Page 8 - A word on Overclocking
Performance & Overclocking
Before we dive into a large series of benchmarks we need to discuss overclocking. With most videocards, we can do some easy tricks to boost the overall performance a little. It's called overclocking the videocard, and by increasing the frequency of the videocards memory and gpu we can make the videocard increase it's calculation clock cycles per second. It sounds hard but it really can be done in less then a few minutes. I always tend to recommend to novice users and beginners not to increase that frequency any higher then 5-10% of the core and memory clock. Example: If your card would run at 300 MHz then I suggest you don't increase that frequency any higher than 330 MHz.
More advanced users push that frequency often way higher. Usually when memory starts to show white dots 'snow' you should go down 10 MHz and leave it at that. The core can be somewhat different. Usually when you are clocking to hard it'll start to show artifacts, empty polygons or it will even freeze. I recommend that you back down at least 15 MHz from the moment you notice an artifact. Look carefully and observe well.
All in all .. do it at your own risk. Overclocking your card too fast or constantly to it's limit might damage your card and it's not covered by your warranty.
You will benefit from overclocking the most with a product that is limited or you may called it 'tuned down'. We know that this graphics core is often limited by tact frequency or bandwidth limitation, therefore by increasing the memory and core frequency we should be able to witness some higher performance results. A simply trick to get some more bang for your bucks.
The GeForce FX 5700 Ultra from Albatron, at standard 128-bit 128 MB DDR2 memory runs at default475 MHz for the core and 450MHz (x2) for it's memory. The 5700 was a decent overclocker with 515 MHz as core frequency and again 987 MHz for the memory.
These settings have been used throughout our entire benchmark suite. That means that each card has been tested on the overclocked conditions in twenty-eight individual tests without corruption or weirdness in the form of system hangs.
Both cards are consistent in all benchmarks and that's 56 single test-runs.
Test system |
Benchmark Software Suite:
* For four of these games we are making use of a custom time demo. Neither NVIDIA or ATI knows what time demo we are using. These are non-public tests which where recorded for us only. We are not going to make them public either as they are and will remain internal material. Therefore the chipset manufacturer will not have the chance of optimizing it that specific benchmark time-demo. We will do our very best now and in the future to keep a close eye on optimizations and cheats, we need to be able to show you objective results. However in the end this should be a responsibility for the chipset designer, if that entity fails to do so, then it'll lose consumer's trust and will dig it's own grave. |
All tests were made in 32 bit colors in resolutions ranging from 800x600 pixels up-to the Godfather of all gaming resolutions, the 1600x1200 several performance/quality settings.