An expertly choreographed work of art, showcasing all of the tricks that make Doom TASes a must-watch. From monster manipulation, projectile manipulation, and abuse of the "suicide exit" (http://doomwiki.org/wiki/Suicide_exit)... this run truly exemplifies what a TAS is really about: Pure. Unadulterated. Raw speed.
Solid 'Yes' vote.
I was going off this Super Demo World any% run that was just published where (I can only guess that) the input ends while Mario is off-screen above Bowser, which causes Mario to fall on top of Bowser and take damage but hit him with the final Mechakoopa, beating the game. I'm guessing that the game "really ends" after 16:52.9 and that the taking damage and hitting Bowser wasn't included in the time or else that wouldn't have been there. I would expect this level of optimization out of all TASes... if not just for the sake of getting a lower time, even if it's just a handful of frames.
DOOM is different though in that stopping the demo/input would cause the game to end prematurely. I do acknowledge this now that it I was reminded of it, so my initial comments and observations were wrong. So now let's just have null input until the game ends. But now the time should stop when input does, which would be when the game takes over. I wouldn't agree with discarding the null input and calling time early while still playing the game, though.
I'm expecting to get my ass thoroughly handed to me on a silver platter about how I don't make the rules around here, I don't know anything about how timing runs work, and that my opinions don't count for jack but that's where I'm coming from...
Maybe it's different in the port I use then...
The inputs afterwards count against the TAS time, though. From what I understand, the TAS is timed until the last input, not the end of the game.
Are the intermission screens supposed to be that slow? I can't tell what source port you're using, but I would expect the intermission screen to immediately show the stats and move on just as fast.
EDIT: Oh, it would help if I read the submission text. Never used GLBoom, but the question still stands. Should source ports even be used instead of "Vanilla Doom"?
I also agree with firing the last rocket at the Cyberdemon at close quarters since you win no matter if you are alive or not. Furthermore, the frame where you fire that rocket should be the last input. You do a short sprint after you fire the last rocket that should be cut out.
Drives the wrong way down the freeway, weaves through cars with laser precision, crazy drifts, gets to the destination late...
...gives the driver $10,000.
Yes.
Finally! This has been one of my most personally anticipated TASs just because of how much I loved playing it when it first came out.
From what I can tell, this looks pretty good. I'm assuming that time spent watching the points total up after collecting a frog is so the moving objects on the map are in the right place when you go back to the start point.
Yes.
If I understand it all correctly, the original "pony" version actually wrote a completely different "game" and loaded it using Yellow whereas this doesn't completely overwrite the game, it just arranges the memory in a way that creates dancing pi symbols and outputs pi to the screen in lockstep with the BGM...
I'm not taking anything away from this run but completely overpowering the game with a new "game" has to account for something at least, IMO.
"Fastest stage 3" is like someone doing a fastest "World 8" run in SMB. It's an arbitrary goal and there's no real reason to stop on Stage 3 if there are more stages afterwards. It's like you gave up on going any further because it got harder.
No.