NVIDIA GTX 580 Review
Reviewed by: ccokeman
Reviewed on: November 9, 2010
Price: $499 - $589
Introduction
Fourteen months ago we got the first look at NVIDIA's latest architectural marvel codenamed "GF 100" at their GPU Technology Conference (GTC) held in sunny San Jose. What we obtained was a view of the next top-of-the-line graphics core from NVIDIA that offered up a promise of some serious firepower with its 512 CUDA cores, 16 Streaming Multiprocessors (SMs) and 1.5GB of GDDR5 memory. As an enthusiast, I couldn't wait to see what this card would deliver in terms of FPS in games and was anxious to see how my folding farm would profit from that kind of computing power!
The problem was that we had to wait and watch while AMD filled up their product stack and stores (and ultimately consumer's rigs) with DirectX 11 hardware. At CES this year, we were invited to a 'Deep Dive' on the Fermi architecture and got a glimpse of what would be coming to consumers when it was released. What we saw was a version that ended up being called the GTX 480 that was delivered with 480 CUDA cores and 15 Streaming Multiprocessors packed into the four Graphics Processing Clusters. A cut-down version of what was presented to us at the GTC. Seven months ago, the GTX 480 was released to the world and in reality, the card kicked ass! The problems were that it was a power hog and ran hot. So hot that one website decided to try and fry an egg on the heat sink (unsuccessfully I might add).
Today, we fast forward to the here and now. I have the latest revision of NVIDIA's Fermi architecture code named GF 110. What we have here is a GPU that has been re-engineered down to the transistor level and comes fully equipped with the full 512 CUDA cores and 16 SMs. The re-engineering has (according to NVIDIA), taken care of the power and heat problems. So let's take a look and find out just how good of a job they have done!
Closer Look:
Since we don't have any packaging to go through, we can just get right to the GTX 580. Measuring 10.5 inches long, the GTX 580 mirrors the GTX 480 in size. The first thing you notice is that the heat sink is not visible from outside the shroud. Considering there are still three billion transistors that need cooling, the cooling solution must be something special! From the front, the only difference between this reference card and the initial offerings will be the manufacturer's sticker affixed to the shroud. The back side does not show anything out of the ordinary. Under the hood is where all the action takes place and NVIDIA has redesigned the core down to the transistors for a cooler running and more efficient product. With the GTX 580 we get the full 512 CUDA cores and increased clock speeds for an added performance boost. This card is meant to be used in motherboards with a 16x PCIe slot.
Connectivity comes in the form of two Dual link DVI ports and a Mini HDMI port. This port type has supported bit streaming for Dolby True HD and DTS HD Master Audio over HDMI since the 260 driver release. Meaning that if you choose to use this card in a multipurpose rig, you can take advantage of this ability for HD sound for a full 3D Blu-ray movie experience. The back end of this card has a small vent and detent to pull fresh air into the fan assembly to keep the card cool when running with two or more cards in an SLI setup (such as when they are installed side by side).
The power connections needed for use with the GTX 580 are one eight pin and one six pin PCIe power connector. The recommended power supply for use with the GTX 580 is 600 watts. The GTX 580 has a TDP of 244 watts or 56 watts less than the GTX 480. Multi-GPU setups are a way to increase your gaming performance and the GTX 580 supports up to a three card Tri-SLI setup. The 580 has the same two bridge connections on the top spine as well as several ventilation slots for removing heat from the shroud. By using two of these cards in an SLI setup, you can take advantage of NVIDIA's Surround technology using three monitors at resolutions up to 7680 x 1600. If that's not enough, you can add one more element from the NVIDIA ecosystem and hook up a 3D Vision system to the surround setup as long as you have 120hz monitors for an out-of-screen experience.
One of the big complaints with the GTX 480 was that it was power hungry and therefore ran hot. You needed to dissipate that energy somewhere. That meant you needed very good airflow and an efficient heat sink design. With the heat pipe based solution on the GTX 480, you had to ramp up the fan speed to push more air. This resulted in more noise so we ended up with complaint number two. The cooling solution on the GTX 580 is quite a bit different from the heatpipe solution used on the GTX 480. NVIDIA fixed both problems on the GTX 580 with the use of a liquid-based "Vapor Chamber" cooling solution. This solution uses a large liquid-filled chamber that functions much like a heatpipe but is of a more efficient design. To cool the power regulation circuit there is a large aluminum plate that serves to both stiffen the card and draw heat from these components. Below is a picture that illustrates how the Vapor Chamber works to remove heat from the GTX 580's core.
