HDTVtoMPEG2 was another example of something that almost died, but the source was released and someone else picked up the torch and took over.

I am having an issue with MPEG2REPAIR on some machines in the last 7 or so days and wondering if anyone else has seen anything.

First I should point out this does not happen on all machines.

When I start MPEG2REPAIR to find the PIDs, on 3 machines it is taking forever (ok, 30-45 seconds) to find the PIDS as opposed to a few seconds.

As its not doing it on all machines, its clearly not a time bomb.

I have hit the machines with heavy duty virus checkers (not the crappy ones out there) and worm detectors - and nothing is out of the ordinary.

I have looked at the Task Bar Processes - it appears it is the System driving this very hard - around 50% for 30-45 seconds as it finds the PIDs.

Once I ask it to check a file, the System uses virtually no resources again - but it does seem it is taking longer to complete.

Any ideas as to what is going on?

I thought that perhaps on these machines that I have captured so many large files, trimmed them and moved them off that possibly the Raid 0 hard drives are just fragmented to no end - and Diskeeper confirms they are almost literally 100% fragmented - and since the files are so large there is really very little way to defragment.

The System files and apps on a 20 Gig C:\ are fully defragmented so that is not the issue - its the D:\ Data drive on the Raid 0 that is about 1TB and fully fragmented.

I will probably move everything off and do a total format on D:\ to see if that cleans it up, but I am curious if I am missing something - or if anyone else has any ideas?

It's just very strange how it started on 3 machines about the same time after a period of time.

Also, fwiw, I did install IE7 about 2 or so weeks ago on the machines - and we know that is a memory hog if you open a number of tabs - but this happens without IE7 open - and looking at the memory use on the task manager, that is well below the amount of RAM I have in the machines (1G - 1.5G in each).

Mark not everything is available to buy, i suppose your refering mainly to films.

It will be a big loss to all M2R users if you decide to end with M2R.

Yes the sorce codes will be invaluable for someone with expert knowledge. To carry on with M2R.

Mark did you have any further communication with DanR (support at videoredo dot com). Or did it all end with you were to busy with other projects. Maybe you could have another chat with DanR, i'm sure he wil be happy to talk to you again. Whether M2R or no M2R, just to give him some pointers. It may sound a bit cheeky me asking that but vrd still cannot see the errors M2R knows about.

M2R is way above many other softwares for finding errors, or could it be it is on the same plain as vrd. Each knowing different errors and how to find them. Both are the best softwares for what each does.

Either way as long as the M2R engine continues with another expert brain coding it. Keeping it alive with updates and new features, i will be very happy.

I can only hope to see M2R and HDTVToMPEG2 merged (maybe a command line feature, etc.).

One of my favorite programs and the most frequently used on my list.

I would love to contribute some $$ through paypal. Where should I send the cash?

I can only hope to see M2R and HDTVToMPEG2 merged (maybe a command line feature, etc.).
If either could handle MPEG2 PS it would be very helpful.

Either M2R to analyze or HDTVtoMPEG to load, and convert them to TS. In the original stream condition as loaded, then able to analyze them with M2R to find all errors.

I know of no program that can convert without modifying the MPEG2 Program Stream structure. Thus making it useless as M2R will not see all the errors in the stream. HDTVtoMPEG2 says it can convert streams without modifaction, alas it cannot convert PS to TS.

Is there a PS to TS converter that can convert, without stream modification ?

Is there a PS to TS converter that can convert, without stream modification ?

VideoRedoPlus.

one main thing I'm missing is cleaning out the timestamps on TS files. Particularly errors caused by removals of commercials...

isn't that what VRD QSF does if you set the secret option for adding timestamps to GOPs?

VideoRedoPlus.
No it modifies the streams, more so it modifies any stream passed through it. In any output or tool output save, Videoredo is useless for this purpose.

No it modifies the streams, more so it modifies any stream passed through it. In any output or tool output save, Videoredo is useless for this purpose.

VideoRedo redoes all the time stamps, and removes or insert frames to insure audio sync to compensate for lost frames due to reception glitches. It also remuxes properly TS files.

But it does not change in anyway the video encoding. The errors reported by M2R remain exactly the same with a file processed with VR.

VideoRedo redoes all the time stamps, and removes or insert frames to insure audio sync to compensate for lost frames due to reception glitches.
It is the glitches data besides other data's that needs to be retained, to find all the errors in a converted PS to TS file.

If only M2R would analyze PS streams.
I never use M2R for error fixing, just need the PS stream to be analysed with a .log file output. Of all errors it finds in a PS stream, as it does for TS streams now.


Videoredo i haveve tested it with with all settings inc advanced switched off, no joy.

Have spoken to Dan about this he says it cannot be changed as hard coded into the qsf engine. It was why Dan and Mark talked about possibilty of M2R intregration into vrd. Mark declined saying would be too much work, to just get the analysis part into vrd, another engine hard coded. Maybe Mark again could take up talking with Dan @ vrd about the same possibilities as before....

If vrd did leave the streams totaly alone for when i convert PS to TS i would be very, very happy. Am still looking around for the best PS to TS converting sollution. Out of six devices four record TS and only two only can record PS only. Not a software problem but hardware and driver keeping them to PS recording. I do like these devices it is why i still look for PS to TS converter. One that will not modify the stream during converting to TS.

It is the glitches data besides other data's that needs to be retained, to find all the errors in a converted PS to TS file.

Some time ago I posted that I could not write repaired files back to my sony hc1 camera. I think I know why. the Sony uses hdv mode hd2. This has the following streams:

Video 0x810
Audio 0x814
Auxillary Video 0x811
Auxillary Video 0x815

Using tsreader I found that I could record with the auxillary video stream.

Does anyone know of a utility to transform the ts files produced by mpeg2repair
into hdv-hd2 transport streams.

Shame its only single-threadded. I've been looking (not intensly) for one that can merge TS files that is multi-threadded.

Some time ago I posted that I could not write repaired files back to my sony hc1 camera. I think I know why. the Sony uses hdv mode hd2. This has the following streams:

Video 0x810
Audio 0x814
Auxillary Video 0x811
Auxillary Video 0x815

Using tsreader I found that I could record with the auxillary video stream.

Does anyone know of a utility to transform the ts files produced by mpeg2repair
into hdv-hd2 transport streams.

I am not sure what kind of conversion you need, but if it to just change the PID numbers, then VideoRedo can do this and probably HDTVtoMPEG2 can also change the PID from the 11,14 that M2R uses.

Mike T

If you can input your raw hdv-hd2 file into M2R, just don't repair it. Use the log mode only, and repair the files elsewhere if needed.

Hqas anyone ever loaded demuxed elementary video stream into m2r.

If you have done what was the extension of the stream.

I could then demux use the video stream only, to scan for all errors. Then edit the original file afterwards. I don't have find sound errors enabled in the .ini file normaly, sound not essential. I find every sound error has a video error associated with it 99.9% of the time, or if not always.

A mind is a terrible thing to waste.

It sure is when you type something that doesn't make sense. Can you enlighten us to what you're refering to.

It sure is when you type something that doesn't make sense

Exactly!!!!!!

IOW, he's referring to the Nexin post immediately above his first remark. Perhaps you could go back and see if you can read it yourself. When you get it figured out, please enlighten the rest of us by editing it into something resembling english. ;)

Ok i've Found This (http://www.avsforum.com/avs-vb/showthread.php?p=4513921&&#post4513921) not exactly above my last post.

Next question, has anyone got this early M2R version that i can download. Or even know the exact version of M2R that can do elementary streams. There is no mention in the changelog.txt i can find.

Also i'm needing a tool, that i can be used to demux PS-SD to elementary streams. Without any stream modification during demuxing - RAW > RAW demuxed.

Is there a command line interface to M2R ?

So I guess there isn't a multi-threadded program that can *properly* merge .TS files together like MPEG2Repair can.

I have been able to capture hdv transport streams from my Sony HC1 using
capdvhs. I then repair these using mpeg2repair which makes the non hdv
complient. I have found very little info on the net about hdv streams but
I have discovered the following streams using tsreaderlite.

Pid
0x0 PAT
0x1f PMT Network
0x81 PMT Program 100
0x810 ES (video)
0x814 ES (audio)
0x811 ES (Private data)
0x815 ES (Private data)
0x134 PCR

A normal transport stream as produced by mpegrepair and other ts
programs has

Pid
0x0 PAT
0x10 PMT Program 1
0x11 ES (video)
0x14 ES (audio)
0x11 PCR

I have written a program to try and convert the normal stream to hdv
stream but it does not record properly on my hc1.

Does any one know of a program which will convert standard mpeg2 ts streams
to hdv streams without rendering the video?. Most things I have tried re render
the video with a loss in quality.

I am not sure what the 0x811 and 0x815 streams are. They may be auxillary
video or audio?

Often the PCR data is embeded in the video stream but hdv has a seperate
stream. I know the format of the packets but I am not sure what values I
should use. The format is detailed in the iso13818-1 spec which I found
on the net.

If anyone has any details of the HDV spec I would be grateful for the info.

I know this is fairly technical are there any other forums I could post this
to.

Thanks

What are you trying to accomplish; edit HDV files or convert other HDTV programs to HDV? Software is most likely available for editing. Converting would require transcoding to the higher data rate transport stream used in the HDV format.
John

I have a .ts file (1080i) that Mpeg2Repair could not find an error, when there were 2 consecutive audio dropouts. I re-ran the file, and still, no error detected. I played the file in 3 different programs, and the audio drops out at the same place. My DD 5.1 receiver flashes a light when the stream is interrupted, so I know there is someting array in the stream.

Would you care to have a look at a small portion of the file?

Thanks, and great program by the way!

I suspect the audio dropout was at the source so it would not be detected.

I suspect the audio dropout was at the source so it would not be detected.

At the source? Please explain. It is dropping out at the exact same spot, when played with various software.

At the source? Please explain. It is dropping out at the exact same spot, when played with various software.

If a tech at HBO kills the audio feed switching routers, the audio drops out on the feed to their encoder - so on your end the audio channel was there - there just wasn't anything on it - so it remains a constant time stamp on the output - thus mpeg2repair would not sense any issue as it was on the original feed that way.

And no matter what you play it with - if there is no audio - there is no audio to be played back.

Let me try it this way:

Look at these 2 logs.....same showing of movie from HBO at 8pm on 12/2/2006.

HBO is HBO-HD on Dish

HBOC is HBO-HD on C Band off 127W


MPEG2Repair: D:\MPEG2REPAIR\A History of Violence 1080iT AC3 HBO_01.ts


Sequence Frame 99778(96-I) / Time 1:09:00 :
VideoError: MPEG2 Intra DCT coefficient index out of bounds. MBA=4717(592,624)
VideoError: Failed to decode macroblock at MBA=4717(592,624)
FileInfo: Last video errors span 1511 bytes at file offset 5534641803

Sequence Frame 105913(85-B) / Time 1:13:16 :
VideoError: Skipped macroblocks following intra coded macroblock on B-Picture. MBA=5796(576,768)
FileInfo: Last video errors span 581 bytes at file offset 5820528741

Sequence Frame 138662(65-B) / Time 1:35:54 :
Info: End of MPEG2 sequence

Sequence Summary:

File Size Processed: 6.80 GB, Play Time: 01h:35m:54s
1920 x 1080, 29.97 fps (24.10 fps Telecine), 18.00 Mbps (9.48 Mbps Average).
Average Video Quality: 48.00 KB/Frame, 0.19 Bits/Pixel.
AC3 Audio: 3/2 Channels (L, C, R, SL, SR) + LFE, 48.0 kHz, 384 kbps.
Dialog Normalization: -27.0 dB, Center Mix Level: -3.0 dB, Surround Mix Level: -3.0 dB
2 of 138662 video frames found with errors.
0 of 179826 audio frames found with errors.
2091 corrupted video bytes in file.
0.000000 seconds of video timestamp gaps.
0.000000 seconds of audio timestamp gaps.

