Is this possible?

While simply putting in a disc and expecting it to work is probably crazy, there might be the option of creating a MD cartridge that can read the VCD and interpret it into video. I'm not sure if the video can be decompressed in real time on the Mega Drive, though.

 

VCD uses MPEG1 for video compression. Since the Sega CD had no dedicated MPEG1 hardware decoder, you'd have to do it software. I highly doubt that a Sega CD has the horsepower to decode an MPEG1 stream.

 

This also depends on whether the Sega CD drive can read the Mode 2 Form 2 ISO9660 format that VCD's are formatted with (removes the error correction from Mode 1 to get another 288 bytes of memory per sector).

This is slightly off topic, but is it possible for the Sega CD to read raw CD-DA data to one of the CPU's (not the audio DAC)?

 

MCD wouldn't even have sufficient video resolution and color output for it to look anything good. There's a reason why all Sega cd fmv games looked like shit.

Saturn needed special decoder cards, but I wouldn't be surprised if it was possible to do in software - very late saturn games used cri sfd videos which had mpeg1 compression. Granted, they weren't running fullscreen, but there is at least potential.

 

Well, the VU meter in the BIOS needs to get its data from somewhere, don't you think? (although I didn't check what could be used to do this...)

 

Perhaps it could be pulled off by emulating MPEG-1 compression on the 32X. You'd have a wider range of colors and more powerful hardware to work with then. It would also put the Sega CD 32X to use. :P

I have no idea if this could actually work. It's all theoretical.

 

Well, there's only one way to find out...

 

Well, here's an example of Night Trap for the CD 32X:

http://www.youtube.com/watch?v=A32FXJIsn6A

It looks significantly better than the standalone Sega CD version, at least.

 

Back in the day, MPEG-1 encoders and decoders weren't well optimized, so it is quite possible the reason they required additional hardware was because the software ones weren't efficient enough. Nowadays, a specially compiled ffmpeg containing ONLY mpeg-1 encoders and decoders is quite optimized and doesn't use nearly as many CPU cycles as old encoders and decoders available.

Additionally, all patents for MPEG-1 have expired, with the exception of MPEG-1 Layer III audio, otherwise known as the MP3 format. So you could even use MPEG-1 (without Layer III) for commercial work and generally be okay.

However, I wonder if it would be possible to use the less complex Ogg Theora and Ogg Vorbis formats for FMV...

 

MPEG-1 can and has played on 386 machines before. Granted, playback SUCKED, but it did work. It would probably be easier to try to use Ogg formats on a 386 because it is computationally simpler than the MPEG-1 format.

 

No way, the Genny's 68k nor SCD's 68k wouldn't be fast enough to do realtime MPEG-1 decoding D=.
I think the only way to do that would be having some hardware decoder... so not. :P

 

You won't get real-time playback of MPEG1 video on the 32X. You'd be looking at 80x60 10Hz video (or worse). There's a reason nearly all 32X FMV is cinepak... the same reason 386's and the like used cinepak or similar level video codecs. That's not to say the 32X isn't capable of better - with the advance in codecs, some techniques used in more recent codecs could be used to make a better codec aimed at the 32X specifically. Just don't fool yourself that you can run VCDs on the 32X... it would be lucky to decode the audio on the VCD in real-time.

 

Wow, no, the 386 is definitely much faster than the 68k =/

 

I think he meant the SH2. The SH2 could be considered about the same power as the 386. I think the SH2 is easier to program, though.

 

I don't know how relevant this is, but I'm linking it anyway because it's amazing.