Any suggestions as to why this clip looks so blended...?

[Lathe]

Sorry about the vague title, but I wanted to see if anyone had any general observations or suggestions. I have this 1969 Telefilm 'FEAR NO EVIL' http://www.imdb.com/title/tt0064319/reference (along with a bunch of other rare and commercially unavailable films that I've collected over the last 30 years - can you say, 'BETA!' anyone...? :) ) Anyway, going by the excellent things that I've learned here so far from all you experienced people, I went ahead and re-encoded it the way BDRB would be if you set it to IVTC it (just copied the script) and I added a couple of VERY simple things which I THINK made it look a little better.

But, it still kind of looks like I'm seeing double frames or blurred frames every once in a while even after I used this code:

DirectShowSource("C:\x\fearnoevil.mkv")
LoadPlugin("D:\EXECUTABLES\BD-RBV05018\BD_Rebuilder\tools\decomb521.dll")
Telecide(order=1,guide=1)
Decimate(cycle=5)
Spline16Resize(960,720).Addborders(160,0,160,0)
LoadPlugin("C:\Program Files (x86)\AviSynth 2.6\plugins\Unfilter\new\UnFilter.dll")
UnFilter(35, 35)
Tweak(cont=1, bright=0, Sat=.75, startHue=80, endHue=200)
ConvertToYV12().AssumeFPS(24000,1001)

Now, I KNOW that I likely did NOT do the colour thing right and it could probably be done a bunch of better ways, but in my clumsy way I TRIED to de-emphasize what I think I see in the print which appears to be a bunch of yellow colour spread about through the film. So, after reading a LOT of posts until I went blind (I got better...) FINALLY I at least got a clue as to how to use the Tweak function to desaturate somewhere HOPEFULLY near the yellow range.

So, I am gonna include the links to both an original MKV snippet and what I re-encoded so far.

Does anyone else see the blurred frames too or am I just hallucinating? Also, if anyone has any EASY ideas to improve this film, that would be great because I really like it but it is not available anywhere.

The MKV file is original; the M2ts file is the re-encoded one.

http://the-lathe-of-heaven.com/fearnoevil.mkv (http://lathe-of-heaven.com/fearnoevil.mkv)

http://lathe-of-heaven.com/fearnoevil.m2ts

Thanks!

[Sharc]

Your source has blended fields (poor framerate conversion).
Instead of IVTC by Telecide().Decimate(), try:

tdeint(mode=1,order=1) #bobbing, TFF
srestore(frate=23.976) #fixes or reduces most of the blends

[magiblot]

tdeint(mode=1,order=1) #bobbing, TFF
srestore(frate=23.976) #fixes or reduces most of the blends

Frames 110 and 111 (repeated):

http://i.imgur.com/ilzH2PY.png

I also tried with:

sanimebob(1) #Uses QTGMC
srestore(24.000/1.001)

But got similar results (some frames still look interlaced).

This worked better:

AnimeIVTC(mode=2, bbob=4, degrain=0)

It seems that there are no repeated frames.

http://i.imgur.com/qHQEx71.png

[FranceBB]

. tfm(mode=1,pp=5,slow=2,micmatching=2,clip2=tdeint(mode=2,type=3))
tdecimate()
srestore(frate=23.976)


Also, consider using FFMPEG or LSMASH instead of DirectShowSource as it's not frame accurate and may "lose" frames, nnedi instead of Spline16 and GamMac instead of manual tweak. Stainless made a very good filter which attempts to repair the original colours automatically. I tried it on old video tapes and I got very good results.

[Sharc]

Frames 110 and 111 (repeated):

Hmmm.... I don't get the repeated frame. Did you index the file for frame accurate seeking?

[magiblot]

Did you index the file for frame accurate seeking?

That might be the problem. I think I don't know how to index the file properly (if FFVideoSource/FFmpegSource2 doesn't do it automatically).

Do you think the interlaced-looking frames from the bobbed clip might be caused by that?

[Sharc]

That might be the problem. I think I don't know how to index the file properly (if FFVideoSource/FFmpegSource2 doesn't do it automatically).

Do you think the interlaced-looking frames from the bobbed clip might be caused by that?
No, the remaining ghosts and jaggies - in addition to the other defects - come from the already damaged source and unblending artefacts.
Some final cleanup with QTGMC(InputType=3) may help.
Your AnimeIVTC picture looks good. I never have tried AnimeIVTC for natural movies, I thought it's only useful for Anime ....

