Need help with results for first backup with AutoGK, Thanks!

[bloomcounty]

So I backed up an episode of Animaniacs. I've done it two ways so far, both using cartoon mode. One I let the program automatically choose the resolution. It converted it at 512x384. I set the target size at 180MB (as that seemed to be the size of all the episodes of Teen Titans off of TV that I have). Properties show 115Kps and 12-bit.

(What determines the 12-bit? I see that my TT are 24-bit when I do properties on them.)

Here's the log:
[2/10/2007 11:32:39 PM] AutoGK 2.40
[2/10/2007 11:32:39 PM] OS: WinXP (5.1.2600).2
[2/10/2007 11:32:39 PM] Job started.
[2/10/2007 11:32:39 PM] Input file: C:\DVD Rips\ANIMANIACS_VOLUME_1_DISC_1\VTS_01_1.VOB
[2/10/2007 11:32:39 PM] Output file: C:\DVD Rips\ANIMANIACS_VOLUME_1_DISC_1\VTS_01_1.avi
[2/10/2007 11:32:39 PM] Output codec: XviD
[2/10/2007 11:32:39 PM] Audio 1: Audio Stream 1 AC3
[2/10/2007 11:32:39 PM] Subtitles: none
[2/10/2007 11:32:39 PM] Format: AVI
[2/10/2007 11:32:39 PM] Target size: 180Mb
[2/10/2007 11:32:39 PM] Audio 1 settings: Auto
[2/10/2007 11:32:39 PM] Standalone support enabled: ESS
[2/10/2007 11:32:39 PM] Started encoding.
[2/10/2007 11:32:39 PM] Demuxing and indexing.
[2/10/2007 11:33:41 PM] Processing file: C:\DVD Rips\ANIMANIACS_VOLUME_1_DISC_1\VTS_01_1.VOB
[2/10/2007 11:33:41 PM] Source resolution: 720x480
[2/10/2007 11:33:41 PM] Found NTSC source.
[2/10/2007 11:33:41 PM] Source aspect ratio: 4:3
[2/10/2007 11:33:41 PM] Color correction enabled.
[2/10/2007 11:33:41 PM] Analyzing source.
[2/10/2007 11:34:40 PM] Source has percentage of interlacing in motion areas: 24.89
[2/10/2007 11:34:40 PM] Source has percentage of telecined patterns: 69.54
[2/10/2007 11:34:40 PM] Source has percentage of progressive patterns: 30.46
[2/10/2007 11:34:40 PM] Source has percentage of interlaced patterns: 0.00
[2/10/2007 11:34:40 PM] Source is considered to be hybrid (mostly FILM).
[2/10/2007 11:34:40 PM] Looking for optimal hybrid thresholds.
[2/10/2007 11:34:50 PM] Found threshold of: 1.95
[2/10/2007 11:34:50 PM] Output will contain 31907 frames
[2/10/2007 11:34:50 PM] Decoding audio.
[2/10/2007 11:35:22 PM] Normalizing audio.
[2/10/2007 11:35:38 PM] Encoding audio.
[2/10/2007 11:38:02 PM] Using cartoon mode.
[2/10/2007 11:38:02 PM] Audio1 size: 19,139,208 bytes (18.25 Mb)
[2/10/2007 11:38:02 PM] Overhead: 1,276,288 bytes (1.22 Mb)
[2/10/2007 11:38:02 PM] Video size: 168,328,184 bytes (160.53 Mb)
[2/10/2007 11:38:02 PM] Running compressibility test.
[2/10/2007 11:38:02 PM] Writing the following script to C:\DVD Rips\ANIMANIACS_VOLUME_1_DISC_1\agk_tmp\VTS_01_1_c omptest.avs
================================================== =========
LoadPlugin("C:\PROGRA~1\AutoGK\DGMPGDec\DGDecode.dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\autocrop.dl l")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\decomb.dll" )
LoadPlugin("C:\PROGRA~1\AutoGK\filters\ColorMatrix .dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\RemoveGrain SSE3.dll")

movie = mpeg2source("C:\DVD Rips\ANIMANIACS_VOLUME_1_DISC_1\agk_tmp\VTS_01_1.d 2v")
cropclip = autocrop(movie,mode=0,wmultof=4,hmultof=4,samples= 10,aspect=0,threshold=34,samplestartframe=0,leftad d=0,rightadd=0,topadd=0,bottomadd=0)
fixed_aspect = 0.888888888888889
c_width = width(cropclip)
c_height = round(height(cropclip) / fixed_aspect)
input_par = float(c_width)/float(c_height)
input_par = (input_par > 1.4) || (input_par < 1.25) ? input_par : (4.0/3.0)
out_width = 640
out_height = round(float(out_width) / input_par)
hmod = out_height - (floor(out_height / 16 ) * 16)
out_height = (hmod > 4) ? (out_height + (16 - hmod)) : (out_height - hmod)
new_aspect = (float(out_width) / float(out_height)) / fixed_aspect
Telecide(movie,guide=1,post=2).Decimate(mode=3,thr eshold=1.95)
autocrop(mode=0,wmultof=4,hmultof=4,samples=10,asp ect=new_aspect,threshold=34,samplestartframe=0,lef tadd=0,rightadd=0,topadd=0,bottomadd=0)
LanczosResize(out_width,out_height)
RemoveGrain(mode=2)
ColorMatrix("Rec.709->Rec.601",opt=0,hints=false)
SelectRangeEvery(214,15)
================================================== =========
[2/10/2007 11:40:50 PM] Duration was: 2 minutes 48 seconds
[2/10/2007 11:40:50 PM] Speed was: 13.28 fps.
[2/10/2007 11:40:51 PM] Compressibility percentage is: 37.30
[2/10/2007 11:40:51 PM] Chosen resolution is: 512x384 ( AR: 1.33 )
[2/10/2007 11:40:51 PM] Predicted comptest value is: 58.29%
[2/10/2007 11:40:51 PM] Running first pass.
[2/10/2007 11:40:51 PM] Writing the following script to C:\DVD Rips\ANIMANIACS_VOLUME_1_DISC_1\agk_tmp\VTS_01_1_m ovie.avs
================================================== =========
LoadPlugin("C:\PROGRA~1\AutoGK\DGMPGDec\DGDecode.d ll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\autocrop.dl l")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\decomb.dll" )
LoadPlugin("C:\PROGRA~1\AutoGK\filters\ColorMatrix .dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\RemoveGrain SSE3.dll")

