battle of the CBR/VBR/CQ modes

[adam]

With the CQ, you do a sample with an arbitrary quality setting, say 75. If the source material is not particularly compressible you will get a larger filesize than you anticipated. You then decide whether to go for more discs or to drop the quality setting. You do not get this choice with 2-pass VBR.

Uhh, yes you do. What is preventing you from taking that same test encode you did with CQ and doing it with 2-pass vbr instead? Doesnt this make MUCH more sense? You will always know what the filesize is, but you can adjust the avg bitrate according to quality. Match this quality level to that of CQ's and you will get an equal quality file in a smaller filesize.

With a CQ setting you can take account of the nature of the source material, and you get adequate warning of this by a simple proportion calculation after doing one or more samples"

Again you can do the exact same thing with x-pass vbr. Why do you keep referring to x-pass vbr and CQ like they are so different? What you keep failing to realize is that CQ is nothing more than 1-pass vbr. It is exactly the same as x-pass vbr except that it doesnt analyze the file first which results in less accurate bitrate allocation. They are the SAME but x-pass vbr does it better! That is why x-pass vbr is intrinsically higher quality.

how many people do you suppose make a 2-pass VBR encode and then decide to do it again because the quality is not as good as they hoped?

Why must you do the entire encode? Why not a sample? Even assuming you do have to encode the whole thing how is this any different than with CQ? Do you really want to use the concept of multiple encoding sessions as a defense for CQ? Isn't that shooting yourself in the foot?

So, although you may not agree, I would say x-pass VBR may be theoretically best if you know how many bits you are going to use, but CQ is practically best because you don't actually know how many bits are going to be needed

But this is totally irrelavant. The best quality mode to use is entirely a subjective opinion. The only question that was ever raised in this thread was which encoding mode was the highest quality. What do you mean you don't know how many bits your going to use? If your comparing CQ and x-pass vbr then YES you do know how many bits you are going to use. How ever many bits your CQ encode used thats how many you use in x-pass vbr. Your using CQ's biggest fault, unpredictability, as its only defense. Is CQ better because it uses at least as much bitrate as needed to maintain your quality setting? Well thats a matter of opinion. But is it higher quality because of this? Of course not. If you want to make an accurate comparison between CQ and x-pass vbr you have no choice but to take compression into account because you can draw no conclusion about a larger mpg vs a smaller mpg except that one uses more bitrate than the other. Again, the best mode depends on your personal needs, the highest quality mode is the mode which produces the highest quality when all other factors remain equal. Please refer back to the original question and please refer back to my prior post. I'll say it again, CQ does not produce higher quality than x-pass vbr and there is simply no way you can justify that statement.

Although you seem to think the reference I gave http://tangentsoft.net/video/mpeg/enc-modes.html
not worthy of consideration, I am not quite sure why.
I certainly agree with the 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."

I never said it wasn't worth consideration I said it wasn't evidence of anything since that author has no more authority than you or I. It is his personal opinion and it is duly noted but I feel that the author, just like you, is failing to realize the relativity of quality. Its perfectly understandable why one would say that a file that uses %30 more bitrate is higher quality, but it is utterly moronic to suggest the file is higher quality because of the encoding method. At the same bitrate x-pass vbr is higher quality than CQ, and at different bitrates no comparison can be drawn at all. The author of that document failed to address this fundamental fact and you yourself said you have never run any specific tests on this matter.

Please just try to read my posts in context of the thread. Yes CQ has practical purposes and depending on YOUR PERSONAL needs it may be more practical to use than x-pass vbr. But this has nothing to do with overall quality. Again refer back to your original post, "CQ will actually give you better quality than x-pass VBR," and just realize that you are wrong. The original poster, in no uncertain terms, asked for the highest quality encoding mode. Are you still trying to argue that CQ is higher quality than x-pass vbr? What exactly are you trying to argue?