[Lathe]

No, the remaining ghosts and jaggies - in addition to the other defects - come from the already damaged source and unblending artefacts.
Some final cleanup with QTGMC(InputType=3) may help.
Your AnimeIVTC picture looks good. I never have tried AnimeIVTC for natural movies, I thought it's only useful for Anime ....

Thanks you guys, appreciate it.

Hmmm, sounds like this one is gonna be a little more complicated. So, I will look for this AnimeIVTC and give it a try. So, what you are saying is that I should use this INSTEAD of the IVTC that I used before?

At least THIS time, I wasn't the one who screwed up the encode :) (or, at least I don't THINK that I was...)

[Lathe]

Wow, sounds like there are a LOT of moving parts (dlls and such) that are needed for this... I hope I can manage it. There was something they talked about in the wiki section of this function. Could this be what you guys are seeing...? (I bloody KNEW I was seeing blurred frames or something...)

"Basic concepts:

Soft telecine - the stream is encoded at 24 fps and the pulldown is applied during playback. Hard telecine - your source was telecined before being encoded on the DVD (or before airing). Double hard telecine - your source has at least one of the "intra-frame variable framerate" situations and the telecining was applied separately on each part of the frame, so the pattern is not constant and many frames are interlaced. Field blended norm-conversion - Instead of going back to 24fps (film rate), the video streams are directly converted from NTSC (29.97fps) to PAL (25fps) or vice versa, creating ugly fieldblending (looks like DHT when inspecting a frame). Hybrid - mix of 24t (24 fps displayed at 30 through telecining) and 30p (progressive movement on every frame) Truly interlaced - each frame contains half the information of two frames, which means that you can bob it to 60p and see movement on every single frame, without any interlacing or blending. Progressive - No interlacing whatsoever at the full framerate. 3:2 pattern - This pattern repeats itself from the very beginning to the very end of your clip: 3 progressive frames followed by 2 interlaced ones. VFR - Variable Frame Rate - The sections of the clip run at different framerates, thus allowing to have less frames for the same movement. The duplicates are removed and played during the same amount of time as if there was the original amount of frames. Lossless rendering pass - Save the file as uncompressed YV12 or Lagarith to apply the script only once and make subsequent filtering/encoding faster. It takes a lot of disk space: around 5GB for a 22 minutes clip @ 23.976fps with Lagarith and 20GB with uncompressed YV12.

Examples:

Double Hard Telecine - At first, your source may look like a typical Hard Telecined one, but after a more thorough inspection, you'll see that it's actually a DHT one: same source, different high motion scene. Interlaced on top of telecined (can be hard or double hard) - In this case, it was a hard telecined source. Image Obviously, since there's no pattern in DHT and interlacing on every frame with pure interlaced material, regular IVTC will output blends and jerkiness."

I don't really understand this, but could this be why the regular IVTC didn't really work fully?

[Lathe]

Okay, I DL'd the 'requirements' for AnimeIVTC and I will include a snapshot of what is in the zip file. This is something that I STILL don't quite grasp yet... When you DL these functions / plugins for Avisynth, they seem to come in all these folders and I don't know if I'm supposed to place all of these folders together directly into my Avisynth plugins folder, or what...???

Here is what I Dl'd - I don't even see any AnimeIVTC.avsi or anything...???

http://lathe-of-heaven.com/animeIVTC.gif

*EDIT

Oh, I see, the avsi is in it's own DL, got it! Uh, what the Avisynth page says about this function is to 'put all Dlls' in the Avisynth plugins folder, but what confuses me (among MANY other things, apparently...) is that you get all this OTHER stuff in the folders above too. So, do I just go into EACH bloody folder and simply copy ONLY the .dll file from each into my plugins folder? Do I just ignore all the other stuff in those folders? So, if that is correct, I put all the .dll's (one from each folder above) and the main .avsi fuction ONLY into my plugins folder, right...?

[manono]

Frames 110 and 111 (repeated):

No, they're not. Don't use DirectShowSource, and bob it and unblend it as already suggested and there's no need at all to mess around with AnimeIVTC.

I don't really understand this, but could this be why the regular IVTC didn't really work fully?
No, your 'regular IVTC' didn't work because it was never telecined to begin with. It was field blended to go from film to interlaced 29.97fps. Only the shadiest fly-by-night companies do this and if there's any other source available for the film, you should use it instead.

[Lathe]

No, they're not. Don't use DirectShowSource, and bob it and unblend it as already suggested and there's no need at all to mess around with AnimeIVTC.


