TMPGEnc templates

[nobspangle]

I'm a fairly new user of TMPGEnc so forgive me if I have entirely missed a few points.
I thought I'd download a few of the templates listed on this site to see if the vast experience of their creators made a great difference to the quality of the encoded video.
Firstly I downloaded KingViper's template who promises, "If you have divx movies and they fit on a cd, more than likely the will fit on a cd in SVCD format." I tried it out and quite often you can fit one whole movie on to a CD but the quality is somewhat crap.
Examination of the settings showed that kingviper had acheived his amazing compression by dropping the bit rate to a CQ 50 and the audio to 128k genius, how did he think of it, it must of taken hours of trial and error and months of lengthy calculation to discover that dropping the bitrate makes the file size smaller.

I thought to myself surely not all these templates can be like this. So I moved on to the "TIbrO VideoCD" template. Tibro says, "I have made two templates, a PAL version and an NTSC version, which give quality that rivals that of Panasonic MPEG Encoder. There are NO MACROBLOCKS and everything looks great"
I thought fantastic a high quality tweeking of TMPGEnc settings to create something truly noticeable. I encoded some video (A DV capture file-served from Premire) and encoded the video again using the default VCD template in TMPGEnc and I was amazed, the two pieces of video looked identical. Wow, I thought, if the video looks identical maybe the template is identical. I went through each setting one by one and found that as far as I could tell the template I had downloaded was identical to the one included with TMPGEnc apart from some settings being grayed out on the bundled template.

So I moved on to Kwag's templates, Kwag says, "352x480 NTSC ( 352x576 PAL ) templates reflecting an optimized GOP and CQ, balanced for quality and size, that enable most 120 minute movies to fit on a 80 minute CD-R."
Testing the template I found the quality on high action scenes to be pretty good but this gave huge filesizes (more than 10Mb a minute). On low action, easy to encode, scenes file sizes went down and so did the quality (right down) The colour gradients that give images depth vanished to be replaced by large blocks of static colour. I investigated Kwag's template settings and saw he had used a CQ_VBR I changed this to a standard two pass VBR and was able to acheive the same or better quality on all scenes.

CQ is great for creating files of the smallest size but the purpose of Kwag's coded is to have the files burnt to CD. If you know how much space you have on the CD why not use the maxium filesize possible giving you a better quality result. If I encode a 100 minute film with Kwag's template and it ends up as 700Mb that's almost 100Mb of wasted space on my CD if I use a two pass VBR I can define the average bitrate and capitalise on this extra 100Mb of file size this represents a 16% increase in file size and therefore average bitrate. For a 100 minute video to compress to 700Mb with a 128Kb audio bitrate would require and average bitrate of 828 kbits/s. If I use two pass instead of CQ I could use an average bitrate of 954 kbits/s giving not a vast increase in quality (due to the non-linearity of the relationship between bitrate and quality) but a noticeable one. However Kwag's template uses a 128k audio bitrate which everyone will admit sounds crap. Using two pass VBR I could bump the bitrate of the audio up to 192 or even 224 kbits/s and still maintain a higher average video bitrate.

[brushswinger]

me really cant imagine any way of getting a full movie (90 mins or more) on a single cdr! And with the price as low as they am now what is the prob? Me often put 78 mins on one disc, then 10 mins on second disc. Bit of a niusance changing discs me suppose, but that is what wives are for! me say, "change disc bitch!" and she changes it for me. Unless the film was crap and she says "Change it yaself *******!"

[Trailryder]

I have been using Kwags template. I modify it a little by setting the output to 352x240, use high quality mode, 74-76 CQ on a 1.5 hour movie, soften block noise box checked, audio at 192, and consistently get conversions of 5-6 mb per minute. Fits on 1 cd and the quality looks great using DVD source. Used this template for converting avi too with good results.