Your personal opinions are your own and its impossible to argue with an opinion. But if you or that author were to look at how x-pass vbr and CQ actually encode then you would realize that they are the same. This is no longer a qualitative issue it is a quantitative one. They both do the EXACT same thing but x-pass vbr can achieve better bitrate allocation. There is nothing up for debate about this, it is academic. Two passes is better than 1, its as simple as that.

Incidentally this thread reminded me of something that might be worth mentioning. I believe DVD2SVCD does a compressability check of your source before encoding in CQ. It basically takes a series of stills at an interval throughout your movie and encodes it in CQ mode and then adjusts the quality level of the actual encode to try to reach the target filesize. Its not perfect but it is more accurate. For those of you who use CQ you may want to use DVD2SVCD to give yourself a little more control.

[dvd2svcd]

Two passes is better than 1, its as simple as that.

Exactly!

[baker]

Its funny this topic came up, I was just playing around with multipass and cq modes myself.

This is the way I see it:
Using a vbr with an average of 1600 will get 60mins per cd in svcd quailty. With a cq of 8 I can get similar quailty and usually get 60mins per cd. Now I should let you know that on a 450mhz system a pass in cce takes 12 HOURS!!!!! So time is of assence to me. I used to use 1 pass+vaf as everybody told me that multipass is good cq is bad. I now know thats not really true. So now i use singal pass, it gives me just as good results (perhaps now on 3cds) its far more worth it time wise for me.


Baker

[man_friday]

Damn this is too good to miss out on.
This whole damn thread is about subjectivity. There are so many more variables to consider other than how you stroke TMPGENC. It's about what pleases you most. if you want to VBR 100 times, do it, if you want 1 dvd on one CD r do that too. Get back to helping those poor bastards who are having problems.

[banjazzer]

Look, I don't want to fall out over this. I respect your views, but I don't think they are right.

Uhh, yes you do. What is preventing you from taking that same test encode you did with CQ and doing it with 2-pass vbr instead?

Nothing, except I don't believe many people do this in real life!

Why do you keep referring to x-pass vbr and CQ like they are so different? What you keep failing to realize is that CQ is nothing more than 1-pass vbr. It is exactly the same as x-pass vbr except that it doesnt analyze the file first which results in less accurate bitrate allocation. They are the SAME but x-pass vbr does it better! That is why x-pass vbr is intrinsically higher quality.

This is nonsense. The only reason for the first pass is because you are constraining the amount of bits to be used. With CQ there is no need for a first pass, because you haven't limited the number of bits to be used!.

If your comparing CQ and x-pass vbr then YES you do know how many bits you are going to use. How ever many bits your CQ encode used thats how many you use in x-pass vbr. Your using CQ's biggest fault, unpredictability, as its only defense. Is CQ better because it uses at least as much bitrate as needed to maintain your quality setting?

Well, if you do a CQ encode and then note the bits used, you will certainly be able to use this for a 2-pass VBR encode. But this is nonsensical, and doesn't happen.

Your using CQ's biggest fault, unpredictability, as its only defense. Is CQ better because it uses at least as much bitrate as needed to maintain your quality setting? Well thats a matter of opinion. But is it higher quality because of this?

No, it is better because it is going for a certain quality regardless of bits used. 2-pass VBR is going for best quality with a constrained quantity of bits used.

I'll say it again, CQ does not produce higher quality than x-pass vbr and there is simply no way you can justify that statement.

I will concede that 2-pass VBR is capable of producing equal quality to CQ if used properly.

Are you still trying to argue that CQ is higher quality than x-pass vbr?

In most cases, in real-life situations, yes. The reality of the situation is that most people who use 2-pass VBR do so to get to a particular filesize. You use CQ because you are going for a quality result foremost, and the size is a secondary consideration.

[banjazzer]

Get back to helping those poor bastards who are having problems.

