I'm new to Intel Communities, and I would like to tell some issues I get with my computer with Intel Mobile 4 Series GL40. It has Pentium Dual-Core T4400 2.20 GHz, 4 GB DDR3 1066 MHz RAM, and Windows 7 Home Premium x64 with the latest generic graphics driver (220.127.116.1169).
Toshiba hosts the drivers on their site but they aren't updated and they are the same drivers that are preinstalled on my computer.
Here are some bugs I've seen with the latest generic driver and an old OEM driver (18.104.22.1683):
- When an application draws with the GDI StrectBlt function repeatedly, and when the destination rectangle is bigger than the source rectangle, there are some horizental lines that get bigger and smaller every time. This occurs when Aero is activated and has more chance to occur if the screen resolution is bigger (I have 1366*768 native resoultion). This may be a bug in some WDDM 1.1 GDI acceleration functions, because I tested a computer with Intel 82945G and WDDM 1.0 driver and that doesn't occur. I made a sample program and I can upload it if wanted.
- When I use screen rotation, the mouse's position is incorrect (1 pixel more in the right and/or down depending on the rotation).
- The files of the latest driver are called "Intel Graphics Accelerator Drivers for Windows XP(R)", even if it's for Windows Vista/7. In the old driver, it's "Intel Graphics Accelerator Drivers for Windows 7(R)"
- In Direct 10.1 and 11, the feature levels 9.2 and 9.3 aren't supported, but 9.1 and 10.0 are. I see that other graphics drivers except Intel's don't have this issue, if they support level 10.0 then they also support 9.1, 9.2 and 9.3.
- The D3DPRASTERCAPS_ZTEST capability flag in Direct3D 9 is not supported. Why?
- The Stream-Out functionality in official drivers is REALLY slower than in some modded drivers.
Some of these bugs may also occur in other graphics chipsets. For example, the Intel 82945G has the mouse position and the D3DPRASTERCAPS_ZTEST bugs, but not the StretchBlt bug (probably because it only supports WDDM 1.0). I don't know if they are also present in new chipsets.
So, can someone fix these bugs?