No, your 'regular IVTC' didn't work because it was never telecined to begin with. It was field blended to go from film to interlaced 29.97fps. Only the shadiest fly-by-night companies do this and if there's any other source available for the film, you should use it instead.

Good, that's easier :)

Hmmm, interesting that the company itself originally produced it that way. I was not aware of this 'travesty'. I knew about interlaced, and I know about telecined, and I myself have screwed up and 'de-interlaced' telecined files, so I know what THAT looks like. But, I sure wasn't aware of this particular odd way to produce a file.

And, sadly, no... there is no official release of this very rare telefilm...

[Lathe]

Your source has blended fields (poor framerate conversion).
Instead of IVTC by Telecide().Decimate(), try:

tdeint(mode=1,order=1) #bobbing, TFF
srestore(frate=23.976) #fixes or reduces most of the blends

I will try this then. I have the Srestore.avsi and the Tdeint.dll in my plugins folder. HOPEFULLY they will run, thanks!

[magiblot]

No, they're not. Don't use DirectShowSource, and bob it and unblend it as already suggested

The problem is, I was already loading the clip with FFmpeg2Source. However, I have already had frame skipping problems with it, and I don't understand why.

[manono]

The problem is, I was already loading the clip with FFmpeg2Source. However, I have already had frame skipping problems with it, and I don't understand why.
Maybe the field order was different in the sample as compared to the complete source. If you were following Sharc's script, he set it for the sample. Maybe it's different for the complete video, but I'm just guessing here. When I tested I did this:

FFVideoSource("fearnoevil.mkv")
AssumeFPS(29.97)
Yadif(mode=1)
Srestore(Frate=23.976)

where he chose a field order. You have to know how to determine the field order. Sometimes AviSynth gets it right, but not always. If you followed his script:

tdeint(mode=1,order=1) #bobbing, TFF
srestore(frate=23.976) #fixes or reduces most of the blends

and you got missed frames, then maybe try:

tdeint(mode=1,order=0) #bobbing, TFF
srestore(frate=23.976) #fixes or reduces most of the blends

and see if the result is any better. In addition, you can't just go right to the place you're checking as Srestore might not work properly right away. You have to kind of 'sneak up' on the place. Either start at the beginning (if checking the sample) or play a hundred or so frames before doing the actual checking for it working properly.

Are Lathe and magiblot the same person? If so, why use two different names? If not, why did you, magiblot, hijack Lathe's thread?

[magiblot]

Maybe the field order was different in the sample as compared to the complete source. If you were following Sharc's script, he set it for the sample. Maybe it's different for the complete video, but I'm just guessing here.

I think the bobber I mainly use (sanimebob.avsi) has no option to choose the field order, so I'll test with tdeint.

(I don't need the source to be a cartoon in order to use sanimebob. I have tested it many times and I have found it to be very good with non-cartoon sources as well. But it's very time-expensive, so I should use it just for the definitive encode, or save the bobbed clip into a lossless file before doing any more edits).

and see if the result is any better. In addition, you can't just go right to the place you're checking as Srestore might not work properly right away. You have to kind of 'sneak up' on the place. Either start at the beginning (if checking the sample) or play a hundred or so frames before doing the actual checking for it working properly.

I think that was pretty much the problem. For more comfortability, I should have exported the clip into a lossless file as well.
(I once had a bad frame-skipping problem when I was loading a script with FFVideoSource into Vegas through avfs. In spite of its bad reputation, DirectShowSource worked better in that case. But this is totally off-topic).

Are Lathe and magiblot the same person? If so, why use two different names? If not, why did you, magiblot, hijack Lathe's thread?

I'm not Lathe; you may check our IPs and connection periods. I'm sorry if I have done bad, but I don't think I have bothered Lathe in having his questions answered. What's more, in such a difficult topic for novices (among whom I include myself) as inverse telecining, there is no experience from which you can't learn. Therefore, not only is what I have asked of interest to Lathe, but I'm also sure that it will be useful in case he ever happens to have the same problem as the one I had.
Finally, it's not like I have just started posting questions in this topic. I have tried to contribute to the solution to Lathe's problem in first place. I don't think it's bad to post secondary questions in a place where all of us want to learn.

[Lathe]

I'm not Lathe; you may check our IPs and connection periods. I'm sorry if I have done bad, but I don't think I have bothered Lathe in having his questions answered. What's more, in such a difficult topic for novices (among whom I include myself) as inverse telecining, there is no experience from which you can't learn. Therefore, not only is what I have asked of interest to Lathe, but I'm also sure that it will be useful in case he ever happens to have the same problem as the one I had.
Finally, it's not like I have just started posting questions in this topic. I have tried to contribute to the solution to Lathe's problem in first place. I don't think it's bad to post secondary questions in a place where all of us want to learn.