We can do that as well. We can have an argument and help people. Some of us are operating in 2-pass multitasking mode. 8)

[adam]

Uhh, yes you do. What is preventing you from taking that same test encode you did with CQ and doing it with 2-pass vbr instead?
Nothing, except I don't believe many people do this in real life!

Of course not. In real life our encodes are dictated by our media size. Compression is pretty much always an issue so there is no reason to try and pit x-pass vbr and CQ against each other, because x-pass vbr always results in better quality at the same filesize. My point is that the ability to run samples to test compressibility is not an advantage of CQ encoding, like you suggested, since the same thing can be done with x-pass vbr.

This is nonsense. The only reason for the first pass is because you are constraining the amount of bits to be used.

Oh really? Then what is the point of a third pass, or a fourth, or a fifth? I have literally mentioned the benefit of added passes in every single post I have made in this thread. In one post I even called you on your inability to understand the root of my entire argument...remember this...

"You think the only point of doing multiple passes is to guarantee a set avg and a set filesize? No wonder your not understanding the point of my post."

The point of an x-pass vbr is NOT to set a desired filesize, that is just an added benefit. The entire point of x-pass vbr is that with each additional pass the encoder becomes more efficient with its bitrate allocation. At the same filesize better bitrate allocation directly translates to better quality. CQ is simply 1-pass vbr so if they were to use the same bitrate allocation, CQ and x-pass vbr would LITERALLY produce the EXACT same file! Bit for bit it would be EXACTLY identical. But since x-pass vbr can scan the source first and allocate bitrate more effectively it will save bitrate on less bitrate intensive scenes and use it where it is needed more. This is the ENTIRE point of vbr encoding. This is the ENTIRE reason one would even use CQ over CBR. So like I have said since my very first post, CQ and x-pass vbr are identical at the same bitrates but x-pass vbr will always have the benefit of better bitrate allocation. This means that x-pass vbr is intrinsically and undeniably higher quality than CQ. Whether or not it is the best format to use is subjective, but the quality issue is not.

You said yourself that you have never actually run any tests to compare CQ and x-pass vbr and in this last post you have clearly shown that you don't even understand the point of vbr even though I have been saying it all along. I would like to respect your opinions as well but they are groundless. You really should not argue so adamantly about things which you don't even understand. You are trying to make this a subjective comparison and it is not. If you take the time to look at mpg encoding and how mpg encoders actually work then you will see why the comparison is so simple. They both do the same thing but one has more information to do it with, and thus can always do it better. Until you actually take the time to test one mode against another, until you actually learn what x-pass vbr is used for, until you actually learn how mpg encoding works...Can't you just take my word for it? I give up, believe what you want.

[vitualis]

@ banjazzer:

I understand where you are coming from and I think that adam does as well. However, I think that you are missing his point.

CQ is just another name for 1-pass VBR where the computer dynamically chooses the bitrate rather than you.

Now, this may be quick and save time and it may appear to yield better quality, but in fact it is not. The reason it is not is as adam has already stated: you are NOT using the same measuring stick.

You do not know the final average bitrate of a CQ encoded clip. In other words, you don't know the final size of the MPEG. If you use the same specs to encode 1-pass VBR, the results would be identical.

However, in terms of maximisation of quality, CQ cannot look better than a standard VBR encoded clip. Why?

The reason is that you DON'T know the final size of the clip BEFORE it is encoded. We live in a world with fixed media sizes. For VBR encoding, anything that doesn't completely fill a CD = lost bitrate potential.

For example, say I have a movie that is 120 minutes long and I want to encode it so that disc one has 65 minutes and disc two has 55 minutes (I want the break to be at a chapter break). With standard VBR encoding, I can encode each section in turn so that it would fill the entire disc (thus, maximising the average bitrate).

With CQ, even if I were very experienced, I still wouldn't get to the degree of precision (or any) I could with standard VBR. This means that the final resulting SVCD would have more "blank/dead space".

