Joined December 2010
- Tweets 3,282
- Following 107
- Followers 3,259
- Likes 3,847
1,545 Photos and videos
Pinned Tweet
Apr 12
Trying to bring a lot of results together.
x.com/noodalls/status/204309…
x.com/noodalls/status/204318…
docs.google.com/spreadsheets…
Apr 12
Reference list of input lag tests (command history)
1
3
2,229
Input lag tests for PS5 2xKO. (repost due to wrong settings last time.
Remember this is command history, animation for 2xKO is 1F earlier.
Only have 4 stages unlocked so can't test the others. @Play2XKO is there something you could do for me to unlock the other stages to test?
4
7
33
9,806
Jun 13
amazon.com.au/dp/B07FMQZVW2?…
amazon.com.au/dp/B07FMQZVW2?…
Would be interesting if someone brought one of these to tournaments to measure temperature at USB port.
1
2
458
For what its worth I've tested and this doesn't add any input lag.
1
175
Jun 17
Got to test the various different stages in SF6 across my PC and PS5.
Full results are here.
docs.google.com/spreadsheets…
Simple results in the graphic.
11100 tests. All saved.
5
26
88
9,619
Jun 17
Sony makes a controller that can be soft to the touch and get hard just before Doa6LR is released.
Hmmmm
Sony has patented a controller with buttons that can change how hard or soft they are based on what happens in a game
They would also trap your fingers if your character was grabbed on-screen
1
1
8
992
Jun 15
That little guy? I wouldn't worry about that little guy.
(Command history harder than usual to test)
2
429
Jun 13
Lazy 10000 tests.
(All results are raw results for top command history in sf6 ps5 version. Input lag reduction on. 120hz vrr auto apply)
4
9
28
3,530
Jun 13
To explain the coloured graphs, these are raw results. Technically they are pixel accurate, so each x coordinate is a test and each y coordinate is the result for that test. So test 1 (blue) is 60ms, and the pixel is at 1,90 (with some smearing for some reason).
2
2
505
Jun 14
Worked out the cause of the smearing. Results are non-integer (0.1ms accuracy) but when you draw pixels using these values they will account for this. Have changed values used to integers when drawing results to prevent the smearing happening.
2
309
Jun 13
So, given we are really interested in when the brightness deviates from the baseline, I just record each time a brightness is recorded for the first time. This way, given brightness is 0-255, there are only 255 values needed to be stored for each test (2x1000x255=510kb).
1
1
257
Jun 13
I've added a bar to make it clearer when the actual tests starts and the zeroing ends.
214
Jun 13
In summary fastest to slowest.
Brook UFB fusion 55.56ms
Gen5X 56.65ms
Dualsense wired (A front/back,C,A-hub,C-hub,extension cable) all around 59.1ms
Dualsense wireless 60ms
1000 tests each.
When testing the hub I added a Brook UFB inputting commands at the same time.
1
5
308