View Full Version : 1.02's "Dynamic Peak Analysis" ????
ron spencer
30th October 2005, 02:29
Saw the notes for 1.02 and the "Dynamic Peak Analysis" feature....is this an option or is it "always on"....saw nothing of it in the OPTIONS menu
thanks
roux
30th October 2005, 02:38
I think it's integrated, don't see a reason to turn it off.
ron spencer
30th October 2005, 03:05
don't want to turn it off....just wondering if it is an option
jdobbs
30th October 2005, 03:18
It's an improvement in the way it processes... not really an "option".
Video Dude
30th October 2005, 05:49
@jdobbs
Added "Dynamic Peak Analysis" which determines peak bitrates from the original stream and uses the info to set segment maximum bitrates.
What happens if the max_bitrate=xxxx hidden setting in the ini file is used? Will each segment get the same specified max bitrate? Or will it be dynamic, but not go over the max bitrate specified?
jdobbs
30th October 2005, 11:06
If you set the max_bitrate -- it will be honored. Whatever bitrate DVD-RB dynamically determines will not exceed that setting.
Rippraff
30th October 2005, 11:52
@jdobbs
Thanks for the new version. :)
I asked myself the same as Video Dude did. So do you recommend to delete the max_bitrate=xxxx hidden setting with the pro version?
Cu Rippraff
jdobbs
30th October 2005, 12:08
I'd leave it out and let DVD-RB determine what is the best max...
Rippraff
30th October 2005, 12:12
Thanks for the advice.
Cu Rippraff
FilipeAmadeuO
30th October 2005, 12:15
Jdobbs:
Is the Special error processing being used any more ? Could you please remove it if not.
Also there is a spelling error in audio track remapping (It is written audo).
The next request that i have is an option not to encode certain parts of the video. Is it on your to do list ?
jdobbs
30th October 2005, 12:26
Whoops. I'll fix the typo. Amazing how you don't see things sometimes.
Yes, that's on the list. I'm hoping to add and release features as I implement them leading into v1.10.
Rippraff
30th October 2005, 12:28
Another thing, are the dynamic peak analysis values stored somewhere?
Cu Rippraff
FilipeAmadeuO
30th October 2005, 12:28
And about the special error processing ?
eriksen76
30th October 2005, 12:31
Thanx for the new version. A few questions:
1) I'm still testing but I get 500-9000 in all the encoded parts, also in the part where I got stutter before.
When using the Dynamic Peak Analysis, will I be able to see that Dvdrebuilder has actually reduced the max bitrate during encode? (In the CCE encoding window)
2) What is the difference between: Dynamic Peak Analysis and the added "Bitrate Sanity Checks"?
3) Does any of the 2 above mentioned changes in this version have any influence on the rebuild step?
/Eriksen76
Rippraff
30th October 2005, 12:35
@FilipeAmadeuO
Beside the typo don't you think this is a little bit off topic? ;)
Cu Rippraff
jdobbs
30th October 2005, 12:47
1) I'm still testing but I get 500-9000 in all the encoded parts, also in the part where I got stutter before.
When using the Dynamic Peak Analysis, will I be able to see that Dvdrebuilder has actually reduced the max bitrate during encode? (In the CCE encoding window)
2) What is the difference between: Dynamic Peak Analysis and the added "Bitrate Sanity Checks"?
3) Does any of the 2 above mentioned changes in this version have any influence on the rebuild step?
1. The default max_bitrate is still 9000. As I mentioned earlier, DVD-RB will still honor the maximum bitrate.
2. Bitrate sanity checks is related to the VBR average bitrate, Dynamic peak works on the maximum.
3. No. These run during PREPARE and influence the ENCODE.
Rippraff
1st November 2005, 15:20
@jdobbs
May I ask you again, if the dynamic peak analysis values are stored somewhere?
Cu Rippraff
jdobbs
1st November 2005, 15:42
May I ask you again, if the dynamic peak analysis values are stored somewhere?