movie = mpeg2source("C:\DVD Rips\ANIMANIACS_VOLUME_1_DISC_1\agk_tmp\VTS_01_1.d 2v")
cropclip = autocrop(movie,mode=0,wmultof=4,hmultof=4,samples= 10,aspect=0,threshold=34,samplestartframe=0,leftad d=0,rightadd=0,topadd=0,bottomadd=0)
fixed_aspect = 0.888888888888889
c_width = width(cropclip)
c_height = round(height(cropclip) / fixed_aspect)
input_par = float(c_width)/float(c_height)
input_par = (input_par > 1.4) || (input_par < 1.25) ? input_par : (4.0/3.0)
out_width = 512
out_height = round(float(out_width) / input_par)
hmod = out_height - (floor(out_height / 16 ) * 16)
out_height = (hmod > 4) ? (out_height + (16 - hmod)) : (out_height - hmod)
new_aspect = (float(out_width) / float(out_height)) / fixed_aspect
Telecide(movie,guide=1,post=2).Decimate(mode=3,thr eshold=1.95)
autocrop(mode=0,wmultof=4,hmultof=4,samples=10,asp ect=new_aspect,threshold=34,samplestartframe=0,lef tadd=0,rightadd=0,topadd=0,bottomadd=0)
LanczosResize(out_width,out_height)
RemoveGrain(mode=2)
ColorMatrix("Rec.709->Rec.601",opt=0,hints=false)
================================================== =========
[2/10/2007 11:58:24 PM] Duration was: 17 minutes 33 seconds
[2/10/2007 11:58:24 PM] Speed was: 30.28 fps.
[2/10/2007 11:58:24 PM] Expected quality of first pass size: 51.06%
[2/10/2007 11:58:24 PM] Running second pass.
[2/11/2007 12:20:43 AM] Duration was: 22 minutes 17 seconds
[2/11/2007 12:20:43 AM] Speed was: 23.85 fps.
[2/11/2007 12:20:43 AM] Job finished. Total time: 48 minutes 3 seconds


And then I did it with the same target size, but setting the resolution to 640x480 (which is that my TT episodes are). Here's the log:
[2/11/2007 1:06:04 AM] AutoGK 2.40
[2/11/2007 1:06:04 AM] OS: WinXP (5.1.2600).2
[2/11/2007 1:06:04 AM] Job started.
[2/11/2007 1:06:04 AM] Input file: C:\DVD Rips\ANIMANIACS_VOLUME_1_DISC_1\VTS_01_1.VOB
[2/11/2007 1:06:04 AM] Output file: C:\DVD Rips\ANIMANIACS_VOLUME_1_DISC_1\VTS_01_1.avi
[2/11/2007 1:06:04 AM] Output codec: XviD
[2/11/2007 1:06:04 AM] Audio 1: Audio Stream 1 AC3
[2/11/2007 1:06:04 AM] Subtitles: none
[2/11/2007 1:06:04 AM] Format: AVI
[2/11/2007 1:06:04 AM] Target size: 180Mb
[2/11/2007 1:06:04 AM] Custom resolution settings: fixed width of 640 pixels
[2/11/2007 1:06:04 AM] Audio 1 settings: Auto
[2/11/2007 1:06:04 AM] Standalone support enabled: ESS
[2/11/2007 1:06:04 AM] Started encoding.
[2/11/2007 1:06:04 AM] Demuxing and indexing.
[2/11/2007 1:07:03 AM] Processing file: C:\DVD Rips\ANIMANIACS_VOLUME_1_DISC_1\VTS_01_1.VOB
[2/11/2007 1:07:03 AM] Source resolution: 720x480
[2/11/2007 1:07:03 AM] Found NTSC source.
[2/11/2007 1:07:03 AM] Source aspect ratio: 4:3
[2/11/2007 1:07:03 AM] Color correction enabled.
[2/11/2007 1:07:03 AM] Analyzing source.
[2/11/2007 1:08:03 AM] Source has percentage of interlacing in motion areas: 24.89
[2/11/2007 1:08:03 AM] Source has percentage of telecined patterns: 69.54
[2/11/2007 1:08:03 AM] Source has percentage of progressive patterns: 30.46
[2/11/2007 1:08:03 AM] Source has percentage of interlaced patterns: 0.00
[2/11/2007 1:08:03 AM] Source is considered to be hybrid (mostly FILM).
[2/11/2007 1:08:03 AM] Looking for optimal hybrid thresholds.
[2/11/2007 1:08:13 AM] Found threshold of: 1.95
[2/11/2007 1:08:13 AM] Output will contain 31907 frames
[2/11/2007 1:08:13 AM] Decoding audio.
[2/11/2007 1:08:39 AM] Normalizing audio.
[2/11/2007 1:09:08 AM] Encoding audio.
[2/11/2007 1:11:31 AM] Using cartoon mode.
[2/11/2007 1:11:31 AM] Audio1 size: 19,139,208 bytes (18.25 Mb)
[2/11/2007 1:11:31 AM] Overhead: 1,276,288 bytes (1.22 Mb)
[2/11/2007 1:11:31 AM] Video size: 168,328,184 bytes (160.53 Mb)
[2/11/2007 1:11:31 AM] Running compressibility test.
[2/11/2007 1:11:31 AM] Writing the following script to C:\DVD Rips\ANIMANIACS_VOLUME_1_DISC_1\agk_tmp\VTS_01_1_c omptest.avs
================================================== =========
LoadPlugin("C:\PROGRA~1\AutoGK\DGMPGDec\DGDecode.d ll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\autocrop.dl l")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\decomb.dll" )
LoadPlugin("C:\PROGRA~1\AutoGK\filters\ColorMatrix .dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\RemoveGrain SSE3.dll")