Last but not least is a look at the GTX 580's GF 110 core. This core is the fully loaded version of Fermi we were expecting when this architecture was originally introduced. Inside is a complete redesign, engineered to reduce the power consumption and heat load all the way down to the transistors. The GF 110 core is equipped with four Graphics Processing Clusters, 16 Streaming multiprocessors, 512 CUDA cores, 16 Polymorph engines, 64 texture units, 48 ROP units with higher clock speeds of 772Mhz on the fixed function units and 1544Mhz on the CUDA cores. The GTX 580 uses the same 1536MB of GDDR5 memory running through a 384bit bus with 768k of shared L2 cache. The 12 memory ICs used by NVIDIA are manufactured by Samsung.
The GTX 580 looks to have the firepower to meet the needs of the enthusiast and hardcore gamer. Based on specifications alone, it's a beast and is what we have been waiting for from NVIDIA. Lets see how it performs.
Specifications:
|
|
Graphics Card
|
GTX 580
|
|
Processing Units
|
Graphics Processing Clusters
|
4 |
|
Streaming Multiprocessors
|
16 | |
|
CUDA Cores
|
512 | |
|
Texture Units
|
64 | |
|
ROP Units
|
48
|
|
|
Clock Speeds
|
Graphics Clock (Fixed Function Units)
|
772 MHz
|
|
Processor Clock (CUDA Cores)
|
1544 MHz
|
|
|
Memory Clock (Clock rate / Data rate)
|
2008 MHz
|
|
|
Memory
|
L2 Cache Size
|
768KB
|
|
Total Video Memory
|
1536 MB GDDR5
|
|
|
Memory Interface
|
384-bit
|
|
|
Total Memory Bandwidth
|
192.4 GB/s
|
|
|
Fillrate
|
Texture Filtering Rate (Bilinear)
|
49.4 GigaTexels/sec
|
|
Physical & Thermal
|
Fabrication Process
|
40 nm
|
|
Transistor Count
|
3 Billion
|
|
|
Connectors
|
2 x Dual-Link DVI-I
1 x Mini HDMI
|
|
|
Form Factor
|
Dual Slot
|
|
|
Power Connectors
|
1x6 pin and 1x 8 pin
|
|
|
Recommended Power Supply
|
600 Watts
|
|
|
Thermal Design Power (TDP)1
|
244 Watts
|
|
|
Thermal Threshold
|
97 C
|
Features:
- Microsoft® DirectX® 11 Support: DirectX 11 GPU with Shader Model 5.0 support designed for ultra high performance in the new API’s key graphics feature, GPU-accelerated tessellation.
- NVIDIA® 3D Vision™ Surround Ready*: Expand your games across three displays in full stereoscopic 3D for the ultimate “inside the game” experience with the power of NVIDIA 3D Vision and SLI technologies. NVIDIA® Surround™ also supports triple screen gaming with non-stereo displays.
- Interactive Ray Tracing: By tracing the path of light through a 3D scene, ray tracing uses the power of the GPU to create spectacular, photo-realistic visuals. Get a glimpse into the future of gaming with ray tracing.
- 3-way NVIDIA SLI® Technology*: Industry leading 3-way NVIDIA SLI technology offers amazing performance scaling by implementing 3-way AFR (Alternate Frame Rendering) for the world’s premier gaming solution under Windows 7 with solid, state-of-the-art drivers.
- NVIDIA PhysX® Technology: Full support for NVIDIA PhysX technology, enabling a totally new class of physical gaming interaction for a more dynamic and realistic experience with GeForce.
- NVIDIA CUDA™ Technology: CUDA technology unlocks the power of the GPU’s processor cores to accelerate the most demanding tasks such as video transcoding, physics simulation, ray tracing, and more, delivering incredible performance improvements over traditional CPUs.
