From: cr88192@gmail.com   
      
   On 9/19/2025 4:50 AM, Anton Ertl wrote:   
   > BGB  writes:   
   >> On 9/17/2025 4:33 PM, John Levine wrote:   
   >>> According to BGB  :   
   >>>> Still sometimes it seems like it is only a matter of time until Intel or   
   >>>> AMD releases a new CPU that just sort of jettisons x86 entirely at the   
   >>>> hardware level, but then pretends to still be an x86 chip by running   
   >>>> *everything* in a firmware level emulator via dynamic translation.   
   >>>   
   >>> That sounds a whole lot like what Transmeta did 25 years ago:   
   >>>   
   >>> https://en.wikipedia.org/wiki/Transmeta_Crusoe   
   >>>   
   >>> They failed but perhaps things are different now.  Their   
   >>> native architecture was VLIW which might have been part   
   >>> of the problem.   
   >>>   
   >>   
   >> Might be different now:   
   >> 25 years ago, Moore's law was still going strong, and the general   
   >> concern was more about maximizing scalar performance rather than energy   
   >> efficiency or core count (and, in those days, processors were generally   
   >> single-core).   
   >   
   > IA-64 CPUs were shipped until July 29, 2021, and Poulson (released   
   > 2012) has 8 cores.  If IA-64 (and dynamically translating AMD64 to it)   
   > would be a good idea nowadays, it would not have been canceled.   
   >   
   > How should the number of cores change anything?  If you cannot make   
   > single-threaded IA-32 or AMD64 programs run at competetive speeds on   
   > IA-64 hardware, how would that inefficiency be eliminated in   
   > multi-threaded programs?   
   >   
   >> Now we have a different situation:   
   >>    Moore's law is dying off;   
   >   
   > Even if that is the case, how should that change anything about the   
   > relative merits of the two approaches?   
   >   
   >>    Scalar CPU performance has hit a plateau;   
   >   
   > True, but again, what's the relevance for the discussion at hand?   
   >   
   >>    And, for many uses, performance is "good enough";   
   >   
   > In that case, better buy a cheaper AMD64 CPU rather than a   
   > particularly fast CPU with a different architecture X and then run a   
   > dynamic AMD64->X translator on it.   
   >   
      
   Possibly, it depends.   
      
   The question is what could Intel or AMD do if the wind blew in that   
   direction.   
      
   For the end-user, the experience is likely to look similar, so they   
   might not need to know/care if they are using some lower-power native   
   chip, or something that is internally running on a dynamic translator to   
   some likely highly specialized ISA.   
      
      
      
   >>    A lot more software can make use of multi-threading;   
   >   
   > Possible, but how would it change things?   
   >   
      
   Multi-threaded software does not tend to depend as much on single-thread   
   performance as single threaded software...   
      
      
   >> Likewise, x86 tends to need a lot of the "big CPU" stuff to perform   
   >> well, whereas something like a RISC style ISA can get better performance   
   >> on a comparably smaller and cheaper core, and with a somewhat better   
   >> "performance per watt" metric.   
   >   
   > Evidence?   
   >   
      
   No hard numbers, but experience here:   
   ASUS Eee (with an in-order Intel Atom) vs original RasPi (with 700MHz   
   ARM11 cores).   
      
   The RasPi basically runs circles around the Eee...   
      
      
   Though, no good datapoints for fast x86 emulators here.   
      At least DOSBox and QEMU running x86 on RasPi tend to be dead slow.   
      
      
      
   ( no time right now, so skipping rest )   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)