Because NBMB is a glitched-out run made for lulz. It has novelty value, but is nowhere near the entertainment level of full runs. In-game category is the only one now that has similar idea in mind, so it would make sense if NBMB will obsolete it, seeing as it will have pretty much the same goals, aside from skipping bosses.
Perhaps a realization that pretty much every old movie is, in fact, realistically improvable — even those deemed perfect. Donkey Kong improvement came right at that time.
[EDIT]
Also, seems like that time also more-or-less coincided with saturation of TASed games' library. Seems like most new games, especially on NES, are nowadays TASed by Chinese and Japanese players (not without a reason). Or by Aqfaq, if talking about Genesis.
Totally offtopic, but this is some awesome gravedigging; with the original post being answered 2.5 years later — and still being a useful answer at that. :)
Arguing semantics. Why bringing up physics textbook definition when we aren't even measuring distance to begin with? The concept we use instead of distance isn't uniformly defined, which is why different goals are there. Aiming for ingame clock = aiming for faster gameplay. Aiming for realtime clock = aiming for lesser overall time. It's a matter of personal taste.
And I repeat, taking less time isn't always perceived as being faster, and perceiving the runs is the purpose of their creation. Not lower numbers.
It's not really "hence", although it does contribute to it. The content management solution we discussed with Warp about half a year ago would take care of the issue, but I'm afraid it's going to take years until it comes (and it may not, after all).
[EDIT]: Heh, only three months ago, actually.
I agree that the issue of difficulty takes place here, but it doesn't mean the easy way is the best way, and it is a recognized problem. And there are precedents when using less frames wasn't the deciding factor in judgement of some movies, but I'm already tired of digging them all up for nth time, so do your own research if you want.
It doesn't sound good mainly because you're measuring speed in time units here. "Takes less time to do" != "faster" — this is the point. And if you want to argue the actual gameplay speed, take a look at the speed addresses, and you'll get your movement speed in the units the game measures it in.
In most cases, "good" means a form of benefit. There will always be some minority that won't get the benefit. What said minority will consist of can depend on anything. Major bias will therefore derive them of this benefit further. This will lead to conflicts.
An example from real life: navigate a busy city on a good car. It will feel faster, but in the end, you'll likely lose to puny subway that doesn't have to stop on crossroads.
Speed is subjective; this is a point I've been advocating for over two years.
And I prefer to think about this site as about one that aims to serve speedy TASes, not lower numbers. It's just that in case with Super Metroid, I stopped caring about it since it's never going to be this way, anyway.
Heh, I've been doing that from the start. I re-evaluate a lot of entries on my rating list over time, each two months or so.
Still, the continual decline in both kinds of ratings of subsequent improvements amuses me a lot. And I'm not even talking about improvements that were questionable from stylistic or plain entertainment-wise point of view like SM64, which had different route changes and camera shots that were questionable at times, and weren't universally loved by the vocal majority. Ratings would make more sense if they weren't so heavily influenced by immediate enjoyment and novelty value.
Right, except it feels as if the current movie is both less awesome than the previous, and technicaly inferior in regards to the current standards (it's not).