Monthly Archives: April 2006

I Have More Games Than You

Well, probably. In my quest for weird and wacky multimedia samples I have amassed quite a collection of game titles. The number is well in excess of 400 right now. Not a hardcore collector by any stretch of the imagination. But it’s more than enough for any member of the opposite sex to write me off as an emotionally stunted, overgrown adolescent, despite the fact that I rarely actually play any of these games.

I catalog the games’ multimedia technology in my Multimedia Exploration Journal (and I presently have 70-80 games to process for the journal). If you care what all the games are, here is the master spreadsheet I maintain:

Mostly, the spreadsheet is to help track which information still needs to go into MobyGames. I have not updated the spreadsheet in a little while so some of the red and yellow cells might be a little inaccurate w.r.t. the database.


Do you have any dusty old games stuffed in the closet? Do you have at least mild obsessive-compulsive tendencies? Check the MobyGames database and see if you have anything to contribute with those old games– box and media scans, game screenshots, reviews, the game itself if necessary– I have dozens of titles that still don’t exist in the database, expansive though it may be (27,478 games as of this post).

The New FourCC Authority?

One of my original motivations which led to the creation of the MultimediaWiki was to expand on the knowledge enumerated at, heretofore the internet’s foremost authority on the curious multimedia concept known as the four-character code (FourCC). With the latest update of the MediaWiki software, I find that I am able to categorize FourCC redirect pages. What this means to the Wiki lay person is that I can use the Wiki to automatically maintain a list of all known video FourCCs, which I have done. I have done the same for audio FourCCs, though the list is not as extensive (mostly applies to QuickTime and Real codecs). I hope to give the same treatment to Microsoft 16-bit audio IDs soon.

So far, I have only catalogued the FourCCs and codecs that I can prove exist, either because we have samples, codecs, or both that correspond to the codec. The list contains dozens of FourCCs for which I can find no samples or codecs. It’s reasonable to believe that they existed, perhaps at the dawn of the consumer multimedia era. It also could be that certain FourCCs were formally registered with Microsoft by ambitious companies that were never able to release their multimedia programs that would have generated the corresponding data.

What to do about these? I don’t wish categorize them along with the provable FourCCs. I may create a different page or category for these strays until they can be claimed either by the discovery of actual media samples in the wild or by codecs (source or binary, coders or decoders) that can handle the data.

Serious About VC-1

I have a plan, I really do. I am still working hard on Understanding VC-1. Just so that you, the reader and open source multimedia enthusiast understand correctly, this is not so that we can build a new implementation that mirrors the reference implementation 1:1. What I am doing right now is more or less the “phase 1” of this particular documentation effort. Phase 2 is developing a more top-down specification in a similar spirit as my old VP3 format document. Phase 3 is to generate a development roadmap with specific tasks and milestones as well as a number of sample vectors for individual modules which should facilitate initial component validation as well as ease development of SIMD-optimized components.

So you see, phase 1 is where I essentially use the MultimediaWiki as a scratchpad. VC-1 is, like, really big and hard to get one’s head around. This process is helping me.

Google Logo

Part of my plan in all of this is that I am hoping to pawn off the actual writing of code on other people. I know there are people out there — even people reading this blog — who are interested in making some material contribution to open source multimedia. On Monday I learned of Google’s second annual Summer of Code initiative where, in a nutshell, college students can get paid to work on approved open source projects. Since the project as outlined above, with the roadmap and deliverables/milestones, seems ideal for this mentoring concept, I submitted FFmpeg as a project for consideration, albeit a few hours after the formal deadline for doing so. Since the SoC is quite popular, it’s likely that they have all the projects they need.

However, if a student is out there following this blog and interested in working on a very important piece of open source multimedia work, you will have the opportunity starting May 1 to submit an application to work on a project not listed with an official mentoring organization, per this SoC FAQ entry. It’s a long shot, but Google has sponsored more ambitious stuff.

A big task? You bet. But if you follow this blog you’re probably at least minimally qualified to tackle the project.