Douglas, so if I'm understadning you correctly, the software/disc capability currently exceeds the [consumer] players' capability at this time. Therefore, until the latter catches up with the former, we need to manually adjust the bitrate within the players' tolerance leve?

Did that make sense???

The numbers you give mean the encoder will essentially encode at the CBR of 9,300. Obviously, if the encoder is told to average 9.3 but at the same time it's told it can't go any higher than 9.3, then it obviously can't go any lower than 9.3 if it wants to maintian an average of 9.3!

I think the concern is that the DVD-R data is somewhat harder to read than the data from a pressed disk. If an error is made, the player will have to go back and try to re-read the data. By doing so it falls behind if the bitrate is already high. If it falls behind enough, it will have to drop frames.

But I think as time goes by the DVD players are getting better. My cheap little Apex player will read anything.

if the encoder is told to average 9.3 but at the same time it's told it can't go any higher than 9.3, then it obviously can't go any lower than 9.3 if it wants to maintian an average of 9.3!

My concern exactly!

I used to have a Sony DVD player that simply couldn't handle the DVD-R discs. Now I have a Pioneer, which does just fine. Too, I bought a cheap little Apex and some other off brand ($29!), and, like you say, they play anything I put in them! Although they did have trouble with the peanut butter and jelly sandwich, but it was worth the try. ;o)

DVDA uses a VARIABLE bitrate as default for a very good reason. That allows scenes that need higher bitrates to get higher bitrates and scenes where it isn't necessary to drop accordingly. Obviously to get a better quality image of some high action scene like at a NASCAR event with autos flying around the track needs more bitrate than 10 seconds of static white on black text.

Again I say, don't mess with the defaults UNLESS you have a reason and know what you're making a change in the internal settings for.

The real reason why some DVD players stumble aside from too high a bitrate is the type of lasers used. Once the shift is made from red to blue things should continue to improve.

My question is this: If most DVD players can't maintain the whatever to play DVDs encoded at a bitrate of 8,000 or higher, what good is it to have a DVD that has a VBR with 4,600 as the minimum, 9,300 as the average, and 9,300 as the maximum?

First, at least according to theory, ALL DVD players are supposed to be able to decode video streams at 9,800 kbps, not just at a peak, but at a constant rate. I get this information from MPEG.ORG which, at least according to its web site address, should know what it's talking about. Here's the link to the page that has the relevant information:

MPEG Video Data Specifications

Second, the whole point of having min, average, and max is for the variable bitrate encoder to, well, vary. Obviously, if you set the average to the same value as either the max or min, there is no room to vary. Thus, as you already concluded, you shouldn't do this. The way to set these values is to set the average to the value that will just make your video fit the 4.7 GB space (single layer, single sided disc), without leaving a lot of empty space, and without running over. Set the maximum to the maximum bitrate (the 9,800 kbps) minus the audio encoding rate, minus a little padding. Set the minimum to some value that is half of the average bitrate (or lower). There is no real reason to pick half -- any number that is a lot smaller will do. I have read that setting it to zero is not a good idea because some decoders (players) may balk at that. I usually set mine to 192 kbps.

The problem is that variable bitrates aren't very effective unless you do a 2-pass encode. I can't even remember the last time I did a VBR, 99.9% of the time a CBR works out just fine, if not better.

From the web site you refered to, John (thanks for that!):

... contrary to popular belief: all DVD players are required to decode video streams up to 9.8 Mbit/sec for indefinite periods of time.

I am not disputing this, however I must ask, if this is true, then why to so many players seem balk at anything over 6,500? A couple of years ago when we started making DVDs available, we encoded a disc at 8,000 (1,800 under the "required" decode stream). The customer's player would play about a third of the disc then it began to choke. They couldn't play the disc completely until I had lowered the bitrate to 6,000. So that's were I encode everything now, to ensure "compatability."

Is it just an issue with the older players?

john meyer is exactly right... any dvd player that won't keep up with a dvd-legal bitrate is defective.

i've burned and sold well over 1200 dvd-r's that had the max vbr set at 8800+... never had an issue with that setting on any player, but then, i don't use cheap crap blank discs either.

i can't see any reason for using single-pass software encoding on any product i sell.

