By: Doug S (foo.delete@this.bar.bar), August 4, 2014 8:14 pm
Room: Moderated Discussions
Exophase (exophase.delete@this.gmail.com) on August 4, 2014 3:20 pm wrote:
> x86 CPUs have a dedicated memory controller, if you want to run another processor at the same
> time it's going to need its own memory. At a minimum. You'd run into other serious problems
> when trying to run them as different clusters of cores under the same OS, like a lack of coherency
> between the two, and not having a unified interface to the same peripherals.
Is that really a showstopper? Couldn't Apple design the memory controller on their SoC to be a buffer for Intel's memory controller? The x86 chip would think it is talking directly to the RAM, but it would be using a go-between. Similar to how buffered DIMMs work (I think)
Its a performance hit, yes, but people aren't buying Macs for top performance anyway, so losing a few percent due to increased memory latency isn't really a big deal.
> x86 CPUs have a dedicated memory controller, if you want to run another processor at the same
> time it's going to need its own memory. At a minimum. You'd run into other serious problems
> when trying to run them as different clusters of cores under the same OS, like a lack of coherency
> between the two, and not having a unified interface to the same peripherals.
Is that really a showstopper? Couldn't Apple design the memory controller on their SoC to be a buffer for Intel's memory controller? The x86 chip would think it is talking directly to the RAM, but it would be using a go-between. Similar to how buffered DIMMs work (I think)
Its a performance hit, yes, but people aren't buying Macs for top performance anyway, so losing a few percent due to increased memory latency isn't really a big deal.