An independent benchmark was recently conducted to determine the results of the graphics processing unit performance between the Xbox One and the PlayStation 4. We had previously known in other benchmarks that the PS4 edges out the Xbox One in central processing performance, but who knew that the extra shader units would play such a significant role in destroying the Xbox One like Godzilla stomping through Tokyo?
Of course, the above question is rhetorical. A higher number of anything usually, in theory, means that the comparison should lean in favor of the device that has that higher number. In the case of a theoretical benchmark taken up by WCCF Tech using Daeguen Lee’s proprietary equation system and VGA calculator, they happened to uncover some amazing results.
The PS4′s GPU, codenamed Pitcairn, went head to head with the Xbox One’s GPU, codenamed Bonaire.
Both the home consoles are running on AMD’s Jaguar APU technology, which actually combines the CPU and GPU into one housing. However, despite both systems having AMD technology under the hood, imagine that the PS4 is running a V10 while the Xbox One is running a V8 (PCs are V12 and above, so we all know where the GPCMR stand). The benchmarks below showcase how well the systems performed, with the Xbox One being run through two tests (one with eSRAM and one without eSRAM). Check it out below.
Oh boy, look at that smokage. That’s like watching a Honda Civic in a drag race with a Nissan Skyline. The ownage is so high that if the PS4 tried to tea-bag the Xbox One from that distance, the balls would only be visible with a pair of binoculars.
According to WCCF Tech, given the structure of the Xbox One’s memory management and eSRAM, the stats middle out given that measuring the GPU performance with eSRAM alone is disingenuous (remember eSRAM on the XB1 only clocks in at 32MB, so it’s only used for special operations) just the same as only measuring the Xbox’s DDR3 RAM would be equally disingenuous. Both memory components work together, it’s like imagining a car that needs both the gear-shift and the clutch in order to accelerate, one or the other alone just doesn’t cut it.
The other thing worth mentioning is that neither the DDR3 or eSRAM individually make up for combined performance. So it’s not taking both the benchmark results and putting the numbers together, but rather finding the average that constitutes how both of them work together in actual usage (as mentioned, eSRAM is usually only reserved for special buffer operations or rendering tasks .)
Here’s the thing: WCCF measures that the Xbox One, realistically, measures on the chart at around 32 through 35 on the chart when you average the performance between the eSRAM and the DDR3, where-as the PS4 consistently hits around 42 on the chart, thanks to the shared GDDR5 memory and shader units.
You can see how both consoles theoretically measure up when compared to high-end PCs.
This isn’t surprising, given that GamingBolt’s benchmark captured from Allegorithmic’s Substance Engine turned out similar results… only it was comparing the CPU performance between the PS4 and Xbox One.
But what does this mean for gamers? How does this affect the games? Well, simply put, it means that in the long term of things the PlayStation 4 has scalability. Developers will be able to squeeze out a bit of extra performance from Sony’s machine over time. The Xbox One? Not so much. In fact, we’re already seeing limits with games like Halo 2: Anniversary Edition and the Destiny beta showing the lack of high-end, full HD scaling for Microsoft’s console.
This doesn’t mean that the Xbox One won’t get good, fun or worthwhile games, it just means that there is more room to scale games with bigger experiences and better performance on the PS4.
TL;DR: The theoretical benchmarks re-confirm that the PS4 is still a lot stronger than the Xbox One. Games coming down the pipeline will eventually showcase the differences.
(Main image courtesy of Melty)