Endless because there are youtubers doing benchmarking. Are they working for Apple?
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
M5 Apple Silicon Chip Reportedly Enters Mass Production
- Thread starter MacRumors
- Start date
- Sort by reaction score
Haven't even got the M4 chips into all their lineup yet, and they're producing the base M5 already!
This is from a roadmap published in 2023:
That Korean news report also referenced that roadmap in the article: 5-10% power savings or 5% performance improvement.
So perhaps expect about a 10-15% performance improvement if we assume slightly higher clocks, possibly slightly higher power utilization, and a small change in the architecture.
My friend says his M4 Mac mini feels noticeably faster for web-based stuff than his M1 Max Mac Studio. That makes sense considering the single-core speed of M4 is roughly 55% faster than M1 Max. He says the M1 Max is still very fast, but M4 is just faster.
In my case I went from a M1 Mac mini to an M4 Mac mini, and I can vouch for the single-core speedup being noticeable, but M1 was more than sufficient already.
It was a joke
Just a little morning humor 😂
Surely you've heard the joke along these lines before?
What about M5 Mini? Why iPad Pro always be the first one to get the latest chip? Its software is way behind hardware already.
Don't worry. There will be people upgrading from M4 iPad Pro.
HobeSoundDarryl
macrumors G5
There is no pattern of releases to be judged in only 2 iterations. For all we know, Apple could finally flip the schedule and roll out M5 Ultra FIRST in Studio & Pro updates in Spring or WWDC.
I can offer this: I bought ULTRA once, only to see much of it "caught up" only about 6 months later with Mnext MAX. So without Apple flipping the release schedule, I don't pay wayyyyyyyy up for ULTRA again. It has to have longer than only a few months as "king of the hill" to get my money. And no, that is not about delaying other chips... but simply flipping the release of Ultra and Base. As is, base first and ultra last. In a flipped scenario, the all-important MBpros could still launch exactly when they do now.
I wish they would go back to upgrading all their Macs at the same time in January every year rather than the current drip drip approach which makes it difficult to decide what to buy and when.
The same goes for the iPad. They should go back to updating all their iPads every March/April.
The same goes for the iPad. They should go back to updating all their iPads every March/April.
Speaking of iPads
Here's a writer I follow who went iPad only for 8 years
And even HE .. a writer .. is now back on macOS
Apple -- the iPad situation is a DUMPSTER FIRE if you are losing "writers" back to a normal Mac
It's a shame that Apple artificially slows down the launch of new products like the Macbook Air, considering it junk, into which you can shove a chip that came out a year ago and get high sales at launch. I guess that's how it works, but I don't see the point in taking such a product at release if it becomes the previous generation in a few months. I hope they return to the previous release strategy, instead of trying to push iPad Pro with an exclusive M5 again, which does not affect this product in any way, because it is still limited by software.
Meh. The MacBook Air sells regardless, as it's their low budget option. It doesn't need the latest and greatest. And no, Apple doesn't consider it junk considering it's one of their best selling Macs. They just realize it doesn't need to have top tier specs to sell, if the price is right.
BTW, I find it amusing that on the one hand we have people saying even their M1 series Macs are still super fast and Apple doesn't need to upgrade the chips so often, but on the other hand we have people saying that the budget-friendly MacBook Air should be an M5 launch product.
M5 already? Huh…
Can't believe I'm still in 2018. 🤣
Can't believe I'm still in 2018. 🤣
Then the M6 and M7 and M8 and so on and so forth. Just enjoy what you have and be thankful you can afford one of Apple's premium computing devices...
I don't think it's a contradiction, the M1 is still very good for doing most tasks comfortably. Macs just don't look as attractive for an upgrade if you release them six months/a year after their chip release. Getting the latest hardware a year earlier just looks better, especially at the same price, because they'll look competitive for a year longer.
Given that the MacBook Air is one of the two best selling Macs, I'd say most people buying the MacBook Air don't care if the SoC isn't upgraded right away. Indeed, I will be buying a MacBook Air this year, and I personally don't give a chit what CPU is in it. The only reason I'm not buying an M3 now is because I predict the M4 MacBook Air will come out in time for the back to school sale, so in that context the small discounts on the M3 aren't enough to push me to purchase a product which will have shorter OS support going forward. To put it another way, I care more about the OS support than I do about the SoC.
Further proof that the Ultra line and its interconnect that did not quite deliver the performance that was expected is a dead end. Now just waiting to see what they do with the Mac Pro and Studio.
The whole MBA situation is weird ... it could probably do just as well without attaching a "number" to it's M chip and still shipping them with M1's in there (in disguise)
Part of me is a little surprised that Apple hasn't just stuck with "M powered" for the Macs and gone with years for the releases.
MacBook Air (2025)
The numbering system actually draws attention to the age of something that, performance and perception wise, huge swaths of customers otherwise wouldn't know about or research (or even notice necessarily)
My M1 Mini is still fast enough for me, my new M4 Pro MacBook Pro is almost too fast.....I'm convinced it's actually doing stuff before I've even thought about doing it...
The 2018 is still fine for a lot of people, yourself included obviously. Even the entry level model is quad-core, it supports hardware H.265 HEVC decode, and the single-core and multi-core performance are both fine for most mainstream usage.
I've said before in another thread that in 2025 I consider bare minimum single-core performance level for a main machine should be perhaps 900+ in Geekbench 6, with 1200+ preferred. The entry-level 2018 Core i3-8100B Mac mini gets 1250 single-core, and about 3800 multi-core. That's almost as fast as my 2017 Core i5-7600 27" iMac, which still feels reasonably fast.
Your i5-8500B is way faster than both of those, at ~1400/5300.
The main issue IMO going forward is OS support. Soon it will no longer get the latest OS updates.
This is why I don’t understand Apple’s update cycle. It introduces certain products with the latest M chip, takes more than half a year to update Studio and Pro, then only months later a newer M chip appears. Does it really take that long to make Ultra chips, that will be dwarfed somewhat by the next generation only months later?
If M5 is already at advanced stages, Apple should skip M4 and have Studio and Pro with M5- but not wait until summer 2026 to release it, as then M6 will be out soon.