@moderators Smb1 All Stars speed ain't right
5 years ago

This ain't right

So, it took me months to finally figure out how in the world Kosmic was being "framerules" faster than me when we seemed to execute at the same level. The key was, it wasn't actually framerules, it was frames. I don't know if it's framerate, or lag, but Kosmic seems to consistently pull ahead as the run goes on

(make sure to ignore my voice for optimal focus) This is the comparison between some random run that I did in which I executed essentially the same speed as Kosmic yet fell behind by frames.

(yet again ignore my voice in this one for optimal focus) (sorry, I liked it when Kosmic said "blazit clip below the bricks") Anyway, in this run I do 1-2g which saves 5 framerules over WR, but I mess up in 8-1 costing me 3 framerules, and everything else I play just as fast as WR Yet, as I should be 2 framerules ahead as we enter 8-3, we're essentially tied instead.

I also compared between mine and RoyLGamer's run as well, and the weird part is that no one is at any framerate or lag disadvantage, essentially that we are playing at equal framerates. Our mistakes add up in the end, no one executes faster than the other, and we finish at basically the same time.

That is quite weird, because if you're playing on console then wouldn't all consoles kinda function the same? Meaning that Kosmic and RoyLGamer should both be going at the same framerate. I'm not suspicious of Kosmic's run, or Kosmic, but I just want to know wut the hecc is going on.

(below the briccs, abbaabbaba)

Like likes this
Antarctica

imdangWut

Edited by the author 5 years ago
Like likes this
Illinois, USA

But...it looks like youre on emu, and i swear i saw lag on your run in 1-2

Like likes this
Germany

I haven't compared video timestamps but I'm just gonna say that the framerule does not apply to

  • the titlescreen to 1-1 transition
  • warp zones
  • the 8-4 to 9-1 transition (lost levels) in All-Stars.

So 1-1 should differ by n frames. 1-2 can differ by anything. 4-1 should then differ by mx21+n frames. 4-2 can differ by anything. 8-1 to 8-3 should differ by px21+n frames.

EDIT: I didn't realize it's emu. snes9x runs at a lower framerate. If you use snes9x, I'd switch to Bizhawk.

Edited by the author 5 years ago
Like, darG and 4 others like this
London, England

Try to find a copy of SMAS on SNES as it has less lag than emulator

Like likes this
New Hampshire, USA

Emu is the slowest, so if you get a snes and a copy of the game (^) you should be equal with kosmic.

Like and kcaj like this

I see But what about when I compared my run to RoyLGamer's run? He's on console, yet we both seem to be at the same framerate, if you compare our runs we execute about the same and we finish at about the same time. Are we both at a framerate disadvantage?

Also, about Bizhawk, how can I reduce input lag so my muscle memory won't be so screwed up?

Like and darG like this
New Hampshire, USA

You probably are at a framerate disadvantage. idk

Like and kcaj like this
Madison, WI, USA

I'd do Bizhawk and SNES9X does have a framerate disadvantage. I'm not sure about ZSNES so I'd stay away.

Like likes this
Germany

"Also, about Bizhawk, how can I reduce input lag so my muscle memory won't be so screwed up?" I think Bizhawk has less input delay than snes9x? At least it felt that way to me.

Definitely go to Config -> Rewind & States... and disable Rewind, that's a massive drain and on by default.

Config -> Cores -> SNES - There are 2 options you can test out.

Config -> Display -> Display Method There are 3 options, experiment with these.

I think that's all.

Like and darG like this

WOW the changes really work very well, thanks for the help!

However, when I compared my run and RoyLGamer's run, we seemed to be at the same framerate as we lose equal amounts of time yet finish at the same time, but he is on a console. How does that happen?

"So 1-1 should differ by n frames." When you say this, do you mean it is possible to save FRAMES in 1-1 depending on how you play the title screen?

Like likes this
United States

"When you say this, do you mean it is possible to save FRAMES in 1-1 depending on how you play the title screen?"

I'm assuming you don't already know this since you're asking, but depending on when you press start on the title screen you can actually save time.

Like likes this

i c So, will I want to press start earlier or later to save time?

Also, in Bizhawk the gameplay every once in a while seems to be laggy, jerky kind of thing that makes it difficult to play the game, is there any way I could fix this?

Like likes this
Portland, OR, USA

would the cutscenes between levels have anything to do with this? idk just throwing it out. you can get a first frame cutscene every time fairly easily but after that time is lost by frame

Like likes this

I highly doubt that, because I'd have to mess up the cutscenes on every single run that I did. I copied the WR splits on my livesplit and I always consistently lose a tenth of a second each level I beat over the WR when I play on Snes9x

Confirmed that Bizhawk does work good, I don't consistently lose time over WR as the run goes on

Like likes this
Germany

Setting high priority to the emuhawk.exe in Task Manager can help with stuttering.

Like likes this
United States

I use snesgt and it works amazingly fine.

Like likes this
Game stats
Followers
7,816
Runs
8,836
Players
1,792
Latest news
Requirements for High-Level Any% Runs

Any% (NTSC) runs below 4:57.000 must now fulfill additional requirements in order to be verified.

  • The run's full session must be included in the submission description.
  • For emulator runs below 4:57.000, some form of input display must be visible for the duration of the run. A hand-cam or input
4 months ago