Choosing a Gaming CPU: Single + Multi-GPU at 1440p, April 2013
by Ian Cutress on May 8, 2013 10:00 AM ESTCivilization V
A game that has plagued my testing over the past twelve months is Civilization V. Being on the older 12.3 Catalyst drivers were somewhat of a nightmare, giving no scaling, and as a result I dropped it from my test suite after only a couple of reviews. With the later drivers used for this review, the situation has improved but only slightly, as you will see below. Civilization V seems to run into a scaling bottleneck very early on, and any additional GPU allocation only causes worse performance.
Our Civilization V testing uses Ryan’s GPU benchmark test all wrapped up in a neat batch file. We test at 1440p, and report the average frame rate of a 5 minute test.
One 7970
Civilization V is the first game where we see a gap when comparing processor families. A big part of what makes Civ5 perform at the best rates seems to be PCIe 3.0, followed by CPU performance – our PCIe 2.0 Intel processors are a little behind the PCIe 3.0 models. By virtue of not having a PCIe 3.0 AMD motherboard in for testing, the bad rap falls on AMD until PCIe 3.0 becomes part of their main game.
Two 7970s
The power of PCIe 3.0 is more apparent with two 7970 GPUs, however it is worth noting that only processors such as the i5-2500K and above have actually improved their performance with the second GPU. Everything else stays relatively similar.
Three 7970s
More cores and PCIe 3.0 are winners here, but no GPU configuration has scaled above two GPUs.
Four 7970s
Again, no scaling.
One 580
While the top end Intel processors again take the lead, an interesting point is that now we have all PCIe 2.0 values for comparison, the non-hyper threaded 2500K takes the top spot, 10% higher than the FX-8350.
Two 580s
We have another Intel/AMD split, by virtue of the fact that none of the AMD processors scaled above the first GPU. On the Intel side, you need at least an i5-2500K to see scaling, similar to what we saw with the 7970s.
Civilization V conclusion
Intel processors are the clear winner here, though not one stands out over the other. Having PCIe 3.0 seems to be the positive point for Civilization V, but in most cases scaling is still out of the window unless you have a monster machine under your belt.
Facebook
Twitter
RSS
242 Comments
View All Comments
Spunjji - Wednesday, May 8, 2013 - link
Crap troll is crap.lorribot - Wednesday, May 8, 2013 - link
Would love to see something like a E3-1230 tested, it is around the same price as a i5-3570K but has no graphics, bigger cache and Hyper threading, but no over clocking and 100MHz lower clock. should be similar to a i7-3770 for around 60% of the price.Spunjji - Wednesday, May 8, 2013 - link
So let me get this straight. The engineers are idiots, yet you want them to go and work for other companies, the best candidate there being Intel. Also, thanks to this apparent mental handicap, they use Intel processors... oh, I get it, you're pro AMD!bikerbass77 - Wednesday, May 8, 2013 - link
Just a note as I keep seeing posts going on about Planetside 2 being CPU limited. It is not. I am saying this from experience having played it just fine on a Core2Duo based system. The reason you are most likely having problems will either be your ping or your GPU. I was running with a GTX 460 1gb card and it was fine for that particular game. I have just upgraded to a new CPU because the main game I play (Mechwarrior Online) is far more CPU bound being based on CryEngine 3.cusideabelincoln - Wednesday, May 8, 2013 - link
I think it should be noted that online multiplayer games are a different beast. Multiplayer is typically more CPU intensive, and if you're looking to maintain completely smooth gameplay without any dips in framerate or stuttering then the CPU becomes more important than it is for single player gaming.Also would you consider benchmarking live, online streaming of games? Would be great to see how much of a benefit the 3930K would have over other chips, and if Piledriver can pull ahead of the i5s definitively.
Markus_Antonius - Wednesday, May 8, 2013 - link
Your sample size is statistically beyond irrelevant which prevents the scientist in me from drawing any conclusions from it. In addition, claiming any sort of causal relationship between results is outright scientifically wrong even if the sample size would be statistically relevant. From an engineering standpoint the X79 systems with ample headroom in every relevant department would be the best choice to avoid any possible bottlenecks / contention issues in the largest possible number of different workloads.Any recent system with a recent CPU and recent midrange graphics card can play a game and can often play it well. Advising a Core i7 3770K based on a statistically irrelevant benchmark while disregarding systems architecture is something that neither the scientist, the software engineer and the hobbyist in me can get behind in any way.
JarredWalton - Wednesday, May 8, 2013 - link
Hyperbole, you have a new friend: meet Markus! "Beyond irrelevant", "any conclusions", "outright scientifically wrong", "ample headroom", "every relevant department", "best choice"....Let me guess: you have an X79 system, and it works great for you, and thus anyone even suggesting that it might not be the best thing since sliced bread is something you can't even think about in any way. This article is focused on gaming, and if you want to do things besides gaming yes, you will need to consider other facets of the system build. At the same time, if all you're looking for is a good gaming setup, perhaps with two or three GPUs, I have trouble imagining anyone recommending something other than i7-3770K right now (unless the recommendation is to "wait for Haswell").
Let me give you a few things to consider that, while the scientist may not necessarily agree, the software engineer and hobbyist definitely would avoid SNB-E and workstations. 1) Overall power requirements (they still matter). 2) Quick Sync (may not be perfect quality, but dang it's fast). 3) Better performance in many games with two GPUs, no matter what paper specs and system architectures might say.
smuff3758 - Thursday, May 9, 2013 - link
And that, Jared, is how to shut down this arrogant, condescending self-titled expert/scientist. I guess he must think the rest of us are bozos who come here for the comic relief?Markus_Antonius - Sunday, May 12, 2013 - link
My comment was about the testing method not being scientifically sound even though the author makes it a point to refer to the "well-adjusted scientist" in himself. There's a huge number of games out there as well as a lot of different mid-range to high-end video cards. Recommending an i7 3770K on the basis of one resolution tested and only 4 games is something that you absolutely cannot call science.I am among other responsibilities a software engineer and I don't actively avoid Sandy Bridge E and workstations.
My criticism of the methods used and the conclusions drawn is valid criticism, especially in the face of the article being given the appearance of being science.
If you're going to do recommendations based on statistics and for whatever reason decide to disregard engineering and the science behind systems design you're going to need a far larger sample size than what was used here.
You can deflect this all you want by quoting power usage and quicksync but while power usage power usage should be a factor, this test was not about quicksync. If it had been they would not have tested X79 systems at all ;-)
From both work and hobby I know a lot of power users and gaming remains one of the most demanding uses and one of the *most prevalent* demanding uses of a modern PC. Throwing a more powerful system aside and disregarding engineering needs to be done with a lot more care and thoroughness, all of which is missing here.
Answering valid criticism with scorn and aggression is also very telling. Perhaps you're more insecure than you thought you were?
Badelhas - Wednesday, May 8, 2013 - link
Great review, congrats!This comes at a perfect time for me, I just ordered a Qnix QX2710 1440p 27 inch monitor from ebay and a couple of 670´s to work along with my 2500k OCed to 4.5Ghz. It seems I will be amazed with that upgrade, lets see!
Cheers