extension is anything else, the avs written has the same naming system, but instead contains
directshowsource("yourmovie.extension")

But for megui, there's only avs and anything that goes beyond that would be a major effort - that time is better invested into something else.
Also keep in mind a directshowsource call without fps is dangerous.. but it's the only thing that can be done because we can't just get fps info from an arbitrary file unless either writing parsers for X different containers, or start messing with directshow (render the file and hopefully the fps can be gotten from that).
But in the input window, you still only have avs shown, and you cannot use the typing facility to manually put a non avs filename there.. that would already turn things too complicated again.

berrinam

11th April 2006, 11:22

Here's the only thing that makes sense from a programmer's perspective: relax the filter to other types. If the file selected has the extension .avs, everything works as normal. If the extension is .avi, megui writes a new avs named yourmovie.avi.avs containing:
avisource("yourmovie.avi")
and loads that avs. The avs is written in the same directory as the avi.
If the extension is anything else, the avs written has the same naming system, but instead contains
directshowsource("yourmovie.extension")This system is what I plan to implement (however, not with AviSource. avisource has problems with non-VfW decoders). It shouldn't be hard (the infrastructure is already there in the AviSynth script creator, which I extended in my last commit to support arbitrary filetypes using DSSource), and MeGUI's error checking for video should avoid some problems (well, it will detect AviSynth errors, which includes problems with not signalling fps, as well as mod16 and colorspace problems), so it can't really stuff MeGUI up.

EDIT: Done. It is a very stupid script generator, and it won't deal with any problems like flipped video. If you run into them that's your problem -- use the Avisynth script creator which is designed to deal with all of that.

faxmactor

18th April 2006, 13:29

"Shutdown when finished" is going onto my mind :mad:

I don't think there's any user who wants to go on with the shutdown when he aborts a job manually. :(

Besides that, a timed interception possibility would be useful, just in case. (Encodes packed with filters take so long that one might forget that the option has been set.)

:thanks:

shon3i

19th April 2006, 16:35

I can't belive MeGUI finally looks like a big app for multiformat encoding, all greetings to MeGUI devs, but you miss only one big thing. That's thing is 5.1 audio encoding. I know MeGUI supports 5.1 encoding for both BeSweet and AviSynth, but encoders is very bad. Nero and FAAC need bitrate like AC3 to sounds normaly. So only one encoder is good for 5.1 HE encoding and we know is that Coding Tehnologies (Winamp). Thanks to Dimzon and his encAACPlus.exe comandline encoder, encoding is very sipmply. New Nero encoder shall to be out in few months which all bugs in 5.1 encoding is cleared but until now you can add support for CT becouse is aslo good in stereo encoding. CT now support both LC and HE profiles, aslo LC encoder support 8-320kbs, HE 8-320kbs, 6ch HE 96-213kbs which all bitrates good for 1cd rips especialy 96-128, and 128-213 is good for 2 or more cd's, aslo LC supports 6ch. Thanks

EDIT: Encoding via BeSweet have no sense, so clear your agony and remove it form MeGUI

dimzon

19th April 2006, 16:42

I can't belive MeGUI finally looks like a big app for multiformat encoding, all greetings to MeGUI devs, but you miss only one big thing. That's thing is 5.1 audio encoding. I know MeGUI supports 5.1 encoding for both BeSweet and AviSynth, but encoders is very bad. Nero and FAAC need bitrate like AC3 to sounds normaly. So only one encoder is good for 5.1 HE encoding and we know is that Coding Tehnologies (Winamp). Thanks to Dimzon and his encAACPlus.exe comandline encoder, encoding is very sipmply. New Nero encoder shall to be out in few months which all bugs in 5.1 encoding is cleared but until now you can add support for CT becouse is aslo good in stereo encoding. CT now support both LC and HE profiles, aslo LC encoder support 8-320kbs, HE 8-320kbs, 6ch HE 96-213kbs which all bitrates good for 1cd rips especialy 96-128, and 128-213 is good for 2 or more cd's, aslo LC supports 6ch. Thanks
AFIAK this feature (CT AAC encoder support) is in TODO list. Unfortunally there are huge work (refactiring) to do before it and we (Doom9 and I) are too busy IRL now

shon3i

19th April 2006, 17:03

@dimzon thanks anyway. When you back to url can you start working on this. I am not sure but i don't now how nero distribute their new aac codec, Ivan told that will be new distribution and maybe limitation over encoding.

cc979

25th April 2006, 16:31

@berrinam
just had time to do some testing (0.2.3.2129 20 April 2006)

good job on the mod16 stuff :) :thanks:

just a little feature when encoding an .avs maybe have script preview window, incase some forget which what they are encoding

chipzoller

28th April 2006, 05:25

I would like to request the addition of MP3 audio support in MP4 muxing operation. I realize AAC is the favored choice and perhaps the better one, but nonetheless if MP3 is supported in the container format, I think megui should be able to do it, and this shouldn't be a big deal to add.

berrinam

28th April 2006, 07:09

I agree, but this will probably have to wait until the refactor is committed.

