Two Pass Encoding in VS8?

[2Dogs]

Can I use two pass mpeg encoding in Video studio 8?

I can't seem to find the check box for this anywhere.

[daniel]

no, that was introduced in VS9.

in VS8 you can only increase quality from standard 70% up to 100 and then increase bitrate until you're happy.

.

[pdunn]

Has anyone done any comparisons of 1 pass (VS8) to 2-pass (VS9) - is it worth upgrading on the basis of improved quality video ?

Cheers,

Peter

[pdunn]

Has anyone done any comparisons of 1 pass (VS8) to 2-pass (VS9) - is it worth upgrading on the basis of improved quality video ?

Cheers,

Peter

[Trevor Andrew]

Hi

How long is your video?
Why use Variable Bit Rate?

Up to 70 minutes of video use Constant Bit Rate. Using variable is a waste of render time with no benefit to quality. At 70 minutes the bit rate will be about the same for each type, the difference if any, insignificant.

If you use Digital Dolby or Mpeg Audio then more space is available for the video file allowing you to either increase the bit rate or video length.

If you have a video in excess of 90 minutes then you may benefit from using Variable, although I am not that convinced.
At 6000 kbps a 90 minute movie using Digital Dolby will fit to disc.

Trevor

[2Dogs]

no, that was introduced in VS9.

in VS8 you can only increase quality from standard 70% up to 100 and then increase bitrate until you're happy.

.

Thanks Daniel! It was beginning to drive me nuts! My VS9 trial has expired, so I wasn't able to look for it in VS9.

All in all, VS9 seems to feature some very worthwhile improvements and additional features.

[daniel]

Has anyone done any comparisons of 1 pass (VS8) to 2-pass (VS9) - is it worth upgrading on the basis of improved quality video ?

Cheers,

Peter

two pass will give you the better result of two algorithms, the standard (pass 1) and an alternative version with better results in specific cases.
You win if your video happens to include those specific cases and if the standard algorithm is unable to get the requested quality (70-100%).

So if you have a bitrate of 8000 and you ask a quality of 70% you are very unlikely to get any difference.
On the other hand if you ask 100% quality with 4000kbs and a "difficult" video the difference will be dramatic for some sequences. At the cost of twice the rendering time.

It's like a joker you have in case you need it.
I always do rendering overnight so I always ask dual, whatever the other conditions, because time doesn't matter, but I'm sure it most of the time is just wasted CPU cycles (and lots of disk thrashing).

[heinz-oz]

It is really only of value, as I understand it, if you have sections of your video with fast movements. The first pass only analyses the video and sets markers for the areas where a higher bit rate is in order to capture the fast movement with good quality and lower bit rates where high ones are not required. It gives you a smaller file size in these cases. You would have to use high bit rates with such videos if you were to use constant bit rate, resulting in a much larger file.

[daniel]

It is really only of value, as I understand it, if you have sections of your video with fast movements. The first pass only analyses the video and sets markers for the areas where a higher bit rate is in order to capture the fast movement with good quality and lower bit rates where high ones are not required. It gives you a smaller file size in these cases. You would have to use high bit rates with such videos if you were to use constant bit rate, resulting in a much larger file.

With all due respect (truly!) Heinz, your discussion mixes variable vs constant with single/dual. Feel free to flame me if I'm wrong, I always wear my asbestos suit when on a forum...

The file size is almost identical with 1/2 pass, only determined by your bitrate including if it's variable or not.

The quality is first standard: pass one is not only analysis but does the usual generation of the output file, you can check that easily, then checks what is the "delta" of the output vs original (supposed AVI).
Then when all is done it tries it again, I THINK/GUESS GOP after GOP and makes another rendering, then measures the new variance and selects the best of the two results. The only difference is quality, in some cases, not size. I agree with you movement is a factor, as are marginal brightness conditions, color variations, used filters and titles etc.
This is what I think I have observed, but I'm no expert (repeat twice)

[heinz-oz]

I'm no expert either and, sure, my explanations were about single/ dual pass, because, if I'm not mistaken, that was the original question. Nobody would do a dual pass with constant bit rate, would they. The software could also not know where to vary the bit rate without running over it twice. So, AFAIK, the two are linked, variable bit rate needs at least a dual pass where constant bit rate does not require a second pass.

Please feel free to correct me if you disagree.

[daniel]

I would not dare to "correct" you since I don't work for Ulead, but I surely disagree.

As I said, my understanding is single/dual pass only improves quality of the output, and is independent of the use of constant or variable bitrate, and relevant to both methods.

It does not influence the file size other than using different algorithms the result will marginally change after compression to MPG.

What it does is trying an alternate compression and check if by chance it improves the result as compared to the original. The best result is kept, the other discarded.
This has the potential of finding a better way in special cases, and its efficiency is proportional to the compression ratio (bitrate) and required quality setting.
It can't fail though: the worst, but typical result, would be that both algorithms give the same result, good or bad. But surely multiplies the time by slightly more than two. So you need to balance the probability of improvement with the time you have to wait.
It is not "the" definitive asset of VS9, but a nice extra touch.

[DiscCoasterPro]

Speaking of beginners ... lol, I am curious about just using a program like DVDShrink after creating a video file. I have backed up some of my dvd collection using that program, at time it showed very large amounts of compression, and the resulting backup was fine.

I haven't run across a situation where I had a real long home video that I didn't or couldnt' split to multiple dvds, but I always had it in the back of my mind that if I did, I'd just render it to best quality and then use dvdshrink to fit it.

Is this a bad way to do it?

dcp

[Black Lab]

I had a project that ran about 90 minutes. I rendered it at 8000 kbps and let DVDShrink do it's thing and it came out great Very easy to use too.

[Ken Berry]

DiscCoaster: this is a gentle slap over the wrist but you really should have started this question about Shrink as a separate thread as it has absolutely (well, almost nothing) to do with the main point of the thread in which it is posted.

That being said, I use Shrink at least a couple of times a week (and have done so for nearly a year), and am quite impressed by the results. It certainly works much, much better, as far as I can see, than the shrink function now included in VS 9. Nero also has a shrink function which works extremely well with non-copyright discs -- though this is not surprising since the guy who invented DVD Shrink now works for Nero! You might also want to investigate DVD Fab, whose Express function does much the same job as Shrink, though with much less flexibility, and does it a fair bit faster and (usually) flawlessly. Personally, though, I use Shrink to reduce the video to a shrunk Video-TS and then burn the latter with Nero as I like Nero's capacity to verify the burn against the original copy. Takes a few minutes extra per disc but I feel more confident about the burn.

[DiscCoasterPro]

Oops, sorry about the thread, that shows how little I admittedly know. I thought rendering twice was a way to make it smaller, and hence I thought I'd ask if shrink was a good alternative. Hmm, getting back in line then with the thread why would you do that twice?

dcp