Originally Posted by Levesque
Why is this so hard to understand? You can't expect their results to be the same when they use a completely different method then Vega.
Vega is using in game benchmarks to compare his results with their results, that they obtained using a totally different methodology. It's totally illogical. They are doing the right thing in using real time actual gaming benchmarks.
Vega is not. They are professionnals reviewers. He is not.
Dunno if he's a proffesional reviever or not, but Hardocp is a joke.
They use ancient system coupled with 3.6 GHz CPU.
Even for high end SLI/CF setups on multi monitor resolutions. How realistic is that? Anyone who knows anything about Fermi, should know this is a nonsense setup.
And no. Slow CPU does not equally affect both NV and AMD because whole Fermi lineup needs CPU cyles/threads, while AMD can rely on QPU much more effectivly in CPU bottleneck scenarios.
In this particular test they went 3xSLI on PCI-E 4x which is mind numbing. GTX 3-Way SLI 580 lost to 6990+6970 in every single benchmark, and this is truly an accomplishment by itself.
Furthermore they test only 4-5 games.
One of which is already in AMD pocket, because they put money in development - Dragon Age 2. Although there are plenty more TWIMTBP games, which cant seem to enter their miniscule games list.
They never test games which don't work with SLI/CF.
They say they test gameplay yet they never mention micro-stuttering. Even with their archaic setup.
They often make blunders and biased benchmark.
Such as regularly comparing Nvidia's Transparency Anti Aliasing vs AMD's Adaptive AA in DirectX 10/11 titles. Which for AMD part is either a straightforward lie or a noobish error, because Adaptive Anti Aliasing works only in DirectX 9 OpenGL.
Im my opinion they have an agenda, because it's just too many errors to dismiss them as accident.