Ciao! (XviD-1.0-Beta2-05122003)

[mf]

Originally posted by Seimour
I downloaded the first beta of XviD, and I couldn't find the credits tab, I'd like to know it won't be no more on the following releases of this codec. Thnx & bye
http://forum.doom9.org/showthread.php?s=&threadid=66206&highlight=zones+credits
*yawn*

[olnima]

...again the B-frames question:
where is the exact difference between

setting B-frames to -1
setting B-frames to 0
disable B-frames ???

Is it right that using B-frames only makes sense with VHQ >0 ?

Thanks in advance

Olnima

[Manao]

Setting b-frames to -1 should not be used.

Setting b-frames to 0 should ( afaik ) not be used either.

Disable b-frames is the right way to disable b-frames

VHQ is only used on P-frames, so it is always useful, because you'll always have P-frames. It have more influence when you don't use b-frames , which doesn't mean you should not use it with them, use VHQ _always_ .

[olnima]

Thanks Manao

Olnima

[vip]

Just read an interesting article on ixbt.com (it's russian software/hardware site) about cpu optimizations and Koepi's build of XviD 1.0 beta 2. They wrote if you disable MMX optimization in codec settings you can get 2x speedup on P4. Thats related only to beta 2 and P4. I cant confirm 'cos i own AthlonXP, but maybe someone could?

P.S. The original story here (http://www.ixbt.com/cpu/pentium4-xvid-opt.shtml). You can use AltaVista's Babel Fish Translation Service (http://babel.altavista.com/translate.dyn) to get a translation from russian...

P.P.S. Sorry for my poor english...

[Tuning]

Originally posted by vip
They wrote if you disable MMX optimization in codec settings you can get 2x speedup on P4.

This is not with latest Gamr build(P4). I tested on a 30sec clip and could not find any increase in speed, instead slow down. With all available optimization it was the fastest encoding than disabling MMX. May be only available to beta2.:o

[Soulhunter]

OK, Ive done a small correction in my test !!! ;)

But for the thing with the MPEG/Trellis encodes...

Have redone the encodes, but results are still same !!! :confused:

Maybe Trellis does not work with MPEG Quants ???

No matter... 2Pass tests are in progress !!!


Bye

[Danzel]

@Tuning, remember that Gamr's builds are done with GCC and Keopis builds are done with the Intel Compiler (iirc) so they will definately have different optimizations, wanna try with keopis 1.0beta2 build to see if their results are correct?

Also, thanks XviD Team!!!

Danzel.

[Manao]

Maybe Trellis does not work with MPEG Quants ???The XviD's dev say the contrary, and I have different results with and without Treillis using a custom MPEG matrix, which should behave like MPEG matrix regarding this parameter ( afaik ). So we'll have to wait for them to confirm either the problem, or how such a behavior could happen.

[Chainmax]

So, the old keyframe decition is no longer used? Not even when disabling b-frames by unchecking the BVOP box?

[Soulhunter]

The XviD's dev say the contrary, and I have different results with and without Treillis using a custom MPEG matrix, which should behave like MPEG matrix regarding this parameter ( afaik ). So we'll have to wait for them to confirm either the problem, or how such a behavior could happen.
Found something funny...

Watch what happens when you put the cursor on "Use Trellis" setting for a while...

"...works with h263-matrix only for now"

So I think the MPEG/Trellis part of my test was a bit unless then...

LOL !!! :D :D :D

Bye

[Manao]

Soulhunter : Naah, the said it was working, I think they just forgot to correct the tooltip in the GUI. Look here (http://forum.doom9.org/showthread.php?s=&threadid=65913&highlight=aloha) for a post from Teegedeck.

Chainmax : I forgot the difference between Old / New keyframes decision mode, so my previous statement on b-frames may be wrong. I would test to confirm, but I can't right now.

[Soulhunter]

Soulhunter : Naah, the said it was working, I think they just forgot to correct the tooltip in the GUI. Look here for a post from Teegedeck.
So.... What then ??? :confused:

Bye

[poochie2]

Is it a bug or whatelse that doesn't allow me to get to q1 frames? setting it in the codec will result in a movie that will never go lower than q2, as the debug window tells me and as the filesize confirms... :scared:
DivX 5 allows it and my sample jumps up to 6 mb VS the 2.1 of the Q2 of both xvid1b2 and divx5 (5.1.1).

Any hint? :confused:

[LoKi128]

Originally posted by Soulhunter
Quantizer2 / No B-Frames / H.263 / VHQ 1
Size: 124.608.512 Bytes / Avg. PSNR: 48.55 / Time: About 06:52 min. (2600XP)

Quantizer2 / No B-Frames / H.263 / VHQ 1 / QPel / Trellis
Size: 125.386.752 Bytes / Avg. PSNR: 48.52 / Time: About 09:59 min. (2600XP)


So, adding QPel and Trellis will lower the Avg. PSNR, increase the encoding time, and only reduce the file by about 1MB.

Soul: is there any visual difference between the two encodes?

I wish I had a faster machine to play with XviD. Makes me wonder... maybe things like GMC, QPel and Trellis only really shine under "tougher" conditions. Tougher as in using more lossy quants. Maybe running the same tests but with Q5 or Q6...

[Manao]

poochie2 : XviD won't let you use quant 1 in a two pass mode. I think it is because it can't interpolate estimated quality of a quant 1 frame from a quant 2.

But in one pass encoding, as soon as you uncap the quantizer, it can use quant 1.

LoKi128 : QPel brings a hge visual difference. But it's up to you to like it or not. Treillis and GMC are far more subtile ( on my tests, in two pass encoding mode ).

Soulhunter : I made some tests, I got different filesizes ( didn't try PSNR ) You may have made a mistake somewhere.