End of Log



MPEG2Repair: D:\MPEG2Repair\A History of Violence 1080iT AC3 HBOC_01.ts


Sequence Frame 99778(989-I) / Time 1:09:00 :
VideoError: MPEG2 Intra DCT coefficient index out of bounds. MBA=4717(592,624)
VideoError: Failed to decode macroblock at MBA=4717(592,624)
FileInfo: Last video errors span 1511 bytes at file offset 5532201522

Sequence Frame 105913(978-B) / Time 1:13:16 :
VideoError: Skipped macroblocks following intra coded macroblock on B-Picture. MBA=5796(576,768)
FileInfo: Last video errors span 580 bytes at file offset 5818037705



Sequence Frame 138662(958-B) / Time 1:35:54 :
Info: End of MPEG2 sequence

Sequence Summary:

File Size Processed: 6.80 GB, Play Time: 01h:35m:54s
1920 x 1080, 29.97 fps (24.10 fps Telecine), 14.25 Mbps (9.48 Mbps Average).
Average Video Quality: 48.00 KB/Frame, 0.19 Bits/Pixel.
AC3 Audio: 3/2 Channels (L, C, R, SL, SR) + LFE, 48.0 kHz, 384 kbps.
Dialog Normalization: -27.0 dB, Center Mix Level: -3.0 dB, Surround Mix Level: -3.0 dB
2 of 138662 video frames found with errors.
0 of 179826 audio frames found with errors.
2090 corrupted video bytes in file.
0.000000 seconds of video timestamp gaps.
0.000000 seconds of audio timestamp gaps.

End of Log


Same Movie from HBO-HD - Errors at same exact time on C Band Distribution and on Dish HBO-HD at the same time.

No matter how you watched the movie on this airing - whether Cable, Satellite or FIOS - all had the same error in the same place as HBO had a very rare transmission error in their uplink.

Now think about it the opposite way. If they killed the audio at the source for a split second, the audio carrier out of the MPEG2 Encoder would still be there - just nothing on it - so it would silence (no 0s or 1s) and thus MPEG2REPAIR would not detect an error.

Right, and as I indicated earlier, my DD5.1 reciever has a light on it, and it flashes on and off if there is an interruption to a stream. It was my own assumption that when I saw my DD5.1 light flash, there is some sort of stream problem, as in the digital signal is null, or with errors.

again, the audio track can be there with no audio - mpeg2repair essentially detects a damaged AC3 frame (it wasn't damaged) or a gap in time in the frame (there was no gap) - the track was there - and the track contained silence.

MPEG2REPAIR detects errors - not silence.

When there is a sequence of different episodes made on the same day.
Each File have have extensions like 01, 02, 03.

How to output each file with numbered extension to a single seperate log file. As it is now, m2r will combine them into one big log file.

What is the ini file setting needed to adjust for this. So i can accept the ok requester and have a one click action for this. For m2r to output a new output log file, for each numbered sequence loaded file.

Hi folks,

I appears I've just hit the time bomb in the last available version. I downloaded again just to check, but the version Mark has available for download is bombing immediately too.

First, mondo thanks go out for having written this tool - I love it. In Australia, we have a host of awesome digital TV and with MPEG2Repair in my armoury, I know I'm only getting the flawless stuff to archive. While this has been rated as beta software, I've never had anything but release quality results from it.

With respect to Mark's loss of interest in needing the program, I'm just wondering if anyone knows any more about a non time-bombed version (or if you have any further thoughts, Mark, if you're reading). I love it exactly as it is - I just need it to not bomb.

Nexin:
I do like these devices it is why i still look for PS to TS converter. One that will not modify the stream during converting to TS.
I realise it's been a couple of months since you said this, but if you're still searching, VLC is the answer to your issues (videolan[dot]org). It's free and it can do a straight copy from a PS container to a TS container. I use it as part of my process for checking all my files before encoding to H264 to add to my library. It can be invoked using a command line, so I'll often feed it a stack of files and then come back and run multiple instances of MPEG2Repair to log the errors.

I've also written PHP code to turn a Windows Command Line listing into a batch file for processing as well as a parser for the log files which tells you if there's errors in a file and if so, where they are. If this is of interest to anyone, I could neaten it up and make it publicly available. Just let me know...

I appears I've just hit the time bomb in the last available version. I downloaded again just to check, but the version Mark has available for download is bombing immediately too.

Yep, it just died on me as well. Finished doing a logging of a file and went to start another one when it bitched about the beta time having expired.

That is a silly thing to do to a program.

I don't see it as a silly thing - the guy was writing free software and choosing to support it for free. If it's the least bit of help to know what version someone is using because they're forced to do so, then that's fine with me.

That said, now that development (and inevitably support) is finishing up, I'd love a non-bombed copy...

I don't see it as a silly thing - the guy was writing free software and choosing to support it for free. If it's the least bit of help to know what version someone is using because they're forced to do so, then that's fine with me.

So, you are fine with that fact that no one is now able to use the program at all?

Putting in time bombs can lead to situations like this, where a new version isn't released and what is out there becomes useless.

That is what version numbers are all about. If someone asks for help with a problem, the version of the program being used is asked for, if not initially indicted, so that support can be given. That might be as simple as telling the person to upgrade to the latest version because said bug was fixed.

Programs that aren't free have time bombs.

Like I said, it was silly to time bomb this program.

I didn't say I'm fine with not having its use any more. Indeed, I said I'd really like a copy that does work. I think that puts us on the same side here doesn't it?

I did say I understand a little of where the guy writing the code was coming from - it was amiss of me to write this response in a manner that made it sound like a contradiction to your view - sorry for that. This doesn't achieve anything.

Right now I have around 50 pieces of video I'm waiting to check with this tool before I can nuke them - don't get me wrong: this is frustrating for me too...

I didn't say I'm fine with not having its use any more. Indeed, I said I'd really like a copy that does work. I think that puts us on the same side here doesn't it?

Yep, it most certainly does.

I did say I understand a little of where the guy writing the code was coming from - it was amiss of me to write this response in a manner that made it sound like a contradiction to your view - sorry for that. This doesn't achieve anything.

I still do not understand the logic of forcing people to keep upgrading versions. Especially when things like this can happen. So someone is using an older version, who cares. If it works for that person, big frakin' deal. If that user has a problem, then download the latest version. At least tons of people all of a sudden won't end up having an extremely useful tool become useless.

Right now I have around 50 pieces of video I'm waiting to check with this tool before I can nuke them - don't get me wrong: this is frustrating for me too...

Thank goodness I don't have than many files. As a workaround, use VideoReDo and run a QuickFix on the file. If there are no problems, then you are safe and can blow away the "fixed" file. If there are problems, you can look at the VRD log and see where the problems are. It is more work to use VRD as a file error tester, but at least you aren't dead in the water.

Thanks for that pointer. Unfortunately I just use the logging side of M2R to test the integrity of the files. I've never had any compatibility issues which required me to actually do any repair.

The idea for me is that if a file has a g-glitch in it, I'll archive it, but then pick it up next time it's on TV. If it's clean then, I'll replace it in my video archive. If it were for repairing, I'd run with the VideoReDo suggestion.

If anyone knows of a similar logger I could be using in the mean time, I'd love to hear about it.

Sorry for the lack of updates. I've been having some internet problems the last few days. Since I can't update my web site at the moment, I attached the latest version to the first post in this thread. Expiration date has been removed since this is probably the final version.

Enjoy. :)

-Mark

Mark - thank you so much for intervening so quickly.

For all your work through the years on this - thanks too. I can't imagine life without this tool.

All the best on the things you're moving on to.

Josh

Yes.. great work... thanks for this essential program

Mark,

Thanks for adding h264 support! Just a few observations from quick testing;

- Sometimes the FPS parameter seems wrong-o... every cap I have is 25fps, but sometimes it'll come up with random numbers like 25.35 and one was even 30.5fps.
- It seems to freak out when processing sequential files (i.e. h264.part1.ts, h264.part2.ts...) by logging massive audio timestamp gaps at the end of each file. For example;


Sequence Frame 19898(0-X) / Time 0:12:46 :
VideoWarning: Discontinuity of (9+) packet(s). First packet ending at offset 564 in file h264.dd51.part2.ts

Sequence Frame 19899(0-X) / Time 0:12:46 :
VideoWarning: Discontinuity of (4+) packet(s). First packet ending at offset 6204 in file h264.dd51.part2.ts
Additional error(s) detected. Increase VerboseLogLevel in INI file for details.

Sequence Frame 19903(0-X) / Time 0:12:46 :
AudioWarning: Timestamp gap of 302.944003 sec. ending at file offset 251174 in file h264.dd51.part2.ts

Sequence Frame 36359(0-X) / Time 0:28:43 :
VideoWarning: Discontinuity of (7+) packet(s). First packet ending at offset 564 in file h264.dd51.part3.ts

Sequence Frame 36360(0-X) / Time 0:28:43 :
VideoWarning: Discontinuity of (14+) packet(s). First packet ending at offset 12032 in file h264.dd51.part3.ts
Additional error(s) detected. Increase VerboseLogLevel in INI file for details.
AudioWarning: Timestamp gap of 303.712003 sec. ending at file offset 1496938870 in file h264.dd51.part2.ts

Sequence Frame 56104(0-X) / Time 0:46:56 :
VideoWarning: Discontinuity of (8+) packet(s). First packet ending at offset 564 in file h264.dd51.part4.ts

Sequence Frame 56105(0-X) / Time 0:46:56 :
VideoWarning: Discontinuity of (2+) packet(s). First packet ending at offset 1880 in file h264.dd51.part4.ts
Additional error(s) detected. Increase VerboseLogLevel in INI file for details.

Sequence Frame 56106(0-X) / Time 0:46:56 :
AudioWarning: Timestamp gap of 293.376003 sec. ending at file offset 118150 in file h264.dd51.part4.ts

Sequence Frame 62229(0-X) / Time 0:55:44 :
Info: End of H.264 sequence

Sequence Summary:

File Size Processed: 5.30 GB, Play Time: 00h:40m:54s
25.35 FPS (Average), 17.70 Mbps (Average).
AC3 Audio: 3/2 Channels (L, C, R, SL, SR) + LFE, 48.0 kHz, 384 kbps.
Dialog Normalization: -27.0 dB, Center Mix Level: -3.0 dB, Surround Mix Level: -3.0 dB
3 of 76717 audio frames found with errors.
900.032010 seconds of audio timestamp gaps.

End of Log


You can see the FPS weirdness there as well. Files edited with the same tool as one long stream don't suffer the same fate (edited using TSPE, FYI).

Just seen someone else test on a German cap and get 150fps!

Thanks for that pointer. Unfortunately I just use the logging side of M2R to test the integrity of the files. I've never had any compatibility issues which required me to actually do any repair.

I don't use M2R for repairing either, just the logging. For many of the programs that give me glitch trouble, I capture both the east and west feeds. I then use the logging to get me a list of the problem areas of both files. I then use VRD to piece together a "clean" copy.

But, as I pointed out, you can use VRD just for the logging. You just have to use the "repair" facility to create the log, throwing away the file that was created as part of the process.

Mark - thank you so much for intervening so quickly.

For all your work through the years on this - thanks too. I can't imagine life without this tool.

All the best on the things you're moving on to.

