iMac Pro RAM

[Chaszmyr]

I'm under the impression that the iMac Pro uses 4 channel memory, but is only equipped with 2 DIMMs in a 32gb configuration.

Can anyone verify this? And if it's true, is the performance boost of 64gb RAM (4 DIMMs) significant enough to consider if 32gb RAM is enough memory?


EDIT: Apparently the 32gb iMac Pro actually uses 4x8gb

 

[FredT2]

I'm under the impression that the iMac Pro uses 4 channel memory, but is only equipped with 2 DIMMs in a 32gb configuration.

Can anyone verify this? And if it's true, is the performance boost of 64gb RAM (4 DIMMs) significant enough to consider if 32gb RAM is enough memory?

Geekbench scores so far show no difference in 32 GB vs 64 GB in the 8-core machines.

 

[jerwin]

Geekbench scores so far show no difference in 32 GB vs 64 GB in the 8-core machines.

Someone by the name of Linus Torvalds once said....

Geekbench is SH*T.

It actually seems to have gotten worse with version 3, which you should be aware of. On ARM64, that SHA1 performance is hardware-assisted. I don't know if SHA2 is too, but Aarch64 does apparently do SHA256 in the crypto unit, so it might be fully or partially so.

And on both ARM and x86, the AES numbers are similarly just about the crypto unit.

So basically a quarter to a third of the "integer" workloads are just utter BS. They are not comparable across architectures due to the crypto units, and even within one architecture the numbers just don't mean much of anything.

And quite frankly, it's not even just the crypto ones. Looking at the other GB3 "benchmarks", they are mainly small kernels: not really much different from dhrystone. I suspect most of them have a code footprint that basically fits in a L1 cache.

Linus

so tell me why more memory would help?

observing Activity monitor, geekbench_x86_64 uses no more than 640 MB (and for most of the tests, no more than 150 MB). A benchmark that shows the benefits of more memory would demand a lot more memory.

I would imagine that a properly designed benchmark would test a multitude of different assumptions and report the best results, so as to help the user identify bottlenecks.