Apple Mac Studio M2 Ultra

In 2011 I bought a fully loaded 27 inch iMac paid 3700 (used it for 9-1/2 years) at today’s prices that would be about $5400, bought a M2 Mac studio M2 ultra that cost about $7200 2-1/2 years ago the price for the new Mac studio M5 ultra just announced by Apple is $11,400. Former owner of Amiga, 1000, 2000, 3000, and 4000 I would have gotten a Next computer at the time if I could’ve afforded it, unfortunately, I was tied to a PC at work using AutoCAD, Revit and Navis Managed a few years later. The Games back in the day on the Amiga computers were ahead of the PC for about 9-10 years, coincidentally Sony bought a British gaming company called Psychosis for their games and gaming technology. They were pretty big back in the day on the Amiga.

It kind of froze my Mac Studio M2 Ultra, but I was able to still move the mouse and force-quit Safari using Command+Q. Once I did, everything immediately went back to normal.

It's not about the feature set. The Music app is the only software I use which randomly locks up my whole mac, just trying to do basic stuff like play a song or navigate albums. Doesn't seem to be related to whether I'm browsing Apple Music or my own library. This is on both my M1 MacBook Pro, and on my M2 Ultra Mac Studio in the office. It's baffling that something that worked so well in 2008 is such a dog in 2025. It's like using winamp in 1998.

The damage is already done, no european corporation is going to rip out Teams because they’re so used to the crap. This should’ve been decoupled when everyone went looking for a chat platform, not in 2024. I’ve worked in many places where no employee had ever opened Slack and they don’t get why Teams is shit. As of this week the “new” version of Teams takes 14 seconds to open from click to usable on my M2 Ultra Mac Studio…

>They just want to see “Running Model X with quantization Y I get Z tokens per second”. Influencers on Youtube will give them that [1] but its meaningless. If a benchmark is not part of an in-depth comparison than it doesn't mean anything and can't inform you on what hardware will run this software best. These shallow benchmarks influencers post on youtube and twitter are not just meaningless but also take days to browse through. And they are influencers, they are meant to influence you and are therefore not honest or reliable. [1] https://www.youtube.com/watch?v=GBR6pHZ68Ho >but I don’t understand how you’re concluding that a Mac is the “second best option” to your $30K machine I conclude that if you can't afford to develop custom chips than in certain cases a cluster of M4 Mac Mini's will be the fastest cheapest option. Cerebras Wafers or NVDIA GPUs have always been too expensive compared to custom chips or Mac Mini clusters, independent of the specific software workload. I also meant to say that a cluster of $599 Mac Minis will outperform a $6500 M2 Ultra Mac Studio with 192GB and be half the price for higher performance and DRAM but only if you utilize the M4 Mac Mini aggregated 100 Gbps networking.

In 1986 we wrote a parallel Mandelbrot program in assembly instructions in the 2K or 4K on-chip SRAM of 17 x T414 Transputer chips linked together with 4 x 10 Mbps links each into a cheap supercomputer [1]. It drew the 512 x 342 pictures on a Mac 128K as terminal at around 10 seconds per picture. I later wrote the quadruple-precision floating-point calculations in microcode [2] to speed it up by a factor of 10 and combined 52 x T414 with 20 x T800 Transputers with floating point hardware into a larger supercomputer costing around $50K. With this cheap 72 core supercomputer it still would have taken years to produce the deep zoom of the Mandelbrot set [4] that took them 6 months with 12 CPU cores running 24/7 in 2010 [3]. In 2024 we can buy a $499 M4 Mac mini (20-36 'cores') and calculate it a a few days. If I link a few M4s together with 3x32 Gbps Thunderbolt links into a cheap supercomputer and write the assembly code for all the 36 cores (CPU+GPU+Neural Engine) I can render the deep zoom Mandelbrot almost in realtime (30 frames per second). That is Moore's law in practice. The T414 had 900,000 transistors, the M4 has 28 billion transistors at 3nm (31.111 times larger at 50% of the price). The M2 Ultra with 134 billion transistors and M4 Max (estimate 100 billion transistors) are larger chips than a M4 but they are relatively more expensive then the M4 so it is cheaper and faster to link together 13 x M4 than buy 1 x M2 Ultra or 6 x M4 instead of 1 x M4 Max. Cerebras or NVid…

This GPU has a 192-bit memory bus. At 32-bit GDDR bus width (well w/ 6, 2x16b data channels per chip), that means you have 6 channels. With regular GDDR6, the largest produced size is 16Gb (2GB), so 12GB is what you get. You could double that up w/ a beefed up PCB if the memory controller supports it to get up to 24GB (in the way workstation cards like W7900 and A6000). Beyond that, you'd have to move to GDDR7 (which has 24Gb/3GB chips incoming) or to HBM stacks, but at that point you're well beyond a "basic GPU". I think the only way you could get to 128GB would be either using regular (LP)DDR or HBM. Note, Apple M chips have weak GPUs with decent MBW and large memory capacities (up to 192GB @ 800GB/s for an M2 Ultra, launched mid 2023) and have not been a major CUDA threat so I don't think your hypothesis actually stands up.