Yeah, hi, I'm OP of that thread.
Originally Posted by SmashTV
I tried to get the OP in that reddit thread to use mouse tester but he refused and spazzed out at the idea of not using paint.
1. I did not refuse. As I said, I am on a holiday trip away from home and I don't have my G203 or a computer on me. That is why I can't redo the tests with mousetester right now. Not because I have some irrational aversion to using it.
2. I also said that I will do mousetester tests as soon as I get home. Should be tomorrow or the day after.
3. I also didn't spazz out at the idea of not using paint. But I don't think that matters, you strike me as the type of person who only hears what you want to hear anyways.
Originally Posted by Alya
It could be related to sensor position and box position, if your wrist rotates and you cause the cursor to go up/down then it will show up in the results slightly, there are also other things like normal amounts of SRAV.
Yeah, I get what you're saying, but the G900 goes through exactly the same (flawed) testing methodology and shows no acceleration. It's not like I'm swiping differently with the G900 than with the G203. And it's hyperconsistent, too. The G203 always shows acceleration regardless of how I swipe. I think there's 20-30 swipes shown in that one video, and all of them had accel without exception. I did probably over a hundred more off-camera outside of that video and never got a single one that didn't have accel. The G900, on the other hand, was also hyperconsistent with not having any accel or decel.
TL;DR Yes, the testing methodology isn't perfect and there is human error introduced, but that's what the G900 control sample was there as a baseline for.