If you have 1 Hr 50 Min total then the average bitrate should be about 5,300,000 bps. Set the low bitrate to 192,000 and the upper bitrate to 8,400,000. Use AC-3 192 encoding. Use 2-pass. Using the Vegas internal encoder, that will give you the absolute best you can get with the material you have.

If you are asking about trading off quality in the outtakes (which I assume are included in the 1:50 total time) by encoding them at a lower bitrate, then the way to do that is to encode them at a MUCH lower bitrate (like 4,000,000 or even 3,500,000 bps). Once you encode them, select them all, and the right-click and select properties. This will give you the file size. You then subtract this from the 4,700,000,000 size for a normal DVD, and enter this into the bitrate calculator as the "custom" size for your DVD. Plug in the remaining time for the tracks that you want to encode at higher quality, and the calculator will give you a recommended average rate.

This is the best you can do within Vegas.

For even better results for ultra-low DVD bitrate encoding, you may want to look into using an external encoder. I use the MainConept external encoder and have previously published (click Here) in this forum the settings you can use for unusually low bitrate encodes. There is no magic -- low bitrate settings will inevitably produce artifacts during motion -- but proper settings can reduce the severity of such problems.

Hiyah Scott!

Understood, I know it depends on how much active and non active movement there is - I understand that. And that you wouldn't want to steer me towards madness! Useful tip on "edge" work too!

BUT, but but . . how about this for a thought. Is there ANY software out there that could scan an AVI and tell me WHAT the bit rate peeks at or what it is overall? I could then frameserve this AVI to this s/w an and get I MAX rate from it.

Only an idea.

Thanks for your thoughts. I've got to that part of my development that I'm wanting just to be aware of what "extra" I can usefully get without the need to jeopardise quality. If MAX8 is silly then I'll drop that.

One of our colleagues can get 2:30hrs by using MAX4.5 and MIN3.5

Grazie

Thanks John.

Good sound advice. No the outtakes are additional. It comprises of 2 sets of roughly 8 mins each. I can do one extra but NOT the other. Pity . ..

Is there any worthwhile drop in size that COULD hurt quality going down to 7.5 max? Obviously you wouldn't know as this is all dependent on the actual requires that the AVI is spewing out.

I really like the idea of a PEAK rate meter - is there such a thing?
I guess it is back to Suck 'n See eh?

Grazie

I would not have believed this had I not seen it myself, but you can make a very good quality DVD by encoding to a max bitrate, creating a DVD VIDEO_TS file and then use DVDShrink to make it fit the disk. DVD Shrink does a very good job at changing bitrates and maintaining high quality. It also save you having to try to calculate.

Dave T2

WHAT?? Really? Excellent! Anybody else tried this???

So I load the DVD VIDEO_TS - from whence I created it within DVDA3 - into Shrink? Do I then tell DVDA3 to go and burn this? Woaaaa.... So how much shrinkage? How much time gained . . .

Grazie

I have used DVD Shrink to shrink a commercial 2 hour movie to a single sided disk.

There are AUTO Settings or you manually set the amount of shrinking.

DVD shrink will burn the disk for you if you have NERO installed, so you want to have a final VIDEO_TS file ready (chapters, etc). IF you don't have nero then I think you create another TS file which you then burn to disk using DVDA or other Video capable software.

Dave T2

OK, sorta off/on the same track. Most of the stuff I burn is less than an hour in length. Any recommendations as to bitrate settings to maximize quality and insure compatibilty? Would CBR be better than VBR for that short material? I'm currently using the default settings for encoding for DVDA and have not been unhappy with the results - just wondering if I could do better.

TIA

Tom

For anything under an hour I always use 8000 CBR. I have not had any problems with this rate on DVD players that I have used them on.

Dave T2

I don't understand how DVDShrink can reduce the output, without re-encoding the mpeg2. Maybe it looks good on a small screen, but if you're squeezing a near two hour movie to fit on a 4.7 GB disc, it has to be re-encoding.
Try playing the disc on a big screen and compare it with the original and you should see some differences.
Keep in mind, that Hollywood uses some major 3 or 4 pass encoding to get the bit rate down and look as good as they possible can. Probably somewhere in the 5,000,000 bps or lower range. So, the re-encoding by DVDShrink, may not be shrinking it that much.
So.......why do we NOT have the option of doing more than 2 passes? True, I'd hardly ever use it, but I'd like the option.

For the ultimate in quality, use CBR at the highest bitrate your player can handle. There is much debate about this, and a lot of urban legends. However, every DVD player is supposed to be able to handle continuous data rates of 9.8 MB/sec (including the audio). Read this document for more information:

mpeg.org (Document describing DVD spec and maximum bitrates)

You might also find useful the Sony MPEG encoding guide:

MPEG Encoding Guide

As for DVD Shrink, I have used it on commercial movies, and it can do wonders. However, I don't know how well it would work with interlaced material encoded with a prosumer encoder. You'll just have to try it and see if you like the results better than simply encoding at the "proper" bitrate that matches the length of your material.

