Chaintech A-GT61 GF4 Ti4600 128M Review (2)
Overclocker 18 Apr 2002

Benchmarking

We conduct the a number of tests on the setup with the following configuration :

Test System:

• AMD AthlonMP 1.2G o/c to 1.33G (10x133)
• ASUS A7V333 running async DDR333 with best timings
• Maxtor 40G 7200RPM ATA133 HD
• AOPEN 40X CDROM 
• Chaintech A-GT61 GF4 Ti4600 graphics card w/Detonator driver 27.30
• Nanya DDR333 256MB DDR 
• Channelwell 375W AMD Approved PSU 

Quake 3 Demo 1 

Quake3 is probably the one of the most important software to be used for testing graphics cards. In our test, we ran the test at least 3 times to obtain a stable sustained result. In our chart, we compared the results of Chaintech A-GT61 in non-oc and oc.

Below are the results of Q3 Demo1 running at various resolutions in 32bit colour depth.

The results are pretty good when running at 150/650 which is the default core clk/memory clk timings of the card. In our tests, we overclock the card using the supplied software TurboGfx. You can also use nvmax and it will generate the same results at 155/728. In fact, as you know GF4 Ti4600 is already tuned to it's maximum and there isn't much room to o/c it further as compared to the GF3 ti200 series of cards. Our maximum stable o/c is at 155 for core and 364Mhz (728Mhz) for memory. If we were to push the core to 160Mhz, the graphics card will not be able to take the high 728Mhz for memory. If we fix the core at 150Mhz default, we were able to run the memory at around 740mhz without artifacts.

As we can observe from the chart above, the difference in results in o/c and non-oc isn't that apparent except that when running it at higher resolutions. The differences are 0.17% ~ 3.6%

Next

(C) Copyright 1998-2009 OCWorkbench.com