[The_Daywalker]

Personally Sefys templates are the best quaility i have seen for vcd but vcd sucks SVCD is the way to go personally i wont go under a VBR of 2320 SVCD gives about 43-48mins of movie wth no blocks and i can say absolutely no blocks no pixelization no washed out colours and Jump on me all you want does rival DVD VHS source material not dvd quality but 98% is good enough for me VHS is the same ANIME Looks fantastic out put is near identical to input only difference is sound as i use 160kbs it sound a little tinny but i always edit sound in a wave editor and convert seperataly resulting in crisp clear stereo sound give SVCD a go IT takes a bit to get it right but it is so much better quality wise most movies fit on to 2 cds long ones if you need three cds thats not so bad try sticking gladiator on to 1cd VCD rip using dvd source then comparing it to a 2-3cd svcd rip using dvd source you cant even begin to compare quality even using a 15 year old VHS source is better than 1cd dvd rips

[kwag]

I'm a fairly new user of TMPGEnc so forgive me if I have entirely missed a few points.
I thought I'd download a few of the templates listed on this site to see if the vast experience of their creators made a great difference to the quality of the encoded video.
Firstly I downloaded KingViper's template who promises, "If you have divx movies and they fit on a cd, more than likely the will fit on a cd in SVCD format." I tried it out and quite often you can fit one whole movie on to a CD but the quality is somewhat crap.
Examination of the settings showed that kingviper had acheived his amazing compression by dropping the bit rate to a CQ 50 and the audio to 128k genius, how did he think of it, it must of taken hours of trial and error and months of lengthy calculation to discover that dropping the bitrate makes the file size smaller.

I thought to myself surely not all these templates can be like this. So I moved on to the "TIbrO VideoCD" template. Tibro says, "I have made two templates, a PAL version and an NTSC version, which give quality that rivals that of Panasonic MPEG Encoder. There are NO MACROBLOCKS and everything looks great"
I thought fantastic a high quality tweeking of TMPGEnc settings to create something truly noticeable. I encoded some video (A DV capture file-served from Premire) and encoded the video again using the default VCD template in TMPGEnc and I was amazed, the two pieces of video looked identical. Wow, I thought, if the video looks identical maybe the template is identical. I went through each setting one by one and found that as far as I could tell the template I had downloaded was identical to the one included with TMPGEnc apart from some settings being grayed out on the bundled template.

So I moved on to Kwag's templates, Kwag says, "352x480 NTSC ( 352x576 PAL ) templates reflecting an optimized GOP and CQ, balanced for quality and size, that enable most 120 minute movies to fit on a 80 minute CD-R."
Testing the template I found the quality on high action scenes to be pretty good but this gave huge filesizes (more than 10Mb a minute). On low action, easy to encode, scenes file sizes went down and so did the quality (right down) The colour gradients that give images depth vanished to be replaced by large blocks of static colour. I investigated Kwag's template settings and saw he had used a CQ_VBR I changed this to a standard two pass VBR and was able to acheive the same or better quality on all scenes.

CQ is great for creating files of the smallest size but the purpose of Kwag's coded is to have the files burnt to CD. If you know how much space you have on the CD why not use the maxium filesize possible giving you a better quality result. If I encode a 100 minute film with Kwag's template and it ends up as 700Mb that's almost 100Mb of wasted space on my CD if I use a two pass VBR I can define the average bitrate and capitalise on this extra 100Mb of file size this represents a 16% increase in file size and therefore average bitrate. For a 100 minute video to compress to 700Mb with a 128Kb audio bitrate would require and average bitrate of 828 kbits/s. If I use two pass instead of CQ I could use an average bitrate of 954 kbits/s giving not a vast increase in quality (due to the non-linearity of the relationship between bitrate and quality) but a noticeable one. However Kwag's template uses a 128k audio bitrate which everyone will admit sounds crap. Using two pass VBR I could bump the bitrate of the audio up to 192 or even 224 kbits/s and still maintain a higher average video bitrate.

Hi nobspangle:

Try a 90 minute movie or longer with 2-pass VBR and try it again with the template at CQ=74.

2-pass VBR does a great job with shorter movies. With longer than about 90 minutes the bit rate is well distributed but the quality starts to degrade.