movie = mpeg2source("C:\DVD Rips\ANIMANIACS_VOLUME_1_DISC_1\agk_tmp\VTS_01_1.d 2v")
cropclip = autocrop(movie,mode=0,wmultof=4,hmultof=4,samples= 10,aspect=0,threshold=34,samplestartframe=0,leftad d=0,rightadd=0,topadd=0,bottomadd=0)
fixed_aspect = 0.888888888888889
c_width = width(cropclip)
c_height = round(height(cropclip) / fixed_aspect)
input_par = float(c_width)/float(c_height)
input_par = (input_par > 1.4) || (input_par < 1.25) ? input_par : (4.0/3.0)
out_width = 640
out_height = round(float(out_width) / input_par)
hmod = out_height - (floor(out_height / 16 ) * 16)
out_height = (hmod > 4) ? (out_height + (16 - hmod)) : (out_height - hmod)
new_aspect = (float(out_width) / float(out_height)) / fixed_aspect
Telecide(movie,guide=1,post=2).Decimate(mode=3,thr eshold=1.95)
autocrop(mode=0,wmultof=4,hmultof=4,samples=10,asp ect=new_aspect,threshold=34,samplestartframe=0,lef tadd=0,rightadd=0,topadd=0,bottomadd=0)
LanczosResize(out_width,out_height)
RemoveGrain(mode=2)
ColorMatrix("Rec.709->Rec.601",opt=0,hints=false)
SelectRangeEvery(214,15)
================================================== =========
[2/11/2007 1:14:22 AM] Duration was: 2 minutes 50 seconds
[2/11/2007 1:14:22 AM] Speed was: 13.11 fps.
[2/11/2007 1:14:22 AM] Compressibility percentage is: 37.30
[2/11/2007 1:14:22 AM] Using softer resizer.
[2/11/2007 1:14:22 AM] Chosen resolution is: 640x480 ( AR: 1.33 )
[2/11/2007 1:14:22 AM] Predicted comptest value is: 39.39%
[2/11/2007 1:14:22 AM] Running first pass.
[2/11/2007 1:14:22 AM] Writing the following script to C:\DVD Rips\ANIMANIACS_VOLUME_1_DISC_1\agk_tmp\VTS_01_1_m ovie.avs
================================================== =========
LoadPlugin("C:\PROGRA~1\AutoGK\DGMPGDec\DGDecode.d ll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\autocrop.dl l")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\decomb.dll" )
LoadPlugin("C:\PROGRA~1\AutoGK\filters\ColorMatrix .dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\RemoveGrain SSE3.dll")

movie = mpeg2source("C:\DVD Rips\ANIMANIACS_VOLUME_1_DISC_1\agk_tmp\VTS_01_1.d 2v")
cropclip = autocrop(movie,mode=0,wmultof=4,hmultof=4,samples= 10,aspect=0,threshold=34,samplestartframe=0,leftad d=0,rightadd=0,topadd=0,bottomadd=0)
fixed_aspect = 0.888888888888889
c_width = width(cropclip)
c_height = round(height(cropclip) / fixed_aspect)
input_par = float(c_width)/float(c_height)
input_par = (input_par > 1.4) || (input_par < 1.25) ? input_par : (4.0/3.0)
out_width = 640
out_height = round(float(out_width) / input_par)
hmod = out_height - (floor(out_height / 16 ) * 16)
out_height = (hmod > 4) ? (out_height + (16 - hmod)) : (out_height - hmod)
new_aspect = (float(out_width) / float(out_height)) / fixed_aspect
Telecide(movie,guide=1,post=2).Decimate(mode=3,thr eshold=1.95)
autocrop(mode=0,wmultof=4,hmultof=4,samples=10,asp ect=new_aspect,threshold=34,samplestartframe=0,lef tadd=0,rightadd=0,topadd=0,bottomadd=0)
BicubicResize(out_width,out_height,0,0.5)
RemoveGrain(mode=2)
ColorMatrix("Rec.709->Rec.601",opt=0,hints=false)
================================================== =========
[2/11/2007 1:35:56 AM] Duration was: 21 minutes 33 seconds
[2/11/2007 1:35:57 AM] Speed was: 24.66 fps.
[2/11/2007 1:35:57 AM] Expected quality of first pass size: 37.46%
[2/11/2007 1:35:57 AM] Running second pass.
[2/11/2007 2:03:18 AM] Duration was: 27 minutes 20 seconds
[2/11/2007 2:03:19 AM] Speed was: 19.45 fps.
[2/11/2007 2:03:19 AM] Job finished. Total time: 57 minutes 15 seconds

So which of this is better to do? What's the difference? (I'll be burning and watching later to test as well...)

I'm going to try one without the cartoon mode. But is cartoon mode generally better?

I also noticed that you can only have ONE VOB file in the folder when converting, or it will do them all together (?) -- is that correct. I wasted like 8 hours until I finally figured that out...

Any help is appreciated -- thanks!

[manono]

Hi-

XviD is always 12bit, no matter what your properties say.

Which is the better to do? In this case 512x384, because the results of the compression test when you forced 640x480 was way too low.

I have no idea what Teen Titans or Animaniacs are, but Cartoon Mode is for very simply drawn cartoons, like maybe The Simpsons, while for any kind of detailed drawing, like for many anime, you shouldn't use Cartoon Mode.

If the Vobs are numbered consecutively, they should all be loaded by picking just the first one. Better would be to decrypt using DVD Decrypter in IFO Mode (up at the top, Mode->IFO), and then load the IFO instead.

[celtic_druid]

The 12/24bit refers to the input. Since the source is generally 12bit, 12bit is better. 24bit means that someone converted to RGB24 along the way, which results in lower quality. Generally a result of using VDub plugins or VFAPI. AutoGK uses AVISynth so the encoding chain is kept as YV12.

[bloomcounty]

The 12/24bit refers to the input. Since the source is generally 12bit, 12bit is better. 24bit means that someone converted to RGB24 along the way, which results in lower quality.

Teen Titans was off of TV then converted to XviD (or maybe DivX). Thanks!

[bloomcounty]

Which is the better to do? In this case 512x384, because the results of the compression test when you forced 640x480 was way too low.

Okay. But when I open the 512x384 files in, say, Windows Media Player on my laptop, it's actually physically smaller than the 640 one (which, I guess, would make sense since it's a smaller resolution). But that means it's going to be "enlarged" when it's played back on my TV through my Philips dvd player. So won't I lose quality by it being enlarged? Or will it still be better than the 640 one (which wouldn't need to be enlarged).

I have no idea what Teen Titans or Animaniacs are, but Cartoon Mode is for very simply drawn cartoons, like maybe The Simpsons, while for any kind of detailed drawing, like for many anime, you shouldn't use Cartoon Mode.

Animaniacs: www.tv.com/animaniacs/show/2077/videos.html

...does that look simple enough to warrant using cartoon mode?

If the Vobs are numbered consecutively, they should all be loaded by picking just the first one. Better would be to decrypt using DVD Decrypter in IFO Mode (up at the top, Mode->IFO), and then load the IFO instead.

I used RipIt4Me, which uses DVDDecrypter. I first tried loading the IFO in AutoGK, which then gave me an option to pick one of the five episodes on the disc. I picked the first one, then let it do its thing. But it actually converted ALL FIVE together in one file (there is a "play all" option, so maybe it was somehow looking at that?). But then I tried just picking the first VOB instead of the IFO, and it loaded the first one, and then ended up doing the same thing (all five together in one). So I just deleted everything but the VOBs for the five episodes, then moved the other four into a sub-folder, then told AutoGK to pick the first (and only) VOB in the folder, and that seemed to work.

So is there a better way to do that?

Thanks!

