its weird having waited all this time to for M3 like it was the next great thing only too have it swept under the carpet for M4 before it even hits a desktop
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.
Apple Likely Planning to Skip M3 Mac Mini for M4 Refresh in Late 2024
- Thread starter MacRumors
- Start date
- Sort by reaction score
For the last 20 years, the laptops have only changed how they look every 3-5ish years, so that was never the annual discussion?
capable of basic tasks and even then swapping. When Apple have to revisit their unwise comment about 8Gb allegedly equivalent to 16Gb on a PC, and have now stepped by suggesting 8Gb is good for basic tasks, you know they are on the back foot. Apple's reputation is slipping fast and I've been with them since Apple Lisa.
Apple's reputation was built on being before the curve not behind it and even that is slipping, and whilst all companies want to maximise profit, Apple are damaging their reputation of late and mass propaganda from them about the 8Gb situation, albeit stepping back from their original assertion, does them no credit whatsoever.
We avoided the M2 completely, and we've not bought on the M3 platform either, and reluctantly took the decision for the first time in our history to offer PC's as well, as Apple's stance on RAM is pure greed.
Gurman has effectively froze the Mac desktop market after his last 2 updates. He insinuated last week that M3 Studio/Pro were not coming - now Mini. Who will bother purchasing a desktop between now and WWDC?
Most people buy computers when they need to, and the schedule for future releases just doesn't matter that much.
Nothing wrong with buying an M2 Pro/Max/Ultra desktop right now, if you need it.
In the past Apple surprised me with how affordable they made some very advanced tech (iPhone 4, for example). Now, it feels like endless rationalization for obvious cash grabs at the expense of function.
M2 was described as a place holder until the first major upgrade to M1, the M3. It will be holding that place as minor incremental update to M1 for another year on the mini/studio.
Maybe Apple is just going through a rough patch with their tech.
I think you’re on to something, come out with a base Mac mini model and provide one slot where a user can put in additional ram or ssd if needed.
Or this is all just normal computer development being normal.
M2 was exactly as expected: A reasonable "same process node" bump. Same as Intel would have ever had.
Intel to M1 was exceptional because Apple's Intel chips sucked and were even bad VS current, at the time, Intel and AMD chips. Expectation management. The "M1 like performance jump" is probably not going to happen until Apple moves to its next processor architecture.
Apple computers. The fans used to blame Intel for untimely CPU upgrades. Turns out it was Apple all along.
HobeSoundDarryl
macrumors G5
I think you’re on to something, come out with a base Mac mini model and provide one slot where a user can put in additional ram or ssd if needed.
That's not what I said at all. I showed a stick of 16GB RAM to show what it can cost at retail... which is FARRRRRRRR from $200. I make no argument for RAM slots, nor have any expectations of Apple going there.
Instead, I suggest it's past time to roll out Macs with 16GB RAM as base spec to completely kill this "whine" & angst at what I'm guessing would cost Apple upwards of about $15 per Mac based on retail pricing of a single stick of 16GB RAM... instead of doing what OP was implying... which is implying that the only way to 16GB base is with a + $200 price increase... as if the Apple price is the only possible price to apply.
The Apple price is many times retail for RAM. Their cost is far from that. They buy in huge volume not in retail quantity 1 at a time. They buy direct, not through an Amazon, who takes a profit on top of the Crucial profit on top of the actual cost. If 16GB of Crucial RAM sold through Amazon can be had for $44, cost is very likely no more than $15 in volume purchasing... very, VERY far from +$200.
$15 per Mac to buy a big shot of customer goodwill is NOT a lot of money at all. $15 to crush this particular topic is a bargain. The marketing spin could be about leading the space to 16GB base and how "we listened to our customers" as opposed to too many moves lately that seem to only have an ear to Wall Street.
And the added RAM even for the light users can be there should they need it at any time in the next 7-10 years... like Ray Tracing even for the people with no current intention to play one game... or AV1 decoders... or the capability to manage XX streams of 8K video at the same time, etc. Light users don't need a LOT of stuff built into Silicon... but perhaps they'll use it at some point. RAM is a core component of any computer... more important- IMO- than Ray Tracing or AV1, etc. Too much RAM is a very good "problem" to have. Too little in a computer with no upgrade options is a "throw baby out with the bathwater" problem (meaning, replace the whole computer should that need arise in the next decade of using that Mac purchased in 2024).
Last edited:
Well, the Intel Skylake decade did suck, and help lead to Intel's fab business losing 7 billion
[corrected]
Last edited:
Intel did not lose $7B this year. It was last year and it was for (still nonexistent) foundry business. As a whole Intel made $7B profits just in Q4.Well, the Intel Skylake decade did suck, and help lead to Intel losing 7 billion this year
Capable of a MR video production workflow. He didn't say that it didn't swap, he said that it wasn't noticeable when it did.
....
Price anchoring on DIMM prices is more delusional than rational here.