In other words, for a fixed medium size (e.g., CD-Rs), the result of CQ encoding = standard VBR but with a lower average bitrate.

This is obviously true for multipass VBR as well.

Now, you could probably make the argument that CQ encoding can deviate from the "average bitrate" more than a standard 1-pass VBR encoding (i.e., is is more "variable") and this may or may not be true. However, this advantage I would imagine be lost for 2+ -pass VBR encoding).

Regards.

[adam]

Well thats another good point Vitualis but not my main arguement. Even if you did use the same avg bitrate and got the same filesize with each encoding mode and filled the cd to its full potential, the x-pass vbr encoded file would still have better bitrate allocation. Your most bitrate intensive scenes would use more of your bitrate and your less bitrate intensive scenes would use less but with no quality loss. The avg is the same in both cases but the distribution of those bits is not. This should be enough to put x-pass vbr over the top in all cases no?

With the same avg and the same filesize is there any possible way to make a CQ encoded file look better than a 2-pass vbr encoded file? No.

With the same avg and the same filesize is there any possible way to make a 2-pass vbr encoded file look better than a CQ encoded file? Yes.

THAT is my point.

Adam I think maybe you dont really understand what you are talking about and maybe you should try making a few samples yourself.
I know I have made many comparisions of 2passVBR, CBR & CQ at various settings.

There are advantages & dissadvantages of each of the above forms of encoding and I dont believe one(VBR) is perfect. I often change which form of encoding I use depending on the source and what I want to achive.

Also with X-pass VBR, with each pass you will introduce new distortions for the space you save, nothing comes for free dude, and no method of encoding except a very high bitrate CBR comes close to perfect.!

[banjazzer]

vitualis

However, in terms of maximisation of quality, CQ cannot look better than a standard VBR encoded clip. Why?

The reason is that you DON'T know the final size of the clip BEFORE it is encoded. We live in a world with fixed media sizes. For VBR encoding, anything that doesn't completely fill a CD = lost bitrate potential.

We don't ever work to one CD with SVCD, nor is it a big deal if part of the last CD in the set is blank. The quality is what is important, and the price of CDs minimal. It seems what you are saying is the important thing is to fully fill every CD - I would disagree.

For example, say I have a movie that is 120 minutes long and I want to encode it so that disc one has 65 minutes and disc two has 55 minutes (I want the break to be at a chapter break). With standard VBR encoding, I can encode each section in turn so that it would fill the entire disc (thus, maximising the average bitrate).

You certainly could do this, but I wonder how many actually do this in practice. However, both chapters would be encoded very likely at different quality settings. I would prefer not to insist on breaking at a chapter, but rather have the movie encoded at a consistent quality throughout.

Now, you could probably make the argument that CQ encoding can deviate from the "average bitrate" more than a standard 1-pass VBR encoding (i.e., is is more "variable") and this may or may not be true. However, this advantage I would imagine be lost for 2+ -pass VBR encoding).

I would definitely make that point. CQ can deviate from the average, because there is no fixed average to start with! 2-pass VBR cannot. The advantage of 2-pass VBR is that it makes best use of the bits you give it, which you choose at the outset. The advantage of CQ is that you are not limiting the bits used at the outset, and hence cannot predict the size. In this case you choose the quality at the outset.

[banjazzer]

adam

I do not agree that 2-pass VBR has any better allocation of bits than CQ. As I stated earlier, the only reason for the first pass is that your total number of bits is constrained. CQ is encoded to a certain quality, bits are used as and when required, because there is no average to work to.

With the same avg and the same filesize is there any possible way to make a CQ encoded file look better than a 2-pass vbr encoded file? No.

With the same avg and the same filesize is there any possible way to make a 2-pass vbr encoded file look better than a CQ encoded file? Yes.