[celtic_druid]

The idea is that given a low enough bitrate, the lower resolution encode will look better when enlarged than the higher resolution one. It is a compromise between artefacts from enlarging and artefacts from the lower bits/pixels ratio of the higher resolution.

[bloomcounty]

Okay, so I burned my three tests and they all kind of looked the same, but there's the same "problem" with each...

The image quality itself, while decent, looks weird when there's motion. It's kind of "stuttery" -- like there's a frame here or there missing or something (not saying that's the case, that's just the only way I can think to describe it). It's not super-stuttery (like a packed bitstreams thing), but it definitely looks off.

(I don't know if converting them with the MediaTek setting instead of the ESS would make any difference with that...?)

So what is causing this? And what settings can I change to get rid of the problem?

Thanks!

[manono]

So won't I lose quality by it being enlarged? Or will it still be better than the 640 one (which wouldn't need to be enlarged).

celtic_druid answered that one. And I agree. In addition, because the compress test result was so low, AutoGK used both a softer resizer and a softer matrix, softening the 640x480 version as compared to the 512x384 one. So, even though the 512x384 one has to be upscaled, thus softening it, the larger resolution one has already been softened to a degree. In addition, the larger one will probably have other kinds of artifacts, such as mosquito noise, color smearing, and blockiness in complex scenes, making it definitely inferior to the lower res one. In my opinion.

The image quality itself, while decent, looks weird when there's motion. It's kind of "stuttery"

That's probably because of it being treated as a hybrid. You could try and force an IVTC in the Hidden Options, but if it's really a hybrid, that may not help. If you like, upload a small sample of the source vob (10 seconds or so) that shows the problem after being reencoded for XviD and we can have a look.

I don't know if converting them with the MediaTek setting instead of the ESS would make any difference with that...?)

No, not on the computer. If you intend to play them on a standalone, you had better know which chipset your player uses.

Yeah, I guess Cartoon Mode is OK for those.

I first tried loading the IFO in AutoGK, which then gave me an option to pick one of the five episodes on the disc. I picked the first one, then let it do its thing. But it actually converted ALL FIVE together in one file.

Which is why I suggested using DVD Decrypter set for IFO Mode, which will allow you to demux your episodes one-by-one. If they're all in the same PGC, you'll have to select the correct number of chapters to demux for each episode. If they're all in separate PGCs (as seems to be the case with yours), then you just select each PGC one-after-the-other, and decrypt each to its own folder. I seriously doubt that you need to use RipIt4Me on these things.

Either that or split them into episodes after they're all done being encoded for XviD.

[bloomcounty]

That's probably because of it being treated as a hybrid. You could try and force an IVTC in the Hidden Options, but if it's really a hybrid, that may not help. If you like, upload a small sample of the source vob (10 seconds or so) that shows the problem after being reencoded for XviD and we can have a look.

How can you have the source vob after it's reencoded for XviD? One's before and one's after, right? I'm not sure what you're asking for here. I'd love to upload something for you to look at, but I don't think I have the tools to just take 10 seconds of either the original vob or the encoded .avi -- or can I do this with one of the ripping programs we've been talking about?

No, not on the computer. If you intend to play them on a standalone, you had better know which chipset your player uses.

I have two players, one is ESS the other is MediaTek -- so I chose the ESS since that should work on both, right? I burned them and played them on both players, and the looked the same on both. (The stuttery-ness isn't as noticeable on the laptop, but it's still there. It's like there's frames missing and then the frames on either side of the missing frame are being quickly dissolved from one to the other... and least, that's the best way I can describe it...)

Which is why I suggested using DVD Decrypter set for IFO Mode, which will allow you to demux your episodes one-by-one. If they're all in the same PGC, you'll have to select the correct number of chapters to demux for each episode. If they're all in separate PGCs (as seems to be the case with yours), then you just select each PGC one-after-the-other, and decrypt each to its own folder. I seriously doubt that you need to use RipIt4Me on these things.

I *think* I understand what you're saying here, but I'd rather not have to rip all five discs again if I don't have to. Do you think that's necessary? They're all ripped with RipIt4Me right now on my laptop's harddrive. Is there anything "wrong" with doing it the way I did it, but just having the VOB for one epsiode as the only thing in the folder and converting it with AutoGK by selecting the VOB? Would it make any difference if I ripped them and converted them the way you're saying? Could that be behind the stuttery-ness? Or is it all the same final product?

Thanks for the help!

[manono]

Hi-

I'd love to upload something for you to look at, but I don't think I have the tools to just take 10 seconds of either the original vob or the encoded .avi -- or can I do this with one of the ripping programs we've been talking about?

Try and cut a piece from the source vob that has a problem after being converted to XviD. Open the vob in DGIndex. Use the [ and ] buttons to isolate a small section. Then File->Save Project and Demux Video. Upload it somewhere and post the link. If you need to know of a place to upload, just ask.

...so I chose the ESS since that should work on both, right?

Yes, you did exactly right.

I *think* I understand what you're saying here, but I'd rather not have to rip all five discs again if I don't have to.

I don't blame you. It's just for future reference.

Is there anything "wrong" with doing it the way I did it, but just having the VOB for one epsiode as the only thing in the folder and converting it with AutoGK by selecting the VOB?

Probably not in this case, but there can sometimes be something wrong with doing it that way. You might sometimes get stuff in the vob that's not part of the episode, such as a studio or production company logo, or even an FBI warning. You might get asynched audio. Also, I'm not entirely sure I understand what you're saying. You have all the vobs on the hard drive in one large file, right? And you're saying each episode corresponds exactly to one 1 GB vob? No more and no less? I must be misunderstanding, because that's patently untrue. Did you cut the vobs somehow, to get the separate episodes?

[bloomcounty]

Okay, here's an update:

1. I chose the Force IVTC in the Hidden Options and converted the episode, and when I played it back on my laptop, it looked to me like the stuttery-effect was gone (I'll have to burn it to cd and play it on my dvd player to confirm -- but there's another problem I need to ask you about first).

2. I'm almost positive I need to rerip all the dvds again the way you said to with DVD Decrpyter. I did the first disc that way, but every time I try to convert an episode that's NOT the first episode in AutoGK, it still just converts the first episode. What exactly should I do, step-by-step, from first opening DVD Decrypter to encoding all five episodes as separate .avi files in AutoGK? I think I might be missing a step somewhere. I don't have to re-rip the dvd each time I convert one episode, do I? (I wouldn't think so...) I'd think I should be able to just rip it once, then be able to convert each episode one after the other, right?

What is happening is that when I rip them in IFO like you said, it's ripping the entire VOB (all five episodes together), even though I'm selecting just one episode to rip. So that's why AutoGK keeps going back to the beginning no matter what, it's because the VOB is all five together. How do I rip the episodes separately so that the VOB that's ripped is ONLY for that paticular episode?

