Posts for Multimedia_Mike

Posted: 8/16/2005 3:44 PM
Quote
Multimedia_Mike
Experienced Forum User
Joined: 8/2/2005
Posts: 6
Location: Colorado, USA
Jyzero wrote:
I'm curious to see how this idea would perform : Dump every data sent to the PPU and compress it. Then, make a video codec that emulate a PPU and process the data back
Interesting idea, but as another poster pointed out, getting timing right can be tricky. Plus, if you carry this approach too far, you may as well be working with an emulator, the ROM image, and the movie-action file. Also, it would be limited to NES data, while we have a good opportunity to design this codec to cope with data from the 8-bit Atari all the to the Neo-Geo. -Multimedia Mike
Posted: 8/4/2005 3:39 PM
Quote
Multimedia_Mike
Experienced Forum User
Joined: 8/2/2005
Posts: 6
Location: Colorado, USA
Warp wrote:
I once experimented with lossless video compression optimized for at-most-256-color tile-based scrolling videos. Here are some results (no sound, just picture information): First 1109 frames of the SMB3 video: 263320 bytes. First 5492 frames of the Snake Rattle&Roll video: 2248127 bytes. First 8646 frames of the Vicedoom video: 3079616 bytes. While the results were promising, they were only a bit smaller than DivX-encoded versions of the same frames with acceptable quality. I decided that it (at least my idea) was not worth the efforts: While you would get lossless video, it would require people to download and install an obscure codec, while at a comparable file size and acceptable quality it would be enough to have the far more popular DivX codec installed. Moreover, with the introduction of H264, which has even a better quality/size ratio than DivX, the idea became even more obsolete.
Is there a website where I can learn more about this optimized for at-most-256-color tile-based scrolling videos codec? Obviously, I am very interested in studying this domain. Lossy codecs (like MPEG-4 & H.264) may be "good enough" for most users. But I know of a few people (myself included) who would like to see an efficient lossless codec for this type of data. I write it all off as an academic exercise. :-)
Posted: 8/4/2005 3:35 PM
Quote
Multimedia_Mike
Experienced Forum User
Joined: 8/2/2005
Posts: 6
Location: Colorado, USA
OmnipotentEntity wrote:
The way you seem to be coding it this will have applications outside of video game movies (most any animation will benefit really, as animations tend to png compress well, which *seems* to be the idea behind your codec). But you don't seem to have implimented a low motion compression based scheme, which will be a great plus to the codec if you can get it hammered out. I'm actually very interested to see where this goes.
That makes 2 of us. :-) The intraframe compression I am experimenting with is quite different from PNG. But as you note, the interframe compression is going to be the most important part of this codec. I am prototyping various ways to exploit the inherent interframe redundancy in these videos (so much of the frame remains the same or just shifts by a few pixels). The patterns are certainly present. We just need to find efficient methods for a computer to notice them.
Posted: 8/2/2005 7:02 PM
Quote
Multimedia_Mike
Experienced Forum User
Joined: 8/2/2005
Posts: 6
Location: Colorado, USA
Multimedia Mike wrote:
Not sure about the AVI-capable DVD player. However, per my understanding, it is tricky to get videos to work in those things in the first place.
Consider, also, that an AVI file encoded with this hypothetical new video codec will retain 100% of the original video information. If an end-user were so inclined, he could download the file and transcode it to a high-bitrate DVD-compliant MPEG-2 video stream for burning onto a DVD. The DVD disc would play in a much larger range of standalone players. -Multimedia Mike
Posted: 8/2/2005 6:44 PM
Quote
Multimedia_Mike
Experienced Forum User
Joined: 8/2/2005
Posts: 6
Location: Colorado, USA
AngerFist wrote:
Hi M-Mike. If you manage to encode a certain avi (movie) file with better picture (the currenct codec, H264 has outstanding picture quality) and sound than the current codec, I think everybody would be much pleased, though Im a bit concerned how windows media player and other media player would react to your "home-made" codec and how a avi playing dvd player would play your movie.
At this point, I am only hypothesizing on a better video codec. However, there may be an opportunity for better audio coding as well, perhaps a lossless codec that takes advantage of certain characteristics of synthesized square & triangle waves, such as those emanating from the NES. Since the goal is 100% lossless compression, I am confident that image quality will be acceptable. The big concern is how it stacks up size-wise. For playing under Windows, someone (likely me) will need to make some downloadable codec modules available. Same for QuickTime under Mac OS. Fortunately, this whole effort will be open source and anyone will be free to jump in. Not sure about the AVI-capable DVD player. However, per my understanding, it is tricky to get videos to work in those things in the first place. Thanks for the feedback. -Multimedia Mike
Posted: 8/2/2005 5:26 PM
Post subject: Custom Video Codec For Console Movies
Quote
Multimedia_Mike
Experienced Forum User
Joined: 8/2/2005
Posts: 6
Location: Colorado, USA
Hi, I call myself Multimedia Mike because I am obsessed with multimedia technology. I also love watching these console time attack videos. One big drawback, I think, is the compression artifacts present with lossy coding algorithms such as MPEG-4 (a.k.a. DivX) and H.264. This is because such codecs are not designed to encode synthetic video very well. To that end, I am experimenting with and prototyping a new video codec that should be better suited for encoding console movies such as time attacks. It is designed to encode synthetic video data (vs. realistic, filmed video data) losslessly. Anyway, I just thought maybe people on this forum would like to know about my little project. I call it the Palettized Animation Video Codec, or PAVC. If you are interested in what I have done so far, I have been writing about my experiments here: http://multimedia.cx/eggs/index.php?cat=20 (Heh, ignore the fact that my latest experiment, from August 1, crashes my computer; I am pretty sure that is a hardware problem.) -Multimedia Mike