If your video is less than about 74 minutes, there is no reason to use VBR, just crank up the bitrate to 8 mbps (or if your video is under an hour AND you're feeling lucky, you could go as high as 9 mbps.) In either case, use 192 kbps AC3 audio and have at it.

BB "The real reason why some DVD players stumble aside from too high a bitrate is the type of lasers used. Once the shift is made from red to blue things should continue to improve."

I'm not trying to start something, but where did you get this information? Could you post a link to a reputable source that backs up this claim?

The incompatibility issue is often a combination of the burner used, the media used and the player. Earlier generation DVD burners calibrated the laser once at the beginning of the burn and used that power level throughout the burn. This could cause problems on "B" grade media. New generation burners optimize the laser power level constantly throughout the burn. Using a quality brand of media is paramount. I used to use Ritek G04 4x media without any problems, but Ritek changed the formulation on the G05 8x media and, after extensive testing with mixed results, I just couldn't be comfortable using the G05 media in the Plextor-based duplicators that we use. I switched to the slightly more expensive Verbatim 8x media and have no problems. I have duplicated literally thousands of discs and have only had one return and that was from someone with a well-used, first generation DVD player.

John

I don't think we're using "cheap crap blank discs," either. We're using Ritek G4 which has a very high rating.

I'm talking G4 as well. Looking at the last 17 blanks all but the last two I have in this opened batch have some slight imperfections on the media side in exactly the same place which probably explains why they're failing now.

Cline, I will no longer respond to you since you have a long history of name calling and losing your temper. No matter what I say, you take issue with it just to take issue with it. So best for everyone for me to simply ignore you. Lets agree to disagree. OK?

Look, I said I wasn't trying to start anything, but you made a definitive statement of fact about "The real reason why some DVD players stumble " and I just wanted to know where you got that information as I had never heard that anywhere before. Maybe someone else can ask where you got your information and you might answer them. If anyone else had made that statement, I would have asked them the same question.

Nevertheless, ignore away. That would be fine.

When I was writing the Ulead DVDWS2 books last year I had some quite detailed discussions with the Ulead engineers on this topic.

I was told that one of the main reasons early players had problems with large bit rate swings (VBR) is do to the fact that the disc speed has to increase and decrease quickly in order to feed enough data to the software before the buffer ran out.

When the player fails it's not that the bit rate is too high, it's that rate of increase of the bit rate was too high for the player to keep up with physically.

Older DVDs used something called "minimum readout rate" so that the player would basically know the lowest speed if would ever need to grab data. Keeping the minimum readout rate higher would NOT allow huge swings in bandwidth and therefore disc rotation speed. Most pressing plants and authoring software still requires minimum readout rates. Commercial discs generally contain this information to help out the player while home grown recordable players do not. Hence the problems with some VBR +-R's on some players.

So, the wider the max/min bit rate the harder it will be for the player to physically adjust the disc ratation rate to keep up with the required data rate.

I was told that most newer players have much better feedback controls in this regard and this isn't so much of an issue these days.

Constant bit rate DVDs are easiest on the player because there is very little change in rotation speed as the disc plays, only that due to changing circumference as the disc plays. Of course this is a very "gentle" change that even the oldest players can keep up with.

I think a lot of people swear by CBR because it is pretty fail safe in that all players will handle it and unless you really need every last scrap of data it does a pretty good job.

That being said, if I need best quality for a given amount of data I use two-pass with max and min set about +-2Mbps. The 4Mbps "swing" is enough to spread out the data efficiently but not so much as to create readout problems for older players.

Also, for bit rate calculation I just keep in mind that 8Mbps is 1 MB/minute. I can usually do the rest in my head, or at least get a pretty close once I know the general behavior of the encoder I'm using. i.e. does it generally overshoot/undershoot?

- Mark

Mark, thanks for that explanation--very detailed and very helpful!

