New device: GILT Game Input Lag Tester

[Oomek]

Sorry for using the wrog term. By saying cable latncy I meant the lag caused by all the electronic components and postprocessing, excluding the mechanical part (LCD crystals)

[Oomek]

On the second thought, using the gray square for the threshold calibration is a bad idea after all. The threshold needs to be gamma independent and the system, gpu and tv can have various gamma settings. My device needs to ignore that, so I can only use the square with the value of 0 or 255 , measure the min/max sensor output values and calculate the threshold by doing simple math
((sensorMax-sensorMin) * threshold) + sensorMin
It's the most foolproof method in my opinion.
I just need to wait for the new sensors to finalize the project. Should have them in the middle of the week.

[AustinClark]

Nice project. I'd like to tinker with something similar soon

What USB-enabled microcontroller are you using, just out of curiosity?

Personally, I'd like to start off using a scope to directly measure everything at once. However, I suspect that every mouse and/or microcontroller emulating a mouse might have variable delay between a "click" and when the data is actually ready to be polled. Then there's the problem that different games/applications might cause their own delay. Lots of variables that are hard to predict and control. Even still, it would be nice to get SOME hard data for my setup.

I wonder if it's possible to create a faux mouse with a polling rate much higher than 1000Hz (or even the overclocked 3000Hz mice I've heard about) with a microcontroller. Just for testing purposes.

[Sparky]

Nice project. I'd like to tinker with something similar soon

What USB-enabled microcontroller are you using, just out of curiosity?

Personally, I'd like to start off using a scope to directly measure everything at once. However, I suspect that every mouse and/or microcontroller emulating a mouse might have variable delay between a "click" and when the data is actually ready to be polled. Then there's the problem that different games/applications might cause their own delay. Lots of variables that are hard to predict and control. Even still, it would be nice to get SOME hard data for my setup.

I wonder if it's possible to create a faux mouse with a polling rate much higher than 1000Hz (or even the overclocked 3000Hz mice I've heard about) with a microcontroller. Just for testing purposes.

Possible, yes, with a microcontroller capable of USB2 High Speed you could presumably get 8khz. Might need custom drivers to get anything out of it though.

Also, I've used an arduino micro for a similar testing setup, and flood is using a teensy 2.0. IIRC, both use an Atmel atmega 32u4.

[Oomek]

Nice project. I'd like to tinker with something similar soon

What USB-enabled microcontroller are you using, just out of curiosity?

Personally, I'd like to start off using a scope to directly measure everything at once. However, I suspect that every mouse and/or microcontroller emulating a mouse might have variable delay between a "click" and when the data is actually ready to be polled. Then there's the problem that different games/applications might cause their own delay. Lots of variables that are hard to predict and control. Even still, it would be nice to get SOME hard data for my setup.

I wonder if it's possible to create a faux mouse with a polling rate much higher than 1000Hz (or even the overclocked 3000Hz mice I've heard about) with a microcontroller. Just for testing purposes.

It's an Atmega32u4 board.
I'm using a serial over USB with a baud rate of 1000000 and Serial.available() as a timer trigger. Thats way faster any mouse can give, emulated or physical overclocked.

[Sparky]

It's an Atmega32u4 board.
I'm using a serial over USB with a baud rate of 1000000 and Serial.available() as a timer trigger. Thats way faster any mouse can give, emulated or physical overclocked.

You're still using an analog read, right? What's the ADC clock set to?

[Oomek]

Yes, I'm using analog read. It's on default 16MHz. I could set a prescaler to 16, but I don't see the point as the default sampling gives me 9.6kHz which is more than enough for my purpose.

[flood]

i'm in the process of (re)building my photodiode circuit
this time it will have bandwidth above 1MHz to allow for true microsecond measurements

@op there's a lot of stuff i can share with you if you'd like. depends on how accurate/precise you want things to be in the end. i'm aiming for <1us for my new setup and sensitivity such that 1 scanline from my crt is detectable.

but to start with:
1. ideally you want a detector that looks at the entire screen, not just one small part. because unless you have vsync, any part of the screen could be the first part to be updated
2. photoresistors tend to be slow but if ~1ms is enough for you it might be ok. but you still want to know how much its speed contributes to systematic error. i'm not sure what would be the easiest way for this though.

[flood]

5-6 ms? u sure m8? that's a pretty quick response for mice > game > display.

uwotm8

0.572ms for my old setup (teensy > custom dx9 program at ~10000fps > crt)

https://docs.google.com/spreadsheets/d/ ... 2006064244

[spacediver]

i'm in the process of (re)building my photodiode circuit
this time it will have bandwidth above 1MHz to allow for true microsecond measurements

@op there's a lot of stuff i can share with you if you'd like. depends on how accurate/precise you want things to be in the end. i'm aiming for <1us for my new setup and sensitivity such that 1 scanline from my crt is detectable.

I'm already drooling.