News:

Willkommen im Notebookcheck.com Forum! Hier können sie über alle unsere Artikel und allgemein über Notebook relevante Dinge disuktieren. Viel Spass!

Main Menu

Post reply

The message has the following error or errors that must be corrected before continuing:
Warning: this topic has not been posted in for at least 120 days.
Unless you're sure you want to reply, please consider starting a new topic.
Other options
Verification:
Please leave this box empty:

Shortcuts: ALT+S post or ALT+P preview

Topic summary

Posted by Daniel K
 - September 23, 2020, 22:34:51
Wasn't there an Apple leak that the fix on the 5nm processors (at least for mac book pros) will make it into the spring mac book pros, not fall processors? If that's true, then it's a processor issue.

Also, Apple wants lots of processing power because (when a similar process) is used for laptops, laptop customers and processors will need to be impressed in order to switch from Intel/Windows to mac. So I doubt the low performance is intentional. I expect to see a commensurate jump in the performance of 2021 5nm chips to get closer to expectation.

It does not strike me at all unlikely that it took awhile to find the optimum layout on the new (truly tiny gate size) architecture. Discovery of physical layout optimizations sometimes won't fit to a desired schedule, but from the rumors, it sounds like the engineers are happy with the performance of [2021] chips that just missed the [2020] cutoff dates.
Posted by vertigo
 - September 23, 2020, 20:06:05
Can't edit my post, but I just changed my preferred language from German to English, so hopefully that will resolve the issue, though the site should still keep the user on the current page when changing the language, not send them back to the main page.
Posted by vertigo
 - September 23, 2020, 20:03:14
I don't disagree with that. If x% is considered good/bad for one, that should be consistent across the board. I think 15% is a reasonably impressive and healthy improvement from one generation to another, and certainly better than what Intel has generally achieved with their typical 7-10% increase (which you then lose a significant portion of due to patches for the multitude of vulnerabilities found later, which is why I generally subtract ~5% from reported gains for more realistic numbers). My issue wasn't in NBC knocking or praising the gains, or lack thereof, of any company, but the fact that Apple appears to have received different treatment by them acting like Apple intentionally had lackluster gains (regardless of how good or bad the gains actually were), whereas I highly doubt they would have made such an excuse for anyone else.

@NBC - Please fix your forum. Every time I click an email link to follow up on a thread, it takes me to the German version, and when I select English in the language dropdown, it redirects me back to the main forum index instead of keeping me in the thread.
Posted by Stephen_L_S
 - September 20, 2020, 08:03:51
The reality is exactly the opposite, people are excited about zen 3's 15 percent improvement and talk about how much it will destroy Intel even in gaming now, yet when the same happens to apple every website that popped up when you search for A14 is complaining about how disappointing it was. This comment illustrates it well.
Quote from: Jordon on September 18, 2020, 15:48:22
Its amazing that we can say "first 5nm chip offers effectively no performance gains for iPhone 11 and iPhone 11 Pro series owners" even when the cpu gets a 15% boost in performance and then in the same breath everyone get all hyped and excited that Intel or AMDs next gen cpu will offer the same 15% performance boost.
Posted by Jordon
 - September 18, 2020, 15:48:22
Its amazing that we can say "first 5nm chip offers effectively no performance gains for iPhone 11 and iPhone 11 Pro series owners" even when the cpu gets a 15% boost in performance and then in the same breath everyone get all hyped and excited that Intel or AMDs next gen cpu will offer the same 15% performance boost.
Posted by _MT_
 - September 18, 2020, 11:16:25
The comparison against A12 makes perfect sense. Why? Because they presented the chip in a new iPad Air. And the outgoing version uses A12. Was it an iPad presentation that happens to have a new chip or a chip presentation that happens to be in an iPad? In the end, consumers care about the product, not the components. We can only speculate that they chose to present iPad first because the gains in an iPhone just aren't that great and it wouldn't look as good.

Personally, if this is indeed indicative for iPhone, I would guess that Apple might have gone for efficiency. Which would allow them to use a smaller galvanic cell for the same endurance and to make the device thinner and lighter, which is something Apple apparently very much cares about.

