ENGTX580 DCUII Review
The Nvidia GTX580 was launched some months ago but for a variety of reasons we never got round to reviewing it. Today we're rectifying that omission with a review of the ASUS ENGTX580 DCU2. This is not a reference design (far from it) and as you'll see later, is an innovative 3-slot solution. That's not a typo - this card does indeed take up 3 slots! It's not so long ago that we grudgingly accepted the standardisation of 2-slot graphics cards so why would anyone purchase one that took up 3 slots? The answer resides with the market segments that the cards unique features benefit, something we will explore in more depth later.
It's been a few years since we reviewed an Asus graphics card (we used to be regular visitors to their offices in Milton Keynes) and are glad we have a chance to develop relations with the new generation of their PR staff.
Here are the specifications of this card:
There are a whole host of features such as Voltage Tweak, Super Alloy Power and not least DirectCU II which are described at the ASUS homepage.
The GF100 was the first Fermi chip and is now followed by the GF110 in the shape of the GTX580 and more recently, the GTX570. As the first member of the 500 series with more released in the last few weeks, this is the second outing for Nvidia's 40nm manufacturing process.
Specifications are impressive and the GTX580 is essentially a faster version of the GTX480 but with everything enabled. The 10% faster clock speeds and FP16 texture filtering capabilities as well as 64-bit floating point precision and better cooling are all significant improvements.
The price point is still relatively high but expect that to fall now that the dual-GPU GTX590 and Radeon HD6990 have been launched for not much more than the price of the GTX580.
The GF110 is much closer in design to the GF100 than to the GF104.
Fundamentally GF110 is the same architecture as GF100, especially when it comes to compute. 512 CUDA Cores are divided up among 4 GPCs, and in turn each GPC contains 1 raster engine and 4 SMs. At the SM level each SM contains 32 CUDA cores, 16 load/store units, 4 special function units, 4 texture units, 2 warp schedulers with 1 dispatch unit each, 1 Polymorph unit (containing NVIDIA’s tessellator) and then the 48KB+16KB L1 cache, registers, and other glue that brought an SM together. At this level NVIDIA relies on TLP to keep a GF110 SM occupied with work. Attached to this are the ROPs and L2 cache, with 768KB of L2 cache serving as the guardian between the SMs and the 6 64bit memory controllers. Ultimately GF110’s compute performance per clock remains unchanged from GF100 – at least if we had a GF100 part with all of its SMs enabled.
Another area of note is power protection. Thanks to the likes of FurMark (which we will see later on) and OCCT AMD cards now monitor the temperature of the VRMs and will downclock if it gets too hot. In practice this doesn't happen very often except during overclocking. Nvidia have gone even further and have sensors monitoring power usage and will slow the card down to keep it within it specified envelope.
Moving on to the ASUS card itself lets start with the box.
Branding is very important for a manufacturer's product positioning, both against the offerings of competitors and to enhance product differentiation. The GTX580 is quite heavy due to the materials used and the box weighs more than out Radeon 6990 box. As for the 20% cooler claim, we will test this later in the review.
It may seem strange to evaluate packaging (some people throw it away or tear it open) but it says a lot about brand positioning. Imparting as much information as possible about the things that are unique about the product will go a long way towards persuading a consumer to buy while browsing at a large store. Few things depress shoppers more than shelf after shelf of plain OEM boxes.
The box is awash with information on all sides and it's nice to see detailed specifications not being ommitted in favour of all-round marketing material. Shoppers at big stores will no doubt appreciate this (particularly those that have ever tried asking a technical question at PC World).
Presentation is also a consideration and here ASUS have a nice foam tray that fits over the card in the box and holds the software and driver CDs in a nice box along with an installation manual. Included is a 2x 6-pin to 1x 8-pin convertor for those PSUs with many 6-pin and no 8-pin plugs and an SLI ribbon cable. Exactly how two of these could be used in SLI mode is puzzling due to the 3-slot design, but then, the target market is not SLI gamers.
Immediately, we can see that the design is very different from the reference one with twin fans and a large heatpipe/fin network. Also visible is a plastic cover over the PCI-E connector - something we have not seen before in a graphics card.
The back view shows the solid PCB construction and covers over the PCI-E, SLI and DVI connectors. Holes in the PCB are a moot point as the shroud does not extend all the way to the back exhaust, and indeed, is open at both ends
The design seems to be open from all sides. That means plenty of airflow but also some potential issues to those of us (like our test rig) that use liquid cooling and rely on a cool case to help air going out through the radiator to cool down the CPU.
The back shows both DVI connectors and mini port and HDMI connectors. Ventilation stretches to one and a half slots and would be more impressive were it not for the open nature of the shroud not being able to take advantage of this and push air out the back rather than also into the case.
The inside edge view shows how open the shroud is and also clarifies understanding of how the heatpipes work in this design.