Ditto. A great program that would be difficult to live without!

ditto :o :o

Mark,

Thanks for adding h264 support! Just a few observations from quick testing;

- Sometimes the FPS parameter seems wrong-o... every cap I have is 25fps, but sometimes it'll come up with random numbers like 25.35 and one was even 30.5fps.
- It seems to freak out when processing sequential files (i.e. h264.part1.ts, h264.part2.ts...) by logging massive audio timestamp gaps at the end of each file. For example;



You can see the FPS weirdness there as well. Files edited with the same tool as one long stream don't suffer the same fate (edited using TSPE, FYI).

Just seen someone else test on a German cap and get 150fps!

The H264 support was a quick hack I put together in less than a day because many people requested this feature. It only checks the MPEG portion of the file which is the PES and TS headers around the H264 elementary data. It's not going to catch all of the errors since it doesn't decode each frame the way the MPEG2 version does. I guess it's better than nothing. It also still checks the AC3 audio track so that may catch some errors.

The FPS reported for H264 is an average based on how many payload_unit_start indicators are present on the video PID. Usually this matches the number of pictures encoded but I guess it's not always the case. I'll try to find a couple free minutes this weekend to add a rudimentary NAL packet parser, at least to extract correct sequence headers for H264. I'll also try to fix the multi-segment support.

BTW, new web server is:
http://homepages.roadrunner.com/mwilczyn/mpeg2

-Mark

amazing.

Mark, is there anything we can do to get you to continue to support the software?

Mark,

Thanks for adding h264 support! Just a few observations from quick testing;

- Sometimes the FPS parameter seems wrong-o... every cap I have is 25fps, but sometimes it'll come up with random numbers like 25.35 and one was even 30.5fps.
- It seems to freak out when processing sequential files (i.e. h264.part1.ts, h264.part2.ts...) by logging massive audio timestamp gaps at the end of each file. For example;


House,

I think the audio timestamp gaps are there because many simple editing tools (HDTV2MPEG2, TSPE, etc.) don't update the timestamps if you remove/cut out video such as commercials. That's probably why you see 5 minute long audio gaps and continuity errors. Video time gaps will not be shown for H264 because the H264 picture headers are not parsed at this time.

It is interesting that many of the errors are at a start of a segment but that could just be coincidence. What tool did you use to split the files? What is the exact size of each segment (in bytes)?

-Mark

Sorry for the lack of updates. I've been having some internet problems the last few days. Since I can't update my web site at the moment, I attached the latest version to the first post in this thread. Expiration date has been removed since this is probably the final version.

Enjoy. :)

-Mark


Figures as I spent 3 hours building a process to bypass the timebomb, lol - and actually finally was able to do it with a simple mpeg2repair.bat file I point the icon to instead of mpeg2repair.exe - and blowing up caps all day long as I advanced the calendar to check it out and the timers figured it was time to stop recording :(

But seriously, I would gladly pay for this program. I really wish you would continue support and development in any spare time.

House,

I think the audio timestamp gaps are there because many simple editing tools (HDTV2MPEG2, TSPE, etc.) don't update the timestamps if you remove/cut out video such as commercials. That's probably why you see 5 minute long audio gaps and continuity errors. Video time gaps will not be shown for H264 because the H264 picture headers are not parsed at this time.

It is interesting that many of the errors are at a start of a segment but that could just be coincidence. What tool did you use to split the files? What is the exact size of each segment (in bytes)?

-Mark

Usually what I do is take my cap (say an hour long with commercial), and edit each segment into it's own file, so in the end I get (for example) about four ten minute long files (for a 40 min show). If you try to splice them together with TSPE (here's (http://ventolin.**********.com/) the website, by the way) then you'll get pretty much a broken file - timestamp will be wrong, you'll get massive glitches at the joined parts... same thing if you do a simple copy /b.

Here are the sizes for those parts I used for the posted m2r log;

part1; 1,877,449,028
part2; 1,496,971,808
part3; 1,881,500,992
part4; 438,757,596

Just today, I tried starting MPEG2Repair and it gives me an error message saying "This beta has expired. Please download a new version."

I'm currently using the August 14, 2006 version. There isn't any new version on the website.

I also sent this message to mpeg2repair@adelphia.net too to make sure I covered all bases.

Thanks.

Just today, I tried starting MPEG2Repair and it gives me an error message saying "This beta has expired. Please download a new version."

I'm currently using the August 14, 2006 version. There isn't any new version on the website.

I also sent this message to mpeg2repair@adelphia.net too to make sure I covered all bases.

Thanks.

You're hosed...

Amazing what you can find if you read the past 2 days of posts...but to sum it up if you don't want to look through the posts, paypal $100 to mpeg2repair@adelphia.net and he will send you a new copy.

You're hosed...

Amazing what you can find if you read the past 2 days of posts...but to sum it up if you don't want to look through the posts, paypal $100 to mpeg2repair@adelphia.net and he will send you a new copy.


Aww, c'mon, give the guy a break. After all, I never respected Wizzywig's reasons for making each ver. have an expiration date. Lawsuit? Pahleese - I guess everyone who writes Freeware better watch out. If someone doesn't like the free software you wrote, you are gonna get sued!!!!!!

I felt that Mark was jealously holding onto his software, but the fact that (if) he released it now, non-shiva, does change my mind a bit. Its great software, but only time will tell if it lasts. I appreciate his efforts greatly, and have been a regular financial contributor (I put my money where my mouth is), but damn me if I'm not going to be honest - it really angered me how he always put that expiration date there.

Thanks Mark for the software. Thanks Mark for not putting an expiration date on it this time.

Usually what I do is take my cap (say an hour long with commercial), and edit each segment into it's own file, so in the end I get (for example) about four ten minute long files (for a 40 min show).

OK, this leads to the obvious question... why are you creating separate files?

If you use a program like VideoReDo, you edit the program, removing unwanted material and then outputting a single, key word SINGLE, resulting file.

You'll save yourself a lot of grief.

OK, this leads to the obvious question... why are you creating separate files?

If you use a program like VideoReDo, you edit the program, removing unwanted material and then outputting a single, key word SINGLE, resulting file.

You'll save yourself a lot of grief.

Sorry to jump right on in, but I break up my files as well. Reason is I archive to DVD-R (140Mb chunks), and I want to fill up each DVD. Since I get DVD's cheap (as low as ten cents), I will continue to use this as my archiving method. Of course, everything is stripped and repaired before I break it up and burn it. One day when Blu-Ray blanks are a dime, I will have plenty of *.TS's to burn! Hope this helps :-)

You're hosed...

Amazing what you can find if you read the past 2 days of posts...but to sum it up if you don't want to look through the posts, paypal $100 to mpeg2repair@adelphia.net and he will send you a new copy.

Sorry, I was busy and wrote a quck reply and went to work. Apparently, the website changed and I wasn't aware.

Sorry to jump right on in, but I break up my files as well. Reason is I archive to DVD-R (140Mb chunks), and I want to fill up each DVD. Since I get DVD's cheap (as low as ten cents), I will continue to use this as my archiving method. Of course, everything is stripped and repaired before I break it up and burn it. One day when Blu-Ray blanks are a dime, I will have plenty of *.TS's to burn! Hope this helps :-)

Ya so :) I too burn these TS files to DVD-R. I have cygwin on my box, so I use the Unix split command to break up the files into pieces that will fit onto a DVD, as a single file, i.e., any piece is no bigger than 4.5GB. If it takes three DVDs, which it can for very high bitrate network captures, I split into three evenly sized files. If I want to work with the files, I just read them off the DVD and "cat" them together.

For those without cygwin on their system, using a program to create RAR files will work. Just split the file into whatever it takes to fit the DVDs.

So, just edit into a single file and then split the file to fit on the DVDs.

You are putting way too much effort into something extremely simple.

I wrote a little addon that converts mpeg2repair to a command-line tool.

Syntax is:

mpeg2repairhelper "c:\input file.ts" "c:\output file.ts"

Details and download at:

ht..||boling.us|ebs|apps|mpeg2repair|

Sorry about the malformed URL; this is my 1st post, so I'm not allowed to put a real one in this message. :-(

Anyone else want to put a real link in a response for convenience? :-)

I wrote a little addon that converts mpeg2repair to a command-line tool.

Syntax is:

mpeg2repairhelper "c:\input file.ts" "c:\output file.ts"

Details and download at:

ht..||boling.us|ebs|apps|mpeg2repair|

Sorry about the malformed URL; this is my 1st post, so I'm not allowed to put a real one in this message. :-(

Anyone else want to put a real link in a response for convenience? :-)

Thanks a bunch, here (http://www.boling.us/ebs/apps/mpeg2repair/) is the link.

Is there a way to create the log without creating the repaired file?

OK, this leads to the obvious question... why are you creating separate files?

If you use a program like VideoReDo, you edit the program, removing unwanted material and then outputting a single, key word SINGLE, resulting file.

You'll save yourself a lot of grief.

Because Videoredo doesn't have h264 support?

I've been using your tool for a long time Wizziwig, and just wanted to thank you for the hard work you've put into it, it truely is invaluable.

Because Videoredo doesn't have h264 support?

Ah, gotchya.

It is being planned, but they do not ever say when new features like this will show up.

All of my transport files are MPEG-2, so it isn't a problem.

Ya so :) You are putting way too much effort into something extremely simple.

How so? I want to fill up each and every dvd as much as possible. Don't want to waste space. 140mb is perfect for 31 of these files, and with a tiny bit of room left over (because I don't like to write 100% to capacity, and in case I need to add a note, ect).

I just keep it plain simple, 140mb per ts, stripped and repaired. No waste of DVD's, and a consistient product.

How so? I want to fill up each and every dvd as much as possible. Don't want to waste space. 140mb is perfect for 31 of these files, and with a tiny bit of room left over (because I don't like to write 100% to capacity, and in case I need to add a note, ect).

I just keep it plain simple, 140mb per ts, stripped and repaired. No waste of DVD's, and a consistient product.

I'm going to assume something here. Please correct me if I am wrong, but it sounds like you are putting more than one show per DVD. If so, I can understand some of your logic. DVD-R media is relatively cheap, so I only put one show per DVD-R set.

90+% of the material I put onto DVD is HD TS. Depending on source, some single shows will fit on a DVD (Fox material for instance because their bitrate is too damn low), while others take 2-3 DVDs. I therefore split accordingly, i.e., if it will take two DVDs, I split the file fairly evenly. If it takes three, I split into thirds, etc.

I only split a file if absolutely necessary.

The next question... how are you splitting?

I'm going to assume something here. Please correct me if I am wrong, but it sounds like you are putting more than one show per DVD. If so, I can understand some of your logic. DVD-R media is relatively cheap, so I only put one show per DVD-R set.

90+% of the material I put onto DVD is HD TS. Depending on source, some single shows will fit on a DVD (Fox material for instance because their bitrate is too damn low), while others take 2-3 DVDs. I therefore split accordingly, i.e., if it will take two DVDs, I split the file fairly evenly. If it takes three, I split into thirds, etc.

I only split a file if absolutely necessary.

The next question... how are you splitting?

Yup. I archive so much, I prefer not wasting any space. We're talking close to 10,000 discs now. I use the only version (1.11.0) of HDTVtoMPEG2 that I know of which doesn't add errors or mess things up. And I have verified the stream is 100% compliant after HDTVtoMPEG editing. But I used to do that, like a 6 gig show I'd either spilt up 3 gigs per DVD, or maybe 4/2-whatever. That was when I actually thought of streaming from the discs real time. Now I have a hard drive dedicated to that, and if I really need a small file, I can always re-encode the mpeg2 to an XviD. So my main concern is to use the DVD potential to its fullest capacity, just for efficiency.