[vlada]

Hi, I have two questions about XviD.
1) In the XviD Status Window I have about twice as much B-VOPs (are they b-frames?) as P-VOBs. Is this right? The movies look good, so I belive it is O.K. I just thought, that there are much less b-frames then p-frames in the movies.
2) If I use a constant quality (quantizer) encoding, does it have the same quality as a 2-pass encoded movie with the same size? I belive it sould, but it doesen§t seem to me so.
I hope You understand what I wanted to ask. My english is not perfect, so excuse me please. Thanks in advance, Vlada

[Manao]

Originally posted by vlada
1) In the XviD Status Window I have about twice as much B-VOPs (are they b-frames?) as P-VOBs. Is this right? The movies look good, so I belive it is O.K.It's totally OK, since you allowed the codec to use 2 b-frames in a row ( it's a default settings when using b-frames ) if he wanted to. That means there can't be more b-frames than two times the amount of p-frames.I just thought, that there are much less b-frames then p-frames in the movies.What do you mean by that ? Can you precise ?If I use a constant quality (quantizer) encoding, does it have the same quality as a 2-pass encoded movie with the same size? I belive it sould, but it doesen§t seem to me so.It doesn't and it's there also totally normal. In a one pass algorithm, the codec can't forsee that a specific area will need a lot of bitrate, and hence it can't allocate enough bits to this area when it has to encode it. In two pass mode, it knows it will have to encode a difficult portion, and so it lowers slightly the bitrate before encoding that portion, in order to have enough bits for it, and still achieve the size given.

[Shalcker]

Originally posted by poochie2
Is it a bug or whatelse that doesn't allow me to get to q1 frames? setting it in the codec will result in a movie that will never go lower than q2, as the debug window tells me and as the filesize confirms... :scared:
DivX 5 allows it and my sample jumps up to 6 mb VS the 2.1 of the Q2 of both xvid1b2 and divx5 (5.1.1).

Any hint? :confused:
Set Quant 1 zone in the beginning of the movie in the first-pass, and remove it in second pass. That will allow codec to select quant 1 in second pass when appropriate.

[Asrial]

Originally posted by nanji

If the 2nd-pass size is greater than 2G, xvid will keep using quant 2 value to compress.I'm experiencing this issue also.

1st pass = 3.11gb

2nd pass = 3.10gb even though I put '2936012' in the kbyte field (~2.8gb)

I haven't gotten through the rest of this thread yet so forgive me if this has been addressed :p

[poochie2]

My cpu is currently busy and will be for other 12 hours (vhq4 you know...:rolleyes:). I shall try as soon as I can... one other thing: while doing this shall I anyway set the minimum q to 1? Why must the zone setting be removed in second pass? Is it just to leave him the choice after it knows that, if needed, it can go down to q1?
:p

[Shalcker]

Originally posted by poochie2
while doing this shall I anyway set the minimum q to 1?
Yes. Xvid always obeys these quantizer restrictions.

Why must the zone setting be removed in second pass? Is it just to leave him the choice after it knows that, if needed, it can go down to q1? :p
Yes. Using this technique XviD gathers information on quant 1 frames in first pass; and in second pass you certainly want to reach desired filesize rather then having same size as in first pass. Since fixed quant zones aren't scaled in second pass, you'll get the same filesize (and a rather huge one), if you will not remove that zone :)

Edit: I think a bit reworded explanation is easier to understand.

With current code XviD never goes in second pass to quantizers lower then those used in first pass. If you will use quant 1 in first pass using zone feature, that will be minimum quantizer used in that zone rather then default quant 2.

[m0rtal]

According to this (http://babelfish.altavista.com/babelfish/urltrurl?url=http%3A%2F%2Fwww.ixbt.com%2Fcpu%2Fpentium4-xvid-opt.shtml&lp=ru_en&tt=url) and this (http://babelfish.altavista.com/babelfish/urltrurl?tt=url&url=http%3A%2F%2Fwww.ixbt.com%2Fcpu%2Fpentium4-32ghz-axp3200-video-encoding.shtml&lp=ru_en) articles (automatic translation from russian, but pretty good, though), current XviD betas are not optimized to work on P4 processors. anyone ever encounter such strangeness?!

[Tuning]

Originally posted by Danzel
@Tuning, remember that Gamr's builds are done with GCC and Keopis builds are done with the Intel Compiler (iirc) so they will definately have different optimizations, wanna try with keopis 1.0beta2 build to see if their results are correct?

Ok I reverted back to beta2, by downloading from koepi's dev site. Again, no difference can be found. With all optimizaton selected I get fastest speed. Somewhat less with MMX disabled. So now I can clearly say there is no fact in that news. :sly:

[Koepi]

mortal:

Cruncher tested some compiler settings, and it was the intel compiler who spit out these strange results using the flag /Ot.

For beta 3 it'll be replaced and it should work fast as expected again.

Regards
Koepi

[gino25]

Originally posted by Koepi
mortal:

Cruncher tested some compiler settings, and it was the intel compiler who spit out these strange results using the flag /Ot.

For beta 3 it'll be replaced and it should work fast as expected again.

Regards
Koepi

Good:D

And when will be released the beta3?

[mikeson]

@gino25:

Don't be impatient. If you are, you are free to download source from CVS and build it yourself.

[Tommy Carrot]

Originally posted by Manao
In a one pass algorithm, the codec can't forsee that a specific area will need a lot of bitrate, and hence it can't allocate enough bits to this area when it has to encode it. In two pass mode, it knows it will have to encode a difficult portion, and so it lowers slightly the bitrate before encoding that portion, in order to have enough bits for it, and still achieve the size given.

Sorry, this is not true. Constant quality mode doesn't have any problem with the difficult scenes, and doesn't necessarily have worse quality than 2-pass mode. The problem is, the target bitrate/filesize is unpredictable.

[Manao]

But in quality / quant one passe approach, the quantizers can only vary between one or two values ( a little more if you use b-frames ). That's not the case if you make a two pass encoding. Hence, the bits are better allocated with two passes ( or at least they are allocated in a different way ).