You'll find that CQ does a better job than 2-pass VBR. I've been though this before. ( too many times, and I substain my facts by many trials ).

The quality of TMPEG, not the template, produces a better result with CQ than VBR.

The only advantage of X-pass VBR is that you can predict the final size very closely.

I also tested against CCE encoder, and there's almost no quality gain in CCE 3-pass to TMPEG 2-pass and encoding times are about the same.

This of course is with TMPGEnc 2.53.

For 2-pass, CCE is much faster, but the flexibility offered from TMPEG with all the tools ( cut, merge, deinterlace, etc etc, ) CCE is not worth it.

I have done a movie with TMPEG with the 2-pass VBR and the file size was 800MB and the same movie with CQ and file size was 700MB.

The result was the CQ was better looking than the 2-pass VBR.

I'll give you a very good example. Try to make the movie "Mission to Mars" with 2-pass and also with the KVCD template at CQ=74 and you'll see what I mean.
The file size will be almost the same. Around 815MB for both.
This is a 2+ hour movie. When you play it back, you'll see the CQ is by FAR better than the one produced with 2-pass VBR.

This has been a practical example, and the results say everything.
When I say it looks better, I mean looking at the burned CD-R's in a 32" HDTV.

The difference is very visible.

kwag

[y2flyy]

trust me nob you would not be complaining about templates if you tried
sefy's template
i use his conveted sevcd enhanced template
you can fit almost 2hrs on one cd in vcd
im telling you its the best template
and to add better quality w/o a file increase and/or encode increase
use the sharpen edge filter in tmpgenc's advanced tab
make sure both sliders are all the way to the right and enable filter
block is CHECKED and the other block is UNCHECKED
report back with results please
Y2Flyy

[nobspangle]

If the ceiling is the same there is no way CQ can produce a better quality than 2-pass VBR without increasing the file size. CQ examines sections of video and detemines how much compression it can apply to acheive the desired quality setting. VBR works by examining the entire video to see what the maximum acheivable bitrate is. Whilst keeping to the average and maximum settings.

The only advantage of X-pass VBR is that you can predict the final size very closely.

This is one of the advantages and is a very important one if you intend on transfering your videos to CD what is the point in sacrificing quality by not filling the CD in its entirety?

If you were to take a film and encode it using CQ and found the filesize to be 800Mb then you calculated the average bitrate and encoded the same film using VBR set to that average bitrate you will find that there is most likely none or very little difference between the two.

CQ has two advantages over 2-pass VBR and quality isn't one of them. CQ is useful because it is faster than VBR (much faster especially on long pieces of video like an entire film). Also CQ is useful for creating files of a given quality that are as small as possible. This is extreamly useful when creating video for multimedia applications when larger filesize=more data=more money.

trust me nob you would not be complaining about templates if you tried
sefy's template
i use his conveted sevcd enhanced template
you can fit almost 2hrs on one cd in vcd
im telling you its the best template

Sounds like an intelligent comment, my point on templates is this,
The template function in TMPGEnc is not designed as way of tweeking TMPGEnc to create better quality files, any amount of this kind of tweeking will only ever produce minimal results. The main improvments in quality in any of the templates I have seen are created by changing the CBR to either CQ or 2-pass VBR, this makes the files non-standard for VCD. This brings me neatly to the point in templates, standards. The author provides us with templates for the VCD standard, the SVCD standard and the DVD standard. If I was making videos for inclusion on the ends of pop albums I might create my own standard, certain resolution, bitrate etc. I could then save my standard as a template so that I could recall it and use it each time I performed this task.

I would be using the template not because it gave a better quality result than the standard templates but because I need the video I was encoding to be different in some way but I needed to use the same settings over and over again.

The template I created would most probably be of no use to anyone else since they are not trying to perform the same task as me. Templates are a useful feature of Microsoft office, but you don't see forums trading the latest letter heading for word.

[kwag]

Hi nobspangle:

Read more about CQ here, and then come back and comment.


http://tangentsoft.net/video/mpeg/enc-modes.html

kwag

[nobspangle]

First can I say Kwag, you go to bed very late and get up very early, you should get more sleep.