Your numbers are disconnected from reality. That DIMM you are pointing at here has eight RAM packages on it. Apple's 16GB implementation for plain Mn series consists of two. There is a 4x difference in packaging density there. That isn't likely going to be a "$8-15" difference. It is pretty similar to saying that HBM RAM should cost as much as generic DIMMs. Furthermore, you are pretending that Apple can just go buy it from someone else. Apple is using semi-custom LPDDR RAM packages. So they can't just pick arbitrary RAM vendors. They have to negotiate a deal for customer work to be done to meet Apple's specs. Even HBM packages are largely physically standard (variations on clock, but don't really need to get physical differences on package build process.).
The Max/Pro memory packages are even more semi-custom ( multiple intra-package stacks ). [ And for better worse that leads to bleed back into the plain Mn packages since Apple is likely going to impose an consistent $/GB curve on both types of packages. Better margins but also spreads out their semi-custom overhead over a broader user base. ]
What Apple is having made is more so a "poor man's HBM" more so than trying to buy the lowest generic LPDDRn packages on the open market. The whole point is not to leverage a generic, commodity RAM package. What Apple is roughly doing is taking basic LPDDRn memory dies and wrapping incrementally more expensive packaging around them . That isn't going to lead to maximum commodity pricing.
Are Apple's margins razor thin? Extremely unlikely. But the commodity RAM of the socketed desktop market and what Apple is doing with RAM is thoroughly comparing Apples to Oranges. The packaging is way past being substantially different. Most likely comparing either substantially different generations of RAM die technology and/or very different packaging technologies. And pretty likely it is differences in both. So it is basically different tech.
That said, Apple probably does use mark ups on RAM and SSD as 'slop' to offset tradeoffs on perhaps other components getting more expensive (e.g., N3 wafers cost more than N5 wafer. If the Mn die is about the same size then the SoC cost is going higher. Previous generations have kept RAM/SSD the same as went to more expensive screens. etc. etc. ) . Part of this is that Apple mainly tries to keep the system prices stable over time even as some components may trend cheaper over the long run. ( blimps like the MP increasing 2x in entry price are rare. And the gen-over-gen price lowerings are relatively rare also. Apple has a relatively fixed average selling price they are shooting for in US dollars. ( Wild ForEx price swings have different impacts in other countries where prices bounce around more volatile fashion. Apple doesn't really want volatility but are not going to 'eat' losses to make it happen. )
Last edited:
HobeSoundDarryl
macrumors G5
OK, I'll be delusional then... and continue seriously considering my next laptop choice being a PC over a MB... something that this long-term Apple everything guy could not even think about 5+ years ago.
And in doing so- per all that- the geniuses there apparently opted to bail on cheaper Intel chips, commodity RAM and commodity SSD to build their own much more expensive versions of all that... so we can rationalize $200 for 8 more GB of RAM in a conversation like this... and $2200 for 8TB of SSD vs. under $700 for quantity 1 m.2 at retail, etc.
But then, in spite of all that added cost they apparently chose to take on instead of continuing to use Intel, they have somehow expanded corporate margin anyway from the traditional 38-40% target to north of 47%. It's truly magical to take on much greater costs and significantly grow margin at the same time. But maybe that's a delusional observation too.
And in doing so- per all that- the geniuses there apparently opted to bail on cheaper Intel chips, commodity RAM and commodity SSD to build their own much more expensive versions of all that... so we can rationalize $200 for 8 more GB of RAM in a conversation like this... and $2200 for 8TB of SSD vs. under $700 for quantity 1 m.2 at retail, etc.
But then, in spite of all that added cost they apparently chose to take on instead of continuing to use Intel, they have somehow expanded corporate margin anyway from the traditional 38-40% target to north of 47%. It's truly magical to take on much greater costs and significantly grow margin at the same time. But maybe that's a delusional observation too.
Last edited:
M2 Mini can already go to 24GB on the M2 chip or 32GB on the M2 Pro. The only advantage the M3 Pro would have is it goes to 36GB.
Now as always the 8GB base is ludicrous, not really because a $600 computer with that isn't reasonable, but the $200 to move past it makes it feel like we're a decade behind.
The M3 Rollout continues to be the weirdest rollout yet, with them prioritizing the iMac over the Mac Mini and now this latest report, it seems every other chip will go to either the Mac Mini or iMac, but not land on both ever again. Also it was pretty believable they would do all the desktops at WWDC? Are they intentionally going to tier the Mac Studio, Mac Pro, and Mac Mini to not match each other at steps anymore?
Given the MacBook Air to MacBook Pro, that's a Pro feature and they want you to upgrade to the Pro chip on the Mac Mini.
Okay, fine, Apple. I had been waiting for an M3 Pro Mini announcement, but now my ancient Intel iMac hangs on another few months before it's put out to pasture.
But the actual HD isn't on the CPU. No desktop computer should come with 256gig HD in 2024 and it certainly shouldn't cost £200 to go from 256gig to 512gig HD.
M3 Pro isn't much of an upgrade over M2 Pro anyways. Certainly not like M3 over M2.
That makes sense.
That makes sense.
I Got 9 years of MacOS support out of the first Retina 15" MBP , thats mid 2012 thru 2019 then 2 additional years of Safari/MacOS security updates to 2021. Given what the M1 offers as far as speed I could see it perfectly usable 10 years later for consumers.