Sign in with
Sign up | Sign in
Your question
Closed

Value In SLI? GTX 260 Core 216 Vs. GTX 280

Tags:
  • Gtx
  • Core
  • SLI
Last response: in Reviews comments
Share
January 5, 2009 8:50:43 AM

We recently built a $2,500 gaming system with three GTX 260 Core 216 graphics cards, noting that these were less expensive than two GTX 280s. In the performance-value game, is 3-way SLI in the cards?

Value In SLI? GTX 260 Core 216 Vs. GTX 280 : Read more

More about : sli gtx 260 core 216 gtx 280

January 5, 2009 9:44:17 AM

....so 2 GTX 280s generally out perform 3 GTX 260s and you count it as a win because its best in Crysis? Also 2 4870 X2s perform a decent amount above 2 GTX 280s and the extra funds could have been relocated. While I did not mind the last build I do see their point. You have 2 options, you focus on gaming performance even though a $2000+ system should do more, or you focus on a balanced system but at the detriment to gaming. Neither option is truly wrong so its up to the builder and their expectations. I think the builders of that system chose what they wanted to focus on and thats it.

I thought that the only problem was that one of the builds performed less than the last one at the same price point. But then again I am used to the comments section of such articles being full of whinners and idiots so its not much of a surprise...
Score
5
January 5, 2009 10:17:42 AM

How about GTX 260 in SLI versus GTX 260 core 216 in SLI,versus a single GTX 280? lol,that'd be single sided.
Score
-8
Related resources
January 5, 2009 10:21:04 AM

interesting to see two be at par with 3! and maybe compare them to 4870x2 to throw into the mix
Score
2
January 5, 2009 11:05:38 AM

I would choose 280 SLI over 260 TRI SLI anyday even if it was cheaper than the 280 SLI. 1 less card to deal with and less chance for driver problems.

285 SLI is looking interesting
Score
1
January 5, 2009 11:09:47 AM

I think this reinforces the idea that if you're thinking of SLI, don't, unless you're getting all the cards now. If you're getting just one card now and waiting until you can afford the next one(s), your money might be better spent on just waiting for the best of the next generation.
Score
8
January 5, 2009 11:27:55 AM

I agree with others, it would benefit the community if a single and pair of 4870x2s were used in this review as well.
This was a good review, it pointed out to us that buying 3 260s is not so beneficial, Crysis is a great game, but it is not the deciding factor for the purchase of my next video card.
Score
8
January 5, 2009 11:49:24 AM

2 really hot girls or 3 above average girls....LOL.

If I really want SLI I would choose as fewer cards as possible. The driver mess and inefficient scaling makes no sense to get more than two cards. GTX 295 > GTX 280 x2 > GTX 260 x3 > 9800GT x4.
Score
8
January 5, 2009 12:06:13 PM

The_Blood_Raven....Also 2 4870 X2s perform a decent amount above 2 GTX 280s and the extra funds could have been relocated....

A 4870 X2 costs around $500. A GTX 280 $330. $500 x 2 = $1000 > $330 x 2 = $660 so what extra funds are you talking about?
Score
5
January 5, 2009 12:44:21 PM

I sortof dislike how much people depend on Crysis as a benchmark. Despite it's awesome gameplay and graphics, there has NEVER, EVER, been a more poorly optimized game. It seems a crap shoot on what sortof performance even high end graphics card will put out when using it...

Better to notice the fact the 280x2 consistently beats the 260x3 except in the one game that sets the standards for erratic performance results.
Score
3
January 5, 2009 12:53:21 PM

Well I would just like to add..... VALUE SHMALUE!! SPEND SPEND SPEND!!! As long as I have enough money left over to make some katsup soup and play all my games on full rez I'm happy.
Score
-1
January 5, 2009 1:00:14 PM

I think this article adresses the criticism from the 2500$ build perfectly. The arguments against trip SLI were all adressed, and even people contradicting themselves got answers that make sense.

