In general, a movie which is slower than the existing fastest movie will be accepted only if it has a sensible and rational secondary goal which explains why it's slower (such goals being typically things like 100% completion or using a suboptimal playable character).
A run in such a different category will be accepted only if the goal choice is sensible and adds to the entertainment value. In general, the amount of different TASing categories for a single game will tend to be minimized in order to avoid clutter, so there should be a really good reason for the goal choice.
It's better to consult the community before spending the time in creating such a run and then have it rejected.
No. The movies marked with stars are recommended movies for first-time visitors. They showcase what TASing is all about.
Some of the best movies in the site are not starred because they are not necessarily deemed as the first introduction to TASing (eg. because they are long and complicated, and someone who isn't so much into gaming in general or speedrunning in particular may find them boring because they don't understand what's going on). Also, starred movies tend to be selected from a wide variety of game genres, rather than strictly taking the top n highest-rated movies.
I suppose not. I just can't see the reason why XML ideology must be imitated, as it only adds useless clutter. (OTOH, even XML supports single tags in the form of <tag />.)
But I suppose that if it would require an extensive refactoring of the phpbb engine to support single tags, it's not worth it.
That may actually be an idea to consider, especially if it introduces significantly more places which require special maneuvering because of the no-coins restriction.
I'm not sure I agree with that.
A goal/restriction in TASing a specific game is ok if:
1) it's completely well-defined and unambiguous (ie. not up to interpretation).
2) Introduces a significant level of challenge and/or entertainment to the run (well, the former is usually sensible only if it also implies the latter).
For example the "maximum kills" goal in the Gradius 3 TAS is sensible because it's both challenging (even with tool-assistance) and introduces a level of entertainment (because it's cool to see everything destroyed in such a hard game).
Additionally, a TAS which uses a special goal or restriction does certainly not need to be as fast as the default "anything goes" run, so using that as an argument for rejection is completely invalid.
That said, the problem with this particular run is that while the goal/restriction is quite well-defined and potentially introduces a significant challenge, unfortunately the game levels have been designed so that there are only two places where the restriction really necessitates special maneuvering. That might not be enough to justify publishing the run under a new category...
I really don't think it harms anybody or anything if even a clear "YES!" submission is published a couple of days later. I don't think anybody would get that impatient. (Besides the vast majority of submissions get pre-encoded to youtube of whatever, so it's not like people would have to wait the 72 hours to see it.)
(And besides, it's not completely unheard of that a supposed improvement to an existing run has been rejected, for whatever reason. "Improvement to existing run" is not, and should not, be some kind of "automatically accepted".)
Nitpicking, but there is no image quality difference between lossless compression and uncompressed video. That's what the term "lossless" means in the first place.
(Of course it might be that "H.264 lossless" isn't lossless after all, unlike the name implies...)
You are assuming that the computer always plays in a deterministic way, with the same answers to all your moves. While that might be the case with very primitive chess programs (probably many of the antiquated chess programs made for the consoles of the 80's), most of the competent chess programs do not answer all your moves deterministically.
(In fact, computer chess programmers deliberately take measures to avoid the program always answering your moves in the same way for a rather obvious reason: It would make playing against the computer rather boring once you have beaten it once, as you could always beat it again with the same moves, which wouldn't make much sense and would greatly diminish the value of the program as a learning tool.)
What do people here think about my suggestion? To recapitulate: Take a "normal" chess program (rather than combat chess), perhaps for a faster console (eg. the snes or even the N64 or the PS1), set the computer difficulty to something which is bearable but still gives the computer a somewhat fair chance of defending itself (eg. something like 10-15 seconds per move), and then beat the computer in as few moves as possible.
(If the chess game supports showing what the computer is "thinking", eg. as a list of moves, definitely turn that feature on. It makes the waiting a bit more interesting.)
Of course the limit of how much the computer is allowed the think will necessarily be rather arbitrary. Perhaps if the game has some pre-defined setting (which ends up with the computer thinking about that 10-15 seconds per move), it would be an obvious choice.
I'm not exactly sure what you are saying there. It seems that you misunderstood what I wrote. I was not talking about the bitrate or bandwidth needed for a video with a large resolution. I was talking about the processing power needed to decode (ie. decompress) a high-resolution H.264 video stream, which AFAIK increases as the video resolution increases (which is why older Pentium 4 computers have hard time playing H.264 videos which have HD resolutions).
(OTOH, I don't know if this is only for HD videos which truly have additional image information in them. It might be lighter to decode if the video consists of a small-resolution original material which has simply been scaled up...)
I have to agree that the run is technically commendable, but that the game choice is poor. In theory it would sound like battle chess would make for a more entertaining run because of all the animations and battles, but in practice the animation (especially moving the pieces) is so slow that it's painful to watch. Unless there's a way to significantly speed up the animations (I suppose there isn't an in-game setting for this) the end result is quite boring.
At least to me it would be even more interesting to see a TAS of a plain regular chess game where the computer is allowed to think eg. something like 10 seconds per move. It would also be more admirable in that the computer is actually given a fair chance to defend itself, rather than the TASer relying on not letting the computer think at all...
Also take into account that the computing resources required for decoding (ie. playing) an MPEG-4 video is, AFAIK, proportional to the resolution of the video (although I have no idea by which function), and decoding H.264 is a surprisingly heavy operation. Which means, in simpler words, that the larger the resolution of the video, the more CPU power is needed to play it.
This can already be seen in HD resolution MPEG-4/H.264 videos, which slower Pentium 4's have a hard time playing in real-time (and even the fastest Pentium 4 struggles to keep up). Not to talk if you were to dramatically increase the resolution even further...
Not everybody has a top-of-the-line quad-core computer.
I suppose you are right. (And given that our encoders go through a lot of trouble to make the video files as small as possible while maintaining maximal image quality, something which not many people bother to do, I'm assuming that our videos are by no means the largest ones people are uploading to archive.org...)