Log in

View Full Version : QuEnc 0.60 Released!


Pages : 1 2 [3] 4 5 6 7 8

Koepi
14th August 2005, 11:48
more interested in getting RC working better first though. ;)

Hi,

I don't know if I already reported back, but the issue isn't solved with your test-versions. It seems you really need to fill the structs for frame size, motion bits and texture bits to make the RC work correctly. I didn't succeed in my (short) try to do that myself, the structs were too complicated to be understandable in half an hour :)

I'd appreciate if you could try to do that!

Cheers
Koepi

mean
14th August 2005, 15:17
_context->me_range=255;

is important for mpeg2 to create 100% compliant stream, it may be the cause of the artefact depending on how strict the decoder is.

Nic
14th August 2005, 17:18
@Mean: The me_range should be fixed to a restricted boundary in all recent builds of QuEnc. So that shouldn't be the problem hopefully.

shirohamada
14th August 2005, 18:12
actually those artifacts look like the trellis bug to me. shirohamada can you try again but this time disable trellis and slow and extreme and see if it happens then ?

and i'm cured
both trellis and extreme are the culprit
though i get a 17% in filesize increase
a little bit more ringing after zooming 300%
and gain of speed ;)

thx all

Shernand
14th August 2005, 20:21
Hi to all,
I've just begun using QuEnc and i'm using version 0.61, as i see above there are settings that can make artifacts to appear, i'm trying to encode anime at 1500 kbps, could anyone tell me the best quality / safe (no artifacts, compatible with all DVD players) settings based on the source i have and the bitrate given?

Thanks in advance and br,
Shernand

Sagittaire
14th August 2005, 20:49
QEnc (and libavcodec) make MPEG2 DVD compliant stream ... ???
Yes or Not ... ???

Nic
14th August 2005, 22:19
@Saggitaire: Yes they do. The compatability from libavcodec has a got a lot better over the last year or so due to some good patches.

@Shernand: Don't use Trellis or "Extreme Settings". Also use the correct GOP size for the format (be it PAL or NTSC) and a maximum of 2 B-Frames. And that should be all you need for good compatible streams. (AFIAK)

-Nic

Sagittaire
14th August 2005, 23:37
@Saggitaire: Yes they do. The compatability from libavcodec has a got a lot better over the last year or so due to some good patches.
-Nic

thank you very much

very good work for your encoder ... QEnc is certainely the best encoder available with multipass encoding (buggy in mencoder and FFDShow).

Libavcodec with best setting (RDO, chroma ...) seem produce better quality than your best setting ... why not an "insame quality mode" in QEnc for the next build ...

Vive QEnc ... !!!

Shernand
15th August 2005, 00:15
Nic -> Thanks for your recommendations, Quenc is the best free mpeg-2 enconder (at least for me) ;)

Now, i have this avisynth script :

----------------------------------------------------------------------
LoadPlugIn("C:\Archivos de programa\AviSynth 2.5\plugins\ffavisynth.dll")
AVISource("C:\Anime\File.Avi")
ffdshow("default")
ffdshowAudio("default")
ChangeFPS(23.976)
BilinearResize(720,480)
------------------------------------------------------------------

I use changefps due that if i leave the 24 original fps in the avi file, pulldown doesn't work.

Note, i use ffdshow since i like the postprocessing options ;)

At the very end of the enconding (2nd pass) qenc closes itself without saying anything and i get a incomplete file, any tips to fix this?

Br,
Shernand

dragongodz
15th August 2005, 00:37
It seems you really need to fill the structs for frame size, motion bits and texture bits to make the RC work correctly. I didn't succeed in my (short) try to do that myself, the structs were too complicated to be understandable in half an hour
yes this is one of the things i am looking at. ;)

QEnc (and libavcodec) make MPEG2 DVD compliant stream ... ???
Yes they do. The compatability from libavcodec has a got a lot better over the last year or so due to some good patches.
well really its yes, the majority of the time. sometimes you may have to set max bitrate a bit lower because libavcodec can still spike rather badly. sorry to nitpick. ;)

Libavcodec with best setting (RDO, chroma ...) seem produce better quality than your best setting ... why not an "insame quality mode" in QEnc for the next build ...
QuEnc does already have an "insane&extreme" mode that turns on a lot of things. it also comes with a nice big speed hit of course.
also the latest versions of QuEnc do not use libavcodecs ratecontrol. so comparison to other libavcodec based programs(except avidemux by mean) is not exactly correct. the change in ratecontrol is to try and produce more consistantly compliant streams. meaning no spiking or underflows etc etc etc which libavcodecs ratecontrol still can do.

Shernand
15th August 2005, 00:58
Nic -> sorry about this, just tested again and it worked! :) :) :)

danpos
15th August 2005, 04:32
@Nic and/or dragongodz

I did see some post where you comment about OPV at constant quantization mode. So I thought that I could set up a Q parameter like is done in NuEnc, TMPGEnc... but I didn't see such parameter in interface and window's registry either. I'm blind or what? How do I set Quenc up for to do OPV at constant quantization ?

Cheers,

Koepi
15th August 2005, 08:28
QuEnc does constant quantizer when setting the bitrate between 1 and 32 -- which then is the quantizer used for the encode.

Cheers
Koepi

P.S.: Thanks for still looking into this dragongodz :)

freelock7
15th August 2005, 09:26
It would be fine Nic to encode in constant quantizer with a max bitrate setting -for the DVD compliance-like HC015.
The 1Pass encoding would be improved. I think.

dragongodz
15th August 2005, 10:59
Thanks for still looking into this dragongodz
no probs mate. ok well except for having to try and learn the Xvid RC and only doing it in small time blocks instead of all at once etc etc etc. :)

It would be fine Nic to encode in constant quantizer with a max bitrate setting -for the DVD compliance-like HC015.
it would be nice to have 2 pass VBR working 100% first. then maybe 1 pass VBR and CBR. THEN maybe max bitrating CQ. but thats just my preference. ;)

danpos
15th August 2005, 12:03
@koepi

Thanks for clarifying me up about CQ (constant quantization) encoding in QuEnc!! http://www.geocities.com/pcdanpos/smilies/thumbsup.gif

Cheers,

Sagittaire
15th August 2005, 14:08
... avcodec DOES NOT create compliant MPEG2 files.
The developers of avcodec acknowledge it, and it's clearly documented in their documentation. And we can testify to that.
QuEnc is just a front end to avcodec. Nothing more :!:
No matter what QuEnc does, FreeEnc does, NuEnc does, the final product is the same ...