People are always gonna critisize choices made in these articles, and make assumptions grabbed out of thin air without any form of proof or facts to back up their statements, imo this article and the 2500$ build made some really interesting and compelling points and pointed out the advantages and disadvantages of both.

Of course I would love to see a 2 x 260 SLI vs. 3 x 260 SLI to see how much that last card really gives, since that's what we are really talking about how much a third GPU really stacks with the others. But since THG already has the GPU charts (http://www.tomshardware.com/charts/graphics-cards,1.htm...) which I assume/hope is getting a updated very soon based on a Core I7 machine, all these speculations will basically be answered by this.

This article is about a 2500$ build and what solution is best for this, and as seen in this article the best decision when going for playable framerates is trip SLI core 216 260's, not dual 280's which only boost your already enormous frame rates on other games.
Score
5
January 5, 2009 1:05:13 PM

AKandyWell I would just like to add..... VALUE SHMALUE!! SPEND SPEND SPEND!!! As long as I have enough money left over to make some ketchup soup and play all my games on full rez I'm happy.

Score
-10
January 5, 2009 1:29:48 PM

yawwnnn.. calm before the storm. (phenom II)
Score
-6
January 5, 2009 1:49:40 PM

it is a great review and all, and I completely agree with the conclusion, but I would've liked to see all the cards in this review, SLI or not. For example, 1 x GTX 280 vs 1 x GTX 260 vs 2 GTX 280 vs 2 x GTX 260 vs 3 x GTX 260 vs 2 x HD4870x2 etc ... In this way we would figure out all the high end GPU's benchmark in one, instead of stitching them all together. Sigh .. now we wait for the next review..
Score
7
January 5, 2009 2:29:20 PM

Vixeit is a great review and all, and I completely agree with the conclusion, but I would've liked to see all the cards in this review, SLI or not. For example, 1 x GTX 280 vs 1 x GTX 260 vs 2 GTX 280 vs 2 x GTX 260 vs 3 x GTX 260 vs 2 x HD4870x2 etc ... In this way we would figure out all the high end GPU's benchmark in one, instead of stitching them all together. Sigh .. now we wait for the next review..


That would be interesting. I know the scaling improved with the i7 for SLI but the comparison Vixe is talking about would be very telling.
Score
1
January 5, 2009 2:34:41 PM

PrangeWayI sortof dislike how much people depend on Crysis as a benchmark. Despite it's awesome gameplay and graphics, there has NEVER, EVER, been a more poorly optimized game. It seems a crap shoot on what sortof performance even high end graphics card will put out when using it...Better to notice the fact the 280x2 consistently beats the 260x3 except in the one game that sets the standards for erratic performance results.


Well. It's one of those few well done games that actually benefit from multible GPU's and multible CPU's. I have a high hope from the Futuremark Games Studios Shattered Horison http://www.shatteredhorizon.com/ because they have allso known how to utilice multible cores...
And yoeh you are right that Crysis is not so good game, but as I mentioned, it is one of those few that scale with multicores guite well, so I think that that is the main reason why it reamains as one of the standard test games. I hope that in the future there will be much more games that can really use all that power that you have in your computer (read scale well). It does not make a good or bad, but tells asomething about the programing.
Score
0
January 5, 2009 2:47:55 PM

I really dont have a dog in this fight... but to get the best comparison of cost/performance of 2-way vs 3-way SLI/Crossfire, I don't think the review should have been with 2 ea 4870 X2 or 2 ea GTX 280s... If you really want to see how the performance scales based on nothing but how many cards are used in an SLI configuration, it should be done with 2 ea and 3 ea of the exact same card.

I know, I know... that doesnt give you a good idea of the performance you can expect for the same dollar spent on a 2-way vs a 3-way. And I understand that the point of this article was to deal with that question posed by the readers and thier comments. Based on that I say...

You should be comparing one of two different scenarios...

1) Performance level per dollar spent. i.e. Spend the exact same (or as close as possible) amount on a 2-way and a 3-way SLI or Crossfire setup. that give you the most performance for your purchase and compare them. For this comparison it does not make any difference if the cards have 2 GPUs or 1 GPU. It is about the performance you can squeeze out of 2 Cards vs 3 Cards... not out of 2 GPUs vs 3 GPUs.

