Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion. Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules. |
21st May 2003, 16:43 | #361 | Link |
Registered User
Join Date: Feb 2002
Posts: 1,195
|
hi,
for sure this option will be added i think i will try to publish a new beta in few days (my exams will be finished soon ) Bye.
__________________
AutoDub v1.8 : Divx3/4/5 & Xvid Video codec and .OGG/.MP3/.AC3/.WMA audio codec. AutoRV10 v1.0 : Use RealVideo 10 Codec and support 2 Audio Streams and Subtitles. |
23rd May 2003, 14:46 | #363 | Link |
Registered User
Join Date: Apr 2002
Posts: 756
|
The new Producer 9.2 M3 is already out. I have read the whole page and spec but i still can not find anything about this new high quality mode. Can somebody point me to the exact place to read about it.
And dark Cracker, where can i get a 1.3 beta version?? |
23rd May 2003, 15:47 | #365 | Link |
retired developer
Join Date: Oct 2002
Location: Canada
Posts: 8,978
|
@Karl
Can you explain what exactly is that HQ mode? @All The lastest mplayer can now play RM files, but the seeking doesn't work wet.
__________________
Detritus Software |
23rd May 2003, 17:20 | #366 | Link |
Moderator
Join Date: Oct 2002
Location: Seattle
Posts: 1,584
|
yes, I will let you know when it is available, which will take a couple of weeks, I am afraid. I just spoke with Producer team, and the next Milestone, M4, should be out in about two weeks. It is not this this feature , which I am pretty much done with, that's holding it up, but other work.
I think there will be three new modes, in addition to what has been default until now, in lack of better terms, let's just call them: Default(0): improved efficiency for high action High(1) : same as (0) + better mode decisions, better representation of high motion Higher(2) : same as (1) + very high accuracy motion representation Highest(3): same as (2) + best possible mode decisions as you can see, there is some improvement even for the default mode. (0) same encoding time as before (1) will take about 2X encoding time compared to default (2) about 2.8X default (3) about 3.5X default The bitstreams will be fully backwards compatible, decoding speed will not be affected. As always, the improvement will depend on content and bitrate, perhaps in some cases hardly noticable, while some cases, quite impressive. More later.
__________________
This information is provided "AS IS" with no warranties, grants no rights, and reflects my personal opinion. |
23rd May 2003, 17:38 | #367 | Link |
Registered User
Join Date: Apr 2002
Posts: 756
|
sorry........... i just reliaze something.... and that is espcially with ainme. Is that sometimes don't change much or even not changing at all for two or 3 secounds. So low framerate is good in this scence. I understand that Rv9 already support variable frame rate. But this apply to the video automatically? And if it is possible have Encoder to decide at what time should the frame rate be...( i.e User defined frame rate @ special time frame.)
And one thing that is good with this new HQ Mode. I have always think that it doesn't really matter if it take a 3+ Ghz and loads of ram to encode to a specfic format. As long as the user are able to watch the end file with a low end system. In this case RV9 has that. And i like it. I hope there is more options and enhancement to be made to the encoder that improve quality ( with higher resources / power to encode ) while staying at the same level amount of resources to decode. |
23rd May 2003, 18:17 | #368 | Link |
retired developer
Join Date: Oct 2002
Location: Canada
Posts: 8,978
|
@Karl
That's great news. I have a little question: What's the commandline option for audio normalization (not gain) ? You said we could use producer instead of besweet for normalization.
__________________
Detritus Software Last edited by Sirber; 23rd May 2003 at 22:24. |
24th May 2003, 05:22 | #369 | Link |
Moderator
Join Date: Oct 2002
Location: Seattle
Posts: 1,584
|
i am afraid producer does not include normalization, even though this would have been possible with 2-pass encoding. Considering how you can now (or very soon) use AC3 tracks as audio source for producer, normalization would have been useful. However, I do suspect the variation in the gain found by for instance Besweet's normalization feature, is not too great, and based on some experiments, you can use about the same 'dB' number for the audio gain in producer all the time, withou actual normalization. Producer includes a very high quality limiter, in case your gain is a little too high, so this should be no problem, rather an advantage.
__________________
This information is provided "AS IS" with no warranties, grants no rights, and reflects my personal opinion. |
24th May 2003, 15:08 | #370 | Link |
retired developer
Join Date: Oct 2002
Location: Canada
Posts: 8,978
|
SO I can ask a 12db gain, and producer will "control" the gain to not go beyond the maximum?
