A Closer Look

The cooling assembly is made of several parts. First you remove the plastic top cooler which houses the dual fans. Please note here that both fans are connected to the same cable which means they are both temperature controlled but always running at the same speed.


Directly below the fans you will find a large heatpipe based heatsink module. It uses three heatpipes and a large number of fins to move heat away from the GPU quickly, no other components are cooled by that part of the cooling system. Further to the back you will find a small black heatsink that is responsible for cooling the power regulation circuitry.


Once the heatpipe heatsink is removed, another piece of metal can be found on the rest of the card. It overs all the other components that need some form of cooling, like the memory chips, the NVIO chip and the memory voltage regulators.


You may combine two or more of these cards to build an SLI rig for even higher performance or image quality settings.


The card has two six-pin power connectors, both are required for operation of the card. Further to the right on the picture is the white input for the SPDIF audio to be streamed into the HDMI output.


The GDDR3 memory chips are made by Samsung and carry the model number K4J52324QH-HJ08. With a latency of 0.8 ns, they are specified to run at 1250 MHz.


Unlike the NVIDIA reference design which uses a Volterra voltage regulator, Palit has chosen to go with a Realtek design which is probably a lot cheaper. Unfortunately this means that there is no software voltage control possible because the controller chip has no I2C interface which is required for communication with the host PC.


NVIDIA has separated the display output logic from the GPU on the latest chips. So in order to drive the display outputs the NVIO chip is required.


Here you can see the GT200b GPU that powers the GTX 260. It is made in a 55 nm process with 1.4 billion transistors.