2) Cost/Performance difference in 2-way vs 3-way SLI. This may sound like the same thing but it is not. This does not compare the best 2-way you can buy for the same amount as a 3-way. This is taking 3 identical cards. Test performance with all three cards. Test performance with only two cards... and if you really want to do the world a favor, test it with only one card as well. What do you get? A scale that show the amount of performance increase as it scales in proportion to the cost of each card.

Just my $.02,
TechDicky
Score
6
January 5, 2009 2:48:32 PM

A software redesign for your charts will be very helpfull. Instead of having separate charts for CPU, Grafix and Motherboard what enthusiasts like to see is the perfromance of various combinations. It could be easily done with first field letting us choose cpu, second Grafix,third Motherboard and fourth Benchmark. It becomes a huge task, but sections could be added part by part and ultimately it wold make a great place to virtually build and test a system.
Score
0
January 5, 2009 2:55:19 PM

The_Blood_Raven....so 2 GTX 280s generally out perform 3 GTX 260s and you count it as a win because its best in Crysis?


Which win matters in game play? You could have 100 games that play at 90FPS on a cheaper setup or 100FPS on the more expensive one, and the difference would be inconsequential since both of them play smoothly. Same goes for comparing RTS games, which don't need to be perfectly smooth, and FPS games, were any lag gets you fragged. It's always the FPS games that fall below 40FPS that take precidence.
Score
2
Anonymous
January 5, 2009 3:13:55 PM

"Same goes for comparing RTS games, which don't need to be perfectly smooth"

That kind of comment shows people that don't take RTS seriously... having a stutter of 0.1 second is enough to break havoc on any carefully planned strategy during an RTS game.

Any "good" RTS player will need to execute between 5 to 10 actions per second (300-600 APM) during the key moments of high. And a stutter of 0.1 or 0.2 seconds those conditions means a dead army. Meaning Frames should NEVER EVER drop below 30 FPS, the same applies to First person Shooters.


Score
2
January 5, 2009 3:31:41 PM

I wonder if the benchmarks would be different if they used a 64 bit os, rather than a 32 bit os. Most of the ram/vidoe ram would not have even been seen during the benchmarks.
Score
0
January 5, 2009 4:24:08 PM

RTS-maniac 01/05/2009 6:13 AM

"Same goes for comparing RTS games, which don't need to be perfectly smooth"

That kind of comment shows people that don't take RTS seriously... having a stutter of 0.1 second is enough to break havoc on any carefully planned strategy during an RTS game.

Any "good" RTS player will need to execute between 5 to 10 actions per second (300-600 APM) during the key moments of high. And a stutter of 0.1 or 0.2 seconds those conditions means a dead army. Meaning Frames should NEVER EVER drop below 30 FPS, the same applies to First person Shooters.
-------------------------------------------------
Very true, on my old pc lag in AOE3 got me creamed a couple of times, lag in Supreme commander delayed victory by over an hour.
Lag in RTS and makes playing them painfull.

And I also agree with "kubes" a 64bit os would be nice to see
Score
1
Anonymous
January 5, 2009 5:03:37 PM

Add to it the amount of energy (electricity bill) one has to pay from 3 cards, and in the end you're better off buying 2 cards.
In 2 months the 2 cards pay themself back in electric bills.
Score
1
Anonymous
January 5, 2009 5:10:50 PM

PrangeWayI sortof I think, that crysis is not a good game to review indeed.
but not because of the 'crappy drivers' or anything.
Have you seen the amount of detail it has?
Or seen the resolutions atr which they play them?
Crysis can be run with an older card as well,on lower resolutions.
However, I would also prefer to see tomshardware aim for more beneficial benchmarks.
I just simply can't afford the hardware to run crysis, so I don't even buy it.
Crysis results really tellme nothing.
It would be nicer if some older games where included... Like prince of persia? or other games that I do play on my older hardware.dislike how much people depend on Crysis as a benchmark. Despite it's awesome gameplay and graphics, there has NEVER, EVER, been a more poorly optimized game. It seems a crap shoot on what sortof performance even high end graphics card will put out when using it...Better to notice the fact the 280x2 consistently beats the 260x3 except in the one game that sets the standards for erratic performance results.