-kwag

who say true ... ???

How to check the real compliance of MPEG2 stream ... ???

IMO QEnc is DVD compliant (work with all my hardware player) but everyone is far from being convinced of it ... !!!

Nic
15th August 2005, 14:39
@Sagittaire: It is compliant. To the best of my knowledge. Once upon a time it wasn't, but several patches made it a lot better. Also it depends on what part you mean. The actual M2V data or the VOB muxing. If people aren't convinced, they should tell me why and i'll fix it.

-Nic

Sagittaire
15th August 2005, 14:57
Only m2v file ... I export it with ifoedit or other ... !!!

but it's clear... I will pass the message ... lol

dragongodz
15th August 2005, 16:16
as i already said libavcodecs rate control can spike occasionally past max bitrate. now if you set max bitrate to a very high amount, such as 9400 and allowing 448 for ac3 audio, and it spikes past the 9400 to say 9700 for example you have a problem. that is when you go to author the m2v and ac3 you will get an underflow error message, basically saying the bitrate is too big for dvd compliance. that is why i said you sometimes have to set max bitrate a bit lower. its also why QuEncs output may be slightly different than other libavcodec encodes, the settings used try to reduce spiking. finally that is also why the change to using Xvids rate control with QuEnc, to try and get a better, more consistant, rate control.

of interest also may be the fact that even CCE has had reports of spiking slightly over max bitrate with just 2 pass encoding. thats one of the reasons jdobbs changed max bitrate to 9000 minus audio and subs with DVD-Rebuilder. i can search for the old thread if you need it.

Sagittaire
15th August 2005, 16:55
as i already said libavcodecs rate control can spike occasionally past max bitrate. now if you set max bitrate to a very high amount, such as 9400 and allowing 448 for ac3 audio, and it spikes past the 9400 to say 9700 for example you have a problem. that is when you go to author the m2v and ac3 you will get an underflow error message, basically saying the bitrate is too big for dvd compliance. that is why i said you sometimes have to set max bitrate a bit lower. its also why QuEncs output may be slightly different than other libavcodec encodes, the settings used try to reduce spiking. finally that is also why the change to using Xvids rate control with QuEnc, to try and get a better, more consistant, rate control.

Yes I understand ... for example 9000 for max bitrate can be very better for DVD compliant stream ... and IMO I think QuEnc produce DVD compliant MPEG2 Stream.

but Kwag, Dialhot, incredible and other in this forum (http://www.kvcd.net/forum/index.php) seem to think the opposite (perhabs bad feedback with hardware player ... I don't know)

it's perhaps interessant to see the problems which they have with QuEnc because they seem to have good competences in MPEG2 encoding ... ???

Nic
15th August 2005, 18:14
Read the Pioneer DVR-220 problem from that forum. Using FreeEnc he had issues, when he redid it with QuEnc he didn't. Anyway, it doesn't matter what they think. I never care for opinion. Someone provide me with a make and model of a DVD Player that doesn't like QuEnc produced files and i'll fix it. Until then, people can use what ever they wish. :)

"it's perhaps interessant to see the problems which they have with QuEnc because they seem to have good competences in MPEG2 encoding ... ???"
In the words of Daffy: "That, my little cherub, is strictly a matter of opinion".

-Nic

shirohamada
16th August 2005, 17:22
do you think Q1 is a waste of space ?
you can see the difference if you zoom enough...

dragongodz
17th August 2005, 04:51
and other in this forum
Sagittaire - that forum is a pay to join one now. some of us wont waste our money so have no idea what they are saying. if they think there is a problem and can give information about it they know fully well this is where to report it. if they do not choose to then they really dont have anything to complain about period.

do you think Q1 is a waste of space ?
in what way? you need to provide more information so people know what you exactly mean.

you can see the difference if you zoom enough...
do you normally watch movies zoomed right in ? :D

Axed
17th August 2005, 09:13
do you think Q1 is a waste of space ?
you can see the difference if you zoom enough...

Its never a waste to have a Q1 option! There are sometimes needs for quality without the boundries of space. But, for real use, if it works the same way as XviD you wont see a difference between Q1 and Q2. If this is wrong, someone please correct me.

Nic and dragongodz thanks for the work on QuEnc, been playing with it making SVCD's of home movies and it doesnt spike anywhere near as much as i found NuEnc and FreeEnc did. All my SVCD's have come out exactly sized as i wanted them to so far. Great programs the lot of them, just those bitrate spikes on NuEnc and FreeEnc hurt when you cant go above a certain limit to stay inside specs.

shirohamada
17th August 2005, 16:53
i think dark scenes could use the quality boost.
but dark scenes are meant to be compressed...

freelock7
17th August 2005, 21:53
Pixels in dark scenes appear when bitrate is too low. That was the main problem with this new RC partially corrected by Nic for the final release -but not perfect.
[What about the curve compression for low motion scenes, Nic -or DGZ?]
BTW I tested once again the QuEnc061-DGZ test on <The flight of the Poenix> .
Avg=4800 (against 5200 for the correct size).
It's still undersized but better than the official Q061 .
Possible underquantization problem (not tested on DVD player yet).
Muxing with Ifoedit: no underflow.

DarkFoon
22nd August 2005, 01:33
Can I request a feature for Quenc here? Or does that go elsewhere?

dragongodz
22nd August 2005, 04:45
sorry its taking so long. i have a lot on my plate at the moment and finding time to spend a couple of hours actually sitting infront of the pc is a bit hard. i will get there though.....eventually. :)

Can I request a feature for Quenc here? Or does that go elsewhere?
there is an actual requests thread
http://forum.doom9.org/showthread.php?t=85039
so you may want to read that incase your request has already been asked.

