programmer

keithw said:

While it's nice that the M3 Ultra is now finally out, why did it take them over a year to release it?  (The M3 line came out on October 30, 2023!) Why didn't they release the M4 Max Studio at the same time as the M4 Max MBP?  If they had, I may have saved a few thousand $$$ since I got tired of waiting and bought the MBP.   And is the single core performance of the M3 Ultra the same as the M4 Max? Enquiring minds want to know...  But I guess with the 512MB memory capacity and the 80 graphics cores on top of the 32 CPU cores, the M3 Ultra should be killer LLM machine.

The thing people don’t seem to understand when whining about not getting this or that is that these products take a huge development effort to create.  And then when they have been designed, there is a complicated balancing act about fab capacities and yields.  This isn’t some blokes with a drill press pumping out aluminum parts from their garage.  This is light years beyond that in terms of complexity.  Apple (and the rest of the industry) are pulling off miracles, and forum trolls pour hate on them because doing that takes no brain cells and you don’t even have to get off the couch.

If I were to speculate wildly, I would suppose that the process node (N3B) used for the M3 series had some issues, and pretty much only Apple used it.  So to address the issues, Apple moved faster on the M4 using the newer process (N3E), and eschewed the ultra connector to get that line out faster.  This may have freed up M3-capable capacity, which they can now use for the M3 Ultra… and IIRC (and this is even more speculative) the first process did have some advantages over the later one (they removed features from N3E to make it work better), which may play better to what high end chips like the ultra need.  So rather than spending the time to make the ultra connector work using N3E, they are probably focused on N3P, which is apparently what comes next.

jvm156 said:

I fail to see how this is any harder to hit than the old ones power button🤷🏻‍♂️

More importantly, it is harder to hit by accident.  I have accidentally hit my 2018 Mac mini's power button while plugging/unplugging cables on the back.  I'm happy that it is moved, and reaching under it to hit the button on such a small and light computer is trivial.  Most rack or other mounting options I've seen will be easier to reach with the button on the bottom.

It is better to leave the machine on.  If power quality is an issue then you ought to have a surge suppressing power strip anyhow (and use its switch), or unplug it from the wall.

My M4 Pro mac mini w/ 48GB RAM and 1TB SSD (plus the external USB3 storage I already had for previous machines) scores a 6.0/5.0 ... it is way beyond what I expected in terms of performance.  Its performance even with Rosetta or other forms of virtualization/emulation is astonishing.  Compile speeds are mind blowing, and its compute capabilities are really amazing, often greatly outperforming big iron servers I use regularly.  And the whole package was quite a bit cheaper than what I've spent on numerous previous computers over the decades.

Oh, and the power button is perfect.  Can't hit it by mistake when fiddling with the ports on the back by feel, and I have pressed it exactly _once_ in over two months.

Sometimes the best way forward is to put something into the market, and increment on it.  I've seen design teams lose their way because they got too detached from market realities.  Just grinding away on a new thing in a lab for years and years doesn't ensure you're going to make forward progress.  Often it can be just the opposite, and what you need is for an exec to step in and put a crisp deliverable in front of you.

danvm said:

And I have work with some of those workstations, and the performance is not as pathetic as you mention.  Some of them have advantages over Apple Silicon, specially when comparing the GPU. 

You say "workstations" whereas what you quoted said "laptops".  Big difference.  Apple Silicon's performance really shines in power-constrained situations.  If you can run plugged into the wall and with big heat sinks, then the Intel/AMD CPUs and the nVidia/AMD GPUs can burn copious power to deliver serious performance.  They're still inefficient in terms of compute per watt, but you don't notice as watts are cheap in such a situation.

It would be interesting to see a large scale data centre built from ARM-based machines and compared to ones build from Intel/AMD-based machines, and compare the operating costs.  Some of the big cloud vendors offer lower cost ARM-based hosts just for this reason -- they greatly reduce energy and cooling costs in the data centre.  Not Apple's focus though, so we aren't likely to see Apple Silicon based data centres (except perhaps for Apple's own, but they are typically very secretive about that).

Bill’s work was tremendously inspiring to me back in the 80s.  One of the greats of the computer industry, gone too soon.  Very sad.  😢

eightzero said:

I don't need a robot or smart technology in my home. I need wifi. Reliable, fast, affordable, secure, easily configurable wifi everywhere. AAPL used to make that. How hard is this? Put it in an Apple TV box, connect it to an antenna and a TV, offer mesh remotes. Pay for it with a subscription to "Apple VPN+" that is part of Apple One. Done.

Not gonna happen. Because robots. 

Not going to happen because there are numerous companies already solving it and therefore pricing has been driven down to minimal margins.