• Notice regarding image hosts and attachments. See here.

  • Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Some more spuriouse fermi benchmarks

They've taken the supposed Fermi results and the 285 results which are known and then attempted to run the same benchmarks with the same settings on a 5870 outside of the control of the original benchmarking session... which means they might not match actual results.


As you look at the charts below, you will notice that we added in “simulated” HD 5870 1GB results and there is a story behind this. NVIDIA was very forward with us and revealed the specifications of the test system they were running: a stock Intel i7 960, 6GB of 1600Mhz memory and an ASUS Rampage II Extreme motherboard running Windows 7. While we weren’t able to copy the exact same system, we did our best to replicate it by using the exact same CPU and OS while estimating the memory latencies and using a Gigabyte X58-UD5 motherboard. The result? Our stock GTX 285 performed within 2% of the results NVIDIA showed us so we are confident in the accuracy of our results.
 
Last edited:
If I've read correctly HardwareCanucks ran their own tests for the 5870 using a similar system to what nVidia used when comparing the GF100 and 285.

Edit: Ahh, Rroff beat me to it! :rolleyes:

Looks positive for nVidia if those results are accurate though.
 
Last edited:
I wonder why they left physics on since havok places a great deal of stress on CPU and system resources which is good if your testing CPU's but horrible if your testing GPU vs GPU.

The numbers are meaningless as far as I'm concerned.
 
Even if havok is on and does stress the CPU those numbers are a key aspect in choosing what CPU to pair with the GPU to avoid bottlenecking and provide maximum performance in short, the CPU will always play a part in the fps of any GPU old or new
 
Even if havok is on and does stress the CPU those numbers are a key aspect in choosing what CPU to pair with the GPU to avoid bottlenecking and provide maximum performance in short, the CPU will always play a part in the fps of any GPU old or new

But they are not key in comparing gf100 to cypress, which is the whole point of this article.
 
Assuming the CPUs were the same or very close - it doesn't make much odds if havok is on or not in this case - as we can see a difference in performance... and quite a large one at that.

If they were showing very close results and the test was heavily CPU bound or run on massively different CPUs then it would be another story.

In the end this benchmark is a good indication of realworld performance in a specific title.
 
Lets not forget that they are probably comparing a £300 card to a £500+ card, so it should come to no suprise that the Nvidia card is a lot faster. They are also cherry picking the benchmarks, just like AMD do, so if they are trying to imply that they GF100 is 30-40% faster then it likely that, on average, it's 20-25% faster.
 
lol a 5870 actuly scores a lot more than what is stated on that on far cry 2 so the bench is flawed

Not with the settings/hardware that was used in those comparisons - a couple of people from these forums with similiar setups also ran it in another thread and got similiar numbers.
 
Back
Top Bottom