A month after the release of the GTX 580, Nvidia is back again but this time to launch their GTX 570 enthusiast video card. The GTX 570 offers a 25% performance increase from the GTX 470, and also a slight performance increase from the last generation GTX 480.
The Nvidia Geforce GTX 570
Just a month ago, we saw the release of the Nvidia GeForce GTX 580, running with the newly optimized and enhanced GF110 chip. This newly redesigned chip proved to be the fastest single GPU on the planet, allowing Nvidia to take the crown for the fastest and quietest video card on the market. With 512 CUDA cores and 16 PolyMorph engines, the GTX 580 was able to pull ahead of the GTX 480 by about 15-20%. When we pushed the card even further by overclocking, it could beat two factory super overclocked GTX 460 (GF104) GPUs, and was about 30-35% faster than the GTX 480. The Nvidia GeForce GTX 580 was designed for the enthusiasts looking for the ultimate performance that money could buy.
Today, Nvidia is releasing their latest and second top of the line video card tailored for the enthusiasts: the Nvidia GeForce GTX 570. The GTX 570 also uses the new GF110 chip, 16 PolyMorph Engines, and has support for FP16 texture filterand and boast improved Z-culling, just like its more powerful sibling, the GTX 580.
But why are we mentioning that the GTX 580 is more powerful than the GTX 570? The GTX 570 has been limited in terms of the GPU performance and memory. 32 of its CUDA cores have been disabled, making it a 480 core GPU, along with a total of 1280MB of 320-bit GDDR5 memory. The texture units, ROP units and streaming multiprocessors count have been dropped as well from 64 to 60 texture units, 48 to 40 ROP units and from 16 to 15 streaming multiprocessors. This at first might sound like a worthless card with the current GTX 480s out on the market, but with the release of the GTX 570s, Nvidia will discontinue the GTX 480 GF100 GPUs.
|So what can we expect from the GTX 570 if it is essentially a GTX 470 with the CUDA cores of a GTX 480? If we look back at the GTX 470′s specifications, we can see that the GTX 470 had 448 CUDA Cores (Shader Units) instead of the 480 the GTX 570 offers. The GTX 470 also had 40 ROPs, and a core clock of 607 MHz, and 837 MHz memory frequency. However, the GTX 570 has a core clock of 732 MHz and 950 MHz memory frequency. Immediately, we can see that the GTX 570 has a lot more performance up its sleeves than the older GTX 470, due to its 32 more CUDA cores, and higher frequencies on the core clock and memory clock.|
We also have a feeling that with these specs, the GTX 570 might be a bit faster than the old GTX 480. If that is the case, then it would not be a problem for the GTX 570 to replace the GTX 480, because the GTX 570 is anticipated to launch at around $349. There will be factory overclocked cards available immediately which might have a slightly higher price, but the stock cards should run around $349. We have also heard that there will be future non-reference card designs of the GTX 570, but we cannot disclose the date and time when those cards will become available.
As a result of the GPU redesign, the GeForce GTX 580 also has a lower temperature threshold than its predecessor. Whereas the 480′s GF100 chip had a temperature threshold of 105C, the 580′s and 570′s GF110 has a threshold of 97C.
The GTX 570 is also the same length as the GTX 480 and the GTX 580. Users should be able to run up to 3-way SLI with the GTX 570s.
Nvidia’s New Demos
Nvidia first released some exciting Demos that showcase Nvidia’s tessellation capabilities on the GeForce GTX 580 video card. The demos have not changed for the GTX 570. The users can still enjoy playing and checking out the features and performance of each 500 series video card on the following demos. While these demos will play on the 400 series video cards as well, to have optimal single GPU performance, users will need a GeForce GTX 580/570 video card. The first picture is from Endless City, a tessellated city landscape generated by the GPU. All the fine detail in the buildings are tessellations and not bump maps that we are well aware of. All the lights in the scene are able to produce excellent shadows because we no longer use bump maps, but rather a higher polygon geometry count.
The second picture shows the Aliens vs. Triangles tessellation demo, in which users can modify the aliens with very fine detail. Once again, instead of having bump maps, the alien is very high detailed in geometry. This allows extra options to be integrated, such as making sure that if something interacts with the alien, the skin would change accordingly. This also allows for much higher quality rendering that was not possible with bump maps in the past.
Click Image For a Larger One
The GF110 Architecture – Improved / Optimized FERMI
The GTX 570 has not changed much from the GTX 580, and our GTX 580 review has a detailed explanation of all the improvements and optimizations the new GF110 went through. We have mentioned the FP16 texture filtering and Z-cull on the first page. Of all the the architectural enhancements the GTX 580/570 went through, these were the two major ones. The FP16 texture filtering helps with texture-intensive applications. The chart below from Nvidia shows how the architectural enhancements improved performance from the GTX 480. The extra performance granted from the faster clock speeds on the core and memory, and the extra 32 cores that were unlocked on the GTX 580, made it a true 512 CUDA core GPU. In the GTX 570′s case, there are 480 CUDA Cores, but it should still have a similar effect in performance if we consider that the GTX 470 has 448 CUDA Cores.