EDIT: OKAY, I'M DUMB -- THAT WAS THE VOB FOR ONLY THE FIRST EPISODE. THE IFO IN AUTOGK JUST LETS YOU CHOOSE ANY OF THE FIVE, BUT THE VOB IS ONLY FOR THE FIRST ONE. GOT IT...

3. What is the force interlace option for hybrid files (or somethink like that), that's right under the Force IVTC that I used? Would this also help or is it something completely different?

Thanks!

[manono]

So, 1 doesn't require an answer, and 2 doesn't either?

3. Force Deinterlace will deinterlace it and keep it at 29.97fps. That's a last resort that you never want to use unless you absolutely have to. IVTC reassembles the fields into frames and then tosses the duplicate frames, restoring it to its original 23.976fps. Admittedly this gets a little screwy with animation, as much of it is drawn at various framerates (8, 12, or 16fps, as well as 24fps), and has duplicate frames already inserted to bring it all up to 24fps. If Force IVTC made it play smoothly again, then I guess that was the correct framerate, and you can use the Force IVTC option for as long as you're working on these episodes.

[bloomcounty]

So, 1 doesn't require an answer, and 2 doesn't either?

Not so far... I'll have to do my new burn test, then maybe...

3. Force Deinterlace will deinterlace it and keep it at 29.97fps. That's a last resort that you never want to use unless you absolutely have to. IVTC reassembles the fields into frames and then tosses the duplicate frames, restoring it to its original 23.976fps. Admittedly this gets a little screwy with animation, as much of it is drawn at various framerates (8, 12, or 16fps, as well as 24fps), and has duplicate frames already inserted to bring it all up to 24fps. If Force IVTC made it play smoothly again, then I guess that was the correct framerate, and you can use the Force IVTC option for as long as you're working on these episodes.

So can I assume all the episodes are hybrid the same way? I'd hate to have to encode them once, check, then re-encode if they're stuttery. Is there some way to make sure without having to do the whole encode?

Thanks!

[manono]

I've never worked with those animations, so I can't be positive, but an educated guess says that if Force IVTC works with one episode, it'll work with them all.

Decomb (the IVTC being used in AutoGK) is being used at default settings. This doesn't always work well with animations. It has a setting to use with animations, but you can't access it in AutoGK unless you apply one of the add-ons which lets you edit the AviSynth script being used. But I guess until you get more experience and can do these things manually, you'll have to hope for the best.

[bloomcounty]

Okay, so I burned them both to cd and watched, and it looks to me that both cartoon mode and not using cartoon mode look pretty much the same. So which should I use?

(I also do notice, that while the video doesn't stutter anymore, it does jump a few frames occasionally... maybe a side-effect of the IVTC since it's not really for animation?

[manono]

If it's any help, I've never used Cartoon Mode, and although I've never encoded any simple cartoons, I've encoded a heck of a lot of anime.

I was hoping I wouldn't have to try and explain this. Decomb first lines up the fields to make 29.97fps progressive, and then it decimates back to 23.976. If a scene was drawn at 8 fps, then after decimation makes the framerate 23.976fps, there should be 3 frames the same to convert 8 fps to 24fps, or 1 1 1 2 2 2 3 3 3. Understand? If it doesn't choose correctly, you may wind up with something like 1 1 2 2 2 2 3 3 3 4 4. Understand? That kind of a result will make it play more jerkily than it should. To find out if that's happening to you, open the completed AVI in VDubMod, scroll to a place where you saw jerkiness, and start advancing a frame at a time, counting the number of times you see the same frame.

Decomb has a Decimate(Mode=2) designed for animation, which is supposed to prevent the kind of thing I just outlined. You can read the DecombReferenceManual for more information.

However, you may just be seeing regular old hand drawn cartoon jerkiness. Since it's often drawn at framerates lower than 24fps, cartoons always play kind of jerky anyway. Again, you can open the AVI in VDubMod and advance a frame at a time. Examining the frames as described before will tell you if it was drawn at 8, 12, 16, or 24fps. Sometimes it's even drawn at, or computers are used to generate, 30fps. If you have any of that in your cartoons (I wouldn't think so for these old cartoons, unless it's the end credits), then it's guaranteed to be partly jerky, as you're removing unique frames needed for smooth playback. Without seeing samples taken from the source, I have no idea what you have. With experience you'll be able to figure it out yourself.

[bloomcounty]

Okay, so I converted all the episodes, and while half of them came out 512x384, the rest came out in various ARs lower than that. I did notice that in those with lower ARs, it doesn't say exactly the same thing about the hybrid stuff, but occasionally says, "Source is considered to be interlaced" -- also, sometimes with the various files (regardless of the AR) it'll say hybrid, but either mostly NTSC or mostly FILM (it varies).

Questions:

1. Why does it vary so much?

2. The file that came out with the lowest AR, 400x304, was the one that said, "Source is considered to be interlaced." So what I did was, I re-ran it in AutoGK and increased the target file size to 220 instead of 180 (doing a little math to make that guess). And it came out this time with an AR of 448x336. (I figure I'll need to increase the target size to about 273MB to get it at 512x384.) However, I'm not positive there's really a picture quality difference between the 400x304 (180mb) version and the 448x336 (220mb) version. But it's hard to tell on my laptop. Would there have to be a difference? Would the 220mb version definitely be a better picture quality?

Here's the logs:

400x304 (180mb)