Mug Funky
26th August 2005, 08:39
Yes I understand ... for example 9000 for max bitrate can be very better for DVD compliant stream ... and IMO I think QuEnc produce DVD compliant MPEG2 Stream.

yeah. just because DVD allows video to go that high, doesn't mean it should.

after lots of testing (and lots of complaints from fans with playback issues on the madboards), madman use a max of 8500 for everything - and this is with spruce mpx3000 encoder boards which are designed specifically for DVD (they don't go higher than 9600).

john3voltas
3rd September 2005, 17:31
Anyway, it doesn't matter what they think.
Of course it doesn't Nic...and that'd explain why you haven't show up there for so many time...
I never care for opinion.
I can see why...
But I'm also a "facts" guy just like you.
Someone provide me with a make and model of a DVD Player that doesn't like QuEnc produced files and i'll fix it.
Just tried a 16:9 Anamorphic encode with QuEnc on my AG DVD Player-336K and caput -> choppy playback all over the place!
Sometimes the movie even stops for 1 second and I get a frozen image with a lot of big 16x16 squares of flashing colors.
And then the choppy playback will resume for a couple of minutes.
It would look like an authoring problem right?
But I'm still using DVDLab, the same that I have used for a looong time.
To proove that this had anything to do with QuEnc (or libav) I took my testing CD and DVD with KVCD and KDVD small clips mostly done with Tmpgenc and CCE.
And all is working ok with these other CDs and DVDs.
I know! Let us all blame it on the KVCD matrix...right?
It's just too damn easy isn't it Nic?
Until then, people can use what ever they wish. :)
Of course we can Nic and that's why I've never used QuEnc (or any libavcodec based encoder) on a regular basis.
On the other hand I have been using HCEnc on a regular basis for quite some time and it's been working fine with all the DVD players I could find on all the TV/HiFi stores around here.
"it's perhaps interessant to see the problems which they have with QuEnc because they seem to have good competences in MPEG2 encoding ... ???"
In the words of Daffy: "That, my little cherub, is strictly a matter of opinion".
Of course it's always a matter of oppinion when it comes from anything starting with K, right? :(
From here you will tell me that I have broken hundreds of D9 forum rules and that I need to go :readrule: again, right?
But I couldn't help answering your clear K remarks ;)

dragongodz
4th September 2005, 12:00
at the risk of a warning or strike i have to answer that. :scared:

Just tried a 16:9 Anamorphic encode with QuEnc on my AG DVD Player-336K and caput -> choppy playback all over the place!
and what settings did you use ? how did the files play from HD ? can you provide a small sample so people can reproduce and/or examine the problem ?

I know! Let us all blame it on the KVCD matrix...right?
no, why dont you provide some real information so that people can try and help determine what is the problem you are having. besides what the hell has Nic saying QuEnc can produce dvd compliant files got to do with that matrix ? nothing so try reading before spouting off.

It's just too damn easy isn't it Nic?
to be a dick that isnt really interested in help ? pretty clearly it IS damn easy for you. :devil:

dragongodz
4th September 2005, 12:05
oh and Koepi, i tried to add the header size in etc for the RC, man what a mess, but i screwed it up. it kept crashing. so i think i will delete the whole folder and start from fresh when i can.

at the moment(last 3 weeks actually) there have been some serious family happenings going on which take precedent over this. as soon as things steady down(hopefully they will) i will get back to it.

john3voltas
4th September 2005, 16:03
at the risk of a warning or strike i have to answer that. :scared:
Can't see why...

and what settings did you use ? how did the files play from HD ? can you provide a small sample so people can reproduce and/or examine the problem ?

OK, here goes nothin':

Avisynth script
-----------------
Mpeg2source("D:\CATWOMAN\D2V\cat.d2v")
LanczosResize(704,432,4,74,712,426)
RemoveGrain().Deen()
AddBorders(0,72,0,72)
Letterbox(72,72,16,16) # To achieve an overlap overscan of 2
Slicer(2,15,2,"ping") # Incredible's slicer function similar to selecrangeevery

Slicer.avsi
----------------
function slicer(clip input, int percent, int Goplenght, int Gopmulti, string "ping")
{
PercCount = (Framecount(input)/100)*percent
period = int(Framecount(input)/PercCount)*(Goplenght*Gopmulti)
input=(ping == "pong") ? input.trim((period/2),Framecount(input)):input
selectrangeevery(input, period, (Goplenght*Gopmulti))
}

QuEnc 0.61 settings
--------------------
Mpeg-2 - check
Bitrate (Kbps) - 2500
Use VBR - check
High quality - check
Use Trellis Quant - check
Extreme slow settings - check
GOP Size - 15
Max B frames - 2
Force closed GOP - check
16:9 Aspect ratio - check
DC Precision - 10
Max bitrate (kbps) - 8000
Load custom Matrix - C:\VIDEO\QuEnc\KVCD_Notch_matrix.xcm

The KVCD_Notch_matrix.xcm has been done using Leigh's wonderful matrix editor and has no corruption as it has been used in other encoders without any issues.
Playing from HDD with MPC v6.4.8.4 or VLC v0.8.2 both the M2V and the VOB files presented no problems at all.
Sorry but no small clip. My webspace is very limited...

no, why dont you provide some real information so that people can try and help determine what is the problem you are having.
Because I would bet a whole lot of money in that the problem lies in libavcodec.
Do I smell a lot of buffer underflows? :p
But you're the expert in this matter, so try to figure it out.

besides what the hell has Nic saying QuEnc can produce dvd compliant files got to do with that matrix ?
That's usually the excuse for people that post here with K issues found in D9 encoders...