According to the MPEG Handbook-
Maximum bitrate is 9.8 Mb/s (but will always be less to allow for audio). The "average" bitrate is 3.5 but depends entirely on the length, quality, amount of audio, etc. This is approximately a 70:1 reduction from original 243 Mb/s CCIR 601 source. (But note that MPEG uses 4:2:0 format, which is a 75% reduction from 4:2:2 before compression)
Raw channel data is read off the disc at a constant 26.16 Mb/s. After demodulation it's down to 13.08 Mb/s. After error correction, the user data stream goes into the track buffer at a constant 11.08 Mb/s. The track buffer feeds system stream data out at a variable rate of up to 10.08 Mb/s. After system overhead, the maximum rate of combined elementary streams (audio + video + subpicture) is 9.8.
The page goes on to explain basically the same thing Mark has posted above. While the post regarding lasers and bitrates could be correct, there is nothing in the standards book on laser capability affecting bitrate, and Jim Taylor's book has nothing about lasers and bitrate, but it does have a lot of information on minimum readout rate being a big factor in holding drive read averages back. Were it laser-related, it would cause me to wonder about the commercial broadcast players that don't have these issues. The laser would have to change frequency speed if it were so, wouldn't it?
Maybe B_JM can drop in on this one.

There are many variables at play, in addition to the extremely technical aspects:

the quality of the media,
the quality/speed at which the DVD is burned,
compatibility of the player with the type/brand of DVD,
whether or not labels have been labels stuck onto them,
where the higher bitrate content physically resides on the DVD, etc.

I usually try to play it safe by never exceeding 8000 bitrate as a general rule of thumb for home-brewed DVDs.

I have not followed the whole thread - but "theForce" is correct ...

ALL dvd players HAVE to play the max. bit rate that is allowed on a dvd, or they are not allowed to be called a dvd player ....

BUT thats is for a pressed DVD, not a burned DVD ...

there are several tools (nero has one) that show how the speed and reading of a commercial dvd differers greatly from a burned dvd ... PLUS encoders have a bit of problem of going over the max sometimes (some more than others) ...

w/PCM audio I max the video at 8Mb/s
w/Dolby Digital 2.0 (at 256Kb/s) I max the vid at 9Mb/s
I've never had an issue or problem... NEVER!

I have, however, had many issues/problems burning 4X -R media in
12X -R/+R Plextor burners. (No issues with 8X -R media.)
My preferred media is presently Verbatim 8X DVD+R

B_JM is right on the money. I've had a few DVDs that certain players would baulk at and all at the same spot, during an iris down / lights out situation. Analysis reveals the bitrate on the video stream hitting well over 9 MB/sec, in other words well outside legal. But also there's a huge spike there as well so it could well have been a combination of issues.
Either way since then set max bitrate to 7.5MB / sec and haven't had a problem since.
Bob.

bob, if there is a spike in the bitrate that goes above the dvd-legal limit, it has to be the fault of the mpeg2 encoder, not the footage... the encoder is simply failing to limit the bitrate, not all encoders are bogus like that, so what you need to do is to switch encoders, rather than crippling your peak vbr limit.

taiwanese media is never as good as the best japanese media, take a look at this testing of the pioneer-based 108 drives:
http://www.dvd-recordable.org/Reviews+index-req-showcontent-id-96-page-2.phtml

notice how the reviewer clearly states at the beginning of the article, "These next tests are to compare the 12x speed burning capabilities of the two drives and for this test I am using the only media I can find that will burn at 12x on both drives. Genuine TYG02".

that is not a fluke, overall testing shows that taiyo yuden media consistently burns quicker and more reliably on most dvd-r burners.

one thing that you could say about ritek is that it's probably the best of the cheap crap media... but you can't expect to make intelligent decisions about bitrate limits and player compatibility issues based on cheapo media.

... one thing that you could say about ritek is that it's probably the best of the cheap crap media...

That's an opinion, not a fact. I've burned hundreds of Ritek DVDs without one single problem due to media. Ritek has, from everything I've read online, consistently received high marks for its performance.

We had exactly the same problems with Ritek and TY media.
Also there's a bit of a mathematical issue with how you determine bitrates from what I can see. If you look at the bitrates with say a 1 second window you get a different picture to what you get with a 100mS window.
Bob.