Hey, that is quite all right; I could tell that you, like the others, were very kindly trying different things to help me with this particularly difficult film. No, I really appreciate ANY input, since just like you say, I am quite new at this and with this particular film I've run into something new as far as how it has been encoded (using the term loosely, apparently...) All that I THINK that I could determine was that after doing what I THOUGHT was the usual IVTC thing to a source that LOOKED like and seem to have the same specifications as the others that I've done, I was pretty sure I could still see some duplicated frames or blurring or something.

I'm still trying to get my mind around exactly what the deal is with this particular file as opposed to the others. Thus my including the comments from the AnimeIVTC page since it talked about what I THINK were things like double hard telecining or combined interlaced AND telecined, and things like that that I do not understand... (yet...)

[real.finder]

I think the bobber I mainly use (sanimebob.avsi) has no option to choose the field order, so I'll test with tdeint.

(I don't need the source to be a cartoon in order to use sanimebob. I have tested it many times and I have found it to be very good with non-cartoon sources as well. But it's very time-expensive, so I should use it just for the definitive encode, or save the bobbed clip into a lossless file before doing any more edits).



I think that was pretty much the problem. For more comfortability, I should have exported the clip into a lossless file as well.
(I once had a bad frame-skipping problem when I was loading a script with FFVideoSource into Vegas through avfs. In spite of its bad reputation, DirectShowSource worked better in that case. But this is totally off-topic).



I'm not Lathe; you may check our IPs and connection periods. I'm sorry if I have done bad, but I don't think I have bothered Lathe in having his questions answered. What's more, in such a difficult topic for novices (among whom I include myself) as inverse telecining, there is no experience from which you can't learn. Therefore, not only is what I have asked of interest to Lathe, but I'm also sure that it will be useful in case he ever happens to have the same problem as the one I had.
Finally, it's not like I have just started posting questions in this topic. I have tried to contribute to the solution to Lathe's problem in first place. I don't think it's bad to post secondary questions in a place where all of us want to learn.

you can use assumetff() or assumebff() before sanimebob

for avc m2ts try mux it in mkv then use the mkv with ffms2 or use seekmode=-1 in ffvideosource if you don't like to mux it in mkv

or try LWLibavVideoSource with m2ts directly

[Lathe]

Okay, I tried the very first step, which was just calling 'yadif'. The dll is indeed in my plugins folder and I used the very simple syntax shown here. But, when I try to run it I get the error (from Virtualdub) that that plugin does not exist. I went and dl'd the newest version and replaced the dll in my folder, but it still throws the error. So, until I know why I cannot run the rest of the script as suggested here...

BTW, I tried using FFVideoSource instead of DirectshowSource and the file choked immediately. I guess I am not set up properly to use that.

For now, I guess I'll try the other script mentioned above...

[Lathe]

Well, for some odd reason the tdeint does run (it looks a LOT better just with that!) But the Srestore will not run and throws an error. The avsi is indeed in my plugins folder, so again some things run and some don't...

Here is the code I tried so far:

DirectshowSource("C:\x\00000.mkv")
AssumeFPS(29.97)
LoadPlugin("C:\Program Files (x86)\AviSynth 2.6\plugins\tdeint.dll")
tdeint(mode=1,order=1) #bobbing, TFF
Srestore(frate=23.976) #fixes or reduces most of the blends

I get 'There is no function Srestore'

But, there is... :)

BTW, the tdeint would NOT run unless I loaded the dll first...

[Lathe]

Ah, I THINK I've made progress (Sharc will be proud of me :) ) I read in a thread from last year that you should load the Srestore.avs in the script. Sooooo, I changed the .avsi to .avs, Imported the function and then tried to run it. The good news (I think) is that I believe it actually began to run and I did NOT get the 'there is no function named Srestore' error. This time it got to line 46 I THINK within the Srestore function and said that 'there is no function named "mt_makediff" So, it actually DID start to run, right?

I am now going to look for a file named mt_makediff...

[Lathe]

Heh, I can see why you guys likely get VERY tired of answering the same things over and over...

It's funny... sometimes you can read for frigg'n HOURS and not find out what you are looking for, and other times you can look at ONE thread and get it :)

Okay, one thing I did was go through my plugins folder and remove all the duplicates that are different versions.

Now, I'm looking for mt_makediff...