If the peak bitrate it less than the amount you have set as "max_bitrate" (or the default) -- you would see it listed as the maximum bitrate in the ECL file.
Just as a note, I found a fault in my algorithm in which the dynamic value will not be considered for non-telecined NTSC sources. So those sources will continue to work the way they always have. I've modified it for the next release.
The point at which the dynamic allocation becomes most important is with higher bitrates and lower reductions... for example, when doing movie-only encodes. When you are doing greater reduction it is less likely that the peak rates will ever reach the original peaks.
That doesn't mean it won't happen, though. The reason I added dynamic allocation was because of an interesting phenomenon you find in reencoding a quantized source through a frame server. Any errors from the original quantization also become the baseline for the new encode. So in a very high action scene the original may have "blocks", for example. When you reencode, the blocks are viewed as sharp contrasting edges and may actually increase the bitrate requirement in order to reproduce what is essentially an error in the original. In the tests I conducted I saw numerous examples of frames that took more space in the reencode than they did in the original -- even though the average bitrate was smaller. That scenario steals space from the rest of the stream... So dynamic allocation limits the peak allocation to no more than the original.
Rippraff
1st November 2005, 15:47
Thank you for this detailed explanation. :)
Cu Rippraff
eriksen76
1st November 2005, 16:25
Just as a note, I found a fault in my algorithm in which the dynamic value will not be considered for non-telecined NTSC sources. So those sources will continue to work the way they always have. I've modified it for the next release.
Is it likely that you also could have overseen a fault in the algorithm on PAL movies? The errors that I have reported are all on PAL movies.
Or did you find anything else interesting in the files I sent you?
A side note is, that my stutter problems not seem to act different on low compression compared with high compression. I think its somehow the way dvdrebuilder compiles it at the 3rd level.
The latest clip I sent you, Jdobbs is apx 62% compression. And I watched the whole dvd yesterday evening. I got about 5-6 times of stutter on the 2 episodes on the dvd. I tried encoding with a max bitrate of 7000 (same problem)
/Eriksen76
jdobbs
1st November 2005, 19:50
@eriksen76
I gotta' tell you... you are one of only a couple of people who are experiencing any stutter at all... I'll continue look at it, of course, because I want DVD-RB to be perfect, and I appreciate the additional reports you're giving me -- but there's always the possibility that it's your player that is the source of the problem, too.
eriksen76
1st November 2005, 20:08
@eriksen76
I gotta' tell you... you are one of only a couple of people who are experiencing any stutter at all... I'll continue look at it, of course, because I want DVD-RB to be perfect, and I appreciate the additional reports you're giving me -- but there's always the possibility that it's your player that is the source of the problem, too.
I know there aren't many people who get this error, but still I have to say that my 2 dvd players are not the source to the problem.
Tonight I took one of the stutter movies to one of my friends for testing purpose. It had similar stutter, but not as much as on my two players.
My parents Sony player was not affected when testing at their place the other night.
When reauthoring with Dvdmaestro, no problem on any of the players.
I understand if you're getting tired hearing about this problem that I experience quite often. Apparently it's very rare for most users.
Hopefully you're not misunderstanding me. (I don't think you did)
All I want to do is help, all I can. And give the inputs on how I think it could be solved.
Do you think it could be related to the fact that I only encode PAL movies. I believe most of the others users of this forum encode NTSC movies?
I mean is it possible that this could somehow have any influence?
/Eriksen76
Rippraff
1st November 2005, 20:18
Do you think it could be related to the fact that I only encode PAL movies. I believe most of the others users of this forum encode NTSC movies?
I don't think so, there are a lot of people in this forum living in PAL country and do not have these problems, like me.
Cu Rippraff
eriksen76
1st November 2005, 20:30
Ok, thanx for the input. It was just a thought.
/Eriksen76
Video Dude
9th November 2005, 20:52
jdobbs,
A question about the new additions:
What if you had a 3.6GB 4:3 LB movie and wanted to use DVD-RB's 4:3 LB -> 16:9 feature and wanted to bump up the size to 4.3GB. Would you have to set "OVERRIDE_BITRATE_CHECK=1" to get the 4.3GB output?
