Into the brains of the creature
Right let's handle a little overview on the 7800 GT. The second product in the Series 7 product range, the GeForce 7800 GT contains 7 vertex units and 20 pixel pipelines. It operates with a core clock of 400MHz, a memory clock of 500MHz (256MB 256-bit 2ns (8x32) GDDR3 Memory (500MHz clock - 1 GHz effective)) and uses 256-bit GDDR3. Memory bandwidth can peak at a lovely 32GB/sec. The GeForce 7800 GT also includes 128-bit floating-point (32 bits/component) through the entire graphics pipeline which is a hundred percent similar to the 7800 GTX. The graphics core was manufacturer at 0.11 micron. Judging from the specs this little creature will even punk out the GeForce 6800 Ultra which is priced similar.
Despite the speculation, the GeForce 7800 GT series has SLI capability.
As stated on the first page, the 7800 GT has a GeForce 7800 GTX core yet with some performance influencing factors disabled. Therefore this core is still is codename G70. Since feature wise we are looking at the same product, let's place all the technical differences in a table.
|
NVIDIA GeForce 6 & 7 Product Lineup Specifications |
|
Product Name |
# pixel processors |
# vertex processors |
Bus width |
Memory Type/Amount |
GPU Speed |
RAM Speed |
| GeForce 7800 GTX |
24 |
8 |
256-bit |
GDDR3/256MB |
430MHz |
1200MHz |
| GeForce 7800 GT |
20 |
7 |
256-bit |
GDDR3/256MB |
400MHz |
1000MHz |
|
GeForce 6800 Ultra |
16 |
6 |
256-bit |
GDDR3/256MB |
400MHz |
1100MHz |
|
GeForce 6800 GT |
16 |
6 |
256-bit |
GDDR3/256MB |
350MHz |
1000MHz |
|
GeForce 6800 |
12 |
5 |
256-bit |
GDDR/128MB |
325MHz |
700MHz |
|
GeForce 6800 LE |
8 |
4 |
256-bit |
GDDR/128MB |
320MHz |
700MHz |
| GeForce 6600 GT |
8 |
3 |
128-bit |
GDDR3/128/256MB |
500MHz |
1000MHz |
| GeForce 6600 |
8 |
3 |
128-bit |
GDDR/128MB |
300MHz |
275(550) |
| GeForce 6200 |
4 |
3 |
64/128-bit |
GDDR/128MB/256MB |
300MHz |
275(550) |
| 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. |
Normally we see little steps though, yet with 24 pipes, 8 vertex units, a huge transistor count combined with some nice internal hardware optimizations, the boost in performance is noticeable.
- 302 Million transistors
- 7 Vertex Shading Units
- 20 Pixel pipelines
- 256MB 256-bit 2ns (8x32) GDDR3 Memory (500MHz clock - 1 GHz effective)
- PCI Express x16 Compatibility (PCI Express Compliant)
- Integrated NVIDIA TV Encoder (HDTV, S-Video, Composite,)
- Integrated NVIDIA VIVO (S-Video, Composite,)
- Dual DVI-I Connectors
- NVIDIA PureVideo Technology
- DVD + HDTV Decode assist up to 1920x1080p resolution
- Integrated Dual 400MHz RAMDACs
- 1 Dual Link TMDS DVI Connector
- 1 Single Link TMDS DVI Connector
- NVIDIA Digital Vibrance Control (DVC) 3.0
- Support for Microsoft Video Mixer Renderer (VMR)
- Advanced adaptive de-interlacing
The first reaction when you look at the architecture of the G70 chip is that it is a prolonged 6800 core. Not entirely true, NVIDIA made some stern changes inside where all that magic happens. According to NVIDIA the internal pipelines were from top to bottom redesigned to make sure there was lower and thus better latency.
This increased performance is for each and any clock cycle. Not only that though, according to NVIDIA the vertex units have been rebuilt to increase geometric performance, a new texture engine accelerates texture processing and there are a number of other things that guarantee better performance. All in all the focus on this design was to increase its shader supremacy as this is where the future of gaming must be found.
Right let's have an overview of today's tested products.
Site
Navigation