But CQ does not have an average to work to. I disagree that if you do a CQ encode, and then do a 2-pass VBR to arrive at an encode with the same filesize, the latter will be in any way distinguishable from the former. Well, it might if you radically change max and min bitrates between the two encodes, but with these factors identical I believe the encodes would look the same. I am sure it is tempting to think, because you are doing the extra pass and taking twice the time, that the quality will be improved. It will not, because the first pass is only needed because you are working with a fixed number of bits, which you control from the outset.

THAT is my point.

Banjazzer I think you hit the nail on the head when you mentioned

. CQ is encoded to a certain quality, bits are used as and when required, because there is no average to work to.

And I think this is the fundamental point ADAM is missing.

[adam]

adam

As I stated earlier, the only reason for the first pass is that your total number of bits is constrained.

the first pass is only needed because you are working with a fixed number of bits, which you control from the outset.

Arghhh! This is like trying to teach a rock to swim! If the additional pass is ONLY needed to set a constrained avg bitrate then why encode in 3, 4, or 5 pass vbr? Did you just miss that question or did you ignore it? What are those extra passes for, shits and giggles? PLEASE before you type another word go to google and do a search on multipass vbr! The WHOLE point of multipass vbr is to achieve better bitrate allocation NOT to be able to set an avg bitrate.

I could literally quote from a thousand different sources but here is a quote taken directly from cinema craft encoder's manual...

"Multipass VBR

...At least two passes are necessary. For the first pass, CBR or One-pass VBR is executed to check the complexity of the images. In this pass, a file, where the complexity of images are described for each frame, is created. For the second or later passes, BIT ALLOCATION IS PLANNED for each frame based on this information on complexity and actual encoding is executed accordingly...IMAGE QUALITY IMPROVES AS THE NUMBER OF PASSES INCREASE."

CQ is 1-pass vbr, nothing more nothing less. x-pass vbr is at least 2-pass vbr. They ARE IDENTICAL IN EVERY WAY except that x-pass has better bitrate allocation. At first I just thought you were stubborn but it is now apparant that you are both stubborn and ignorant. Its perfectly ok that you don't understand how vbr works. Whats not ok is that you keep arguing under the guise that you do, without even taking 2 secs to verify what you BELIEVE to be true. This is like the sixth time I have stated that additional passes achieve better bitrate allocation. Rather than disregarding this as false based on your own personal theories why not just verify it for yourself? You have repeatedly said that 2-pass vbr will not achive better bitrate allocation than 1-pass vbr. PLEASE read ANY article on multipass vbr and you will see just how wrong you are. Instead of rehashing your same old inaccurate arguments just educate yourself.

D_Head, no offense but that post really made me laugh. Trust me, with additional passes you do not introduce new distortions. I was referring to your statement that CQ produces higher quality but that the filesizes are larger. As I have explained throughout this thread the only reason the quality is better is because you used more bitrate, which caused the larger filesize. The only thing a test like this proves is that you used more bitrate in one encode than another, it has absolutely nothing to do with quality. Its like saying a pebble is heavier than a boulder, but oh by the way I have a million pebbles and only one boulder. I would suggest that you also go to google and search for "multipass vbr."

Adam- first off `CHILL WINSTON`


Secondly I think you should be the one to search Google to understand exactly what compression is, whether Audio or Video, .Exactly what is CBR & VBR is, and then the variation of VBR(CQ)

You will see that using VBR the more passes you make or(another way of explaining this) The more you optimise the bitrate to fit the properties of the source, again Audio or Video, you will introduce distortions.......as you are dropping information(bits)........that you had before you started.

The next point is when this is (VBR encoding) done well, the visable (or audable distortions are kept to a minimum. So the the resulting encoded file(eg mpeg) appears visably the same when played but the filesize is smaller.

BUT in many cases these distortions(eg lack of sharpness&blocks) are visable, and I have found using CQ in Tmpgenc with Planet of the apes I can fit all 1hr54mins on 3 CDs with a better quality at least to my eyes than a 2passVBR (3CD).