to be a dick that isnt really interested in help ? pretty clearly it IS damn easy for you. :devil:
Wrong!
I have always tried to help anybody that asks if it is inside my knowledge domain.
All the above post by me was "a message to Nic".
In such sense there was no special reason for an answer coming from you.
In fact you are the usually the boring dick that spends all the time kissing Nic's butt for no apparent reason.
But you are a cool dick, and especially one that tries to help ;)
And that's why I elaborated a bit more than I would usually do, on the settings that I used in QuEnc.
You know this is a public forum and you have the right to answer anything that is posted here even if it's not addressed to you.
You did answer to my post when it wasn't addressed to you and I answered Nic's post when it was not directly addressed to me.
But by being a public forum, and since Nic is a Mod here, he might have been able to choose a more appropriate way of refering to kvcd.net.
But then again, why would he choose more appropriate words when D9 himself speaks as highly of kvcd.net as he does...
At kvcd.net there is no big reason for loving D9 forum, but we acknowledge and respect what is accomplished in here.
Sometimes we would like to see the same coming from here but it's just too damn hard for some of the big shot develppers in here, some that even are (or were) kvcd.net users, to understand that people have the right to choose.
Just by choosing to use K methods, one should not be seen as a hobo or an outlaw.
Look at Prodater64 here (http://forum.doom9.org/showthread.php?p=682007#post682007).
He is a cool guy. He developed a tool that is in fact very interesting and he posted it in July.
Do you know who else posted there and cheered Prodater64 for his work to the community? Zyphon, a really cool guy that is registered both here as well as in kvcd.net and some other video encoding forums.
Why do you think that happens DGz? Why do you think the same has happened with so many other kvcd users, for instance VMesquita?
VMesquita only started to take credit for his work here after he built his own forum...Now, what a coincidence!
So see it this way: my problem was in how Nic addressed to us, and since he says that he doesn't care about oppinions, I don't think he will be bothered with mine on this subject ;)
But anyway thanks for caring about what the issues with QuEnc, that is, with libavcodec.

DarkFoon
4th September 2005, 17:39
less drama, more development please :D

Nic
4th September 2005, 22:43
@john3voltas: An interesting post. Apart from your obvious anger and venom, you could have a good post.

I have no problem at all with the kvcd matrix, or the forum itself. Myself and kwag seem to get on fine too. My problem is that all I get when told libavcodec isn't producing good streams is hearsay and rumour. But in this instance you actually gave me a make of DVD player that has a problem. Which no one else has done, so I thank you for that.

I stand by my statements, I don't care for opinion or what people "think", no matter what forum they are from. Only fact and useful criticsm interest me. Your critique was useful because you've given me a DVD player that doesn't work with QuEnc. I honestly can't tell you why that would be though, every verification I can do of the MPEG video from QuEnc seems fine (unless the bitrate spikes). Did it in your case? Which version did you use? Can you do tests again for us to help with the development?

I hope that clears it up a little, I'll be looking into your issue soon.

Cheers,
-Nic

ps
Any more of the use of words like d*ck will result in strikes (this may have already taken place, I've only just come back from being away)

john3voltas
4th September 2005, 23:20
@john3voltas: An interesting post. Apart from your obvious anger and venom, you could have a good post.
No Sir, no anger and especially no Venom.
What you may have noticed on my words it's a bit of sadness for the way some things are said or done around here once in a while.
I have no problem at all with the kvcd matrix, or the forum itself. Myself and kwag seem to get on fine too. My problem is that all I get when told libavcodec isn't producing good streams is hearsay and rumour. But in this instance you actually gave me a make of DVD player that has a problem. Which no one else has done, so I thank you for that.
No rumours at all Nic.
I truely believe that libavcodec has the best bitrate allocation algos from all the encoders we have right now.
You can see that with your eyes, no need for special equipment.
And I was truely happy to hear that you had finally moved on to XVID ratecontrol because I knew how buggy libavcodec was in that field.
But unfortunately in my case I have an old player that doesn't like it.
That's all.
Small note: bitrate allocation with HCEnc has seen some very good progress and is very close to what libavcodec can offer.
But that's just MHO.
I stand by my statements, I don't care for opinion or what people "think", no matter what forum they are from. Only fact and useful criticsm interest me. Your critique was useful because you've given me a DVD player that doesn't work with QuEnc. I honestly can't tell you why that would be though, every verification I can do of the MPEG video from QuEnc seems fine (unless the bitrate spikes). Did it in your case? Which version did you use? Can you do tests again for us to help with the development?
I don't have much means of analising the stream encoded by QuEnc 0.61.
All I could do was throw it at bitrateviewer and yes, as usual with libavcodec, I could seem some massive bitrate spikes.
When I say massive I do mean massive.
I mean that it blows up for more than 200 frames up to the 10.000~12.000 Kbps.
But I've seen other encoders doing the same and this old player never had any problems with that.
On the other hand this movie has some pretty long dark scenes and bitrateviewer also shows the bitrate going way below the 100Kbps for several seconds.
But I don't "think" it would have anything to do with that since the choppiness is almost systematic.
But hey, you are much more skilled than me to draw such an oppinion.

ps
Any more of the use of words like d*ck will result in strikes (this may have already taken place, I've only just come back from being away)
Ah! you bet it has already taken place.
Only it was not strikes, instead warnings.
Only thing is such word was used by more than one user and I can only speak for myself...

dragongodz
5th September 2005, 02:31
Do I smell a lot of buffer underflows?
doesnt dvdlab tell you if it detects underflows ? try running the clip through muxman or dvdauthorgui or Tmpgenc dvdauthor as they should tell you if this is detected.
actually just for the sake of completness try authoring with one of them and burning the result to a dvdrw(assuming your player plays them) and see if the problems are reproduced. this will rule out dvdlab not liking QuEncs output.