We're talking close to 10,000 discs now.

Damn, that is a lot of discs :D

It is hard enough finding the time to capture and save what I do now, let alone create a library of that many DVDs.

Sheesh. Brings to mind Ismelda Marcos and closets-full of shoes that she never could possibly have lived long enough to wear. :o

I'm new to M2R, but I scanned this forum and couldn't find an answer to this question. How do I fix a file, but leave the ac3 stream as an ac3 stream? When I fix a file it turns a 384 kbps ac3 file into a 176 kbps MPEG2 file. Any ideas?

I wrote a little addon that converts mpeg2repair to a command-line tool.

Syntax is:

mpeg2repairhelper "c:\input file.ts" "c:\output file.ts"

Details and download at:

ht..||boling.us|ebs|apps|mpeg2repair|

Sorry about the malformed URL; this is my 1st post, so I'm not allowed to put a real one in this message. :-(

Anyone else want to put a real link in a response for convenience? :-)
Thank chuckybaby you for posting this m2r batch utility. As said many of us need the output a log file only, no repair is needed or wanted.

Also the ability to give each file the pids for processing. M2R for me, 90% of the time will not autofind them correctly. I select each time from the drop down program numbers list. This is after it fails the initial scan.

Sheesh. Brings to mind Ismelda Marcos and closets-full of shoes that she never could possibly have lived long enough to wear. :o

Haa-Haa. Yes, its an addiction ;)

I'm new to M2R, but I scanned this forum and couldn't find an answer to this question. How do I fix a file, but leave the ac3 stream as an ac3 stream? When I fix a file it turns a 384 kbps ac3 file into a 176 kbps MPEG2 file. Any ideas?
Can anyone help out on this audio file question?

Thanks!

Can anyone help out on this audio file question?

I don't use M2R for repair work. I just use it for getting logging info (looking for glitch locations).

I use VideoReDo for editing, which also can do glitch repair.

Can anyone help out on this audio file question?

Thanks!

M2R does not have the ability to change audio or video format. How do you know the original contained AC3 and repaired has MPG? Were there any warnings or errors during/after the "Find PID" stage? Some channels contain both AC3 and MPG audio on different tracks. In such cases, you may need to manually select the correct audio track. Please post or email some of the log file (including summary).

Usually what I do is take my cap (say an hour long with commercial), and edit each segment into it's own file, so in the end I get (for example) about four ten minute long files (for a 40 min show). If you try to splice them together with TSPE (here's (http://ventolin.**********.com/) the website, by the way) then you'll get pretty much a broken file - timestamp will be wrong, you'll get massive glitches at the joined parts... same thing if you do a simple copy /b.

Here are the sizes for those parts I used for the posted m2r log;

part1; 1,877,449,028
part2; 1,496,971,808
part3; 1,881,500,992
part4; 438,757,596

I don't think there is anything I can do to help you. Since you're missing all the data that was part of the commerical, the remaining pieces will not fit back together without gaps/corruption. As I said before, none of the freeware editors correctly update timestamps after removing video. My tool will only warn you of these errors, there is nothing in there to modify or fix timing issues.

I don't know of any H264 editors that will do what you need correctly. I recall seeing something about H264 on the last Womble update patch. Maybe they have something that will work?

Usually what I do is take my cap (say an hour long with commercial), and edit each segment into it's own file, so in the end I get (for example) about four ten minute long files (for a 40 min show). If you try to splice them together with TSPE (here's (http://ventolin.**********.com/) the website, by the way) then you'll get pretty much a broken file - timestamp will be wrong, you'll get massive glitches at the joined parts... same thing if you do a simple copy /b.

I don't know TSPE and what it does. I do know, and you might not, how MPEG-2 is basically formated. In a nutshell, because of how MPEG-2 is formatted, you just can't join two pieces together, like you can two audio files (at the same sample rate).

To join two MPEG-2 files, they have to be electronically joined, where the joint has to be electronically rebuilt. A program to do that correctly is one like VideoReDo.

When editing a program, it is best to create a single edited file. If you then have to split it up to fit on CD/DVDs, the you want to split the file. I have the "Unix" Cygwin tools, which has the Unix split program, which can split a file any size that is needed. Then, all you have to do is "cat" them back together to get the original file back together. Even copy /b will work.

M2R does not have the ability to change audio or video format. How do you know the original contained AC3 and repaired has MPG? Were there any warnings or errors during/after the "Find PID" stage? Some channels contain both AC3 and MPG audio on different tracks. In such cases, you may need to manually select the correct audio track. Please post or email some of the log file (including summary).
I'll post a log file later, but what I'm trying to repair is an off-air ATSC transport stream. The audio is AC3 (as it must be for ATSC) 384 kbps, 5.1 audio. PIDS were correctly located, but when I checked the fixed file with TSReader the PID numbers were changed and the audio was 176 kbps MPEG audio and could not be played on VLC or HDTV Fusion.

The original plays, but it has a lot of dropouts. Maybe it is a bit too far gone?

The original plays, but it has a lot of dropouts. Maybe it is a bit too far gone?

Get VideoReDo and use it to do the repair.

Mark I still have to use v1.1.3

The v1.1.4 and agian the v1.1.5 log formating is harder for me to read. Can you please release a updated v1.1.5 with the same, log formatting that v1.1.3 has.

Info: Please make again the nice clean streamlined formatting.
Info: It is so untidy and hard to read in the v1.1.4 and v1.1.5 newer versions
Info: As each video is stored on dvd the original log file accompanies it.
Info: Mark i'm hoping you would make this needed change before you dissapear from Mpeg2Repair.
Info: I would prefer to be using a more recent version, with clean easy to read log formatting.
Info: Thank You :>

I prefer the information in the newer versions. It's much more detailed and gives a much better idea of what is going on in the file. If you are using that older version then it must not be as important to you.

The v1.1.4 and agian the v1.1.5 log formating is harder for me to read. Can you please release a updated v1.1.5 with the same, log formatting that v1.1.3 has.

I have no problem understanding the log entries.

I'm not saying change the ammount of details that are in any recent versions.

Its not the log file understanding that i have difficulty with at least not with v1.1.3. It is with reading the log itself. Just the tail part, to me it looks like a jumbled mess, a hard to read mess.

Info: This formatting is very much prefered for reading.

Its not the log file understanding that i have difficulty with at least not with v1.1.3. It is with reading the log itself. Just the tail part, to me it looks like a jumbled mess, a hard to read mess.

The current format is not hard to read or understand. At least, not to me.

I'm not saying change the ammount of details that are in any recent versions.

Its not the log file understanding that i have difficulty with at least not with v1.1.3. It is with reading the log itself. Just the tail part, to me it looks like a jumbled mess, a hard to read mess.

Info: This formatting is very much prefered for reading.


Info: Learn to live with it.

I have a dissabilty with reading the log looks like a jumbled mess. At least my dissabilty does not affect reading and understanding of forum threads. You both must have the same dissabilty, to reply to a post not directed at either of you.


My original post was directed at Mark the developer of M2R only.

Nexin,

If I remember correctly, I made this format change in order to keep the line width as short as possible and thus avoid word wrapping on some displays. Most people seem to prefer a compact design they can easily copy/paste into other documents. You will never make everyone happy. :(

It's just a plain text file. If format bothers you so much, why not just open the file in notepad and edit it anyway you like before saving to DVD?

-Mark

Thanks for the reply Mark, this is what i did try to do using note or wordpad. But it became to time consuming. I produce a log file for eveything i record, most are 10-12 mins each to produce. It all adds up having to do it manualy, baby sitting to input the next one.

Would be nice if you could put out two final versions. One with Info in the log output one without. I may not be the only one, still having to use an older version because of this.

The word wrap you mentioned above. Using Info: in defining each new line in log output report, this would not be a problem, would be very clean and unjumbled.

If you publicly don't want to publicly output another version. You can pm me a url from an upload site, so i could download a M2R Info: adjusted version. It will stay non public.

Thanks for the reply Mark, this is what i did try to do using note or wordpad. But it became to time consuming. I produce a log file for eveything i record, most are 10-12 mins each to produce. It all adds up having to do it manualy, baby sitting to input the next one.

Would be nice if you could put out two final versions. One with Info in the log output one without. I may not be the only one, still having to use an older version because of this.

The word wrap you mentioned above. Using Info: in defining each new line in log output report, this would not be a problem, would be very clean and unjumbled.

If you publicly don't want to publicly output another version. You can pm me a url from an upload site, so i could download a M2R Info: adjusted version. It will stay non public.

You get a program for which you paid nothing and its too much trouble for you to spend 15 seconds in a word editor.

ROFLMAO.

Amazing attitude.

Good you have no problem, though i do have a problem. Wordpad editing is not my thing, make many mistakes. Then there is the time in doing so.

Not everyone in this world is the same, you seem not to realise this. You are implying what you can do, everyone else is able to do.
That is not how it is, not one person is ever totaly the same as another person, ever.

Or are you just another forum trawler, writing inane remarks here and there at will.

Amazing how all your posts are complaining about support for free products.

As I have said before:

http://www.avsforum.com/avs-vb/showthread.php?p=7891050&&#post7891050

You seem to have a personel grudge with my request or me. Could you use another thread and topic on the forum somewhere, to voice your opinions about this, i'm sure i will ignore it.

You would think after 7 thounsand plus posts, you would know better. Are the last few posts in this thread, indicitive of the quality of your postings on this site. With comments out of place, critisim and complaints where they are not in context.

Do you just spend your time here trolling the forums to see who you can antagonise. So if or when they retaliate and tell you to sod off or similar. You can the press the report button , then sit back and wait to see if that persecuted user dissapears. What a sad flucking life you have, the only one left is you. Mr no mates, Mr Lonely thats is all you are, a waste of space.

Go on then post your (HDTVFanatic loves the) 'last word' reply, after all you have nothing better to do with your existance.

My original post was directed at Mark the developer of M2R only.

This is a public forum. No posting is ever only directed to a particular person.

If you don't want others to particpate in your forum postings, send a private message to that forum member.

Do not bitch because others made a comment to your public posting.

You seem to have a personel grudge with my request or me.

I seriously doubt that.

I have a problem with the request in that it will end up dumbing down what the log contains. It contains lots of useful info, of which I look for the pieces I need, ignoring the rest.

While you don't want to use wordpad, or whatever, to "fix" the log the way you want it, I have a suggestion that will allow you to automate the process. But, it will require learning on your part.

Windblows doesn't have the tools, Unix does. So, load Cygwin onto your system and then learn how to build a script that uses "awk" to parse the original log file and produce the output you want. Then all you have to do is run your script against the log file and you'll have what you want, without having to affect the log for the rest of us.

You aren't going to get the author to change the latest version to match your particular needs (I hope the author doesn't cave). That is not fair to the rest of us who love the log just the way it is. Nor do I think the author should spend his time creating a special version just for you.

Go develop tools on your own to produce what you want.

You seem to have a personel grudge with my request or me. Could you use another thread and topic on the forum somewhere, to voice your opinions about this, i'm sure i will ignore it.


Again, your 46 posts are essentially in 2 threads - both free software where you do nothing but complain about them.

I, unlike you, have been willing to put financial support Mark in this venture to try to keep him developing the software. All you keep doing is complaining the log format isnt to your liking.

For God's sake, if it were an error, I could understand it - but a fwaking log format.

Get off your lazy ass and use notepad.

guys,

take a break & give us one too.

I probably shouldn't be doing joining this discussion, but here is my build:

AWK is a great tool to be used for manipulating text lines and outputting the data into a new file. So ie SED. I think you can get MS-DOS or "Windows Command Prompt" versions of these tools WITHOUT the need to install the cygwin program. I know for a fact that SED does.