Score
0
Anonymous
January 5, 2009 5:14:41 PM

And man,I don't know what's wrong with these tomshardware pages.
They know of hardware, but their website really sucks!

Too many ads, and in firefox the commentbox works more not then it does; and I can't enjoy this page when disabling flash and java!

Why can't they just make this webpage non-java compatible, and get rid of all these moving annoying banners? And just stick with the passive banners?
Of all the commercial banners I hate the most are the ones that popup in the middle of the screen; often not even directing to a computer or hardware related page, and the banners that just flip over the text!
those are the most annoying!
Only for those I wish not to visit this site; and especially not click those banners.
Score
0
January 5, 2009 5:16:34 PM

PrangeWay said:
I sortof dislike how much people depend on Crysis as a benchmark. Despite it's awesome gameplay and graphics, there has NEVER, EVER, been a more poorly optimized game. It seems a crap shoot on what sortof performance even high end graphics card will put out when using it...

Better to notice the fact the 280x2 consistently beats the 260x3 except in the one game that sets the standards for erratic performance results.



It's an unfortuate artifact of using the "wrong games" for the SBM that Crysis was the only game to benefit. I'd expect to see similar gains in other gpu-handicapped FPS titles like Far Cry 2.
Score
0
January 5, 2009 5:39:27 PM

My Strategy at this point: Forget about the games I would love to play now and wait until a single mainstream card can handle Crysis at full rez with AA enabled @ or above 60fps. Gaming doesn't pay the bills for me and waiting for such an occurence will ensure I don't get screwed $-wise. Sacrifices oftentimes reward you with equal or greater payback. 'Course I'm still sitting @ home with a Pentium 4 @ 1.8G and a FX-5200 card playing SC: BW and Fallout 1 & 2.
Score
1
January 5, 2009 5:47:35 PM

BrueBixMy Strategy at this point: Forget about the games I would love to play now and wait until a single mainstream card can handle Crysis at full rez with AA enabled @ or above 60fps. Gaming doesn't pay the bills for me and waiting for such an occurence will ensure I don't get screwed $-wise. Sacrifices oftentimes reward you with equal or greater payback. 'Course I'm still sitting @ home with a Pentium 4 @ 1.8G and a FX-5200 card playing SC: BW and Fallout 1 & 2.

You can't find a mainstream system that can play SC: BW and Fallout 1 at full resolution and all eye candies?
Score
1
January 5, 2009 6:12:39 PM

I would like to see this test performed with 3x2GB RAM and a 64-bit OS. Putting together a system this high-end and using anything less is unrealistic in my opinion.

Other than that, good test, interesting results.
Score
0
January 5, 2009 7:12:46 PM

3 cards also produce more heat to the system, also they overclock less cause of the heat and the chance of having a poor overclocking card increases to 50% over 2way sli + u re gonna have less compatibility than 2 cards. But i think u should have included more games in the test. Games in the bench are old. I also think that in World In Conflict the victory is for the 3way-sli, anyone with this cards is going to activate 4x AA and there 3way is 5% + faster.
The more variables u include in the ecuation the more accurate the result is gonna be.
I would love to see gtx 260, gtx 260 core 216, gtx 280 + new gtx 285 and 295 in 2 way, 3 way and quad sli. It has been a long time since we dont have a review like that, since gtx 280/260 launch.
Score
-1
January 5, 2009 7:56:37 PM

Pei-chenA 4870 X2 costs around $500. A GTX 280 $330. $500 x 2 = $1000 > $330 x 2 = $660 so what extra funds are you talking about?


