Going first base with an XFX member
Let's talk swiftly about the graphics core of the the GeForce 6800 GS.
The GPU (graphics processor unit, the core) itself is build upon a 0.11 Micron fabrication process technology. Transistor talk then, the 6800 has a lot of transistors not as much as the new 7800 series have but still, it's impressive. There are 222 Million transistors. This is a 12 pipe version so part of that is inactive. Imagine all that on a 40mm x 40mm FlipChip GPU running 425+ MHz. Amazing. Since we are on the topic of the graphics core, inside it there are precisely five vertex units active. The number of pixel pipelines are identical to the 6800 model as there are twelve configured in 4 quads.
Today's card has been equipped with 256 MB of gDDR3 memory, also called framebuffer.
| What does "frame buffer" mean? |
| The amount of memory that holds the entire bit-mapped image that is sent to the monitor. The frame buffer usually is stored in the memory chips on the video adapter. The video chip set can also be integrated into the motherboard design, and the frame buffer stored in the main memory. |
$ffffffffff ---------------------------------------------------
$ffffffffff Display adapter information
$ffffffffff ---------------------------------------------------
$0000000000 Description : NVIDIA GeForce 6800 GS
$0000000001 Vendor ID : 10de (NVIDIA)
$0000000002 Device ID : 00c0
$0000000003 Location : bus 5, device 0, function 0
$0000000004 Bus type : PCIE
| What is MHz (Megahertz) ? |
| The clock frequency of a processor is usually measured in MHz or GHz for the latest CPUs. The clock frequency of a CPU determines how many times per second the transistors can change state, or, rather, switch between 1 and 0. |
$000000000f PCIE link width : 16x supported, 16x selected
$0000000009 Base address 0 : f2000000 (memory range)
$000000000a Base address 1 : e0000000 (memory range)
$000000000b Base address 2 : none
$000000000c Base address 3 : f3000000 (memory range)
$000000000d Base address 4 : none
$000000000e Base address 5 : none
$ffffffffff ---------------------------------------------------
$ffffffffff NVIDIA specific display adapter information
| What does "clock" frequency mean ? |
| The clock rate measures the number of pulses emitted from a computer's clock in one second. The clock rate is often measured in MHz (or GHz) and a higher clock rate usually means a faster CPU or Graphics Processor. (This is not always true though, because the CPUs architecture is also important.) |
$ffffffffff ---------------------------------------------------
$0100000000 Graphics core : NV42 revision A1 (12x1,5vp)
$0100000001 Hardwired ID : 00c0 (ROM strapped to 00c0)
$0100000002 Memory bus : 0-bit
$0100000003 Memory type : DDR (RAM configuration 07)
$0100000004 Memory amount : 262144KB
$0100000005 Core clock : 11.322MHz
$0100000006 Memory clock : 548.438MHz (1096.875MHz effecti...
$0100000007 Reference clock : 27.000MHz
Above is a BIOS dump from the graphics card taken with RivaTuner, the software which you can observe needs a small update when you look at the discrepancies. Let me enlighten briefly what happens in the pixel pipeline for you to understand its importance.
The LCD you are staring at for example has 1280x1024 dots, thus 1.3 million individual pixels that need a 32-bit or lower color. Each pixel that is rendered on your screen goes through a pipe where it'll receive its complex color/effect etc. Each time that pixel is altered it'll pass through the pixel pipeline, one pass is one clock cycle. You can imagine going from 8 towards 12 pipes can bring you a nice performance increase. The Series 7 7800 GT(X) actually has 24 of them. They are scalable, each pipe is available at any time in sets of 4, which we call quads. Today's product as stated has 12 pixel pipelines running at a clock frequency of 485 MHz.