So you could then write a quick logstrip.bat file and have the tool "simplify" your logfiles.

Or if you want to stay in the MSWindows world, there is a Microsoft scripting language called VBScript that you can also use to parse text files. This scripting language comes with most modern MS Windows systems by default. Otherwise, it can be installed for free from the Microsoft download site.

Duming down - The information would be exactly the sames as you see now.
The Info: adjustment will then have the "Info: comment" (without quotes) in front of each statement. Making the formatting much easier to read.

Feeware or payware. It make no difference to the question i asked to Mark. Before you champion Marks defence why not first ask Mark. If he wants or needs you, as his spokesman.

A simple adjustment by Mark, would do away with having to find a new editor, or having to learn one. The adjustment for Mark is a very simple one to do, as said it was there upto v1.1.3.

Duming down -


You might think about learning how to spell "dumbing down" - but I do think that explains why you must have someone do it for you :D

I have a .ts file recorded OTA from CBS.
It plays fine on every Player..zoom,wmp,mpc etc.
MediaInfo reports it contains:
-----------------------------------------------------------
General #0
Complete name : V:\Videos\MP4Tests\Cold Case.ts
Format : MPEG-2 Transport
Format/Family : MPEG-2
File size : 44.2 MiB
PlayTime : 20s 679ms
Bit rate : 18 Mbps

Video #0
Codec : MPEG-2 Video
Codec profile : Main@High
Codec settings/Matri : Standard
PlayTime : 20s 679ms
Bit rate : 19 Mbps
Bit rate mode : CBR
Width : 1920 pixels
Height : 1080 pixels
Aspect ratio : 16/9
Frame rate : 29.970 fps
Chroma : 4:2:0
Interlacement : Top Field First
Bits/(Pixel*Frame) : 0.303

Audio #0
Codec : AC3
PlayTime : 20s 480ms
Bit rate : 384 Kbps
Bit rate mode : CBR
Channel(s) : 6 channels
Sampling rate : 48 KHz
Video0 delay : -294ms
ChannelPositions : Front: L C R, Rear: L R, Subwoofer
-----------------------------------------------------------------------
When I run it through MPEG2Repair it finds the Video and Audio PIDs and reports:
-----------------------------------------------------------------------
MPEG2Repair: C:\Documents and Settings\Owner.GTW\My Documents\Videos\MP4Tests\Cold Case.ts

Sequence Frame 609(11-P) / Time 0:00:20 :
VideoError: Invalid motion vector code. MBA=536(896,64)
VideoError: No start code at end of slice. MBA=536(896,64)
VideoError: Failed to decode macroblock at MBA=536(896,64)
VideoError: Missing 7625 macroblocks in picture slice(s) at MBA=535(880,64).
FileInfo: Last video errors span 16 bytes at file offset 46333524

Sequence Frame 610(11-P) / Time 0:00:20 :
Info: End of MPEG2 sequence

Sequence Summary:

File Size Processed: 44.19 MB, Play Time: 00h:00m:20s
1920 x 1080, 29.97 fps, 18.80 Mbps (16.88 Mbps Average).
Average Video Quality: 68.63 KB/Frame, 0.27 Bits/Pixel.
MPEG Audio.
1 of 610 video frames found with errors.
0 of 0 audio frames found with errors. <<<<<------!!!!!!
16 corrupted video bytes in file.
0.000000 seconds of video timestamp gaps.
0.000000 seconds of audio timestamp gaps.

End of Log
--------------------------------------------------------------

M2R could not find the audio!! so the output file contains no audio
but the video plays fine.

-------------------------------------------------------------------

Following a tip in the forum above, I loaded the original file
into VideoReDo and Saved it without changes.

Now the file processes correctly and M2R reports:
-----------------------------------------------------------------
MPEG2Repair: C:\Documents and Settings\Owner.GTW\My Documents\Videos\MP4Tests\Cold Casevrd.ts

Sequence Frame 598(13-B) / Time 0:00:19 :
Info: End of MPEG2 sequence

Sequence Summary:

File Size Processed: 42.41 MB, Play Time: 00h:00m:19s
1920 x 1080, 29.97 fps, 18.80 Mbps (16.88 Mbps Average).
Average Video Quality: 68.68 KB/Frame, 0.27 Bits/Pixel.
AC3 Audio: 3/2 Channels (L, C, R, SL, SR) + LFE, 48.0 kHz, 384 kbps.
Dialog Normalization: -31.0 dB, Center Mix Level: -3.0 dB, Surround Mix Level: -3.0 dB
0 of 598 video frames found with errors.
0 of 623 audio frames found with errors. <<<<----!!!
0 corrupted video bytes in file.
0.000000 seconds of video timestamp gaps.
0.000000 seconds of audio timestamp gaps.

End of Log
-------------------------------------------------------
Audio is now present but the errors in the original file are gone.

Does this mean they were fixed by VRD or did VRD just bury them?..if buried
then M2R should be able to handle this file correctly.

I also have a second file recorded from the Grammy's with the same
problem but VRD does not fix the problem and M2R still finds no audio.

If these files all play then M2R should process them correctly.

Has everyone but me deserted this program??

I'd sure appreciate some help getting audio back.

Ran a Criminal minds episode thru M2R las night..8 gigs..no audio out!

Ran fine thru VideoReDo..edited it and output to DVD using ConvertX at 8mbps, but can't run thru M2R

Log says:

MPEG2Repair: V:\Videos\Total Media ts\KCBS-DT_02_04_2007_19_26_20.ts

Sequence Frame 2867(3-B) / Time 0:01:35 :
VideoError: Motion vector going out of range at MBA=7961(656,1056)
FileInfo: Last video errors span 357 bytes at file offset 221306668

Sequence Frame 12173(9-B) / Time 0:06:46 :
VideoError: Missing 4800 slice macroblocks at MBA=3360(0,448)
VideoError: Missing 4815 macroblocks in picture slice(s) at MBA=3345(1680,432).
FileInfo: Last video errors span 14 bytes at file offset 931241878

Sequence Frame 12174(9-B) / Time 0:06:46 :
FileInfo: Last video errors span 128 bytes at file offset 931241892
VideoWarning: Unexpected MPEG2 start code 000001b2 skipped
FileInfo: Last video errors span 56 bytes at file offset 931242024
.
.
.
Sequence Frame 122492(0-I) / Time 1:08:07 :
VideoError: Invalid Huffman code in intra MPEG2 block. MBA=22(352,0)
VideoError: Failed to decode macroblock at MBA=22(352,0)
VideoError: Missing 8139 macroblocks in picture slice(s) at MBA=21(336,0).
FileInfo: Last video errors span 104 bytes at file offset 8912516648

Sequence Frame 122493(0-I) / Time 1:08:07 :
Info: End of MPEG2 sequence

Sequence Summary:

File Size Processed: 8.30 GB, Play Time: 01h:08m:07s
1920 x 1080, 29.97 fps, 18.80 Mbps (16.48 Mbps Average).
Average Video Quality: 67.13 KB/Frame, 0.27 Bits/Pixel.
MPEG Audio.
26 of 122493 video frames found with errors.
0 of 0 audio frames found with errors. <====
113855 corrupted video bytes in file.
0.000000 seconds of video timestamp gaps.
0.000000 seconds of audio timestamp gaps.

End of Log

Out of that entire log, and the program doens't list video/audio PIDs it detects? Hah.

Guess you don't use the program.
The log does not show PID's just the GUI.

Anyway I fixed the problem by stripping off the first half meg of the file.
Seems the program should be smart enough to have found the audio without me having to do this. Must be giving up pretty early.

Log now shows:

MPEG2Repair: V:\Videos\MP4ColdCaseTests\coldcase-500K.ts


Sequence Frame 609(11-P) / Time 0:00:20 :
VideoError: Invalid motion vector code. MBA=536(896,64)
VideoError: No start code at end of slice. MBA=536(896,64)
VideoError: Failed to decode macroblock at MBA=536(896,64)
VideoError: Missing 7625 macroblocks in picture slice(s) at MBA=535(880,64).
FileInfo: Last video errors span 16 bytes at file offset 45833524

Sequence Frame 610(11-P) / Time 0:00:20 :
Info: End of MPEG2 sequence

Sequence Summary:

File Size Processed: 43.71 MB, Play Time: 00h:00m:20s
1920 x 1080, 29.97 fps, 18.80 Mbps (16.88 Mbps Average).
Average Video Quality: 68.63 KB/Frame, 0.27 Bits/Pixel.
AC3 Audio: 3/2 Channels (L, C, R, SL, SR) + LFE, 48.0 kHz, 384 kbps.
Dialog Normalization: -31.0 dB, Center Mix Level: -3.0 dB, Surround Mix Level: -3.0 dB
1 of 610 video frames found with errors.
0 of 635 audio frames found with errors.
16 corrupted video bytes in file.
0.000000 seconds of video timestamp gaps.
0.000000 seconds of audio timestamp gaps.

End of Log

I don't have access to any ATSC transport streams, but I ran it on a local HDTV capture with 1440x1080 mpeg2 @ ~13mbit and AAC audio, and it, also, showed same thing. The GUI displayed Video/Audio PIDs properly, and when I clicked start, it went through and made a copy of the video, with same 0 of 0 audio frames, however the resulting file DID have audio - it just wasn't checked.
Yes, probably a bug.

very strange. I had the same thing happen to me once or twice in the past.

I hope someone picks up the code and uses it elsewhere (would be an awful shme if they didn't). HDTVTOMPEG2 seems like a potential choice.

Also TSREMUX might work (although it's written in C#, I believe).

A new version of my hack to turn mpeg2repair into a command-line tool has been released.

It now allows you to write a log file without actually writing a repaired version of the video stream. It also handles streams with only a single audio program now.

Further Details at:
ht..||boling.us|ebs|apps|mpeg2repair|

(I still haven't posted 5 times, so am not allowed to include URLs in my posts...)

Just to update.

I fixed the no audio problem by stripping off the first meg or so of the file.
Others can do this using the trim feature of VideoReDo, but MPEG2Repair should be able to find the video too. Sample files have been provided to the author, and hopefully he will find the time to fix this very valuable program.

Just to update.

I fixed the no audio problem by stripping off the first meg or so of the file.
Others can do this using the trim feature of VideoReDo, but MPEG2Repair should be able to find the video too. Sample files have been provided to the author, and hopefully he will find the time to fix this very valuable program.

Thanks for the samples. I found the bug. Somehow the AC3 headers on your samples are triggering my MPEG audio detection code before the AC3 parser has had a chance to realize it's actually AC3 audio. So your audio is being misidentified as MPEG audio. Since I don't do any decoding/testing of MPEG audio, it reports 0 audio frames in the log.

The reason this doesn't affect other applications is because they blindly trust the PMT tables to tell them what audio format is in the file. I found that method too unreliable with certain european streams so I use something more complicated. Unfortunately, this bug somehow fell through the cracks even after being there for a couple years. :(

I wasn't planning to release any more versions of MP2R so you'll have to wait before I can provide a fix. I'll try to finish a couple other new features to justify making a new version. Maybe in a few weeks, depending on my work schedule.

-Mark

Same bug happens with a lot of MPEG2TS/AAC streams that I have as well.

Same bug happens with a lot of MPEG2TS/AAC streams that I have as well.

I don't see how this bug matters for AAC since I don't do any testing of AAC audio no matter what. Are you saying it's reported as MPEG in the log? Please put a sample on a public upload site and email me the link.