Doom9

28th April 2006, 09:40

@berrinam: if not already done, all that needs to be done to support that is adding the mp3 audio type as supported audio type.. I may have even already done it. Then of course there's the matter of making the mux windows muxer aware to they allow you to select input types that the muxer supports..

chipzoller

28th April 2006, 22:12

Also, another request....
I think perhaps it's time megui had its own icon. It seems to be in a worthy enough state :)

Or is this feature in the works?

As always, many thanks for the outstanding program.

Morte66

3rd May 2006, 16:56

Small ease of use request: when using file dialogues to open souce audio files for encoding, please treat Avisynth files (.avs) as an audio format along with .ac3, .dts etc. I create quite a few combined audio/video scripts to retime PAL movies to 24fps, so I often have avisynth as an audio source.

ChronoCross

3rd May 2006, 16:59

That however would be undetectable as to you having audio in the avs script. so it would be pointless.

you can't encode audio and video together and change the audio format. just won't work. Audio must be encoded separately.

if you had a completely separate avs for audio it would be another matter.

Morte66

3rd May 2006, 17:16

* confused *

I do mean to encode the audio separately. Perhaps I should give an example. Say I have externally created an avisynth script called deadwood3.avs which has 24fps audio and video streams:

LoadPlugin("C:\Program Files\DGMPGDec\dgdecode.dll")
loadplugin("C:\Program Files\avisynth 2.5\plugins\NicAudio.dll")
video=mpeg2source("C:\MyDox\My Videos\Staging\Deadwood 3\VTS_01_1.d2v",cpu=4)
audio=nicac3source("VTS_01_1 T01 3_2ch 448Kbps DELAY 0ms.ac3",2)
audiodub(video,audio)
AssumeFPS(24,1,true)
SSRC(48000,false)

I open it once as a video source and again as a sound source, maybe using "AutoEncode" to do the mux as well, to get 2 or 3 related jobs. I open it as video as normal on th emain input tab, no problem. Then I in the main input tab I click on the "..." button to the right of the audio input filename. That launches the "Select your audio input" dialogue. In the bottom drop-list on that dialogue, labelled "Files of type", I can choose "Supported audio formats" {default}, "AC3 audio", "DTS audio", "MPEG audio", "WAV audio", and "All Files (*.*). At them moment, I have to choose "All Files" to open an avisynth file like deadwood3.avs as my source sound file. I would like to see "Avisynth Audio" added to the list of audio file types, and for it to fall within "Supported audio formats" so that avisynth files appear when I first launch the "Select your audio input" dialogue. {edit} It would also be quite useful if the audio input box on the main Input tab accepted avs files via drag and drop, like it accepts ac3 etc.

[I'm referring to versions 0.2.3.2065 and 0.2.3.2127, if that matters. I've encoded over 100 files this way, generally to x264/FAAC.]

ChronoCross

3rd May 2006, 18:17

why don't you just encode the AC3 Seperately? you can do the same SSRC transform in the audio config dialog. Your method seems strangely wrong to me.

dimzon

3rd May 2006, 19:02

why don't you just encode the AC3 Seperately? you can do the same SSRC transform in the audio config dialog. Your method seems strangely wrong to me.

LoadPlugin("C:\Program Files\DGMPGDec\dgdecode.dll")
loadplugin("C:\Program Files\avisynth 2.5\plugins\NicAudio.dll")
video=mpeg2source("C:\MyDox\My Videos\Staging\Deadwood 3\VTS_01_1.d2v",cpu=4)
audio=nicac3source("VTS_01_1 T01 3_2ch 448Kbps DELAY 0ms.ac3",2)
audiodub(video,audio)
AssumeFPS(24,1,true)
SSRC(48000,false)
trim(123,456)

In this case this method is good, isn't it? :)

Morte66

3rd May 2006, 19:10

why don't you just encode the AC3 Seperately? you can do the same SSRC transform in the audio config dialog. Your method seems strangely wrong to me.

I can? Where, and how? I see no option for the slowdown/resample, just delay/gain and channel mix options. What am I missing? Can you walk me through how you would do it, if you don't mind?

I actually do it this way becuase I've got Robot4Rip set up to write the avisynth script for me in its "Finalize" stage. In R4R I can spend a minute setting up 4-6 jobs for a TV DVD then wander off for 40 minutes while it does all the rip-index-vobsub-avs cycles. Then I spend another minute queueing up the avs files for encoding in MeGUI. Every other workflow I've tried requires repeated attention during the preparation process. OCE is close, but (AFAIK) it won't do PAL->24 or subtitles. So I just use it when I don't need either, which generally means US NTSC IVTCd to 23.976fps or some UK TV DVDs made from 25fps video.

But if you know a better way (i.e. requires less of my time per DVD encode), I'm all ears.