***EDIT

Nope, can't find it anywhere. I copied a page that LOOKED like it was the 'source code' for the function, so I copied the text and created an mt_makediff.avs file, but no go...

Oh well, I sure can NOT get this Srestore to work. I guess I will have to do something else...

[Lathe]

Nope...

Tried hundreds of things in the last 4 hours.

Nada...

http://lathe-of-heaven.com/mt_makedifferror.gif

[Groucho2004]

Tried hundreds of things in the last 4 hours.
Had you only spent 2 of all these hours brushing up on the very basics of Avisynth scripting, you would have succeeded. Looking at your bizarre attempts to get things working shows that you don't have the faintest clue of what you are doing. Your approach reminds me of the "Infinite monkey theorem (https://en.wikipedia.org/wiki/Infinite_monkey_theorem)".

So, how about trying to get at least some basics sorted by starting with the section "New to AviSynth – Start Here" on this excellent site (http://avisynth.nl/index.php/Main_Page)?

Edit: While I'm still ranting - Change the thread title to something meaningful.

[StainlessS]

Lathe, you can always try the search option at the top of every forum page (as a last resort before harakiri, or maybe even as the first resort
before even considering Seppuku).
I suggest that you nearly always select "Show Posts" rather than "Show Threads", and maybe try the "Advanced Search" or "Tag Search" on occasion.
If you are ever stuck for searching any site (not just doom9), you can also try in Google, eg


Mt_Makediff site:forum.doom9.org


or if you want to look for an exact phrase eg 'big dog', then try


"Big Dog" site:forum.doom9.org

[raffriff42]

While I'm still ranting - Change the thread title to something meaningful.Right on. It's not just a good idea, it's a rule (http://forum.doom9.org/forum-rules.htm).

[manono]

Letting that goofy thread title stay up for so long is my fault. If he doesn't change it in a day or so, I'll do it (if I can figure out how).

When trying to make a filter work, it's often a good idea to look for its page at the AviSynth site (http://avisynth.nl/index.php/Main_Page). Had you done so, you'd have discovered one of the requirements for SRestore (http://avisynth.nl/index.php/Srestore) is MaskTools2.

[Lathe]

Letting that goofy thread title stay up for so long is my fault. If he doesn't change it in a day or so, I'll do it (if I can figure out how).

When trying to make a filter work, it's often a good idea to look for its page at the AviSynth site (http://avisynth.nl/index.php/Main_Page). Had you done so, you'd have discovered one of the requirements for SRestore (http://avisynth.nl/index.php/Srestore) is MaskTools2.

Okay, okay...

BUT... DLing (your) MaskTools 2 was something that I did months ago, thanks...

AND... FWIW, I've read that page until I've gone sterile, okay...?

Now, I will try to change the (yes) stupid thread title, my apologies...

And, thank YOU stainless for being a tad more kind about it. It's not like I've been posting every 5 minutes ASKING what to do, have I...??? In just about EVERY other thread from a Newbie, all they do is sit there and constantly ask questions. Well, as per Sharc's sound suggestion, I spent 4 friggn HOURS last night TRYING to figure it out INSTEAD of bothering all of you, okay?

How about some love for the effort, huh...?

At this point, screw it... I'll just use the 'bob' thing and live with it.

Please excuse me for actually wanting to learn...

[manono]

If that didn't work then, did you get your SRestore from that same page? Different SRestore versions require different filters. The version of SRestore I'm using doesn't use that same MaskTools2. And do you have the Microsoft Visual thing it mentions?

[Lathe]

If that didn't work then, did you get your SRestore from that same page? Different SRestore versions require different filters. The version of SRestore I'm using doesn't use that same MaskTools2. And do you have the Microsoft Visual thing it mentions?

I was coming back here to request that you delete this thread, but, since you took the time actually to ask me something meaningful, perhaps I won't give up quite yet.

Yes, that was precisely what I was thinking, since there doesn't seem to be any other answer left (uh, yeah, that's after all that reading last night that I was talking about, NOT sitting here with my head up a dark, smelly place doing nothing...)

I'm thinking too by now that the versions MUST not be matching. I even tried to go back and use the older Mrestore and interestingly even with a completely different function I got exactly the same error, but just on a different line. EVERY plugin or function that I try to add I TRY to make bloody sure to load the other 'requirements' that are listed on the page. Like I SAID above last night, I did indeed learn a lot like getting rid of duplicate functions / dlls in the plugins folder.

