The star of the Snapdragon X Elite is its all-new custom CPU architecture, codenamed Oryon. The X Elite includes a total of 12 Oryon cores, rather than a mix of large and small cores like Qualcomm has used in past designs. When all cores are active, they can run at peak speeds of up to 3.8 GHz, though when just one or two cores are boosting they can go up to 4.3 GHz.
Qualcomm uses some Apple-esque performance charts to demonstrate the X Elite’s performance and power efficiency; the company claims that the X Elite will run up to twice as fast as an Intel Core i7-1355U or Core i7-1360P at the same power level, or it can match their performance while using 68 percent less power. Qualcomm also says the X Elite can match the performance of a beefier Core i7-13800H using 65 percent less power—providing roughly the same multi-core performance at 30 W that the Intel chip provides at 90 W. The X Elite’s power consumption appears to max out at around 50 W, and to go as low as 10 W, at least according to these charts.
Qualcomm says that the X Elite has “50 percent faster peak multi-threaded performance” than the Apple M2, based on a multi-threaded Geekbench 6.2 test. Qualcomm’s charts imply a peak power usage of around 50 W for the X Elite, which suggests that Qualcomm can roughly match the CPU performance of Apple’s M2 Pro or M1 Max at roughly the same power consumption—this would be impressive for a first outing. But remember that this is just one test that Qualcomm itself has provided and that Apple is gearing up to release M3-series processors on a new 3 nm manufacturing process, which ought to help Apple keep its power-efficiency edge.
The performance/power numbers are not too surprising given they are comparing a TSMC N4 device to an Intel_7 device. The arm design will also lose performance in some applications due to the need for x86 emulation.
I suspect if they compared to the similar TSMC N5 AMD dragon range CPU they would not have faired so well.
Thanks. And we might see how the Apple M3 compares on Monday at the next Apple event, though I’m not anticipating a huge increase in performance over the M2.
Yeah… What’s the main market for native high performance apps these days? Games, and perhaps media creation. A lot of people would probably be happy with one of these Arm systems if they get behind the major apps like browsers performing well. Gamers will be last in line I think to get their needs well served.
I’m not too worried esp. with the assertion the other day that AMD can and probably will do Arm/Windows in 2025 if this becomes a must have. Qualcomm had a ten year exclusive on Windows on Arm and look where it got them.
Wild-assed idea: in the age of chiplets could we see a mixed instruction set CPU pairing Arm and x86 cores in one package? One x86 Zen5c chiplet or something, alongside a bunch of Arm cores for the true efficiency core?
It was a long time ago microsoft and google (Chromebook) made a play here with ARM. Intel basically shut this down with the low end celeron and pentium chips that sold for less than $50 with good enough performance and little additional required R&D. We now see some Arm chips with excellent performance rivaling the best X86, but they are still going to have a significant application problem, along with a marketing hill to climb. While Apple has some great products with ARM, their market share remains tiny.