ChronoCross

3rd May 2006, 19:43

LoadPlugin("C:\Program Files\DGMPGDec\dgdecode.dll")
loadplugin("C:\Program Files\avisynth 2.5\plugins\NicAudio.dll")
video=mpeg2source("C:\MyDox\My Videos\Staging\Deadwood 3\VTS_01_1.d2v",cpu=4)
audio=nicac3source("VTS_01_1 T01 3_2ch 448Kbps DELAY 0ms.ac3",2)
audiodub(video,audio)
AssumeFPS(24,1,true)
SSRC(48000,false)
trim(123,456)

In this case this method is good, isn't it? :)

true but your also assuming this hasn't already been done in the ripping or d2v creation stage. I suppose this could be a good thing for people who do avisynth editing for AMV's. I'll throw my support in for this feature.

CobraX

4th May 2006, 19:18

Can you guys implement the new free Nero AAC Encoder? You can find it here (http://www.nero.com/nerodigital/eng/Nero_Digital_Audio.html).

You can also read more about it here (http://www.hydrogenaudio.org/forums/index.php?showtopic=44275).

fatboyfin

4th May 2006, 19:53

ChronoCross

4th May 2006, 20:31

it doesn't work right now as a valid option. it only accepts wav files. and apparently only ones created using a microsoft encoder cause a .wav I have from one of my import DVD's doesn't work in the encoder.

I've also been pouring over their forums and it seems it doesn't work on 98% of the machines people are using. so until any of these bugs are resolved and they add support for other input formats I wouldn't recommend using it.

jpl

4th May 2006, 21:15

Can MeGUI be expanded to include support for the CT AAC encoder? This would provide a nice alternative to Nero for HE encoding.

Thanks
JPL

Kostarum Rex Persia

5th May 2006, 02:13

And newest Nero AAC encoder, also.

Doom9

5th May 2006, 08:04

@jpl: can and will.. I don't know why it's not in the list but it has long since been an approved future feature.
@KPR: striked for rule 1.. two posts above Chronocross already explained why the encoder in question cannot be supported.

Latexxx

5th May 2006, 09:20

it doesn't work right now as a valid option. it only accepts wav files. and apparently only ones created using a microsoft encoder cause a .wav I have from one of my import DVD's doesn't work in the encoder.

Use -ignorelength -q 0.4 -if - -of file.mp4 for stdin source.

I've also been pouring over their forums and it seems it doesn't work on 98% of the machines people are using. so until any of these bugs are resolved and they add support for other input formats I wouldn't recommend using it.

It currently requires sse2 which isn't available in athlon xp and pentium 3 machines. Newer machines work and they've already said that they are working on a version which works on all machines.

dimzon

5th May 2006, 09:54

I've also been pouring over their forums and it seems it doesn't work on 98% of the machines people are using. so until any of these bugs are resolved and they add support for other input formats I wouldn't recommend using it.
AFAIK problem is in SSE2 (it reques SSE2 capable CPU)
And MeGUI doesn't need nothing except WAV via STDIN support
So it will be extreme easy to replace and use this encoder instead of NeroRaw.Exe + dll's
But we need to wait for support of non SSE2-capable CPU's first

shon3i

5th May 2006, 10:10

Guy's why just now clear some things, why just completly remove support for BeSweet and FAAC, there is not need anymore for audio encodings. Instead FAAC add support for CT. And we get MeGUI with tree most uses audio codecs, NERO,CT,MP3 and what you think about implementing four codec like OGG. This lastest beta of ogg is very good in everything.

Doom9

5th May 2006, 10:37

why just completly remove support for BeSweetIt's already gone.. ask berrinam.. some code is still there but the encoder shouldn't be registered anymore.. so.. when it's done, besweet support will be gone.
Ogg will come, too. And BeLight already supports various other encoders that could easily be integrated. FAAC will remain though because only faac allows to have a 100% open source solution.

shon3i

5th May 2006, 10:40

FAAC will remain though because only faac allows to have a 100% open source solution.Yes but i think nobody uses him becouse can disturb quality

New NERO encoder just realised. non SSE2 machine works

Latexxx

5th May 2006, 10:50

FAAC will remain though because only faac allows to have a 100% open source solution.
I've nothing against keeping FAAC besides Nero but Nero has one great advantage: They pay the appropriate license fees and thus downloading their coder is more legal than downloading faac binaries at least in some countries.

Doom9

5th May 2006, 12:00

and thus downloading their coder is more legal than downloading faac binaries at least in some countries.well.. who pays the licensing fees for x264, xvid and lmp4? As private user you're not eligible to pay licensing fees unless you start mass distributing.

dimzon

5th May 2006, 12:43

Anycase I'm voiting for
-Keep FAAC (everything including configuration dialog is already done, we does't need any significant resources to support it into the future so why not)
-Replace NAAC by new free CLI encoder (lets call it ND AAC)
-Add CTAAC (from WinAmp) (lets call it CT AAC)

shon3i

5th May 2006, 12:54

Latexxx

5th May 2006, 13:01

By the way, the encoder has been updated and the zip file (http://www.nero.com/nerodigital/eng/Nero_Digital_Audio.html) now includes two versions, one with sse2 and one without.