If it is helpful, the last thing I THINK I did (it was SO damn late and I actually hurt my back I was sitting in a frozen position so long) was I believe that I DL'd QTGMC_plugin_package_AVS26 because on one thread a person kindly provided ALL of the necessary plugins, etc. that went with it and detailed instructions as to where to put what. So, since I was GUESSING that all these plugins that go with this are fairly recent and up to date, I thought it might be a good idea to comprehensively replace all the plugins across the board that matched the ones in the folder above, for the sake that I was again GUESSING that at least all those plugins go together and were of contemporary versions. Perhaps that screwed the pooch, at this point I sure don't know.

I even bloody copied what LOOKED like the source code for the page that LOOKED like it was for mt_makediff and created an avs file and put it in my plugins folder, but that didn't work at all, so I took it out.

Anyway, like you say and after all this, I am thinking that the versions of whatever I have (I THOUGHT they were the newest) likely simply don't match the version of Srestore that I have (although I'm pretty damn sure as with ALL the others, I went to the Srestore avisynth page and OF COURSE tried to Dl whatever was supposed to go with it. Maybe I should go back and start over with the plugins actually given on THAT page.)

And, YES, BTW... whoever the hell said above that I should probably just go read that Avisynth basics... you are probably right...

After this, I don't think I'm going to bother you guys anymore... I'll frigg'n figure it out if it bloody takes me the rest of my pathetic life...

Oh, and yes, I will doublecheck, but I believe that I have the Visual C thing for Microsoft.

[Lathe]

I have this version: ###### srestore v2.7h ######

I'm going back carefully to check again whether I have the 'requirements' that specifically go along with it...

[Lathe]

Okay, I went carefully through the 'requirements' for Srestore on it's page. All of the plugins I had were the same versions except I had the mt_masktools-26 in my folder and it asked for the basic Masktools2. So, I took the -26 out and replaced it with the Masktools2.dll (the reason why I had the other version is that the QTGMC_plugin_package_AVS26 said to remove all other versions of Masktools, so I had) Also, the Removegrain version's subfolder dlls were the same except for the RSharpenSSE2.dll. The version that it provided was an older version (2005) than the one in my plugins folder already (2007) So, until coming here and telling you that, I didn't replace that one yet.

Otherwise, I SHOULD have all of the correct versions of the 'requirements' that Srestore is asking for in my plugins folder. But, I am getting the exact same error.

I wonder WHERE the hell the mt_makediff function is...???

I will keep searching as suggested above...

[manono]

I wonder WHERE the hell the mt_makediff function is...???

It's part of MT_MaskTools. But now we're getting somewhere. I use MT_MaskTools-25.dll. I'll include it and you can try. No guarantees, though.

[Lathe]

It's part of MT_MaskTools. But now we're getting somewhere. I use MT_MaskTools-25.dll. I'll include it and you can try. No guarantees, though.

Thanks. So, I'm guessing I need to remove the other one (MaskTools2, right?) and replace it with this one. I think I had one of those, but because of the QGTMC requirements, I think I moved it to another folder to get it out of my plugins folder.

I'll try it...

[Lathe]

Nope, same error as the picture I posted earlier. So, the function IS started, right? But, according to the error it reached 'line 52' WITHIN that function and at that point cannot find the function 'mt_makediff'.

Hmmm... Well, if the trouble resides in the MaskTools setup, I think I've tried this with Masktools2.dll, and now MT_Masktools-25.dll (on one thread the guy even suggested removing the '-' before the '25' and I tried that too.)

I think I even have MT_MaskTools-26 somewhere, but I doubt that will help.

BTW, just in case I have to give up on Srestore and I will just live with the 'Bob' thing, which did make it look a lot better, after running the tdeint(mode=1,order=1) don't I have to do something because it has a double frame rate at that point? If I just have to leave it at that, is there another step to finish it off so that I can play it on my OPPO player? Do I need to make it Progressive somehow, or just play it as it is after Bobbing it with the code above?

[manono]

Someone around here has a program you run which tells you what dependencies are missing. I forget but maybe someone will see this and respond. QTGMC is difficult to get running. SRestore isn't supposed to be. I'm sorry you're having trouble. Maybe switch SRestore versions and try again.

I use SRestore for virtually all my projects (damned Indian blended field DVDs) and can't live without it. And I would never be satisfied with simply bobbing such a video. Since you have a bunch of funky video sources also, you want to be able to use it also.

...don't I have to do something because it has a double frame rate at that point?
That's what bobbing does - converts fields into full frames, in the process doubling the framerate.