[Soulhunter]

Seems something went damn wrong...

After re-installing beta2, I got complete other results !!! :eek:

I remember I haven't un-installed older XviD version before installing newer...

Could this cause the problem ??? :confused:

No matter !!!

I will redo my tests till next week... ;)

Bye

[Nibor]

@Soulhunter:
Maybe you forgot to load default options?

Is this 'issue' already in the FAQ?
If it isn't, it really should be added!!

[Tommy Carrot]

Originally posted by Manao
But in quality / quant one passe approach, the quantizers can only vary between one or two values ( a little more if you use b-frames ). That's not the case if you make a two pass encoding. Hence, the bits are better allocated with two passes ( or at least they are allocated in a different way ).

I don't think 2-pass has _clearly_ superior quality, it's just able to encode to an exact target filesize, while constant quality cannot. If the filesize is not important, it's not worth the double time.

[MajinMarc]

Ah been a few since I posted but I have some interesting results. While Encoding the Menshin Movie Samurai X Beta 2 did some interesting things. First off when it see's total black Spots it seems to make them gray or blocky. Secondly it made several High motion scens so pixelated I wanted to throw Xvid beta 2 out the window. I redid the test with the last unstable build same avs, same settings and welllets just say that it looked perfect. no pixelation, no nothing, absolutely perfect. I have a feeling that it might have something to do with the bitrate distribution of beta 2 but I think this could be a serious problem. I'm looking further into it by doing a scene to scene bitrate comparission so I'll have more details in a few days.

[Leak]

Originally posted by MajinMarc
I have a feeling that it might have something to do with the bitrate distribution of beta 2 but I think this could be a serious problem. I'm looking further into it by doing a scene to scene bitrate comparission so I'll have more details in a few days.

IIRC, the CVS build from gamr I had a look at yesterday had both "max overflow improvement" and "max overflow degradation" default to 5% instead of 60% - maybe that made the difference? Or did you set your own values for it?

np: Autechre - Left Blank (EP7)

[Koepi]

Leak:

that is PreBeta3 with many, many changes. Don't use such little values with beta1/2, you'll rn into serious problems i fear.

Await Beta3 very soon, with a huge changelog...

Regards
Koepi

[Leak]

Originally posted by Koepi
Leak:

that is PreBeta3 with many, many changes. Don't use such little values with beta1/2, you'll rn into serious problems i fear.


Oh, okay. I'm using 30% with beta 2 and am quite happy with that. :)


Await Beta3 very soon, with a huge changelog...


*awaits beta3 with huge changelog*

Dum dee dum dee dum...

*twiddles thumbs*

;)

np: Autechre - Garbagemx36 (Garbage EP)

[Nibor]

@Koepi

I have a little suggestion to the installer of Beta 3..
There could be an option in the 'components page' of the installer to choose if you want to reset to the default settings after the installation.
This would be enabled by default, so everyone who had an older build installed before and doesn't know he has to load defaults doesn't run into problems when he first encodes something!
It's a very simple thing that solves quite a few problems which else get posted on the forum!

What do you think about this?
(Hey, if you want to, I could even write this code for you ;))

.: Nibor :.

[Soulhunter]

OK, here is the corrected version of my test... ;)


Source:

The Matrix - Lobby Shootout / 3:07 min. @ 25fps / 1024x576 pix. (Lanczos)


Results:

Quantizer2 / No B-Frames / H.263 / VHQ 0
Size: 123.623.424 Bytes / Avg. PSNR: 47.97 / Time: About 04:58 min. (2600XP)

Quantizer2 / No B-Frames / H.263 / VHQ 1
Size: 124.694.528 Bytes / Avg. PSNR: 48.57 / Time: About 06:29 min. (2600XP)

Quantizer2 / No B-Frames / H.263 / VHQ 2
Size: 123.168.768 Bytes / Avg. PSNR: 48.57 / Time: About 09:48 min. (2600XP)

Quantizer2 / No B-Frames / H.263 / VHQ 3
Size: 122.978.304 Bytes / Avg. PSNR: 48.56 / Time: About 13:05 min. (2600XP)

Quantizer2 / No B-Frames / H.263 / VHQ 4
Size: 121.458.688 Bytes / Avg. PSNR: 48.54 / Time: About 16:43 min. (2600XP)


Quantizer2 / No B-Frames / MPEG / VHQ 0
Size: 142.090.240 Bytes / Avg. PSNR: 47.89 / Time: About 04:59 min. (2600XP)

Quantizer2 / No B-Frames / MPEG / VHQ 1
Size: 143.366.144 Bytes / Avg. PSNR: 48.16 / Time: About 06:53 min. (2600XP)

Quantizer2 / No B-Frames / MPEG / VHQ 2
Size: 140.619.776 Bytes / Avg. PSNR: 48.18 / Time: About 10:53 min. (2600XP)

Quantizer2 / No B-Frames / MPEG / VHQ 3
Size: 140.079.104 Bytes / Avg. PSNR: 48.18 / Time: About 14:47 min. (2600XP)

Quantizer2 / No B-Frames / MPEG / VHQ 4
Size: 136.732.672 Bytes / Avg. PSNR: 48.17 / Time: About 19:43 min. (2600XP)


Quantizer2 / No B-Frames / H.263 / VHQ 0 / Trellis
Size: 122.200.064 Bytes / Avg. PSNR: 48.00 / Time: About 05:30 min. (2600XP)

Quantizer2 / No B-Frames / H.263 / VHQ 1 / Trellis
Size: 122.939.392 Bytes / Avg. PSNR: 48.57 / Time: About 07:10 min. (2600XP)

Quantizer2 / No B-Frames / H.263 / VHQ 2 / Trellis
Size: 121.387.008 Bytes / Avg. PSNR: 48.56 / Time: About 10:10 min. (2600XP)

