NVIDIA's shady trick to boost the GeForce 9600GT

CDsDontBurn · Feb 29, 2008, 07:45 PM

When we first reviewed NVIDIA's new GeForce 9600 GT (review here), we noticed a discrepancy between the core clock speed reported by the driver and the core clock speed read from the clock generator directly.

RivaTuner Overclocking and GPU-Z read the clocks from the NVIDIA driver, displaying whatever the driver returns as core clock. Rivatuner Monitoring however accesses the clock generator inside the GPU directly and gets its information from there. A PLL to generate clocks works as follows. It is fed a base frequency from a crystal oscillator, typically in the 13..27 MHz range. It then multiplies and divides this frequency by an integer value to reach the final clock speed. For example 630 MHz = 27 MHz * 70 / 3.

Source: techPowerUp!

Judas · Feb 29, 2008, 08:00 PM

so....how does this work when you try and overclock.. if you bump it up only 5Mhz.. it's actually a bigger overclock?

while i'd rather have higher clocks... then stated..... that's quite a large difference in fequencies.

reno · Feb 29, 2008, 08:53 PM

That's quite a misleading title.

Nvidia responded that the Crystal is set to 25Mhz, so my bet is that Rivatuner Monitoring is reading it wrong or set wrong.

Besides, a few MHz overclock isn't going to account for the drastic performance improvement anyway. So I don't see how this is a "shady trick."

edit: After reading the rest of the article (bad web design FTL) and the mention of it generating clocks per PCI-E bus, I still don't see any "shady tricks."

Erroneus · Mar 1, 2008, 01:15 AM

Well it can great some problems, if a user is overclocking his pci-e, he should also be aware he is clocking his GFX a lot more then normal.

mkk · Mar 2, 2008, 10:26 AM

Let's see if they keep denying it in the coming week.

Matth · Mar 2, 2008, 06:05 PM

Interesting, so if this is detected as "linkboost" capable, giving the 25% boost to 590 chipset PCI-E and SPP-MCP links, then it will be "overclocked" by 25%.

You would then ask, though, if it is capable of those speeds, why not deliver them as standard, unless the idea is to make magic when you use an Nvidia card and an Nvidia chipset.

Feb 29, 2008, 07:45 PM	#1
CDsDontBurn 127.0.0.1 Join Date: May 2003 Location: Southern California Posts: 22,458 Rep Power: 202 System Specs	NVIDIA's shady trick to boost the GeForce 9600GT When we first reviewed NVIDIA's new GeForce 9600 GT (review here), we noticed a discrepancy between the core clock speed reported by the driver and the core clock speed read from the clock generator directly. RivaTuner Overclocking and GPU-Z read the clocks from the NVIDIA driver, displaying whatever the driver returns as core clock. Rivatuner Monitoring however accesses the clock generator inside the GPU directly and gets its information from there. A PLL to generate clocks works as follows. It is fed a base frequency from a crystal oscillator, typically in the 13..27 MHz range. It then multiplies and divides this frequency by an integer value to reach the final clock speed. For example 630 MHz = 27 MHz * 70 / 3. Source: techPowerUp! __________________ Her First Day \| A year later

Mar 1, 2008, 01:15 AM	#4
Erroneus Get off my lawn! Join Date: Aug 2002 Location: Denmark Posts: 13,342 Rep Power: 114 System Specs	Well it can great some problems, if a user is overclocking his pci-e, he should also be aware he is clocking his GFX a lot more then normal. __________________

Mar 2, 2008, 06:05 PM	#6
Matth Flash Banner Hater Join Date: Jun 2002 Location: UK Posts: 3,409 Rep Power: 91 System Specs	Interesting, so if this is detected as "linkboost" capable, giving the 25% boost to 590 chipset PCI-E and SPP-MCP links, then it will be "overclocked" by 25%. You would then ask, though, if it is capable of those speeds, why not deliver them as standard, unless the idea is to make magic when you use an Nvidia card and an Nvidia chipset. __________________ Mary had a little lamb, Her father shot it dead Now Mary takes her lamb to school, Between two crusts of bread

Feb 29, 2008, 08:53 PM	#3
reno Painlord of Ichor Join Date: May 2002 Location: bloinkin! Posts: 1,610 Rep Power: 0	That's quite a misleading title. Nvidia responded that the Crystal is set to 25Mhz, so my bet is that Rivatuner Monitoring is reading it wrong or set wrong. Besides, a few MHz overclock isn't going to account for the drastic performance improvement anyway. So I don't see how this is a "shady trick." edit: After reading the rest of the article (bad web design FTL) and the mention of it generating clocks per PCI-E bus, I still don't see any "shady tricks."

Mar 2, 2008, 10:26 AM	#5
mkk Cthulhu/Dagon 2012 Join Date: Oct 2003 Location: Gefle, Sweden Posts: 4,375 Rep Power: 120 System Specs	Let's see if they keep denying it in the coming week.