|
|
Post by cloudmcshort on Apr 14, 2017 9:56:37 GMT
I think since the recent patch and physics change, fps boosting is close to being non-exsistent. How recent? I've done a bit of testing as well, but that was several weeks ago so if it was a very recent update then i'd have to re-do everything... |
|
|
|
Post by jeffreyb1990 on Apr 14, 2017 10:18:43 GMT
I think since the recent patch and physics change, fps boosting is close to being non-exsistent. How recent? I've done a bit of testing as well, but that was several weeks ago so if it was a very recent update then i'd have to re-do everything... It's the fix for the Infernus Classic. That changed the physics. |
|
|
|
Post by cloudmcshort on Apr 14, 2017 10:33:31 GMT
How recent? I've done a bit of testing as well, but that was several weeks ago so if it was a very recent update then i'd have to re-do everything... It's the fix for the Infernus Classic. That changed the physics. ok so i did the tests before that... ima do one more test on each setting and see if my tests are still valid. I'll be back. |
|
|
|
Post by cloudmcshort on Apr 14, 2017 14:57:59 GMT
Well, ok, i can confirm that my tests are still accurate and relevant. I re-did my tests on all settings and got similar results to before the update. Unlocked FPS is still faster. However, it only makes a difference on the higher spectrum (120+), and only when it fluctuates.
Solid high framerates make no significant difference. (See my spreadsheet for those tests)Moved to my next post. About your test adam10603 . I think the reason you didn't see a difference was because your FPS was too low and didn't fluctuate enough. I think the real boosts come when the FPS goes really high and it seems like ~100 isn't enough. That's why i decided to do the same test, but i had higher and more inconsistent framerates on the "high FPS" run. These were the results: Now, i'm not saying your test is invalid or bad or anything (or that my test is superior). The other way around actually. It's really helpful to see that you can run those framerates unlocked and not get a speed difference. However i feel like it's important to point out that if the FPS goes higher, differences will start to show. |
|
|
|
Post by cloudmcshort on Apr 14, 2017 18:28:02 GMT
Just a little bit of self-promotion here. This is my testing video showing all my fastest laps on all the different settings i used. Locked 60, Locked 120 and Unlocked FPS. EDIT: Just gonna quote my previous post... Well, ok, i can confirm that my tests are still accurate and relevant. I re-did my tests on all settings and got similar results to before the update. Unlocked FPS is still faster. However, it only makes a difference on the higher spectrum (120+), and only when it fluctuates. Solid high framerates make no significant difference. (See my spreadsheet for those tests) I hope this'll serve as good data to perhaps answer a few questions about FPS-boosting. |
|
|
|
Post by adam10603 on Apr 14, 2017 22:38:39 GMT
About your test adam10603 . I think the reason you didn't see a difference was because your FPS was too low and didn't fluctuate enough. I think the real boosts come when the FPS goes really high and it seems like ~100 isn't enough. Interesting. Yeah your lap seems to be a little faster when unlocked. I assumed that if there's no real difference from going form 60 to 100-ish, then there's probably isn't any difference above it either. That's interesting though. |
|
|
|
Post by adam10603 on Apr 14, 2017 22:52:42 GMT
cloudmcshort can you do the same thing on Spa, but with say 90 fps? I know that's a weird frame rate to lock to, but it could confirm if the differences only start to show at ~120
|
|
kasumi
Member
wat
Posts: 114 
Registered on: August 2015
|
Post by kasumi on Apr 14, 2017 23:03:20 GMT
imo differences should show up when u not lock your fps to your refresh rate and have unlocked fps. 60 120 144 seems to be ok. idk about 165hz monitors tho. needs to be tested. also I don't think for fluctuating fps between 60 and 144 give u any weird boosts. only unlocked fps that go sky high 150 160 ++
|
|
|
|
Post by cloudmcshort on Apr 14, 2017 23:37:23 GMT
cloudmcshort can you do the same thing on Spa, but with say 90 fps? I know that's a weird frame rate to lock to, but it could confirm if the differences only start to show at ~120 Depends on what you mean. Locked or Unlocked? I would say that going locked isn't gonna make any difference from 60 & 120, but going unlocked might be the interesting thing to look at here. I'll try that anyways at some point, see what happens. |
|
scenic
Member
Posts: 126
Registered on: February 2016
Social Club: Scenic_route
Discord: scenic_route#6404
|
Post by scenic on Apr 15, 2017 22:00:53 GMT
