Quick noob question: what's the rationale for going for in-game time rather than real time?
Like with most games where in-game time is given priority (c.f. Sonic, Super Metroid), the reason is that there's a community of non-TAS racers for this game, and it's good to be able to compare the TAS results to what those racers get.
Another reason is if in-game time leads to more interesting, varied strategies that would be slower in real-time.
Tis calculated at the NES' frame rate of ~60.098814 fps (or more likely a less precise number).
Ye of little faith.
For NES, the site uses 60.098813897440515532 for NTSC and 50.006977968268290849 for PAL
How did you calculate it that exact? Did you look at it and say it was about 60.098813897440515532 fps?
In all seriousness how did they calculate that?
The NES CPU has an exact Hz and there's an exact number of cycles between each frame on the NES CPU. (Some more knowledgeable in the field than me could provide the numbers, I'd check nesdev for example)
The reason why the same TAS can be different speeds when converted to different emulators is because they emulate lag and loading times differently - not because of (or not necessarily because of) differing FPSes.
Also, the reason why SM64 TASes sync so well on console is because the N64 only polls during input frames, so you can program your bot to only send input when it polls - then it doesn't matter how lag emulation differs from the real thing!
A similar question is 'Is pi a Normal number ?'
Here is an interesting quote:
"Determining if numbers are normal is an unresolved problem. It is not even known if fundamental mathematical constants such as pi (Wagon 1985, Bailey and Crandall 2003), the natural logarithm of 2 ln2 (Bailey and Crandall 2003), Apéry's constant zeta(3) (Bailey and Crandall 2003), Pythagoras's constant sqrt(2) (Bailey and Crandall 2003), and e are normal, although the first 30 million digits of pi are very uniformly distributed (Bailey 1988)."
Though one could imagine a non-normal number that does have every possible series of digits within it. For example, concatenating the digits of every number one after the other, and doing it twice if it contains only 1s, is a non-normal number containing every possible series of digits.
However, I would expect every 'non-constructed, transcendental' normal number to also contain all subsequences in it.
Is there a hotkey that toggles frame advance on/off? If not, why not set the non-frame advance speed to something slow (50%, 25%, whatever) and also bind to your controller this hotkey?
I was aware that bizhawk and its ilk are very demanding, it just seems strange that it can bog down my computer where newly-released games on high-quality settings do fine. I know this is a really dumb statement so forgive me, but how hard can it really be to emulate an SNES?
Harder than you think to emulate with 100% accuracy.
The main reason is that the drawing to the screen is not done after all the CPU cycles for that frame are over - it is done in parallel. A truly accurate emulator must process the screen being drawn in progressive scanlines simultaneous with the CPU cycles advancing.
http://arstechnica.com/gaming/2011/08/accuracy-takes-power-one-mans-3ghz-quest-to-build-a-perfect-snes-emulator/ is a really great three page explanation of what bsnes does other emulators don't and why.
Tasvideos would rather ban snes9x submissions entirely, as it is lower accuracy than bsnes core (Bizhawk, higan, bsnes/lsnes, etc). However, the requirements to run this core at 60FPS are very high and not met by all computers currently in use. On top of that, the developer of bsnes does not have much interest in improving the performance (and especially it will never be made more performant at the cost of accuracy). As such, I think it will still be possible to make snes9x 1.51-1.53 submissions until a few years from now.
Here are some reasons why zeromus may not like the voxel drawing mode to be in the main app and would prefer it to be a separately maintained branch and/or plugin:
-If it is part of the main app that everyone gets, there is an expectation that it not be broken, that it remain polished and up to date and 'not crufty', etc. If it is made entirely by and to the ideals of a separate developer who may not continue working on the project, it may just age and deteriorate.
-It is out of scope/not part of the intended functionality the developers want the program to serve.
-If it has high coupling (is reliant on many implementation details and parts of the source being just-so) requiring it to be supported could slow down refactors and rearchitectures of the code.
This happens in a lot of places, for example many plugins for Notepad++ that are super useful like TextFX, even when they get bundled with Notepad++, they still aren't a core part of Notepad++ because of reasons such as the above.
Isn't sonic 3's engine slightly different? Maybe that could be abused for some faster speedruns at parts, if one were to do a Sonic 3 TAS? (I know this is Sonic and Knuckles, just stating) After all, do we know if Knuckles' boss was added for Marble Garden?
Sonic 3 & Knuckles changed some Sonic 3 levels and physics (Some glitches only work on Sonic 3 alone, not in S3&K), but I think the reverse is not true (an S&K level or physics being changed by playing it in S3&K instead)
https://www.mediafire.com/view/zbrgv7mr87oi9nq/Bunny_Must_Die_WW.xlsx
The glitch that was found was where you select "return to title" a frame before a text box appears. The text boxes can be anything from boss encounters, item acquisitions, signs you read or even the text that appears when you first get a weapon (very relevant). If you do that, you don't end up returning to the title, but you change the value of the starting position of that room to something else. If you then die and use a bunny doll you will wrong warp to another location in the game. Each different text box wrong warps you to a different location. Some warps will put you into a wall and subsequently give you a game over, some will put you in a good position, and some will put you in a good position depending on the tile you initiated the glitch (being an extra tile up, to the left etc. will let you not die in the room you warp to). The excel sheet posted above shows all of the current known wrong warps. It's fairly exhaustive from what I can tell.
The most useful wrong warp is funnily enough the sword acquisition text box. The most relevant position to do it in is the first sword you get in a glitchless run, but you have to get it before the sword falls or you'll get stuck after the warp. This warp puts you into the right side of the room of the back door to the final area. As long as you have the cutscenes triggered (first the purple area then the first boss area) you can fight Chelsea (who is the final boss for those of you who haven't played the game).
