I'm assisting a composer who asked me to design him a super kick-ass PC slave for VSL. He wants to replace his three quad-core PC slave machines with a single PC slave, and he's willing to spend some dough.
What is the core speed VS thread count performance trade-off for VSL PC slave machines like the one I'm building?
For those who are curious, here is what the new build is currently looking like. (I've bolded the elements that I tried to max out that normally bottleneck system performance):
- CPU: AMD EPYC 7302
- Cores: 16
- Threads: 32
- Base Clock: 3.0GHz
- Max Boost Clock: 3.3GHz
- L3 Cache: 128MB
- RAM: 256GB 2666MHz Crucial
- LDIMM
- 8x 32GB modules
- Motherboard: AsRock EPYCD8-2T
- Includes dual 10Gbe Ethernet ports
- Eight Channel DDR4 memory
- Boot Drive: SAMSUNG 970 PRO 500GB
- Storage: 4x 2TB Samsung 860 EVO
- 8TB total
- Configured in software RAID 0 (striped storage for increased speed)