[2/14/2007 11:51:07 PM] AutoGK 2.40
[2/14/2007 11:51:07 PM] OS: WinXP (5.1.2600).2
[2/14/2007 11:51:07 PM] Job started.
[2/14/2007 11:51:07 PM] Input file: C:\DVD Rips\Animaniacs\22 (21_16)\VTS_01_0.IFO
[2/14/2007 11:51:07 PM] Output file: C:\DVD Rips\Animaniacs\Encodes\Anim22.avi
[2/14/2007 11:51:07 PM] Output codec: XviD
[2/14/2007 11:51:07 PM] Audio 1: English AC3 2ch
[2/14/2007 11:51:07 PM] Subtitles: none
[2/14/2007 11:51:07 PM] Format: AVI
[2/14/2007 11:51:07 PM] Target size: 180Mb
[2/14/2007 11:51:07 PM] Audio 1 settings: Auto
[2/14/2007 11:51:07 PM] Standalone support enabled: ESS
[2/14/2007 11:51:07 PM] Started encoding.
[2/14/2007 11:51:07 PM] Demuxing and indexing.
[2/14/2007 11:51:43 PM] Processing file: C:\DVD Rips\Animaniacs\22 (21_16)\VTS_01_1.VOB
[2/14/2007 11:51:43 PM] Source resolution: 720x480
[2/14/2007 11:51:43 PM] Found NTSC source.
[2/14/2007 11:51:43 PM] Source aspect ratio: 4:3
[2/14/2007 11:51:43 PM] Color correction enabled.
[2/14/2007 11:51:43 PM] Analyzing source.
[2/14/2007 11:52:40 PM] Source has percentage of interlacing in motion areas: 83.71
[2/14/2007 11:52:40 PM] Source has percentage of telecined patterns: 3.37
[2/14/2007 11:52:40 PM] Source has percentage of progressive patterns: 0.79
[2/14/2007 11:52:40 PM] Source has percentage of interlaced patterns: 95.84
[2/14/2007 11:52:40 PM] Source is considered to be interlaced.
[2/14/2007 11:52:40 PM] Output will contain 38273 frames
[2/14/2007 11:52:40 PM] Decoding audio.
[2/14/2007 11:53:06 PM] Normalizing audio.
[2/14/2007 11:53:25 PM] Encoding audio.
[2/14/2007 11:55:44 PM] Using cartoon mode.
[2/14/2007 11:55:44 PM] Audio1 size: 18,392,688 bytes (17.54 Mb)
[2/14/2007 11:55:44 PM] Overhead: 1,530,880 bytes (1.46 Mb)
[2/14/2007 11:55:44 PM] Video size: 168,820,112 bytes (161.00 Mb)
[2/14/2007 11:55:44 PM] Running compressibility test.
[2/14/2007 11:55:44 PM] Writing the following script to C:\DVD Rips\Animaniacs\Encodes\agk_tmp\Anim22_comptest.avs
===========================================================
LoadPlugin("C:\PROGRA~1\AutoGK\DGMPGDec\DGDecode.dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\autocrop.dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\ColorMatrix.dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\RemoveGrainSSE3.dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\leakkerneldeint.dll")

movie = mpeg2source("C:\DVD Rips\Animaniacs\Encodes\agk_tmp\Anim22.d2v")
cropclip = autocrop(movie,mode=0,wmultof=4,hmultof=4,samples=10,aspect=0,threshold=34,samplestartframe=0,leftadd=0,rightadd=0,topadd=0,bottomadd=0)
fixed_aspect = 0.888888888888889
c_width = width(cropclip)
c_height = round(height(cropclip) / fixed_aspect)
input_par = float(c_width)/float(c_height)
input_par = (input_par > 1.4) || (input_par < 1.25) ? input_par : (4.0/3.0)
out_width = 640
out_height = round(float(out_width) / input_par)
hmod = out_height - (floor(out_height / 16 ) * 16)
out_height = (hmod > 4) ? (out_height + (16 - hmod)) : (out_height - hmod)
new_aspect = (float(out_width) / float(out_height)) / fixed_aspect
LeakKernelDeInt(movie,order=1,sharp=true)
autocrop(mode=0,wmultof=4,hmultof=4,samples=10,aspect=new_aspect,threshold=34,samplestartframe=0,leftadd=0,rightadd=0,topadd=0,bottomadd=0)
LanczosResize(out_width,out_height)
RemoveGrain(mode=2)
ColorMatrix("Rec.709->Rec.601",opt=0,hints=false)
SelectRangeEvery(250,15)
===========================================================
[2/14/2007 11:58:02 PM] Duration was: 2 minutes 17 seconds
[2/14/2007 11:58:02 PM] Speed was: 16.66 fps.
[2/14/2007 11:58:02 PM] Compressibility percentage is: 20.86
[2/14/2007 11:58:02 PM] Using softer resizer.
[2/14/2007 11:58:02 PM] Chosen resolution is: 400x304 ( AR: 1.32 )
[2/14/2007 11:58:02 PM] Predicted comptest value is: 55.66%
[2/14/2007 11:58:02 PM] Running first pass.
[2/14/2007 11:58:02 PM] Writing the following script to C:\DVD Rips\Animaniacs\Encodes\agk_tmp\Anim22_movie.avs
===========================================================
LoadPlugin("C:\PROGRA~1\AutoGK\DGMPGDec\DGDecode.dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\autocrop.dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\ColorMatrix.dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\RemoveGrainSSE3.dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\leakkerneldeint.dll")

movie = mpeg2source("C:\DVD Rips\Animaniacs\Encodes\agk_tmp\Anim22.d2v")
cropclip = autocrop(movie,mode=0,wmultof=4,hmultof=4,samples=10,aspect=0,threshold=34,samplestartframe=0,leftadd=0,rightadd=0,topadd=0,bottomadd=0)
fixed_aspect = 0.888888888888889
c_width = width(cropclip)
c_height = round(height(cropclip) / fixed_aspect)
input_par = float(c_width)/float(c_height)
input_par = (input_par > 1.4) || (input_par < 1.25) ? input_par : (4.0/3.0)
out_width = 400
out_height = round(float(out_width) / input_par)
hmod = out_height - (floor(out_height / 16 ) * 16)
out_height = (hmod > 4) ? (out_height + (16 - hmod)) : (out_height - hmod)
new_aspect = (float(out_width) / float(out_height)) / fixed_aspect
LeakKernelDeInt(movie,order=1,sharp=true)
autocrop(mode=0,wmultof=4,hmultof=4,samples=10,aspect=new_aspect,threshold=34,samplestartframe=0,leftadd=0,rightadd=0,topadd=0,bottomadd=0)
BicubicResize(out_width,out_height,0,0.5)
RemoveGrain(mode=2)
ColorMatrix("Rec.709->Rec.601",opt=0,hints=false)
===========================================================
[2/15/2007 12:09:54 AM] Duration was: 11 minutes 51 seconds
[2/15/2007 12:09:54 AM] Speed was: 53.78 fps.
[2/15/2007 12:09:54 AM] Expected quality of first pass size: 53.09%
[2/15/2007 12:09:54 AM] Running second pass.
[2/15/2007 12:25:58 AM] Duration was: 16 minutes 3 seconds
[2/15/2007 12:25:58 AM] Speed was: 39.71 fps.
[2/15/2007 12:25:58 AM] Job finished. Total time: 34 minutes 50 seconds

