F5F Stay Refreshed Hardware Desktop Planning a PC Build: ASUS ROG STRIX X870E-E & DDR5-6600 CL32 — Seeking Compatibility Feedback & Advice

Planning a PC Build: ASUS ROG STRIX X870E-E & DDR5-6600 CL32 — Seeking Compatibility Feedback & Advice

Planning a PC Build: ASUS ROG STRIX X870E-E & DDR5-6600 CL32 — Seeking Compatibility Feedback & Advice

H
hrgriff
Senior Member
Find
573
06-02-2023, 04:18 AM
#1
Hi everyone,
I’m planning to set up a
workstation at home
. Until recently, I’ve been relying on a cloud server for compilations and heavy multi-threaded simulations, but the costs are adding up, so I’ve decided to build something similar locally.
Current cloud specs I use:
AMD Ryzen™ 9 7950X3D (16C/32T)
128GB ECC DDR5-5600
2×1.92TB NVMe (RAID1)
My workloads:
Building large C/C++/Rust codebases (high RAM capacity is critical).
Running heavy simulations across all cores for long periods — stability at max clocks/overclock, cooling, and reliability matter a lot.
Memory bandwidth and latency are important to speed up mapping/loading of large datasets.
Some generic ML training and data pipeline processing (GPU VRAM matters).
Mostly Linux, with some Windows use for browsing, documents, and
casual gaming
(smooth 1080p, acceptable 4K).
Proposed build so far:
CPU:
AMD Ryzen 9 9950X3D – 16 cores / 32 threads (AM5)
GPU:
ASUS PRIME GeForce RTX 5070 Ti 16GB GDDR7 OC Edition (DLSS 4)
Motherboard:
ASUS ROG STRIX X870E-E GAMING WIFI (ATX, AM5)
RAM:
Corsair Dominator Titanium RGB 96GB (2×48GB) DDR5-6600 CL32
Storage:
Samsung 9100 PRO 2TB PCIe 5.0 NVMe M.2 SSD
Cooler:
Corsair iCUE H170i Elite LCD XT 420mm Liquid Cooler
Case:
Fractal North XL Mid-Tower (Mesh Side Panel, Black)
PSU:
ASUS ROG STRIX 1000W 80+ Platinum, Fully Modular (ATX 3.1)
Extra Cooling:
Fractal Design 140mm ARGB Fan
It’s been a while since I last built a PC, and a lot has changed. I’d really appreciate feedback:
Does this configuration make sense for my workloads?
Anything overkill, wasteful, or potentially problematic?
Any changes you would recommend?
Budget (including currency):
3-4000 EUR
Reply
Reply
H
hrgriff
06-02-2023, 04:18 AM #1

Hi everyone,
I’m planning to set up a
workstation at home
. Until recently, I’ve been relying on a cloud server for compilations and heavy multi-threaded simulations, but the costs are adding up, so I’ve decided to build something similar locally.
Current cloud specs I use:
AMD Ryzen™ 9 7950X3D (16C/32T)
128GB ECC DDR5-5600
2×1.92TB NVMe (RAID1)
My workloads:
Building large C/C++/Rust codebases (high RAM capacity is critical).
Running heavy simulations across all cores for long periods — stability at max clocks/overclock, cooling, and reliability matter a lot.
Memory bandwidth and latency are important to speed up mapping/loading of large datasets.
Some generic ML training and data pipeline processing (GPU VRAM matters).
Mostly Linux, with some Windows use for browsing, documents, and
casual gaming
(smooth 1080p, acceptable 4K).
Proposed build so far:
CPU:
AMD Ryzen 9 9950X3D – 16 cores / 32 threads (AM5)
GPU:
ASUS PRIME GeForce RTX 5070 Ti 16GB GDDR7 OC Edition (DLSS 4)
Motherboard:
ASUS ROG STRIX X870E-E GAMING WIFI (ATX, AM5)
RAM:
Corsair Dominator Titanium RGB 96GB (2×48GB) DDR5-6600 CL32
Storage:
Samsung 9100 PRO 2TB PCIe 5.0 NVMe M.2 SSD
Cooler:
Corsair iCUE H170i Elite LCD XT 420mm Liquid Cooler
Case:
Fractal North XL Mid-Tower (Mesh Side Panel, Black)
PSU:
ASUS ROG STRIX 1000W 80+ Platinum, Fully Modular (ATX 3.1)
Extra Cooling:
Fractal Design 140mm ARGB Fan
It’s been a while since I last built a PC, and a lot has changed. I’d really appreciate feedback:
Does this configuration make sense for my workloads?
Anything overkill, wasteful, or potentially problematic?
Any changes you would recommend?
Budget (including currency):
3-4000 EUR

