Can someone make a HumanBenchmark test

Talk to software developers and aspiring geeks. Programming tips. Improve motion fluidity. Reduce input lag. Come Present() yourself!
Post Reply
User avatar
lexlazootin
Posts: 1251
Joined: 16 Dec 2014, 02:57

Can someone make a HumanBenchmark test

Post by lexlazootin » 07 Mar 2015, 02:06

Can someone make a fullscreen non-vsync app that I can run to test my reaction time, possibly configurable colors and settings.

Or is there a reliable one out there already made? I don't really like the one on the website because i'm using windows 8.1 with aero and also because it's in a web browser.

flood
Posts: 929
Joined: 21 Dec 2013, 01:25

Re: Can someone make a HumanBenchmark test

Post by flood » 07 Mar 2015, 19:39

you can throw together one with love2d pretty easily :P

User avatar
lexlazootin
Posts: 1251
Joined: 16 Dec 2014, 02:57

Re: Can someone make a HumanBenchmark test

Post by lexlazootin » 07 Mar 2015, 22:41

Never really coded before and this looks easy enough... just going to copy and paste from everyone else projects till it works, normally what I do when I need to code.

User avatar
lexlazootin
Posts: 1251
Joined: 16 Dec 2014, 02:57

Re: Can someone make a HumanBenchmark test

Post by lexlazootin » 08 Mar 2015, 00:03

I think I did it. I get much better times then I ever did on humanbenchmark but it might be because my code doesn't work.

Edit: http://pastebin.com/m7uKfGWh fix2
Last edited by lexlazootin on 09 Mar 2015, 02:42, edited 1 time in total.

flood
Posts: 929
Joined: 21 Dec 2013, 01:25

Re: Can someone make a HumanBenchmark test

Post by flood » 09 Mar 2015, 02:24

math.random only needs a min and a max...
i think on the website the time when the color changes is a random num between 2 and 5 or somethign like that

also you should use fraps or something to make sure it's not running with vsync

User avatar
lexlazootin
Posts: 1251
Joined: 16 Dec 2014, 02:57

Re: Can someone make a HumanBenchmark test

Post by lexlazootin » 09 Mar 2015, 02:58

flood wrote:math.random only needs a min and a max...
i think on the website the time when the color changes is a random num between 2 and 5 or somethign like that

also you should use fraps or something to make sure it's not running with vsync

Whyyyy, I had vsync off and then when I turned it back on for testing it now no longer turns back off... why would you do this too me..

flood
Posts: 929
Joined: 21 Dec 2013, 01:25

Re: Can someone make a HumanBenchmark test

Post by flood » 09 Mar 2015, 03:03

lol
https://love2d.org/wiki/love.graphics.setMode

set it off or force in nvidia

User avatar
lexlazootin
Posts: 1251
Joined: 16 Dec 2014, 02:57

Re: Can someone make a HumanBenchmark test

Post by lexlazootin » 09 Mar 2015, 03:55

I have AMD and for some reason when forcing it off in the driver I have to run the love.exe then run the batch file or the change wont be made :|

Also using the ingame FPS counter seams to work fine for telling if vsync is off for every combination I tried using fraps. I don't know if you had problems or something.

And I just assumed that math.random() would pick one only one of the numbers I gave it and not a random decimal in between.

Thanks Btw, i'm very happy with the end result.

greenenemy
Posts: 35
Joined: 11 Mar 2015, 04:45

Re: Can someone make a HumanBenchmark test

Post by greenenemy » 11 Mar 2015, 05:07

Nice program, thanks.

I'm getting around 50ms better results with this than with humanbenchmark.com, both @120hz.

For v sync off change line #2 from

Code: Select all

love.window.setMode(0, 0, {fullscreen = true}) 
to

Code: Select all

love.window.setMode(0, 0, {vsync=false, fullscreen = true})
Also i would change state 1 to black and state 2 to green because right now it's confusing.

Post Reply