- 32x Anti-aliasing Technology: Lightning fast, high-quality anti-aliasing at up to 32x sample rates obliterates jagged edges.
- NVIDIA® PureVideo® HD Technology: The combination of high-definition video decode acceleration and post-processing that delivers unprecedented picture clarity, smooth video, accurate color, and precise image scaling for movies and video.
- PCI Express 2.0 Support: Designed for the new PCI Express 2.0 bus architecture offering the highest data transfer speeds for the most bandwidth-hungry games and 3D applications, while maintaining backwards compatibility with existing
All information courtesy of NVIDIA
Testing:
Testing of NVIDIA's GF 110 based GTX 580 Fermi derivative will consist of running the card through the OverclockersClub.com suite of games and synthetic benchmarks. This will test the performance against many popular competitors. Comparisons will be made to cards of equal and greater capabilities to show where the performance of these cards stand. The games used are some of today's newest and most popular titles to give you an idea on how the cards perform relative to each other. The system specifications will remain the same throughout the testing. No adjustment will be made to the respective control panels during the testing with the exception of the 3DMark Vantage testing where PhysX will be disabled in the NVIDIA control panel. I will test the card at stock speeds, then overclocked in order to see how much additional performance is available and to determine if it can run with other single and mulit-GPU cards on the market. The drivers used in this test will be the 10.10 Catalyst drivers for the old ATI lineup, the latest launch driver for the HD 68XX series and 260.89 Forceware drivers from NVIDIA for the GTX 480, 470, 465 and GTX 460 and 450. The GTX 580 will use driver 262.99. Tests will be conducted at both stock and overclocked settings to gauge performance when an increase in clock speed is applied.
- Processor: Intel Core I7 920 200x18 3.6Ghz
- Cooling: Noctua NH-U12P SE 1366
- Motherboard: MSI X58 Eclipse SLI
- Memory: Mushkin 996805 Redline PC312800 6-8-6-24 1600MHz
- Video Card(s): NVIDIA GTX 580
- Power Supply: Mushkin 1000 watt Joule Modular power supply
- Hard Drive: 1 x Seagate 1TB SATA
- Optical Drive: LG DVD-RW
- OS: Windows 7 Professional 64-bit
- Case: Cooler Master HAF 932
Comparison Video Cards:
- PowerColor HD5770
- PowerColor HD5770 CrossFire
- Sapphire HD 6870
- XFX HD 6870
- XFX HD6850
- PowerColor HD 6870
- NVIDIA GTS 450
- EVGA GTX 460 FTW Edition
- ASUS ENGTX465
- NVIDIA GTX 480
- Galaxy GTX 470 GC
- BFG GTS 250 1GB OC
- Sapphire HD 5970 4GB Toxic
- Sapphire HD 5970 2GB
- Sapphire HD 5870
- Sapphire HD 5850 Toxic 2GB
- Sapphire HD 5770 VaporX
- Sapphire HD 5750 VaporX
Overclocking:
- NVIDIA GTX 580 827/1654/2110MHz
Overclocking the GTX 580 is pretty much the same as any of the Fermi derivatives with the fixed function units tied to the shader domain so that you cannot unlink the two clock speeds to gain more clock speed on the core. I was hoping that I would be able to increase the clock speeds for added performance up to the levels seen on the GTX 460 and even the GTS 450 cards but that really was a short lived dream. The base clock speeds on this card are 772/1544/1002Mhz. By using a few different utilities, I was able to ramp up the clock speeds to 827/1654/2110Mhz. Not much, but it is still something and is a bit better than what I was able to get out of the GTX 480. In a press briefing I was told that the voltage monitoring hardware mounted on board would not preclude the end user from using voltage adjustments to overclock the card. However, it will preclude you from using utilities such as Furmark or OCCT as a way to load the GPU. This was tested and proved true in my Kombustor testing. While the GPU would ramp up to 99% usage, the temperatures and power consumption numbers were much lower than they should have been. By running Unigine's Heaven benchmark or Crysis Warhead, the power consumption and temperatures returned to the expected levels. At this time (as is almost always the case with a new launch) there are no utilities out to allow for voltage adjustments to the core so that type of testing will need to come at a later date. When I push the clocks speeds, the first thing I do is turn up the fan speed to the maximum level possible through available utilities such as MSI's Afterburner. In this case, the maximum fan speed was listed as 85%. NVIDIA was specific in their briefing that the noise signature of the GTX 580 was examined as a user experience item. By experience I mean the gaming experience. You don't want the noise equivalent of a vacuum cleaner running within a few feet of your ears. The noise signature is still there but is nowhere near what it has been. The fan does not have the high pitched whine associated with most blower style fans and it sounds more like air rushing through the shroud. So at this point, the card has the ability to clock a little better than my reference GTX 480 and is quieter. Both are wins in my book.
