A little while ago we delivered this bouncing baby boy of an X79 system with a E5-2450 V2 Xeon processor and an RX 480, and it was performing quite well for the £175 it cost to build. I reckoned it would be able to handle an upgrade to a 1080Ti, and seeing as I’ve recently bought one and now have better testing data from recent videos, we can put that theory to the test. You don’t have to join me in this endeavour, but it will be marked on your report card if you don’t.
Welcome to Esk Computers, I’m Scott and today on the desk of Esk, let’s see if this build can handle a GPU that costs as much as the rest of the system put together.
Nothing too exciting to show for the setup here, a quick run of the Display Driver Uninstall utility to do what it says on its tin, a swap of GPU, and installation of the NVidia drivers. I’ll run through the usual battery of testing, and compare the results with the RX480, and as we’ve also got data from the pairing of this Xeon with the 7800XT, we can see how close we’re getting to its limits. If the 1080Ti’s performing significantly worse than we saw the testing with the newer Xeon, we can point that out too.
A link to the build video should be popping up now if you want the full system details, but the important parts for today’s experiment are this Qiyida X79, or so it claims, motherboard, 16GB of 1600MHz DDR3 RAM and of course the E5 2450 V2 Xeon, an eight core, sixteen thread part from 2013, with a base clock of 2.50 GHz, turboing up to 3.3 GHz. Let’s get cracking.
Running through 3DMark first, and Fire Strike gives us 17543 points, which is a solid 50% better than the RX480 result. Decent, but it’s 18% lower than the 1080Ti gets with the more modern Xeon, the E5-2680 V4.
Time Spy gives us 8987 points, a massive 103% better than the RX480 result of 4421, but 9% lower than the V4 Xeon result.
Rounding off the synthetics, Superposition at 1080p High gives us 11846, another big win of 111% over the RX480 numbers. Still, the 1080Ti can give more, as the 18% drop from the V4 Xeon result shows.
Shadow of the Tomb Raider at 1080p High Settings is sitting at 88 average fps. That’s only 28% better than the RX480, and essentially the same result as the 7800XT. The more powerful V4 Xeon gets another 30-odd percent out of the 1080Ti at 130 fps, so a concerning showing for my “won’t bottleneck it” theory.
The Counter-Strike 2 test results I’ve been gathering for videos lately across multiple systems are… weird – to the point that I no longer really trust it. But, for completeness, here the FPS Heaven benchmark at Very High settings gave 102 fps, nicely between the 480’s 76 fps and the 7800XT’s 120 fps. The V4 Xeon gets 123 fps.
Dropping to Low settings, with dynamic shadows again gives that inexplicable drop in performance, which is what has me questioning either its sanity or mine. Here we see only 94 fps, lower than the 480’s 119 fps. Absolute nonsense. The 7800XT gets 160 fps, and the 1080Ti/V4 Xeon combo got 149 fps, but I’m not putting much faith in any of that at this point.
Returning to Night City, we’ve got the in-built Cyberpunk 2077 benchmark at 1080p Ultra getting 60 fps, a 70% gain on the 480’s 35 fps. The 7800XT gave us a similar 64 fps, so we’re pretty much at the CPU limits for this game. There’s about 20% of additional performance left in the 1080, as the V4 Xeon result of 75 fps proves.
It’s a similar story in actual gameplay on the mean streets of Kubuki, where the 38 fps result is functionally the same as the 7800XT’s 40 fps, with the V4 Xeon again squeaking roughly 30% more out of the 1080Ti at 53 fps.
We descend into Unreal Engine 5 next with the Black Myth Wukong benchmark at 1080p High, without any RT or upscaling, where the 38 fps result is again roughly halfway between the 480’s 25 fps and the 7800XT’s 69 fps. It’s also exactly the same 38 fps result for the V4 Xeon.
Marvel Rivals at 1080p Medium settings, without any RT or upscaling, gives us a similar 35 fps result, again 50-odd percent better than the 480’s 23 fps, but 50% off the 7800XT’s 53 fps. It’s seemingly wasting at least 40% of the 1080Ti’s potential in this game, if the V4 Xeon’s 60 fps is anything to go by.
Assassin’s Creed Mirage at 1080p High returns 70 fps, again 50-odd percent better than the 480’s 46 fps, and within touching distance of the 7800XT’s 79 fps. Still, the V4 Xeon can get 90 fps out of it, so about 22% difference.
While we’re getting a very playable result on Forza Horizon 5 at 1080p High at 73 fps, it’s only a 16% increase on the RX480 numbers. The V4 Xeon can get 102 fps, so that’s a 28% drop from where it could be.
Horizon Zero Dawn at 1080p Favour Quality settings gives us 80 fps, another decent 50%-ish jump from the 480’s 52 fps and reasonably close to the 7800XT’s 91 fps, but again, the V4 Xeon result of 110 fps shows a near 30% performance drop.
Red Dead Redemption 2, also at 1080p Favour Quality settings will give us 64 fps, which is, say it with me now, about 50% better than the 480’s 42 fps. This is also one of the rare occasions where the single Xeon result is much the same, at 66 fps.
Total War: Warhammer III’s Mirrors of Madness benchmark at 1080p Ultra gives a slightly disappointing 27 fps, although this is more of a reflection on the CPU, as shown by the 7800XT’s identical 27 fps result. The V4 Xeon gets 48 fps.
We round out the testing with a wander around Baldur’s Gate 3’s Lower City, where we’re getting much the same story. With this V2 Xeon, the 1080Ti gets 40 fps, the 7800XT gets 41 fps, and there’s about a 30% benefit in going to the V4 Xeon at 55 fps.
Much as I hate to admit it, I was wrong. This processor cannot properly handle the 1080Ti, at least on the bulk of these games tested. We are on average about 20 to 25% down on the performance we can get out of the 1080Ti with a more powerful CPU, and there may be even more potential as there’s no guarantee those V4 Xeon results are the absolute best the 1080Ti can deliver.
Obviously, it’s able to deliver a sizable jump up from the RX480, fifty to sixty percent better performance, but you’ll have to judge for yourself whether that jump in performance is worth paying about five times the price.
I’d say not, unless you intend on keeping the 1080Ti around for a future platform upgrade or just see a particularly good deal on one. Under normal circumstances, you’d be better off moving a bit down the stack. If you’re looking to keep the CPU and GPU in perfect harmony, I’d guess a 1070 or a 1660 would fit the bill, or maybe one of the AMD Vega cards which I’d really love to test sometime.
Those would cost about half the £160 I paid for this 1080Ti, and should perform roughly the same given the CPU bottleneck, so better value for money, although I still think the winning play for this setup is to keep it super-budget, as the value that £27 8GB RX480 delivers is phenomenal, and somehow still clinging on to relevancy a decade on from its release.
That’s all I have for you now, but if you have any questions or want further details, please leave a comment down below, and if you enjoyed this videotronic missive then consider subscribing. Until next time, take care of yourself, and each other.