Page 1 of 1

Testing Latency of Controllers

Posted: Thu Apr 28, 2022 9:04 pm
by lifewithmatthew
Hello all! I'm new to the MiSTer community (still waiting on my first one in fact) and I was learning about controller latency. I've done a bit of an overhaul of the code used for the Demonbite dual sega controllers and wanted to test the latency of the new code to see if I made it any better, if it's the same, or if I made it much MUCH worse :lol: How is testing done to find out controller latency?

Re: Testing Latency of Controllers

Posted: Thu Apr 28, 2022 9:34 pm
by Xbytez
This is the latency testing procedure used in producing the MiSTer Input Latency spreadsheet:

https://docs.google.com/spreadsheets/d/ ... =369482991

Re: Testing Latency of Controllers

Posted: Thu Apr 28, 2022 10:06 pm
by lifewithmatthew
Thanks!

Re: Testing Latency of Controllers

Posted: Fri Apr 29, 2022 7:15 am
by clarkkent
In my opinion the input latency only matters if you have a very low output latency… 🤪

I‘m extremely happy with the 8bit Arcade Stick because of the special capabilities but the speadsheet says it has high latency - I do not notice it. Maybe because I’m not the youngest one or my reaction is not the highest…😎

Re: Testing Latency of Controllers

Posted: Sat Apr 30, 2022 1:43 am
by lifewithmatthew
To be honest this is more of a "climbing the mountain because it's there" thing for me. I followed the instructions (with a couple tweaks) to make a dual sega to usb adapter from demonbite (I even found a nice 3D printer STL file to use for the case), but looking at the code I saw a couple places I thought I could tweak to get some better performance. It's works well using an emulator on my PC, but I don't have a mister to test and see if I actually improved anything.

Even once I do get my MiSTer, I'm not sure if I could get it set up to test latency 🤷‍♂️ I'm more than willing to share the code with anyone who has a setup for testing that would be interested in taking a look.

Re: Testing Latency of Controllers

Posted: Sat Apr 30, 2022 2:11 am
by FoxbatStargazer
Stuff is more than playable with the 1-2 frames of lag you're going to get with a decent TV and wireless controller... but I can still feel the difference between that and a CRT and 1ms wired. I don't know that it even makes me any better at games, but in some games it just feels more satisfying to get that instant response when you press a button, a feeling that's very hard to get from any software emulator. So it's nice to have this kind of resource if you want to chase down that feeling without getting into SNAC.

Re: Testing Latency of Controllers

Posted: Sat Apr 30, 2022 8:32 pm
by PikWik
I can say, micguyver has optimized his daemonbite code to the best of his abilities, and a couple months ago, optimized his SNES adapter even further to get an average reading of 0.9ms, versus the previous reported 1.1ms average

but, that doesnt mean the code couldnt get better, and it would be cool if someone could optimize it even further (another great thing about an open source project)

Re: Testing Latency of Controllers

Posted: Sat Apr 30, 2022 8:40 pm
by PikWik
i have played with SNAC on a modern HDTV, a HD computer monitor, and also have a SNES & CRT.
i play alot of kaizo super mario hacks, and the input response is "identical" to my CRT setup using the MiSTer with a wired daemonbite, vsync2, and a fast computer monitor @ 1080p

Re: Testing Latency of Controllers

Posted: Sun May 22, 2022 1:23 am
by El_Syd
PikWik wrote: Sat Apr 30, 2022 8:40 pm i have played with SNAC on a modern HDTV, a HD computer monitor, and also have a SNES & CRT.
i play alot of kaizo super mario hacks, and the input response is "identical" to my CRT setup using the MiSTer with a wired daemonbite, vsync2, and a fast computer monitor @ 1080p
What monitor do you use?

Re: Testing Latency of Controllers

Posted: Mon May 23, 2022 3:55 am
by PikWik
i use an Acer Nitro XF243Y, which is a reasonably priced gaming monitor in the $160 range
https://www.rtings.com/monitor/reviews/ ... y-pbmiiprx