GeForce GT 230M tested as Generic VGA

[markbyrn]

Here's my score info and Futuremark Sys Info; any suggestion on making sure the test is using the correct GPU?

3DMark Score P2548 3DMarks
CPU Score 16736
Graphics Score: 1987
OS: Windows 7
CPU: Intel(R) Core(TM) i7 CPU Q 720 @ 1.60GHz
CPU Speed: 1596 MHz
GPU: Generic VGA

The Futuremark System Information notes the correct display/graphics adapter:

-Display device
Description NVIDIA GeForce GT 230M
Manufacturer NVIDIA
Total Local Video Memory 1018 MB
Total Local Texture Memory 1018 MB
AGP Aperture Size 2.75 GB
Driver File nvd3dum.dll
Driver Version 8.15.11.8644
Driver Details
Driver Date 7-23-2009
Driver WHQL Certified true
Max Texture Width 8192 px
Max Texture Height 8192 px
Max User Clipping Planes 8
Max Active Hardware Lights 8
Max Texture Blending Stages 8
Fixed Function Textures In Single Pass 8
Vertex Shader Version 4.0
Pixel Shader Version 4.0
Max Vertex Blend Matrices 0
Max Texture Coordinates 8
PCI NVIDIA GeForce GT 230M0x10de0x0a280x363c103c0x00a2
Bus Type: PCI Express
Revision: 2.0
Enabled: true
Rate: 16
Sideband Addressing: Not Supported 0x00000001
Fast Write: Not Supported 0x00000001

[ristoc]

Added the chip, should be visible in few minutes

[markbyrn]

It's showing the correct card now but the testing still doesn't appear to be taking advantage of it with an FPS of no better than 6. Changing the card settings has no effect, whether disabling v-sync or even disabling physx which would presumably slow it down even further. Right after the last test, I ran the Cryostasis TechDemo and measured in at an average FPS of 32.0 and 25 when I disabled physx.

[Jarnis]

Originally Posted by markbyrn

It's showing the correct card now but the testing still doesn't appear to be taking advantage of it with an FPS of no better than 6. Changing the card settings has no effect, whether disabling v-sync or even disabling physx which would presumably slow it down even further. Right after the last test, I ran the Cryostasis TechDemo and measured in at an average FPS of 32.0 and 25 when I disabled physx.

The detection has nothing to do with the way the chip performs in the benchmark - the score would be the same either way. Only your graphics driver is really the critical bit.

3DMark Vantage is a fairly demanding benchmark (to ensure that it can stress all kinds of setups, even those quad-SLI monsters), so it is quite feasible that a laptop chip will get a fairly low FPS - it's normal.

You can't really compare two completely different benchmarks or applications and just stare at the framerate - 3DMark Vantage barely breaks 20fps on high end desktop cards.

As a rough guideline, you can use our videocard performance list

http://service.futuremark.com/hardware

Your system managed a graphics score of 1987 which sounds about right to me. In comparison:

NVIDIA GeForce 8700M GT - 1388
NVIDIA GeForce 130M - 1426
NVIDIA GeFroce 9700M GT - 1591
ATI Mobility Radeon HD 3850 - 1629
NVIDIA GeForce 8800M GTS - 2419
NVIDIA GeForce 9700M GTS - 2475

(These are estimates of average *graphics score* (not the overall score) based on all benchmark submits using these graphics chips)

NVIDIA GT 230M is not yet on the list (not enough submits, it is fairly new) and with about 2000 in graphics score would be slower than 8800M GTS (older high end laptop chip) but considerably faster than GeForce 130M (previous generation chip from the same level).

If you expected better performance in demanding 3D, it should be noted that laptop chips in general are still quite a bit slower than desktop cards and the naming conventions can make things confusing. Based on your results, your system is still pretty good for a laptop and comparable to such desktop cards like ATI Radeon HD 4650, ATI Radeon HD 2900 GT and NVIDIA GeForce 9600GS.

[Meaker]

http://service.futuremark.com/compare?3dmv=1284100

Thats an overclocked mobile 4650. Also my machine was running a 2.8ghz Core 2 duo (which really ran a bit too hot for the machine).

We get about the same graphics score. Your chip is about the same as an 8800GS (AKA 9600GSO) just clocked a little lower.