CGTalk - Quadro vs. GeForce (actual tests)

View Full Version : Quadro vs. GeForce (actual tests)

sykosys

07-08-2009, 11:32 PM

Hello, all. This is a book, but necessary to cover points, and is a duplicate of what I posted over in the area... but I figured more would get use of it here. Couple follow on posts with additional images.

Recently, I bought a new machine, and wanted to determine, once and for all, if it’s really worth buying quadro video cards over geforce, as well as HD speed testing between SSDs and traditional drives in RAID. This is relvant to everyone who reads this forum. I did this for my own reasons, wanting to have the best performing machine possible, because I don’t have time left for waiting unnecessarily. I own my own studio (just me ATM), and have 10 machines in my farm. When I buy new, I want to spend my money where it will get me the most bang for the buck.

Life happens, and while I was well intentioned on making a very in-depth article about this, I just don’t have the time. However, I do think that it would be of benefit to all to read this. I will likely post it on my site at some point so it can be found after it stale dates here. My software is Max (2008 in these tests), and I tested on both x64 & Vista 64, using a GTX 285s in single and SLI, a Quadro 5600 and 4500, together and apart. On the hard drive side, I compared 6 OCZ Core SSD drives, and 16 500GB Seagate 7200RPM drives on a Highpoint RocketRaid card. (Full system specs later.)

All tests were done on an i7 extreme system. If anybody is interested in the full spreadsheet, just ask. I did each test 3 times, on a clean install, with nothing other than OS, necessary drivers, and software tested. I used the benchmarks noted and two of my own scenes. the 9.1MP scene is free to anyone who asks. These are not truely exhaustive tests, I admit, but represent a lot of the type of work that I do.

HD performance: (see hdperformance.jpg) Tests were done with IO meter in different configurations. As you can see from the graph, you can get a lot more performace out of HDs as more drives are added, as you would expect. Probably more performance is available with some other (more expensive) RAID cards. This test determined, for me, that SSDs just are not raid-friendly enough to compare to a larger traditional raid array. Especially when you consider that a 6 disk raid 5 array in this case was $600 (excluding card) for a 160GB array. The same array in traditional drives would cost $540 for a 2.5TB array… prices from 5 months ago. (I was seeing if it made sense to have 6 SSDs, and 10 7200.11 drives was faster than 16 7200.11 drives… because I also need at least 2TB of working space.) Ultimately, I chose 16X 7200.11 drives, with 4 RAID 5 arrays, because I didn’t want to chance my system to a drive failure. (Even though I have extensive Ghost backups.) Which is fortunate, because a drive failed on me two months ago in the middle of a project.

Card performance: I was most interested to see how the Quadro cards stacked up against the GeForce cards; I’ve bought many Quadro cards over the years (always used, on ebay), but even then, a pair of brand new GeForce cards would always have been 40% less than the used pro hardware. I also did this test with Vista, which, for the most part, was so slow it wasn’t worth doing more tests. Exception being DX10, which doesn’t work in max, anyway.

Specview: (see specview.jpg) As you can see, in this benchmark, the Quadro cards all significantly outperform their gaming counterparts. The benchmark obviously hit a wall with Maya, where all the Quadro configs rated at maximum. I wasn’t satisfied with this, however, as Spec doesn’t necessarily represent what I would see… and the ‘real world’ test below confirmed that.

Max specific: (see bismarck.jpg & 9pt1mil.jpg) At the time, the only relevant 3D software I had was max. (I now have SolidWorks, but no time for testing.) These were done on basic setups, the first being a large scene of the battleship bismarck that you can see on my site, copied once, with the camera spinning around. The second was a flythrough of copies (not instances) of a piece of geometry, totalling 9.1 million polys. Vista performance was crap, but we know that. The most surprising thing for me was that the margin between the 285s and the 5600 was minimal, ultimately only made larger by maxtreme drivers. Around 2% without drivers, 18% with. The 285 SLI setup was marginally slower, unless I had one monitor plugged into each card (which makes most sense.) You can, therefore, decide if it’s worth an extra $2400 for an 18% gain… there may be more of a gain from going with SLI 5600s, but I wouldn’t want to finance that. :) I averaged wireframe and shaded viewport runs.

Gaming: (3dmark.jpg, ut.jpg, crysis.jpg) Not that this is relavant to a work machine, but I figured it worth doing the tests when I was at it. No big surprise here - the gaming cards cleaned the clock of the quadros.