Anyone else: if you have any major outstanding bugs to report, please do so now or forever hold you peace. :) I want this to be the last version.

-Mark

Hi Mark,

Anyone else: if you have any major outstanding bugs to report, please do so now or forever hold you peace. :) I want this to be the last version.

Here is an AVC file:

http://www.savefile.com/files/906831


MPEG2Repair cannot find the PIDs inside.

Thanks.

I don't see how this bug matters for AAC since I don't do any testing of AAC audio no matter what. Are you saying it's reported as MPEG in the log? Please put a sample on a public upload site and email me the link.

Anyone else: if you have any major outstanding bugs to report, please do so now or forever hold you peace. :) I want this to be the last version.

-Mark

Mark have you ever thought of... erm no, maybe not.

Ah a new thought, yes it did hurt.. ;)


A-Word: With a colon and a space, after each first word of each line. Now this is only one extra character being a colon: and it would align all the details. :)

Chuckybaby off to try your updated mpeg2repairHelper hack.

To: test if it allows for manualy input of the program stream id.
See: if it can batch a few files all for individual log output.
Check: if it can batch for example. Filename 1, Filename 2 and Filename 3 etc to individual log files, from the same batch script.

All this would be for nothing if Mark decides to better your hack. It's Mark`s prerogative to do so. not that any of us here would ever argue with that or want to. ;)

Off now to test it bbl.

Example with AAC:

Sequence Frame 2427(13-B) / Time 0:01:20 :
Info: End of MPEG2 sequence

Sequence Summary:

File Size Processed: 158.89 MB, Play Time: 00h:01m:20s
1440 x 1080, 29.97 fps, 20.00 Mbps (13.62 Mbps Average).
Average Video Quality: 55.38 KB/Frame, 0.29 Bits/Pixel.
AAC Audio.
0 of 2427 video frames found with errors.
0 of 0 audio frames found with errors.
0 corrupted video bytes in file.
0.000000 seconds of video timestamp gaps.
0.000000 seconds of audio timestamp gaps.

Somewhere in one of your changelogs you said you added 'AAC audio detection'.

By the way, if you plan to have that as "the last version", perhaps it would be beneficial to make the source available as well, so that a useful utility isn't wasted.

Oh.. i forgot how it was for the recent versions my idea above could never work. Only this could and still no line wraps happen in a txt log, normal reduced window.

Maybe a switch in the .ini to turn it on and off for user to choose which formatting prefered.



Sequence Summary:

Info: File Size Processed: 158.89 MB, Play Time: 00h:01m:20s
Info: 1440 x 1080, 29.97 fps, 20.00 Mbps (13.62 Mbps Average).
Info: Average Video Quality: 55.38 KB/Frame, 0.29 Bits/Pixel.
Info: AAC Audio.
Info: 0 of 2427 video frames found with errors.
Info: 0 of 0 audio frames found with errors.
Info: 0 corrupted video bytes in file.
Info: 0.000000 seconds of video timestamp gaps.
Info: 0.000000 seconds of audio timestamp gaps.


End Of Summary.


Thanks go to timecop for use of his summary report.

Example with AAC:

Sequence Frame 2427(13-B) / Time 0:01:20 :
Info: End of MPEG2 sequence

Sequence Summary:

File Size Processed: 158.89 MB, Play Time: 00h:01m:20s
1440 x 1080, 29.97 fps, 20.00 Mbps (13.62 Mbps Average).
Average Video Quality: 55.38 KB/Frame, 0.29 Bits/Pixel.
AAC Audio.
0 of 2427 video frames found with errors.
0 of 0 audio frames found with errors.
0 corrupted video bytes in file.
0.000000 seconds of video timestamp gaps.
0.000000 seconds of audio timestamp gaps.

Somewhere in one of your changelogs you said you added 'AAC audio detection'.

By the way, if you plan to have that as "the last version", perhaps it would be beneficial to make the source available as well, so that a useful utility isn't wasted.

I guess it wasn't clear enough in the changelog. By supporting AAC, I just meant that it would allow you to open and save ts files with AAC audio. It will not analyze or test any of the AAC (or MPEG) audio data so it has no idea how many frames are present. Only errors you would see are continuity errors. Only AC3 audio is checked for decoding and timestamp errors. I'll try to make this clearer in the new log summary.

Sorry, I don't plan to release the source code. I've had offers to include this tool in several commercial software packages so I prefer to keep it private in case I decide to sell it in the future.

I'll see what I can do to make the error log format more flexible.

-Mark

Should mpeg2repair be able to log errors in MyTheatre recordings?

Should mpeg2repair be able to log errors in MyTheatre recordings?

mpeg2repair only works with .ts transport streams, and I believe that MyTheater only records mpg program streams, so no it will not work unless you convert the recording to a transport stream, I recommend VideoRedo for the conversion.

Mike T

But VRD will remove any errors in that process, won't it? In which case running M2R will be a waste of time. ;)

But VRD will remove any errors in that process, won't it? In which case running M2R will be a waste of time. ;)

It will remove time-stamp gaps, but not video errors.

mpeg2repair only works with .ts transport streams, and I believe that MyTheater only records mpg program streams, so no it will not work unless you convert the recording to a transport stream, I recommend VideoRedo for the conversion.

Mike T

I have not used MyTheatre in years, but I remember it having some method to save in TS format. Check all the options/preference menus. Also, I think there was something called PID grabber or saver which used TS format.

Looking for some help. I've browsed the thread but didn't really see anything that applied to my problem.

I've downloaded the file as linked to in the first post, extracted the files to a directory on my C: drive, but, when I doubleclick the executable "mpeg2repair.exe" I get a warning box popup with the file location listed and the words "A device
attached to the system is not functioning".

I have no idea why it's doing that. I noticed last night that my AV program, BitDefender, thought the file was a virus and quarantined it, I halted BitDefender from loading today, but I'm still getting that warning popup I noted above.

I have also DL'ed and placed the dll files noted in the FAQ to both the MPEG2Repair directory and the windows/system directory....no luck...BTW, the standard Windows executable icon is used with the file and I thought I've seen a graphical one used with the program..maybe a clue, I don't know, at this point I'm completely lost on what the problem may be.

Is there something else I need to allow this program to work, some VB application(s) or something..??

Thanks.

The below is what I'm getting. I'm guessing it's something specific to my system as I didn't see any other mention of it in the thread, but I have no idea why it's doing that.


http://img.photobucket.com/albums/v637/keenanj/SP32-20070728-130805.gif

Looking for some help. I've browsed the thread but didn't really see anything that applied to my problem.

I've downloaded the file as linked to in the first post, extracted the files to a directory on my C: drive, but, when I doubleclick the executable "mpeg2repair.exe" I get a warning box popup with the file location listed and the words "A device
attached to the system is not functioning".

I have no idea why it's doing that. I noticed last night that my AV program, BitDefender, thought the file was a virus and quarantined it, I halted BitDefender from loading today, but I'm still getting that warning popup I noted above.

I have also DL'ed and placed the dll files noted in the FAQ to both the MPEG2Repair directory and the windows/system directory....no luck...BTW, the standard Windows executable icon is used with the file and I thought I've seen a graphical one used with the program..maybe a clue, I don't know, at this point I'm completely lost on what the problem may be.

Is there something else I need to allow this program to work, some VB application(s) or something..??

Thanks.

The below is what I'm getting. I'm guessing it's something specific to my system as I didn't see any other mention of it in the thread, but I have no idea why it's doing that.



Yeah, that is definitely something specific to your setup. I have never seen that error before. You don't need to download any additional files (they were only needed for very old versions). My guess is that your virus checker did something to the file to keep you from running it since it thinks it's a virus. I would try disabling your virus checker, download again, unzip to new location, and try running again.

I mentioned the virus checker issue in the FAQ - have no idea why it keeps coming up for some people. I think it has something to do with the way I compile the code for this tool - it creates smaller exe files but confuses some virus checkers.

-Mark

Yeah, that is definitely something specific to your setup. I have never seen that error before. You don't need to download any additional files (they were only needed for very old versions). My guess is that your virus checker did something to the file to keep you from running it since it thinks it's a virus. I would try disabling your virus checker, download again, unzip to new location, and try running again.

I mentioned the virus checker issue in the FAQ - have no idea why it keeps coming up for some people. I think it has something to do with the way I compile the code for this tool - it creates smaller exe files but confuses some virus checkers.

-Mark
I tried it with the AV off, it wasn't even loaded, and it still comes up with that error...

I'm going to try on a different machine and see what happens...it's weird..

Thanks for the response BTW. :)

Sorry, I don't plan to release the source code. I've had offers to include this tool in several commercial software packages so I prefer to keep it private in case I decide to sell it in the future.

-Mark

We hold out for vrd to get it, off course at a high price, it is worth it. Though they want you also i remember. And the code dismantled for intergration wasn't it.

Mark shame you couldn't take their money, if enough has been offered. And be a background advisor for a while. Until the new owners get upto speed with the code, and possibly a little longer.

This would make vrd a blinding better app at error correcting than it is now. As well as just reporting the errors, befor and after processing with vrd. With vrd now there is no way of knowing what errors exists in a video. It is why we use M2R for this, before editing with vrd or other editor.

For others if your recordings are precious to you. Run each recording through m2r first before any loading or repair with vrd.

If you use vrd first you will not see all that is wrong with the file after vrd processed it. Test it out for yourself with a known ts file m2r has reporeted errors with.

M2R is the best tool and only tool of its kind. It is why we want Mark to stay with it. It is why Mark knows he can get a fortune for it. Mark don't let it go, for nothing less than a skip full of platinum.

Mark if ever someone buys your code, can you post here who has bought it. Just so we can go and use their software, i'm hoping it will be vrd. Because they do have more beta fixes and update releases. Though all vrd betas are now timebombed, a bad management decision. Leaving many users waiting for the next official releases only. Less users around to report bugs for fixing because many will not install timebomed crippled softwares. I had this problem with M2R at first, thereafter no timebombs, thankfully. Vrd also Have a better public and customer repore than other softwares, through their forum. Also because i would like to think, you are a videoredo user yourself. :D

Note: I have no professional or other relationship with videoredo. I'm just happy to use the best mpeg editor (my opinion) around, at this time.

Mark it would be very helpfull if m2r could support mpv (video) elementary files or vid/audio if possible together. I can load them but cannot run them to log all errors.

But only needed if known that i can after an mpg demux, run through m2r to pick up errors.

I have a substantial number of data disks with mpgs ready for dvd authoring. These go back some years. I'm not sure if it would have the info in te demuxed mpg files for m2r to find any errors.

I know i should have saved them as data in ts format. Though after vrd saving any usefull information for m2r is lost anyway. So i decided way back to save from vrd in mpg dvd authour ready format.

Since using m2r on ts stream files i have noted all errors that remain in each file. And are saved to each disk with the data mpg files. Problem is the 1000s of data mpg disks i have previous to using m2r. These do have errors, but is not known for each file where they all are. Are you able to advise, or help with this with using m2r. :)

@ Chuckybaby


I've just had a chance to try your updated version today v1.3.0.7
Sorry about it taking little longer than planned to trying new version out. Pc's have been busy, more the need for m2r to batch files for processing.

mpeg2repairHelper
http://www.boling.us/ebs/apps/mpeg2repair/

I tried mpeg2repairHelper from a .bat file. Which is east to use and edit compared to the xp cli window.