Maximum Clock Speeds:
MSI's Kombuster utility was used to test stability and to put a constant load on the GPU for the purposes of testing maximum power draw and temperatures. The stability test was used to find a range of settings that are stable. The stable condition was determined through a 15 minute run at 1920 x 1200 8xAA. The reported clock speeds are those that proved stable over a 15 minute test at 1920 x 1200, 8x AA and the run through the benchmarks suite.
- Gaming Tests:
- Far Cry 2
- Metro 2033
- Crysis Warhead
- Call of Duty Modern Warfare 2
- Just Cause 2
- Unigine Heaven Benchmark 2.0
- Batman: Arkham Asylum
- Resident Evil 5
- 3DMark 06 Professional
- 3DMark Vantage
- Usage:
- Temperature
- Power Consumption
The maximum clock speed graphs above show the overclocking potential of all the cards tested.
Testing:
Featuring a new game engine named Dunia, this game looks to be another one to stress your video card. Built especially for Far Cry 2, this engine allows for real time effects and damage. This next generation first person shooter comes to us from Ubisoft surprisingly - not from Crytek. The game is set in a war-torn region of Africa where there is a non-existent central government and the chaos that surrounds this type of social environment. If you have seen the movie Blood Diamond, you know the setting. Ubisoft puts the main storyline of the game into focus with these statements: "Caught between two rival factions in war-torn Africa, you are sent to take out "The Jackal," a mysterious character who has rekindled the conflict between the warlords, jeopardizing thousands of lives. In order to fulfill your mission you will have to play the factions against each other, identify and exploit their weaknesses and neutralize their superior numbers and firepower with surprise, subversion, cunning and, of course, brute force." In this version of the game, you don't have the beautiful water, but instead the beauty and harshness of the African continent to contend with. Most games give you a set area that can be played through, while Ubisoft has given the gamer the equivalent of 50 square kilometers of the vast African continent to explore while in pursuit of your goals. The settings used are just a few steps below the maximum in-game settings and offer a good blend of performance vs. visual quality.
Settings
- 4x AA
- 16xAF
- Ranch Small
- Global Setting = High
Higher = Better
The GTX 580 is the fastest single GPU card in the Far Cry 2 comparison. It even beats the 2GB HD 5970 in three out of the four stock tests.
Testing:
Part first person shooter, part survival horror, Metro 2033 is based on the novel of the same name, written by Russian author Dmitry Glukhovsky. You play as Artyom in a post-apocalyptic Moscow, where you'll spend most of your time traversing the metro system, with occasional trips to the surface. Despite the dark atmosphere and bleak future for mankind, the visuals are anything but bleak. Powered by the 4A Engine, with support for DirectX 11, NVIDIA Physx and NVIDIA 3D Vision, the tunnels are extremely varied - in your travels, you'll come across human outposts, bandit settlements, and even half-eaten corpses. Ensuring you feel all the tension, there is no map and no health meter. Get lost without enough gas mask filters and adrenaline shots and you may soon wind up as one of those half-eaten corpses - chewed up by some horrifying manner of irradiated beast that hides in the shadows just waiting for some hapless soul to wander by.
Settings:
- DX 11
- AAA
- 16xAF
- Global settings = High
- Physx = off
Higher = Better
Metro 2033 is a demanding DX 11 game when you turn up the eye candy. The GTX 580 responds by outperforming the HD 5970 2G and even the 4GB Toxic version at 2560x1600. All things considered, the GTX 580 is a beast!
