R580+: Higher Clocks and GDDR4
ATI’s answer to the gaining popularity of the multi-GPU GeForce 7950 GX2 is the X1950 XTX — still a single GPU card but one that performs amazingly close to NVIDIA’s flagship part.
We try to be topical at PC Perspective, but sometimes its hard. While I’d love to have more Core 2 Duo benchmarks for you to gander at, ATI went and threw a wrench in the works by sending along a pair of new GPUs for me test. The R580+, that is now officially known as the X1950 XTX, is a minor upgrade to the existing ATI flagship card that plans on taking away the thunder that NVIDIA has been building over its 7950 GX2 GPU and Quad SLI.
We’ll see here today if they are able to accomplish that.
The Radeon X1950 XTX GPU (codenamed R580+) is, with only a couple minor exceptions, the same GPU that was seen when the X1900 series was released all the way back in January of 2006. With it we saw 48 pixel pipelines, a new programmable memory architecture and several other features that ATI developed to take on NVIDIA’s GPUs. Below is a short summary of the technical details included in our previous X1900 launch article.
R580 Architecture – Radeon X1900
That mash of lines and squares is actually showing 48 pixel pipelines in the new X1900 architecture, three times as many as we had in the X1800! Those 48 pixel pipes are still grouped together in sets of four and each pixel pipeline also has a corresponding flow contol unit, bringing them up to 48 total as well.
This quick table shows you the raw processing power that the additional 32 pixel processors bring to the table for the X1900 architecture. It is indeed quite impressive.
An important note here is that besides the additional pixel processing power, no other components in the GPU have really been changed all that much. We’ll get into the reasoning behind ATI’s decision to leave the vertex and rendering stages alone in the R580. It is still based on the 90nm technology that the X1800 used, but tweaked and fixed so that they should be getting much better yields this time around. Some other quick comparison specs on the die and transistor count are here:
R580 vs. R520
90nm process technology
Transistors: 384 Million vs. 320 Million
Die size: 315 sq mm vs. 264 sq mm
You can see that the R580 is a pretty big chip, though it isn’t quite as large as the G70 built on the 130 nm process.
The ATI R580+
So what makes this new X1950 XTX card so much better than the previous generation from ATI? First, it is the first graphics card to introduce GDDR4 memory technology (the GeForce 7-series and X1800/X1900 used GDDR3); thanks to the versatile and programmable memory architecture that ATI built into the core, the transition was very smooth. The only other change is a shift from the noisy fan system the X1900s used to a new, quieter design.
The GPU clock stays at 650 MHz, just as it was on the X1900 XTX, and the memory buffer remains at 512MB as well. The new GDDR4 memory clocks now run at 2.0 GHz, though a big increase from the 1.55 GHz that was on the high end X1900.
The new heatsink on the X1950 XTX is much quieter and more efficient than the previous ATI design. Similar to the 7900 GTX cooling design, this two-slot design uses a heatpipe to transfer the heat from the GPU to a larger heatsink. The fan is located on the rear of the card as ATI claims this will lower the noise level even more, keeping the fastest moving air inside the case.