Mouse Issues

[Julian]

Hi,

Might anyone know why I'm getting the following strange mouse problems in all the demos, I'm running PB 5.10 64bit in windows 8 (if that helps diagnose the problem) (same thing happens in 32bit PB)

Left and right mouse movement works fine no matter how fast I move the mouse.

If I move the mouse slowly diagonally it works ok, but as soon as I start going very fast diagonally, the program locks up while I keep moving diagonally, until I stop, when life returns to the program.

--- update ---

ok, I just tried a different mouse and it seems to be ok

The problem mouse is a razer naga hex, which has 5600dpi set in the drivers, and motion set in windows to setting 2 with pointer precision turned off.

The mouse that works ok is a logitech mx518. Im guessing that it works ok because there is no driver based mouse acceleration unlike the razer.

Now the razer works fine in windows, and all other games, its just causing problem with the demos running via PB so I'm going to assume that its a problem here, rather than something in windows.

If someone could help me get to the bottom of this I'd appreciate it, logs etc can be provided where needed, or test code run.

Thanks in advance,

Julian

[Julian]

Update

I've narrowed it down to the following:

Polling Rate in the razer driver set to 1000 causes freezes in the mouse handling events in PB

If I change the polling rate to 500 or 125 (the three options are 1000,500,125) in the razer driver, all the problems stop, and the mouse is handled correctly.

I'm assuming that the frequency of mouse events that are being fired cant be handled by ExamineMouse, causing the freeze.

[Julian]

Here you go, I've made a little unlisted video showing the problem:

http://youtu.be/RAhDyEUDp14

[Grunz]

We got similar problem reported from windows users.
For us it happened when the mice have a too high sample rate (resolution was not the issue).
Our game was reading input 100 times per second, which is more than fast enough for human input.....
BUT high sample rate mice will cause some overflow issues if mouse data is not read often enough.
So try lowering the sample rate of the mouse from 1000 MHz or whatever too high value it uses.
If that fixes the problem, then you can make your code working by reading the mouse data more frequently.

[Julian]

We got similar problem reported from windows users.
For us it happened when the mice have a too high sample rate (resolution was not the issue).
Our game was reading input 100 times per second, which is more than fast enough for human input.....
BUT high sample rate mice will cause some overflow issues if mouse data is not read often enough.
So try lowering the sample rate of the mouse from 1000 MHz or whatever too high value it uses.
If that fixes the problem, then you can make your code working by reading the mouse data more frequently.

Thanks for posting and showing a 2nd confirmed sighting of this problem. As I suspected its to do with the 1000hz poll rate causing a flooding/data rate issue with the way PB handles the mouse.

How do you suggest that I read the mouse more often, my loop is pretty tight with one ExamineMouse in there.

I'm not sure its acceptable to ask users to change the polling rate of their hardware, considering my mouse works fine with every other program/game I run on the system? This would lead me to believe there is some flaw in the mouse handling code somewhere. Any thoughts on this Fred?

Thanks in advance,

Julian

[Thade]

http://www.purebasic.fr/english/viewtop ... f=4&t=4701

.

[Julian]

http://www.purebasic.fr/english/viewtop ... f=4&t=4701

.

Yeah, I've read that, I didn't know it was a "crime" to post additional information to try and aid narrowing down a bug... or to discuss ways of getting around a bug in the bug thread.

Could you do me a favor and highlight which point you think I should re-read on the link you provided?

Thanks

Julian

[Demivec]

Yeah, I've read that, I didn't know it was a "crime" to post additional information to try and aid narrowing down a bug... or to discuss ways of getting around a bug in the bug thread.

Could you do me a favor and highlight which point you think I should re-read on the link you provided?

IMHO, maybe points 1, 4, 5, (6), 7, 8?

Your problem (not bug) seems to revolve around your choice of hardware and not about PureBasic functions as documented. That would seem to make the resolution of your problems a feature request and not a bug report.

[Julian]

Yeah, I've read that, I didn't know it was a "crime" to post additional information to try and aid narrowing down a bug... or to discuss ways of getting around a bug in the bug thread.

Could you do me a favor and highlight which point you think I should re-read on the link you provided?

IMHO, maybe points 1, 4, 5, (6), 7, 8?

Your problem (not bug) seems to revolve around your choice of hardware and not about PureBasic functions as documented. That would seem to make the resolution of your problems a feature request and not a bug report.

IMHO, maybe people want to lighten up and spend time actually being a bit more constructive...

Since when does a problem with mouse input (which is pretty core to most games) become a feature request, are you for real?