Conclusions: Unless you’re in a ‘money doesn’t matter’ or ‘when every second counts regardless of cost’ situation, don’t waste your time with Quadros. If you do, save 66% and buy on ebay. I did do some non-scientific testing of scenes that used more extensive features, like real-time shadowing and the like, but I didn’t have time to do definitive tests. (Damn the need to make a living.) It appeared that the Quadros had a little more headroom here, amplified by drivers, but you could just get yourself VRay RT, and use your old machines (or even buy a couple of small, cheap systems) to solve that problem for less than the cost of the higher-end card. And then have a bigger render farm to boot.

Notes: I’ve got the Quadros (both) in my machine right now, due the benefits of maxstreme in 2008. I’ve just installed 2010, and I will be testing the 285s, and see how they work out there. In talking with a SolidWorks rep recently, he told me how he has had personal experience with gaming cards overheating to the point of fire when put into heavy SW environments. I don’t know what other mitigating factors may be there, like insufficient airflow, but my guess is if you’re dealing with scenes that heavy, you likely have the budgets for Quadros.

As far as hard drives, large arrays of 7200RPM Sata2 drives are the most cost effective. SCSI drives are just so expensive, and the RAID controllers are even more. SSD have issues with write/rewrite which can cause many other issues, and simply are not as RAID ready right now. The newest generation may be better than these. As it is, with my current system, I see VERY fast load/transfer times, due in large part into the way that the arrays are divided.

System:

Intel i7 extreme /w CoolIt Freezone Elite Peltier Cooler
Biostar X58 motherboard
12GB OCZ Platinum DDR3 (4X3GB)
16 Seagate Barracuda 7200.11 drives
SilverStone Strider 1200W 80Plus PSU
Highpoint RocketRaid 2340 16 Channel Raid
Lian-Li V2000Plus Chassis, custom printed front vented/fan mount front grille
3X120mm 140CFM fans, 2X60MM 50CFM fans.
I hope that some of you will find this useful. Just as a matter of interest, I paid $1250USD for the 5600 ($3200 retail), and $660 for the two GeForce Cards.

Kris.
Mafic Studios, Inc.
http://www.maficstudios.com (http://maficstudios.com/)

sykosys

07-08-2009, 11:34 PM

More images...

sykosys

07-08-2009, 11:35 PM

Last image, and follow on comment...Just wanted to add my tests of 2010 - viewport performance is, in all cases, slower. Attached is a graph; /M is maxtreme, /D is directx, using the Quardo 5600 and GeForce 285s. This is not on a clean system like previous tests - rather my working system that’s been under use for a couple of months. In lieu of the fact that there are no longer performance drivers for quadros (per nVidia), there’s even less reason to go Quadro now. I used current drivers.

With the quadros, an average of a 28% decrease in speed from Maxtreme to DX/2010, though only a 3% decrease from DX to DX/2010. GeForces suffer a 14% decrease. In 2010, the GeForces are 11% slower than the quadros in 2010.

The other image is a screenshot of the Bismarck scene used in these tests.

Kris.
Mafic Studios, Inc.
http://www.maficstudios.com (http://maficstudios.com/)

SanjayChand

07-09-2009, 12:25 AM

Did you try using the SSD as a boot drive with the OS and hence supplementing with 7200rpm drives for data storage and what not?

sykosys

07-09-2009, 12:35 AM

My base speed tests were as drives off of the RocketRaid. After those tests were done, I loaded the OS there, to see what kind of performance gain there might/might not be. The reality is there was none. Windows loads faster on the 7200.11 drives, period. Keeping in mind that these were OCZ Core drives; the new Vertex drives might be a little better... or some of the higher end drives.

Kris.
Mafic Studios, Inc.
www.maficstudios.com (http://www.maficstudios.com)

imashination

07-09-2009, 02:08 AM

So, what youre saying is, theres hardly any difference between a gaming card and a quadro when used in real life scenes in max, but that the autodesk + nvidia funded specview benchmark shows a much much bigger difference?

And that if you use a geforce card on a production machine it will burst into flames through the sheer awesome power of cad?

sykosys

07-09-2009, 02:42 AM

Yes, that the specview bench, best I can see, is a load of crap. The card bursting into flame was an experience from a SW rep. Any card could burst into flame if a) there was something wrong with the card to begin with b) there was almost no airflow, and extremely heavy scenes were under use constantly and/or c) there was 1/2" of dust/lint/lighter fluid on top of the card.