Testing:
Crysis Warhead is a standalone expansion pack situated in time with the story line of the original Crysis. As Sergeant "Psycho" Sykes, you have a secret mission to accomplish on the far side of the island. Along the way there are EMP blasts and aliens to contend with as you hunt down the KPA chief. This game uses an enhanced version of the CryEngine 2.
Settings
- Frost Map
- 4XAA
- 16X AF
- Gamer
- DX 10
Higher = Better
In this game, the Multi GPU HD 5970 has an advantage over the GTX 580. The performance advantage over the GTX 480 is 12.5% or four FPS at 2560 x 1600.
Testing:
Call of Duty: Modern Warfare 2 is the latest iteration of the venerable first person shooter series, Call of Duty. Despite its long, successful pedigree, the game is not without substantial criticism and controversy, especially on the PC. Aside from the extremely short campaign and lack of innovation, the PC version's reception was also marred by its lack of support for user-run dedicated servers, which means no user-created maps, no mods and no customized game modes. You're also limited to 18-player matches instead of the 64-player matches that were possible in Call of Duty 4: Modern Warfare. Despite all this, the game has been well received and the in-house IW 4.0 engine renders the maps in gorgeous detail, making it a perfect candidate for OCC benchmarking. You start off the single player missions playing as Private Allen and jump right into a serious firefight. This is the point where testing will begin. Testing will be done using actual game play with FPS measured by Fraps.
Settings
- In Game settings to Maximum
- 4xAA
- 16xAF
- Fraps measured Gameplay
Higher = Better
In Call of Duty Modern Warfare 2, the GTX 580 is able to handle pretty much any card save the HD 5970. The advantage over the HD 5870 and GTX 480 are quite substantial.
Testing:
Just Cause 2 is a third-person shooter that takes place on the fictional island of Panau in Southeast Asia. In this sequel to 2006's Just Cause, you return as Agent Rico Rodriguez to overthrow an evil dictator and confront your former boss. When you don't feel like following the main story line, you're free to roam the island, pulling off crazy stunts and causing massive destruction in your wake, all beautifully rendered by the Avalanche Engine 2.0. In the end, that's what the game basically boils down to. Crazy stunts and blowing things up. In fact, blowing things up and wreaking havoc is actually necessary to unlock new missions and items.
Settings
- 4xAA
- 16xAF
- Global Settings = High
- Water Detail = Medium
- Dark Tower Time Demo
Higher = Better
Again, the GTX 580 is stronger than any single GPU card in this comparison by a wide margin. At 1920 x 1200, the performance differential over the fastest single GPU card from AMD is 22 FPS.
Testing:
Unigine Heaven Benchmark 2.0 is a DirectX 11 GPU benchmark based on the Unigine engine. What sets the Heaven Benchmark apart is the addition of hardware tessellation, available in three modes. Moderate, Normal and Extreme. Although tessellation requires a video card with DirectX 11 support and Windows Vista/7, the Heaven Benchmark also supports DirectX 9, DirectX 10 and OpenGL. Visually, it features beautiful floating islands that contain a tiny village and extremely detailed architecture.
Settings
- 4xAA
- 16xAF
- Shaders = High
- Tesselation = Normal
- Measurement = FPS
Higher = Better
Unigine 2.1 is a benchmark that uses varying degrees of tessellation. This makes it an ideal benchmark to test DX 11 video cards and their tessellation performance. While the GTX 480 was strong and outperformed all the single GPU cards from AMD, the GTX 580 takes that performance to another level.
Testing:
Batman: Arkham Asylum is a new game that brings together two bitter foes, the Joker and Batman. The Joker has taken over Arkham Asylum, Gotham's home for the criminally insane. Your task is to rein in the Joker and restore order. This game makes use of PhysX technology to create a rich environment for you to ply your trade.
Settings
- 4xAA in CP
- 16xAF in CP
- Global settings = High
- Ambient Occlusion = On
Higher = Better
In the Batman testing, the only card faster than the GTX 580 is the HD 5970 4GB Toxic from Sapphire. The GTX 580 is faster than or equal to it in two out of the eight tests.
