• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

x2 or xt?? for second card

Associate
Joined
12 Oct 2005
Posts
850
Location
left abit no your other l
hi guys

i have an x2 at the mo and want to run to more monitors on my rig to bring it up to 4 also having a second card is always fun but dont know is it worth the £250 price tag this week for the x2 or if i should just go with the the cheaper xt what you think??
 
if you can afford it, X2... i say that because we could well continue to see driver optimisations that oost it's Xfire performance later down the line. but be aware, though, GPU scaling is best when it's 3 GPUs atm... depends on how long you'd want to keep them, then
 
thanks well i am planning to keep the rig for the best part of a year before a full upgrade as that seams to be the trend in my computer spending one big upgrade a year that is almost a hole new rig.
 
A single XT will be your best bet, as the 4th GPU on a 2nd X2 hardly makes a difference, your only talking about a couple of extra frames or something, so not worth adding another X2, 3x GPU's is the sweet spot. :)
 
Yep 3 GPU's scale better, I have it myself, gains are had on high rez though with AA and AS applied, dont expect miracles at lower rez's you will be disapointed.
 
Yep 3 GPU's scale better, I have it myself, gains are had on high rez though with AA and AS applied, dont expect miracles at lower rez's you will be disapointed.
What rez you at Banny?

I'm @ 1920x1200 and contemplating an additional card for crossfireX.

gt
 
I went for 2 X2s because I didn't want to take the reduced GPU speed from the XT across to the X2...which is probably not a valid justification assuming the XTs clock well.
 
I went for 2 X2s because I didn't want to take the reduced GPU speed from the XT across to the X2...which is probably not a valid justification assuming the XTs clock well.
How do you find the performance Mud?

Did you get them both at the same time or did you buy a 2nd when you needed/wanted more performance? How would you say the performance compares 1x X2 vs 2x X2's?

Sorry for all the Q's!

gt
 
right well the price difrence is about £100 so for the extra cash should i got for an x2 as i am not planning another upgrade for ages and would rather have something thats not being used yet but when it is being used will save me from an upgrade??
 
How do you find the performance Mud?

Did you get them both at the same time or did you buy a 2nd when you needed/wanted more performance? How would you say the performance compares 1x X2 vs 2x X2's?

Sorry for all the Q's!

gt

I had a single X2 for a week or so before getting a second one...it tied up nicely with an order I was making anyway, so I impulse bought. Frankly I'm not sure there's much difference between 1 X2 and 2 X2s, mostly because all the games I've been playing either ran maxed out on a single X2 or are Crysis.

I'm a bit too CPU-bound to artificially benchmark my performance (I'll download 3dmark03 now for science's sake), and I'm not hitting the kind of framerates SimonMaltby was on his overclocked (watercooled) X2s in Crysis. I'm banking on the drivers improving, but the hit on minimum framerates caused by the additional overhead of 4GPUs does concern me somewhat. I'll report back with the 3dmark03 runs with crossfire enabled and disabled. I've not really toyed with benches relative to a single X2, is there a particular game benchmark that you'd recommend?
 
That's not an issue in CrossfireX as far as I know.

I thought I read in an interview with an ATI representative that the cards would run as fast as the slowest card, though I may be mistaken. My understanding was though the core speeds do not change, since the other cores are effectively waiting for the slowest core to catch-up then their speed is equivalent to the slowest core.
 


Got weird strobing vertical bars when I disabled crossfireX, though a restart seems to have cured it. Looks like this bench advocates tri-fire if that's scaling well.
 
Cheers Mud.

Actual FPS seem to increase quite nicely - if drivers improve the scaling with 4 it could be even more viable. :)

What games do you have? I suppose fraps and a comparison of the same area in game would be more than acceptable but only if you have time!

Have given up on Crysis as i found it's framerate far too choppy in the later chapters of the game.

Really appreciate the 3DMark03 scores.

gt
 
I am using 1680 rez, reckon it would scale well with 1920 too. didn't feel the need for another x2
price doesn't justify it and heat aswell.
Mud seems to have pretty identical system to me, so his results show it all
On 3dmark06 I get 14100 at 1680x1050 4aa 16as so the scaling is clearly visable cards are all clocked though.
I agree with Mud, I see no increase in crysis bench at all
 
Last edited:
Cheers Mud.

Actual FPS seem to increase quite nicely - if drivers improve the scaling with 4 it could be even more viable. :)

What games do you have? I suppose fraps and a comparison of the same area in game would be more than acceptable but only if you have time!

Have given up on Crysis as i found it's framerate far too choppy in the later chapters of the game.

Really appreciate the 3DMark03 scores.

gt

I'm hideously busy at the moment, but an excuse to take a break is always nice! I've only got Call of Juarez, Crysis, GTR2, Live for Speed and Unreal Tournament 3 installed. GTR2 does show a decent improvement in fps when you have a lot of cars on the grid :)
 
I'm hideously busy at the moment, but an excuse to take a break is always nice! I've only got Call of Juarez, Crysis, GTR2, Live for Speed and Unreal Tournament 3 installed. GTR2 does show a decent improvement in fps when you have a lot of cars on the grid :)

GTR2 works great for sure, can run nice high setting in it.
 
Back
Top Bottom