Quantizer2 / No B-Frames / H.263 / VHQ 3 / Trellis
Size: 121.141.248 Bytes / Avg. PSNR: 48.56 / Time: About 13:35 min. (2600XP)

Quantizer2 / No B-Frames / H.263 / VHQ 4 / Trellis
Size: 119.625.728 Bytes / Avg. PSNR: 48.53 / Time: About 17:26 min. (2600XP)


Quantizer2 / No B-Frames / MPEG / VHQ 0 / Trellis
Size: 139.116.544 Bytes / Avg. PSNR: 47.85 / Time: About 05:42 min. (2600XP)

Quantizer2 / No B-Frames / MPEG / VHQ 1 / Trellis
Size: 140.228.608 Bytes / Avg. PSNR: 48.11 / Time: About 07:35 min. (2600XP)

Quantizer2 / No B-Frames / MPEG / VHQ 2 / Trellis
Size: 137.605.120 Bytes / Avg. PSNR: 48.14 / Time: About 11:40 min. (2600XP)

Quantizer2 / No B-Frames / MPEG / VHQ 3 / Trellis
Size: 137.111.552 Bytes / Avg. PSNR: 48.14 / Time: About 15:34 min. (2600XP)

Quantizer2 / No B-Frames / MPEG / VHQ 4 / Trellis
Size: 133.941.248 Bytes / Avg. PSNR: 48.13 / Time: About 20:30 min. (2600XP)


Quantizer2 / No B-Frames / H.263 / VHQ 0 / QPel
Size: 129.626.112 Bytes / Avg. PSNR: 47.80 / Time: About 06:56 min. (2600XP)

Quantizer2 / No B-Frames / H.263 / VHQ 1 / QPel
Size: 129.744.896 Bytes / Avg. PSNR: 48.62 / Time: About 09:03 min. (2600XP)

Quantizer2 / No B-Frames / H.263 / VHQ 2 / QPel
Size: 128.079.872 Bytes / Avg. PSNR: 48.61 / Time: About 16:32 min. (2600XP)

Quantizer2 / No B-Frames / H.263 / VHQ 3 / QPel
Size: 128.034.816 Bytes / Avg. PSNR: 48.60 / Time: About 23:10 min. (2600XP)

Quantizer2 / No B-Frames / H.263 / VHQ 4 / QPel
Size: 127.442.944 Bytes / Avg. PSNR: 48.59 / Time: About 26:42 min. (2600XP)


Quantizer2 / No B-Frames / MPEG / VHQ 0 / QPel
Size: 152.709.120 Bytes / Avg. PSNR: 47.52 / Time: About 07:05 min. (2600XP)

Quantizer2 / No B-Frames / MPEG / VHQ 1 / QPel
Size: 158.093.312 Bytes / Avg. PSNR: 48.13 / Time: About 09:16 min. (2600XP)

Quantizer2 / No B-Frames / MPEG / VHQ 2 / QPel
Size: 154.347.520 Bytes / Avg. PSNR: 48.14 / Time: About 19:00 min. (2600XP)

Quantizer2 / No B-Frames / MPEG / VHQ 3 / QPel
Size: 153.978.880 Bytes / Avg. PSNR: 48.13 / Time: About 27:03 min. (2600XP)

Quantizer2 / No B-Frames / MPEG / VHQ 4 / QPel
Size: 152.594.432 Bytes / Avg. PSNR: 48.12 / Time: About 31:23 min. (2600XP)


Quantizer2 / No B-Frames / H.263 / VHQ 0 / QPel / Trellis
Size: 128.229.376 Bytes / Avg. PSNR: 47.82 / Time: About 08:00 min. (2600XP)

Quantizer2 / No B-Frames / H.263 / VHQ 1 / QPel / Trellis
Size: 127.983.616 Bytes / Avg. PSNR: 48.62 / Time: About 09:43 min. (2600XP)

Quantizer2 / No B-Frames / H.263 / VHQ 2 / QPel / Trellis
Size: 126.255.104 Bytes / Avg. PSNR: 48.60 / Time: About 16:12 min. (2600XP)

Quantizer2 / No B-Frames / H.263 / VHQ 3 / QPel / Trellis
Size: 126.142.464 Bytes / Avg. PSNR: 48.59 / Time: About 23:52 min. (2600XP)

Quantizer2 / No B-Frames / H.263 / VHQ 4 / QPel / Trellis
Size: 125.497.344 Bytes / Avg. PSNR: 48.58 / Time: About 27:23 min. (2600XP)


Quantizer2 / No B-Frames / MPEG / VHQ 0 / QPel / Trellis
Size: 150.673.408 Bytes / Avg. PSNR: 47.49 / Time: About 08:19 min. (2600XP)

Quantizer2 / No B-Frames / MPEG / VHQ 1 / QPel / Trellis
Size: 155.584.512 Bytes / Avg. PSNR: 48.09 / Time: About 10:34 min. (2600XP)

Quantizer2 / No B-Frames / MPEG / VHQ 2 / QPel / Trellis
Size: 151.709.696 Bytes / Avg. PSNR: 48.10 / Time: About 19:50 min. (2600XP)

Quantizer2 / No B-Frames / MPEG / VHQ 3 / QPel / Trellis
Size: 151.121.920 Bytes / Avg. PSNR: 48.09 / Time: About 28:07 min. (2600XP)

Quantizer2 / No B-Frames / MPEG / VHQ 4 / QPel / Trellis
Size: 149.676.032 Bytes / Avg. PSNR: 48.08 / Time: About 33:45 min. (2600XP)


Bye

[Manao]

Bug report : when importing a YV12 clip ( created with avisynth ) with a mod4 width, the YV12 decompressor inside XviD shifts the chroma position. It works normally if the width is mod8, or if the colorspace is YUY2. It works with the last unstable Koepi's build ( 240603 ). It doesn't work with the latest decoder ( pre beta 3 ) nor with the decoder of beta 2.0.