1. Latest public release, as posted in the original bug report (as explained in the bug report)

4. This isn't a case of lack of knowledge or using commands incorrectly, as this is happening in example code (as explained in the bug report).

5. The mouse input isn't working as it should.

6. Replication is tricky without access to another PC at the time of posting.

7. The problem was occurring in "all the demos" as stated in the initial post, so code doesn't need to be posted.

8. Full description of bug reported, the problem is with evident with example code, so no need to post code. Video uploaded to show the problem happening. Cant really get much more helpful than that?

Your attitude to "my hardware" "my problem" is pretty sad, it seems like a very strange way of thinking about PB as a platform in general. I just can't imagine releasing a game with the mouse input issues described above. The fact that its showing up in the example code which is included with PB concerns me and that any 1000 mhz polling mouse would probably display the same issues.

You do realise than when I'm saying "my mouse" or "my hardware", I'm actually meaning the people who could be using the programs / games made by people using PB? This isn't about fixing my hardware issue, its fixing a bug that could cause end users issues, issues that are seemingly out of our control to remedy on mass.

[Demivec]

Yeah, I've read that, I didn't know it was a "crime" to post additional information to try and aid narrowing down a bug... or to discuss ways of getting around a bug in the bug thread.

Could you do me a favor and highlight which point you think I should re-read on the link you provided?

IMHO, maybe points 1, 4, 5, (6), 7, 8?

Your problem (not bug) seems to revolve around your choice of hardware and not about PureBasic functions as documented. That would seem to make the resolution of your problems a feature request and not a bug report.

IMHO, maybe people want to lighten up and spend time actually being a bit more constructive...

I am actually indifferent to your feature request. If a solution is found it would be great but it doesn't affect me much at the moment.

Since when does a problem with mouse input (which is pretty core to most games) become a feature request, are you for real?

Didn't you post a link earlier which described individuals all experiencing problems with the razer mouse and its high polling rate and none of them were using PureBasic? Weren't the issues reported in it 3 or 4 years old?

Polling Rate in the razer driver set to 1000 causes freezes in the mouse handling events in PB

If I change the polling rate to 500 or 125 (the three options are 1000,500,125) in the razer driver, all the problems stop, and the mouse is handled correctly.

Change the driver settings for the polling rate to less than 1000Hz.

[Thunder93]

Julian, don't take it to heart when people suggests to read the bugs forum policy. They are ultimately trying to help Fred and freak and whomever. Most of the posts screaming bugs aren't bugs but user-errors. And more time Fred and freak and whomever spends going over them all, is time lost when it could have been time well spent.

Now that more information about the your experiences is in, one could now say that this topic is more appropriate for the feature requests forum. You wanting PB to support the updated technology, but PB was designed and tested for definitely mice with the default USB polling rate of 125 Hz – 8ms response time. They might of have designed it to handle faster USB polling rates but ... I think its safe to say 1000 Hz wasn't one of those. http://www.codinghorror.com/blog/2007/0 ... -rate.html

[Julian]

Cheers Thunder93. A good nights sleep, some more reading and I can now see the problem more clearly.

I'll thread off the mouse events to make sure they don't get held up by the main loop.

Apologies for getting a bit antsy and wasting peoples time.

Could someone move this out of bugs and put it in questions, or just nuke it from orbit?

Thanks

J

[Thunder93]

I like this thread, brings awareness to the problem and how it'll have to be addressed with a little extra work.

So no nuking please.

[Raybarg]

It appears I am having similar problem in Windows 7.

Depending how much I use mouse (mostly related to pressing mouse buttons) the program will crash after a while. Before crashing mouse locations are not updated regularly, mouse pointer is moving jerky.

Installed my mouse own applications (logitech) to manage the update speeds (dropped it to lowest setting), I also ran some registry modifications I found from interwebs with decent amount of googling. Only the logitech application had an impact, it fixed the mouse jerkyness, but app still crashed after a while.

But thats not all, you know how we all tend to circle our mouse pointer on windows desktop while waiting something to happen... ever since I had this issue with PB, I also have not been able to enjoy that desktop pass time activity, because the selection area is similarly jerky as mouse was jerky in PB app.

So, I will continue trying to figure out where the problem is. But so far, PB seems just showing the problem (better than some java apps) and does not really seem like thats where the problem is.

(one other change that happened when these problems started to appear was my motherboard I/O chips starting to make that high pitch high frequency "whining" sound, or continuous beep sound... very long long time ago I was fixing CRT monitors in one electronics company and learned what "upset" I/O chip might sound like, yet, remain mostly functional... so who knows if I am just dealing with broken/dieing chip)

I try to remember to update here if I actually figure out something that would have been worth of posting here in the first place.