Discussion in 'NVIDIA' started by cooler, Jun 1, 2008.
Is this what you meant to show?
These were done by Nvidia and we know that Nvidia and ati always send bs benchmarks out so I say bullsh*t
haha, yes theyre using older drivers to help them but they arent gonna just lie. As the original XS poster showed in his modified graph, if the 4870x2 improves over the 3870x2 by 65% in between the gtx260 and gtx 280.
The 4870x2 may end up cheaper though.
Im also surprised that the gtx 280 doesnt do better. Its only about 50% better in some games.
as it's coming out on my birthday, I might try to preorder me a 4870...prepare for some decent CF results
but I dunno..I'm normally an Nvidia kinda guy..but as I try to maintain, whatever is best = use
Look at the games. COD4, UT3, FEAR. Actual framerates are probably 150 vs. 250 or whatever, but extremely high. Extremey high framerates = MASSIVE CPU bottleneck.
Funny how the AMD scores don't move....all i know is these cards are fast, very fast.
That is how the control is supposed to react. The Y axis on the graph is the number of times better each card scored over the 3870x2. That is why the 3870x2 sits at 1 the entire time.
Alright, can someone draw up an easier to read graph? They do it like this for people like me to lazy to read the legends...
yes the hd card is the baseline.
anyone have more ditels about it , i need Texture Units, Raster Units Stream Processor Clock,Memory Clock
If it had 24ROPs the memory bus would be only 384bits and capacity 768MB.
calvary1980 ninja-deleted/edited the post
According to that, Crysis performs better than FEAR
I also like how the 4870x2 is the same speed in every game
Reading graphs is not that difficult...
Sadly it seems IT IS. This is the 7th thread (split between many forums) where I see the same confusion with that graph. People blaming Nvidia for trying to fool them with it or even claiming they made it up (reason? Crysis performing better than... etc.), is very common on many of the threads too.
1 is the baseline that the ati card is measured at .. everything else is a percentage of that.
Yeah it's easy to undestand. Anyone can figure that out at the first glance. Except MANY can't from what I've seen, and instead they blame the maker of the graph for trying to fool them. It's both sad and hilarious. What did people do when they had to be at class?
Opps, i didn't notice that.
there is no 4870 on the graph.
But im glad for the graph, as teh 4870 is supposed to be as fast as a 3870x2 ... so make your inferences from the graph about the 4870X2 ....
Oh shit...I read it wrong again. Wow I'm off today.
I would expect Crysis to perform better than Fear, apart from the fact that fear's game architecture is pretty old now, at the time of release there were elements of it's architecture that were pretty unique, current cards and their architecture largely determine how well they play todays games, we all know the different methodology and architecture between ATI's and NVidia's current crop, that in some way will determine how well a current game performs on that particular card.....dependant on of course the games architecture.
They seem to "perform the same" as the fps have been standardized to the score of a 3870X2.
If a 3870X2 gets 40 fps in Game A, it is 1.0 in that bar.
If a 3870X2 gets 10 fps in Game B, it is 1.0 in that bar.
Separate names with a comma.