Alright, let's start off with a generic chat on the graphics core of this graphics card. ATI's new flagship chips were developed under codename R580. Radeon X1900 (or R580 core) is in numerous ways what the R520 core should have been when it was launched. R580 uses the same memory controller found on R520, with that wicked 512-bit internal ring bus controller. Externally, the chip addresses memory with 256-bits. The R580 core still features 16 pixel pipelines yet now has 48 Pixel Shader processors. R580 is produced on a 90nm process.
Let's talk transistor count, make an estimated guess, how many transistors do you think the X1900 chip has? The X1900 chip my fellow guru's has just over 380 million transistors, that makes you think huh? NVIDIA's 7800 for example already passed 304 million and the X1800 XT had 321 million. But yeah, we are now at a stage where we are nearing 400 million transistors. Smaller fabrication processes help the companies do this. That manufacturing process is at the newer 90nm right now.
Why the massive increase of Pixel Shader cores you ask? Simple, rendering cinematic effects as fast as possible because these are being created by pixel shaders.
Here we see the pixel engine .. 12 x 4 - 48 Pixel units, each unit is organized in a quad.
So the increased Pixel Shader capacity of course is the biggest change in the design of this graphics chip architecture. A pixel pipeline pretty much consists of a pixel shader processor, a texture mapping unit (TMU) and a Raster OPerator unit (ROP). ATI has sort of designed their pixel pipelines such that the pixel shader processors are more independent, while NVIDIA has their ROP units independent.
Really you can compare this X1900 product with the X1800 series in almost every way, yet the focus here is absolutely on the 48 Pixel Shader units. Each Pixel Shader processor can handle anything from 1 to 5 shader instructions per clock cycle in its various ALUs. With 48 pixel processors the 1900 series has three times more arithmetic shading processing power when compared with the previous flagship product, the X1800 XT. ATI believes that the 3:1 ratio of arithmetic to texture units provides the ideal balance for current and future 3D performance.
To give you an more understandable example: the X1800 could manage 60 million Pixel shader operations per second. The X1900 series can do 166 million. With the focus on Shader Model 2 and 3, that alone is the money shot right there, let's not forget that Shader Model 3.0 has finally become very important.
| What is a shader ? |
| What do we need to render a three dimensional object; 2D on your monitor? We start off by building some sort of structure that has a surface, that surface is being built from triangles and why triangles? They are quick to calculate. How's each triangle being processed? Each triangle has to be transformed according to its relative position and orientation to the viewer. Each of the three vertices the triangle is made up of is transformed to its proper view space position. The next step is to light the triangle by taking the transformed vertices and applying a lighting calculation for every light defined in the scene. At last the triangle needs to be projected to the screen in order to rasterize it. During rasterization the triangle will be shaded and textured.
Graphic processors like the GeForce and Radeon series are able to perform a certain amount of these tasks. The first generation was able to draw shaded and textured triangles in hardware. The CPU still had the burden to feed the graphics processor with transformed and lit vertices, triangle gradients for shading and texturing, etc. Integrating the triangle setup into the chip logic was the next step and finally even transformation and lighting (TnL) was possible in hardware, reducing the CPU load considerably. The big disadvantage was that a game programmer had no direct (i.e. program driven) control over transformation, lighting and pixel rendering because all the calculation models were fixed on the chip. And now we finally get to the stage where we can explain Shaders. Vertex and Pixel shaders allow developers to code customized transformation and lighting calculations as well as pixel coloring functionality. Each shader is basically nothing more than a relatively small program executed on the graphics processor to control either vertex or pixel processing. |
SM3 allows the programmer to fire off some very nice shader programs that in certain cases can speed up your game. The world has moved on to SM3, people expect it to be integrated and so it has and had to be been done. Very good integration I might add because SM3 seems to work pretty darn efficiently for ATI, it has to do with dynamic branching, although that matter is too complicated to explain for this article. What you need to know is that it works really well. More efficiency, that really is what the new card is all about. I'll be using that word in this review a lot. According to the chip designers, every transistor in that core is constantly put to use to push the results onto your screen. Yes... efficiency.
Another feature in the X1000 series (and no, it's not new to our ears at all) has had a little upgrade has to do with texture compression capabilities. Almost every, well... any graphics card nowadays makes use of texture compression technology. It's been discussed here on more than one occasion, so i'm sure you recognize terms like S3TC and DXTC. Basically you reduce the byte-size of a texture whilst maintaining the best quality as possible. However, compression equals artifacts and thus image degradation at some point. 3Dc is a compression technology designed to bring out fine details in games while minimizing memory usage. It's the first compression technique optimized to work with normal maps, which allows fine per-pixel control over how light reflects from a textured surface. With up to 4:1 compression possible, this means game designers can now include up to 4x the detail without changing the amount of graphics memory required and without impacting performance.
3Dc was upgraded a little and on the X1000 series of cards we now have 3Dc+ available to us. Let me just get it out of the way and move on. High quality normal map compression can (and could) be handled up to a 4:1 ratio and works on any two-channel texture format.
This updated + version of 3Dc adds support for single-channel textures with 2:1 compression, which is good enough for stuff like luminance maps, shadow maps, HDR textures and more.
What about the cards then?
Let me present to you the entire X1900 range of cards as they will be available today. My bet is that you'll see a lot of reviews as of today.
- Radeon X1900 All-in-Wonder 500 MHz core / 500 MHz (x2) memory
- Radeon X1900 XT 625 MHz core / 725 MHz (x2) memory
- Radeon X1900 Crossfire (master card) 625 MHz core / 725 MHz (x2) memory
- X1900 XTX 650 MHz core / 775 MHz (x2) memoryD
Guru3D fact: One MHz represents one million cycles per second. The speed of microprocessors, called the clock speed, is measured in megahertz. For example, a graphics processor that runs at 600 MHz executes 600 million cycles per second. Each computer instruction requires a fixed number of cycles, so the clock speed determines how many instructions per second that processor can execute.
Site
Navigation
