Why aren't there any high performance desktop ARM CPUs?

>>58902891
The market is already saturated. There is not enough desktop and enterprise support for ARM architecture for it to be viable, yet.

>>58902891
>It seems like they have been shitting on x86 in most of the important categories such as efficiency, price, and heat

The ISA makes no difference in efficiency and heat, it's just that ARM designs have always targeted those areas over performance and x86 has always been for performance.

>>58903334
>x86 has always been for performance

What about the Atoms and Core M?

Power consumption does not matter in the desktop. So, ARM does not matter in the desktop.

>>58903334
>x86 has always been for performance.
No, x86 has always been for compatibility and legacy support.

>>58902891
CPU's don't drive profit, GPU's do tho

>>58903456
yea but what if I put 8 of them on a single motherboard?

>>58903511
8 ARM chips on a desktop?

Not him, but the vast majority of software can't take advantage of multiple processing cores, so you probably won't notice a difference after 2 or 3 arms

File: intel-xeon-phi-compare-table-1.jpg (89KB, 660x339px) Image search: [Google]

89KB, 660x339px

Because they can't compete, at all, especially to xeon phi chips.

Currently intel has knights landing which can output over 3,000 GFLOPS of FP64 and consume less than 300 watts. No ARM chip has gotten anywhere near that leven of performance and efficiency.

ARM processors have been and always will be angry birds processors.

>>58903457
see >>58903570

>>58903570
>1.05 GHz

what did they mean by this

>>58903631
>3,000+ GFLOPS of FP64 and less than 300 watts of power consumption

what did they mean by this

>>58902891

RISC is not really suited for high performance applications

>>58902891
Because the very essence of ARM is deliberately limited instruction set.