The stuff Hacker News actually recommends.

GPUs and CPUs

AMD Ryzen vs Intel Xeon

Owners are split: the signals below point different ways.

AMD RyzenIntel Xeon
Owners with an opinion35543
Positive verdicts77% of 26589% of 36
Positive/negative since 2024109/2416/2
Happy after 3+ years54
Longest happy owner7 yrs10 yrs
Praised forperformance and reliabilityperformance and value
Criticized fornoisenoise
Newest comment

Updated

What owners wrote about both

AMD Ryzen: Likes itIntel Xeon: Doesn’t like it
On the hardware side of things not using server grade stuff really isn't as big of a deal these days. I'd happily take a decent Ryzen 5 or 7 series over a "new" Xeon that has twice the power consumption and mysteriously the same specs as an older Xeon made a decade ago. Even ECC - for 99% of applications (and especially on low-end VPS servers) its less likely to be a problem. The only thing I have found to be an issue with Hetzner is on dedicated servers, and specifically the hard drives. I've had new servers provisioned and they've given me decade old drives that are on the verge of failure - it's less of an issue now as most of their servers are shipping with new nvme drives but I dare say in 3-4 years time it'll be a problem when they reuse those and have instant non-recoverable failures for some of the hardware range.
Switched from the Intel Xeon to the AMD RyzenAMD Ryzen: Likes it
By virtue this is nice and noble. But objectively you're likely not doing you a favor. To answer your question, and to also give you an example: In early 2020 I passed my 2012 Xeon machine down to my SO to replace her 2007 Celeron. Technically, the machine was "good enough" for what I did, but for WFH I needed more RAM if I wanted to avoid VNCing (via VPN) into my office computer to do my work - which really annoyed me. Upgrading 4x4GB DDR3 to 4x8GB DDR3 would have made no economical sense (even used), so I switched to AM4 (MB/CPU/RAM). Because I hate waste I already planned for a future upgrade path (= decent mainboard) and got a really cheap Ryzen 1600 AF (still 50% faster than my old CPU) instead of a more expensive 2800X. Back then I planned to eventually get a 3800 (wasn't released then), now the 5800X or better are an option. That was for work. The machine is actually my personal computer (work data is encrypted on Linux), so that also does gaming duty (got new GPUs 2007, 2014 and now 2021; old GPUs have been passed down to family/friends). For hobby I still have a 2009 Xeon in the NAS (got it from a friend), which will probably inherit the Ryzen 1600 CPU once I upgrade the desktop. For the sake of your eyes: Get a decent TFT. Also more screen estate will make you more productive. I usually recommend 2x27" WQHD or 4k for pure work stuff. At home I use 1x34" UWQHD instead (because two screens are not good for gaming).
AMD Ryzen: Likes itIntel Xeon: Likes it
Programs whose performance is dominated by array operations, as it is the case for most scientific/technical/engineering applications, achieve a much better energy efficiency on the AMD or Intel CPUs with good AVX-512 support, e.g. Zen 5 Ryzen or Epyc CPUs and Granite Rapids Xeons, than on almost all ARM-based CPUs, including on all Apple CPUs (the only ARM-based CPUs with good energy efficiency for such applications are made by Fujitsu, but they are unobtainium). So if you want maximum energy efficiency, you should choose well your CPU, but a prejudice like believing that ARM-based CPUs are always better is guaranteed to lead to incorrect decisions. The Apple CPUs have exceptional and unmatched energy efficiency in single-thread applications, but their energy efficiency in multi-threaded applications is not better than that of Intel/AMD CPUs made with the same TSMC CMOS fabrication process, so Apple can have only a temporary advantage, when they use first some process to which competitors do not have access. Except for personal computers, the energy efficiency that matters is that of multi-threaded applications, so there Apple does not have anything to offer.
AMD Ryzen: Likes itIntel Xeon: Likes it
> Biggest jump would be AVX512 support on the newer CPUs. Didn't this get removed from the "consumer" 12th gen Intel models? Regarding the speed of the AMD 5000 series, I have to agree. I only touched a laptop part, and an U at that, but it was the first new CPU that impressed me. My daily driver before that was a 3rd or 4th gen Xeon, and it was faster than the i5 desktops we had a work.
AMD Ryzen: Doesn’t like itIntel Xeon: Doesn’t like it
Haswell Xeon is way too outdated in 2022, but there are loads of newer ones. There's a fair inventory of Skylake and Cascade Lake Xeon on ebay, like HP Z4 G4 w/ 32GB and a GPU for ~$750. It's a pretty square deal considering. Since Ice Lake Xeon workstations are so thin on the ground, this is the most recent Xeon you can get second-hand. The Ryzen story is a joke if you are actually in the market for a workstation, meaning you want ECC that works. The only way to get it is to buy integrated OEM workstations with the TR Pro.
AMD Ryzen: Likes itIntel Xeon: Likes it
Depending on what you do, you may be impressed with zen 3 u cpus. I was in the same boat, with an old 8-core Xeon something from like 2013 (it takes DDR3 RAM, and was one of the first to support PCIE 3). It was so fast, it ran circles around my current work computers (basic enterprise i5 HP). Then I insisted on a zen 3 laptop at the end of last year, and for rust compiles, it's almost twice as fast as the Xeon. It only has a 5650U (6 cores + HT), and a teeny-tiny heat sink. It does get somewhat noisy under load, but it's dead silent most of the time, including in Teams video chats!
AMD Ryzen: Likes itIntel Xeon: Likes it
>which is insecure Splitting IOMMU groups is not inherently insecure; doing so without ACS is insecure. It is true that's not the motherboard vendors fault - it is because Intel designed it that way. ACS (the part of PCIe which makes sane IOMMU group splitting possible) works great on my Zen products (Ryzen, Threadripper, EPYC, etc.) It also works fine on my Intel Xeon products. Intel client silicon is the odd man out here, and if you ask me that's hardly an accident. It falls neatly in line w/ their other artificial market segmentation. (They do the exact same thing w/ ECC RAM.) I'm mostly just jaded because my use-case (PCIe passthrough for a 'gaming VM') is hardly conducive to enterprise grade hardware (Xeon SKUs w/ low base clocks, headless GPUs designed for compute, etc.) yet vendors like NVidia and Intel choose to make this use-case difficult/impossible to implement on their consumer grade offerings. It gets old after a while. I, for one, am really glad AMD shook things up in this regard in the CPU space.