Doom9

5th May 2006, 13:32

@dimzon: That's exactly how I see it. Once berrinam commits, you'll be good to go :)

Latexxx

5th May 2006, 14:27

And one other feature request. Make it possible to adjust the delay of an audio track in the mp4 muxer.

Doom9

5th May 2006, 16:00

Make it possible to adjust the delay of an audio track in the mp4 muxer.The reason that hasn't already been done is the following: except for using the mp4 muxer manually from the tools menu, your audio is assumed to have been normalized wrt delays (since megui does that automatically if there's a delay in the filename).. and the manual muxer is the only way you could ever get to enter a delay.. everything else works automatically and you don't get a chance to interfere manually.
So is the separate muxer useful enough to warrant the time?

ChronoCross

5th May 2006, 16:18

Also according to discussion that went on in #x264 apparently delay of any type must be supported by the splitter. They weren't sure if haali's splitter supported it yet so that is norther issue. You should pretty much always fix the delay when transcoding.

shon3i

8th May 2006, 13:21

Becouse Teegedeck write beautiful presents for XviD, he and we XviD users have little barring to use this presents in MeGUI. The problem is using Custom Quantization Matrices, becouse they locked into MeGUI. All of this presents use different custom matrices.

Teegedeck original question (http://forum.doom9.org/showpost.php?p=824526&postcount=183)

Presets (http://forum.doom9.org/showpost.php?p=796900&postcount=54)

Thanks for advice XviD users

berrinam

8th May 2006, 13:40

I'll look into it.

Mtz

8th May 2006, 13:50

Many standalones have problems with XViD and Custom Matrices.

enjoy,
Mtz

shon3i

8th May 2006, 13:55

Many standalones have problems with XViD and Custom Matrices.

enjoy,
Mtz
This is a quality presets, but i used this presets with my standalone and i don't have problems, btw sharktooth, soulhuners and didee matrices are supported on most standalones, some presets use qpel and gmc but this is not usefull for most standalones

@berrinam thanks

Mtz

8th May 2006, 14:27

With this, many of them have problems:

Quant type: MPEG Custom

Custom intra matrix:
8 9 10 11 12 14 16 18
9 11 12 13 15 16 18 20
10 12 14 16 17 19 21 22
11 13 16 18 20 22 24 26
12 15 17 20 23 25 28 30
14 16 19 22 25 29 34 38
16 18 21 24 28 34 46 52
18 20 22 26 30 38 52 72

Custom inter matrix:
16 16 17 17 18 19 21 23
16 17 17 18 19 20 22 24
17 17 19 20 21 23 24 27
17 18 20 22 24 25 28 32
18 19 21 24 26 29 36 40
19 20 23 25 29 40 48 52
21 22 24 28 36 48 60 64
23 24 27 32 40 52 64 106

What is the name of your standalone?

enjoy,
Mtz

shon3i

8th May 2006, 17:02

Doom9

8th May 2006, 18:22

becouse they locked into MeGUI.come again?

Teegedeck

8th May 2006, 22:14

I think he means I could not load any CQMs (http://forum.doom9.org/showthread.php?p=824526#post824526). I'm still not sure whether this is only me because I haven't been following MeGUI nor xvid_encraw development closely. I just updated to new versions of both last weekend and wasn't very successful with what I wanted to do.

berrinam

9th May 2006, 14:19

CQMs should be fixed in 0.2.3.2135 (it's on the dev thread).

b9AcE

9th May 2006, 15:50

Ok, I'll start by admitting that I did not read all the 430+ posts before mine, so if my suggestions have already been rejected or accepted... ...sorry.

I very much miss two features from GordianKnot.
Firstly, the very nice resolution-tab with the dynamic previewing window.
I want to be able to make a manual case by case decision on which pixels I can lose or which ones are to vital to be cropped.
If I get an unacceptable aspect ratio deviation from my desired cropping I resize one step down and try again until I achieve an acceptable compromise.
This is harder with the more automatic approach of MeGUI.

Secondly I would like to have the "bits per pixel"-calculation somewhere.
I know this is not a proper measurement of quality, but if I combine that number with previous experiences and my guesstimation of the complexity of the video I do get quite a bit of help from the bpp-number.

As it is now, I load the movie into GK first to get the desired resolution/cropping and approximate bitrate and save the generated AVISynth-script.
Then I load the movie into MeGUI and use the AVISynth-creator to get input on what (if any) deinterlacing I need to apply and save that script.
Finally I merge the appropriate parts of the different scripts, load it back into MeGUI, select a Sharktooth-profile and set the inloop filter/bitrate according to video complexity.