Frankly, if any one could cause a card to burn, it would be a gamer. IMHO, anyway. I'm just passing along the story I was told. :)

Kris.
Mafic Studios, Inc.
www.maficstudios.com (http://www.maficstudios.com)

sykosys

07-09-2009, 02:51 AM

A quick add - I've been using the GeForces for a couple of days now, and I'm starting to get a lot of weird behavior. In Photoshop, OpenGL only works on the primary monitor (this is apparenlty a known bug), that didn't occur with the Quadros. Max seems to be acting really wierd, mostly viewport problems, again on non-primary monitors. (I've got three running ATM.) I'm not sure if it's related, because I've installed a couple of plugins and Mudbox, which is also being wonky.

It could be a driver issue, but then I don't nVidia would see it to be in its best interst to fix these types of issues... :/ Keeps the Quadros justified w/o having to write card-specific software.

Kris.
Mafic Studios, Inc.
www.maficstudios.com (http://www.maficstudios.com)

dmeyer

07-09-2009, 10:58 AM

biliousfrog

07-09-2009, 12:37 PM

Should probably test Quadros and GeForces from the same generation if you're looking for an apples to apples comparison. The FX 5800 is the workstation equivalent of the 285. The 5600 is equivalent to the 8800 Ultra.

I kinda agree but most people that would be considering the two types of cards would primarily be looking at price comparisons and which offers the best performance for the price...in which case even the 5600 would have to be f*@£in' amazing when a GTX 295 can be between a third and a tenth of the price.

sykosys

07-09-2009, 12:43 PM

Technically, you're right. However, when I bought the cards, the 5800 wasn't yet available. Further, I chose not to go bleeding edge - which would have entailed the 295 and, later the 5800, and I took the next step down in both lines. My biggest factor was having cards that had CUDA enabled, because I wanted to access the GPU side of Photoshop.

My other reasoning is that I am not afraid to spend big bucks on a workstation (my current rig is $7K, and I've spent as much as $10K) - but I could not justify (or stomach) the price of that card, which would have been ~$4800 at release. I had asked PNY/nVidia if they would be willing to provide a sample card for me to test - remember, I paid for all of that hardware out of pocket - but they didn't get back to me on it, so it didn't make it into my review.

In spite of that, I do think this review is a valuable assessment of current cards, when one considers price. Had I had time for a complete review, I would have mentioned it. You may be able to extract another 10-20%, but the cost is staggering. One could also add a second 5600...

Kris.
Mafic Studios, Inc.
www.maficstudios.com (http://www.maficstudios.com)

sykosys

07-09-2009, 01:16 PM

The max problems were a result of monitor order - I had switched the desktops around for mudbox, and that cause those problems. Photoshop is still an issue, though, but goes away if you turn off open GL... so if you need it, you have to run PS on your primary, or your working window will look like the attached...

Just as an FYI. Still driver issues, though, didn't happen with the Quadros.

Kris.
Mafic Studios, Inc.
www.maficstudios.com (http://www.maficstudios.com/)

imashination

07-09-2009, 01:44 PM

Glad im not the only one who thinks specview is a steaming pile ;-)

On the PS issue, personally the difference between ogl or non-ogl I dont see the point, does anyone really care whether PS instantly zoom to where you want it (no ogl) or whether it plays a small zooming in animation first? (with ogl)

One thing to try, go into the gfx card prefs and look for the multiple monitor acceleration option. I have mine set to single display performance mode, I use this with 2x screens and all my 3d apps and even PS in ogl mode are all fine even when spread across both screens. Ive never seen that noise pattern in your screenshot

sykosys

07-09-2009, 02:41 PM

Single monitor performance helped a bit, but the funkiness seems to still be there at the edges. (Thanks just the same.) To tell you the truth I don't so much care about openGL now; the only thing that might be slightly useful is canvas rotating... but I have Painter for that if I really need it (and would likely need it there more, anyway). It's moreso that I expect in future releases that Adobe will include more GPU related tasks. Then it may matter more than it does now.

Kris.
Mafic Studios, Inc.
www.maficstudios.com (http://www.maficstudios.com)

aglick

07-09-2009, 03:13 PM

Kris's findings are right in line with our's here at BOXXlabs.

I generally agree that for 3dsMAX users, modern Geforce cards are probably a "safe" option.