Read more about CQ here, and then come back and comment.

Cheers Kwag this was most informative I'll have to read it many more times before I can make a final comment (it takes me a long time to take in information) but I can see that about x-pass VBR the author does say,
"This asymptotically approaches the best possible video stream that the encoder is capable of."

For the benefit of everyone else reading this thread who is too lazy to click Kwag's link and read the info for themselves I'll quote the portion about CQ

CQ can stand for "constant quality", or "constant quantization". The effect is the same, though they come at the same issue from different sides. The quantization (Q) level of an MPEG stream is a measure of the amount of data the encoder threw away to bring the bit rate down. Higher Q values mean the encoder threw away a lot of data, implying a low bit rate, but also low visual quality. A low Q level implies a high bit rate, and high visual fidelity. Coming at the issue from the other direction, setting a "constant quality" implies a particular MPEG Q level, with higher quality numbers implying lower Q numbers. Usually an encoder that expresses the issue as one of setting the quality doesn't show you the resulting MPEG quantization (Q) level. Because every encoder is different, there is no standard mapping from one "quality level" to an MPEG Q level. Indeed, the Q level for a given quality level might well depend on the video characteristics as much as the encoder.

(By the way, in MPEG, if you see Q written alone, it always means quantization, and not quality. Some would therefore say it's confusing to use "CQ" to mean "constant quality." CQ as shorthand for "constant quantization" has a very long history. CQ as shorthand for "constant quality" is a more recent coinage.)

In regular VBR modes, the quality level is varied continuously to keep to a desired bit rate. CQ modes let you pick a desired quality/Q level, and let the bit rate vary to compensate. This means that you cannot predict the bitrate output from a plain CQ mode: the complexity of the video, the encoder's settings, the encoder's particular algorithm, and the pre-filtering you've done on the video all affect the output bit rate. You can't possibly take all this into account and make an accurate guess at the output bit rate when using a CQ mode.

If quantization level is a measure of how much data is thrown away then it is directly related to bitrate and therefore a constant quantization level would imply a constant bitrate. This leads me to belive that somewhere along the line the author of this article has become confused.

If, as is written, regular VBR modes vary the quality level to maintain a desired bitrate how are they variable bitrate modes? Surely if you are maintaining anything at one level then it is at a constant level, in this case a constant bitrate.

The quantization level can not be described as a measure of the amount of data thrown away, but is more acurately described as measure of the quality of the video.

As the samples on your website show, maintaining a constant quality on high action scenes results in high bitrates and larger filesizes. Maintaining constant quality on low action scenes requires lower bitrates and therefore lower filesizes. If I encode the sames scenes using 2-pass VBR and aim for the same filesizes the quality will be the same because the quantization will be the same.

One thing I noticed is that nowhere in this article, or as far as I have noticed elsewhere on the site, does it say that CQ is better than 2-pass on a file of the same size.

2-pass VBR on TMPGEnc asks for an average bitrate to be entered (ie the average is not mearly the arithmetic mean of the two extremes) this means you can use 0 bits/s as the minium rate. Doing this allows the encoder a great deal of freedom in decideing the bitrate used for a paticualr sequence.

I am currently a bit snowed under with work but ASAP I will mock up some samples using test videos (not bits of film) to show how file sizes and quality are varied with bitrate, encoding modes and source material.

[kwag]

Hi nobspangle:

You left the most important part out!
The conclusion:

Conclusion
VBR modes are worth using, when you're not forced to use CBR. And when you can get away with it, pure CQ modes are the best VBR types of all. You have to work harder to get good video, but you can achieve absolutely stunning results with a properly-tuned CQ encoder.

kwag

[Heywould3]

I jsut wanted to make a comment about the teplate settings. I tried all kinds of them and found that kwags is by far the best, in order to get the file size to be exactly 833megs you will need to mess with the CQ setting a bit but its absolutly great. only work i have to do which is by choice not nessisity SP is to use bit rate calc to find the frame rate and resolution taht the original file ( i re-encode screeners) is in to adjust them in tmpg so that the movie dont look streached or squished. BUT i was able to take panic room which is a 2 hour movie and compress it down to 848 megs and the quality is so damn close to the original that you would not be able to tell teh difference between the two if i showed you. For the record before i get FLAMED a mpeg file that is 833 megs in winexp will burn at 703 megs which is the size of a 80 min cdr. and i over burn to 720 megs all the time with no problem. i have 60+ screeners and all of them are great quality and all of them will NOW fit on 1 80 min CDR. i didnot believe it myself but its TRUE i have done it and it looks the exact same or very close.