Sorry it was poorly worded. I meant the extra funds required for the 4870 X2 crossfire could relocated from other components.
Score
0
January 5, 2009 8:03:53 PM

CrashmanWhich win matters in game play? You could have 100 games that play at 90FPS on a cheaper setup or 100FPS on the more expensive one, and the difference would be inconsequential since both of them play smoothly. Same goes for comparing RTS games, which don't need to be perfectly smooth, and FPS games, were any lag gets you fragged. It's always the FPS games that fall below 40FPS that take precidence.


Which win matters? When 1 setup out perfoms the other in everything except 1 extremely nvidia optimized game then I would say that the setup that loses in Crysis is the winner. While you are very right in that the difference between the 2 setups can only be now seen in Crysis, the important word is NOW. You want the setup that will perform the best in years to come. No one is going to spend $2500 and not think of the future. The inherent better performance, easier and more reliable driver support, and less heat makes the dual SLI setup far superior in future titles.
Score
1
January 5, 2009 8:08:06 PM

I would also like to know why you're using a 32-bit OS instead of 64-bit. Seems kind of a no-brainer to just go ahead and use 64-bit, doesn't it?
Score
0
January 5, 2009 8:44:55 PM

sebojI would also like to know why you're using a 32-bit OS instead of 64-bit. Seems kind of a no-brainer to just go ahead and use 64-bit, doesn't it?


They used a 32-bit in the SBM, so they are recreating it here as well.
Score
1
January 5, 2009 8:47:16 PM

Lurker87They used a 32-bit in the SBM, so they are recreating it here as well.


I know that - it was more of an overall question, for all their testing situations.
Score
0
January 5, 2009 8:58:17 PM

sebojI know that - it was more of an overall question, for all their testing situations.


The same point was brought up ad nauseum in the SBM. It was because they couldn't get some of the benchmark programs to work on the 64-bit OS, IIRC. Many had the same problem with toms not going 64-bit, just for benchmarking, but it was their call, in the end.
Score
2
January 5, 2009 9:05:03 PM

If I was forced to choose between these two setups I would choose three GTX260s over two GTX280s.
This is because as a gamer, I always tune my graphics so that I am getting the best Performance to visuals ratio, which for me is around 60FPS avarage for any game. When looking through the results, It seems where the framerates land in that sweetspot of playability and performance, the GTX 260s win.
For Example if I go through the results and take the best playable Frame Rates and compare them directly: (GTX260s results shown first).

Crysis 1680x1050 no AA = 66.9 vs 52.1 = +28.4%
Unreal 2560x1600 4x AA = 115.5 vs 97.9 = +18.0%
SupCom 1680x1050 no AA = 55.3 vs 63.2 = -12.5%
W.I.C. 2560x1600 4x AA = 66 vs 58 = +13.8%

It would seem that looking at it from this perspective in specifically the gaming benchmarks, the three GTX 260s are on avarage 12% faster.

I would also like to see unreal with a higher level of AA, as from the trend I'm seeing, I find it likley the GTX260s would gain a further lead over the GTX 280s. (Although it could be argued the 260s would hit a memory limit if you cranked the AA too high, but it may be the case the frame rates would then be unplayable)

Anyway, I'd sooner buy just a pair of GTX260 216s, as looking at SLI vs TRI SLI comparisons on other sites, the returns from the extra card is not justifiable for the added cost (note that a more expencive mobo would be required for tri sli too).
Score
2
January 5, 2009 10:59:24 PM

I think this comparison is fundamentally flawed. However, I could be basing this on some wrong thinking so bear with me. Please let me know if I'm mistaken.

I believe that the motherboard used, the EVGA 132-BL-E758-A1 LGA, is the reason these results are so close.

The PCI-E slots on this mobo run at 1 x PCIe x16, 1 x PCIe x8/x16, and 1 x PCIe x8. I believe this means the 1st slot runs at x16, the 2nd slot can run at 16x or 8x, and the third slot runs at 8x.

The third slot cannot run at 16x.

If in Tri-SLI, on an interface that has a maximum bandwidth of 8x, I believe all 3 cards may be throttled down to 8x.

So what I think is happening, is we have 3 gtx260's running at 8x. VS. 2 gtx280's running at 16x. This is hamstringing the bandwidth of the 260's and they're not actually performing as they could.

