Sapphire Radeon HD 5850 CrossfireX

Games Testing

Crysis Warhead
 
Crysis Warhead is without a doubt one hard nut to crack, especially at higher resolutions and a dash of Anti Aliasing and Anisotrophic Filtering. Typically, this game is heavy on both CPU and GPU however the limitation at high resolutions skews the dependancy towards the latter. Depending on how well the drivers are optimised, Multi GPU scaling could vary greatly. But anyway, given the game's "tough" characteristics, we were eager to find out the end result.
 
 
 
In single card mode, the Sapphire Radeon HD 5850 has a 4 frame per second deficit over the previously tested HD 5870. This may not seem like a lot but every little bit matters in the sub 25fps range as we steadily approach game play which could be described as choppy.
 
FarCry 2
 
Far Cry 2, a slightly older hit but based on a fairly demanding engine also has a fair level of GPU dependancy at least particularly so at higher resolutions. This particular game features a fabulous benchmarking tool, which loops an intensive Far Cry 2 scene and gatherers minimum, maximum and average framerates accordingly. This allows for a more accurate comparison from one test item to another. As mentioned, we have set all Quality Settings to the highest possible within the tool.
 
 
Unlike Crysis Warhead, all of our graphics card configurations played Far Cry 2 without so much as a stutter. This shouldn't detract from the fact that the scaling on both graphics cards in Multi GPU mode was excellent. The Radeon HD 5870 maintains a fair lead over the HD 5850 once again.
 
Microsoft Flight Simulator X
 
Microsoft Flight Simulator X is part of a handful of games that are heavily CPU reliant. If you thought that this was bad, at initial release this game did not have multicore cpu support. I need not say anymore on that front, but with this in mind, the results that followed were not too surprising.
 
 
The game's CPU dependancy is quite clear from the level set of results.
 
Call of Duty 4
 
Call of Duty 4 is one of the older games within our testing. Once again, all quality settings are set as high as possible and Dual Graphics Card support has been checked in game. Our particular CoD4 test is not terribly strenuous but the performance scaling is quite clear.
 
 
In both configurations, the Radeon HD 5870 outpaces the 5850 by considerable margins. Once again however, the HD 5850 plays the game flawlessly and the quality of game play would be no different to the other graphics card configurations.
«Prev 1 2 3 4 5 Next»

Most Recent Comments

01-11-2009, 21:25:50

tinytomlogan
Ask any man, 1 is never enough. Mul puts a pair of fast women on the bench to see how 2 Sapphire 5850 shaped ladies perform under preasure.

Continue Reading

02-11-2009, 14:12:16

MeltedDuron
Great review as usual Mul. The 5850's look like a good buy! They should be awesome around christmas when the price drops kick in, ATI's cards are the real working class hero's of graphics

02-11-2009, 14:28:23

Mul.
Quote:
Originally Posted by name='MeltedDuron'
Great review as usual Mul. The 5850's look like a good buy! They should be awesome around christmas when the price drops kick in, ATI's cards are the real working class hero's of graphics
Cheers MeltedDuron.

As for price drops, I fear it may well go the other way until nVidia has a direct competitor. Given that the production of GeForce GTX 260/275/285's, VAT increases and a (currently) low supply of Cypress cards to meet demand, prices could even rise. We'll see how things pan out though.

02-11-2009, 15:07:35

Rastalovich
Great review.

Got a few questions I'm not sure of:

  • Is the difference between the 5850 & 5870 purely clock speed, or is there something capability-wize the 5870 can do that the 5850 can't ?

  • Why does the i7 cpu score in the FutureMark benchie vary so much with the differing 4 setups ?



Unless there's a non-clock answer to the 1st question, I don't see any point in any1 buying 2x 5870 to xfire. There's a difference of some %age of fps in games that the 2x 5850 are already scoring over 100s. To that end, I don't see the cream.

Need to revisit this once there are some/few Dx11 comparisons to be had. Making comparisons of these cards in Dx10 at best doesn't seem to present much visible differences.

Decent card on the face of it. Nothing being made of AMD scoring nice pricing, and perhaps more to come, from the fact they have the only Dx11 out in this window. We know it would, and will, be different in a months time, and I'm sure there will be some shouting from the roof tops then - especially when these cards get cut.

02-11-2009, 15:24:58

Mul.
Cheers for your thoughts Rastalovich,

The answer to the first question is no unfortunately. Aside Frequencies, the 5850 variant of the Cypress core has 8 5 Way Shading Units disabled, resulting in a 1440 Stream Processors overall, which is 10% less than the HD 5870 at 1600. The consequence is that you'd need to push a HD 5850 to slightly higher core frequencies in order to match the HD 5870.

As for the second question, I'm unsure as to why the CPU Score varied so much. Identical testbeds were used, and the same goes for the drivers.

With regards to revisiting these cards with DX11 hits, I completely agree.

02-11-2009, 15:28:55

Rastalovich
The variation of the cpu score concerns me.

If u were to lay out a flat comparison of 5850/5870/5890, I'd personal need a serious answer to why the likes of the i7 cpus are either being held back or being used to bolster deficiencies.

Moreover, would it mean that if u had

02-11-2009, 15:39:54

Mul.
It's a shame I didn't publish the results on the Futuremark ORB so at least the scores could've been dissected further. It's also interesting that the HD 5850 CPU scores are similar to each other, as are the HD 5870 CPU scores. One couldn't rule out an anomalous result for one of the two but I couldn't see how the differences could be so considerable. Prior to testing, I had confirmed the system's CPU and DRAM frequencies at 3.40GHz and as mentioned, the same driver sets and testbed was carried forward for both tests. Strange.

02-11-2009, 17:42:56

iceie
the temp on the 5850 is really impressive

04-11-2009, 12:54:31

Talonaer
"Straight off the bat, a quick Google Shopping search reveals a 25% saving incurred in opting for Radeon HD 5850's in Single or Multi GPU format over the equivalent HD 5870 configuration."

300 - > 200

or 600 -> 400 ?

Thats a 33% saving or a 50% increase in price, dependant on which way oyu look at it :P. Sorry to be picky just annoyed me

04-11-2009, 13:07:04

Bungral
Yeah I saw that but couldn't be arsed to say anything.

Was more interested in the results.

08-11-2009, 12:07:14

Mul.
Quote:
Originally Posted by name='Talonaer'
"Straight off the bat, a quick Google Shopping search reveals a 25% saving incurred in opting for Radeon HD 5850's in Single or Multi GPU format over the equivalent HD 5870 configuration."

300 - > 200

or 600 -> 400 ?

Thats a 33% saving or a 50% increase in price, dependant on which way oyu look at it :P. Sorry to be picky just annoyed me
School boy error quite frankly. Fixed
Reply
x

Register for the OC3D Newsletter

Subscribing to the OC3D newsletter will keep you up-to-date on the latest technology reviews, competitions and goings-on at Overclock3D. We won't share your email address with ANYONE, and we will only email you with updates on site news, reviews, and competitions and you can unsubscribe easily at any time.

Simply enter your name and email address into the box below and be sure to click on the links in the confirmation emails that will arrive in your e-mail shortly after to complete the registration.

If you run into any problems, just drop us a message on the forums.