• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

HD4890 Xfire with HD4850

Joined
Aug 3, 2008
Messages
1,644 (0.27/day)
System Name Black Killing Machine
Processor Delidded NO IHS mount Intel i7-4770k
Motherboard Gigabyte Z87X-OC
Cooling Swiftech DIR655 pump, Watercool HEATKILLER® GPU-X³ 79X0 Ni-Bl gpu, DT SNIPER CPU block, UT60 420 RAD
Memory Gskill Trident 2400mhz CL9 Samsung IC9
Video Card(s) SAPPHIRE HD7950 OC 950mhz Edition, VaporX HD7950
Storage 2xM4 Crucial 64GB RAID 0, 1 OCZ AGILILITY 3 60GB, WDCB 500GB x2RAID0, WD Green, Hitachi 1TB
Display(s) Samsung SyncMaster 226BW and 24" 120hz BenQ gamer
Case SilverStone Raven rv-02
Audio Device(s) Audio Technica ATH-AD900 headphones, Topping D2 DAC/Headphone AMP
Power Supply Seasonic Platinum 860
Software Windows 7 Pro 64bit
Benchmark Scores http://hwbot.org/user/sonda5/ Delidding is magic. http://valid.canardpc.com/2878462
New Xfire config for my RIG.

About to boot into OS.




HD4850xFireHD4890small.JPG

Diamond OC edition HD4890 with VisionTek HD4850.

Not sure if 2 HD4890s would fit in Xfire on my MB.

Still running Corsair TX650W PSU. Hope it holds up.
 
why the 4850 as a primary card?. also is the 4850 512mb or 1gb?
 
why the 4850 as a primary card?. also is the 4850 512mb or 1gb?


For cooling. If I put the HD4890 on top it will be suffocated. I've tried different combinations with a Toxic HD4850 and VisionTek HD4850 in the past and it didn't really seem to matter.

My MB by default goes to the top slot for display as you have guessed.



Here is a view of my system.

XfiretempsHD4850withHD4890.JPG
 
37% scaling loss from different GPU's.

Is it a 1GB 4850?
 
37% scaling loss from different GPU's.

Is it a 1GB 4850?

No. It's a 512MB.


I'll be gaming to see if there are any advantages to this set up over my HD4850 Xfire set up I had before.

As of right now It's really not looking like it's a good match. The HD4890 by itself scored over 20,000 points on 3dMark06.


Here is a furmark benchmark with the HD4890+HD4850. The Furmark WEB site reports my set up to be 2 HD4850s. That is incorrect. Scored a little over 8,300.

http://www.ozone3d.net/benchmarks/score.php?id=17ce10c456b948c2f12cc20c073367c2

With my HD4850 Xfire set up I was just under 8,000.
 
well that does make sense... sell the 4850 and pick up another 4890. They would fit, but you might have to do a bit of modding to make them run cooler.
 
well that does make sense... sell the 4850 and pick up another 4890. They would fit, but you might have to do a bit of modding to make them run cooler.


I'm not sure if I want to get sale my DFI P45 JR T2RS mATX MB just to be able to fit another HD4890.

I'm probably just going to go back to HD4850 Xfire. Why? Well for one thing I could get a higher 3dMark06 score. Not sure how actual game play works out. I'll be playing my favorite games (mainly UT3) over the next few weeks with this HD4850 and HD4890 Xfire configuration.

Maby the next CCC 9.5 drivers will improve Xfire performance. Right now this match is a total waste of money.


HD4890 by itself rocks for 3dmark06. Scored over 20,000 with single HD4890 in my RIG.

Add a HD4850 in Xfire and everything goes south. :cry:

terriblematchHD4890andHD4850.JPG
 
From what I understand it's supposed to clock down to the lowest card, only reason crossfire is showing such a small performance difference. If I'm wrong someone correct me?
 
From what I understand it's supposed to clock down to the lowest card, only reason crossfire is showing such a small performance difference. If I'm wrong someone correct me?

That is what I have read.

On the good side overall game play has greatly improved.

Slaughtering on UT3 right now. Cozworld.
 
How does everything play with just your single 4890? If it clocks high enough it may be more beneficial to just leave that card single for the time being.
 
How does everything play with just your single 4890? If it clocks high enough it may be more beneficial to just leave that card single for the time being.

Single HD4890 is awesome with the 3D power during game play. Better than my Hd4850s in crossfire. Sad that I can't get as good 3dMark06 score though. If I had dual Hd4890s maby Crossfire would work on my MB.

As it stands I'm not sure I can run 2 HD4890s in my MB. I'd probably have to upgrade my psu from Corsair TX650 to something else.
 
Well if it runs better in games, what's the question? Are you worried about 3dmarks or are you worried about fps?
 
as people are saying, the clock down to the lowest common denominator.

Amount of shaders, GPU/shader/RAM clocks, even ram amount.

as far as anything is concerned, you ARE running two 4850's there. the only differences would be in the memory bus and ram timings.

Just run the single 4890.
 
Crossfire does not down clock the faster card. Crossfire will run async. It does reduce the ram to the lowest available amount though. So an HD 4890 would be reduced to 512Mb with an HD 4850 512Mb.

For the heck of it you might want to try switching the card's positions.
 