Many times I find the situation reversed and I prefer VBR-Xpass. Again depends on the source.

[SatStorm]

This post is a dead end.

If someone want to believe that CQ_VBR (1 pass vbr) is better multipass VBR, it's okey. Let him believe what he wants.

All moderators, advance users, enthusiasts, etc from this forum and the svcd/vcd scene in general, stated at least once that multipass vbr is better. Well, it seems that we are all wrong. Okey, we are all wrong. Let as be wrong, as we let you be correct....

Meanwhile, let me copy to my friend the movie "Highlander 4: Endgame", at 2:11:1 picture format, encoded @ 352 X 576 mpeg 2, interlace, 2 Pass VBR, min 420, average 1100, maximum 3500. I have it on one CD and looks like the original DVD...

[adam]

Well I leave for a 2 week vacation in a few hours so I guess I'll let banjazzer and D_Head have the last word on this one. Honestly I think they can prove my point better than I can.

Yeah I guess your right SatStorm. Every single document ever written on multipass vbr must be incorrect. After all, that guy on tangentsoft said so. I don't know what I was thinking...oh well at least I can admit when I'm wrong, HINT HINT.

[banjazzer]

Arghhh! This is like trying to teach a rock to swim! If the additional pass is ONLY needed to set a constrained avg bitrate then why encode in 3, 4, or 5 pass vbr?

We were talking about 2-pass VBR versus CQ. I don't know about 3,4 or 5-pass VBR, nor do I know anyone who actually uses them. Please don't keep moving the goalposts. I do know that 2-pass has distinct limitations because the filesize/bits used is constrained at the outset.

Some of you guys are making a damned religion out of this, and no doubt in a different time and a different place I would have been by now hunted down for heresy. I don't mind you having a different view to me, even if it is wrong.

I know about TMPGenc, not other encoders. From the TMPGenc helpfile,
my emphasis:-

[2-pass Variable Bitrate (VBR)]

Every scene in a movie has a different optimal bitrate. In this mode, a scene is analyzed to determine the best bitrate before it is encoded. One can set the maximum, average, and minimum bitrates to use.

[Constant Quality (CQ)]

When one is more concerned about picture quality than file size, one should use this mode. Using the CQ mode, one can set all of the aspects of quality to use in the encoding. This mode will guarantee a high quality movie; however, movies with many scene with rapid motion can become rather large.

Note that in 2-pass VBR you start out with a fixed number of bits.
Note that in CQ you don't.

I also suspect the creators of TMPGEnc know about their encoding modes. 8)

[SatStorm]

You lucky Adam...
I have to wait till Augost for my vacations. I have a month and a half ahead! ^*(#**
Have nice vacation and do me a personal favor: Have fun!

[tenders]

i have ran a test with the same source,bitrate,and from a quality standpoint,cq is better[/u] than 2 pass vbr

Tenders good to see someone else is also running a few tests.
Just out of interest what was the difference in size between the 2 files.% wise.
Also what min/max bit rate etc did you use.

cheers

[tenders]

the cq file was actually smaller by about 1mb.

cq=2520 max cq=65

vbr=0 min 2520 max ave 1150

with cq the file came out very nice! :P
with vbr 2-pass there was alot of pixilation

well...ok not alot but it was noticable

[smokingweed3000]

thats because although u think u had the same bitrate u didnt. follow this method and then post your results:

-encode the file with CQ at any settings u want
-run that CQ encoded file through a bitrate viewer
-write down the max, min, and avg bitrates
-take the same file that u used for the CQ and use 2-pass VBR and the same settings (i.e. resolution, gop, and all that other stuff) except for bitrate i should put in those max, min, and avg bitrates u wrote down earlier.
-now u can enocde
-when the second file is done ull have two files with exactly the same settings and bitrates except for one is CQ and the other is 2-pass VBR.

only now u can compare the two methods (CQ and 2-pass VBR) cause u've created two files with the same settings.

