HD 3850 Crossfire vs. 8800GTS 512MB G92

Is an ATI HD 3850 Crossfire setup a match made in heaven? - We compare to NVIDIA's new G92 based 8800GTS to find out.

Published
Updated
Manufacturer: none
11 minutes & 39 seconds read time

Introduction





We've been given an indication of the pricing which is going to accompany the new 8800GTS; when it came to figuring out what exactly it would compete against in the red corner, we soon after came to the following conclusion.

The HD 3850! Oh, but not just one of them, rather two working together in Crossfire. So it was clear what we had to do; grab a pair of HD 3850 cards, stick them into our beloved P35 motherboard and let them go at it. Two setups enter with one setup to come out on top.

What we have with us today is a pair of GECUBE HD 3850 cards under the new X-TURBO III naming scheme. With an increased core and memory clock, and without much of a difference in price these could be the 3850s to pick up.

It's been a while since we've seen anything from GECUBE, and though we're not going to cover these cards in detail today (except for their performance), we will take a deeper look in a future article.

What we will be doing now is to briefly look at the clock speeds of the cards (below our test system details), and then we'll just simply get straight into it with the benchmarks to see if HD 3850s in Crossfire have what it takes to beat out the singular G92 based 8800GTS.

Test System Setup and 3DMark05


Test System Setup

Processor(s): Intel Core 2 Quad Q6600 @ 3GHz (333MHz x 9)
Cooling: Corsair Nautilus500 (Supplied by Corsair) with Arctic Cooling MX-2 Thermal Compound (Supplied by Arctic Cooling)
Motherboard(s): ASUS Blitz Extreme (Supplied by ASUS)
Memory: 2 X 1GB Corsair XMS3 DDR-3 1066MHz 7-7-7-21 (Supplied by Corsair)
Hard Disk(s): Hitachi 80GB 7200RPM SATA-2
Operating System: Windows XP Professional SP2, Windows Vista
Drivers: Catalyst 7.11, Forceware 169.06 (8800GTS 512MB)

With a pretty good idea of what an 8800GTS 512MB is going to set you back (providing stock levels aren't a problem), for similar money you're able to find yourself with a pair of Radeon HD 3850s. Not only this, but you can grab a pair of pre-overclocked GECUBE variants with an even better than standard cooler on them.

The GECUBE X-TURBO III ATI Radeon HD 3850 manages not only to pack some extra speed, but also get an upgrade from 256MB of DDR3 to 512MB; this to match the G92 based 8800GTS. Looking at clockspeeds, the GECUBE offerings come in with a core clock of 725MHz and 1800MHz DDR on the memory. The 8800GTS 512MB on the other hand packs a default clock speed of 650MHz with a 1650MHz shader clock to boot. It also runs with a relatively high 1940MHz DDR memory clock.

Across our normal line up of real world and synthetic benchmarks we will see how a two card Crossfire setup competes against the soon to be released 8800GTS 512MB. Where is your money best spent? Since we already have the 8800GTS' overclocked results in hand from a previous article, we thought we may as well throw them into the mix as well and see what comes out in the wash.


3DMark05

Version and / or Patch Used: Build 130
Developer Homepage: http://www.futuremark.com
Product Homepage: http://www.futuremark.com/products/3dmark05/
Buy It Here




3DMark05 is now the second latest version in the popular 3DMark "Gamers Benchmark" series. It includes a complete set of DX9 benchmarks which tests Shader Model 2.0 and above.

For more information on the 3DMark05 benchmark, we recommend you read our preview here.




Straight away we can see the Crossfire setup takes a pretty good lead. We see it doesn't take as big a hit at the higher resolution as the single 8800GTS 512MB.

Benchmarks - 3DMark06


3DMark06

Version and / or Patch Used: Build 110
Developer Homepage: http://www.futuremark.com
Product Homepage: http://www.futuremark.com/products/3dmark06/
Buy It Here




3DMark06 is the very latest version of the "Gamers Benchmark" from FutureMark. The newest version of 3DMark expands on the tests in 3DMark05 by adding graphical effects using Shader Model 3.0 and HDR (High Dynamic Range lighting) which will push even the best DX9 graphics cards to the extremes.