See, this is just me be a perfectionist, but what use is there doing a backup "almost as good as you can? :cool:
Now, let me end by saying that I just love both of these programs.
So much easier than doing each step manually the way I did before GK was released...

ChronoCross

9th May 2006, 17:38

bits per mixel is useless. I'm highly against it. As for cropping...it is already dynamic, in the avisynth script creation window. The only thing that isn't dynamic is resize.

ariga

10th May 2006, 10:33

A delay for the shutdown with an option to abort the shutdown as in Virtualdub would be nice.

If the encoding finishes sooner than expected and you still happen to be working on something...

kurt

10th May 2006, 15:27

@ariga: in console shutdown -a doens't work?

cc979

10th May 2006, 17:54

quick suggestion about cropping, i've had some files that have multiple borders - could it be possible to crop from a certain frame using the video preview ?

Doom9

10th May 2006, 17:58

could it be possible to crop from a certain frame using the video preview ?You mean crop 20 pixels from the top from frame 0 - frame 100 and then crop 22 pixels from frame 101 till the end? I'm not aware that avisynth supports that kind of thing and I'm opposed to splitting a source into different scripts.. that would give a huge mess and complicate matters beyond good reason.

asdfsauce

10th May 2006, 18:50

Is there any way you guys could give the "AVISynth script generator" window its own taskbar entry? It's really annoying to have to move all the other windows around to find it when you want to do cropping and what not, especially when working with HD material.

Doom9

10th May 2006, 20:09

Is there any way you guys could could yes.. but then we'd have other people complaining and wanting us to go back.. initially every dialog was in the taskbar and it wasn't well received.

ChronoCross

10th May 2006, 20:30

You mean crop 20 pixels from the top from frame 0 - frame 100 and then crop 22 pixels from frame 101 till the end? I'm not aware that avisynth supports that kind of thing and I'm opposed to splitting a source into different scripts.. that would give a huge mess and complicate matters beyond good reason.

the method for doing this is complex. I do believe it is possible as long as the resolution is the same. For example:

orig = MPEG2SOURCE("file.d2v")

#first cropping
firstcrop = orig.trim(0,100).crop(0,4,4,4).resize(finalheight,Finalwidth)

secondcrop = orig.trim(101,end).crop(0,6,8,6).resize(finalheight,Finalwidth)

Final = firstcrop + secondcrop

return final

This however leads for possible distortion of the videoes and is not really something I think anyone uses. I would be very weary of using it at all. I would be interested in the explaination of in what instances he uses it.

asdfsauce

10th May 2006, 20:44

could yes.. but then we'd have other people complaining and wanting us to go back.. initially every dialog was in the taskbar and it wasn't well received.

Could someone enlighten me as to why they disliked this? The window is open for only a small while.

How hard would it be to make an option Doom9?

Doom9

10th May 2006, 21:10

How hard would it be to make an option Doom9?It's one click. But.. one click I'm not willing to make because of the assured backlash it will create. There's a divide over what an app should look like and that is not going to go away. You said you have problems with HD sources.. I guess it depends on the source size.. if you have a DVD source, then it's nice to have all the windows side by side.. and if you have a big enough screen (I have the Dell 30 incher ;) ) even 1080p sources are not an issue at all and multiple entries in the task bar just fill up my taskbar that I don't want filled up. But it's really not about my likes or dislikes. Specifically, you can search this forum to see that people objected to what you're suggesting, especially when it comes to configuring codecs.

sp@rrow

10th May 2006, 22:05

asdfsauce

10th May 2006, 22:34

@Doom9

I meant an option in settings to have all windows have their own entries. I kinda feel like that was unclear from reading your reply.

Thanks for the explanation.

ChronoCross

10th May 2006, 22:55

It is possible to make in MeGui test for compressibility - like GordianKnot? :helpful:
Not everyone can estimate bitrate "in eye".
It very much does not suffice me (each time it is necessary to make it manually) :(

compressibility tests are rather useless and generally do not show the true compressibility of a source. It's not even a close estimate. Generally a good majority of the action in a movie takes place near the end with complex scenes that will not be caught unless you do a 100% first pass analysis pass. which is not really feasible with newer codecs.

sp@rrow

11th May 2006, 00:31

I think - sample in 10-15 % can give objective enough information on film.

As I do:
I create a script with such line:
SelectRangeEvery (1760, 294) - 12 seconds through ~ 1 minute - and not important at the end of film there will be dynamic stages or in the beginning ;)
Further - pass 1 - qp 18 (turbo) - and after of the received data I calculate final bitrade.
If you know faster (wait for the end of the first pass does not approach) and an effective way - it is very interesting to me.

ariga

11th May 2006, 10:04

@ariga: in console shutdown -a doens't work?
You mean open a cmd window and type it out just as soon as I see windows shutting down ? :confused: The shutdown completes even before I can open a command window. I would need to learn the five point palm-abort shutdown technique from sensei Pai Mei to move any faster :p

cc979

11th May 2006, 18:23

cc979

11th May 2006, 18:27

I would be interested in the explaination of in what instances he uses it.

i've seen in some early dvd 'bad' editing between different scenes border differences

Kayaker

11th May 2006, 19:31

I totally agree with a former post of b9AcE that I miss a lot some metrics display ( bpp or a better one if you like ) while resizing.
So I have something to play with to know how much to resize down.
It doesn't matter if I later aim at a specific filesize or with a constant quality.

