A while ago we examined the new dual-GPU graphics card NVIDIA GeForce 295, based on two 55nm GT200b chips. And today we are going to review a single-chip GeForce GTX 285 card that is to replace GTX 280, the previous single-GPU top card. First of all, it's a usual upgrade to a better process technology that leads to corresponding improvements in power consumption, heat release, and a little performance gain.
GeForce GTX 280 hasn't had a single-GPU competitor from AMD since its release. RADEON HD 4870 X2 is more expensive and reaches higher frame rates in popular applications, and HD 4850 X2 is not widely spread. However, the latter card is the best choice for comparison with the new GTX 285, if we consider their recommended prices. It conforms well with AMD's concept of manufacturing High-End dual-GPU graphics cards. Unlike AMD, which already manufactures only multi-GPU solutions for the High-End segment, NVIDIA is not going to give up the idea of single-GPU top cards, and GeForce GTX 285 continues this good strategy.
Unlike GTX 295, our product under review is a very pragmatic solution for NVIDIA, designed to reduce manufacturing costs and to ensure large sales volumes. Together with the overhauled GTX 260 with its 216 stream processors based on GT200b, this model should improve financial situation of the company, which has not been doing very well of late. And that's what NVIDIA got used to. Manufacturing costs of the new cards, which have come to replace GTX 260 and GTX 280, are significantly lower owing to 55nm GPUs and a simplified PCB design. Of course, the company should have launched the overhauled products before the beginning of 2009. But to all appearances, there were some problems with GT200b.
Theoretical part of our GeForce GTX 285 review will be very short, because we already examined the GT200 architecture (GT200b is not different) and even wrote several articles about the GT200b. And the only differences from GeForce GTX 280 that matter to common users are increased operating frequencies and reduced power consumption.
If you are not familiar with the architecture of GeForce GTX 200 (GT200), you can read about it in our baseline review. This architecture developed from G8x/G9x with some modifications. Before you read this article, you should study the baseline theoretical articles -- DX Current, DX Next, and Longhorn. They describe various aspects of modern graphics cards and architectural peculiarities of products from NVIDIA and AMD.
These articles predicted the current situation with GPU architectures, and confirmed many of our assumptions about future solutions. The detailed information about NVIDIA G8x/G9x unified architecture is provided in these articles:
We assume that you are already familiar with the GT200(b) architecture. Now let's examine the detailed characteristics of a new card from the GeForce GTX 200 series that has a 55nm GPU.
GeForce GTX 285
- Code name: GT200b
- process technology: 55nm
- 1.4 billion transistors
- Unified architecture with an array of common processors for streaming processing of vertices and pixels, as well as other data
- Hardware support for DirectX 10, including Shader Model 4.0, geometry generation, and stream output
- 512-bit memory bus, eight independent 64-bit controllers
- Core clock: 648MHz
- Higher than doubled ALU frequency: 1476MHz
- 240 scalar floating-point ALUs (integer and floating-point formats, support for FP32 and FP64 according to IEEE 754(R), two MAD+MUL per cycle)
- 80 texture address and filtering units, support for FP16 and FP32 components in textures
- Dynamic branching in pixel and vertex shaders
- 8 wide ROPs (32 pixels) supporting antialiasing with up to 16 samples per pixel, including FP16 or FP32 frame buffer. Each unit consists of an array of flexibly configurable ALUs and is responsible for Z generation and comparison, MSAA, blending. Peak performance of this subsystem is up to 128 MSAA samples (+ 128 Z) per cycle, in Z only mode - 256 samples per cycle
- Multiple render targets (up to 8 buffers)
- Interfaces (2 x RAMDAC, 2 x Dual DVI, HDMI, DisplayPort, HDTV) are integrated into a separate chip.