[smokingweed3000]

ofcourse the cq file would look better cause it had a higher bitrate, just follow the method i posted in my last post. u can get the bitrate viewer at the tools section at the bottom. u have to know whats the min, max, and avg bitrates of the CQ file and then enter those bitrates in for the 2-pass vbr, thats the only way u can compare. remember to have all the other settings exactly the same for both files. when u r done ull be amazed. then post yr result and u will see that adam was right all along.

[q1aqza]

For encoding DV camcorder footage I have found CQ 100 produces the best quality picture and obviously it is faster than x-pass VBR and smaller filesize than CBR. I don't have the concern of filesize for camcorder stuff as I usually only create 20-30 mins films.

For DVD back-up where you really need to control filesize then without doubt I use x-pass VBR and I don't see much difference in picture quality between CQ and VBR when using a DVD rip as the source.

Therefore for me it is CQ100 for DV camcorder and x-pass VBR for DVD rip. I also disagree with an earlier post saying that CQ100 is basically CBR. I found a 25 minute DV camcorder encode (max 2520) was 50MB smaller using CQ100 compared to a CBR at 2520.

[smokingweed3000]

first to q1aqza, u can make yr file even smaller cause CQ at 90 is the same as CQ 100 and u can even use 85, there is not much difference.

all of u that say that CQ is better have to remember that 2-pass VBR uses the bitrate that u gave it more efficently therefore making the file better quality and also making it a lot smaller in size

[adam]

Time to squeeze one more post in, CQ %100 is CBR, literally. No wonder you get best results at this level.

Banjazzer, its amazing how powerful denial is when your backed against the wall. Your still missing the point. Have you checked google yet because your still avoiding all the hard questions. For instance...

"This is like the sixth time I have stated that additional passes achieve better bitrate allocation. Rather than disregarding this as false based on your own personal theories why not just verify it for yourself?"

We were talking about 2-pass VBR versus CQ. I don't know about 3,4 or 5-pass VBR, nor do I know anyone who actually uses them. Please don't keep moving the goalposts.

I know about TMPGenc, not other encoders.

No we were comparing CQ to x-pass vbr. X means fill i the blank and I can assure you than many many people use at least 3 pass vbr in their encodes. Also TMPGenc does support more than 2 passes in the latest full release. Way to dodge the question but Im not trying to get you to give me a real answer here why one would use more than two passes. You don't have to know a damn thing about encoding to see the illogic of your statements when viewed in light of the fact that 3, 4, and up passes DO exist. PLEASE just check your facts, your really going to kick yourself later.

Hey ADAM ,pull your head out of the sand for a moment, stop getting all personal, its suposed to be a Forum to discuss things, you cant be right all the time, other people often have some good infos if you

LISTEN

[banjazzer]

Adam

I was specifically talking about 2-pass VBR, so it was wrong of me to mention x-pass VBR. However, if x-pass still works to the constraint of the average bitrate, then quality will be limited by that average bitrate, which is fixed from the outset.

Basically, 2-pass VBR implies constant filesize and fixed number of bits used, however you want to phrase it. These bits are then used in the most efficient way. But the bits used may not be enough for the quality you desire. CQ means constant quality, and as many bits as it takes to provide that quality are used. This is not fixed from the outset, but is determined by the quality setting and the source material. Hence, the quality is guaranteed, but the filesize/bits used aren't.

All I can say is that you must know more than the creators of TMPGEnc do about their own encoder. I notice, unless I missed something, that you haven't commented on the piece I quoted from the TMPGEnc helpfile. I can only see one interpretation of the piece about CQ, namely - "When one is more concerned about picture quality than file size, one should use this mode. " What do you suppose that means, or have they made a mistake? 8)

[baker]

ANYOBY HERE WHO KNOWAS ANYTHING ABOUT CQ GO HERE:

http://forum.vcdhelp.com/viewtopic.php?p=270457#270457

Baker