Playback of RRS 2.0 videos with Marvel G400

[Maximilian]

Hello videofans,
early 1998 I bought a Mil2/RRS and recorded up to now about 100 CDs with MJPEG video clips (704x576, audio CD quality, 2MB/s = 5:37 min video per CD (6:02 on 700 MB CD), Matrox MJPEG driver 2.0). They play excellent on my 32x CD-drive, the quality is very good and there seems to be no problem until the day when I tried to render my files in another codec.

The problem is as follows: the files recorded seem to contain corrupted parts which are accepted by the Matrox 2.0 codex and played without any visible fault but only by this codec, not by any other MJPEG codec I have tested up to now. E.g. the Matrox 2.1 driver displays a change of the field sequence from time to time which result in inacceptable frame shake. Trying to render in e.g. 352x288 Indeo 5 yields black frames after the first corrupted part is reached, and playing the file with other software MJPEG players gives the same result.

As I want to buy a Marvel G400 in the next time, it am afraid that I have to record every clip once more. Does anybody have a better idea? The Matrox service in Germany has not.

My system: First W95a/Gigabyte TX3 board/P233MMX/64MB/Quantum Fireball SCSI 3,2 GB 3 MB/s, normally 0 dropped frame, Matrox PD 4.11. Later W98SE/K2-366/IBM DDRS SCSI 8 GB, 6 MB/s, no change noticed.

[Maximilian]

I have expected that I get the message that the Marvel, of course, has a tool enclosed which converts older matrox file formats to the new format as this is executed by Pinnacle systems. This seems not to be the case. On the other hand I found about 20 older messages in this forum from users having the same problem so I am at least happy that Im not the only one having this problem.

From this older messages I got the idea to override the RRIcm.dll and RRIcmExt.dll of version 2 with the ones of version 2.1. The result is that the 2.0 videofiles can now be converted in other formats. However, there is sometimes a change in the field order sequence which has to be corrigated e.g. with Premiere which allows field order reverse for these parts. As this method is possible but laborative it is in most cases easier to record the files once more.

For this reason I am happy that my RRS is working fine and that I don't need a new system at once.

[Captain Marvel]

RRS is field order "a" Marvel G200 & 400 is field order "b". SUonds like maybe evil frame described in other posts.

[Maximilian]

At the Cebit in Hannover I had the opportunity to play old RRS files with the Marvel 400. The results are as follows: Astonishing, but true, although the Marvel has an other field order than the RRS, files recorded with driver 2.1 play perfect in the correct field sequence with the Marvel.
Some files recorded with the 2.0 driver played also perfect with the Marvel. Some files recorded with the 2.0 driver showed an inversion of field order during playing. These files can be repaired if the morgan-multimedia software mjpeg-codec is installed. Method one: changing the field order of the corrupted parts. Method two: recompress them in the matrox mjpeg codex. The morgan-multimedia software mjpeg codec is the only codex I found which allows transforming avi-files of Matrox driver 2.0 in other video codecs.