If the movie peaks at 5.8 Mb, and if max_bitrate=8500 is set in the ini, would the max_bitrate still be respected over the 5.8? In your previous post you said Dynamic Peak Analysis will not go over max_bitrate, but is the opposite also true?
Edit: Just wanted to say thanks for the 4:3 LB -> 16:9 feature. Its my favorite part of DVD-RB.
Capsbackup
10th November 2005, 18:30
I experienced some random problems a while back that I thought were caused by DVD-RB. I also tried movie only backups with DVDMaestro and it did not have the stutter at the same times as the DVD-RB backup, but on a couple backups I had random stutter even with DVDMaestro. I purchased a new Plextor 716a burner to test to see if my Sony DRU530a was the cause and sure enough I have not had a stutter since. I even went back and did a couple of the suspect movies with stutter and burned with Plextor drive and the stutter was gone. I could not believe I stuck with the Sony drive as long as I did and thought DVD-RB was the cause because of a few reported posts in this forum that I trusted. I have not had a stutter or glitch since changing burners. Thanks JDobbs for this program and your patience for trying to track down potential bugs, and I will try to do my part to make sure the problem isn't my own.
jdobbs
10th November 2005, 21:16
jdobbs,
A question about the new additions:
What if you had a 3.6GB 4:3 LB movie and wanted to use DVD-RB's 4:3 LB -> 16:9 feature and wanted to bump up the size to 4.3GB. Would you have to set "OVERRIDE_BITRATE_CHECK=1" to get the 4.3GB output?
If the movie peaks at 5.8 Mb, and if max_bitrate=8500 is set in the ini, would the max_bitrate still be respected over the 5.8? In your previous post you said Dynamic Peak Analysis will not go over max_bitrate, but is the opposite also true?
Edit: Just wanted to say thanks for the 4:3 LB -> 16:9 feature. Its my favorite part of DVD-RB.If you wanted to make the overall output larger than the original, yes, you would have to set OVERRIDE_BITRATE_CHECK. The only time I can think where you'd want to do that (rather than keep the original intact) would be for filtering like in your 4:3 -> 16:9 example.
As to you second question, no. If the original movie peaks at 5.3Mbs -- then any bitrate applied that is over 5.3 in the output would be wasted bits (you can't make the copy better than the original). In fact, what you would in most cases be doing is using a lot of extra bandwidth in order to try and repeat the errors of the original encode. So max_bitrate is just that -- it will never be exceeded... but DVD-RB will now also never peak higher than the original.
Video Dude
11th November 2005, 00:25
As to you second question, no. If the original movie peaks at 5.3Mbs -- then any bitrate applied that is over 5.3 in the output would be wasted bits (you can't make the copy better than the original). In fact, what you would in most cases be doing is using a lot of extra bandwidth in order to try and repeat the errors of the original encode. So max_bitrate is just that -- it will never be exceeded... but DVD-RB will now also never peak higher than the original.
Thanks, I understand how it works now.
The main reason I asked about this was that in the case of 4:3 LB -> 16:9, I thought the new anamorphic encode would need a higher bitrate since about 1/3 of the picture is no longer static black bars.
jdobbs
11th November 2005, 01:39
I don't think you'll find that many originals that peak as low as 5.3Mbs... most of the ones I see are in the 8 - 9Mbs range. If you do, it probably means it was a pretty easy-to-encode stream. Remember, though, that if you are doing and NTSC non-telecined encode, the bitrate listed in the ECL file is only 80% of the actual because of the common-denominator-frame-rate adjustment to accomodate hybrid sources. So if you saw a maximum bitrate of 7Mbs in one of those sources you are actually getting 8.75Mbs after reconversion at REBUILD.
vBulletin® v3.8.5, Copyright ©2000-2012, Jelsoft Enterprises Ltd.