AT BEST, we have 3 gtx260's running at 16x 16x 8x VS. 2 gtx280's at 16x 16x.
In this best-case scenario, the 3rd gtx260 is still only at half bandwidth, so it's true possible performance gain can not be realized.

I would propose that this same contest is run on the ASUS P6T6 WS Revolution LGA 1366 Intel X58 ATX Intel Motherboard.

This motherboard is capable of running 3 PCI-E 2.0 interfaces at true 16x speeds. I would put my money on the 3 gtx260's consistently over the 2 gtx280's on this motherboard, that will actually let them run flat out.

-Donald
Score
3
January 6, 2009 12:36:06 AM

Again, can some one please run BadaBOOM on a tri-SLI set up?
Score
0
January 6, 2009 1:06:42 AM

This is definitely a good follow-up on the comments from the $2500 build. Thanks for your efforts, Tom's.
Score
1
Anonymous
January 6, 2009 1:17:04 AM

you guys really need to keep the colors that you choose for each set-up the same in the graphs through out the article. It makes it very confusing to read and interpolate the data easily
Score
2
January 6, 2009 1:19:17 AM

To be honest I'm not exactly excited about either result. One high end videocard would be enough performance to max out any game except for Crysis and still get decent fps.

With the money saved on going with one high end card you will have enough to buy the next generation high end card that will beat your now aged SLI/Crossfire setup anyways.
Score
3
January 6, 2009 1:35:17 AM

you rly need to fix the light gray text for the good of my old eye's! just something that I have not seen maybe do a comparision on tri 4870's see how it fiars :p 
Score
0
January 6, 2009 1:43:57 AM

sebojI would also like to know why you're using a 32-bit OS instead of 64-bit. Seems kind of a no-brainer to just go ahead and use 64-bit, doesn't it?


Just an FYI, all testing in the Tom's Hardware US lab is being transitioned to 64-bit Vista. We hear your feedback and are responding accordingly.

Just an interesting aside, though--Windows XP x32 remains the favorite for overclocking. We even received the heads-up from AMD not to expect Phenom II to overclock as aggressively under a 64-bit OS. Remember that when our Phenom II vs. i7 overclocking comparison launches.

Just something to keep in mind for the enthusiasts clamoring for a native 64-bit environment with 32-bit tests ;) 

*duck*
Score
3
January 6, 2009 4:02:10 AM

oh good! now that's an article looking forward to. :D 
Score
0
January 6, 2009 8:08:39 AM

ginbong46I would choose 280 SLI over 260 TRI SLI anyday even if it was cheaper than the 280 SLI. 1 less card to deal with and less chance for driver problems.285 SLI is looking interesting

Definetly so would I. Two cards mean they'd work in any c2d nvidia system and not only in a completely new platform with still unknown quirks, and also most quality power supplies may have enough juice to power two 280s or three 260s but won't nessecarily actually carry enough connectors to power the last of the three 260s

Anyway, on the bottom line I'd very likely go for two 4870 x2's instead. I'm very much inclined to picking nvidia cards for their better drivers, but I just can't ignore the potential of the ati cards.
Score
0
January 6, 2009 8:59:43 AM

A cuda benchmark would have been nice too for application like TEMPGenc Xpress or Folding to know power calculation between 3 Sli and 2.
Score
0
January 6, 2009 9:39:11 AM

Far as I remember, cuda doesn't support sli? they'd treat it like 2 or 3 individual gpu's - so the performance would simply be a matter of multiplying the performance of one gpu.
Score
0
January 6, 2009 10:46:36 AM

One thing I don't see mentioned with regards to 32bit or 64bit is the affect the amount of video memory would have on usable memory. I would think SLI on 32bit (i.e. 3 x 512mb cards) would cut into the amount of available memory (due to memory addressing limitations) and thus have an overall reduction in the performance of OS. I have not SLI'd so maybe this is not an issue, but I would not think this limitation has been overcome has it?
Score
0
      • 1 / 2
      • 2
      • Newest
!