Tuesday, March 21st 2023
Apple A17 Bionic SoC Performance Targets Could be Lowered
Apple's engineering team is rumored to be adjusting performance targets set for its next generation mobile SoC - the A17 Bionic - due to issues at the TSMC foundry. The cutting edge 3 nm process is proving difficult to handle, according to industry tipsters on Twitter. The leaks point to the A17 Bionic's overall performance goals being lowered by 20%, mainly due to the TSMC N3B node not meeting production targets. The factory is apparently lowering its yield and execution targets due to ongoing problems with FinFET limitations.
The leakers have recently revealed more up-to-date A17 Bionic's Geekbench 6 scores, with single thread performance at 3019, and multi-thread at 7860. Various publications have been hyping the mobile SoC's single thread performance as matching that of desktop CPUs from Intel and AMD, more specifically 13th-gen Core i7 and 'high-end' Ryzen models. Naturally the A17 Bionic cannot compete with these CPUs in terms of multi-thread performance.
Apple has an excellent reputation for its chip designs, and a good portion of their customer base are not too concerned with hardware specifications, so the rumors of slightly lowered performance expectations for next's years flagship devices could be less of a headache for the engineering team. The current generation A16 Bionic outperforms Qualcomm's latest Snapdragon 8 Gen 2 in terms of pure processing power, and only lags slightly behind with its GPU's capabilities.
Sources:
MacWorld, Revegnus Twitter
The leakers have recently revealed more up-to-date A17 Bionic's Geekbench 6 scores, with single thread performance at 3019, and multi-thread at 7860. Various publications have been hyping the mobile SoC's single thread performance as matching that of desktop CPUs from Intel and AMD, more specifically 13th-gen Core i7 and 'high-end' Ryzen models. Naturally the A17 Bionic cannot compete with these CPUs in terms of multi-thread performance.
33 Comments on Apple A17 Bionic SoC Performance Targets Could be Lowered
I don't see a 13600k going into an ipad mini and running cool.
Go to semianalysis for a full review of TSMC's 3nm node and how poorly it rates.
Geekbench 5
Geekbench 6
My 1100T is still fine at 4.2Ghz and my 12 year old hard drives are still working, there are however a small number of games that use newer software that my CPU doesn't support natively and they either run poorly or not at all. But I have been through many, many, many phones since I built this machine. My wife has had multiple Ipads and Iphones as they "aren't supported" for general use anymore and they cannot be upgraded, tweaked, or tuned.
If you are fine paying thousands for a 3 year life cycle device, be my guest, but never whine about the cost, how bad things are for the enviroment, workers rights, or tell me that Apple is amazing and not a company making a product for profit with a lifecycle decided on before it for sale.
I can't really find a workload that requires more than what the A16 already does because Apple doesn't allow for that workload to exist (the pro's are using M series and even them are massivily over powered for what they are able to do within their walled gardens)
Battery is getting a bit old though..
The name also implies to customers that it's a generation newer. Really the difference isn't even going to be on par with CPU or GPU mid-generation refresh gain. Apple is implying to customers that there will be big gains when in reality they are getting an extremely tiny bump.
But it's Apple so people will defend it tooth and nail.
But there is always room for improvement in the performance per watt , or indeed the „do the work asap, go to idle asap“ method. And whatever other battery extending tricks can be found.
Are screens, wifi, 4/5G, gps the weaker link? Should mire effort be made in making those more power efficient?
There are definitely gains to be made by the OS thread scheduler. When i put my phone down , if i leave too many apps open , esp navigation apps, they drain the battery really fast. They could make better use of sensor awareness to put the device into idle; app specific.
Apple and Nvidia are claiming ARM's just due.
But regarding work, anything bigger than sending e-mail is painful, it is less than 1% of users. Same for photo and video editing...
However, faster "mobile" CPUs are very welcome for tablets and light laptops... Apple already shaken the market with M1, so now they are facing bigger and bigger expectations :)
A lot of mecanism to improve performance have been well documented and many people know how to implement them. This is why the first phase of catching up is always quick.
The problem start when you have to find news ways of improving performance. You can't just reuse already knows trick and you need to come up with your own things.
And not just one time, but every revision of your architecture.
This is where Apple is right now. Doesn't means they won't provide big boost of IPC in the future, but they will have to find new ways of doing it. But they are now in uncharted territory.
And if you look carefully, you have more powerful CPU on flagship phones than on tablets.
Also, if you're filming and taking photos on your phone, it makes sense to edit them on the phone.
If you're buying a high end phone just for emails, well, you do you then;)
Sure, if you cherry pick the worst efficiency Intel processor it might look bad. Then again that's cherry picking so any argument with that as it's basis is easily dismissed.
And yes, x86 is a much older architecture that has to maintain comparability with decades of applications. That requires ingenuity. You'd expect X86 gains per generation to be lower, not higher than a brand new uArch that dropped backwards compat in order to squeeze out maximum performance and min power consumption.I did say it was impressive as you even quoted in your first reply to me. It was the M2 that I said was disappointing.
There's a lesson to be learned here, those that fling insults are often met with ironic consequences frequently beset on themselves.