I'm not speaking about encoding, just about importing the clip.

[cipher]

Originally posted by Manao
Bug report : when importing a YV12 clip ( created with avisynth ) with a mod4 width, the YV12 decompressor inside XviD shifts the chroma position. It works normally if the width is mod8, or if the colorspace is YUY2. It works with the last unstable Koepi's build ( 240603 ). It doesn't work with the latest decoder ( pre beta 3 ) nor with the decoder of beta 2.0.

I'm not speaking about encoding, just about importing the clip.

when dev-api-4 was still in alpha stage, I encontered that once, too.

And the chroma shifting still existed after encodings, when being playbacked.

[Kb_cruncher]

I have just finished some tests with beta2 and noticed some thing about BFrames.I restricted Quants in both passes, I,P frames to 2 and BFrames to 4.

I noticed that BFrames restricted to 2 Quant 4 will actually be used, when restricted to quant 4 quant 5 and 6 are still allowed and used in approx equal ratio.

If i add an avisynth filter things change.Using MipSmooth 1.11(movieHQ)and BFrame at quant4, mipsmooth restricts the quant to 4 only, no 5 and 6.

If i use c3d then quant 6 is used for the majority of BFrames, quant 5 and 6 where used but only a very small percentage.

With both filters size was smaller than without

Only - / H.263 / VHQ1 / BFrames / - where used, no other settings enabled and no other filters.

I use aSharp somtimes and i also noticed that with low thresholds it actually gives better compression than without, which had me scratching my head as i would have thought a sharpening filter would do the opposite:confused:

[sysKin]

Originally posted by Kb_cruncher
I have just finished some tests with beta2 and noticed some thing about BFrames.I restricted Quants in both passes, I,P frames to 2 and BFrames to 4.

I noticed that BFrames restricted to 2 Quant 4 will actually be used, when restricted to quant 4 quant 5 and 6 are still allowed and used in approx equal ratio.Restrictions are applied before b-frame is scaled by ratio/offset. It's best to keep the same restrictions as for p-frames.

Radek

[symonjfox]

Hi, I just tried to encode a DVB capture from an interlaced channel. I wanted to keep it as it is and test the interlacing feature.

Well, what I have to say? Good work.

It seems there are no troubles, it works (with b frames, not like dev3) and the result is fine.

It would be nice if anybody could test it on a standalone and an interlaced TV, just to say if fields are shown correctly and if there are any hidden issues.

I tried to use FFDshow to deinterlace, but using Tomsmocomp it shows green lines and colors messed up, but using FFMPEG deinterlacer, everything is just fine :D

I also tried to mux it into MP4 stream, the 3ivX decoder doesn't support interlacing so I just see green blocks on the top ... they sayd next version will support it.

Nero Direct Show filter only decodes I frames and skip everything else. Now I'm reinstalling other Dshow filters, just to check compatibility with Interlaced material.

[cweb]

@Koepi,

please check the Tmpgenc forum on this site, because your latest build was causing Tmpgenc Plus 2.521 to bomb out whenever Tmpgenc terminated (with an error in XVID.DLL!) (because it's loading an AVS file with avisource, apparently, due to XviD handling YV12 perhaps?).

I removed XviD - Tmpgenc didn't crash when terminating.
I installed a different build - Gamr's build dated 200312220040.
It doesn't have the same problem.

Could you or someone else please look into the problem? There must be something with your build which causes this, or some other side-effect triggering XviD's dll.

I didn't (don't) have the problem with the previous beta 1's.

[mazzo]

Hi, I have been here before telling you that the encoding here is very slow (like 15 hrs on sc. pass on 2CD encode). This goes for captured avi files, not for ripped dvd-s, where the speed is much more normal.

I don't know why. It also seems to me that Virtualdub works with audio on the second pass and that this could slow down the process.

All my experiences are with XVid 1.0 beta 2 with GKnot 0.28.7.

15 hrs is a very long time.

[Koepi]

mazzo,

if the speed with dvds is normal, then it's your setup for captured files which slows things down. on the one hand you're compressing audio the same time, on the other hand you apply denoise filters which usually are very slow.

so you can't blame xvid for being slow, if you use divx it'll be just as slow or slower.

Regards
Koepi

[mazzo]

How can I avoid processing audio - encode it first and choose just mux?

[Leak]

By the way, Koepi, since at some time in the future there's going to be a bitrate calculator integrated into the GUI, which will probably work quite similar to MiniCalc - could you please add support for giving it the exact length of the video _including_ seconds? (i.e. 0:24:21 or something like that)

With minicalc, you can kludge your way around this by converting the seconds to a fraction of a minute and adding that after the seconds, but that's quite clumsy and not really obvious... (i.e. 0.2450 if you've got 24:30, or 0.2475 for 24:45)

np: Autechre - Tewe (Chiastic Slide)

[Leak]

Originally posted by Koepi
EDIT:
New standalone decoder is online - fresh "from CVS". Included: new, very fine postprocessing (but it's slow, have a 1GHz+ machine at hands ;) ). Also the nice brightness slider. Find it on my site in my sig.


I just did a 2-pass encode of the first episode of Last Exile, and while the first pass plays just like it should the second pass is upside down using this decoder. :confused:

Using ffdshow, both videos are right side up - but the postprocessing of the beta decoder gives a much better picture when compared to ffdshow, especially for the scenes with a huge storm. :(

Also, playing the 2nd pass in Zoom Player using the VMR9 renderer, I get absolutely no video (though I hear the sound) and my system periodically grinds to a halt for almost a second.

Of course, once I select Overlay Mixer in Zoom Player, it works as expected, and if I open the file in GraphEdit (even though the VMR9 renderer is used in the graph it builds) it still works.

Very strange.

EDIT: Okay, scratch half of that - this only happens with DirectVobSub 2.23 loaded. Still, why does it only happen with the XviD decoder?

