nullpomino has very good TAS-detection. So while it will let you cheat, it will show TAS-PLAY on the screen if you do.
There's no such thing as a perfect TAS detector. If you cheat modestly there's no way to tell no matter how you code it.
The TAS detection was pulled out in 6.3 alpha. Even then, it's not *that* hard to get around the TAS detection in the last NullpoUE for 6.2... And it's far from perfect. It only sets off in two cases (detecting >5% slowdown and use of rerecord). Only I know how to turn off the TAS PLAY thing, though (as far as I know) I suppose with a bit of digging it wouldn't be all that difficult to find out how to do it though. This is sort of the problem with replays one can freely edit, but whatevs...
You don't even have to cheat modestly - even in the latest Texmaster / Nullpo it's very easy to slow the game down to as slow as 50% and still fool the TAS detection. Honestly, when the TAS detection is so easy to negate, it makes it completely redundant, epecially considering that experienced players can detect TASes by eye way more accurately than any machine ever could. This is a rather obvious example: http://www.youtube.com/watch?v=jQcEdJbDdcY
The point wasn't about TAS'es though, it's about getting slowdowns and not even noticing yourself. When I started playing Heboris, my average FPS would be in the range of 50-55fps. Because I just started out and didn't know what the game was supposed to feel like delay-wise, I didn't notice until I turned on FPS display. Playing with accidental slowdown can make you last 100 levels longer than you would normally.
As I said before, the latest version of NullpoMino seems to have gutted the TAS detection I wrote in... some of it is still there, but a lot of it isn't...
That is from before the TAS detector was put in, so that doesn't really count. I've also found that responsiveness is not very good when i fall below 60 fps.
Zaphod, the point I was making is that TAS detection is not needed, the fact that that video was made before Report put it in is irrelevant; any experienced player can tell that it's a TAS within the first few seconds. Okay, so that video is one of the more blatant examples, but even when they're subtle, the human eye is still the most reliable method of TAS detection. Anyway, back on topic, Xkeeper was asking about the best TGM simulator. Without a doubt, Eirstt is the most accurate in terms of timing, score, grading, and randomizer (pretty much every aspect). Heboris C7 is all over the place in terms of accuracy. NullpoMino and Texmaster are somewhere in between; Texmaster has very good timing (apart from the DAS charging aspect) but the score, grading and randomizer are significantly off; Nullpomino has less good timing, but the score, grading and randomizer are more accurate. If you're interested in SRS, currently the only games that implement it 100% accurately are Eirstt and NullpoMino.
Nullpomino is practically the spiritual successor to Heboris, only more accurate and much more mod-friendly. Only problem is that since it runs in Java, it eats up memory and has input lag every now and then.