[Lathe]

Someone around here has a program you run which tells you what dependencies are missing. I forget but maybe someone will see this and respond. QTGMC is difficult to get running. SRestore isn't supposed to be. I'm sorry you're having trouble. Maybe switch SRestore versions and try again.

I appreciate the thoughtful reply. I will try that.

[Lathe]

Someone around here has a program you run which tells you what dependencies are missing. I forget but maybe someone will see this and respond. QTGMC is difficult to get running. SRestore isn't supposed to be. I'm sorry you're having trouble. Maybe switch SRestore versions and try again.

I use SRestore for virtually all my projects (damned Indian blended field DVDs) and can't live without it. And I would never be satisfied with simply bobbing such a video. Since you have a bunch of funky video sources also, you want to be able to use it also.


That's what bobbing does - converts fields into full frames, in the process doubling the framerate.

Interesting... I just rendered that snippet with the bobbing and then checked it frame by frame afterward. It's kind of hard to tell, but I THINK what it looks like is this. On a panning movement: One clear frame, one blurred, one clear frame, one blurred, 2 clear frames, one blurred, repeat...

What does this mean?

Oh, and MediaInfo now calls it 'Progressive', so does the frame doubling then make it 59.xx Progressive?

I'm wondering if it APPEARS to run more smoothly because of the doubled frame rate? Can I just play it that way in a BDMV folder?

Yeah, the snippet actually did play on my OPPO and it looked pretty good except for the generally slight blurry quality of the picture itself. But, it doesn't look like it has all the blurring and ghosting going on like before.

[Lathe]

From further reading about 'poorly interlaced sources' on VideoHelp, they were mentioning taking the 59.xx progressive video (like mine after bobbing it, I think) and simply re-interlacing it so as not to lose too much information.

Would that basically have a similar effect as Srestore converting it back to a progressive 23.xxx Could that be an alternative to using Srestore?

[StainlessS]

I have just come in from the pub, and appologise that I did not help more at some earlier instance.
I guess some people (including me) wanted you to find it yourself', despite the fact that we all knew exactly what you were looking for.
If you have further problems suggest that there may be a lack of required runtimes involvled, see here, hope you get it sorted.

Here , an all-in-one installer, first posted by (I think) Groucho2004):- http://repacks.net/forum/viewtopic.php?t=125

Good luck

[Lathe]

I have just come in from the pub, and appologise that I did not help more at some earlier instance.
I guess some people (including me) wanted you to find it yourself', despite the fact that we all knew exactly what you were looking for.
If you have further problems suggest that there may be a lack of required runtimes involvled, see here, hope you get it sorted.

Here , an all-in-one installer, first posted by (I think) Groucho2004):- http://repacks.net/forum/viewtopic.php?t=125

Good luck

Ah, the pub... sounds nice...

Thanks mate; I'll give it a try.

[Lathe]

Nah, no difference...

Thanks anyway. I think I'll just watch the PHk'n movie the way it is. I'll keep reading and if there is some other function that actually will f'n RUN as it is supposed to, then I'll find it. Otherwise, at 59.xx fps on my OPPO it doesn't look that bad for now.

That's all, over and out, I frigg'n give up, we're done here...

Cheers all!

[StainlessS]

OK, firstly, you should not have multiple versions of any plugin in the plugs directory (last one in directory order would be used I think).
'-' Hyphen in dll file names can cause some problems, notably in AvsPMod, also small problems using long version name eg "mt_masktools-26.dll"
long name would be "mt_masktools-26_Mt_Makediff" which might fail due to '-' not being valid in plugin name (I think AVS+ tried some work around to get
that to work correctly but suggest that any such dll be renamed to remove the hyphens '-').
Plugins in the plugins auto load directory do not need you to LoadPlugin(), that is the whole point of it, so you can leave out those lines in script.
v2.5 cannot see v2.6 plugins, that would explain the problem that you seem to be having, but you seem to be using Avisynth v2.6 so that
would rule that out.
Can you say what version of Avisynth v2.6 you are using, is it final version of standard 2.6 or AVS+, it would likely have problems if pre v2.6 alpha 3 I think.

Also, this is a handy little gadget AvsMeter by Groucho2004:- http://forum.doom9.org/showthread.php?t=165528&highlight=avsmeter
Shows info on your setup and points out things that can mess you up.
For checking dll dependencies can use Dependency Walker:- http://www.dependencywalker.com/
although AvsMeter should be just fine and do same.