For batching of .ts files i cannot progress, with the .ts files i record. When i run a file through m2r i need to manually select the program stream number.
An example how i use m2r.

load .ts file
press scan
cancel scan
choose the program stream number
start scan
scan ends
Result = anyfilename.txt log file ouput from m2r with errors reported
It seems that the /SAT switch would be similar, if not the same process. How the program number stream is selected, from a drop list. If you could you add a /switch for the the program number stream selection. Maybe even do the center drop list selection, so they are all available. Adding the program stream number selection /switch would be helpful. ;)

GUI any thoughts on one, for a quick and easy multi file selections. With quick control of switches, for each file batched.

Attached is a command line utility that will automate MPEG2Repair and supports selecting one of the program ID's.

You need to have MPEG2Repair.exe in the same directory as the utility.

The DOS window will display the status of MPEG2Repair as it processes.

Just run the app without any parameters from a DOS prompt and it'll tell you how to use it.

I wrote it in Delphi since it was pretty easy to hook into the MPEG2Repair windows messages. I can't guarantee it's perfect yet. But it's good enough to release it.

Enjoy!

Thanks!

@ shawnph thanks i will try this.

BTW Not tried it yet all pc's busy recording. Except this one that has no sd ts files on it.

Had a look at the cli coomands, what is the cli command for it just to log the m2r scan only.

Sorry misunderstood cli tool.

It is for extraction of sinle programe stream from multi stream file. I thought it was something along the lines of what Chuckybaby was doing.

BTW what do you use to make a cli tool. As you said m2r coomands are easily accessable. Is that tool gui driven and self contained, whilst beign small and not bloated. Is it freeware public domain available, and has support help files to learn it. Did you have to disasemble m2r first. Or have its source code to work with.

Curious, any ETA on last version? Or which product will receive the benefits of your labor?

TIA

Sorry, I don't plan to release the source code. I've had offers to include this tool in several commercial software packages so I prefer to keep it private in case I decide to sell it in the future.

...got into some troubles with my AV,had to re-download it...but ok,
it was mentioned in the FAQ that there are false warning from some AV products.
Then,I saw it was packed with PESpin,and got curious...
why would anyone use PESpin,which is a semi-"hackish" packer,
that has been widely used for packing malware,in a legitimate tool?
So I dumped/semi-unpacked it...few nice stuff I found under it's strings:

usage: xport <-pavtdszn> <infile> <program number> <video stream number> <audio
and later...
Usage: mpeg2decode {options}

...Anyway,I see at least 2 open source programs here,
not only without releasing sources,
not only without mentioning it in the faq/readme/whatever other way,
and without accompanying gpl licences etc....
What I do see though,it's an attempt of..."security by obscurity"...

In short...maybe you should think again,
before trying to sell it's sources to a commercial company? ;-)

In short...maybe you should think again,
before trying to sell it's sources to a commercial company? ;-)

nice, I was wondering why this thing was so damn slow.
usage of opensource explains it.

Just thought it would be useful to post a couple of notes,
say in case anyone wants to have a more detailed look at it...
as PESpin is a quite tricky packer to get fully unpacked...
someone needs to have quite a bit of knowledge regarding PE format,
in order to repair/get a working executable after general "dumping"...
But for basic examination,the repair part is not really needed...

1)The manual (and preferred) way,using OllyDbg...and your mind ;-)
http://www.reversing.be/article.php?story=20050811173217736

2)The semi-automatic way,using pre-made generic unpackers,
you'll might need to play with their settings a bit...
Either with RL!dePacker...
http://ap0x.jezgra.net/unpackers.html
Or alternatively,with QuickUnpack...
http://qunpack.ahteam.org/

In both cases,you'll have to fix the unpacked result,
in order to get a working executable...
but as already said,for basic examination,this isn't really needed...
That's just the basic workflow that I generally use,
when I come across packed executables that alarm my AV software,
and give suspicious results when scanning online at VirusTotal service...

Explains why av protectors i tried stick on the executable for ages.

Mark did say he was going to make a version with usable command line switches. It never happened, we are still waiting for it. Maybe a version from Mark will surface soon, we hope. As a few posts back he was still showing interest.

Given the license side of things that you have pointed out. If Mark has given up on it, the kindest thing he could do is. Release the source or hand the project over to someone else. Just hate it when an active project that works, stops completely. Another coder for future development and implementation. Of the needed requested features, from the past remarks hee for a start. Because it cannot be legaly sold, if is gpl. Mark has nothing to lose by handing it over. As other coders have done in the past, to keep the best softwares alive.

Every possible m2r command line switch available, with full batch processing parsed to .bat file if needed. With a gui to select which files are for processing, so we run your tool instead of m2r initially maybe. As full switches it could do just logging of multi chosen files, unattended scanning.

I'm not a coder but hope that maybe someone. Who loves m2r, as we do can make it happen.

He can certainly sell it if its GPL (or sell support for it, etc) but he must include the source code for it.

Hi,

Sorry I have not checked this thread in a few weeks. Maybe I can clear up a few things about the origin of this tool.

The project started out using the reference decoder provided by MPEGLA. It's not under any open source license. It was very slow and not really practical for any real-world usage. I had to make many changes to get the speed and stability of decoding to a point useful in this tool and to support in-place bitstream repair. Any commercial usage of this tool/code would require an MPEG license (as does any MPEG decoder). This is why the tool is provided as freeware.

If you were around this forum a couple years ago, you would know the origins of this tool. It was intended to repair recordings from the old 169time recording system. At the time, it produced files with many small errors that would cause glitches when watching on anything but JVC DVHS decks. With the help of Ron (Dr1394) we eventually found the source of the glitches and I decided to write a tool to repair them. You can search the archives if you're interested in the full history.

When I first started this tool, it was all command-line driven and only supported MPEG2 elementary video streams as input (.mpv, m2v, etc.). I only used this version for a short time to debug the basic algorithm. Once I decided to release this to the public, I added support for TS input and a Windows GUI. The command-line interface and mpv input is no longer supported although parts of that old code may still be in the source. I have tried to add command-line support again but it just wasn't practical with all the other changes. Luckily some other folks found their own work-arounds.

When I first added TS input, I didn't have access to the ISO/IEC 13818-1 specs, so Ron sent me the source to his xport utility. I used that as a starting point/reference to write a demuxer. Eventually I got the specs, so I wrote my own demuxer and added support for more exotic TS variations and codecs. It also allowed me to write a mini-muxer that was needed to perform the repairs without changing the surrounding error-free packets. The AC3 decoder was written from scratch using the A52 specs on the ATSC web site.

The exe was packed because I didn't want users to easily remove the expiration date on the beta versions and to make the file smaller. Everyone is entitled to release their software however they like. I didn't want buggy versions to live forever on the net so I made sure they would expire. If you don't like it, don't use it. It's freeware.

I stand by my decision to keep the source code private. If Ron requests, I can send him my changes to xport but it's virtually unchanged and no longer used except for the initial PID scan. I just never got around to removing the cpp file from the project.

-Mark

P.S. No ETA on a new version. I just can't find the time. This is why I stopped development in the first place. I will still try to finish it but it's going to take a while.

Mark sometime since your last post, and it is an enjoyable read. So much i have saved it and is in the v1.0.1.3 folder, the version stil only use.

I love the fact that you will try and finish m2r, as much as i love m2r. Thats presently my greatest love of all. Without m2r i would not bother recording from dvb sources, there would be no point. As i need to check each and every recording for errors and log them.

Request: I Still have a huge need for a full batch m2r abilty, m2r logging is the slowest part of all. If could batch while i'm away from pc or asleep overnight. Then comeback to a folder full of m2r log files for each video, ready to go to editing. I would need to manualy set program numbers for every video that are not found automatically, for m2r to log the errors. M2r does not auto find the program numbers for most videos, so with batching m2r abilty i need to be able to manualy input for each video.

Giving m2r batch abilty would give me a better life again. Right now all my spare time is with logging by hand every video with m2r. Slow, painful and verrry boring to do, but has to be done.

Mark sometime since your last post, and it is an enjoyable read. So much i have saved it and is in the v1.0.1.3 folder, the version stil only use.

I love the fact that you will try and finish m2r, as much as i love m2r. Thats presently my greatest love of all. Without m2r i would not bother recording from dvb sources, there would be no point. As i need to check each and every recording for errors and log them.

Request: I Still have a huge need for a full batch m2r abilty, m2r logging is the slowest part of all. If could batch while i'm away from pc or asleep overnight. Then comeback to a folder full of m2r log files for each video, ready to go to editing. I would need to manualy set program numbers for every video that are not found automatically, for m2r to log the errors. M2r does not auto find the program numbers for most videos, so with batching m2r abilty i need to be able to manualy input for each video.

Giving m2r batch abilty would give me a better life again. Right now all my spare time is with logging by hand every video with m2r. Slow, painful and verrry boring to do, but has to be done.


When I have multiple files to scan, I run multiple instances of m2r, it is true they all run at the same time, but I don't think it takes any longer, especially if you run them overnight. I have a dual core, so I can run 2 instances with very little speed slowdown.

I do the same with Quickpar as I create par recovery files for any recording I choose to archive to DVD rom disks. This way if they cannot be 100% read after 100 years, I can recover with Quickpar.

Mike T

@mtallent
I to have tried this, but for me on single intel cpu core, it stalls all open m2r's. It takes forever to complete, even tried using task manager to adjust priority. I also tried on a friends dual core p4 intel 800fsb pc again no a good sollution even using affinity to try help it logging many gb files.

Nothing beats a working in software sollution. Mark i think, may have a near working version for batching. As i recall from memory (please don't quote me). Batching i think is the only, major feature that is missing from m2r.

Nothing that can compare to m2r, and may never be any other. I feel very saddned when Mark speaks of not continuing. And my world is filled with joy and happiness, when Mark says another release will happen.

TBH.. it's totaly pointless capturing dvb streams without having m2r to log any stream errors. This says how dependable, needed and essential m2r is for working with dvb streams.

I was ready to give up with dvb pc recording until i found m2r. I was going to sell the pc and all, because finding errors by watching the video was a pointless exercise. This is why m2r is the greatest software there is for me.

Thank you Mark and i hope m2r development continues. :)

...

The exe was packed because I didn't want users to easily remove the expiration date on the beta versions and to make the file smaller. Everyone is entitled to release their software however they like. I didn't want buggy versions to live forever on the net so I made sure they would expire. If you don't like it, don't use it. It's freeware.

I stand by my decision to keep the source code private. ...

P.S. No ETA on a new version. I just can't find the time. This is why I stopped development in the first place. I will still try to finish it but it's going to take a while.


I can respect all that, but PLEASE don't let this code be lost due to lack of interest. M2R is unique piece of code that many of us cannot live without. It would be a tragedy if the source was lost after you burn out. Please find a good home for it when you decide your done.

I have a 27GB ts file produced by TSRemux from a Blu-Ray rip that plays fine but produces many discontinuity errors and eventually crashes when running in ConvertX2DVD. Trying to run it through MPEG2Repair it produces a stream of "No TS sync byte" errors and eventually produces a fatal error.

Can anyone explain please? As I say it plays beautifully in WMP, Crystal, VLC or PowerDVD

I think M2R is one of the greatest tools I have seen in quite some time. Agreed with others on the forum I would not be without this tool as part of making TS recordings and verifying them afterwards.

One thing that would be nice is if it was ported to Linux or at least command line support on Windows, Specifically if it was supported on a Dreambox. Then I would be able to make recordings of a TS stream directly off the Satellite or Cable, then verify that the TS stream was valid afterwards and correct any errors that were made. The only downside is the Dreambox CPU is pretty underpowered (250Mhz, 350MIPS) so analyzing and repairing TS's could be a time consuming exercise.