We have two main test systems at present and recently upgraded our Intel one to an i7-2600K we received from Intel for the Sandy Bridge launch in January. The AMD system has the most powerful processor currently available in the AMD range. All cards are tested at their default speed but in the case of the Radeon HD6990 we are able to get it to run at full 2x 6970 speeds by flicking a switch on the card. Therefore we will show results as HD6990 and HD6990 OC to distinguish between them.
1.5GB is an odd memory configuration to use and only time will tell if 2GB should have been the norm. We used GPU-Z to measure temperatures and gather various bits of information.
Firstly we will start with DX10 testing (the range of DX11 games is still limited at present although we benchmark two of them in the next section.
3DMark Vantage is a full DirectX 10 compliant synthetic benchmark. It tests out all of the DirectX 10 abilities of each card to give the user an idea of where they rank with their graphic card in a collective pool of many users all over the internet.
The testing was done with the default performance setting for all of the cards to give an idea where each one stands in the GPU line up.
The demanding test punishes all comers with the HD5850 at a crawl and the GTX580 putting a good show against the dual-GPU ATI Radeon HD6990
Setting aside the middle bars (the CPU performance is fairly level as the only variable during testing is the graphics card) it can be seen that the performance of the high end cards scales with their product positioning. The Radeon HD 5850 which has served us so well for over a year is still capable of putting up a good fight but is now end of line and is about the same price as the better performing Radeon HD 6870. The Sapphire Radeon HD 6950, currently priced about 20% higher than the two others shows a nice boost in performance and we'll see how that translates into real-world benchmarks. Both the GTX580 and the HD6990 manage to reach around the 30000 mark which is no easy thing.
Far Cry 2 is still a firm favourite and was taxing on any system. It's now broken by the HD6990 and is CPU limited at all resolutions. Considering that our i7-2600K Sandy Bridge processor is the most advanced available, that is quite an achievement. On the other hand the GTX580 performs well all the way up to full HD resolution.
HAWX seems to be easy for most systems to handle with playable rates from all contendors at all resolutions. We'll have to see how that changes when we move to HAWX 2. The GTX580 is not far short of the Radeon HD6990.
The current darling of horror games provides lots of eye candy and furious action. As with Far Cry 2 performance is very good but only the 6XXX series cards can provide triple digit frame rates at all resolutions. The ASUS ENGTX580 is just as fast as the more expensive HD6990.
We're in the process of changing our benchmarks to more demanding ones and appreciate any feedback suggestions.
One of the key features of the DCU2 range is better noise management. We cannot give exact decibel levels due to lack of sufficiently sensitive equipment but can confirm that the ASUS ENGTX580 DCU2 is very quiet. Thanks to the large fans spinning at lower than normal speeds the fans are inaudible at idle performance levels unless we turn off all the fans in our test rig (we use water cooling so the CPU is safe). Even at performance levels we can only barely make out the sound of the graphics card above that of the system fans (running at slow speed - cannot make out the graphics card at all at medium and high speeds).
Power consumption is a big issue these days with GPUs sometimes consuming 2 or 3 times as much as a CPU. Performance returns never seem to scale anything like power increases and heavy users can find that the cost of increased electricity mounts up over time. The idle loads shown are all perfectly acceptable but under load (running Resident Evil 5 fixed benchmark) the HD6990 jumps to around 400W while the GTX580 is closer to 300W. Using FurMark the HD6990 overclocked (cores running at full HD6970 speed) uses 614W of power versus 485W at normal setting. That's a 130W increase for about a 7% increase in clock speed.
Temperatures determine how long a component will last and can affect other components in a case. The HD6990/HD6990OC temperatures are 44/48 idle and 86/93 in FurMark despite an extremely well ventilated case. The ASUS ENGTX580 DCU2 literature claims that it can make components last over twice as long. We can see that this is down to good design with an idle temperature of only 38 and a FurMark load of 76 - 10 degrees lower than the default HD6990.
Because of the longevity of our standard tests we have introduced Alien vs Predator and Stalker: Call Of Pripyat as two new benchmarks. We wanted to add Crysis2 as well but DX11 issues with that title mean we will have to wait until a patch is released.
Aliens VS Predator
As with many of the already released DirectX 11 benchmarks, the Aliens vs. Predator DirectX 11 benchmark leverages DirectX 11 hardware to provide an immersive game play experience through the use of DirectX 11 Tessellation and DirectX 11 Advanced Shadow features. In Aliens vs. Predator, DirectX 11 Geometry Tessellation is applied in an effective manner to enhance and more accurately depict HR Giger’s famous Alien design. Through the use of a variety of adaptive schemes, applying tessellation when and where it is necessary, the perfect blend of performance and visual fidelity is achieved with at most a 4% change in performance.
All four cards provide playable framerates with the GTX580 being above 60fps at 1920x1080.
