I'm pretty sure Apple knows what machines it sells, including to video and audio pros, and knows how many of them buy the 32GB options where they're available (in Intel).
Besides "Pro" is just a branding, not a class of machines designed for specific tasks with any specific guarantees.
Up until now, "pros" did just fine with 16GB or up to 32GB max on their Intel Mac laptops, so I'm not sure how "twice as fast" machines, at lower price, with the same 32GB RAM (but much more efficient use of it, closer to the CPU) will suddenly be a problem.
Especially if hardly any buy PC laptops with 64GB of RAM anyway -- and Apple doesn't, and has never, competed with custom tower PC rig makers, but to people looking at the sweet spot of portability, power, battery life, size, weight.
I agree, and to be fair, there are plenty of professional use cases that are totally serviceable with one external monitor and 16GB of RAM. Not every professional is a YouTuber or big data engineer.
I can't speak for anyone else, but I do like it when the name of a product actually means something, because it gives me confidence in the honesty of the vendor.
With NVMe, I'd rather have 16GB LPDDR4X than 64GB low-JEDEC, high latency DDR3 memory. More RAM is still good in many cases, but it's not vital like it was 15 years ago when you'd end up in swap hell or OOMing constantly over a GB or two.
I'm not quite sure where why people seem to be clamoring for more RAM, it seems like a non issue these days. I don't really game these days, and when I do it's on older "steam sale" titles so maybe gamers need it, but even for dev stuff, I just don't see the issue.
It's not my job to help people pick out computers better suited to do the work they were hired for, but I really enjoy seeing the characteristics of and where the bottlenecks are in new generations of hardware.
People seem to be under the mistaken impression that supporting higher memory capacities is hard. It's not. It's trivial. The only reason the M1 doesn't is because of how it's packaged, and they only chose to package it that way because the products they were intending to replace it with only need 2 memory chips. They did what made sense for the M1, and they will do what makes sense for the M1x.
Do we have perf counters that show how memory starved is M1 under pressure?
Granted you'll feel bad if this leak is wrong and Apple shows an M1x with more than 16GB of RAM.
From my experience with both, I think you might be very unhappy in retrospect if you bet that more RAM is going to make the most recent Intel one come within shouting distance of even an 8GB M1 Mini. The old one was one of the best for single-core performance within its generation, but they're not even the same thing, not even close. Again, I own both machines with the Intel Mini maxed out as far as it can go with RAM.
My gut level reaction is, you could give the new one lots more RAM (in theory) and it wouldn't end up using it. So they're capping them, otherwise people will assume there's a benefit and that the 8GB and 16GB ones aren't workable for heavy processing.
I've also got a suspicion (possibly just a weird cynicism) that this is why the previous-gen Mac Pro runs so huge and expensive: they've overpriced and overdesigned it so hard that it's unattainable for nearly anybody, making it a sort of status object with Veblen pricing, specifically to avoid class action lawsuits when they release newer machines that outperform it at a tenth of the price.
You were never supposed to actually buy the monstro cheese grater Mac Pro or its screen with the thousand dollar stand, and if you did, (a) it's of symbolic value to you and (b) it's your own fault. When the real high performance computer comes out of Apple it'll be at Mac Mini prices, and meant to iPhone-ify desktop computing as a category.
They just started the Apple Silicon transition. A bit early for the doom and gloom, no?
I started computing in the days when you could add your own instructions to your computer. I’m not suing anyone though that capability is no longer available
http://users.telenet.be/kim1-6502/6502/2ksa/2ksa.html
Cover:Schematics for a 5V, 3A regulated power supply and a 1K x 8 read/write memory block. The power supply and three such memory blocks can be added to the basic KIM-l microcomputer to provide the 4K RAM required by this assembler. Parts are available from Jameco Electronics.
It was a pleasant shock. I ended up returning it as it turns out I really do need at least 32GB of RAM, but it was VERY difficult to take it back.
Even if this rumor is way overhyped, I still can't wait for a MBP that can take at least 32GB of RAM. I'll be there with bells on!
Definitely glad other people are having a better experience.
Would be much better if the base model was 16/512, but that's $500 of nearly pure profit apple would lose.
I've recently used a 100 GB partition on my personal laptop for client work and it was fine.
(I'm not trying to defend the price of Apple's SSD upgrades.)
Instead of a DIMM, LPDDR basically puts an entire DIMM on one chip soldered to the board (or package on the M1). LPDDR topologies tend to be limited in how many ranks of memory you can have, which limits the total capacity.
I think if Apple really wanted to mac a "HEDT SoC", they would switch to "quad channel" (256bit wide) memory, which along with the jump to LPDDR5 would let them sell 64GB/128GB machines. The only existing SoCs with wide memory controllers is the Tegra Xavier though.
Best. Typo. Evar.
These things already don't act as if it's RAM as we're used to considering it. They could probably have shipped the first ones with 4G RAM and had 'em act relatively normal… with heavy stuff like 4K video editing. Not great, but normal. So they did 8G RAM to have 'em act more impressive… like doing said editing smoother with fewer dropped frames than far more powerful Intel machines with a lot more RAM.
M1: https://images.anandtech.com/doci/16252/latency-m1_575px.png
Zen3: https://images.anandtech.com/doci/16214/lat-5950.png
If anything the M1's ram has _higher_ latency than the Zen3.
(Can't reply to the sibling comment but Intel's numbers are even better: https://images.anandtech.com/doci/16214/lat-10900.png)
Is it random accesses you want to test? Or TLB thrashing?
The R per RV prange benchmark accesses memory randomly, but in a TLB friendly pattern (TLB thrashing real world workloads aren't particularly common). The M1 gets 30ns, the 5960X is 55ns and still climbing at the max array size.
The 5950X does better on "full" random, although that's unclear if that's a faster TLB, or the testing with a different page size. With Linux for TLB thrashing workloads you can always use HUGE pages, not sure if the M1 hardware supports that though.
Just noting Zen3 is not Intel.
Please stop this non-sense on HN. DRAM is DRAM. L2 Cache is L2 Cache. Doesn't matter how good its 8-Channel LPDDR4X Memory Architecture sitting next to the Die in the same package it is still just DRAM.
It takes my M1 a little while longer to stitch together 4K videos than my 2017 MPB 16". My 2017 MBP screams and throttles down the CPU down to 60%-75% as soon as editing starts.
Then buy something else, full stop.
Even if the RAM were as fast as L1 cache, it still wouldn't do a thing for the SSD bottleneck.