Yes, I can imagine Apple reaching a point where they're happy with the performance and they'd rather invest elsewhere. But I would be more careful with comparisons against x86. x86 processors excel in general computing. The kind of computing for which there is no specialized hardware acceleration. Yes, they still have many specialized instructions. But they're very fast in the generic stuff. Mobile chips are largely built on hardware acceleration (because they're slow and it would cost too much power otherwise). Scaling is also a problem. The cores are just not designed to run at high frequencies. High frequencies are inherently less efficient. Which is why phone chips gravitated towards higher core counts faster. Why design a core for something nobody is going to use in a phone? So, I'm really looking forward to what Apple manages to do. Personally, I was hoping for Mac mini and MacBook Air to be the first ARM systems. I'm not keen on laptops smaller than 13 inches, even for travel. I wouldn't be worried about performance in those systems. I was hoping for proper passive cooling in the Air and while the 6 core Intel mini might have faster CPU, it only has UHD 630 for GPU. Come to think of it, passively cooled Mac mini - that might be nice (dust is evil). Renoir might eat the outgoing mini for breakfast but I'm sure the ARM would do as a mini. It's the bigger systems where things start getting interesting. I really wouldn't fancy standing up against a Threadripper. AMD did a stellar job.
Posted by Avinash
 - September 18, 2020, 07:43:51
I beg to differ on that.

In my opinion, Apple is mostly focused/concerned on the SoC that they are bringing with the ARM Mac (rumored A14X) and because that is where Apple wants to show the significant improvements/jumps they have intentionally trimmed on the performance of the A14 Soc which is intended for iPhones and iPad Air.

Also, a much powerful SoC (A14x) would also compel buyers to go for the iPad Pro

Clearly Apple wants to keeps its best performing SoC for the Ipad Pros and the ARM based Macbooks because iPhones are already fast and there's one could do to make use of all that processing power that comes with the newer Apple silicons.
Posted by toven
 - September 17, 2020, 19:48:36
No. Next gen SD with X1 and Adreno7 will be a big jump ìn performance.
Posted by ike
 - September 17, 2020, 18:26:11
Isnt it way way to early to draw conclusions. Did you write the review of the ipad generation last year. The graphics are way stronger then the A12 op ipad air. Its almost equal to A13. gfxbench 4.0 offscreen is around or well over 60 if i am not wrong that would mean 80 fps or over on that. That is a very sizeable upgrade to gfxbench on the A13 not a few percentages. And apple has thusfar never lied about its performance upgrades usually it even underdid the performance improvements. And there is no reason for them NOT to make the A14 strong it is a statement to their other products.

Lets see many more benchmarks before we draw a conclusion on one benchmark especially if its not the final product.

Posted by Samunosuke
 - September 17, 2020, 18:20:45
Quote from: vertigo on September 17, 2020, 17:43:41
If anybody else (Intel, AMD, Qualcomm, Samsung, etc) had done this, this article would be about how disappointing the new processor is, how the marginal improvement is a failure by the company to offer a significant, or even worthwhile, upgrade. But of course Apple does it and it's suddenly ok and possibly intentional, etc. Yeah, no bias there.

I actually do believe CPUs, both in phones and laptops, are plenty fast enough, and that more attention needs to be spent on making them more efficient for better battery life than in making them more powerful. But this article doesn't even state that as a reason, and it's ridiculous that it gives Apple a total pass on it, when there's no doubt the same leniency wouldn't be given to other manufacturers.

The Reality Distortion Field did not follow Steve to the other side. It remains in full effect.
Posted by vertigo
 - September 17, 2020, 17:43:41
If anybody else (Intel, AMD, Qualcomm, Samsung, etc) had done this, this article would be about how disappointing the new processor is, how the marginal improvement is a failure by the company to offer a significant, or even worthwhile, upgrade. But of course Apple does it and it's suddenly ok and possibly intentional, etc. Yeah, no bias there.

I actually do believe CPUs, both in phones and laptops, are plenty fast enough, and that more attention needs to be spent on making them more efficient for better battery life than in making them more powerful. But this article doesn't even state that as a reason, and it's ridiculous that it gives Apple a total pass on it, when there's no doubt the same leniency wouldn't be given to other manufacturers.
Posted by Redaktion
 - September 17, 2020, 17:09:38
Apple's A14 Bionic SoC reveal indicated what might be the smallest generation on generation leap we've seen til date from Cupertino. Is the near-static performance due to issues with the 5nm process? Or did Apple just fire the last shots in a smartphone spec war that's spiralled out of control?

https://www.notebookcheck.net/Conservative-performance-gains-on-the-Apple-A14-Bionic-could-mean-one-thing-smartphones-are-fast-enough-and-the-spec-war-might-almost-be-over.494566.0.html