Reply
Reply
R
R3kty
Member
Find
133
06-03-2023, 08:21 AM
#2
Welcome to the forums, newcomer! Discussing stability at maximum clocks/overclocking isn't a straightforward process. Overclocking usually comes with some risk, especially in a server setup. Your main goal should be minimizing downtime—keeping things running smoothly at stock speeds, maybe even slightly undervolting. Does this setup suit your needs? Sharing the app names you plan to use will help us understand better. Is there anything excessive or risky about this configuration? Please format your questions here; this thread is meant for clarifying details requested in the discussion. *How to Ask for New Build or Upgrade Advice* This guide is based on the original *How To Ask For New Build Advice* post from Animal. It was made to address the growing number of users seeking upgrade guidance both here and on CPU forums, so we can offer helpful input. Forums. adds context to your situation, encouraging community support with relevant tips. Consider any adjustments you'd like to see—this site is based in the USA, and using local currency would be appreciated. We also recommend moving this thread to the Systems section for better organization.
Reply
Reply
R
R3kty
06-03-2023, 08:21 AM #2

Welcome to the forums, newcomer! Discussing stability at maximum clocks/overclocking isn't a straightforward process. Overclocking usually comes with some risk, especially in a server setup. Your main goal should be minimizing downtime—keeping things running smoothly at stock speeds, maybe even slightly undervolting. Does this setup suit your needs? Sharing the app names you plan to use will help us understand better. Is there anything excessive or risky about this configuration? Please format your questions here; this thread is meant for clarifying details requested in the discussion. *How to Ask for New Build or Upgrade Advice* This guide is based on the original *How To Ask For New Build Advice* post from Animal. It was made to address the growing number of users seeking upgrade guidance both here and on CPU forums, so we can offer helpful input. Forums. adds context to your situation, encouraging community support with relevant tips. Consider any adjustments you'd like to see—this site is based in the USA, and using local currency would be appreciated. We also recommend moving this thread to the Systems section for better organization.

Reply
Reply
P
Pokyyy
Junior Member
Find
17
06-07-2023, 09:02 PM
#3
Hi
@Lutfij,
Overclocking and stability often conflict. Overclocking usually comes with some risk. For a server setup, your priority should be minimizing downtime—keeping the system running smoothly at stock speeds, maybe slightly undervolting.
I want to be clear: overclocking will only be applied during specific testing periods for fun and experimentation, aiming to boost performance on certain benchmarks. In regular development or daily use, the system will mostly run in its default mode, focusing on stability. So, from my perspective, overclocking isn’t a major concern here. I’m looking for a reliable, smooth server for everyday tasks.
Does this setup align with your needs?

My workloads are mainly for development and research. I use it for compiling code with GCC/Clang, running data processing in Python, some ML pipelines with CUDA-optimized code, and long simulations of historical financial data. I also handle heavy tuning of computer vision and signal processing models.

Are there any changes you’d suggest?
The final location is in the UAE, but I’m open to importing parts from other regions. Here are some main vendors:

In the UAE:
https://gccgamers.com/
In Germany:
https://de.pcpartpicker.com/
Reply
Reply
P
Pokyyy
06-07-2023, 09:02 PM #3

Hi
@Lutfij,
Overclocking and stability often conflict. Overclocking usually comes with some risk. For a server setup, your priority should be minimizing downtime—keeping the system running smoothly at stock speeds, maybe slightly undervolting.
I want to be clear: overclocking will only be applied during specific testing periods for fun and experimentation, aiming to boost performance on certain benchmarks. In regular development or daily use, the system will mostly run in its default mode, focusing on stability. So, from my perspective, overclocking isn’t a major concern here. I’m looking for a reliable, smooth server for everyday tasks.
Does this setup align with your needs?

My workloads are mainly for development and research. I use it for compiling code with GCC/Clang, running data processing in Python, some ML pipelines with CUDA-optimized code, and long simulations of historical financial data. I also handle heavy tuning of computer vision and signal processing models.

Are there any changes you’d suggest?
The final location is in the UAE, but I’m open to importing parts from other regions. Here are some main vendors:

In the UAE:
https://gccgamers.com/
In Germany:
https://de.pcpartpicker.com/

Reply
Reply