In such sense there was no special reason for an answer coming from you.
actually as the other person who works on QuEnc, infact currently looking at its ratecontrol, there is every reason for me to be interested in any problems. i know it just says "by Nic" in the title bar but what people dont see is a lot of PM's that Nic and i have had over both QuEnc and Rejig(which i also helped with, mainly engine tuning). and no we dont always agree on things but we accept that and continue anyway. believe me Nic is one of the easiest people to work with and thats not kissing up to him. ;)

Sometimes we would like to see the same coming from here but it's just too damn hard for some of the big shot develppers in here, some that even are (or were) kvcd.net users, to understand that people have the right to choose.
my problem is with certain people from that forum that have come here in the past to only flame and abuse rather than ask or give help. yes there are others who post on both forums and i have no problems at all with them. i also am no fan of the fact it has gone to a pay system where you can not even view posts without paying to join first but i do not blame or take that out on users.

bitrateviewer also shows the bitrate going way below the 100Kbps for several seconds.
But I don't "think" it would have anything to do with that since the choppiness is almost systematic.
actually some players(some models of philips and panasonic for example) have been reported as not liking bitrates to drop below about 300Kbps but since its systematic as you say then this is not likely the problem unless it happens at all problem points.

if you can encode a small clip, say 5 minutes worth, and author that and have it still show the problem then that would be good for you doing some testing. if that short is no good then you would have to see how much needs to be encoded to constantly reproduce it for tests.

so lets assume authoring with another program didnt fix the problem, this is what i would suggest trying.
1. disable trellis and extreme settings. encode small clip.
2. disable custom matrix. encode small clip.
3. disable closed GOP. encode small clip.
4. do not filter input(deen). encode small clip.
of course enable the disabled settings for the other encodes to isolate each setting/change.

hmm i just noticed you didnt mention have 2 pass ticked. i assume you are using 2 pass though yes ? 1 pass encoding doesnt use the xvid RC. i plan to work on 1 pass after 2 pass undersizing is fixed or atleast much better.

Only thing is such word was used by more than one user
yes i used it first and am sorry but if you reread your post you should see it can come off rather aggressive and provided no clues to what the problem may be. so yes it irritated me at a very stressful time.

less drama, more development please
yes boss. :D

Nic
5th September 2005, 11:50
@john3voltas:
Hmmm. Then I'm guessing it's the bitrate spikes causing the problems with your player rather than any other bitstream problem. I've never seen bitrate spikes like that using the latest version and two pass. Just to repeat what Dragongodz said really, if you didn't use two pass, please try again using two pass mode (this enables the XviD ratecontrol). Without it, you are using libavcodec's ratecontrol, and that would explain the spikes (and hence the problems with your player).

Let us know how you get on, if that's the case :)

-Nic

Zarxrax
5th September 2005, 18:05
I have been doing some encodes with quenc, trying to author a dvd in Adobe Encore, and Encore complains that the bitrate is going too high for dvd. I set the max bitrate to 8000, average bitrate to 4500, and there is no audio. The too-high bitrate doesnt seem to occur until towards the very end of the stream though. I've encoded some shorter clips (< 15 minutes) and they didn't have this problem. I am using 2-pass encoding. I dont have trellis or extreme enabled, I'm not using a custom matrix, and I have tried with and without closed GOP.

Nic
5th September 2005, 18:28
@Zarxrax: How close to the end of the stream? The very very end (like the last few hundred frames?) Or is it just generally towards the end of the encode. I assume this is with QuEnc 0.61 (?)

Zarxrax
5th September 2005, 18:57
Yes, 0.61. One of my files is about 1 hour 11 minutes in length, and Encore says that the too-high bitrate occurs around 1 hour 03 minutes.

Do you know of a tool that can display the max bitrate that is used in the file, so I can see if maybe its just a bug in Encore, or if the bitrate really is spiking?

danpos
5th September 2005, 19:56
@Zarxrax

You can to use Bitrate Viewer or alternatively to use a tool present in DVD-Lab Pro.

Cya!

Zarxrax
5th September 2005, 20:29
Ok, the peak bitrate of my file went way to high all over the place, not just near the end. The Bitrate viewer in DVD-Lab is saying the peak bitrate is 10230 kbps.

While I am encoding the 2nd pass, it reports that there are many possible overflows. Is that referring to bitrate overflows? At less than 10% through, it already reports 6 possible overflows!

kwag
6th September 2005, 07:43
my problem is with certain people from that forum that have come here in the past to only flame and abuse rather than ask or give help. yes there are others who post on both forums and i have no problems at all with them. i also am no fan of the fact it has gone to a pay system where you can not even view posts without paying to join first but i do not blame or take that out on users.

@All

Hi.

Before I get banned, or this post gets deleted, I just hope several people have a good chance to read it and digest it.
If it gets deleted, I'll just repost it in the main public page at KVCD.Net.
I'm sure it will spread worldwide in a matter of hours.
I said to myself a long time ago that I was not going to post at this site anymore, but it seems that I must in order to clear up several comments about me and KVCD.Net.
Since KVCD.Net went to a "pay-once" (dirt cheap $2.99) model for forum access, some people around the net immediately started whining.
Why?, because most sites are free, and don't require any payment, so anyone can join.
Some use advertizing (very bad model in my opinion, unless it's google, CNN or other huge sites ;) ), and others are free for some time, before finally folding up after some time because of not making any revenues to maintain the site.
Well, since this model was put into operation, KVCD.Net has been the cleanest and most joyful site to visit and to talk about advanced video conversion, because the people that join truly want to learn, and the pay model keeps the rats out of the site.
Now, I've been refered here, because specifically dragongotz has made several comments about us, on different occasions, and I'd like to show the community exactly what KVCD.Net does with some of the money that comes in via new registrations.
Below is our PayPal record of some transactions that have been made specifically to the open source community, and some of them with dates that are prior to the time we started the payment registration model. Back then, the money came out of my pocket.



Web Accept Payment Sent (ID # 4B877749NU657345W)

Original Transaction
Date Type Status Details Amount
Feb. 24, 2004 Payment To Laurent DUVEAU Completed ... -$25.00 USD

Related Transaction
Date Type Status Details Amount
Feb. 24, 2004 Transfer From Credit Card Completed Details $25.00 USD

Name:
Laurent DUVEAU (The recipient of this payment is Non-U.S. - Verified)
Email:
laurent@aldweb.com

Total Amount:
-$25.00 USD

Item Amount:
$25.00 USD
Shipping:
$0.00 USD
Handling:
$0.00 USD
Quantity:
1
Item Title:
Donation to support the GuppY project
Time:
06:59:11 PST
Status:
Completed
Shipping Address:

No Address Provided

Funding Type:
Credit Card
Funding Source:
$25.00 USD - MasterCard Card XXXX-XXXX-XXXX-xxxx
This credit card transaction will appear on your bill as "PAYPAL *LAURENT".