And to answer my own question - forcing the output to YV12 fixes the vertical flip, YUY2 looks like crap (green and purple instead of color, the picture is much too wide, you name it), and RGB24/32 causes the flipped image.

So the question is - why does it seem to default to RGB24/32 without forcing a color space when YV12 would probably be the best choice? And why only when using Zoom Player, as it works in GraphEdit? :eek:

np: Fluke - Hang Tough (Puppy)

[HarryM]

Will be any Xmas edition of XviD? :D

[Koepi]

Leak:

ask blight (author of zoomplayer IIRC). :)

Regards
Koepi

[mazzo]

I made a test. 3 minutes of video, encoded with picvideo MJPEG qual 19 768x576 and run through GKnot. It took 27 minutes. This would indicate an encoding time of approx. 18 hours for a 2 hour movie.

My setttings:

Profile AS @ L5 / H.263 / BVOPs max consec. 3, Quantizer ratio 1.50 offset 1.00, Closed GOV


Here is my avs, generated by Gordian Knot:


#
# Created with Gordian Knot
#
# http://gknot.doom9.org
#
# PLUGINS
LoadPlugin("D:\video\DivX\GordianKnot0.28beta\mpeg2dec3.dll")
LoadPlugin("D:\video\DivX\GordianKnot0.28beta\undot.dll")
#LoadPlugin("D:\video\DivX\GordianKnot0.28beta\decomb.dll")
#LoadPlugin("D:\video\DivX\GordianKnot0.28beta\KernelDeInt.dll")
#LoadPlugin("D:\video\DivX\GordianKnot0.28beta\dgbob.dll")
#LoadPlugin("D:\video\DivX\GordianKnot0.28beta\Convolution3d.dll")
#LoadPlugin("D:\video\DivX\GordianKnot0.28beta\FluxSmooth.dll")
#LoadPlugin("D:\video\DivX\GordianKnot0.28beta\TomsMoComp.dll")
#LoadPlugin("D:\video\DivX\GordianKnot0.28beta\VSFilter.dll")
#LoadPlugin("D:\video\DivX\GordianKnot0.28beta\SimpleResize.dll")
#
# SOURCE
avisource("D:\video\vitester.avi")
#
# TRIM
#trim(startframe,endframe)
#
# IVTC
#Telecide(order=1,guide=1).Decimate()
# or use
#IVTC(44,11,95)
#GreedyHMA(1,0,4,0,0,0,0,0)
#
# CROPPING
crop(2,2,764,568)
#
# DEINTERLACING (1)
#FieldDeinterlace()
#FieldDeinterlace(blend=false)
#TomsMoComp(1,5,1)
#
# DEINTERLACING (2)
#KernelDeInt(order=1,sharp=true)
# or maybe
#DGBob(order=1,mode=0)
# DEINTERLACING (3) - special requests
#GreedyHMA(1,0,0,0,0,0,0,0)
#Telecide()
#SeparateFields()
#
#
# SUBTITLES
#VobSub("FileName")
#
# RESIZING
LanczosResize(624,464)
#
Undot()
#
# DENOISING: choose one combination (or none)
# 1) little noise
#Temporalsoften(2,3,3,mode=2,scenechange=6)
#mergechroma(blur(1.3))
#FluxSmooth(5,7)
#
# 2) medium noise
#Temporalsoften(3,5,5,mode=2,scenechange=10)
#Convolution3d("moviehq")
#FluxSmooth(7,7)
#
# 3) heavy noise
#Temporalsoften(4,8,8,mode=2,scenechange=10)
#Convolution3d("movielq")
#FluxSmooth(10,15)
#
# BORDERS
#AddBorders(left,top,right,bottom)
#
# COMPRESSIBILITY CHECK
# !!!!Snip Size now has to be 14 for use in GKnot!
#SelectRangeEvery(280,14)
#
# FOOL CCEnc
#empty = BlankClip()
#AudioDub(last,empty)




I'll post the log in pt. 2

[mazzo]

My log:

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


And here is the log:



23.12.03 20:57:07: Job "dangerous" started.

o Encode WAV -> MP3: D:\video\vitester.wav
o Xvid - First Pass: D:\video\vitester-full.avs
o Xvid - Second Pass: D:\video\vitester-full.avs
o Mux Audio.
End of Job 6 (dangerous).
_______________________________________________________________________________________________________________________

Gordian Knot -
Encoding Job Data:

Type: EncXvid_All
Number: 6
Name: dangerous
Platform: Win2000 (5.0.2195).2

Files:
VirtualDubModExe: D:\video\DivX\GordianKnot0.28beta\VirtualDubMod\VirtualDubMod.exe
NandubExe: D:\video\DivX\GordianKnot0.28beta\Nandub\Nandub.exe
AviFinal: D:\video\vitester-full_1.avi
AviMovie: D:\video\vitester-full_1_Movie.avi
AviCredits: D:\video\vitester-full_1_Credits.avi
MovieFS: D:\video\vitester-full.avs
CreditsFS:
Log : D:\video\vitester-full_1.log
Stats: D:\video\vitester-full_1.stats
Ecf : D:\video\vitester-full_1.ecf

Options:
Quality/DRF: 0
CreditsStart: 0
UseEcf: 1
CreditsIVTC: 0
CreditsAppend: 0
DeleteInterm.: 0
EnforceBitrate: 0
AntiShit: 0

Calculator:
VideoSize: 32963 kbyte.
Audio1Size: 2937 kbyte.
Audio2Size: 0 kbyte.
Overhead: 184 kbyte.
AviSize: 36084 kbyte (35 mbyte).
FilesSize: 0 kbyte.
TotalSize: 36084 kbyte (35 mbyte).
Interleaving: vbrmp3
Bitrate: 1434 k(=1000)Bits/s
Frames: 4709
FPS: 25.000
Duration: 03:8
Quality: 0.198 bits/(pixel*frame).
CompCheck3: 0.000 bits/(pixel*frame).
CompCheck5: 0.406 bits/(pixel*frame).
Resolution: 624 x 464

Audio 1:
Mode: wav -> mp3
SourceFile: D:\video\vitester.wav
WavFile: D:\video\vitester.wav
Mp3File: D:\video\vitester.mp3
Azid: -c normal
Lame: -h --cbr -b 128
Delay: 0
Interval: 1
Preload: 500
Mux: 1
Recalc: 1
DeleteWav: 0

Audio 2:
none.