3DMark06 also focuses on not just the GPU but the CPU using the AGEIA PhysX software physics library to effectively test single and Dual Core processors.




We can again see the Crossfire setup takes a pretty serious lead. The Crossfire setup manages to score the same at 1920 x 1200 as the 8800GTS scores at the default 1280 x 1024 resolution.

Benchmarks - Half Life 2 (Episode One HDR)


Half Life 2 (Episode One HDR)

Version and / or Patch Used: Latest from Steam
Timedemo or Level Used: Custom Timedemo
Developer Homepage: http://www.valvesoftware.com
Product Homepage: http://www.half-life2.com
Buy It Here




By taking the suspense, challenge and visceral charge of the original, and adding startling new realism, responsiveness and new HDR technology, Half-Life 2 Episode One opens the door to a world where the player's presence affects everything around him, from the physical environment to the behaviors even the emotions of both friends and enemies.

We benchmark Half Life 2 Episode One with our own custom timedemos as to avoid possible driver optimizations using the "record demo_name" command and loading the timedemo with the "timedemo demo_name" command - For a full list of the commands, click here.




We can see in our first real world game performance test that things don't look quite as attractive for the Crossfire setup. With that said though, Source based games really don't like dual card setups, be it SLI or Crossfire.

Benchmarks - Prey


Prey

Version and / or Patch Used: Unpatched
Timedemo or Level Used: HardwareOC Custom Benchmark
Developer Homepage: http://www.humanhead.com
Product Homepage: http://www.prey.com
Buy It Here




Prey is one of the newest games to be added to our benchmark line-up. It is based off the Doom 3 engine and offers stunning graphics passing what we've seen in Quake 4 and does put quite a lot of strain on our test systems.




The OpenGL based prey manages to favor the single NVIDIA card across all resolutions..However, when it comes to the higher resolution the difference between the cards aren't as far apart.

Benchmarks - F.E.A.R.


F.E.A.R.

Version and / or Patch Used: Unpatched
Timedemo or Level Used: Built-in Test
Developer Homepage: http://www.vugames.com
Product Homepage: http://www.whatisfear.com/us/
Buy It Here




F.E.A.R. (First Encounter Assault Recon) is an intense combat experience with rich atmosphere and a deeply intense paranormal storyline presented entirely in first person. Be the hero in your own spine-tingling epic of action, tension, and terror...and discover the true meaning of F.E.A.R.





F.E.A.R. has always managed to make great use of multi card setups, and here is no different. You can see big gains on the CF setup when compared to the single 8800GTS 512MB.

Benchmarks - Company of Heroes


Company of Heroes

Version and / or Patch Used: 1.7
Timedemo or Level Used: Built-in Test
Developer Homepage: http://www.relic.com
Product Homepage: http://www.companyofheroesgame.com
Buy It Here




Company of Heroes, or COH as we're calling it, is one of the latest World War II games to be released and also one of the newest in our lineup of benchmarks. It is a super realistic real-time strategy (RTS) with plenty of cinematic detail and great effects. Because of its detail, it will help stress out even the most impressive computer systems with the best graphics cards - especially when you turn up all the detail. We use the built-in test to measure the frame rates.






The CF setup manages to stay ahead of the single 8800GTS 512MB, but you can see when overclocking comes into play the 8800GTS manages to get a bit of a leg up on the competition.

Benchmarks - World in Conflict


World in Conflict (DX9)

Version and / or Patch Used: Demo
Timedemo or Level Used: Built-in Test
Developer Homepage: http://www.massive.se
Product Homepage: http://www.worldinconflict.com




World in Conflict is a real-time strategy video game by Massive Entertainment and to be published by Sierra Entertainment for Windows (DX9 and DX10) and the Xbox 360.

The game is set in 1989 where economic troubles cripple the Soviet Union and threaten to dissolve it. However, the title pursues a "what if" scenario where, in this case, the Soviet Union does not collapse and instead pursues a course of war to remain in power. It is an intensive new game is sure to put plenty of stress on even the latest graphics cards and we use the built-in benchmarking for our testing.





