***OFFICAL X1900 XT REVIEW THREAD***

Cookie Monster

Diamond Member
May 7, 2005
5,161
32
86
Its only faster than the R520/G70 in pixel shader heavy games. Well we all expected this, but the G71 probably will be fast through ALL benchmarks.

However, i want more benchmarks because it lacks a more thorough benchmark like xbitlabs. More games, power consumption, noise, etc.
 

Jagercola

Senior member
Aug 23, 2001
384
0
76
Nice refresh... I'm still counting on my X800 Unlocked at 560/560 to carry me till the new designs come out this fall. Mmmm DX10 and hardware accelerated aero glass!
 

n7

Elite Member
Jan 4, 2004
21,281
4
81
Holy technical jargon batman! :Q

That was enough babbling for my eyes to glaze over...

Also, goddamn horrible graphs...so hard to see differences, etc.

But at least some good games tested.

Frankly, it looks to me like the X1800XT is going to become an awesome value, even better than it is now IMO.

If there's enough X1800XTs left to sell, i can see them being dumped out at fantastic prices.

Otherwise, the X1900XT looks like a great card.

I think it's pretty obvious ATi's 16 pp scheme is going to get raped by the G71 though...hopefully they'll keep prices cheaper accordingly :)
 

ElFenix

Elite Member
Super Moderator
Mar 20, 2000
102,389
8,547
126
why buy 900mhz ram to only run it at 750? or worse 725?
 

nitromullet

Diamond Member
Jan 7, 2004
9,031
36
91
Their graphs either aren't accurate or their percentages are off...

From the FEAR test.

X1900 XT is 17% faster than X1800 XT at the highest resolution with IQ on, X1900 XTX over 46% faster than GTX 512. Yay for lots of ALUs!

at 1024x768, the GTX looks like it gets about 70ish and the XTX gets about 90. Well, I don't see how 20 more FPS than 70 is "over 46% faster".

 

M0RPH

Diamond Member
Dec 7, 2003
3,302
1
0
Originally posted by: nitromullet
Their graphs either aren't accurate or their percentages are off...

From the FEAR test.

X1900 XT is 17% faster than X1800 XT at the highest resolution with IQ on, X1900 XTX over 46% faster than GTX 512. Yay for lots of ALUs!

at 1024x768, the GTX looks like it gets about 70ish and the XTX gets about 90. Well, I don't see how 20 more FPS than 70 is "over 46% faster".

They said at the highest res. At 1920x1200 it looks like roughly 45 vs 30. That's about 50% just like they said.
 

n7

Elite Member
Jan 4, 2004
21,281
4
81
Originally posted by: nitromullet
Originally posted by: videoclone
I'm more interested in a 7900GTX VS a X1800XTX that would be a nice review.
Well yeah, but that will be a while.

:Q

May just be forever, unless they've started production of the X1800s again :laugh:

 

RobertR1

Golden Member
Oct 22, 2004
1,113
1
81
According to the guys at Beyond3d the 1900xt/x has lower voltage settings on the core and mem than the x1800xt, likely to make it more appealing to OEM's (power/heat/noise/cooling issues) so it's quite possible that with some added voltage you can really OC these cards, esp the mem since it's 1.1ns meaning it should easily push 1800mhz.

I'm hoping ASUS or Sapphire will have an "OC'd" version of this card and you can simply flash you current 1900xt/x to these speeds. There's a pretty good chance of this.
 

chinkgai

Diamond Member
Apr 4, 2001
3,904
0
71
Originally posted by: n7
Holy technical jargon batman! :Q

That was enough babbling for my eyes to glaze over...

Also, goddamn horrible graphs...so hard to see differences, etc.

QFT!!! i thot i was the only one :confused:
 

chinkgai

Diamond Member
Apr 4, 2001
3,904
0
71
Originally posted by: nitromullet
These guys ever heard of numbers?

also QFT

wtfuxxor i want to compare hard digits. not the height of the bar and approximating numbers on dinky graphs. rawr!
 

darXoul

Senior member
Jan 15, 2004
702
0
0
The FC benchmark is apparently BS. The GTX 512 was MUCH faster in this game than all the other cards like 7800 GTX, especially in high resolutions. Less than 40 fps in 1600*1200? ROFL sure.
 

lopri

Elite Member
Jul 27, 2002
13,310
687
126
In most game benchmarks GTX 512 is losing to X1800XT.. by large margins (compared to X1900 vs X1800). The selected benchmarks/games and settings (i.e. AA/AF) are somewhat questionable. Well, the author sort of admitted it (needs software to really shine.. etc.), but I think this review shows only half of the picture, at best. I'll wait for AT's review.
 

Skott

Diamond Member
Oct 4, 2005
5,730
1
76
Definetly a poor choice of graphics to show their testing. Anyone else notice the CoD2 and FX-60 problem? I wonder if its the chip or they just too lazy to figure it out?