Discussion in 'Reviews' started by W1zzard, Aug 21, 2006.
To read this review go to: http://www.techpowerup.com/reviews/ATI/X1950XTX/
Thumbs down for "paper launch"?
How can ATI be given a thumbs-down for a so-called "paper launch"? They came right out and said the X1950XTX would not be available in mass quantities until the middle of September.
If they're still hard to find after that date has come and gone, then fine. Thumbs down. But until then, I don't see how they can be given a poor grade on availability.
Same arguement can be said for the Tumbs up on a card that isn't released yet.
Which thumbs up are you referring to?
Competitive price - MSRP is already set unless something drastic changes things between now and mid-Sept. Can't see that happening, though.
Quieter operation - I don't see that anything here that can change between now and mid-Sept.
Performance improved - See above.
Lower power consumption - Again, see above.
Works with X1900 Crossfire Master - Once again, see above.
DDR4 should offer nice overclocking head room - Yet once again, see... Oh heck, you know what to do.
It looks pretty good, would have really liked to see it running on a faster test system but it seemed to do just fine
imo not worth buying, even if your using a x1800xt, clock it and theres your near identical performance really. still, at least the stock cooler looks cool, and surprisingly, effective (once you dismantled the thing to remove that odd "support bar" going round the fins that is)
The only thing i like about it is the cooling, other then that its not much faster then my 1900xt
yeah i have a sapphire 1900 XT, i think that will last me
why do you use for the test setup a 1800+@2000mgh isnt this the thing that hold you back in tests?
could you plz re run this tests with a conroe or at leats a 2800mgh a64 cpu??
otherwise this test is ridiculous so plz do this better
these lamers ripped of your hard work
go get em w1zz!
ed- if anyone wants to see less bottlenecked scores i have an incoming system that will do it nicely, so if someone wants to pass a sample on to me ill do a review on it.
they didn't rip it off, ripping it off would be not showing any credit to techpowerup, look at the bottom of the page
boo wooti doo, they included a link to the source... ooooh how nice; as if anybody will ever visit the source ... the whole article is on their site anyway.
it's their "tradition" to generate "content" this way :/
Can you set the cooler to spin @ 100%? Any ideas as to how it would compare to an Accelero? Also, what about system temps...how much might they rise, eh?
nice job Wizz.
Is it possible to get the script info you need from an ATi partner? (IE, Sapphire, or another 3rd party thats more likely to give you info.)
Nice review, but I doubt this is the processer you actually used.
Im going out on a limb and saying he meant a 4800+, because its not a 1800 (Don't exist) and its not a 2800+ because that would not be used in a review. But it could be a 3800+
Well I was hoping he would read this at some point.
Lets ask W1zzard himself. What processor did you use W11zard?
Thankfully they didn't take credit for the review, and they re-hosted the images, so they're not leaching our bandwidth. As long as they asked permission to put it on their forum they're cool
Well spotted, thanks for telling us though.
But what's with all the ?????????'s ? ????? ?? x1900xtx???? ??? - Different language? Chinese?
Higher is better ??
I think this has to be changed to lower is better
lmao, well spotted
cpu was 3000+ @ 2000 mhz
of course it has to say lower is better .. i just came back from games convention and will fix it tomorrow
w1zz? isnt the cpu holding the cards back in this test?
I agree that it isnt worth buying, simply because when directx 10 comes out everything from the old voodoos to this x1900xtx will be void.
BTW the stock x1950XTX ramsink and Vcore mosfet strip sink are not copper.
The ramsink is a copper plated aluminum casting and the Vcore strip heatsink is a copper
plated aluminum extrusion.
The heatsink in the GPU cooler is all copper though.
Separate names with a comma.