But I would like to remind folks of a few things to remember...

-Remember that this (nicely done) comparison is limited in scope to 3DSMAX, and isn't meant to be used as a general comparison between Geforce and Quadro when using other apps such as Maya or XSI. There are confirmed viewport and menu "issues" for these two apps specifically under Geforce.

-NVIDIA/PNY have NEVER positioned Quadro as being any faster than Geforce cards -only as a more refined and better option for pro artists, designers and engineers.

-QUADRO cards still provide custom application-specific drivers and Geforece cards don't. I havn't heard anything from NVIDIA or PNY stating otherwise.

-SPECViewperf uses ancient versions of all the apps' viewport engines and can't necesarilly be compared to today's verions. This said, Viewperf is a perfectly acceptible benchmark utility for determining relative performance between different machines or GPUs.

Nice work Kris!

Adam
BOXXlabs

sykosys

07-09-2009, 04:02 PM

Thanks Adam. I agree whole-heartedly about not being able to have comparisons on Maya and XSI. I suppose I could have downloaded the trials had I thought about it. If I didn't have to re-do all the tests in order to get a fair comparison, I'd do that now.

As far as refinement, I certainly agree that there is a refinement to the cards, and generally, there is better quality. However, when one considers the premium paid, it's just not something I think most can justify. If a Quadro card entered into the market at the $1200 price point, it would be far more compelling. (And I think nVidia would sell far more of them.) Beyond which, it appears that there will no longer be custom max drivers (based on what I able to find, reported), which really decreases the value of these cards in the Max environment; even when there was, the Maxtreme driver was always very slow to be released, and often buggy... certainly not a premium product.

Kris.

okmijun

08-01-2009, 07:30 PM

Hey man, thnx on info on newer Quadro cards, personally I finished with quadros within GF4 Ti4200 <> Quadro 780 ;) that were the last gen of amazing quadro speed....any way please try this test

http://rapidshare.com/files/3108339/bmark4.rar.html

unpack and start the old scholl 3dsmax test, FPS is still FPS ;) , if script won't start, please try again, it works tested in any max

THANX!

sykosys

08-02-2009, 04:12 AM

I guess I should have looked into maxscript solutions. Would have made my testing life way easier.

In any event, I ran the tests. On the 5600, 285 and 285SLI. Net result: if you exclude maxstreme drivers (which have been phased out 2010 forward), the 285s are just as fast, or faster, just not in SLI. The caveat is that the comp low for the 285s was abysmal. I don't know if that's a one-off thing, I didn't realize it was there until I'd swapped cards. I've attached the excel sheet, that has in it two graphs that showcase the differences.

It's probably best not to compare the 2010 and 2009 benchmarks too closely - they did not come out the same. Almost all the scenes looked very different (i.e. camera position seemed to have changed, textures not rendering). However, this also means that the performance of 2010 is much worse.

Since I posted this, I ended up using the 5600 again, because I was getting some very flaky behaviour out of the 285s in max. Likely driver-caused, but for the test you requested, I ran the latest drivers (190.38), so I don't know if that's changed. There is one situation that I have found that there is definitely a difference - in Photoshop, using OpenGL, as previously reported. When you're using OpenGL, and move any image window to a non-primary monitor, it goes to crap. This is a known issue, on the adobe knowledgebase. It doesn't seem to be as prevalent on the 5600, and only happens occasionally. Right now, it doesn't really matter as there's not a whole lot of GPU goodness going on; it may matter in the future. It's certainly not a killer issue; just needs to be known.

Also, as another aside to the first post, the Seagate drives: man, they're crap. I've had 5 drives fail out of that array so far. I've only sent in one for RMA to Seagate, but I'm seriously unimpressed. Especially when one failed in my NAS and my workstation at the same time. Looking online, this problem is not uncommon, and, unfortunately, seems to be a problem across many different drives & manufacturers. Good to know my livelihood is sitting on flaky hardware that I can't do anything about... just have lots of spares, I guess!

Kris.
Mafic Studios, Inc.
www.maficstudios.com (http://www.maficstudios.com/)

The4thAggie

08-02-2009, 05:27 AM

ThE_JacO

08-02-2009, 06:17 AM

-Remember that this (nicely done) comparison is limited in scope to 3DSMAX, and isn't meant to be used as a general comparison between Geforce and Quadro when using other apps such as Maya or XSI. There are confirmed viewport and menu "issues" for these two apps specifically under Geforce.
No there aren't really for XSI, none that you wouldn't see on a quadro with the wrong drivers either, and I think I can safely claim I've been using XSI for longer than most people and have used every single version from twister to the recent ones.