If there any chance that you could add something like that ? Some metric while resizing ?

Also I don't quiet understand why vorbis support is rejected. ( In red in the first page )
Maybe I'm kinda lost here, but for two channel enconding is the best I've found so far.

Thanks a lot guys, MEGui rulez ;-)

ChronoCross

11th May 2006, 19:38

I do believe that vorbis support will be added to the current revisions. once some of the other post-refacro bugs are fixed. It's on the list.

foxyshadis

12th May 2006, 02:56

One way to do a comp check in a reasonable time might be to run it while making adjustments to the video, as a background thread. It could continuously update the compressibility and confidence as the results of each 200 frame segment show up, chosen in a totally random order. The longer you let it run the more confidence you have!

But whether to do it in xvid or x264. Hmm, I guess whichever is selected, even if they select another afterward it'll still give a reasonable approximation for all but the lower bitrates.

Perhaps too much effort, however.

Sharktooth

13th May 2006, 21:48

compressibilty with modern codecs depends too much on settings, so a compressibility test would require the knowledge on how to use it...
ironically that knowledge is exactly the same that you require to choose how much to resize and what bitrate/quantizer to use to get a good quality backup.
so, im sorry but it will be useless for half of the video codecs supported by MeGUI and that's enough to not consider it at all.

ChronoCross

18th May 2006, 08:55

are there any plans to allow more than two audio streams?

quake74

18th May 2006, 09:45

are there any plans to allow more than two audio streams?

I tried already http://forum.doom9.org/showthread.php?p=764833#post764833

foxyshadis

18th May 2006, 15:50

Also, now that the ultimate refactor is in and ready for prime time, will it be easy to swap out a codec with another executable without big upheavals or hacks? I'm referring to integrating ELDER. Not directly, as the commandlines are different, but through a small wrapper, so only the exe name would change and the extra parameter of # of segments (could be zero as a default, when you enable elder it gets set to the real default, and back to 0 if disabled, so the codec commandline isnt tweaked out). Right now it's mp4 joining only, but it could be extended to support mkvmerge internally easily, so no need for hacks around that.

If this is the case I may give it a crack myself, then perhaps try for a minimalist x-platform gui.

berrinam

18th May 2006, 23:36

It is now quite easy to add extra video and audio encoders (which elder would be classed as). However, there is ATM no management for multiple encoders for the one codec (there is SOME infrastructure, but the GUI itself is not aware of it.... it was way too annoying when we had mencoder AND x264.exe for x264 encoding, so we removed it).

It shouldn't be too hard to integrate ELDER into MeGUI, but the main problem would be working out the pausing/restarting and all of the advanced job management ELDER does. I haven't used it, so I don't actually know what it is capable of, but I'm assuming one of the main reasons for including it in MeGUI would be for its safe stopping function (ie you can stop and restart later without losing anything).

ATM MeGUI assumes that jobs can be Started, Abortted, or Paused -- there is no support for the safe stopping that ELDER is designed for.

To add support for safe stopping is not infeasible, but it requires:
A stop button (or perhaps you could override the Abort function for ELDER with safe stopping)
Another job status indicating stopped, but can be restarted.
Adding a restart function, and adding something in the job settings to indicate how much has been completed, etc

Not directly, as the commandlines are different, but through a small wrapper, so only the exe name would change and the extra parameter of # of segments (could be zero as a default, when you enable elder it gets set to the real default, and back to 0 if disabled, so the codec commandline isnt tweaked out). Right now it's mp4 joining only, but it could be extended to support mkvmerge internally easily, so no need for hacks around that.I don't understand this bit. What do segments refer to in ELDER? Does ELDER also manage appending the completed files?

foxyshadis

19th May 2006, 07:51

kurt

19th May 2006, 08:30

You mean open a cmd window and type it out just as soon as I see windows shutting down ? :confused: The shutdown completes even before I can open a command window. I would need to learn the five point palm-abort shutdown technique from sensei Pai Mei to move any faster :p
hehe, didn't try this out with megui - but I believe you :)
so a delay of maybe 10seconds before shutting down windows would be nice. Then there is no need for the five point palm-abort shutdown technique :p

berrinam

19th May 2006, 09:03

It does all the hard work of splitting it into n pieces and parceling them out to cpus (on the machine or network), and remerging them all, so that the input and output looks identical to x264's.Do you mean to include this in MeGUI? I imagine that this could be done transparently, so I suppose this is what you were talking about earlier?