WIC performance is a bit all over the place; performance doesn't really move down but just fluctuates with the Crossfire setup. You can see at the highest resolution though that the CF setup does come out ahead.

High Quality AA and AF




Our high quality tests let us separate the men from the boys and the ladies from the girls. If the cards weren't struggling before they will start to now.


3DMark06



Back to 3DMark06 and we again see big gains on the Crossfire setup.


Half Life 2 (Episode One HDR)



Moving back to the Source engine, we see the cards put out some disappointing numbers again. Source and multi card setups just don't play along at all.


Prey



Throw in AA and an OpenGL based game and you just know it's just not going to look good for an AMD based graphics card. You can see here that the stock 8800GTS is in excess of 30% faster.

Benchmarks - 3DMark06 - Vista


3DMark06

Version and / or Patch Used: Build 110
Developer Homepage: http://www.futuremark.com
Product Homepage: http://www.futuremark.com/products/3dmark06/
Buy It Here




3DMark06 is the very latest version of the "Gamers Benchmark" from FutureMark. The newest version of 3DMark expands on the tests in 3DMark05 by adding graphical effects using Shader Model 3.0 and HDR (High Dynamic Range lighting) which will push even the best DX9 graphics cards to the extremes.

3DMark06 also focuses on not just the GPU but the CPU using the AGEIA PhysX software physics library to effectively test single and Dual Core processors.




3DMark06 continues to favor the Crossfire setup even under Vista; which doesn't come as any real surprise to us.

Benchmarks - Call of Juarez (DX10) - Vista


Call of Juarez (DX10)

Version and / or Patch Used: DX10 Benchmark Demo
Timedemo or Level Used: Built-in Test
Developer Homepage: http://www.techland.pl/en/
Product Homepage: http://www.coj-game.com

Buy It Here



Call of Juarez is a Western-themed FPS from Techland. The North America release was one of the first games to utilize DX10.

Call of Juarez is loosely based on a number of Western movie hits from the last four decades, particularly the Spaghetti Westerns that became popular in the sixties and early seventies. The game features two different characters: Billy "Candle", a young fugitive accused of murder and the Reverend Ray, an ex-outlaw who turned to religion, believing that God himself has chosen Ray to be the instrument of His wrath.





The generally AMD favored Call of Juarez benchmark surprisingly shows the G92 based 8800GTS win out at every resolution this time round.

Benchmarks - Lost Planet (DX10) - Vista


Lost Planet: Extreme Condition

Version and / or Patch Used: Demo
Timedemo or Level Used: Built-in Test
Developer Homepage: http://www.capcom.com
Product Homepage: http://www.nzone.com/object/nzone_lostplanet_home.html




Driven to the brink of extinction on ice-covered wastelands, humankind fights to survive. From extreme conditions and insurgent Snow Pirates to hostile gargantuan alien Akrid, all that you can trust are your instincts.

Combining a gripping single player campaign and intense multiplayer modes with support for up to 16 players online, Lost Planet is an epic gaming masterpiece. Enormous world maps unfold as players' battle across vast snow fields and deserted cities either on foot or in armed, robotic Vital Suits.



Lost Planet across the board continues to favor the NVIDIA offering. Considering an option for Multi-GPU is available in Lost Planet, it doesn't really seem to do a whole lot.

Benchmarks - Company of Heroes (DX10) - Vista


Company of Heroes

Version and / or Patch Used: 1.7
Timedemo or Level Used: Built-in Test
Developer Homepage: http://www.relic.com
Product Homepage: http://www.companyofheroesgame.com
Buy It Here




Company of Heroes, or COH as we're calling it, is one of the latest World War II games to be released and also one of the newest in our lineup of benchmarks. It is a super realistic real-time strategy (RTS) with plenty of cinematic detail and great effects. Because of its detail, it will help stress out even the most impressive computer systems with the best graphics cards - especially when you turn up all the detail. We use the built-in test to measure the frame rates.




We can see the 8800GTS 512MB comes out ahead at the lower resolution, but as more and more load is placed on the card(s) the Crossfire setup manages to come out ahead.

