Support forum for DG Tools

Hi.

First of all let me say that I think this is a problem with the SUP files on the disc, but maybe something can be done about it, and maybe not.

What I see is the following.

Using dgindexnv/dgdemux to demux the playlist and then muxing them into mkv file, mediainfo claims that the subtitles are more than 13 hours long.

Using tsmuxer to demux the playlist and then muxing into mkv file, mediainfo says the same thing. 13+ hours.

Using eac3to to demux the playlist and then muxing into mkv file, mediainfo claims the correct length, in this case 1h37m.

This is a Scandinavian disk, so I am not sure that you will be able to buy it. Can you use just the demuxed SUP files for anything?

The playback of the resulting file is just fine. All subs are in sync, and the player itself is reporting the correct length. Just mediainfo is reporting 13+ hours length (could be an issue with mediainfo as well).

Not sure where it goes wrong. If you can't use the demuxed SUP files themselves is there anything I can do to test something?

renols

Yes, please provide the SUP files. I can do a binary compare to try to find the cause.

Hi.

I have sent the SUP files from dgdemux, eac3to and tsmuxer to the FTP.

I have binary compared the dgindexnv and dgdemux files, and they were identical as expected.

renols

Thank you. Investigating...

OK, pretty sure I know what is going on. Please retry after I release the next build of DGDemux. Then I'll port it to DGIndexNV.

Please try DGDemux/DGDemuxGUI slipstream 38.

Hi.

Sorry about delay. Just got home.

It is defenately different now.

Tested with dgdemux 38.

This is what mediainfo shows under "General" without the sup file muxed in:

Format : Matroska
Format version : Version 4
File size : 4.36 GiB
Duration : 1 h 37 min
Overall bit rate : 6 387 kb/s
Encoded date : UTC 2020-08-19 21:50:01
Writing application : mkvmerge v40.0.0 ('Old Town Road + Pony') 64-bit
Writing library : libebml v1.3.9 + libmatroska v1.5.2

This is what mediainfo shows when the sup file is muxed in:

Format : Matroska
Format version : Version 4
File size : 4.36 GiB
Duration : 1 h 46 min <- here is said 13+ hours before. Not sure where the extra 10 minutes comes from now
Overall bit rate : 5 836 kb/s
Encoded date : UTC 2020-08-19 21:51:40
Writing application : mkvmerge v40.0.0 ('Old Town Road + Pony') 64-bit
Writing library : libebml v1.3.9 + libmatroska v1.5.2

ID : 4
Format : PGS
Muxing mode : zlib
Codec ID : S_HDMV/PGS
Codec ID/Info : Picture based subtitle format used on BDs/HD-DVDs
Duration : 1 h 36 min <- here it also said 13+ hours but now shows correct length
Bit rate : 12.6 kb/s
Count of elements : 1189

I am not sure where the extra 10 minutes in length under General comes from. The only difference is with or without 1 sup file muxed in.

Btw, is there no progress indicator in dgdemux? The green bar at bottom left just moves forth and back in the window, but one have no feeling of wether it will take 5 or 10 minutes for the job to complete.

If you need a dump of the entire disc I can send it to the ftp. I have the physical disc here and have bought it. As I said it is a scandinavian bluray, so it is probably not possible for you to buy it. The .us disc will probably be mastered totally different.

renols

PM coming your way. Thank you for your testing.

I've been just muxing in only the SUP file for testing. Can you do that and tell me if my files act differently than EAC3TO files?

Full progress will come back but it's not high priority right now. Expect a disk to take 5-10 minutes and be patient.

Hi.

When I just add the sup file and create a .mks file with mkvmerge it looks exactly the same in mediainfo wether I user the sup file from dgdemux 38 or the sup file from eac3to. I see absolutely no difference here. With the old sup file in the mks file is says 13h+.

But as soon as I mux the sup file from dgdemux 38 together with video and audio, the time in the General section in Mediainfo says 1h 46m, and not 1h 37m.

If I use the sup file from eac3to the same time under General shows 1h 37m.

Really strange.

About progress bar. Roger that. Not a big deal.

renols

That is indeed strange. I'll do some tests with disks that on hand and see if I can replicate that. I will also try to purchase the disk. Thank you for your report, renols.

Rocky wrote: ↑

Thu Aug 20, 2020 5:27 am

That is indeed strange. I'll do some tests with disks that on hand and see if I can replicate that. I will also try to purchase the disk. Thank you for your report, renols.

No problem.

Just happy to help make the software even better and more sturdy to "weird" discs.

Seems like every time we believe that we have it in the bag, some disc appears that for some reason creates new issues

renols

Can you re-download and test build 38? I made a small change that may be relevant. Anyway, using the same disk I muxed with SUP alone, vid+DTS+SUP, and vid+AC3+SUP. All show 1:37 with this build now. I have mediainfo 20.03.

Hi Rocky.

I can confirm that the 10 minute difference in General section is now gone, and all now show the saem as if I use the eac3to sup files.

Were there something weird with the sup files on this specific disc?

renols

Thank you for the testing and happy to hear things are OK.

The disk while a bit strange is not illegal. It's a bit complex to describe and I am not done with it quite yet as I want to understand why my timings are a little off from EAC3TO (few milliseconds and inconsequential). When I have it fully understood I'll expound. I am thinking that EAC3TO is offsetting the times by the IN_time while I offset by the first video PTS that I see. Things are fine either way so far. More later...