4 ms·

Boring Is What We Wanted

4 points by CharlesW 1y ago

PaulHoule 1y ago
It is so different form the Windows and Android world where you go 10 years and performance goes sideways but maybe you have some accelerators you don’t use and if you are lucky you get some SIMD instructions that nobody uses except for hyperscalers and the national labs because they are just so slow to roll them out. X86 is increasingly like the VAX and ARM ex. Apple is increasingly asking us to believe they aren’t going to disappoint us the way they’ve disappointed us in each generation for the last decade. Apple is so lucky that they can quit supporting old chips and their fanbois celebrate but when Microsoft wants to use a 20 year old instruction the “who moved my cheese” brigade that keeps Microsoft relevant [1] accuses them of creating e-waste. [1] you can still run Office 97 binaries on Windows 11, you wouldn’t even try doing the same with MacOS classic or Linux binaries of that era.
- bigyabai 1y ago
  Those are tall words for describing a chip with... checks clipboard ...four discrete hardware-based SIMD accelerators.
  - PaulHoule 1y ago
    Which one are we talking about? Apple is successful at getting the capabilities of their latest chips used in library code that applications use. For instance, back when Apple was running the legacy x86 architecture they packed in the Intel MKL libraries but Microsoft and Linux distributions didn't. MacOS comes with libraries that use the accelerators and applications often use those libraries so I've got some confidence that I'll actually use an accelerator that is on the M5 chip. In Wintel my expectation is that they will include some accelerator or other feature (say TSX or SGX) but it might be fused off or there won't be actual OS or library or application support or we might find out a year or two later that the feature doesn't actually work or leaks information. It's a nightmare we've been living for more than a decade. The fixed-width SIMD model that we're stuck with is a regression from the vector processors of the 1980s in that you have to rewrite your code with each generation of the instruction set. Couple that with Intel thinking AVX512 is just a "national labs and hyperscaler" feature but still adding it into consumer chips with it fused off -- too many people think this is business genius but it is not... if the customer isn't paying for the manufacturing of those fused off features the shareholder is. They would be better off not putting them in so ordinary customers get cheaper chips and telling the national labs they have to build a 2x bigger cluster.
    - bigyabai 1y ago
      > Which one are we talking about? You tell me. AVX and SSE are at-most two discrete hardware blocks and more often one. Apple Silicon, in it's genius RISC glory, possesses: 1) A Neural Engine 2) ARM NEON as Accelerate 3) Generic GPU SIMD as MPS 4) MatMul units in M5+ hardware It's an impressive amount of fragmentation for a single vendor that supposedly knows what they're doing. Even Nvidia doesn't suffer from this level of identity crisis, despite burning the compute/raster candle at both ends. Apple's approach explains their NASDAQ performance but not much else. It's a bit baffling.
- bediger4000 1y ago
  Given Linux resting on source code, I'm not sure the proper comparison is "running a 20+ year old executable". Can you compile an app from 20+ year old source and run it, might be the right choice.
  - PaulHoule 1y ago
    It's a general approach to life. A certain kind of customer wants to be able to run their old binaries, another kind of customer is happy to rebuild their software when they move a new platform. DEC for instance decided to give up on the VAX platform and figure that its customers would rebuild their software for the new chip. Problem is that if they can rebuild their software to run on Alpha/VMS, they also will consider rebuilding it to SPARC/Solaris or something. Apple switched from 68k -> Power PC -> x86 -> ARM and kept one generation of binary compatibility for each time, Microsoft's sorta kinda tried to an x86 -> ARM transition like that and has always failed. The 8086 and the IBM 360 are the only instruction set architectures which have maintained compatibility to before 1980. Part of the Wintel appeal is that stability.
bigyabai 1y ago
Boring isn't even what you got. I don't know a single person who considers a proprietary chip with proprietary graphics APIs is "boring" in 2025. My brothers in Christ, you are using the least-standardized chip out of everything commercially availible to consumers. You could switch your desktop to a PowerISA machine and still daily-drive a more boring chip than Apple Silicon. But this is what Mac owners wanted. No standards, no respect for third-party platforms, and zero documentation. A clean slate for a magic CPU, one filled with mystery and excitement that may persist beyond your lifetime.