Temperature and Sound Tests


Temperature Tests



With the TES 1326 Infrared Thermometer literally in hand we found ourselves getting real-world temperatures from the products we test at load (3D clock speeds).

There are two places we pull temperature from - the back of the card directly behind the core and if the card is dual slot and has an exhaust point we also pull a temperate from there, as seen in the picture.




As you would expect, the second card (furthest from the CPU) manages to heat up a bit more than the first, but really it's not by a huge amount and certainly nothing we would really worry about.


Sound Tests



Pulling out the TES 1350A Sound Level Meter we find ourselves quickly yelling into the top of it to see how loud we can be.

After five minutes of that we get a bit more serious and place the device two CM away from the fan on the card to find the maximum noise level of the card when idle (2D mode) and in load (3D mode).




Noise levels are only slightly up for the dual card setup, and still better than the smaller 8800GT which pumps out 70+dB when under load.

Power Consumption Tests






Using our new PROVA Power Analyzer WM-01 or "Power Thingy" as it has become quickly known as to our readers, we are now able to find out what kind of power is being used by our test system and the associated graphics cards installed. Keep in mind; it tests the complete system (minus LCD monitor, which is plugged directly into AC wall socket).

There are a few important notes to remember though; while our maximum power is taken in 3DMark06 at the same exact point, we have seen in particular tests the power being drawn as much as 10% more. We test at the exact same stage every time; therefore tests should be very consistent and accurate.

The other thing to remember is that our test system is bare minimum - only a 7,200RPM SATA-II single hard drive is used without CD ROM or many cooling fans.

So while the system might draw 400 watts in our test system, placing it into your own PC with a number of other items, the draw is going to be higher.




This was quite interesting, we really expected power draw to be significantly greater than the single 8800GTS. However, the G92 is quite a power hungry card whilst the new 38XX AMD offerings are a huge improvement over the older HD 2900 series in terms of power draw.

Final Thoughts




Deciding which route to go isn't going to be an easy one as both setups offer some pretty good numbers. If you're gaming around the 1280 or 1600 mark you're probably better off with the 8800GTS 512MB as it seems to handle the lower resolutions better. With that said though, if you're gaming around that area you're probably better off with the cheaper 8800GT.

If you're interested in high resolution gaming we would probably opt for the Crossfire setup as it tends to offer similar performance to the 8800GTS 512MB, if not a little faster in some cases. With that said though, if you're going to venture into the land of AA and AF we would probably suggest the 8800GTS.

Originally we thought that the power was going to be a drawback for the Crossfire setup, but the power figures between it and the 8800GTS 512MB are so close to each other that it doesn't really come into the equation.

The biggest let down with the 3850 Crossfire setup is the AA and AF performance; AMD and the driver team just can't seem to get that side of things right. What will be interesting now though is the performance we get out of the Catalyst 7.12s which should be due out soon. AMD really need these drivers to be good so they can get over the line.

Both setups tend to be pretty good and it would be hard to say a particular one is better than the other. As we mentioned before though, both setups have their strengths and weaknesses. OpenGL along with AA performance favors the 8800 while higher resolution gaming and especially games that have the ability to take advantage of a multi card setup tend to favor Crossfire.

PRICING: You can find products similar to this one for sale below.

USUnited States: Find other tech and computer products like this over at Amazon.com

UKUnited Kingdom: Find other tech and computer products like this over at Amazon.co.uk

AUAustralia: Find other tech and computer products like this over at Amazon.com.au

CACanada: Find other tech and computer products like this over at Amazon.ca

DEDeutschland: Finde andere Technik- und Computerprodukte wie dieses auf Amazon.de

Shawn takes care of all of our video card reviews. From 2009, Shawn is also taking care of our memory reviews, and from May 2011, Shawn also takes care of our CPU, chipset and motherboard reviews. As of December 2011, Shawn is based out of Taipei, Taiwan.

Newsletter Subscription
We openly invite the companies who provide us with review samples / who are mentioned or discussed to express their opinion. If any company representative wishes to respond, we will publish the response here. Please contact us if you wish to respond.