I do however agree that many would be lost without this tool, so I hope it does not disappear!

shawnph: I just checked out your mpeg2repaircli-auto utility. I like your idea of using the windows messages; although I had a couple of problems where I'd launch it and it would never type anything into the mpeg2repair window, I think it has the potential to increase the speed and reliability quite a bit over my delay-and-blindly-type technique. Unfortunately, I've little experience interacting with MS Windows, so I'm not capable of much more than that. :-)

If we can get the features my program (mpeg2repairHelper) offers into yours and it proves to be fairly reliable, I wouldn't at all mind seeing mine disappear!

mpeg2repairHelper now allows you to select which audio track you want to use.

See http://www.boling.us/ebs/apps/mpeg2repair/


PS: I don't normally hang out on this forum, so if anyone expects a reasonably quick response from me on anything, I recommend emailing me at the address shown on the web page.

Thanks for the info Mark, i use M2R at the moment only to analyze the streams, not for repair. I have tried repair previously, but disliked the (black) blocks it placed on the video. Times i may see an error is likely to be few frames. M2R locates any errors, which i then replace using clips, from a repeat showing. This way i get nice video, as shown.


Nexin, how do you go about replacing the bad frames? What tools do you use for this? Is there an automatic way of combining two imperfect streams to take the best frames of each and put them into a single stream? I'd love to know, as that would make a nice addition to my recording/processing setup with MythTv.

For anyone wondering about the black blocks that mpeg2repair uses to replace corrupt macroblocks, that feature can be turned off in the .ini file, as mentioned earlier in this thread or in the documentation.

P.S. I have posted a proposal on the doom9 forums to create a library of broken mpeg2 transport stream .ts samples. This library would be useful in testing the robustness of software and hardware stream repairers and players. If anyone finds this proposal interesting, feel free to post your comments (and links to your broken streams). Preferably all can be posted on my doom9 thread so that that discussion can stay consolidated. My proposal: h**p://forum.doom9.org/showthread.php?t=134693
(can't post links yet; replace *'s with t's)

Just thought it would be useful to post a couple of notes,
say in case anyone wants to have a more detailed look at it...
as PESpin is a quite tricky packer to get fully unpacked...
someone needs to have quite a bit of knowledge regarding PE format,
in order to repair/get a working executable after general "dumping"...
But for basic examination,the repair part is not really needed...

1)The manual (and preferred) way,using OllyDbg...and your mind ;-)
h**p://www.reversing.be/article.php?story=20050811173217736

2)The semi-automatic way,using pre-made generic unpackers,
you'll might need to play with their settings a bit...
Either with RL!dePacker...
h**p://ap0x.jezgra.net/unpackers.html
Or alternatively,with QuickUnpack...
h**p://qunpack.ahteam.org/

In both cases,you'll have to fix the unpacked result,
in order to get a working executable...
but as already said,for basic examination,this isn't really needed...
That's just the basic workflow that I generally use,
when I come across packed executables that alarm my AV software,
and give suspicious results when scanning online at VirusTotal service...

I was recently trying to get mpeg2repair to work under Wine 0.9.4.6 and 0.9.54 on Ubuntu. Nothing I tried worked. The error prompt is:
Titlebar: Microsoft Visual C++ Runtime Library
Window contents: Runtime Error!
Program: This application has requested the Runtime to terminate it in an unusual way. Please contact the application's support team for more information.

Could this be related to the PESpin packer? Would an unpacked release of the last version of mpeg2repair be possible, to hopefully make it compatible with Wine?
Wine Appdb entry:
h**p://appdb.winehq.org/objectManager.php?sClass=application&iId=6629
(can't post links yet; replace *'s with t's)

Nexin, how do you go about replacing the bad frames? What tools do you use for this? Is there an automatic way of combining two imperfect streams to take the best frames of each and put them into a single stream? I'd love to know, as that would make a nice addition to my recording/processing setup with MythTv.

I use VideoRedo+ or VideoRedo TVSuite only. I log only the errors, i never repair with M2R.

If you have a reshow with same resolution that is clean, all well and good. Otherwise i edit scenes with VideoRedo for both sources and use the joiner funtion to make a good file.

If no second reshow is available i cut what isn't talking, Or if Audio dropouts a cut those also. And keep the M2R log updated with the new video times. By using VideoRedo preview you can see the output video time to copy into the m2r log. This i save onto dvd, i can then edit at a later date for a clean recording.

The only bug bear there is With M2R is that there is no batch for running many files through at once. You have to baby sit the thing to get all files logged. It does take forever with many files, and is Very Boring doing it. Oh the option to batch them and come back later, all automaticaly done.
- dream on....

And if someone decides to do this, not all my ts streams are seen by M2R i have to select from the M2R Program drop list for half of them.

I get this stupid runtime error dialog not only with WINE, which could be a quite normal WINE incompatability, I also get it in a fresh W2k installation on VMware! How's that for being totally incompatible to anything, VMware usually doesn't have any problems like that at all!

I don't own any (usable) Windows machines here, so this stupid packer makes your otherwise excellent software completely useless for me and many others.

If you want to protect your IP, I can accept and respect that easily, but please repack it with some kind - or version! - of packer/protector/whatever that keeps the software actually executable!

I get this stupid runtime error dialog not only with WINE, which could be a quite normal WINE incompatability, I also get it in a fresh W2k installation on VMware! How's that for being totally incompatible to anything, VMware usually doesn't have any problems like that at all!

I don't own any (usable) Windows machines here, so this stupid packer makes your otherwise excellent software completely useless for me and many others.

If you want to protect your IP, I can accept and respect that easily, but please repack it with some kind - or version! - of packer/protector/whatever that keeps the software actually executable!

It's apparently only useless to you, not "many others".

Of course you can always ask for your money back.

Note: Version 1.0.1.5 is now available (02-15-07).

Please download the attached file or visit:

http://homepages.roadrunner.com/mwilczyn/mpeg2


What's new:

1.0.1.5 / 02-15-07:

• Added "bad-edit" detection to show inconsistent telecine flags. Use "IgnoreFieldRepeatFlagWarnings" ini setting to disable.
• Added "ReadBufferSize" and "WriteBufferSize" ini settings. Might improve file i/o performance if you're not CPU limited.
• Improved performance of initial PID scan on slow wireless networks.
• Added "FileWildcard" ini setting to allow custom input/output file extensions.
• Added limited H.264 video support. Only scans for Transport/PES header errors.
• Removed beta expiration date. Barring any major bugs, this is the final version of MPEG2Repair. Moving on to other projects...



Hello,

I'm pleased to announce the public release of my new Mpeg2 error detection/repair program called "MPEG2Repair". It's designed for testing and repair of HDTV transport streams. Features:

-Error Detection:

Find file errors resulting from signal drop-outs, faulty tapes, encoding errors, etc. Detection code checks transport layer for continuity errors and faulty headers. Code also checks the MPEG2 PES layer for parsing errors (corrupted macroblocks, slices, headers, etc.). All errors are logged with time stamps and file positions so you can easily locate them inside your original recording. Very useful for verifying files prior to permanent archiving.

-Error Repair:

Automatically repair minor Mpeg2 transport and PES errors which may cause problems for certain decoders. Tool will remove erroneous data from the file by either cutting it out or replacing it with valid bits. The goal of repair is to create a fully Mpeg2 compliant file that will not crash, hang, or stall decoders which were not designed to deal with errors. Repair is currently not intended to improve the visual appearance of errors but simply to remove them.

-Improved Decoder Compatibility:

Automatically fix other issues that may affect decoder compatibility. These include partial GOP's, missing headers, incorrect continuity, etc. This feature is especially useful for those of you with old recordings from the 169time system. They should now play smoothly on other decoders besides the JVC 30K deck.
my anti-virus stop me going to the site said it was a hacked site
is this program a scam

It's apparently only useless to you, not "many others".

Of course you can always ask for your money back.

Well, to be fair, I'll bet there are many others that have attempted to use this app and have run into the same problem and just haven't bothered to mention it, myself included.

I'm not sure why the creator has chosen an executable/packer that has a habit of triggering AV protection so much, why not use something more commonly used/accepted?

But it's his to do with what he wants, and you have to respect that, even if it means you don't get to use it.

Well, to be fair, I'll bet there are many others that have attempted to use this app and have run into the same problem and just haven't bothered to mention it, myself included.

I'm not sure why the creator has chosen an executable/packer that has a habit of triggering AV protection so much, why not use something more commonly used/accepted?

But it's his to do with what he wants, and you have to respect that, even if it means you don't get to use it.

Suggest you get a better AV as yours obviously sucks.

Its a friggen .zip file for god's sake. A Zipfile is very widely accepted and has been for 20 years.

Inside is a simple 1 file .exe program that runs without an install. Its not much simpler than that.

Again, get a better AV that doesnt produce false positives!

NOD32 comes to mind - and it will also give you 6%-20% of your processor ing power back.

my anti-virus stop me going to the site said it was a hacked site
is this program a scam

Goes triple for you.

Suggest you get a better AV as yours obviously sucks.

Its a friggen .zip file for god's sake. A Zipfile is very widely accepted and has been for 20 years.

Inside is a simple 1 file .exe program that runs without an install. Its not much simpler than that.

Again, get a better AV that doesnt produce false positives!

NOD32 comes to mind - and it will also give you 6%-20% of your processor ing power back.



Goes triple for you.

It's not my AV that's doing it, besides, I don't even have any AV on the machine I just tried it on. I understand it's a zip file, but something about it throws flags, and it's not exclusively due to on-machine AV software.

The below is what pops up from Google. Maybe the site owner can request to be removed from the blacklist...?

Safe Browsing
Diagnostic page for homepages.roadrunner.com/

What is the current listing status for homepages.roadrunner.com/?

Site is listed as suspicious - visiting this web site may harm your computer.

Part of this site was listed for suspicious activity 13 time(s) over the past 90 days.

What happened when Google visited this site?

Of the 1756 pages we tested on the site over the past 90 days, 147 page(s) resulted in malicious software being downloaded and installed without user consent. The last time Google visited this site was on 09/23/2008, and the last time suspicious content was found on this site was on 09/23/2008.

Malicious software includes 196 trojan(s), 101 adware(s). Successful infection resulted in an average of 35 new processes on the target machine.

Malicious software is hosted on 31 domain(s), including viacodecright2.com, viacodecright1.com, 91.203.92.0.

1 domain(s) appear to be functioning as intermediaries for distributing malware to visitors of this site, including siski.info.

Has this site acted as an intermediary resulting in further distribution of malware?

Over the past 90 days, homepages.roadrunner.com/ did not appear to function as an intermediary for the infection of any sites.

Has this site hosted malware?

No, this site has not hosted malicious software over the past 90 days.

How did this happen?

In some cases, third parties can add malicious code to legitimate sites, which would cause us to show the warning message.

Next steps:

* Return to the previous page.
* If you are the owner of this web site, you can request a review of your site using Google Webmaster Tools. More information about the review process is available in Google's Webmaster Help Center.

So just stay away from the hacked website (for now) and download the zip from the attachment in Post #1. ;)

hehe...the most obvious answer of course. :D

That's how I got it the first time, although it kept "encountering problems" part way through using it. I'll grab it again and see what happens. :)

My AV McAfee says theere is a trojan in the actual zip file and quaranteens it.

Russell

Almost certainly a false positive, but you can submit the file to one of the online scanning services to double check it.

Thank you!!!!!!!!!!

Is anyone still working on this program? :confused:

Every so often I get an error message that the beta has expired and to download the latest version. If it's not an active project having a timeout isn't very helpful.

Is anyone still working on this program? :confused:

Every so often I get an error message that the beta has expired and to download the latest version. If it's not an active project having a timeout isn't very helpful.

The time out was removed in the last version 2-15-07 see first page of this thread.

Mike T