GeForce GTX 285 reference specs
- Core clock: 648MHz
- Frequency of unified processors: 1476MHz
- Unified processors: 240
- 80 texture units, 32 blending units
- Effective memory frequency: 2484 (2*1242) MHz
- Memory type: GDDR3
- Memory: 1024 MB
- Memory bandwidth: 159GB/s
- Maximum theoretical fill rate: 20.7 gigapixel per second
- Theoretical texture sampling rate: 51.8 gigatexel per second
- 2 x DVI-I Dual Link, 2560x1600 video output
- Double SLI connector
- PCI Express 2.0
- TV-Out, HDTV-Out, HDCP support, HDMI, DisplayPort
- Power consumption: up to 183W (two 6-pin connectors)
- Dual-slot design
- Recommended price: $349
As GT200b chips are manufactured by the 55nm process technology, NVIDIA can actually launch an overhauled solution with a new name and improved performance, instead of simply replacing the GPU in GeForce GTX 280. Along with improved performance, this solution consumes much less power than its predecessor (GTX 280) and its main competitor (RADEON HD 4850 X2).
In other words, the overhauled card from NVIDIA offers similar performance at a lower power consumption level. Even though GT200b GPUs are manufactured by the 55nm process technology just like RV770 chips, they have much larger surface area and complexity (the number of transistors), but these solutions consume less power than their competitors.
Note the name of this product. Unlike GeForce GTX 295, NVIDIA decided to name the new product GTX 285, as it's a tad faster than GTX 280. Users understand this minor change in product number. It shows that the new card is little different from GTX 280 -- only a tad faster and with better characteristics (power consumption, operating temperatures, power connectors).
Architecture and features
We have nothing interesting to tell you, because GT200b is no different from GT200, it's just smaller and consumes less power. According to NVIDIA, the improved GT200 architecture based on the 55nm process technology makes NVIDIA GeForce GTX 285 faster, quieter, and less power hungry than GTX 280. We might have finished our theoretical part here, but we'll try to give you more details.
GT200 architecture was announced last summer. And if we take into account that it's actually the improved G8x/G9x architecture, it appeared even earlier -- in 2006. The main difference between G92 and G80 was the 65nm process technology, GT200 featured mostly quantitative changes, and GT200b is no different from GT200. You can read about it in our previous articles.
From the user's point of view, nothing has changed in GTX 285 compared to GTX 280, except for the GPU, a new PCB design, and slightly better characteristics. But along with a minor increase in operating frequencies, GTX 285 features significantly reduced power consumption. GeForce GTX 280, based on the 65nm GT200 GPU, consumes up to 236W. Having upgraded GT200 to the 55nm process technology, the company reduced maximum power consumption to 183W, almost by one third!
As a result, we've got not only reduced requirements to power supply, but also changes in extra power connectors. Relatively low power consumption allows to install two 6-pin connectors instead of 8-pin and 6-pin ones. They are prevalent, especially in power supply units.
As we've touched upon the topic of power consumption and energy efficiency, let's look back to AMD's assurance that High-End solutions with a single large GPU had problems with high power consumption. In order to verify this assumption, we can compare GeForce GTX 285 and its competitor from the same price segment -- RADEON HD 4850 X2. Both graphics cards use 55nm GPUs, only GTX 285 uses one complex GPU, while HD 4850 X2 uses two small GPUs operating at lower frequencies.
We'll evaluate relative performance of these solutions in the next parts of the article. And now we'll compare maximum power consumption values: 230W and 186W for HD 4850 X2 and GTX 285 correspondingly. As you can see, it's the single-GPU solution that demonstrates an almost 25% advantage in power consumption and energy efficiency (with a similar performance level), not the dual-GPU card, as AMD had claimed.
GeForce GTX 285 supports the SLI technology. Two corresponding on-board connectors allow to unite up to three such cards in a system to form 3-way SLI. Such a system will use AFR mode to process three frames in parallel. Each GPU in this configuration raises the frame rate by up to 80-90%.
It goes without saying that GeForce GTX 285 supports NVIDIA PhysX, which is actively promoted to games now, e.g. Cryostasis and Mirror's Edge. However, hardware support for computing physics in games hasn't reached the level yet to bring significant improvements in performance or gameplay.
Support for external interfaces in GeForce GTX 285 does not differ from what we saw in GeForce GTX 280. The graphics card is equipped with two Dual Link DVIs supporting HDCP. It's also possible to use an adapter to connect the card to HDMI, a popular connector in consumer electronics.
That's the end of our brief theoretical part of the article. Everything was already covered in our previous reviews. The second part will analyze synthetic performance tests of the new single-GPU card from NVIDIA, based on GT200b. We'll compare its performance with that of GeForce GTX 280 and RADEON HD 4850 X2.
Write a comment below. No registration needed!