Starting and stopping would be quite cool, definitely. Do you have a general design you'd like it to follow, code-wise?I would say that the IVideoEncoder should have a new function added:
ISafeStopData safe_stop(out string);
and also
bool CanDoSafeStop(); // this is used to en/disable the button mentioned below
both of whose implementations are obvious enough. Then, VideoJob also gets a SafeStopData property, and there is a button added to the queue tab, whose event handler calls safe_stop, and saves the results in SafeStopData. The status is then changed to the new status called, eg, safe_stopped, and the queue is stopped.

That's the design that's most obvious to me, but also mention this on the dev thread and see if Doom9 or dimzon suggest any other ideas. I am very interested in easily being able to start/stop x264 encodes.

EDIT: I looked through the ELDER thread and it seems that the parallel part of ELDER could be added very easily to MeGUI (even without modifying MeGUI! just by making a new x264.exe which wraps ELDER). The start/stop functionality is a different issue and could be dealt with later (like when we have decided on a way of dealing with multiple encoder apps for a single codec).

cc979

19th May 2006, 14:07

excuse for posting here, but what is this ELDER ?

dimzon

19th May 2006, 14:37

I looked through the ELDER thread and it seems that the parallel part of ELDER could be added very easily to MeGUI (even without modifying MeGUI! just by making a new x264.exe which wraps ELDER).
Fine idea!

cc979

19th May 2006, 20:05

just tested the auto-update works well, if on the first use could there be a choice where to install the external apps, eg in a subfolder 'tools' instead of the root of megui

berrinam

20th May 2006, 00:00

just tested the auto-update works well, if on the first use could there be a choice where to install the external apps, eg in a subfolder 'tools' instead of the root of megui
I'm looking for ideas as to how this should be managed. You see, what it currently does is checks if the file exists in the location where the settings currently says it should. If not, then it installs it to meguiroot/appname/appname.exe. It would be easy to change that to meguiroot/tools/appname/appname.exe (and I will do that), but there might be a better way of managing it.

cc979

20th May 2006, 16:18

I'm looking for ideas as to how this should be managed. You see, what it currently does is checks if the file exists in the location where the settings currently says it should. If not, then it installs it to meguiroot/appname/appname.exe. It would be easy to change that to meguiroot/tools/appname/appname.exe (and I will do that), but there might be a better way of managing it.

maybe set it up like

tools/audio-tools
tools/video-tools
tools/muxer-tools

or something

berrinam

21st May 2006, 07:03

Anyone interested in MeGUI supporting VC-1? Nic's WMNicEnc (http://forum.doom9.org/showthread.php?t=111385) could perhaps be integrated into MeGUI if someone is interested in adding a CLI.

Doom9

21st May 2006, 13:05

hmm... VC-1 means WMV and WMA so it's another video and audio codec, and having to worry about yet another muxer.

berrinam

21st May 2006, 13:07

hmm... VC-1 means WMV and WMA so it's another video and audio codec, and having to worry about yet another muxer.
But all of the new muxpathfinding means that this can be done in stages, without breaking anything. It is actually really easy to add. Even though I don't plan to use it, it is probably the easiest way to test the new codecs.

chipzoller

21st May 2006, 18:29

berrinam

21st May 2006, 23:22

How about 'unlocking' the deinterlace feature of the avisynth script generator for those that are aware of the source type and wish to choose the deinterlace method? I agree that this could potentially be a problem for inexperienced users, so perhaps it would be possible to enable a popup message box when selecting a deinterlace method without doing an analysis pass warning the user this is generally not recommended.

I find it annoying having to do an analysis of the source (which is 2 passes) when 90% of the time the fields need to be blended. Megui is powerful enough now to cater to both the inexperienced and experienced users, and I think this is a needed feature.
There are many different source types:

(Each of these can also be anime/non-anime, and TFF/BFF/variable)
Progressive
Pure Interlaced
Pure Film
Hybrid Film/other, mostly film
Hybrid other/film, mostly other

This means it would need a checkbox for anime, a dropdown for TFF/BFF/variable, a dropdown for source type, and a dropdown for the filter itself. That takes up a lot of space.

Granted, it is worth thinking about, but it still needs that: to be thought about (and worked out).

check

23rd May 2006, 08:32

Hi, I've checked the last three pages of this thread and I couldnt see the question..etc etc
On managing the files installed with the autoupdater, an option to download optional components such as alternate / standalone GUIs or related components would be extremely useful. An "optional downloads" section where you can download additional files such as mplayer, the mmg.exe help file (mkvmerge-gui.hlp) and so on to go in with the programs they are paired with that are part of meGUI. I suggest this because it will be a useful addition. You could even let people add in their own tools hosted on external servers, a la Steam.
This would be for files that are not directly related to meGUI, but somewhat related. Sorry this got so wordy, I can't seem to make it more compact without cutting detail.