And here's the 448x336 (220mb) version

[2/15/2007 10:36:53 AM] AutoGK 2.40
[2/15/2007 10:36:53 AM] OS: WinXP (5.1.2600).2
[2/15/2007 10:36:53 AM] Job started.
[2/15/2007 10:36:53 AM] Input file: C:\DVD Rips\Animaniacs\22 (21_16)\VTS_01_0.IFO
[2/15/2007 10:36:53 AM] Output file: C:\DVD Rips\Animaniacs\Encodes\Anim22a.avi
[2/15/2007 10:36:53 AM] Output codec: XviD
[2/15/2007 10:36:53 AM] Audio 1: English AC3 2ch
[2/15/2007 10:36:53 AM] Subtitles: none
[2/15/2007 10:36:53 AM] Format: AVI
[2/15/2007 10:36:53 AM] Target size: 220Mb
[2/15/2007 10:36:53 AM] Audio 1 settings: Auto
[2/15/2007 10:36:53 AM] Standalone support enabled: ESS
[2/15/2007 10:36:53 AM] Started encoding.
[2/15/2007 10:36:53 AM] Demuxing and indexing.
[2/15/2007 10:37:32 AM] Processing file: C:\DVD Rips\Animaniacs\22 (21_16)\VTS_01_1.VOB
[2/15/2007 10:37:32 AM] Source resolution: 720x480
[2/15/2007 10:37:32 AM] Found NTSC source.
[2/15/2007 10:37:32 AM] Source aspect ratio: 4:3
[2/15/2007 10:37:32 AM] Color correction enabled.
[2/15/2007 10:37:32 AM] Analyzing source.
[2/15/2007 10:38:30 AM] Source has percentage of interlacing in motion areas: 83.71
[2/15/2007 10:38:30 AM] Source has percentage of telecined patterns: 3.37
[2/15/2007 10:38:30 AM] Source has percentage of progressive patterns: 0.79
[2/15/2007 10:38:30 AM] Source has percentage of interlaced patterns: 95.84
[2/15/2007 10:38:30 AM] Source is considered to be interlaced.
[2/15/2007 10:38:30 AM] Output will contain 38273 frames
[2/15/2007 10:38:30 AM] Decoding audio.
[2/15/2007 10:38:57 AM] Normalizing audio.
[2/15/2007 10:39:35 AM] Encoding audio.
[2/15/2007 10:41:56 AM] Using cartoon mode.
[2/15/2007 10:41:56 AM] Audio1 size: 18,392,688 bytes (17.54 Mb)
[2/15/2007 10:41:56 AM] Overhead: 1,530,880 bytes (1.46 Mb)
[2/15/2007 10:41:56 AM] Video size: 210,763,152 bytes (201.00 Mb)
[2/15/2007 10:41:56 AM] Running compressibility test.
[2/15/2007 10:41:56 AM] Writing the following script to C:\DVD Rips\Animaniacs\Encodes\agk_tmp\Anim22a_comptest.avs
===========================================================
LoadPlugin("C:\PROGRA~1\AutoGK\DGMPGDec\DGDecode.dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\autocrop.dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\ColorMatrix.dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\RemoveGrainSSE3.dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\leakkerneldeint.dll")

movie = mpeg2source("C:\DVD Rips\Animaniacs\Encodes\agk_tmp\Anim22a.d2v")
cropclip = autocrop(movie,mode=0,wmultof=4,hmultof=4,samples=10,aspect=0,threshold=34,samplestartframe=0,leftadd=0,rightadd=0,topadd=0,bottomadd=0)
fixed_aspect = 0.888888888888889
c_width = width(cropclip)
c_height = round(height(cropclip) / fixed_aspect)
input_par = float(c_width)/float(c_height)
input_par = (input_par > 1.4) || (input_par < 1.25) ? input_par : (4.0/3.0)
out_width = 640
out_height = round(float(out_width) / input_par)
hmod = out_height - (floor(out_height / 16 ) * 16)
out_height = (hmod > 4) ? (out_height + (16 - hmod)) : (out_height - hmod)
new_aspect = (float(out_width) / float(out_height)) / fixed_aspect
LeakKernelDeInt(movie,order=1,sharp=true)
autocrop(mode=0,wmultof=4,hmultof=4,samples=10,aspect=new_aspect,threshold=34,samplestartframe=0,leftadd=0,rightadd=0,topadd=0,bottomadd=0)
LanczosResize(out_width,out_height)
RemoveGrain(mode=2)
ColorMatrix("Rec.709->Rec.601",opt=0,hints=false)
SelectRangeEvery(250,15)
===========================================================
[2/15/2007 10:44:25 AM] Duration was: 2 minutes 28 seconds
[2/15/2007 10:44:25 AM] Speed was: 15.51 fps.
[2/15/2007 10:44:25 AM] Compressibility percentage is: 26.05
[2/15/2007 10:44:25 AM] Using softer resizer.
[2/15/2007 10:44:25 AM] Chosen resolution is: 448x336 ( AR: 1.33 )
[2/15/2007 10:44:25 AM] Predicted comptest value is: 56.13%
[2/15/2007 10:44:25 AM] Running first pass.
[2/15/2007 10:44:25 AM] Writing the following script to C:\DVD Rips\Animaniacs\Encodes\agk_tmp\Anim22a_movie.avs
===========================================================
LoadPlugin("C:\PROGRA~1\AutoGK\DGMPGDec\DGDecode.dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\autocrop.dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\ColorMatrix.dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\RemoveGrainSSE3.dll")
LoadPlugin("C:\PROGRA~1\AutoGK\filters\leakkerneldeint.dll")

movie = mpeg2source("C:\DVD Rips\Animaniacs\Encodes\agk_tmp\Anim22a.d2v")
cropclip = autocrop(movie,mode=0,wmultof=4,hmultof=4,samples=10,aspect=0,threshold=34,samplestartframe=0,leftadd=0,rightadd=0,topadd=0,bottomadd=0)
fixed_aspect = 0.888888888888889
c_width = width(cropclip)
c_height = round(height(cropclip) / fixed_aspect)
input_par = float(c_width)/float(c_height)
input_par = (input_par > 1.4) || (input_par < 1.25) ? input_par : (4.0/3.0)
out_width = 448
out_height = round(float(out_width) / input_par)
hmod = out_height - (floor(out_height / 16 ) * 16)
out_height = (hmod > 4) ? (out_height + (16 - hmod)) : (out_height - hmod)
new_aspect = (float(out_width) / float(out_height)) / fixed_aspect
LeakKernelDeInt(movie,order=1,sharp=true)
autocrop(mode=0,wmultof=4,hmultof=4,samples=10,aspect=new_aspect,threshold=34,samplestartframe=0,leftadd=0,rightadd=0,topadd=0,bottomadd=0)
BicubicResize(out_width,out_height,0,0.5)
RemoveGrain(mode=2)
ColorMatrix("Rec.709->Rec.601",opt=0,hints=false)
===========================================================
[2/15/2007 10:57:38 AM] Duration was: 13 minutes 12 seconds
[2/15/2007 10:57:38 AM] Speed was: 48.28 fps.
[2/15/2007 10:57:39 AM] Expected quality of first pass size: 54.25%
[2/15/2007 10:57:39 AM] Running second pass.
[2/15/2007 11:16:03 AM] Duration was: 18 minutes 24 seconds
[2/15/2007 11:16:03 AM] Speed was: 34.66 fps.
[2/15/2007 11:16:03 AM] Job finished. Total time: 39 minutes 10 seconds

Thoughts? Thanks for the help!

