06-05-2017, 08:53 AM
You don't even need benchmarks, it's a matter of very simple maths:
Ryzen7 1800X - 3600 Mhz, 8 cores, 2-way SMT, 128-bit registers == 3600000 * 8 * 2 * 4 == 230400000 32-bit IOPS
Core i7-7700K - 4200 Mhz, 4 cores, 2-way SMT, 256-bit registers == 4200000 * 4 * 2 * 8 == 268800000 32-bit IOPS
Core i7-7740X - 4300 Mhz, 4 cores, 2-way SMT, 256-bit registers == 4300000 * 4 * 2 * 8 == 275200000 32-bit IOPS
Core i7-7800X - 3500 Mhz, 6 cores, 2-way SMT, 256-bit registers == 3500000 * 6 * 2 * 8 == 336000000 32-bit IOPS
Core i7-5960X - 3000 Mhz, 8 cores, 2-way SMT, 256-bit registers == 3000000 * 8 * 2 * 8 == 384000000 32-bit IOPS
Core i7-6900K - 3200 Mhz, 8 cores, 2-way SMT, 256-bit registers == 3200000 * 8 * 2 * 8 == 409600000 32-bit IOPS
Core i7-7820X - 3600 Mhz, 8 cores, 2-way SMT, 256-bit registers == 3600000 * 8 * 2 * 8 == 460800000 32-bit IOPS
So it's literally the slowest of all the high-end desktop CPUs. And of course if you go to the workstation/server side of things, AMD can't even begin to compete with the E5-2630v4, let alone shit like the E5-2699v4. But we were only talking about mainstream CPUs so I won't get into those -- but come on man, the 2699v4 is like 6x faster than the 1800X, so let's not act like AMD is doing anything exciting or innovative.
Also re: the CCX interconnect, I can't believe you would tout that as a positive attribute of the chip architecture. For most hash algorithms we keep everything in registers and in cache, we don't use off-chip memory for the hash calculations. This usually means memory type and speed do not play any real role in the actual hash calculation. HOWEVER, because the Infinity Fabric speed is linked to the off-chip memory speed, you MUST use the fastest DDR4 possible and OC the hell out of it when using more than 4 cores to get the most performance out of the chip. That is ABSURD.
Ryzen7 1800X - 3600 Mhz, 8 cores, 2-way SMT, 128-bit registers == 3600000 * 8 * 2 * 4 == 230400000 32-bit IOPS
Core i7-7700K - 4200 Mhz, 4 cores, 2-way SMT, 256-bit registers == 4200000 * 4 * 2 * 8 == 268800000 32-bit IOPS
Core i7-7740X - 4300 Mhz, 4 cores, 2-way SMT, 256-bit registers == 4300000 * 4 * 2 * 8 == 275200000 32-bit IOPS
Core i7-7800X - 3500 Mhz, 6 cores, 2-way SMT, 256-bit registers == 3500000 * 6 * 2 * 8 == 336000000 32-bit IOPS
Core i7-5960X - 3000 Mhz, 8 cores, 2-way SMT, 256-bit registers == 3000000 * 8 * 2 * 8 == 384000000 32-bit IOPS
Core i7-6900K - 3200 Mhz, 8 cores, 2-way SMT, 256-bit registers == 3200000 * 8 * 2 * 8 == 409600000 32-bit IOPS
Core i7-7820X - 3600 Mhz, 8 cores, 2-way SMT, 256-bit registers == 3600000 * 8 * 2 * 8 == 460800000 32-bit IOPS
So it's literally the slowest of all the high-end desktop CPUs. And of course if you go to the workstation/server side of things, AMD can't even begin to compete with the E5-2630v4, let alone shit like the E5-2699v4. But we were only talking about mainstream CPUs so I won't get into those -- but come on man, the 2699v4 is like 6x faster than the 1800X, so let's not act like AMD is doing anything exciting or innovative.
Also re: the CCX interconnect, I can't believe you would tout that as a positive attribute of the chip architecture. For most hash algorithms we keep everything in registers and in cache, we don't use off-chip memory for the hash calculations. This usually means memory type and speed do not play any real role in the actual hash calculation. HOWEVER, because the Infinity Fabric speed is linked to the off-chip memory speed, you MUST use the fastest DDR4 possible and OC the hell out of it when using more than 4 cores to get the most performance out of the chip. That is ABSURD.