Sorry for my english, I just woke up
__________________
Detritus Software |
24th May 2003, 15:39 | #371 | Link |
Registered User
Join Date: Feb 2002
Posts: 280
|
@Karl,
Is there any plan that Real will provide some type of frame capture option like that in BSplayer or Vdub that can export frame/frames to either bmp/png/jpg? The work around with hardware is getting quite old As for quality wise, current 9M3 is doing just fine; However, if 9M4 is promissing to be better, then all the best! |
24th May 2003, 16:47 | #372 | Link |
! - User - !
Join Date: Nov 2001
Posts: 1,081
|
A subtitle question:
The subtitles displayed with smil files are not part of the hardware overlay, and consequentially are not displayed on TV when using matrox' Dvd max feature. Is there any possibility to work around this?
__________________
Keep your tone warm and your sigs decent! |
24th May 2003, 16:48 | #373 | Link |
retired developer
Join Date: Oct 2002
Location: Canada
Posts: 8,978
|
9.2M4 will kick ass. I saw a preview and it's SO GREAT!!!
__________________
Detritus Software |
28th May 2003, 17:40 | #375 | Link |
Moderator
Join Date: Oct 2002
Location: Seattle
Posts: 1,584
|
EHQ update
Just to let you know the status for the new Extra High Quality mode. I am still struggling with a very hard to catch bug. It is so non-reproducible, I can't tell how long it will take either.
__________________
This information is provided "AS IS" with no warranties, grants no rights, and reflects my personal opinion. |
28th May 2003, 23:14 | #376 | Link |
retired developer
Join Date: Oct 2002
Location: Canada
Posts: 8,978
|
If it's non-reproductible and hard to catch bug, may we have an alpha test?
__________________
Detritus Software |
29th May 2003, 06:48 | #377 | Link |
Moderator
Join Date: Oct 2002
Location: Seattle
Posts: 1,584
|
hey, maybe that's a good idea, considering how until now I have done all the development, optimization, testing and QA all by my lonely self, for this EHQ mode. I could use some help with the testing and QA part . I did however make great progress today, so hopefully it should be resolved soon.
__________________
This information is provided "AS IS" with no warranties, grants no rights, and reflects my personal opinion. |
29th May 2003, 22:18 | #379 | Link |
retired developer
Join Date: Oct 2002
Location: Canada
Posts: 8,978
|
Yeah!
Send it to hell!! Can't wait for the next built
__________________
Detritus Software |
31st May 2003, 02:07 | #380 | Link |
Moderator
Join Date: Oct 2002
Location: Seattle
Posts: 1,584
|
bug ( ) terminated! a misfit compiler setting led to a missing __chkstk() call in a P4 specific function, and the stack mangling that would then sometimes occur, resulted in these instances of sudden death.
The next Producer build that passes BVT should include an EHQ mode that I am quite happy with.. For certain content a quality improvement corresponding to a 30% bitrate reduction can be seen, and this is with a fully backwards compatible bitstream. On average though, and for very high bitrate content, it will be less, but for my current test encodings I can reduce the bits/pixel number quite significantly, and have the result still look very pleasing. It will be very interesting to get some feedback. Remember though, every improvement is incremental, and it takes some time to adjust to the new mode. For instance, if you previously used such a high bitrate relative to resolution, that no compression artifacts could be seen, using the same bitrate now, you will still see no visual degradation. The improvement then would have been that you could have gotten away with a lower bitrate and still have seen no problems. I know you will ask if it can be made faster, and the answer is that every CPU cycle is spent trying to find more optimal compression parameters, and every CPU intensive function is fully MMX/SSE/SSE2 optimized. I would rather have my computer's 2.4 GHz P4 do that instead of searching for aliens in its free time Will the improvement be worth the slower encoding speed? Well, certainly for me it will, but if your CPU is on the slow side, it may be too much. I have used it enough now to see the improved visual quality. I can now lower the bitrate more without seeing "stuff going on" both in the background and around edges. Of course, being the implementor, it is not possible to get more biased, but I am very happy with the result. More later. EDIT (typo only)
__________________
This information is provided "AS IS" with no warranties, grants no rights, and reflects my personal opinion. Last edited by karl_lillevold; 31st May 2003 at 06:24. |
Thread Tools | Search this Thread |
Display Modes | |
|
|