4 points by CharlesW 1y ago

It is so different form the Windows and Android world where you go 10 years and performance goes sideways but maybe you have some accelerators you don’t use and if you are lucky you get some SIMD instructions that nobody uses except for hyperscalers and the national labs because they are just so slow to roll them out. X86 is increasingly like the VAX and ARM ex. Apple is increasingly asking us to believe they aren’t going to disappoint us the way they’ve disappointed us in each generation for the last decade. Apple is so lucky that they can quit supporting old chips and their fanbois celebrate but when Microsoft wants to use a 20 year old instruction the “who moved my cheese” brigade that keeps Microsoft relevant [1] accuses them of creating e-waste. [1] you can still run Office 97 binaries on Windows 11, you wouldn’t even try doing the same with MacOS classic or Linux binaries of that era.

bigyabai 1y ago
Those are tall words for describing a chip with... checks clipboard ...four discrete hardware-based SIMD accelerators.
- PaulHoule 1y ago
  Which one are we talking about? Apple is successful at getting the capabilities of their latest chips used in library code that applications use. For instance, back when Apple was running the legacy x86 architecture they packed in the Intel MKL libraries but Microsoft and Linux distributions didn't. MacOS comes with libraries that use the accelerators and applications often use those libraries so I've got some confidence that I'll actually use an accelerator that is on the M5 chip. In Wintel my expectation is that they will include some accelerator or other feature (say TSX or SGX) but it might be fused off or there won't be actual OS or library or application support or we might find out a year or two later that the feature doesn't actually work or leaks information. It's a nightmare we've been living for more than a decade. The fixed-width SIMD model that we're stuck with is a regression from the vector processors of the 1980s in that you have to rewrite your code with each generation of the instruction set. Couple that with Intel thinking AVX512 is just a "national labs and hyperscaler" feature but still adding it into consumer chips with it fused off -- too many people think this is business genius but it is not... if the customer isn't paying for the manufacturing of those fused off features the shareholder is. They would be better off not putting them in so ordinary customers get cheaper chips and telling the national labs they have to build a 2x bigger cluster.
  - bigyabai 1y ago
    > Which one are we talking about? You tell me. AVX and SSE are at-most two discrete hardware blocks and more often one. Apple Silicon, in it's genius RISC glory, possesses: 1) A Neural Engine 2) ARM NEON as Accelerate 3) Generic GPU SIMD as MPS 4) MatMul units in M5+ hardware It's an impressive amount of fragmentation for a single vendor that supposedly knows what they're doing. Even Nvidia doesn't suffer from this level of identity crisis, despite burning the compute/raster candle at both ends. Apple's approach explains their NASDAQ performance but not much else. It's a bit baffling.
bediger4000 1y ago
Given Linux resting on source code, I'm not sure the proper comparison is "running a 20+ year old executable". Can you compile an app from 20+ year old source and run it, might be the right choice.
- PaulHoule 1y ago
  It's a general approach to life. A certain kind of customer wants to be able to run their old binaries, another kind of customer is happy to rebuild their software when they move a new platform. DEC for instance decided to give up on the VAX platform and figure that its customers would rebuild their software for the new chip. Problem is that if they can rebuild their software to run on Alpha/VMS, they also will consider rebuilding it to SPARC/Solaris or something. Apple switched from 68k -> Power PC -> x86 -> ARM and kept one generation of binary compatibility for each time, Microsoft's sorta kinda tried to an x86 -> ARM transition like that and has always failed. The 8086 and the IBM 360 are the only instruction set architectures which have maintained compatibility to before 1980. Part of the Wintel appeal is that stability.

bigyabai 1y ago

Boring isn't even what you got. I don't know a single person who considers a proprietary chip with proprietary graphics APIs is "boring" in 2025. My brothers in Christ, you are using the least-standardized chip out of everything commercially availible to consumers. You could switch your desktop to a PowerISA machine and still daily-drive a more boring chip than Apple Silicon. But this is what Mac owners wanted. No standards, no respect for third-party platforms, and zero documentation. A clean slate for a magic CPU, one filled with mystery and excitement that may persist beyond your lifetime.