Is this recording using autofire or slowdown?

greenspun.com : LUSENET : MARP Editors : One Thread

Here's GK's recording, i think this is clearly slowdown but it does not use chronic fire so it's not detected by analinp, but i am hesitant (even though it's from the famed german krol) to declare it slowdown because there are no concrete rules that we can use to determine if 39% of 1 frames means autofire.... do we have some test inps to statistically and reliably say a certain is slowdown and this is probably not? If not how vote ye to this being dqed? gc_cosmogng_7623530_m36b10.zip -------- analinp Report --------
p = 3 f = 4
Bit# frames:on/off Presses BPresses Bursts highVar LowVar Speed%
250 21666/55490 9695 4134 712 0.31 0.396 49.6%
p = 10 f = 9
Bit# frames:on/off Presses BPresses Bursts highVar LowVar Speed%
250 21666/55490 9695 7693 200 1.09 1.271 68.2%

..........
Bit 250, length 1: 7507 ( 39.87%)
Bit 250, length 2: 4445 ( 23.61%)
Bit 250, length 3: 3461 ( 18.38%)
Bit 250, length 4: 1949 ( 10.35%)
Bit 250, length 5: 676 ( 3.59%)
Bit 250, length 6: 335 ( 1.78%)
Bit 250, length 7: 187 ( 0.99%)
Bit 250, length 8: 149 ( 0.79%)
Bit 250, length 9: 65 ( 0.35%)
Bit 250, length 10: 53 ( 0.28%)
Bit 250, length >10: 563
Bit 252, length 2: 1 ( 50.00%)
Bit 252, length 3: 1 ( 50.00%)
Bit 252, length >10: 2
Bit 254, length 1: 1 ( 16.67%)
Bit 254, length 2: 1 ( 16.67%)
Bit 254, length 4: 2 ( 33.33%)
Bit 254, length 6: 1 ( 16.67%)
Bit 254, length 8: 1 ( 16.67%)
Bit 254, length >10: 16
-------- End Of InpHisto Report --------


-- Anonymous, July 10, 2000

Answers

I'm pretty sure that there's now ay in hell that Krol's computer could have played this game at 90% of the game speed or above, let alone slowdown.

Fry it.

Q.T.Quazar

-- Anonymous, July 10, 2000


Solely going by these numbers, I'd immediately say "YES!". Add to that the identity of the person who submitted it, and it becomes even more certain that slowdown or autofire was used. I'd say the former.

But let's take a step back. Clearly, this game (although I don't know it) is in the 56-64 fps range, or else analinp wouldn't have done an autofire analysis. And the kinds of statistics as shown by InpHisto are generally not human if done at, say, 60 fps. I have analysed .inps of known fast pressers like Bubble, BBH, Steve Krogman, and myself, and all of our speeds are very similar and nowhere near the statistics shown in this recording.

However, I always perform one sanity check before relying on the data shown by InpHisto. And that is: I play the game myself and press the fire button like crazy. Then I compare my statistics to the one in the suspicious recording and if mine are a lot (s)lower, then I have no doubt that slowdown/autofire is used.

For the sake of references: I usually get an average frame length of 4 on games where I have to press fast, and well over 60% of all my presses will then be in the 3-5 frame length range. I could get a LITTLE bit faster, if I just pressed like crazy and did not really try to get far in the game, but anything well below a 3 frame average on a 60 fps game always makes me very suspicious.

Oh, and that does not only go for button presses, it also goes for the directions on a joystick.

Cheers, Ben Jos.

-- Anonymous, July 10, 2000


By the brief evidence by looking at the game - I see this is a 60 FPS Namco System 2 game. It is also a game where autofire can be detected - as it's a shooter(a nice looking one as a matter of fact)

My rule is if lengths 1 & 2 in INPHistogram combine for a total of 50 % or greater, and length 1 is 10 % or higher - then it's autofire / slowdown. Sometimes you can catch autofire w/o the autofire detecters by looking at the game.

My final opinion is it's slowdown(very possible with this type of game) or autofire - it needs to be DQed. I'll leave it to the initial accuser to do so :) GB9

-- Anonymous, July 11, 2000


Moderation questions? read the FAQ