[manono]

Hi-

1. I have no idea, except it does mean that I was wrong when I said you can handle them all the same way as you handled the first one, by forcing an IVTC. It looks like they really do have different amounts of interlacing. Animation by it's very nature isn't interlaced. Having as much interlacing as the 2 examples you showed is consistent with a bad PAL2NTSC conversion. Or maybe they're just noisier and grainier than the others, and the analysis mistakenly diagnoses all the grain as interlacing. But I'd need a sample to be sure.

As for getting different resolutions, it means different episodes compress differently, and there may be something fundamentally different about the ones that got the lower resolutions and had the higher amounts of interlacing, as compared to the 512x384 hybrid mostly film ones.

So, if you'd like to upload a small sample somewhere, I and maybe others will be happy to have a look.

2. Not necessarily. They come out with a similar quality (as measured by the percentage), so they probably look about the same.

[bloomcounty]

1. I have no idea, except it does mean that I was wrong when I said you can handle them all the same way as you handled the first one, by forcing an IVTC. It looks like they really do have different amounts of interlacing. Animation by it's very nature isn't interlaced.

But the IVTC option says it does that only if it's detected as a hybrid, right? And the one in question that came out with the smallest resolution didn't say hybrid, it said, "Source is considered to be interlaced". I checked, and that's the ONLY one what says that. The rest either say "hybrid (mostly NTSC)" and "hybrid (mostly FILM)". (The two examples I showed were the same file, only done at a two different target sizes.)

So half of the hybrid ones (either mostly FILM or mostly NTSC) came out as 512x384 when set to a target size of 180mb.

The other half (minus one) came out to various sizes: 464x352, 496x368, 448x336, 480x386, and 432x320. Within the logs for those, the "percentage of interlaced patterns" (if that's important) varies all over the place.

Then that one 400 one said "interlaced" and not hybrid at all.

So, considering all that, is there a different way I should do each of these? Or just to the "interlaced" one differently (minus the IVTC option -- although, like I said, I don't think it actually did that option, even though it was checked, because it wasn't considered "hybrid", so having that set probably doesn't matter...? I'm pretty sure the two logs I posted don't how it as having used that option, even though it was set to.).

I don't have the means to take a snippet and upload, unfortunately.

2. Not necessarily. They come out with a similar quality (as measured by the percentage), so they probably look about the same.

Even though one file is a lot bigger in size than the other? (180 vs. 220)

So is there no point/reason to go back and re-encode the ones that aren't 512 by increasing the target size until they are 512? I won't get a better picture quality?

Thanks!

[manono]

Like I said, the only way for me to be able to offer some better advice would be to have a look at the real thing. Unless you're on dial-up, I don't see why you can't upload a 10-15 MB piece somewhere, like:

http://www.badongo.com/
or http://rapidshare.de/

I'm tired of guessing what they're like and how best to treat them.

Even though one file is a lot bigger in size than the other? (180 vs. 220)

The size doesn't have much to do with it, as different sources compress differently. The resolution could have something to do with it, but if you make them all 512x384 with the same file size, some of them will have so much artifacting they would have been better off at the lower resolution that AutoGK chose for you. If you increase the file sizes so that they're all 512x384 with roughly the same percentage quality, then no matter the resulting file sizes, they should have roughly the same quality and appear generally the same.

[bloomcounty]

Like I said, the only way for me to be able to offer some better advice would be to have a look at the real thing. Unless you're on dial-up

I'm on dial-up.

The size doesn't have much to do with it, as different sources compress differently. The resolution could have something to do with it, but if you make them all 512x384 with the same file size, some of them will have so much artifacting they would have been better off at the lower resolution that AutoGK chose for you. If you increase the file sizes so that they're all 512x384 with roughly the same percentage quality, then no matter the resulting file sizes, they should have roughly the same quality and appear generally the same.

You said size doesn't matter, then at the end you said increasing the file size will make them have the same quality. I'm not talking about changing the resolution and keeping the 180 file size. I'm talking about just increasing the target file size until the resulting encode has a resolution of 512x384. So...

1. The two logs I posted are for the SAME file. One was done like the rest (target size 180), and came out 400x304. The second was redone, increasing the target size to 220, resulting in an AR of 448x336. I would increase it again, probably to about 273mb, which should give me an AR of 512x384 for THIS file. So, will that file with the 273mb size to get to 512 AR (and, thus, any file I increase the target size to get to 512 AR) have a better picture quality than the original encode with the smaller AR and target size of 180mb?

2. I see that the "Expected quality of first pass size" for each is pretty close to the same, even though I increased the target file size of the second. What would this be?

2a. And is the higher percentage better?

3. Or will I not see a better image quality if I increase the target sizes to whatever it takes to get the 512 AR? Meaning, is it not worthing re-encoding to do this?

Just forget about guessing what they're like. I don't think you need to in order to address these questions above. Thanks again!

[manono]

This will be my last post on the subject. I can't say anything intelligent without a sample of the source to go on.

You said size doesn't matter, then at the end you said increasing the file size will make them have the same quality. I'm not talking about changing the resolution and keeping the 180 file size. I'm talking about just increasing the target file size until the resulting encode has a resolution of 512x384.

When I said file size doesn't matter, that was in response to your question about 220 MB being better looking than 180 MB at a lower resolution. Didn't you yourself say that you couldn't tell any difference between the 2? If they're both from the same episode, and if the resulting quality is similar, in terms of the compression percentage, or in terms of the average quant, then the higher resolution one should be slightly better quality, be slightly sharper, or hold slightly more detail. So, you're right about that. Among different episodes, though, file size alone is no indication of relative quality.

By the way, if you really want to make them all the same resolution at a certain quality, then choose a 512 width in the Advanced Settings, and a Target Quality in percentage. The default is 75%, which may give you larger file sizes than you might like. If you experiment with 50, 60, and 70%, you might find one that you find suitable for both file size and quality, and then you can do them all that way. The added benefit is that you'll be running only 1 pass, and the encoding will go much more quickly. Doing it that way, you might be surprised by the differences in file sizes you get for the different episodes, because for some reason they seem to compress very differently, some of them. That will save you from slowly upping the file size until AutoGK finally gives you the 512x384 you want.