I have samples if you all dont believe it BUT i dont have web space to post and the files are too large to email :)

its taken me a long time to learn this stuff enough to be able to create good quality movies (VCD) that will fit on 1 CDR but it was well worth it.

thanks to people like KWAG that gave me a place ( template) to start from

If anyone wants or need proof email me and we'll talk

[kwag]

I jsut wanted to make a comment about the teplate settings. I tried all kinds of them and found that kwags is by far the best, in order to get the file size to be exactly 833megs you will need to mess with the CQ setting a bit but its absolutly great. only work i have to do which is by choice not nessisity SP is to use bit rate calc to find the frame rate and resolution taht the original file ( i re-encode screeners) is in to adjust them in tmpg so that the movie dont look streached or squished. BUT i was able to take panic room which is a 2 hour movie and compress it down to 848 megs and the quality is so damn close to the original that you would not be able to tell teh difference between the two if i showed you. For the record before i get FLAMED a mpeg file that is 833 megs in winexp will burn at 703 megs which is the size of a 80 min cdr. and i over burn to 720 megs all the time with no problem. i have 60+ screeners and all of them are great quality and all of them will NOW fit on 1 80 min CDR. i didnot believe it myself but its TRUE i have done it and it looks the exact same or very close.

I have samples if you all dont believe it BUT i dont have web space to post and the files are too large to email

its taken me a long time to learn this stuff enough to be able to create good quality movies (VCD) that will fit on 1 CDR but it was well worth it.

thanks to people like KWAG that gave me a place ( template) to start from

If anyone wants or need proof email me and we'll talk

8) Hey man, those are the comments that make my day! :P

Thanks,
kwag

[bilbogod]

It hardly needs to be reiterated, but actions always speak louder than words 60,000 + topic views is testament alone to either the sheer stupidity of people on the forum or the success of this particular (CQ) template. Now we're not gonna suggest that anyone is stupid are we
In short, any advice as to a better template must surely warrant samples first ???

[kwag]

KVCD templates have been updated.

Read changes here:
http://www.kvcd.net/forum/viewtopic.php?t=43

Regards,
kwag

[Sefy]

nobspangle, i have two things to say to you:
1 - Show some respect to other people
2 - The whole idea of a Template is that you have optimized settings already saved and you don't need to do it every time and again, and optimized settings mean tweaked, for either higher quality playback or smaller file size, whatever you set it for.

[vidkid®]

I'd only add that if you go on to read the individual reviews of the encoders, for TMPGEnc the author says:
"I tried the 2-pass VBR mode, and although the flaws I reported on in earlier versions of this review have been fixed by now, 2-pass VBR is still inferior in image quality to CQ mode. This behavior is consistent with that of other encoders, so I can only conclude that the CQ algorithm is inherently superior."

see http://tangentsoft.net/video/mpeg/reviews/ and then

http://tangentsoft.net/video/mpeg/reviews/tmpgenc.html

seems that kwag is not the only one to have come to this conclusion

[playndirty101]

can u guys explain to me why my audio is out of sync and video hangs certain frames...it only happens like every 3-5 mins.

what am i doing wrong? i used vdub to extract the audio (AC3) and used headache to convert to .wav, them tmpgenc to convert to mpeg-1

downloaded from edonkey "shallow hal"

anyone? your help is appreciated

[thehire]

Are the resulting VCDs created after making all these tweaks playable on just as many standalone DVD players as leaving the standard VCD template settings alone?