CBR vs. VBR is really only a question when you are encoding at low bitrates (for me the dividing line between low and high is somewhere between 5,000,000 and 6,000,000 bps). As you approach the maximum bitrate allowed (somewhere north of 8,000,000 bps), the two become identical. I am not aware of any advantage to using CBR rather than VBR, even at higher bitrates, so I always use VBR.

As far as Grazie's idea of scanning the AVI to find the peak bitrates, this is exactly what the 2-pass VBR encoding does, and it does it all for you automatically. 2-pass makes a BIG difference as you go to lower encoding rates. You definitely want to use it when you go below the "threshold" that I defined above.

It does some re-encoding, but I am hardpressed to see any major loss of quality. I am not sure what the author is doing.

Give it a try and see for yourself. It is free software.

Like I said in my first post, I was skeptical when I heard other people talk about doing this, but I have used it a few times and was quite impressed.


In some situations I have found that the quality was better than encoding at the "proper" bitrate. This probably depends a lot on the quality of the encoder that you are using.

Guide to using DVDShrink

http://www.doom9.org/index.html?/mpg/dvdshrink31-main.htm
Dave T2

Actually, it doesn't re-encode. I have been trying to find a long-lost post by the elusive author where he actually described in detail how it worked. It all has to do with I, B, and P frames.

For a really wonderful visual depiction of what happens to the material when you run it through DVD Shrink, take a look at this:

DVD Shrink Visual Comparison

If I find the technical description of how it works, I'll post it here.

[Edit]

Here's the link to the technical description in a rare post by the author himself:

DVD Shrink Explained

Scroll about 2/3 of the way down the page until you get to the long post by "DVD Shrink."

[Further Edit]

Please note this text from the above link:

Note that at low compression ratios, DVD Shrink will never need to touch I and P pictures. The exact % compression where this becomes necessary depends largely on the DVD, or more accurately, on the video encoder software used to encode it.

Thus, with a lesser encoder (Vegas' Mainconcept encoder compared to "Hollywood" encoders used on major motion pictures), you will much more quickly get into a situation where DVD Shrink will "break down."

I've fitted over 3 hours of video onto a SL DVD, results were very good. Pans and zooms looked fine, the two dissolves though were a little sad but then again I was looking hard. Footage was from 1/2" camera on good sticks in daylight.
Put simply, your biggest enemy is NOISE.
That's why I get pretty hostile about those who rave about cameras and mojo etc. Sure it might look fine played out uncompressed, that's not how your viewers will see it.
I've had a DVD stop playing because of noise, 90% of players stopped at a fade up from real black, the fade up was actually an iris up and the noise was real bad i.e. the bitrate went from 192K to 8M in a few frames and the players simply could not track the disk.
So my first word of advice, don't set the minimum bitrate too low or you can have problems.
My second word of advice, do all you can to get rid of noise in your video.

Most PC DVD players will display bitrate dynamically, look in the options to turn it on.
Also Bitrate Viewer will graph the bitrate of a mpeg-2 file, can be very handy.

Better encoders such as the Cinema Craft one are available that do upto 400 passes. They COST and I really wonder if they could do any better with noisy video.

As some have said the biggest impact on your final output is detemined by the quality of the input. If you create that input with how temporal encoding works in mind you can fit an amazing amount onto a DVD without quality going south.
One other trap, serious CC can push the noise floor up badly, try to get the color right in camera.

Bob.

The obsession videographers have with cramming everything on a single DVD is mind boggling. Customers don't care if you give them two DVD's with nicely done artwork in a double DVD case as long as they don't have to pay more for it.

Render two DVDs at a CBR of 8,000,000 and you will have the best possible quality from DVDA.

John

Bob,

Good advice, all of it.

Is there a problem with a dual layer disc?
JJK

Is there a problem with a dual layer disc?

None, except that early reports on compatibility were not promising. I think there have been some posts recently indicating that with the proper media, this may not be as big a problem as once thought, although clearly dual layer will not be as compatible with as many players as single layer +R or -R media. The other problem with dual layers is media cost. Good single layer media is $0.50 to $2.00 a disc. The cheapest dual layer media I've seen was a recent promotion for $2.00 a disc, but most of the media is closer to $10/disc. Thus, there is, at the moment a 5-10x increase in media cost for dual layer compared to single layer.

Of course, there is dual sided, but labeling is an issue ...

I am not sure what the author is doing.

