Jump to content

NVIDIA "G80" GeForce 8800GTX Performance Benchmark


toyo

Recommended Posts

NVIDIA is set to launch its upcoming G80 GeForce 8800GTX and 8800GTS graphics cards next week, however, DailyTech snagged a GeForce 8800GTX board to run a couple quick benchmarks on. The GeForce 8800GTX used for testing is equipped with 768MB of GDDR3 video memory on a 384-bit memory bus as previously reported. Core and memory clocks are set at 575 MHz and 900 MHz respectively. Other GeForce 8800 series features include 128-bit HDR with 16x anti-aliasing and NVIDIA’s Quantum Physics Engine.

Previous NVIDIA graphics cards in single card configurations were limited to lower levels of anti-aliasing. With the GeForce 8800 series, users can experience 16x anti-aliasing with only a single card. DailyTech has verified the option is available in the NVIDIA control panel.

The physical card itself is quite large and approximately an inch and a half longer than an AMD ATI Radeon X1950 XTX based card. It requires two PCI Express power connectors and occupies two expansion slots. An interesting tidbit of the GeForce 8800GTX are the two SLI bridge connectors towards the edge of the card. This is a first for a GeForce product as SLI compatible graphics cards typically have one SLI bridge connector.

Having two SLI bridge connectors onboard may possibly allow users to equip systems with three G80 GeForce 8800 series graphics cards. With two SLI bridge connectors, three cards can be connected without any troubles. NVIDIA is expected to announce its nForce 680i SLI and 650i SLI chipsets with the GeForce 8800 series. NVIDIA nForce 680i SLI and 650i SLI based motherboards are expected to have three PCI Express x16 slots.

Full benchmark specs

Link to comment
Share on other sites


  • Replies 6
  • Views 2.8k
  • Created
  • Last Reply

o man, what will i have to do to own one of these babies???

8800gtx+SLI board=1000$+...

and there will be the need for proc, ram, etc, etc...

how happy i would be to throw away my 2400 sempron+6600+512ram...if i sell thw whole system maybe i end up with enough money to buy a 8800gts :))

Link to comment
Share on other sites


It’s a known fact that GPU architecture is in some ways more advanced than CPU related one. And it’s also known the fact that - in order to release all the computational power residing in your GPU - you would need something like a GPGPU (general purpose GPU) and by that I mean a GPU with architecture capable of interpreting codes written in a specific language. Ati made the first move by optimizing its architecture in order to run Folding@Home project. At the moment Nvidia stood still.

Then the launch of G80 GPUs arrived. These chaps use a unified shader architecture incorporating 128 independent stream processors which actually run at a higher speed than the core itself (1.35GHz vs. 575MHz) making 8800 GTX somewhat similar to NetBurst CPUs who had the ALU units working at twice the speed of the CPU.

Nvidia is full of surprises. The latest came in form of a C compiler for G80 GPUs integrated into their latest software suit named CUDA (Compute Unified Device Architecture). What this actually does is transform the C code into instructions that the G80 GPU can execute. And, if you take into account the massive power this GPU has, you’ll get the idea.

“CUDA gives us a whole new level of computing capability and enables closer access to the hardware. CUDA makes it possible for Acceleware’s electromagnetic simulation and geophysical processing products to continue to double in speed each year, and, with our OEM partners like SPEAG, will enable us to address the needs of new markets such as biomedical imaging and reservoir modeling. The latest advancements from NVIDIA are helping to quickly push the boundaries of product development and commercial science," said Ryan Schneider, CTO of Acceleware Corp.

“CUDA opens the door for new ways to analyse and interpret seismic data, allowing for interaction with multi-terabyte prestack surveys. With this compelling new GPU architecture from NVIDIA, we can accelerate some of the most computationally intensive algorithms in oil and gas exploration – far beyond the performance CPUs are capable of delivering,” stated Alex Krueger, president of Headwave, Inc.

Last heard that it might be incorporated on a INTEL mboard as well..may be by next spring..

Link to comment
Share on other sites


Now Nvidia wants to produce an IGP for INTEL CPUs in order to counter the attack coming from AMD’s RD600.

They are trying to further develop more G80 derivatives, with the purpose of releasing no less than 9 different models supporting DirectX 10 extensions. There are 3 products being launched in February and each will be split into 3 separate flavors.

From Softpedia...

Link to comment
Share on other sites


  • 2 weeks later...
o man, what will i have to do to own one of these babies???

8800gtx+SLI board=1000$+...

and there will be the need for proc, ram, etc, etc...

how happy i would be to throw away my 2400 sempron+6600+512ram...if i sell thw whole system maybe i end up with enough money to buy a 8800gts :blink:)

:blink:

guess what.

I got mad when I read about the new GeForce 8800GTX. because I've had bought recently the Geforce 7900GT. Grrrr. Now I'll try to sell it and get the new one.

Link to comment
Share on other sites


  • 1 month later...
boulder omen

you bought the Geforce 7900GT? the AGP version? the 8800gtx is PCI-e, you better make sure you have PCI-e before you seel your 7900 lolol

Link to comment
Share on other sites


Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...