Just a little bit of self-promotion here. Kudos for tests well done and shared/documented! Your findings are in-line with previous findings, no meaningful difference 60 to 144fps interval, but cars pick up more boost with 150+ fps, and the Nero Custom on a glitch-friendly road is prone to very high top speed. Things are definitely being changed behind the scenes with R* updates, I've seen people recently get the Re7B to glitch on the Spa-straight, surpassing 811 by many seconds, but still far from the 7b mega-glitch last summer (Arifim did Spa in 2:19 I think). I've gotten the Turismo Classic to speed-boost on airport tracks (I can hold 155-180 fps alone there) with suspension glitch clearly audible, but not any more. All the magic appears to happens on high fps, 150+. The Comet Retro also seems semi-nerfed, a fairly recent development. There are also speed differences that happen without a clear pattern, some lobbies/hosts or group of players (p2p factor) just produce better lap-time than others. This might very well be an issue with the Roachstar time tracking itself, rather than the speed. Muscle cars seem to be quite unaffected by fps, at least the people who are experts in glitching the Comet Retro, 811 and whatnot, cannot do the same in the Pisswasser Dom. |
|
|
|
Post by CHILLI on Apr 17, 2017 19:37:25 GMT
cloudmcshort can you do the same thing on Spa, but with say 90 fps? I know that's a weird frame rate to lock to, but it could confirm if the differences only start to show at ~120 Depends on what you mean. Locked or Unlocked? I would say that going locked isn't gonna make any difference from 60 & 120, but going unlocked might be the interesting thing to look at here. I'll try that anyways at some point, see what happens. Maybe Adam can also chime in on this aswell, but I've been running my game unlocked with settings that end up averaging out performance around 85-95fps with the odd areas creeping their way into the 100's. When doing hotlapping on that Epsilon track, the qualifier for TEPCOTT, I dont recall ever going at higher speeds than Adam apart from when it's expected from a normal scenario. The best laptime I achieved around there was like :12.8 or a low :13, cant remember exactly, with minimal short-shifting (only in low speed corner exits). In some later sessions I set a few other laps somewhere around similar pace so it has remained consistent between computer restarts spanning weeks. I dunno though, Adam, do you recall my car ever doing something odd? EDIT: Offtopic but what the hell, is my signature absurd levels of huge for anyone else? That gif is only like 150x150 resolution, what's this witchcraft? Hopefully it's ok now. |
|
|
|
Post by cloudmcshort on Apr 17, 2017 19:45:03 GMT
Depends on what you mean. Locked or Unlocked? I would say that going locked isn't gonna make any difference from 60 & 120, but going unlocked might be the interesting thing to look at here. I'll try that anyways at some point, see what happens. Maybe Adam can also chime in on this aswell, but I've been running my game unlocked with settings that end up averaging out performance around 85-95fps with the odd areas creeping their way into the 100's. When doing hotlapping on that Epsilon track, the qualifier for TEPCOTT, I dont recall ever going at higher speeds than Adam apart from when it's expected from a normal scenario. The best laptime I achieved around there was like :12.8 or a low :13, cant remember exactly, with minimal short-shifting (only in low speed corner exits). In some later sessions I set a few other laps somewhere around similar pace so it has remained consistent between computer restarts spanning weeks. I dunno though, Adam, do you recall my car ever doing something odd? Oh, right, forgot to update on that.. I did some tests (just quick ones, didn't record or anything) and it seems like what you're saying is true. I upped my settings to get those fluctuations and i didn't notice any difference. I also tried locked 90, which gave me about the same speeds as 60 (if not slightly slower, can't confirm that though, not enough tests). |
|
|
|
Post by CHILLI on Apr 17, 2017 20:11:47 GMT
I also tried locked 90, which gave me about the same speeds as 60 (if not slightly slower, can't confirm that though, not enough tests). I've gotten a similar impression at times, but then in other moments it almost seems like I'm a little faster. So far I've assumed it to be natural differences going from 60fps to something else, causing the car to react to certain kinds of bumps a little differently but remains pretty close on average. Hm... So going by the most recent posts in here it sounds to me like R* has improved stability in the <120fps region compared to day 1. As my game barely peaks above 100-110fps I rarely go above that framerate threshold of ~120 I cant really comment on what my experiences with that have been. Pretty much everyone I've raced with so far have been on very close pace to myself but I also have no idea what FPS those guys have been running, so it seems to me like sub-100fps is a safe framerate to use without causing the game to freak out. That said I dont use every little trick in the book to go as fast as possible so there's still the off chance that I've overlooked something. |
|
|
|
Post by adam10603 on Apr 17, 2017 20:55:35 GMT
I dunno though, Adam, do you recall my car ever doing something odd? Not at all. In fact I had no idea you were running above 60 until you told me, which I think was after that. |
|
|
|
Post by SteveJEWBZ on Apr 19, 2017 14:42:11 GMT
So I'm now on my new PC (finally kasumi lol) I have awful screen tearing, wondering if anyone else is using an AMD card and how they lock to 60. Rivatuner gives me awful awful tearing and the "wait for vertical refresh" setting does nothing when I turn it on. The ingame Vsync is amazing, very fluid but I would like to test for myself if it does slow you down and idk if I can stand the screen tearing while I'm doing testing lol |
|