-----------------------------------------------------

Transaction Details

Web Accept Payment Sent (ID # 4A840097MM512301D)

Original Transaction
Date Type Status Details Amount
Feb. 3, 2004 Payment To Jason Rohrer Completed ... -$5.00 USD

Related Transaction
Date Type Status Details Amount
Feb. 3, 2004 Transfer From Credit Card Completed Details $5.00 USD

Name:
Jason Rohrer (The recipient of this payment is Verified)
Email:
jcr13@cornell.edu

Total Amount:
-$5.00 USD

Item Amount:
$5.00 USD
Shipping:
$0.00 USD
Handling:
$0.00 USD
Quantity:
1
Item Title:
The MUTE Project
Time:
08:38:44 PST
Status:
Completed
Shipping Address:

No Address Provided

Funding Type:
Credit Card
Funding Source:
$5.00 USD - MasterCard Card XXXX-XXXX-XXXX-xxxx
This credit card transaction will appear on your bill as "PAYPAL *JCR13".

Description:
The MUTE Project

----------------



Transaction Details

Web Accept Payment Sent (ID # 11L981748U312622X)

Original Transaction
Date Type Status Details Amount
Feb. 11, 2003 Payment To Videotools.net Completed ... -$25.00 USD

Related Transaction
Date Type Status Details Amount
Feb. 11, 2003 Transfer From Credit Card Completed Details $25.00 USD

Name:
Videotools.net (The recipient of this payment is Non-U.S. - Verified)
Email:
payments@videotools.net

Total Amount:
-$25.00 USD

Item Amount:
$25.00 USD
Shipping:
$0.00 USD
Handling:
$0.00 USD
Quantity:
1
Item Title:
professional_package
Time:
13:22:19 PST
Status:
Completed
Shipping Address:

No Address Provided

Business Contact Information

Customer Service URL: http://www.videotools.net
Customer Service Email: supportrequest@videotools.net

Funding Type:
Credit Card
Funding Source:
$25.00 USD - MasterCard Card XXXX-XXXX-XXXX-xxxx
This credit card transaction will appear on your bill as "PAYPAL*VIDEO TOOLS".

Description:
professional_package

------------------------------

Jun. 25, 2005
Transaction Details

Web Accept Payment Sent (ID # 2X453905167394053)

Name:
OSDN / VA Software (The recipient of this payment is Verified)
Email:
salexander@valinux.com

Total Amount:
-$10.00 USD

Item Amount:
$10.00 USD
Shipping:
$0.00 USD
Handling:
$0.00 USD
Quantity:
1
Item Title:
Donation
Invoice ID:
239485
Time:
18:34:49 AST
Status:
Completed
Shipping Address:

No Address Provided

Business Contact Information

Customer Service URL: http://www.sourceforge.net
Customer Service Email: staff@sourceforge.net
Customer Service Phone: 510-687-7181

Funding Type:
PayPal Funds
Funding Source:
$10.00 USD - PayPal Account

Description:
Donation

Note:
Great work Kris :)
And btw, thanks for changing the license to BSD. I'm not very fond of GPL ;)

Thanks,
-Karl
http://www.kvcd.net
-----------------------------

May 4, 2005
Transaction Details

Web Accept Payment Sent (ID # 4H264470XL3323333)

Original Transaction
Date Type Status Details Amount
May 4, 2005 Payment To DVDREasy Completed ... -€5.00 EUR

Related Transactions
Date Type Status Details Amount
May 4, 2005 Currency Conversion (debit) From U.S. Dollar To Euro Completed Details -$6.62 USD
May 4, 2005 Currency Conversion (credit) To Euro From U.S. Dollar Completed Details €5.00 EUR

Name:
DVDREasy (The recipient of this payment is Non-U.S. - Unverified)
Email:
dvdreasy@netcabo.pt

Total Amount:
-€5.00 EUR

Conversion From:
-$6.62 USD
Conversion To:
€5.00 EUR
Exchange Rate:
1 U.S. Dollar = 0.756307 Euros

Item Amount:
€5.00 EUR
Shipping:
€0.00 EUR
Handling:
€0.00 EUR
Quantity:
1
Item Title:
DVDREasy €5 Donation
Time:
02:42:50 AST
Status:
Completed
Shipping Address:

No Address Provided

Business Contact Information

Customer Service URL: http://pwp.netcabo.pt/dvdreasy
Customer Service Email: dvdreasy@netcabo.pt

Funding Type:
PayPal Funds
Funding Source:
$6.62 USD - PayPal Account

Description:
DVDREasy €5 Donation
-------------------------------------------

Mar. 10, 2005
Transaction Details

Web Accept Payment Sent (ID # 6CB399479B321621K)

Name:
Ratajik Software, LLC (The recipient of this payment is Verified)
Email:
sales@ratajik.com

Total Amount:
-$14.99 USD

Item Amount:
$14.99 USD
Shipping:
$0.00 USD
Handling:
$0.00 USD
Quantity:
1
Item Title:
StationRipper-FullVersion
Item Number:
100
Time:
19:28:36 AST
Status:
Completed
Shipping Address:

United States
Unconfirmed Help

Business Contact Information

Customer Service URL: http://www.ratajik.com
Customer Service Email: support@ratajik.com

Funding Type:
PayPal Funds
Funding Source:
$14.99 USD - PayPal Account

Description:
StationRipper-FullVersion

--------------------------------------------------

Feb. 28, 2005
Transaction Details

Web Accept Payment Sent (ID # 2U720486V85903434)

Name:
Laurent DUVEAU (The recipient of this payment is Non-U.S. - Verified)
Email:
laurent@aldweb.com