Also, I've no doubt this was mentioned before, so please just add my vote to the pile: using ctrl-a to open the AVS window has caused far more harm than good :P

EDIT: rewritten for clarity

berrinam

23rd May 2006, 09:07

On managing the files installed with the autoupdater, an option to download optional components so you can use the standalone files would be extremely useful.I don't understand you.

check

23rd May 2006, 10:05

updated previous post, sorry about the confusion.

asdfsauce

23rd May 2006, 20:29

Couple requests here-

Instead of only being able to load AV scripts, wouldn’t it be better if you could load any file and have MeGUI go to d2v generation, AVS generation, “video preview” or whatever depending on the file type? I think audio should be just the same. Load audio, generate AV script and then load the script for job config.

Second request,

Add an option to the job queue’s right-click context menu to open a job’s output directory.

Thanks guys.

Doom9

23rd May 2006, 20:33

Instead of only being able to load AV scripts, wouldn’t it be better if you could load any file and have MeGUI go to d2v generation,That has been possible for ages.. try the power of drag&drop. It's just that nobody has ever bothered to adopt the video input field to offer the same kind of functionality.
Add an option to the job queue’s right-click context menu to open a job’s output directory.And why would that be so useful as to consider it?

dimzon

23rd May 2006, 20:45

Add an option to the job queue’s right-click context menu to open a job’s output directory.
Really usefull trick, let's add it (for fast navigation, when I want to view encoding output after finished)
I really like such function in fb2k!

asdfsauce

23rd May 2006, 20:52

It's just that nobody has ever bothered to adopt the video input field to offer the same kind of functionality.

Well then, that's my feature request. :-)

And, will audio ever be able to accept avis scripts?

And why would that be so useful as to consider it?

So you can get to your encodes more easily when they're finished.

dimzon

23rd May 2006, 20:55

And, will audio ever be able to accept avis scripts?
Yes. Actually there just absent *.avs in fileOpenDialog only but if you open AVS it will be encoded properly ;)

asdfsauce

23rd May 2006, 20:56

Yes. Actually there just absent *.avs in fileOpenDialog only but if you open AVS it will be encoded properly ;)

Oh, Okay, I thought I tried that. :P

Thanks.

berrinam

23rd May 2006, 22:42

Many more input files were supported before the refactor (and MeGUI would in fact send the files to the correct places), but the refactor introduced a different way of managing input files, so I left it that way.

cc979

24th May 2006, 00:39

i got a couple feature requests if possible

when auto updating have tick box on the settings tab to delete the backup files, and a shut down count down

cheers

Doom9

24th May 2006, 16:25

For the autoupdater: a "Check/uncheck all" right click menu. I don't mind it updating filters but I want to remain in charge of all video encoders without having to uncheck each one manually..

fields_g

24th May 2006, 21:18

1) I mentioned something in another conversation that might be useful for this discussion. It has to do with allowing access to changelogs.
http://forum.doom9.org/showthread.php?p=831822#post831822

2) By going too fast, I noticed that there is no way to reinstall/redownload something that has been already updated.
When I got to the ST video profiles, I accidently selected only one. In order to get the others, I had to open the XML file and erase the current version and it then did a reinstall. How about something in the right-click menu, or just allowing someone to manually check something that has already been updated?

berrinam

24th May 2006, 22:17

For the autoupdater: a "Check/uncheck all" right click menu. I don't mind it updating filters but I want to remain in charge of all video encoders without having to uncheck each one manually..
Shift-select them all, right-click, uncheck.

cc979

28th May 2006, 00:20

i got suggestion for avs script preview have video preview have the same preview as the on the main preview with 'show PAR' - it might be handy for some folks

check

28th May 2006, 10:07

Process status window:
The 'projected filesize' is a useful display, but for larger files it would be nice to see it measured in mb or gb. Failing some sort of intelligent mechanism to swap between them, is it possible to make the fields in the window selectable?

berrinam

28th May 2006, 10:19

Failing some sort of intelligent mechanism to swap between them, is it possible to make the fields in the window selectable?It is no more work to add a GUI component than to add an intelligent display mechanism.

check

28th May 2006, 11:15

is that a yes? ;)

berrinam

28th May 2006, 12:49

No, it's an annoying comment so I can avoid committing to anything.;)

fields_g

28th May 2006, 14:45

I have a issue issue that can be fixed it two ways.

1) Get a better system (This will be done by the end of the summer) with a more stable OS (once it works on Linux).

OR

2) Have an additional megui feature that saves me A LOT of work.

My computer seems to want to lock up occasionally. This is really noticed when I'm pushing my computer hard for hours at a time (eg. Encoding). If I have a sizeable queue built up, it's lost after a restart. Is there any way that you can make queue status update between processes so that if the program crashes, the queue list could be reloaded and the process that the system crashed on could be immediatly restarted and the rest of the list preserved?

Thanks