X1800 XT/Pro Review

[BrianG]

http://www.hardwareanalysis.com/content/article/1813.1/

Benched in HL2, Doom3, Far Cry and Splinter Cell.

Edit: There have been several threads around the hardware web discussing these results. ATI reps continue to convey messages to certain websties that the actual release cards will be faster than the competition, but there have not been any benchmarks released.

Concerning CrossFire and the R520, it is best to continue to wait for official reviews.

[oldsk00l]

More of a 7800GT competitor, if it's for a good deal it might be a decent buy to get.

I just really hope their price point is good, otherwise ATI got pwnt.

[oldsk00l]

if the article is true....ATI is pulling an Intel....a part that's not really all that competitive in the high end...but would be a great deal with a lower tag...

I have to say, if those parts went for like $350, consider me sold, at the prices mentioned though I would avoid those like the plague :(

[skoprowski]

Makes me happy I chose to stick with my 6800GT SLI setup for now :D

[Riptide]

Wow this is not encouraging at all... It barely matches and in many cases loses to the GTX. Only in FC does it do, marginally, better.

This is what we waited all this time for? Weak.

Hopefully they'll be able to tweak the drivers or something because if this is representative of the true picture they just got their ass handed to them.

[Ninjaman09]

Feels good to be an early 7800GTX adopter. *phew*

[Tygerwoody]

Nvidia > all

[superklye]

Nvidia > all
QFT

[Ninjaman09]

Nvidia > all
Word, got my second 7800GTX coming! Can't wait to set up SLI! :D

[Acid Rain]

Holy froggen crap! (http://img210.imageshack.us/img210/4947/115850op.jpg) A 6800 Ultra running pretty close to even with ATi's new baby? It is Doom 3, but, jimmeny christmas!!

I think I'll be playing Quake IV and Prey on my new setup, thank you very much! Far Cry aside, those extra pipes on the GTX really seem to be paying off.

I'd like to see Anand and everyone else benchmarks. Can this complete owneage be true?

[Red_Shift]

Wow X1800XT loosing to 7800GTX in HL2 with a product that'll launch 3 months after :eek: :eek: and it even has larger (which is about meaningless) and faster memory :eek: :eek: it only wins in Far Cry, I guess the faster memory is giving X1800XT the edge there.
For the sake of us consumers I hope this is not true.

[Riptide]

For the sake of us consumers I hope this is not true.Not to mention ATI stockholders...

[anzak]

it only wins in Far Cry, I guess the faster memory is giving X1800XT the edge there.

Heh, vertex performance is what would give it a lead in Far Cry. That does make sense since ATI is faster in the vertex department. No benchmark seems too high or too low in comparison to the other benchmarks.

[DEaD_IRiS]

Feels good to be an early 7800GTX adopter. *phew*


seconded :)

[oldsk00l]

Heh, vertex performance is what would give it a lead in Far Cry. That does make sense since ATI is faster in the vertex department. No benchmark seems too high or too low in comparison to the other benchmarks.

right, however the price makes it a ****ty deal, if it's true

[anzak]

right, however the price makes it a ****ty deal, if it's true

I agree. Although you should take into account the 512mb of memory.

We really only need one thread for this topic.

[anzak]

I've found one serious flaw in the benchmarks. Its the gap between the X850XTPE and 6800GT in Doom 3.

This guy's benchmarks show the 6800GT is 30% faster.
http://www.nvnews.net/vbulletin/attachment.php?attachmentid=13366&stc=1

In X-Bit labs latest roundup the 6800GT is a little slower at the same settings.
http://www.nvnews.net/vbulletin/attachment.php?attachmentid=13365&stc=1

[-=DVS=-]

They used different maps anzak.

[oldsk00l]

I've found one serious flaw in the benchmarks. Its the gap between the X850XTPE and 6800GT in Doom 3.

This guy's benchmarks show the 6800GT is 30% faster.
http://www.nvnews.net/vbulletin/attachment.php?attachmentid=13366&stc=1

In X-Bit labs latest roundup the 6800GT is a little slower at the same settings.
<anzak's handy quoted charts>

Nice find, I'd be highly surprised too if the 6800GT got a magical burst of speed since then too.

Would be nice to see all this ratified. I don't rule out the possibility that the numbers are fudged just trying to smear ATI either. This just kinda puts a monkey wrench in it.

Funny thing is, reading the news over at Rage3d, it seems we're being more friendly to ATI than most of the Rage3d users are.

...ironic isn't it? :D

[Darkoz]

I've found one serious flaw in the benchmarks. Its the gap between the X850XTPE and 6800GT in Doom 3.

In X-Bit labs latest roundup the 6800GT is a little slower at the same settings.


The settings aren't quite the same, different maps were used and the tests were done on two different systems.
You certainly can't make an equal comparison under those different circumstances.

[anzak]

The settings aren't quite the same, different maps were used and the tests were done on two different systems.
You certainly can't make an equal comparison under those different circumstances.

But thats a massive difference. Something is fishy about those benchmarks, if you disagree with that then your kidding your self.

They used different maps anzak.

I've used both cards and played through at least half of the game on both of them. The performance difference is next to nothing.

[OWA]

I think the benchmarks are supposed to just give you a rough idea. They were done by passing along the scripts/benchmarks, etc.. It's possible the guys that did them had their settings slightly different or whatever and considering all the variables, I don't think you can try to compare percentages and say they're fake based on that. Not saying this is legit though since I have my doubts.

One of the hardocp benchmark shows the GT outpacing the XT PE so I don't think that part is difficult to believe.

[SH64]

OWA : Those were in older drivers .. in the newer drivers ATi's performance in Doom3 improved.
however i dont find the numbers fishy since they used a "custom" benchmark not the regular timedemo or one of Xbits benchies.
that custom benchmark probably stress the parts where nVIDIA's advantage can be experianced more than other benchies & stress the cards overall (just like VIA VGA's benchmark).. hence the GTX/SLI performance which looks to be weak if compared to other map/timedemo benchmarks.

[retsam]

you notice today ati had two scandels...ati only commented on one... which one is true and which one is false....at least we do know one thing....one is right, one is wrong...

[anzak]

One of the hardocp benchmark shows the GT outpacing the XT PE so I don't think that part is difficult to believe.

That is the X800XT PE not the X850XT PE. That was also before the Catalyst AI was introduced which put the X800XT about on par with the GT.