#17,
*drool* I'll one up you here and get a second 1GB 5200, run them in SLI, then I'll have 2GB of GPU bandwidth !!!
That would kick, ermmm, no .. waitaminute ...
Maybe the sole purpose of the card is to get some honest benches under Doom 3 where ATI ties/beats a comparably-priced offering from Nvidia, and get as much PR out of it as possible. Sounds like there's a lot to sort out, but maybe this does happen under some tests in uncompressed mode.
Anyway, the purpose wouldn't be to sell the particular card, as to squelch the "OMG, Nvidia pwns in Doom3!" so that those who buy a card based on one game/anecdote won't shy away from current ATI products. Fanboys for either company are a steady source of revenue, so it won't do to miss out on a new (vid chip/FPS) generation of them.
Lot of speculation on my part, but the product is definitely befuddling otherwise.
More data points: Beyond3D got the same results as Anandtech; DriverHeaven got an increase in speed, but only because its 256MB speed was much lower than the rest.
Right, #11. Shows I should read more carefully. That's really strange. T-Break also got the same results as Anandtech. Perhaps Carmack made a mistake and the modes are mislabeled? I mean, it's not as if he could test them. :)
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
70 Comments
View All Comments
WT - Wednesday, May 4, 2005 - link
#17,*drool* I'll one up you here and get a second 1GB 5200, run them in SLI, then I'll have 2GB of GPU bandwidth !!!
That would kick, ermmm, no .. waitaminute ...
MuhahahaAaAHAHAA !!!
*sarcasm sensors malfunctioning*
fishbits - Wednesday, May 4, 2005 - link
Maybe the sole purpose of the card is to get some honest benches under Doom 3 where ATI ties/beats a comparably-priced offering from Nvidia, and get as much PR out of it as possible. Sounds like there's a lot to sort out, but maybe this does happen under some tests in uncompressed mode.Anyway, the purpose wouldn't be to sell the particular card, as to squelch the "OMG, Nvidia pwns in Doom3!" so that those who buy a card based on one game/anecdote won't shy away from current ATI products. Fanboys for either company are a steady source of revenue, so it won't do to miss out on a new (vid chip/FPS) generation of them.
Lot of speculation on my part, but the product is definitely befuddling otherwise.
Phantronius - Wednesday, May 4, 2005 - link
Ahhh silly silly ATI, when will you learn, in Soviet Russia, GPU memory RAMS you!!Sunbird - Wednesday, May 4, 2005 - link
I'm wondering when the fx 5200 with 1GB memory (64bit memory interface of course) is coming out :pET - Wednesday, May 4, 2005 - link
More data points: Beyond3D got the same results as Anandtech; DriverHeaven got an increase in speed, but only because its 256MB speed was much lower than the rest.ET - Wednesday, May 4, 2005 - link
#13, don't judge the waste of money on Doom3 results. There's obviously something fishy here.ET - Wednesday, May 4, 2005 - link
Right, #11. Shows I should read more carefully. That's really strange. T-Break also got the same results as Anandtech. Perhaps Carmack made a mistake and the modes are mislabeled? I mean, it's not as if he could test them. :)bob661 - Wednesday, May 4, 2005 - link
I'm really surprised at these 512MB results. I thought at least the Doom 3 benchmarks would be better. This is just a waste of money.CrystalBay - Wednesday, May 4, 2005 - link
I think nVidia's recently launched 6200/512MB is even a more peculiar release.mbhame - Wednesday, May 4, 2005 - link
Guys, Extremetech said they *originally* tried Ultra but the graph is based on High Quality mode....which further confuses things here. :(