The only and sole difference is on a geforce you have many more drivers releases and you have to hit the right one, on quadros the pachidermic testing pace normally makes it a non-issue.
Whether freedom FROM choice is an advantage or not remains up to the user I imagine.

On a related note, I've had some absurd freaken problems with quadros being inconsistent and horribly performing with GLSL and CUDA at work. Ironically enough, I have none of those problems on geforce, and nowadays real time shaders and CUDA are becoming rather important for VFX work.

Can't speak much for Maya as I haven't used it at home on geforce much, only at work on quadros, but again when I was using it on both platforms I never really had any issues.

The last time and app I saw a legitimate reason to justify a quadro for was Houdini 6.5. After that the so called glitches have always been just people trying to stay on the bleeding edge of driver releases or cheaply manufactured cards.

Your other points I agree with though.

sykosys

08-02-2009, 03:04 PM

Your OCZ Cores are 1st generation economy models with a known fault with their controller. With your 2x OCZ RAID 0 you look to get 180MBs max. I've got two OCZ Vertex drives in RAID 0 and they max out 250 MBs at 80% full (or 290 not as an OS drive).

BTW (this (http://www.evga.com/forums/upfiles/559052/5FF0B68C5BFF416AB888D4F4038CB047.jpg)) is a benchmark someone did with an onboard RAID controller with 4 vertex ssds (620MBs+). Your 4x raid 0 test was around 350MBs. Then again, once you got to 6, it was all over. Could be a problem with the junk Jmicron controller since the Vertex uses the "Bigfoot" controller.

You may be right, and I am certainly willing to concede the fact. When I did this test, the vertex drives weren't out yet. The only alternatives were so expensive that they weren't in consideration. If I have time/money, I may do a test with some of the newer drives. With my recent drive failures, I am thinking I may want my mission-critical data on a technology that will be more stable.

Kris.
Mafic Studios, Inc.
www.maficstudios.com (http://www.maficstudios.com)

sykosys

08-13-2009, 05:17 PM

I've found out from NVidia that they will be releasing a 2010 performance driver for the quadros, and it is currently in the final stages of performance tweaking. How long to release is unknown. That should make the quadros faster in max than the GeForce Cards, but not by a significant margin. If you're using this card mainly for max/maya, it probably won't mean much in most cases.

Secondly, new Quadro 800 class cards (3800,4800,5800) will be getting software from Elemental technologies that improves speed in Adobe significantly for encoding (much like the CX), which may be a consideration for some. These are available currently by some resellers; I don't know if exising owners will be receiving the software.

Lastly, I got a Quadro 4800 (for the reason above), and ran benchmarks in max. It averages 10% lower performance than a 5600. Being that all specs are equal, and the 4800 has more cores, I suspect it may be a driver issue; I'm using non-current drivers because of compatibility with a beta I'm on. I'm getting a second 4800 to test SLI performance.

Kris.
Mafic Studios, Inc.
www.maficstudios.com (http://www.maficstudios.com)

ha-dou-ken

08-21-2009, 02:55 PM

First of all, thanks for the real world comparision as I'm in the process of considering which card to get, specifically in Max, as that's what I use almost exclusively.

My pivot for comparision is price point so I'm looking at the Quadro 1800 and GTX 295 as they're both around $450 right now.

This (http://www.gpureview.com/show_cards.php?card1=603&card2=558#) is a good website for comparing sheer numbers between cards. it could prove useful to some. Unfortunately they haven't updated their database to include the 1800 so I compared the 295 with the 1700 since they cost about the same.(1800 is actually cheaper and better performing than the 1700)

Now clearly, from a numbers standpoint, the 295 utterly destroys the 1700 when you compare the two side by side, but one of my biggest issues when working on 26 million poly scenes is stability and viewport accuracy. I want to know what's the most important factor for responsiveness in the viewport. Is it GPU memory...or is it FLOPS?I think this scene is a good benchmark and will bring any card to it's knees and right now the 1700 in my work machine is chugging at about 2fps when I tumble slowly in perspective view.it also has crazy shading errors. I guess I'm trying to predict the average fps jump I'd get by going to a 295. I'm also worried that if the quadro is intended to be more "stable" than geforce while I'm getting crazy viewport quirks, then does that mean that I can expect even more ensuing craziness when switching to a "less stable"geforce series card?