20:57:07: Started Transcoding Audio.

20:57:07: Using "BeSweet" to transcode D:\video\vitester.wav
CommandLine: D:\video\DivX\GORDIA~1.28B\BeSweet.exe -core( -input "D:\video\vitester.wav" -output "D:\video\vitester.mp3" -logfile "D:\video\vitester.log" ) -ota( -hybridgain ) -azid( -c normal ) -lame( -h --cbr -b 128 ) -profile( Gordian Knot 0.28 )
20:57:25: Finished. Duration: 17 seconds.

20:57:25: Audio Done.
20:57:25: MP3-File-Size is 2943 kb.
20:57:25: Audio 1, determined: 2943 kb
20:57:25: Audio 1, calc: 2937 kb
20:57:25: New Bitrate for Movie = 1434 k(=1000)Bits/s

Adding this script to D:\video\DivX\GordianKnot0.28beta\VirtualDubMod\LastJob.vcf :

// $job "Xvid - First Pass"
// $input "D:\video\vitester-full.avs"
// $output "D:\video\vitester-full_1_Movie.avi"
// $state 0
// $start_time 0 0
// $end_time 0 0
// $script

VirtualDub.Open("D:\\video\\vitester-full.avs",0,0);
VirtualDub.video.SetDepth(24,24);
VirtualDub.video.SetMode(1);
VirtualDub.video.SetFrameRate(0,1);
VirtualDub.video.SetIVTC(0,0,-1,0);
VirtualDub.video.SetRange(0,0);
VirtualDub.video.SetCompression(0x64697678,0,10000,0);
VirtualDub.video.SetCompData(2932,"AQAAALwCAACQsggAXHZpZGVvLnBhc3MALgBwAGEAcwBzAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQVMgQCBMNQAgAEwANQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAANAAAAAAAAAAgREhMVFxkbERITFRcZGxwUFRYXGBocHhUWFxgaHB4gFhcYGhweICMXGBocHiAjJhkaHB4gIyYpGxweICMmKS0QERITFBUWFxESExQVFhcYEhMUFRYXGBkTFBUWFxgaGxQVFhcZGhscFRYXGBobHB4WFxgaGxweHxcYGRscHh8hAAAAAAAAAAAAAAAAAAAAAAAAAAABAAAAAwAAAJYAAABkAAAAAAAAAAEAAAABAAAAAAAAAAAAAAAAAAAAZAAAAPQBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEAAAAGQAAABkAAAAAQAAAAAAAAABAAAAFAAAAAAAAAAAAAAACgAAADwAAAA8AAAABgAAAAEAAAABAAAAAAAAACwBAAAAAAAAAgAAAB8AAAACAAAAHwAAAAIAAAAfAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAAAAzwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA==");
VirtualDub.video.filters.Clear();
VirtualDub.subset.Delete();
VirtualDub.SaveAVI("D:\\video\\vitester-full_1_Movie.avi");
VirtualDub.Close();

// $endjob
//
//--------------------------------------------------
// $done

20:57:25: Started Xvid - First Pass: D:\video\vitester-full.avs
21:08:10: Finished Xvid - First Pass: Duration: 10 minutes, 44 seconds.
21:08:10: Trying to open Log-file.
21:08:10: Success: Log-file open.
21:08:10: Encoded: 4709 Frames.
21:08:10: Speed: 7.305 Frames per Second.
21:08:10: New target size = 32957 Kb

Adding this script to D:\video\DivX\GordianKnot0.28beta\VirtualDubMod\LastJob.vcf :

// $job "Xvid - Second Pass"
// $input "D:\video\vitester-full.avs"
// $output "D:\video\vitester-full_1_Movie.avi"
// $state 0
// $start_time 0 0
// $end_time 0 0
// $script

VirtualDub.Open("D:\\video\\vitester-full.avs",0,0);
VirtualDub.video.SetDepth(24,24);
VirtualDub.video.SetMode(1);
VirtualDub.video.SetFrameRate(0,1);
VirtualDub.video.SetIVTC(0,0,-1,0);
VirtualDub.video.SetRange(0,0);
VirtualDub.video.SetCompression(0x64697678,0,10000,0);
VirtualDub.video.SetCompData(2932,"AgAAALwCAAC9gAAAXHZpZGVvLnBhc3MALgBwAGEAcwBzAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQVMgQCBMNQAgAEwANQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAANAAAAAAAAAAgREhMVFxkbERITFRcZGxwUFRYXGBocHhUWFxgaHB4gFhcYGhweICMXGBocHiAjJhkaHB4gIyYpGxweICMmKS0QERITFBUWFxESExQVFhcYEhMUFRYXGBkTFBUWFxgaGxQVFhcZGhscFRYXGBobHB4WFxgaGxweHxcYGRscHh8hAAAAAAAAAAAAAAAAAAAAAAAAAAABAAAAAwAAAJYAAABkAAAAAAAAAAEAAAABAAAAAAAAAAAAAAAAAAAAZAAAAPQBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEAAAAGQAAABkAAAAAQAAAAAAAAABAAAAFAAAAAAAAAAAAAAACgAAADwAAAA8AAAABgAAAAQAAAAAAAAAAAAAACwBAAAAAAAAAgAAAB8AAAACAAAAHwAAAAIAAAAfAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAAAAzwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA==");
VirtualDub.video.filters.Clear();
VirtualDub.subset.Delete();
VirtualDub.SaveAVI("D:\\video\\vitester-full_1_Movie.avi");
VirtualDub.Close();

// $endjob
//
//--------------------------------------------------
// $done

21:08:10: Started Xvid - Second Pass: D:\video\vitester-full.avs
21:23:55: Finished Xvid - Second Pass: Duration: 15 minutes, 45 seconds.
21:23:55: Speed: 4.982 Frames per Second.

