Fixed the problem with the lab computer
Aug. 7th, 2011 03:39 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
shockwave77598Those of you following for a bit know that I've had a bizarre problem with my garage computer. This is the quadcore that used to be my main computer. Since it is the last of the XP machines, and it holds and runs programs that either can't be bought for the Win7x64 box (either it's not legit, or the company is no more, or it simply won't work there) I need it to run forever. And you've heard me ponder and wonder why they hell the computer could not run any sort of VR Viewer when the only noted change was moving the computer to the garage -- even the keyboard and mouse are the same!
Well, I solved the problem. And not only that, but I got the thing fixed and now it can run SL just fine. And the problem was nothing I expected at all... not even remotely. Okay, you techie types out there know how the 3d graphics work. Tell me, how can having one device plugged into USB cause a problem with the GPU? I'll let you chew on it for a second before I give the answer...
...
The answer showed up while doing a DXDiag test. The test would die every time it got to Direct Input, which is the providence of Dinput8.dll, one of the DirectX9c drivers. Now, Direct Input is solely for haptic devices, joysticks, keypads, and the like. It had no reason to die because there simply aren't any devices plugged in except the mouse and keyboard. But then I got to thinking, and you know that's not really the case. I have midi interfaces, synth keyboards, programming wands, logic analyzeres... perhaps one of them was the culprit. So I unplugged devices one at a time to find which might be killing DXDiag, and thus SL. Finally, I found a USB device that when unplugged, allowed DXDiag to run properly. And a quick install later to confirm it, SL ran correctly too! TADA, a number of games and a number of things would not work properly because a USB device was screwing up Dinput8.dll
And that USB device, is the PICKIT3 programming wand for the pic processors.
This is not a device that even 1/1000th of the computer users out there will possess, thus it's a very strange glitch for the internet overall. But when it's unplugged, I have my regular game box back. Plugged it, certain stuff stops working because it screws up one of the directX dlls.
So I'll just leave it unplugged unless I'm in Pic programming operations. Tada, problem solved.
Well, I solved the problem. And not only that, but I got the thing fixed and now it can run SL just fine. And the problem was nothing I expected at all... not even remotely. Okay, you techie types out there know how the 3d graphics work. Tell me, how can having one device plugged into USB cause a problem with the GPU? I'll let you chew on it for a second before I give the answer...
...
The answer showed up while doing a DXDiag test. The test would die every time it got to Direct Input, which is the providence of Dinput8.dll, one of the DirectX9c drivers. Now, Direct Input is solely for haptic devices, joysticks, keypads, and the like. It had no reason to die because there simply aren't any devices plugged in except the mouse and keyboard. But then I got to thinking, and you know that's not really the case. I have midi interfaces, synth keyboards, programming wands, logic analyzeres... perhaps one of them was the culprit. So I unplugged devices one at a time to find which might be killing DXDiag, and thus SL. Finally, I found a USB device that when unplugged, allowed DXDiag to run properly. And a quick install later to confirm it, SL ran correctly too! TADA, a number of games and a number of things would not work properly because a USB device was screwing up Dinput8.dll
And that USB device, is the PICKIT3 programming wand for the pic processors.
This is not a device that even 1/1000th of the computer users out there will possess, thus it's a very strange glitch for the internet overall. But when it's unplugged, I have my regular game box back. Plugged it, certain stuff stops working because it screws up one of the directX dlls.
So I'll just leave it unplugged unless I'm in Pic programming operations. Tada, problem solved.
