AMD Ryzen Threadripper PRO 7995WX Linux Performance Benchmarks

[DavidBrown]

sophisticles is just trolling. Calling server CPUs a scam and also criticizing EPYC I/O subsystem when in fact Intel server platforms have a weaker I/O shows either explicit trolling or a lack of fundamental hardware knowledge. Either way no point to engage.

I always try to assume people are genuine until the evidence shows otherwise, and even with trolls it can be useful to correct matters of fact - after all, there can be many other people reading these posts who don't know the real information (we all start ignorant before we've learned about a particular topic). I rarely care much about the opinions or posts of single individuals (though it is always satisfying when you know you have helped someone learn or solve a problem), but I do care about others reading the thread. Someone coming along later, having found this thread because he or she was searching for information about high core count processors to decide what kind of system they need for a demanding task, should not be left with the impression that HPC cpus are all a scam.

But I think those points have been made now by page 7 of this thread, and most other readers will already have dropped off and gone somewhere else!

[sophisticles]

sophisticles is just trolling. Calling server CPUs a scam and also criticizing EPYC I/O subsystem when in fact Intel server platforms have a weaker I/O shows either explicit trolling or a lack of fundamental hardware knowledge. Either way no point to engage.

Threadrippers are not server CPUs. they should be renamed RipperOffers because that's what they are for, ripping people off that do not know better.

They are a scam, perpetrated by AMD to sell as much worthless product as possible,

Why do they even exist? It's so that AMD can re-badge defective EPYC CPU's and sell then at a somewhat cheaper price point.

As for I/O, it stands for Input / Output and it's limited by your drive speed, AMD and Intel should have similar I/O performance all other things being equal.

[drakonas777]

Threadrippers are not server CPUs.

I was referring to this specific part of the discussion:

Displaying things on a screen actually has a higher latency than writing to a high performance nvme device. Most servers don't really write all that much data to disk, most of the data is actually served over network. The notion that "eventually the data has to be copied from volatile memory to nonvolatile and this takes time" and that this is somehow detrimental to HPC CPUs is outright ridiculous.

Have you considered migrating your screen name to something more befitting your reality, like for example simplisticles?

Reread what i said and try to understand what i mean, i am saying that in many ways, these HPC C{U's are a scam because they rely on vast amounts of memory to mask the limitations of I/O.

Yes, obviously Threadrippers are not server CPUs, but you responded about server/HPC CPUs in the same manner, calling them a scam. Does not make your statements about Threadripper CPUs being a scam any less absurd though.

they should be renamed RipperOffers because that's what they are for, ripping people off that do not know better.

They are a scam, perpetrated by AMD to sell as much worthless product as possible

There are use cases where these CPUs make perfect sense. I agree that these use cases are not mainstream, so perhaps they are worthless for you.

Why do they even exist? It's so that AMD can re-badge defective EPYC CPU's and sell then at a somewhat cheaper price point.

Yes, though most of them (TR chiplets) are not defective, just not good enough to pass validation for EPYC line. Also, for some niche cases they are preferable for cost savings and higher frequency at a cost of power usage.

As for I/O, it stands for Input / Output and it's limited by your drive speed, AMD and Intel should have similar I/O performance all other things being equal.

Good to know, I had no idea what I/O means (sarcasm). Both Intel server and HEDT/WS platforms have less PCIe lanes so overall per-socket I/O is a bit weaker all other things being equal. You can combine multiple disks and/or NICs to improve IO performance using those lanes. Because, you know, people who buy 64 cores CPU usually don't use it with a single fucking Silicon Power value series 256G SATA drive.

[qarium]

i really feel sorry for you. Years ago i worked in a PICU, an involuntary commitment unit and i have seen people going through episodes like the one you are going through.
I mean this, if you should experience a moment of clarity, please check yourself into a hospital, I believe you may be a danger to yourself or others.
I will not be responding to you any longer, and I urge everyone else in this forum to also not engage you, you are clearly having trouble distinguishing between reality and fantasy and I don't know the cause of it but I do worry that you may be triggered in real life by something you read on this forum and possibly injure yourself.
Very sad case, I hope you get better.

CVE-2023-6212

https://github.com/advisories/GHSA-4cv2-qh42-x2j4

"
CVE-2023-6212
Memory safety bugs present in Firefox 119, Firefox 115.4,...

Unreviewed Published Nov 21, 2023 to the GitHub Advisory Database • Updated Nov 23, 2023

Package
No package listed— Suggest a package

Affected versions
Unknown

Patched versions
Unknown

Description
Memory safety bugs present in Firefox 119, Firefox 115.4, and Thunderbird 115.4. Some of these bugs showed evidence of memory corruption and we presume that with enough effort some of these could have been exploited to run arbitrary code. This vulnerability affects Firefox < 120, Firefox < 115.5, and Thunderbird < 115.5.0.
References

https://nvd.nist.gov/vuln/detail/CVE-2023-6212

https://bugzilla.mozilla.org/buglist...%2C1829252%2C1 856072%2C1856091%2C1859030%2C1860943%2C1862782

https://www.mozilla.org/security/advisories/mfsa2023-49/

https://www.mozilla.org/security/advisories/mfsa2023-50/

https://www.mozilla.org/security/advisories/mfsa2023-52/

https://www.debian.org/security/2023/dsa-5561

Published by the National Vulnerability Database Nov 21, 2023

Published to the GitHub Advisory Database Nov 21, 2023

Last updated Nov 23, 2023

Severity
Unknown

Weaknesses
No CWEs

CVE ID
CVE-2023-6212

GHSA ID
GHSA-4cv2-qh42-x2j4

"