Testing:
Resident Evil 5 is the sequel to one of the best selling video games of all time. You play the game as Chris Redfield a survivor of the events at Raccoon City who now works for the BSAA. Sent to Africa to find the genesis of the latest Bio Organic agents, you meet up with another BSAA operative and work together to solve the problem. The game offers incredible 3D effects and a co-op gaming style.
Settings
- 4xAA
- 16xAF in CP
- Global settings = High
Higher = Better
The performance trends of the GTX 580 show it again trading scores with the HD 5970 cards from AMD. At 1920 x 1200 and above, the GTX 580 is faster in this game than the 2GB HD 5970.
Testing:
3DMark06 is one of the benchmarks that always comes up when a bragging contest has begun. 3DMark06 presents a severe test for many of today's hardware components. Let's see how this setup fares. The settings we will use are listed below.
Settings
- Default test settings
- Resolutions 1280-1024-2560 x 1600
- SM2.0 Graphics Tests: GT1- Return to Proxycon, GT2- Firefly Forest
- CPU Tests: CPU1- Red Valley, CPU2- Red Valley
- HDR/SM3.0 Graphics Tests: HDR1- Canyon Flight, HDR2- Deep Freeze
Higher = Better
In the FuturMark 3DMark06 testing, the GTX 580 has no competition from a single GPU video card. The HD 5970 is still the only thing capable of beating it in this test.
Testing:
Featuring all-new game tests, this benchmark is for use with Vista based systems. "There are two all-new CPU tests that have been designed around a new 'Physics and Artificial Intelligence-related computation.' CPU test two offers support for physics related hardware." There are four preset levels that correspond to specific resolutions. 'Entry' is 1024 x 768 progressing to 'Extreme' at 1920 x 1200. Of course, each preset can be modified to arrange any number of user designed testing. For our testing, I will use the four presets at all the default settings.
Settings
- Settings (each preset): Default Testing
- Entry: 1024 x 768
- Performance: 1280 x 1024
- High: 1680 x 1050
- Extreme: 1920 x 1200
Higher = Better
In the Vantage testing, the GTX 580 has the highest Entry score to-date here at OCC, at almost 50,000 points. When the resolution scales upwards, the HD 5970 cards have the edge in performance. When compared to the GTX 480, the performance difference is huge.
Testing:
Temperature testing will be accomplished by loading the video card to 100% using MSI Kombuster which is paired with MSI's afterburner overclocking utility for temperature monitoring. I will be using the stability test set to a resolution of 1920 x 1200 using 8xAA. I will use a 15 minute time frame to run the test ensuring that the maximum thermal threshold is reached. The fan speed will be left in the control of the driver package and video card's BIOS for the first test, with the fan moved to 100% to see the best possible cooling scenario for overclocking. The idle test will be a 20 minute cool down with the fan speeds left on automatic.
Settings
- Monitoring with MSI Afterburner
- GPU Stability test in MSI Kombuster
- 15 minute load duration
- 15 minute idle duration
- Temperature measured in degrees Celsius
Lower = Better
The temperatures delivered by the GTX 580 are an indication of the refinement of the Fermi architecture that has taken place since its introduction seven months ago. Both idle and load temperatures are much better. But to put that into context, the testing method used to test the GTX 580 was slightly different as well. How so? In the drivers as well as in the power monitoring hardware, NVIDIA has set this card up so that you cannot run a traditional stress test such as Furmark, Kombustor or OCCT. When one of these applications are used, the power usage is throttled and you get much lower temperatures and power consumption than you normally see in this kind of test. Run a stressful game and the temperatures and power consumption reach levels you would expect from this card. This is kind of disappointing as I have not killed a video card yet using these utilities but our stress testing matrix will need to be modified.
Testing:
Power Consumption of the system will be measured in both idle states and loaded states and will take into account the peak voltage of the system with each video card installed. I will use MSI Kombuster to load the GPU for a 15 minute test and use the peak load of the system as my result for the maximum load. The idle results will be measured after 15 minutes of inactivity on the system.