Do you think there's any advantage to getting a 1800 over a gtx 295? Again, I know the gtx is a faster card, but since the're basically the same price point, I want to know if the quadro has any justification.

Sorry for the long post.Thanks for the great community service!

aglick

08-21-2009, 03:04 PM

For you, a Geforce card will probably provide better bang/buck.

Do note that in 3dsMAX there is no appreciable difference difference between the Geforce 285 and Geforce 295. (3dsMAX cant make use of SLI to improve viewport performance)

I've heard good feedback from 3dsMAX users running modern Geforce cards with the drivers and driver settings.

sykosys

08-21-2009, 03:13 PM

I think the biggest issue around accuracy is drivers. Without the performance driver, I really don't think there is much of a difference. After having watched all these benchmarks, I can comfortably say that the GeForce Cards look better most of the time. That being said, I have not tested anything with a polycount that high.

I just got my second FX4800, and it performs better than a single, but not by much. I haven't tried SLI, but that's only useful if you use a single monitor. (I'm having the missing SLI option issue ATM.)

But if you're comparing between the relatively low end 1800 and a 285, I don't see how there could be a comparison when the 285 flogs a 5600 or 4800.

Looking at the 1700/1800, the specs are roughly doubled. Somewhere in the neighbourhood of the Quadro 4500 I have. (Which I'll be selling on ebay in the next week or so. :) )

I'm still waiting for NVidia to release those aforementioned performance drivers, and will report back results when they do.

Kris.

ha-dou-ken

08-21-2009, 03:35 PM

That's interesting about the 285 and 295. I see here (http://www.gpureview.com/show_cards.php?card1=603&card2=605) in the comparison that the 295 is more buff, but you're saying that in max there's not much difference? I honeslty don't care about gaming anymore since I'm a retro gamer so if the 295 is just more awesome for gaming then I could care less and will go with a 285.

Yeah, there was a BIG jump between 1700 and 1800. my coworker here has one.But it looks like 285 is the way to go right now if 295 is really not that much better in max.

I hear that ATI is coming out with DX11 capable cards to coincide with the release of Windows 7 when it will release the internal fury of DX11 features of the OS. This is also something I've been thinking about.I just upgraded to Windows 7 and don't plan on using xp anymore so I'd like to have that DX11. Should I wait another 2 months for a DX11 card? I want this card to last me for about 2 years.I just don't want to miss the opportunity of being able to utilize DX11 features in MAX 21...whatever. then again, I don't even know if autodesk will support DX 11 in their next couple versions.

sykosys

08-21-2009, 07:14 PM

I wouldn't go purchasing a card based on DirectX; max still only uses DX9...

Kris.

ha-dou-ken

08-21-2009, 07:47 PM

That's not true. There's a DX10 option in there under the driver setup settings. I'm using 2009

ha-dou-ken

08-21-2009, 07:59 PM

Yeah, I've been reading that it's BAD.I can't do anything with it. I'm hoping all their efforts will go towards 11 and they'll just skip 10 support...like vista to 7.

sykosys

08-21-2009, 08:00 PM

I stand corrected - it's there, but it's not really a useful driver. From what I've been told, it's only there for those in games to test DX10 shaders. As an every-day viewport engine, it's painfully slow. I'd done a few benchmarks when I had Vista installed (I'm using X64 now), and it was so bad, I didn't do all the tests.

Kris.

imashination

08-22-2009, 08:35 AM

The 295 is just 2x 280's stuck together. 3D apps dont use multiple gfx cards for the display, so half of the 295 card would sit there idle.

In other words, take the 285 and save some money.

sykosys

08-22-2009, 03:17 PM

The 295 is just 2x 280's stuck together. 3D apps dont use multiple gfx cards for the display, so half of the 295 card would sit there idle.

In other words, take the 285 and save some money.

The caveat is while the software doesn't use both cards, having the second card for a second display improves the performance of the first. Not in a massive way, but 10-15% - though this doesn't justify a 295, as the 285 is at least 10% faster than a 280. If you want the extra edge in the dual display situation, you're far better off with two 285s. Which should rock games nicely, as an added benefit.

Kris.

CGTalk Moderation

08-22-2009, 03:17 PM

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.