Review NVIDIA GeForce GTX 570: All the Features for Less

We’ve witnessed the awesome performance of GeForce GTX 580, NVIDIA’s latest flagship intended to replace the previous throne holder, the venerable-yet-insufficiently cool GeForce GTX 480. NVIDIA redesigned their GF 100 chip, optimizing its design down to every transistor, and came up with the brand new GF 110. This new chip has a better performance/ watt ratio and a much lower operating temperature. It appears that NVIDIA had done a good job with their homework this time.

So, after previously carrying out extensive tests on the GTX 580, now we bring you yet another exclusive review on an interesting new graphics specimen: the GeForce GTX 570. The graphics card had just been officially introduced by NVIDIA two days ago. It shares the same GF110 chip as NVIDIA’s current top-of the-line model, GTX 580. What are the differences between the two then? Have a look at the table below to learn about the technical specifications of NVIDIA’s current top-of the-line models.

Specification

	GTX 470	GTX 480	GTX 570	GTX 580
Codename	GF100	GF100	GF110	GF110
Fabrication Process	40nm	40 nm	40 nm	40 nm
Transistors	3 Billion	3 Billion	3 Billion	3 Billion
Die Size	320 mm²	320 mm²	320 mm2	320 mm²
Core Clock	607 MHz	700 MHz	732 Mhz	772 MHz
Stream Processors (CUDA Cores)	448	480	480	512
Shader Clock	1215 Mhz	1400 MHz	1464 Mhz	1544 MHz
Texture Units	56	60	60	64
ROPs	40	48	40	48
Memory Type	GDDR5	GDDR5	GDDR5	GDDR5
Memory Size	1280 MB	1536 MB	1280 MB	1536 MB
Memory Clock (effective)	3348 MHz	3696 MHz	3800 MHz	4008 MHz
Memory Interface	320-bit	384-bit	320-bit	384-bit
DirectX Version	11	11	11	11
Shader Model	5	5	5	5
Power Connectors	6-pin + 6-pin	8-pin + 6-pin	6-pin + 6-pin	8-pin + 6-pin
Min Recommended Power Supply	550 Watt	600 Watt	550 Watt	600 Watt
Bus Support	PCIe 2.1 x16	PCIe 2.1 x16	PCIe 2.1 x16	PCIe 2.1 x16
TDP	225 Watt	250 Watt	219 Watt	244 Watt
GPU Thermal Threshold	105 C	105 C	97 C	97 C

The GeForce GTX 570 is based on the GF110 chip, also used by its “bigger brother” the GTX 580. As such, the two share a similar set of specification. Of course, some cut-downs are applied to the GTX 570 to position it a step lower than the flagship GTX 580. We can see that the GTX 570 has fewer shader processors and ROPs, along with lower frequencies and memory interface numbers. Those differences are offset by more affordable price and lower power consumption. GTX 570 needs only two 6-pin connectors, as opposed to the GTX 580 that needs one 6-pin and one 8-pin connectors. GTX 570’s TDP, therefore, should be lower as well.

Now, let’s talk about the technical specification. By deactivating one Streaming Multiprocessor cluster on the GTX 570, NVIDIA reduced the number of shader processors to 480. That amount is similar GTX 480’s. When both the GTX 480 and GTX 570 are put side-by-side on paper, the former may look slightly more powerful due to higher amount of ROP (48, versus 40 on the GTX 570) and wider memory interface. However, GTX 570 has slightly higher core and memory frequencies, at 732 MHz and 3800 MHz respectively. So, our main question would be something like: “Can the GTX 570 compete against the GTX 470 or even the GTX 480 in terms of performance?” The answer is intriguing, but before moving on to that, let’s see how the graphics card looks like!