[Lathe]

OK, firstly, you should not have multiple versions of any plugin in the plugs directory (last one in directory order would be used I think).
'-' Hyphen in dll file names can cause some problems, notably in AvsPMod, also small problems using long version name eg "mt_masktools-26.dll"
long name would be "mt_masktools-26_Mt_Makediff" which might fail due to '-' not being valid in plugin name (I think AVS+ tried some work around to get
that to work correctly but suggest that any such dll be renamed to remove the hyphens '-').
Plugins in the plugins auto load directory do not need you to LoadPlugin(), that is the whole point of it, so you can leave out those lines in script.
v2.5 cannot see v2.6 plugins, that would explain the problem that you seem to be having, but you seem to be using Avisynth v2.6 so that
would rule that out.
Can you say what version of Avisynth v2.6 you are using, is it final version of standard 2.6 or AVS+, it would likely have problems if pre v2.6 alpha 3 I think.

Also, this is a handy little gadget AvsMeter by Groucho2004:- http://forum.doom9.org/showthread.php?t=165528&highlight=avsmeter
Shows info on your setup and points out things that can mess you up.
For checking dll dependencies can use Dependency Walker:- http://www.dependencywalker.com/
although AvsMeter should be just fine and do same.

Thanks. Appreciate the thoughts. I did try removing the '-' in mt_Masktools-26.dll, no diff. I tried -25 (as per Manono) no diff. However, you brought a new thought to mind... Since Srestore is bloody looking for the 'mt_makediff' function and not finding it, you said that the file COULD HAVE been called mt-Masktools-26_Mt_Makediff which SUGGESTS to me that there are the same thing. If so, perhaps I should just simply rename 'mt_masktools-26' to 'mt_makediff' and then at least the bloody program can 'find' it.

I hadn't thought of the possibility that there were 2 different names for the same function. Interesting...

BTW, as I posted on the other thread last night, MeGUI is a frigg'n LIFESAVER! I did ask whether I should rely on it, but no one commented, so I would appreciate just a 'yay' or 'nay' to that please. BUT... the great thing is that whatever script it calls from it's own folder is clearly all harmonious and consistent, so it WILL definitely bloody RUN! And, after MeGUI suggests the script, I can simply copy and paste it into my own CLI and run my own x264 stuff and other simple filters I want to use. SUPER easy!

And YES, as I stated before, I had indeed gone through and removed all duplicate dlls, etc from the plugin folder.

As far as THIS pestiferous film here, I just ran it on my OPPO at it's current 59.xx rate (progressive I guess) apparently that is one of the Blu-ray standards, and it looked just fine. I might just go ahead and plug it into MeGUI just to see what it recommends and run that to compare it. So, I am truly DONE with this one.

However, I appreciate the thoughts about trying to get Srestore working properly. Actually, conceptually... Isn't using Srestore more intrusive than just simply running the video at a smooth doubled, bobbed rate of 59.xx natively. Wouldn't it be interpolating or something and screwing more with the source? Just curious...

http://lathe-of-heaven.com/avsversion.gif

[Groucho2004]

I did try removing the '-' in mt_Masktools-26.dll, no diff. I tried -25 (as per Manono) no diff. However, you brought a new thought to mind... Since Srestore is bloody looking for the 'mt_makediff' function and not finding it, you said that the file COULD HAVE been called mt-Masktools-26_Mt_Makediff which SUGGESTS to me that there are the same thing. If so, perhaps I should just simply rename 'mt_masktools-26' to 'mt_makediff' and then at least the bloody program can 'find' it.
The file names of the DLLs in the auto-load directory are irrelevant. Avisynth enumerates the functions they export upon loading. Masktools2 exports "mt_makediff". If Avisynth doesn't find that function, the DLL is missing from the auto-load directory. It's as simple as that.
If you don't follow the advice given to you (mainly using AVSMeter to check for problems), you'll never solve the problem.

Anyway, it seems you're all set now by using megui so good luck with that.

[manono]

BTW, as I posted on the other thread last night, MeGUI is a frigg'n LIFESAVER!
In what sense did it save your life?
I did ask whether I should rely on it, but no one commented.
Rely on it for what? To analyze your source and come up with a script? No, you analyze with your eyes. It hasn't a clue how to detect and then fix blended sources properly.
Isn't using Srestore more intrusive than just simply running the video at a smooth doubled, bobbed rate of 59.xx natively.
What does 'intrusive' mean in this context? That it removes the blended frames and leaves you with an equally smooth playing but (mostly) unblended video at the correct unbobbed framerate? Then, yes, it's more intrusive.