Crossfire does not down clock the faster card. Crossfire will run async. It does reduce the ram to the lowest available amount though. So an HD 4890 would be reduced to 512Mb with an HD 4850 512Mb.

For the heck of it you might want to try switching the card's positions.

i'm not so sure on the async part, but its definately a good idea to swap the cards around. the primary (master) card may have a larger impact.
 
I'll try swapping them to see what if there are any improvements.


3D eye candy is much better with this set up over my HD4850 Xfire set up.

The 3dmark06 score doesn't bother me. Gaming is more important.
 
Crossfire does not down clock the faster card. Crossfire will run async. It does reduce the ram to the lowest available amount though. So an HD 4890 would be reduced to 512Mb with an HD 4850 512Mb.


I didn't notice any peformance problems when I went from single HD4890 to xfire with VisionTek HD4850. I did notice performance gains in 3dMark06 and UT3.
 
i'm not so sure on the async part, but its definately a good idea to swap the cards around. the primary (master) card may have a larger impact.

My system didn't take the swapping very easy. Had to completely unistall drivers and kill them 100% with CCC, Driver Sweeper, Registry removal, system hardware device removal, ect....

After all that I have gotten it to boot up with a single HD4890 in the primary PCI slot. I'm going to run a quick 3dMark06 with it with my CPU at 3.6 GHZ. Then add the other HD4850 in the bottom slot and run the same 3dMark06 test (specs) that I did yesterday with the CPU@4.25 GHZ.
 
Last edited:
My system didn't take the swapping very easy. Had to completely unistall drivers and kill them 100% with CCC, Driver Sweeper, Registry removal, system hardware device removal, ect....

After all that I have gotten it to boot up with a single HD4890 in the primary PCI slot. I'm going to run a quick 3dMark06 with it with my CPU at 3.6 GHZ. Then add the other HD4850 in the bottom slot and run the same 3dMark06 test (specs) that I did yesterday with the CPU@4.25 GHZ.



Just ran the 3dMark06 with the HD4890 in the primary slot by itself with the CPU@3.6GHZ and scored

SM 2.0 7347
SM 3.0 8436
CPU Score 5698

Total Score 18653.


Seems on track for the same peformance score of 20,333 that I scored yesterday with the Hd4890 in the bottom (non primary) slot with my CPU@4.25 GHZ.

I'm going to pop the HD4850 in the bottom slot now and go for another 4.25 GHZ CPU Xfire run.

I'm expecting close to 23,000. If there is an improvement it will be because I have the HD4890 in the primar slot.

It's an extremely tight fit and I think the heat may effect peformance because there isn't going to be much room for the HD4890 to breathe.


HD4890primaryHD4850bottomXfire.jpg
 
Game over for HD4890 on the primary with HD4850 on bottom. Did a fresh install with 9.4 drivers and drivers don't recognize the set up. :(

I can do Xfire set up with HD4850 on the top (primary) and HD4890 on the bottom.

Also I was able to monitor my temps a bit and as expected the temps are much hotter with the Hd4890 on top. About 20 degrees hotter. No air.


So...

HD4890 by itself is awesome.

With HD4850 in Xfire a little more. Not much more. Not worth the money IMO. Better off with HD4890 by itself.

So my experience with HD4890 is done.
Good stuff. I love the smooth play and 3d eye candy compared to my 2 Hd4850s with 512mb of GDDR3 each.

2 HD4890s doesn't seem like a good build for my MB because of heat issues. Would work with liquid cooling but I'm not willing to spend that much money for a set up like that.

Especially with better ATI 3D technology right around the corner.
 
Alright so you're sticking with the single 4890? Sounds like a good plan if so.
 
Alright so you're sticking with the single 4890? Sounds like a good plan if so.


I don't think so. Because I have an extra PCI slot for Xfire. I'm not willing to fork out the cash for liquid cooling to get 2 HD4890s to work. I'll probably just stick with my Hd4850 Xfire set up till ATI comes out with the better GPUS later this summer.

Or... I might sell the HD4850s for HD4770 Xfire. Provided they stomp all over HD4890 performance.
 
I don't think so. Because I have an extra PCI slot for Xfire. I'm not willing to fork out the cash for liquid cooling to get 2 HD4890s to work. I'll probably just stick with my Hd4850 Xfire set up till ATI comes out with the better GPUS later this summer.

Or... I might sell the HD4850s for HD4770 Xfire. Provided they stomp all over HD4890 performance.

I'm relatively sure you'll get very similar performance. Between the two crossfire set-ups there isn't quite a fair test to look at reviews as only review of 4770 crossfire I can find is with an i7. But assuming they are a bit better than a single 4890 and your two 4850's are a bit better than your 4890, one can only figure out that it'll be close.
 
I'm relatively sure you'll get very similar performance. Between the two crossfire set-ups there isn't quite a fair test to look at reviews as only review of 4770 crossfire I can find is with an i7. But assuming they are a bit better than a single 4890 and your two 4850's are a bit better than your 4890, one can only figure out that it'll be close.

My advantage is that I have experience gaming with HD4890. I have a few settings that I use when I play UT3 that I tried with the HD4890 that my HD4850 xfire set up couldn't pull off.

So if the HD4770 in crossfire trully is better than a single HD4890 I will know. I am certain. :rockout:


If it doesn't work out I may end up going HD4850x2 in X fire with HD4850 for Tri fire.
 
Back
Top