SoylentNews is people
SoylentNews
Apple's A12Z Under Rosetta Outperforms Microsoft's Native Arm-Based Surface Pro X
Apple's Developer Transition Kit equipped with an A12Z iPad Pro chip began arriving in the hands of developers this morning to help them get their apps ready for Macs running Apple Silicon, and though forbidden, the first thing some developers did was benchmark the machine.
Multiple Geekbench results have indicated that the Developer Transition Kit, which is a Mac mini with an iPad Pro chip, features average single-core and multi-core scores of 811 and 2,871, respectively.
As developer Steve Troughton-Smith points out, the two-year-old A12Z in the Mac mini outperforms Microsoft's Arm-based Surface Pro X in Geekbench performance, running x86_64 code in emulation faster than the Surface Pro X can run an Arm version natively.
So the DTK with a two year old iPad chip runs x86_64 code, in emulation, faster than the Surface Pro X runs it natively 😅 Oh boy Qualcomm, what are you even doing? https://t.co/UAlZiwSsF8 — Steve Troughton-Smith (@stroughtonsmith) June 29, 2020
(Score: 2, Insightful) by darkfeline on Wednesday July 01 2020, @06:24AM (13 children)
The last time I cared about CPU performance was, uh, can't remember. Certainly many years ago. Battery performance though, I care about all the time.
People need to stop reporting benchmarks so companies can stop increasing them, it's retarded like the push for thinner, bigger phones.
Join the SDF Public Access UNIX System today!
(Score: 0) by Anonymous Coward on Wednesday July 01 2020, @06:35AM (1 child)
Use an AGM deep cycle battery, then?
(Score: 0) by Anonymous Coward on Wednesday July 01 2020, @05:36PM
Not sure if you were kidding, and you probably were, but battery technology aside, darkfeline's point was in favor of lower power CPUs and RAM.
(Score: 1) by petecox on Wednesday July 01 2020, @10:04AM
One might consider a real-world benchmark of Chromebooks [pcmag.com], where both x86 and ARM architectures are established. At a price point in a form factor I might consider, disappointingly the ARM64 Lenovo is beaten by Celerons. One might well say that's an entry level Mediatek phone SoC but ARM still has some catching up to do if it's to blow away x86 in the back-for-buck stakes.
Still, with the 8GB Rpi4 heralding the year of the ARM64 Linux desktop, there's hope.
(Score: 5, Insightful) by takyon on Wednesday July 01 2020, @10:43AM (1 child)
Higher performance (at no increase in power consumption) can lead to higher battery life. If a task is completed in less time, the device goes back to idling that much sooner.
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 1, Insightful) by Anonymous Coward on Thursday July 02 2020, @12:52PM
Agreed.
But also, CPU performance does matter. My wife has a $400 Acer (or maybe Asus? I don't remember) Chromebook and web pages that pop up instantly on my mid-range desktop take at least 5 seconds longer to load on the Chromebook. My kids' school district supplies $200 Dell Chromebooks, and they're even slower. If the low end Chromebooks were the only computing device in the house, I probably wouldn't care because I wouldn't know what I'm missing. But if you spend two weeks on the desktop and then switch to the Chromebook you will lose your mind in frustration over all of the delays.
I don't have a personal laptop right now. I wouldn't get a Chromebook because I'm not a fan of Google. But if I was to buy a laptop, even though I wouldn't plan to game on it I would get something with an NVMe SSD and a pretty fast processor, battery life be damned.
(Score: 0) by Anonymous Coward on Wednesday July 01 2020, @11:46AM
The thinner, bigger phones fold easier.
(Score: 2) by driverless on Wednesday July 01 2020, @12:41PM (5 children)
I know it's fun to laugh at Microsoft but with a performance disparity this strange has anyone considered that there's a problem with the benchmarking? These figures don't make any sense, which makes it more likely the benchmark values are wrong than that 1 + 1 = 7.
(Score: 0) by Anonymous Coward on Wednesday July 01 2020, @12:53PM (4 children)
It's not actually that big a disparity - the benchmark was NOT emulating amd64, they were both running natively.
(Score: 2) by driverless on Wednesday July 01 2020, @01:06PM (3 children)
So the summary is wrong?
(Score: 0) by Anonymous Coward on Wednesday July 01 2020, @05:01PM (1 child)
Yes, see the Twitter thread
(Score: 2) by FatPhil on Wednesday July 01 2020, @06:07PM
Conclusion: the arm-targetting compiler is shit, and the intel-targetting compiler's better?
It's just as valid a conclusion as "ipad's chip is better than surface's chip" given the available data.
Having said that, I've worked a fair bit with Arm SoCs, and Apple's were very well designed for speed compared to many.
Great minds discuss ideas; average minds discuss events; small minds discuss people; the smallest discuss themselves
(Score: 2) by FatPhil on Wednesday July 01 2020, @06:01PM
"... than the Surface Pro X can run an Arm version natively.", and
"... runs x86_64 code, in emulation, faster than the Surface Pro X runs it natively.",
where "it" should refer to "x86_64 code".
Blame macrumors, not the eds.
Great minds discuss ideas; average minds discuss events; small minds discuss people; the smallest discuss themselves
(Score: 0) by Anonymous Coward on Wednesday July 01 2020, @08:48PM
You went to all the effort of posting a comment that makes you look stupid. Turn off luddite mode.