| Refence Specs |
GeForce 6600 |
GeForce 6600 GT |
GeForce 6800 |
GeForce 6800 GS |
GeForce 6800 GT |
GeForce 6800 Ultra |
GeForce 7800 GT |
GeForce 7800 GTX |
| Codename |
NV43 |
NV43 |
NV42 |
NV42 |
NV40GT |
NV40U |
G70 |
G70 |
| Transistors |
? |
? |
222 million |
302 million |
302 million |
| Process, GPU maker |
110nm |
110nm |
130nm |
110nm |
130nm |
110nm |
110nm |
| Core clock |
300 MHz |
500 MHz |
Up to 400 MHz |
425 |
350MHz |
400MHz |
400 MHz |
430 MHz |
| Memory |
128MB DDR1 |
128MB GDDR3 |
128MB DDR1 |
128/256MB gDDR3 |
256MB GDDR3 |
| Memory bus |
128-bit |
256-bit |
| Memory clock |
Up to manufacturer |
2x500 MHz |
2 x 325MHz |
2x
500MHz |
2 x 500MHz |
2 x 600MHz |
2 x 500MHz |
2 x 600MHz |
| PCB |
P212 |
P212 |
P2?? |
P210 |
P210 |
- |
| Pipelines |
8 |
8 |
12 |
16 |
16 |
20 |
24 |
| FP operations |
FP16, FP32 |
| DirectX |
DirectX 9.0c |
| Pixel shaders |
Pixel Shaders 3.0 |
| Vertex shaders |
Vertex Shaders 3.0 |
| OpenGL |
1.5+ (2.0) |
2.0 |
| Availability |
Now |
So what are the major advantages of the Series 6 6800 products ? Well, feature wise we are looking pretty much at the same technology we have known for 14-15 months now. What you need to remember is that any Series 6 graphics card can achieve what a modern game expects from it. Obviously the keywords over the past couple of years has been "Shader technology." It really changed the way we look at games from a graphical point of view. It allows the game programmers to take games to a next level in both a visual and performance terms.
As always, that's the point where we land and quickly discuss on Shader model 3.
| So Hilbert, 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 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 (GeForce 256). 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. |
Talking about Shader Model 3
If you program or play computer games or even recently attempted to purchase a video card, then you will have no doubt heard the terms "Vertex Shader" and "Pixel Shader". The step from 2.0 towards 3.0 was a small one and most Shader Model 2.0 games can easily be upgraded towards Model 3.0, which can bring more performance to that gaming experience. DirectX 9 was recently updated and we are going to see more and more support for 3.0 Shaders.
Is SM 3.0 technology a huge visual advantage over 2.0? Nope, not even the slightest bit. Yet any technological advantage is always welcome and preferred over a previous generation's development. What you need to remember about Shaders 3.0 it that it can and will be used only in several critical places where it can give a performance boost and graphics cards are all about performance my friends. Both ATI and NVIDIA now offer shader model 3 support in the new products. GeForce series 6 and upwards and for ATi their X1000 series and upwards.
Talking about HDR
Another big trendy implementation that will bring games closer to a movie like quality experience is HDR.
Both ATI and NVIDIA
have been focusing extremely hard on HDR. They put a lot of money into their technology to support HDR in the best possible way, and they should as it just is a fantastic effect that brings so much more to the your gameplay experience. HDR is something you all know from games like Far Cry. It's extremely bright lighting that brings a really cool cinematic effect to gaming. This effect is becoming extraordinarily popular.
Valve recently released a new HL2 level in the form of Half Life 2: Lost Coast. Go download it as it'll show and amaze you what HDR can do. The difference is obvious. HDR means High Dynamic Range. HDR facilitates the use of color values way beyond the normal range of the color palette in an effort to produce a more extreme form of lighting rendering. Typically this trick is used to contrast really dark scenery. Extreme sunlight, over-saturation or over exposure is a good example of what exactly is possible. The most simple way to describe it would be controlling the amount of light used present in a certain position in a 3D scene.
Half Lide 2 - Lost Coast level. If you bought the game, available for free on Steam.
HDR is already present in Far Cry, Splinter Cell: Chaos Theory and in Half Life 2: Lost Coast. It will be available in Unreal 3 and likely a large number of other games. Let the screenshots do the talking.
Site
Navigation