Settings
- MSI Kombuster Stability test
- 1920 x 1200 Resolution
- 8XAA
- 15 minute load test
- 15 minute idle test
- Measurement is in Watts
Lower = Better
NVIDIA has listened to all the criticisms they received for building a video card that some said needed nuclear reactors to run (GTX 480). Throughout that storm, they have worked hard and put together a refined chip with better power consumption characteristics. At this point delivering power consumption levels close to that of an overclocked HD 5870 under load. At stock speeds, the picture is not quite as rosy but still runs using 60 watts less than the GTX 480. This set of numbers is lower than it should be by comparison but it's what we have with the power monitoring hardware and driver recognition of stress testing apps like Kombuster, Furmark and OCCT.
Configuration:
There is no doubt that NVIDIA was late to the DX 11 party, allowing AMD to stack the deck against them. Here they are seven months later with their competition getting ready to drop yet another high end DX 11 card into the market. NVIDIA has struck first and set the benchmark to beat. They have done this in a convincing fashion with the GF 110 based GTX 580. This is the card we hoped the GTX 480 would have been when it was introduced. Through refinements to the Fermi architecture, NVIDIA has finally upped the ante in the video card wars. The architectural improvements come in the form of full speed FP-16 texture filtering and new tile formats for improved Z cull efficiency. The hardware was redesigned down to the transistor level for improved efficiency that allowed for increased clock speeds with lower power consumption. On top of that, the GTX 580 gets the full compliment of 512 CUDA cores (up from 480), an additional Streaming Multiprocessor and an extra Polymorph Engine. During the game testing, it was evident that there is nothing but a high-end dual GPU setup that can compete with it in heads-up fashion. In many of the games, the GTX 580 was even able to deliver performance higher than the reference based HD 5970. That is some serious firepower!
All this graphics firepower allows you to increase the eye candy in the games you play and to take advantage of NVIDIA's arsenal of technologies. Technologies such as: PhysX in titles like Mafia II, Batman Arkham Asylum and Metro 2033; the strength of its tessellation engines in games such as in Hawx 2; and then its parallel processing strengths in Just Cause 2 where CUDA is used to simulate realistic water. It does not stop there. NVIDIA's 3DVision system and Surround technology for a 3D Surround experience needs to be seen to be believed. The GTX 580, much like the GTX 480, is Tri-SLI capable for an immense increase in graphics horsepower. This should allow for excellent frame rates in resolutions up to 7680 x 1600. If that much power is not enough there is always the ability to overclock the GTX 580. I was concerned that the voltage monitoring and driver package would let the performance degrade when you start overclocking due to higher than expected power demands when increasing the voltage to the core. In our press briefing, I was told that would not be the case. So, as soon as utilities are available with the ability to increase the voltage on the core, we will take another stab at overclocking. When it came to this overclocking exercise, the amount of headroom left on this card was a meager 56Mhz on the core and 106Mhz on the memory. Not really what I was expecting after seeing the improvements in the overclocking ability of the GF 106 and GF 108 silicon but it's still something. Overclocking did allow for increased performance when push came to shove but at this level, it's just icing on the cake.
By using less power, NVIDIA went back to the drawing board for the cooling solution and decided on a more efficient Vapor Chamber heat sink assembly much like the one used by AMD on the HD 5970. This allowed the card to stay relatively cool under load while still keeping reasonable noise signatures. Under load, I saw 81 degrees Celsius when the card controlled the fan at stock speeds and a cool 59 degrees Celsius when overclocked and the fan speed was set to the maximum 85% ratio. This is a big improvement over the GTX 480. Pricing on a card of this caliber is going to run on the high end of the scale but when you buy a Ferrari you know it's going to cost you a few pennies. What you get with the GTX 580 is the 'Bionic Man' version of the GTX 480 that has been rebuilt to be stronger, faster, quieter and cooler running - that can play any game out there and defend its territory with impunity. With NVIDIA's second winner at the top end of the single GPU market, I can't wait to see what happens in the next few weeks. It's gonna be interesting.
Pros:
- Fastest Single GPU card on the planet
- High performance
- Lower Noise
- Lower Heat
- Vapor Chamber Cooling
- PhysX
- Tri-SLI capabilities
- Surround
- 3D Surround
Cons:
- Pricing