Last I heard, he was hired by Nero which is why DVD Shrink hasn't been developed any further.
Similar to the way the guy who wrote DVD Decrypter was hired by Macrovision.
Some software companies figure they'd rather have these guys working for them than against them. Good thinking on their part :-)

Mike

We have used DVDshrink occasionally for creating raw footage copies for clients, and on my 57" Sony, I can tell no difference. It is marvelous.

Back on topic now Grazie.
I've done several 2 hr. encodes before and been happy with the results.
I tried a 3 1/2 hr. one once (teacher at a blackboard) and decided to go with 2 DVDs instead.
I ran your times (1 hr. 50 min. + 2x8 min. extras) through the bitrate calculator (links to a zipped file) I use and came up the following settings:
Max: 7900
Avg: 4500
Min: 2700
BTW, I've used DVD Shrink as well and been quite pleased with the results. It really is an amazing piece of software.

Mike

Put simply, your biggest enemy is NOISE.

That makes perfect sense. Noise is by definition high frequency data, which will drive up the bitrate all for some useless garbage. I have a plugin for PhotoShop made by Kodak that does the most beautiful noise reduction I have ever seen. It can take an otherwise useless image and make it look good (although a bit soft, which is actually good for portraits). After taking out the noise, the file size goes down dramatically. (Solids are easier to compress than random artifacts, which will hold true for video just as much as for stills).

Thank you for the good analysis, Bob! We learn from this forum every day...

Just got outtaa bed . .and have seen and read ALL your posts . . I'm really grateful . . most helpful indeed - Good People!

OK . ..

John M - I didn't truly appreciate that 2-pass was employing this feature to determine where from AVE to MAX my video was performing? So would it therefore be applying/disregarding more to the AVE/MIN when stuff was more static? Is that how it works? So .. by having a HIGHER max nothing is gained when 2-pass is used in conjunction with a lower MAX - say moving from 8 down to 7.5? Is it that the overall "bandwidth" is in itself reduced THEREFORE making for a "tighter" and more controlled area for the 2-pass application? I thought 2-pass was ONLY working out stuff that was moving a lot and stuff which was more static. Which maybe is exactly what you've implied by your comments. Or have I got this wrong?

Craftech "The obsession videographers have with cramming everything on a single DVD is mind boggling. Customers don't care if you give them two DVD's with nicely done artwork in a double DVD case as long as they don't have to pay more for it".

As to customers wanting and accepting 2 DVDs this is NOT an option on this occasion. But one I would have jumped at if it were. I'll tell you just exactly what "this videographer" is obsessed with is - making life easy for myself. Quite frankly what I really SHOULD do is tease out yet another 10% off the edit. But I tell you what? Life is too short. And just hitting the 1hr 50min I thought would give me JUST enough for the Menu structures and maybe a wee bit of outtakes. So, this videographer had kinda done the "soul-searching" already - Craftie - and had come up with a rational, non-obsessed rational and logical solution. 2 DVDs weren't/isn't an option and I'd carved out just enough to feel comfortable. Obsessed? Nah . . . Easy? Oh yes . . . yes . . Beer? Yes . . Good curries . .yes! There is too much more important than counting "Angels on Pinheads" - but of course this is NOT what you were implying. Is it? I just wish to understand the elegance and import of what I'm doing to make my life easier - for me. 2 DVDs is one way. But getting an extra 4>8mins on an already complete 1hr 50min "down and dusted" video, without the need to include an extra DVD is, by my reckoning not being obsessive it is just finding a place for some outtakes. Often I get to these junctions and crossroads in this craft only out of necessity - it isn't out of choosing.

"Render two DVDs at a CBR of 8,000,000 and you will have the best possible quality from DVDA. Kinda knew that . . Or are you just reminding others? Yup, folks - if you have too much stuff use 2 DVDs!

Bob - Yup noise/CC and getting the shot as right as possible in the first place IS my holy grail. But it isn't gonna slow me down either. It is all called compromise. If I want a shot that is kinda dodgy then I'll CC it. I've seen pre-CCed work done by others. I guess we, as vidfolk, EITHER one-man bands OR multi-staffed companies have to have a thought to the WHOLE workflow procedure and process. And that includes getting the STUFF outta the door too!

MIKE! Excellent . . plus some Qs for you too:

Your recipe:

Max: 7900
Avg: 4500
Min: 2700

I notice you've raised substantially the Min. Why? What exactly does this do? The AVG to MAX bandwidth is lowered - i.e. the whole area for operating on is dropped. Plus the increase in the MIN makes the BAND between AVG and MIN much smaller too. Is this thought to be advantageous? Maybe others would like to respond.

Anyways - good people - it would appear that I'm not doing anything abnormal as fitting 1hr 50 to almost 2 hours on a single DVD.

Best regards,

Grazie