tsMuxeR Blu-ray output and Fixclpi

[laserfan]

Thanks again deank, I was thinking not about FC-type compare (or WinMerge) but rather comparing SEI/SPS/PPS flags, but you answer part of this too ("x264 produces elementary stream and it does not contain anything else but the video itself - no SPS/PPS/SEI.") So I guess my remaining ? might be how to see in the file structure the effects of tsMuxeR's insertSEI, contSPS which is almost back where we started w/the OP asking what fixclpi does exactly.

Hey deank can you write a better blu-ray muxer maybe than smlabs' Roman!?

Re: chapters I was referring simply to xx:xx:xx.xxx timecode specifications to the --custom-chapters option. The tools from Donald Graft like DGAVCDecNV here always seem to ID timecodes by frame exactly right!

[setarip_old]

@quantum

Hi!

I just tested PowerDVD 7.3 and played some test Blu-ray muxes (as Blu-ray disks using the "open movie file on hard disk" feature).

Forgive my all-too-obvious ignorance regarding this procedure, but since I've had no success doing this myself, so I'll ask - when using "open movie file on hard disk" for BluRay hard drive "packages", what file or folder of the "package" do you click on in order for the file selector to make its "OK" radiobutton available?

[quantum]

Originally Posted by setarip_old

...when using "open movie file on hard disk" for BluRay hard drive "packages", what file or folder of the "package" do you click on in order for the file selector to make its "OK" radiobutton available?

Either the folder above BDMV, or the BDMV folder. However, I've heard that this functionality may have been removed in later versions. I'm using 7.3. Some complex Blu-ray disks may not have started for me, but the simple movie-only muxes from tsMuxer always work fine.

[setarip_old]

@quantum

Thanks for the information ;>}

**EDIT** Strange, it only works for standard DVD "packages" on the hard drive, not BluRay "packages" (I also have v.7.3)

[laserfan]

Originally Posted by setarip_old

[B]**EDIT** Strange, it only works for standard DVD "packages" on the hard drive, not BluRay "packages" (I also have v.7.3) [/COLOR]

Should work, I have same version as you & quantum and it works well for everything.

[setarip_old]

@laserfan

Should work

Would that it were so. Just another anomaly in the world after punchcards ;>}

[anode]

I have been analysing the output *clpi from tsMuxeR 1.8.8b for my own. This files contain a table with PTS (timecode) to SPN (source packet number in m2ts-file) relationship for all I-IDR(?)-frames.
So if you skip (chapters) or ff/rw these tables are used to determinate where is the location of the next i-frame to start decoding.
tsMuxeR generates these tables wrong. The tables have first a COARSE section, and second a FINE section. The full PTS and SPN Numbers are divided in this coarse and fine tables. If the fine entry of PTS and/or SPN rolls over, you have to make a new coarse entry. tsMuxeR does this for PTS, but not for SPN.

djobbs fixclpi fixes this, but as far i can see, introduces another error: As PTS fine and coarse entries share one bit together, you have to generate a new coarse entry for PTS when the 2nd most significant bit of the fine PTS rolls over, since the most significant bit of the finePTS is the least significant bit of the coarsePTS (shared). So here should a new coarse entry be created and fixclpi does not do this.
(Sorry for this technical stuff, maybe it is of use for somebody)

I am trying to fix this for myself to see if seeking/ff/rw behaviour is changing. Weren't there some Panasonic Players who reject discs with fixclpi'ed-files?
Has somebody researched this topic,too?

[laserfan]

Well, of course jdobbs has looked hard at clpi files and if you are correct about fixclpi you should somehow direct him to this thread. Here are a couple of his posts on this:

http://forum.doom9.org/showthread.ph...05#post1223405

http://forum.doom9.org/showthread.ph...92#post1207492

I'm guessing the latter thread would surely get his attention, but maybe instead you could PM him and direct him here in case he misses your post.

[rack04]

Originally Posted by anode

I am trying to fix this for myself to see if seeking/ff/rw behaviour is changing. Weren't there some Panasonic Players who reject discs with fixclpi'ed-files?

Yes, the Panasonic BD-35 for example.