Total Amount:
-$10.00 USD

Item Amount:
$10.00 USD
Shipping:
$0.00 USD
Handling:
$0.00 USD
Quantity:
1
Item Title:
Donation to support the GuppY project
Time:
13:18:41 AST
Status:
Completed
Shipping Address:

No Address Provided

Funding Type:
PayPal Funds
Funding Source:
$10.00 USD - PayPal Account

Description:
Donation to support the GuppY project
-----------------------------------

Feb. 16, 2005
Transaction Details

Web Accept Payment Sent (ID # 0EA20842775885503)

Name:
jeffrey weisberg (The recipient of this payment is Verified)
Email:
jaw+paypal@tcp4me.com

Total Amount:
-$10.00 USD

Item Amount:
$10.00 USD
Shipping:
$0.00 USD
Handling:
$0.00 USD
Quantity:
1
Item Title:
argus contribution
Time:
19:34:52 AST
Status:
Completed
Shipping Address:

No Address Provided

Funding Type:
PayPal Funds
Funding Source:
$10.00 USD - PayPal Account

Description:
argus contribution

Note:
Thanks for a wonderful piece of software!
Argus rocks!!! :)

Thanks,
-Karl
http://www.kvcd.net
-------------------------------------------------

Jan. 20, 2005
Transaction Details

Web Accept Payment Sent (ID # 46284855BL779473E)

Name:
The NetBSD Foundation, Inc. (The recipient of this payment is Verified)
Email:
paypal@netbsd.org

Total Amount:
-$10.00 USD

Item Amount:
$10.00 USD
Shipping:
$0.00 USD
Handling:
$0.00 USD
Quantity:
1
Item Title:
The NetBSD Foundation
Time:
00:55:35 AST
Status:
Completed
Shipping Address:

No Address Provided

Business Contact Information

Customer Service URL: http://www.netbsd.org
Customer Service Email: admin@netbsd.org

Funding Type:
PayPal Funds
Funding Source:
$10.00 USD - PayPal Account

Description:
The NetBSD Foundation

Note:
Thank you all for a wonderfull and stable OS!

Karl Wagner
http://www.kvcd.net
--------------------------------------------------

Dec. 19, 2004
Transaction Details

Web Accept Payment Sent (ID # 9H769888B8268731Y)

Name:
Hubert Feyrer (The recipient of this payment is Non-U.S. - Verified)
Email:
paypal@feyrer.de

Total Amount:
-$5.00 USD

Item Amount:
$5.00 USD
Shipping:
$0.00 USD
Handling:
$0.00 USD
Quantity:
1
Item Title:
g4u
Time:
14:35:24 AST
Status:
Completed
Shipping Address:

No Address Provided

Funding Type:
PayPal Funds
Funding Source:
$5.00 USD - PayPal Account

Description:
g4u

Note:
Hi Hubert,

Thanks for such a wonderful utility, and for basing it on one of my favorite OS'es ;)

Cheers,
-Karl
http://www.kvcd.net
------------------------------------------
Dec. 17, 2004
Transaction Details

Web Accept Payment Sent (ID # 6FV13308CC485771W)

Name:
OSDN / VA Software (The recipient of this payment is Verified)
Email:
salexander@valinux.com

Total Amount:
-$5.00 USD

Item Amount:
$5.00 USD
Shipping:
$0.00 USD
Handling:
$0.00 USD
Quantity:
1
Item Title:
Donation
Invoice ID:
223316
Time:
20:55:47 AST
Status:
Completed
Shipping Address:

No Address Provided

Business Contact Information

Customer Service URL: http://www.sourceforge.net
Customer Service Email: staff@sourceforge.net
Customer Service Phone: 510-687-7181

Funding Type:
PayPal Funds
Funding Source:
$5.00 USD - PayPal Account

Description:
Donation

Note:
Thanks for a wonderful piece of software.

-Karl
http://www.kvcd.net
----------------------------------------

Dec. 4, 2004
Transaction Details

Web Accept Payment Sent (ID # 6PH284168E1303916)

Name:
Manuel Kasper (The recipient of this payment is Non-U.S. - Verified)
Email:
mk@neon1.net

Total Amount:
-$10.00 USD

Item Amount:
$10.00 USD
Shipping:
$0.00 USD
Handling:
$0.00 USD
Quantity:
1
Item Title:
m0n0wall
Time:
01:14:54 AST
Status:
Completed
Shipping Address:

No Address Provided

Funding Type:
PayPal Funds
Funding Source:
$10.00 USD - PayPal Account

Description:
m0n0wall


------------------------------


Donation of $25.00 (if I recall the amount correctly) was made to DVD2SVCD about 3 years, but I lost track of that one.
Also donation to Baldrick at vcdhelp.com some time ago.

Also contributions made by code review/fix, specially to the FreeBSD Team as shown below:


Page 180
The example in the middle of the page should read:

For example, to generate a second set of 32 pseudo-terminals, enter:

# cd /dev
# ./MAKEDEV pty1

You can generate up to 256 pseudo-terminals. They are named ttyp0 through ttypv, ttyq0 through ttyqv, ttyr0 through ttyrv,

ttys0 through ttysv, ttyP0 through ttyPv, ttyQ0 through ttyQv, ttyR0 through ttyRv and ttyS0 through ttySv. To create each

set of 32 terminals, use the number of the set: the first set is pty0, and the eighth set is pty7. Note that some processes,

such as xterm, only look at ttyp0 through ttysv.

Thanks to Karl Wagner for pointing out this error.
Page 197, first line

link here: http://www.lemis.com/errata-2.html

More:
http://www.openbsd.org/donations.html


And too many others I lost track of, or were payments I did on an old PayPal account I no longer have.

This is just the way I show appreciation to some developers and people I value, so KVCD.Net DOES give back a lot to the community.
So anyone that joins KVCD.Net can be assured that some of the money he gave when joining, eventually goes to some selected open source projects, and that is a feeling of satisfaction.