Stalker: Call of Pripyat
This is a special version of S.T.A.L.K.E.R.: Call of Pripyat Having a PC tested with this benchmark will enable the gathering of detailed information about its performance on various graphical modes and finds out how well the system is balanced for gaming and S.T.A.L.K.E.R.: Call of Pripyat in particular.
This benchmark is based on one of the ingame locations – Pripyat . The testing process includes four stages, those utilizing various weather and time of the day settings (day, night, rain, bright sun). In order to test the system thoroughly the benchmark is provided with a number of presets and options including different versions of DirectX (9.0, 10.0, 11), screen resolutions, antialiasing etc.
We use the Ultra (highest details) setting and DX11 mode. The only switching we did was between 1680x1050 and 1920x1080 resolutions and to switch the DX11 Hardware Tesselation and Contact Hardening Shadows.
While the HD6990 is a clear winner at maximum frame rates, the GTX580 is just as good at average frame rates and often has much higher minimum frame rates. This translates into a smoother performance than its more expensive rival.
Performance hit for going up to 1920x1080 is negligible for all except the Radeon HD5850.
Enabling HW Tesselation and Contact Hardening Shadows provides a huge boost in visual quality. This alone is worth the small performance hit (good programming from the developers - compare that with the hit from doing the same in Dragon Age 2).
Now with everything maxed out (we need a better test monitor to achieve 2560x1600) we are still getting perfectly playable (usually over 100fps) frame rates from the ASUS ENGTX580 DCU2.
Having the latest graphics card is not just about speed but also other benefits. One of these is hardware acceleration in video playback and this is illustrated with Adobe's Flash player with support for hardware acceleration.
The difference is startling and shows how important this feature can be if widely implemented. Let's hope that people keep pestering the relevant software developers until they all have this level of support by default.
To see the benefits of DX11 that are unique to the DX11 series cards we need to dig further. It’s tricky to test to see how much of a speed difference DX11 makes as there are few reliable DX11 benchmarks but as a showcase of DX11 quality the situation is much better. We used Unigine’s Heaven Benchmark and can only convey the quality to our readers via YouTube.
The improvement in quality is apparent for all to see.
This next video shows the effects of Hardware Tessellation. The technical white paper on this feature is enough to send the most avid technophile to sleep but there's no denying the benefits of the end result.
If you have a DirectX 11 card then you can download the demo for yourself at http://unigine.com/download/
The ASUS ENGTX580 DCU2 is part of a unique series. Using a 3-slot cooler ASUS are able to overcome limitations in reference designs that plague users with high noise pollution and temperatures. The space available for large heatpipte based cooling and big fans allows for a sizeable (at least 10 degrees Celsius) reduction in temperature and profound reduction in irritating noise.
Size is another plus point and at a full 1" less than the Radeon HD6990 in length we were able to fit it into our Antec 902 case without any problems. The same cannot be said of the HD6990 which required us to have one of the front bays unlocked and protruding out of the front of our case during testing. The weight of the ENGTX580 DCU2 is quite hefty and this is largely down to the extra large cooling system with expanded heatpipes and numerous large fins but in practice, once the card is installed, it is secure and cannot be made to wobble by pushing the corner.
The reduction in noise is quite profound and is no louder than our Radeon HD5850 (a card renowned for its accoustics under load) - even while running FurMark. The Radeon HD6990 by comparison sounded like a dustbuster during FurMark. While our ears are particularly sensitive, and noise is accentuated because we use water cooling for testing rather than loud CPU fans, we feel that most home users will not deem the extra performance offered by a card such as the Radeon HD6990 a good trade-off compared to the low noise output of the custom cooling system on the ASUS ENGTX580 DCU2.
The design is not without flaws - we would have preferred a closed shroud so that hot air could only exit from the back plate thereby keeping our case cooler. SLI is a big problem and its hard to envisage two of these cards side by side.
Despit these minor reservations, the benefits are very appealing and ASUS have managed to distinguish their product from the GTX580 solutions of other manufacturers in a way that makes their offering more enticing. The build quality is very good and the manufacturing accuracy is spot on with the PCI-E connector (some cards need a struggle to insert them into the slot). The use of covers over the connectors and back ports is a nice touch that protects contacts against oxidisation in the journey from factory to end user (supply chains can be many weeks or even months in duration).
Performance wise, the card is superb and only bested by a dual GPU rival (the GTX580 is still the fastest single GPU card available) and anyone purchasing one now will know they have things covered for demanding software for the next 2-3 years and will even get faster performance as new CPUs are released.
We'd like to thank ASUS UK for the review sample ENGTX580 DCU2
For its innovative design, solid build quality and low noise under heavy workload we are awarding the ASUS ENGTX580 DCU2 our Hardware Review Gold Award.
All trademarks are the property of their respective owners.