Adding this script to D:\video\DivX\GordianKnot0.28beta\VirtualDubMod\LastJob.vcf :

// $job "DivX5 Append"
// $input "D:\video\vitester-full_1_Movie.avi"
// $output "D:\video\vitester-full_1.avi"
// $state 0
// $start_time 0 0
// $end_time 0 0
// $script

VirtualDub.Open("D:\\video\\vitester-full_1_Movie.avi",0,0);
VirtualDub.RemoveInputStreams();
VirtualDub.stream[0].SetSource("D:\\video\\vitester.mp3",0x00000202,0);
VirtualDub.stream[0].SetMode(0);
VirtualDub.stream[0].SetInterleave(1,500,1,0,0);
VirtualDub.stream[0].SetClipMode(1,1);
VirtualDub.stream[0].SetConversion(0,0,0,0,0);
VirtualDub.stream[0].SetVolume();
VirtualDub.stream[0].SetCompression();
VirtualDub.video.SetDepth(24,24);
VirtualDub.video.SetMode(0);
VirtualDub.video.SetFrameRate(0,1);
VirtualDub.video.SetIVTC(0,0,-1,0);
VirtualDub.video.SetRange(0,0);
VirtualDub.video.SetCompression();
VirtualDub.video.filters.Clear();
VirtualDub.subset.Delete();
VirtualDub.SaveAVI("D:\\video\\vitester-full_1.avi");
VirtualDub.Close();

// $endjob
//
//--------------------------------------------------
// $done

21:23:55: Started Muxing Audio.: D:\video\vitester-full_1_Movie.avi
21:24:07: Finished Muxing Audio.: Duration: 11 seconds.

21:24:07: Done.
21:24:07: Movie = D:\video\vitester-full_1.avi
Total Encoding Time: 26 minutes, 59 seconds.
23.12.03 21:24:07: Job "dangerous" finished.
_______________________________________________________________________________________________________________________




Is something wrong here?

[mf]

Doesn't junk like that rather belong in the GKnot forum? I can't see how it could be of interest to XviD forum readers.

[sysKin]

And more importantly, GKnot still can't configure XviD 1.0 - so it can't be used.

Radek

[mazzo]

OK guys, last time I had a similar problem, I was invited to post the logs into this forum, so don't call it junk, please. I DON'T do this for fun.

And IS it really a GKnot problem, as long as DivX 5.x through the same forum encodes with a lot higher speed. Many XVid users also use GKnot, don't they.

And BTW: This is the first time I see clearly stated that GKnot doesn't work with XVid 1.0

[Leak]

It definitely looks like a GKnot problem, as the configuration data for the XviD codec in the VirtualDub script you posted is too short - it should be 2956 bytes, not 2932 as in your script:


VirtualDub.video.SetCompData(2956,"AQAAALw...")

vs.

VirtualDub.video.SetCompData(2932,"AgAAALw...")


If what GKnot created here was configuration information for an older version of XviD it'll totally screw up XviD's configuration.

How about dumping GKnot and just encoding something with VirtualDub alone? If it then is still as slow as you say then there *is* a bug in XviD - but not if it gets configured incorrectly by some wonky application.

np: Jah Wobble & The Invaders Of The Heart - Banks Of The Sweet Primrose (English Roots Music)

[outlyer]

Originally posted by mazzo
Many XVid users also use GKnot, don't they.Can't speak for others, I do use it but only for AR calculations and AVS generation (and I've always used it the same way :P).

Originally posted by mazzo
And BTW: This is the first time I see clearly stated that GKnot doesn't work with XVid 1.0Then, maybe, you should have read the two whole beta threads.

[mf]

I'll only speak for myself too: I avoid any all-in-one turd. I can do anything(don't take my word for it, "anything" is probably too broad of a term) encoding-related better than automated programs, so I do so as much as I can.

[powerslave]

Go to the gknot forums here on doom9. Lenox has said that because the new xvid settings are different in the registry then previous alpha versions, then he would not implement full compatability with xvid 1.0 until it ended its beta run and became an official release. I believe there are some workarounds for gknot related problems with xvid 1.0 in the gknot forums so check them out.

[mazzo]

Yes, thank you, and merry christmas to all of you. I wonder: If I let GKnot make the avs and then run the avs and the whole encoding directly in VirtualDubMod, then I will be safe, won't I?

Still - there is one thing that causes some thoughts: While a ripped dvd takes about 8 hours through GKnot to finished (2 cd's), a captured avi from tv takes 20 hours. That is my experience. So I say to myself: Could there be an issue with XVid 1.0 and MJPEG avis?

[sysKin]

Originally posted by mazzo
Yes, thank you, and merry christmas to all of you. I wonder: If I let GKnot make the avs and then run the avs and the whole encoding directly in VirtualDubMod, then I will be safe, won't I?Yes. You might want to read first post on this page: http://forum.doom9.org/showthread.php?s=&threadid=66235&perpage=20&pagenumber=2
Still - there is one thing that causes some thoughts: While a ripped dvd takes about 8 hours through GKnot to finished (2 cd's), a captured avi from tv takes 20 hours. That is my experience. So I say to myself: Could there be an issue with XVid 1.0 and MJPEG avis? As explained here:
http://forum.doom9.org/showthread.php?s=&threadid=66806
one of the effects of using gknot is horribly low encoding speed. Load default settings to clear gknot's registry entries.

Radek