Yeah, I know 9.52 preview works, I just don't feel like downgrading.
Anyway, what particular advantages will this have over YouTube/Google Videos? Last I checked Theora didn't hold up to H.264 well, if at all.
I measured from entering the room until the appearance of refill items. Could you re-check?
[EDIT]
Checked again with VirtualDub, with starting point = first frame of motion in the room, and ending point = first frame of refills appearing. In Saturn's 14% run the time between these two points is 6036 AVI frames, in your AVI, it's 4214 frames at half the framerate. When adjusted to the same framerate, it's 2392 frames, or ~40 seconds, longer.
[EDIT 2]
I see the reason for disrepancy: you've measured the difference in in-game frames, and your movie has more lag due to using one more powerbomb.
So, according to my crude calculations, this fight is roughly 41 seconds slower than the Ice route counterpart.
Unfortunately, you won't be able to make MB fight this exciting (if only because running around during it is counterproductive, not to mention the complete lack of ammo).
Then don't use it.
There was no sarcasm whatsoever in my previous post. The responsibility for using an unstable emulator is yours, and yours alone.
(Don't worry, you look offensive either way.)
Maybe it's worth waiting for stabilization of the emulation core then. You knew full well it was far from final when adelikat released the first rerecording version. No point in complaining that it ruins something now.
For the record, this was the precise reason I wasn't too big on the idea of porting rerecording features to immature emulators.
Right. But if the goal is to create a small encode, it is an acceptable condition (after all, watching four extra seconds for once isn't going to kill you).
You don't need deldup, as I believe it will lead to a/v desynchronization. And yes, if there are more keyframes, more bits get spent on them and less on the other frames. Actually, a good keyframe interval is the one that lets you seek in the intervals of 5 to 10 seconds (multiply this by frame rate to get the keyint value).
You can use AVI, but it doesn't handle some options (like b-frames) well. As for WMP problems, I believe it largely depends on the decoder you're using for H.264 video. Last I checked, WMP used DirectShow filters for decoding video streams, so you'd have to install something — be it a filter like ffdshow, a codec pack, or a standalone player — either way.
Much better. The difference is more pronounced if the original recording is clean (like emulator output) since H.264 excels at preserving shapes.
Large part of the problem is lesser/harder customizability.
To be honest, at the moment H.264 is about the closest thing I've seen to such kind of balance. Compatibility is drastically improving each year, and ease of use problems are largely exaggerated.
I guess it won't be much of a surprise as to what made you two minutes slower than this four year old publication, which aimed for the same goals as your submission.
Not using tools given to you is not a means of displaying your skill, but a good way for a submission to end up in gruefood. Read our rules and guidelines, otherwise you will be bound to repeat your mistake again.
Somehow I got unsubscribed from this topic, and it's the gazillionth time it has happened to me on this forum, RAWHGRHG.
Well, this might indeed be a borderline case… But I would still suggest "Castlevania – Symphony of the Night" for consistency. And definitely not "Castlevania – Nocturne in the Moonlight", because it's quite literally a mishmash of either name.
This arcade game is widely known among shoot-em-up enthusiasts, but is rarely mentioned outside. Its PS2 port has four game modes, the last one of which is near-impossible. Here are two of the hardest attacks in the game played with invincibility (via Pro Action Replay) just so that you could admire the bullet count.
http://dic.nicovideo.jp/v/sm7682508http://dic.nicovideo.jp/v/sm7682351
I don't expect anyone to TAS this game, so this is probably the only occasion you will see of a player not bombing or dying every 10 seconds of these attacks.
For the record, this mode has been completed in unassisted conditions, there are superplay DVDs and YouTube videos out there.
It will probably make sense to make a feature-freeze branch at one point (say, after that cycle-PPU thingie gets finished), and only update it once per year or so. On one hand, it will let you update the main branch as often as you see fit, on the other, it will reduce the version conflicts to acceptable minimum.
Keep in mind that tricks don't automagically appear after some time, they're always there waiting to be discovered. And it won't be a mistake to say that it's TASers' duty to discover these tricks, and they have a lot of leeway compared to the unassisted runners in doing so. If a TAS failed to explore an opportunity at saving a large chunk of time, well, then it's exactly that: it failed. Much like the first 3-4 Super Metroid TASes failed because they were since beaten by unassisted runs and realtime savestate-assisted runs which didn't include any arcane knowledge that couldn't be obtained at the time.
29.91 fps? >_>
Anyway, the movie was pretty enjoyable. I was a bit surprised you decided to go with F for the last boss, was it not available earlier? Or is there something I'm missing? Shame this Contra doesn't offer much in the way of optimization (or even dodging, apart from the fish boss), but at least the presentation is nice, and you've managed to take advantage of some creative shortcuts. I'm voting yes.
I'm not in favor of unpublishing runs due to entertainment values and such. However, I do agree that runs that are violating the rules, particularly "submission must complete a game" and "submission must beat all comparable records", might need either rejudging or subjection to quarantine, even if the violation technically happened post factum.
The entertainment ratings are heavily influenced by popularity of games. Two different games may both be boring as fuck in their own right, but if one of them was played by a lot of people who remember it fondly (most often only because they do remember them), it will a priori get higher ratings. It's been seen, and it's been expected. So, to answer the "should we unpublish the low-rated movies" question, I guess we should first ask, "do we want only popular content here". I know I don't.
The other reason I suggest using kanji instead of romaji is that we — Westerners — don't really need either, but Asian audience prefers that over both romaji and English translations of the titles.
The original name will be used as in case with Japanese titles, the rest will depend on the alphabet used for it (i.e., there's no need to transcribe Latin letters, but there may be a need to transcribe, say, Cyrillic).
Stop and consider for a minute how you are going to determine which one is more "widely" used. There's no definitive answer. Arguably, we could always give priority to European titles simply because there are more countries and more general population in PAL regions compared to NTSC-U. Using release dates or de-facto prioritizing USA titles are the only solutions that seem feasible to me.
This is more difficult than it seems because there are multiple ways to romanize kanji/kana (think l/r, o/oh/ou, silent "u", etc.), and little to ne means to determine which is correct/closer to intended pronunciation.
There have been both a discussion and attempts to do an updated non-glitched S3&K run, but nothing came out of it. It has the potential to become a top-10 movie easily, though, so it's even surprising there's no competition anymore. I guess players are somewhat afraid of not holding up to Nitsuja's and Upthorn's standards?..