Have a G'day!,
-kwag :thanks:

Zyphon
6th September 2005, 08:25
Well, since this model was put into operation, KVCD.Net has been the cleanest and most joyful site to visit and to talk about advanced video conversion, because the people that join truly want to learn, and the pay model keeps the rats out of the site.
Hi Karl,

I have to agree with you there. Since the paid model has come into action KVCD.net has very extemely clean indeed.

In the past it used to get very frustrating with all the AVI warez kiddies that used to hound the board with ways to convert their illegally gotten movies and got to such a state that it was becoming unbearable.

Since the change it has been a clean site whith genuine people who love video/audio encoding and who are always on hard to help people out and to be honest I think $2.99 is such a small amount to ask and is not hardly going to break the bank.

KVCD.net was one of the first places I found and I have learned a lot from there and Karl and the other members some of whom are members here also have taught me a lot. So thanks Karl for that. :D

@All

Please don't feel that I am putting this site and forum down, I an not. I like being a member here also and equally there are many many great members here that are also so helpful and I have learned a lot from this site also.

This is a great site to, but the beauty of the www is that you can have more than one source for from which to get knowledge and this is a good thing.

I just wanted to help defend Kwag's decision to go the way of a pay site. :)

Thanks for listening. :D

john3voltas
6th September 2005, 12:28
@Nic,
As far as I can recall, I mentioned all the settings that were enabled.
If I didn't mention 2-pass it's more than sure that I had it disabled.
And if what you're saying is that in single pass the ratecontrol is managed by libavcodec, then it only gives me more reason to be angry at libavcodec...
@Kwag,
Nice seeing you around.
Glad you made your point very clear to anyone ;).
@all
Doom9 is one of the best places to find video encoding information.
And is also the places where one can find more free open source developpers coding video enconding tools.
And that's very cool.
The encoding community owes a lot to Doom9 and to it's users.
For that I thank you all.
But there are other places where people can find help and where one can find a lot of information gathered from lots of testing.
Kvcd.net is one of those places and so it is Vcdhelp (or whatever name is used nowadays) which is another good example.
I just would like to know why some people here won't acknowledge that and why they will never stop sending sarcastic messages when someone else says he/she has a problem with an encode that used the Notch matrix or that was encoded with some help at the Kvcd.net forum.
At the end of the day what I mean is that "Knowledge and good will" are not an exclusive of Doom9.
And for me this would close the argument that I raised.
Just my 0.02€ before I get another warning or get striked for not being nice or for going way too off-topic...
Thanks

dragongodz
6th September 2005, 13:23
since this model was put into operation, KVCD.Net has been the cleanest and most joyful site to visit and to talk about advanced video conversion, because the people that join truly want to learn, and the pay model keeps the rats out of the site.
In the past it used to get very frustrating with all the AVI warez kiddies that used to hound the board with ways to convert their illegally gotten movies and got to such a state that it was becoming unbearable.
hmm ok lets assume the "rats" refered to are warez kiddies. how exactly does disabling reading stop them from fouling up the forum ? it doesnt. disabling posting by needing to be a member does. so the arguement that reading by anyone has to be members only aswell to stop them is plain false.
now you may say why would anyone just want to read. ok i will give you an easy example.
Bill hears about kvcd and after looking around thinks he may be interested in testing it out for his vcd/dvd. now he hears there is an actual website with a forum. good he thinks i can see what experiences people have had and any problems and work arounds etc. goes to the website, goes to read the forum and......he must pay.
now other may be fans of that all they like but i am not.

because specifically dragongotz has made several comments about us, on different occasions
funny but i dont seem to remember me ever going to your forum and flaming away at people as you have done before here. infact me saying things like i dont like a forum being pay-per-view is nothing to the attitude you have brought here in the past. as i also said i get on fine with people that are seriously trying to give or get help here but not those who just want to rant and flame. oh and so we are clear on this it was john3voltas who brought kvcd in to this topic with his mini-rant about how kvcd matrix always gets blamed etc. instead of just explaining what the problem was(with details) and seeing what we would say.

Below is our PayPal record of some transactions that have been made specifically to the open source community
well good on you. still has nothing to do with what i have said.

I just would like to know why some people here won't acknowledge that and why they will never stop sending sarcastic messages when someone else says he/she has a problem with an encode that used the Notch matrix or that was encoded with some help at the Kvcd.net forum.
now thats a big stretch. go back and read that post you did. you assumed people would say something. maybe in future wait and see what they actually say instead.

this is all way off topic and i have no interest in reading more justifications of peoples actions so i wont bother to reply to anymore such posts here.

john3voltas
6th September 2005, 14:34
Correction: it wasn't me who introduced the kvcd.net into this topic.
That was Sagittaire's doing, not me.
I just commented on Nic's comments about kvcd!

Note: I have lurked D9 for a very long time and I don't recall Kwag ever flaming in here.
I do remember some argumentations (both here and at kvcd.net) with a lot of people from both forums flaming the opposite end.
If for instance we were talking about honorable moderator Mr. Neo Neko flamings at kvcd.net then I would easily remember and agree.
Maybe you could give us some evidence of such flaming behaviour from Karl?...

Note2: BTW we both used the word d*ck a few posts above.
I got a warning for that.
Did you get a warning too :devil: ?
Forget it. I can guess it :D.

Mug Funky
6th September 2005, 15:51
Note: I have lurked D9 for a very long time and I don't recall Kwag ever flaming in here.

well, that thread was in fact delteated. it got quite nasty, and i'm not too proud of the contribution from either "side". i totally don't understand inter-forum rivalry at all. we're all just netizens after all, but the whole thing just left a bad taste in both kvcd and doom9's collective mouths. stupid really, and since then kvcd people have been immediately on the defensive as soon as they post here, which of course means they don't get particularly useful responses from doom9 members - it's the text equivalent of poor body language, and it's hard to avoid thinking some people are just trying to start a fight.

[edit]

btw, i think i read the first 2 sentences of kwag's post. please put all that in a code block so people who are interested can read it, but people who are interested in quenc development can read on without wearing out their scroll-wheel.