How Does AMD’s New Ryzen 5 2400G Fare Against Intel in GPGPU Compute?
When AMD’s R5 2400G launched in mid-February, it quickly became clear that Intel had a formidable new competitor on the market. While Intel still leads on CPU performance, the gap is much smaller than it used to be, and the 2400G’s GPU blows Intel out of the water in gaming.
Now, PC Perspective has examined the other half of that coin: How does Intel compare in GPGPU compute? There’s a bit more variance to that question than you might think — it depends on the workload and the application and the degree of baked-in support for AMD’s Vega graphics.
When it comes to synthetic tests like SiSoft Sandra, the R5 2400G makes hash out of the Intel. In other applications, however, like DVDFab, the tables are reversed, with Intel decisively outperforming AMD. This is likely an issue related to the age of AMD’s APP SDK, which hasn’t been updated in two years. AMD does have a newer SDK, the Advanced Media Framework, but no applications appear to use it.
This does point to one of the struggles AMD faces that Intel and Nvidia don’t. When you own the overwhelming majority of the market, you can take software optimization for your hardware more-or-less for granted, provided you offer developers some resources and support. AMD, as the minority company in both CPUs and GPUs, has to work harder to get support for its own products baked into software.
AMD outperforms Intel in Cinebench, Luxmark (see above), and can accelerate workloads in V-Ray, which doesn’t even detect the Intel OpenCL solution as an acceleration option. That said, you can still get much better performance out of a basic solution like the RX 550 — but with the RX 550 also selling for well over MSRP, that’s just not a solution we can particularly recommend. As before, the APU is a uniquely better value thanks to the overheated state of the GPU market.
It’s going to be very interesting to see how Intel responds to AMD’s GPU performance surge. From 2011 – 2015, Intel made consistent improvements to its GPU architectures on a year-on-year basis. They may never have been fast, but the company was gaining more performance than it was losing to increased fidelity and visual quality. Even strictly talking Intel, the situation is far better than it was 10 years ago, when Intel integrated graphics meant you might be able to play Quake 3 or something equivalent.
Skylake marked the end of Intel’s push to improve integrated graphics performance, at least on the desktop. To be fair, back in 2015, Intel didn’t exactly have much to worry about; the Bulldozer family was too slow to provide effective competition. But now we’re in a new era, playing a new game, and Intel needs to show it can compete on both fronts. Its partnership with AMD to add Radeon graphics to an Intel CPU will definitely help. But given the cost of those chips, they aren’t going to be going toe-to-toe against a $170 APU, unless Intel is willing to eat a heck of a hit on margins.
Check PC Perspective for more information and additional benchmarks.
Continue reading
Google Pixel Slate Owners Report Failing Flash Storage
Google's product support forums are flooded with angry Pixel Slate owners who say their devices are running into frequent, crippling storage errors.
Astronomers Might Finally Know the Source of Fast Radio Bursts
A trio of new studies report on an FRB within our own galaxy. Because this one was so much closer than past signals, scientists were able to track it to a particular type of neutron star known as a magnetar.
How to Build a Face Mask Detector With a Jetson Nano 2GB and AlwaysAI
Nvidia continues to make AI at the edge more affordable and easier to deploy. So instead of simply running through the benchmarks to review the new Jetson Nano 2GB, I decided to tackle the DIY project of building my own face mask detector.
Apple’s New M1 SoC Looks Great, Is Not Faster Than 98 Percent of PC Laptops
Apple's new M1 silicon really looks amazing, but it isn't faster than 98 percent of the PCs sold last year, despite what the company claims.