With the new GF110 chip, PolyMorph and Raster Engines have been added to help with tessellation. While the new PolyMorph engine helps with tessellation performance in games, the extra Raster Engine helps with the conversion of polygons to pixel fragments. Now with 16 PolyMorph Engines (15 PolyMorph Engines on GTX 570) and 512 CUDA cores (480 CUDA Cores on GTX 570), the cards are able to achieve a stunning 2 billion triangles per second. That is a tremendous amount of polygons, something we would only see in Hollywood blockbuster movies. Now all of this can easily be rendered real-time with the GTX 580/570 GPUs. Nvidia’s new demo Endless City shows this off, rendering and playing back everything in real-time.
The Radeon HD series video cards still have a much harder time with tessellation based benchmarks, which means that when games start incorporating extensive tesselation into their geometry, the Nvidia cards will have an advantage over their AMD counterparts. There are some games that already take advantage of tessellation, like H.A.W.X 2. The Unigine Heaven 2.1 benchmark also tests tessellation capabilities. While the tessellation visual improvement is very limited at the moment, we believe that tessellation will be taken much further in the future, making it possible to make characters, terrain, and objects much more belieavable than they are now.
For the GF110 design, Nvidia completely re-engineered the previous GF100, down to the transistor level. The previous chip had to be evaluated at every block of the GPU. To achive higher performance with lower power consumption, Nvidia modified a very large percentage of the transistors on the chip. They used lower leakage transistors on less timing sensitive processing paths, and higher speed transistors on more critical processing paths. This is why Nvidia was able to add the extra 32 cores to the final Fermi architecture (GTX 580/570), while also adding another SM to the chip (GTX 580).
For many of Nvidia’s previous video cards, the GPU’s thermal protection features meant that the GPU would be downclocked when at extreme temperatures. This would protect the cards from unwanted damage. However, with the release of stressing applications such as FurMark, MSI Kombustor, and OCCT, the latest video cards can reach dangerously high currents, potentially causing damage to components on the card. Nvidia integrated a new power monitoring feature into the GTX 580/570, which will dynamically adjust performance in certain stress applications if the power levels exceed the card’s specifications. These dedicated hardware circuitries run real-time, monitoring the current and voltage on each of the 12V rails. These rails include the 6-pin, 8-pin (GTX 580) or 6-pin, 6-pin (GTX 570), and the PCI-Express edge connector.
Nvidia made improvements when developing the GTX 570 based on what consumers said about the GTX 470. The thermal characteristics of the GF110 chip are also much better than of the GF100. What we see on this chart is that the GTX 480 is roughly about 9-10 dBA higher than the GTX 580. Generally, a human perceives each 10 dBA increase as being twice as loud as the previous noise level. The GTX 580 will perform much quieter than any high-end card Nvidia has released in the past few years.
Based on the tests we did in our labs, the GTX 580/570 does indeed perform very quietly during high loads. We tested the thermal improvements and acoustic improvements on the card, and with our Silverstone TJ-10 chassis and some acoustic dampaning on each side panel, the GTX 580/570 was totally inaudible during gaming. The other fans in the system were a bit louder than the GTX 580/570. When we ran Furmark, the fan speed starts getting faster. However, Furmark is not a real-life based application because it actually pulls more power and heat out of the video card than a real-life application would. Also, if we push the fan speed on the GTX 580 to 100%, we can definitely hear the fan loud and clear, but the fan is limited to go only up to 85%. Depending on the manufactorer of the card, there might be tweaked BIOSs, which allow the cards to push the fan speed all the way up to 100%. During our testing period, we played Metro 2033 for about an hour in a closed chassis with no side ventilation, and the fan speed only reached up to 58%, which kept a very quiet environment for gaming.
The new cooling solution on the GTX 580 uses a special heatsink design, including what is called a vapor chamber. Nvidia explained that the vapor chamber on the GTX 570 is slightly different from the vapor chamber on the GTX 580, due to the less heat the card needs to deal with. Think of the vapor chamber as a heatpipe solution, but instead of just contacting the heatsink fins in certain areas, the vapor chamber has 100% contact with every fin of the heatsink. This helps tremendously by spreading the heat out over a large block of a heatsink.
The GTX 580/570 also has a new adaptive GPU fan control, and the card is designed for great cooling potential in SLI setups. The fan has been redesigned to generate a lower pitch and tone, which allows for lower acoustic noise. The back of the cover is designed to route the air towards the rear bracket, improving SLI temperature performance.
The vapor chamber is a sealed, fluid-filled chamber with thin layered copper walls. When the heatsink is placed on the GPU, the GPU quickly boils up the liquid inside the vapor chamber, and the liquid evaporates to vapor. The hot vapor spreads throughout the top of the chamber, transferring the heat to the heatsink fins. Finally, the cooled liquid goes around and returns to the bottom of the vapor chamber, allowing the whole process to restart again. The hot heatsink fins are cooled by the air being pushed through the fins of the heatsink.
Continue onto the next page, where we examine the Nvidia GeForce GTX 570 in more detail.
What is CUDA?
CUDA is NVIDIA’s parallel computing architecture that enables dramatic increases in computing performance by harnessing the power of the GPU (graphics processing unit).
With millions of CUDA-enabled GPUs sold to date, software developers, scientists and researchers are finding broad-ranging uses for CUDA, including image and video processing, computational biology and chemistry, fluid dynamics simulation, CT image reconstruction, seismic analysis, ray tracing and much more.