[nwg]

What is meant by bad ff/rw? I have done some Blu Rays and the Sony S350 is not ff/rw very well. It does work for a few minutes then freezes the image until I press play again. This is with fixclpi used and add picturing timing and insert sps is greyed out so I don't know how to select them

[quantum]

Originally Posted by anode

...significant bit of the finePTS is the least significant bit of the coarsePTS (shared). So here should a new coarse entry be created and fixclpi does not do this.
(Sorry for this technical stuff, maybe it is of use for somebody)

I'm grateful that you are taking an interest in this, and even better, that you have the technical knowledge of the Blu-ray format. If I can offer any help at all (especially as a programmer), I'd be glad to be of service.

[jdobbs]

Originally Posted by anode

I have been analysing the output *clpi from tsMuxeR 1.8.8b for my own. This files contain a table with PTS (timecode) to SPN (source packet number in m2ts-file) relationship for all I-IDR(?)-frames.
So if you skip (chapters) or ff/rw these tables are used to determinate where is the location of the next i-frame to start decoding.
tsMuxeR generates these tables wrong. The tables have first a COARSE section, and second a FINE section. The full PTS and SPN Numbers are divided in this coarse and fine tables. If the fine entry of PTS and/or SPN rolls over, you have to make a new coarse entry. tsMuxeR does this for PTS, but not for SPN.

djobbs fixclpi fixes this, but as far i can see, introduces another error: As PTS fine and coarse entries share one bit together, you have to generate a new coarse entry for PTS when the 2nd most significant bit of the fine PTS rolls over, since the most significant bit of the finePTS is the least significant bit of the coarsePTS (shared). So here should a new coarse entry be created and fixclpi does not do this.
(Sorry for this technical stuff, maybe it is of use for somebody)

I am trying to fix this for myself to see if seeking/ff/rw behaviour is changing. Weren't there some Panasonic Players who reject discs with fixclpi'ed-files?
Has somebody researched this topic,too?

Hmmm... I thought that was how I did it, and it was done to fix exactly the situation you are describing. Have you dumped it out and seen this on "fixed" clips?

I'll go back and look. Thanks for the hint... I was doing a logical OR of the COARSE/FINE portions together when I was testing the output, so I may have missed it. It definitely fixed the FF/RW issues I had -- but maybe my player's firmware doesn't notice this because it works the same way I did. I even went so far as to put a "don't fix the clip" option in BD Rebuilder so the Panasonics could still use it.

[laserfan]

Originally Posted by jdobbs

I'll go back and look.

Thanks for checking. I hope anode comes back here and is not just a "one-post wonder"!

Your fixclpi repairs chapter-seek problems with 2 out of 3 playback options I have (a stb, TMT, and PDVD). But with PDVD I still sometimes have odd seek problems and I dunno if it is PDVD or maybe this thing anode suspects in the clpis (or a different tsMuxeR bug for that matter).

[jdobbs]

He's right. I was checking for rollover on the MSB of the fine portion of the PTS, when I should have been looking for change in the coarse part or rollover on the 2nd MSB... I've fixed it and am testing it. I also added a check that would correct any previously "fixed" CLPI files.

[quantum]

Let me be the first to say THANK YOU jdobbs and anode

Also I'll be ready to test the new update as soon as it comes out on my Standalone and on PowerDVD.

[jdobbs]

Well, not yet... I adjusted it so the it creates a COARSE entry when the 2nd MSB rolls over. But that caused the output FF/RW and chapter jumping to fail completely (with picture freeze). What anode says make sense, but my Sony SAP certainly doesn't like it. Still experimenting/testing. I may be missing something.

[jdobbs]

It turned out to be the source I was working with. You can now download a new version of FIXCLPI (v2.1) from this link.

I'm hoping this one fixes the previously reported issues with Panasonic players. I'd appreciate it if someone would report back on that.

I added a new mode in which if you pass the program a directory name (rather than a file name) it will correct any CLPI files it finds within that directory tree. That should save a little time.

I tested it on a couple of sources and it appears to work correctly -- but like all freeware, there are no assurances made.

My thanks to anode for pointing out the issue.

[MadMonkey57]

Sounds promising !!! Thanks anode and jdobbs.
I'll report to let you know what my Pana BD35 thinks about it...

[rack04]

Burning a disc now to test on my BD35. Will report back soon. Thanks jdobbs.

[jdobbs]

Just updated it again, download from this link. The folder update option was only updating the first CLPI in each subfolder. It will now do all of them. I also updated the link above, so it will also get this version.