A call for modern CPU benchmarks

[altenbach]

OK, I changed the download to a shared folder on google drive. That should satisfy the https requirement 😄

[altenbach]

So for fun I added a few more columns (e.g. max turbo frequency and TDP) and added two derived columns:

• S/GHz: Serial single core performance normalized to the boost clock frequency. A better CPU architecture will give a higher score. This assumes that the serialized speed runs at max boost. Once all cores run in parallel, the clock might be reduced for thermal reasons.

• P/TDP: Parallel Spectra/seconds normalized to TDP. Higher is more efficient.

[Craig_]

Here's some more results 🙂

[AeroSoul]

---

@altenbach wrote:

---

---

Not sure what you need. What kind of error are you getting with the current link?

---

No error, it just wouldn't download. No message about it being blocked either, opening the link manually just closes the tab.

The google drive link worked tho, thanks.

[altenbach]

---

@Craig_ wrote:

Here's some more results 🙂

---

Wow, the 3970X single-handedly (or is that 32 core-ly) dethroned the leader and landed in first place with room to spare!

Of course at $3k+ for the CPU alone, we don't expect anything less. 😄

[Intaris]

Xeon E-2286G @ 4GHz, while performing a local ISE FPGA compile (1 Core pegged to 100% already).

[Intaris]

It may be unrelated, but I had to install LV 2020 runtime for the benchmark and since them my LVCompare (LV 2015 SP1) won't work...... Anyone had similar experiences?

[wiebe@CARYA]

LVCompare.exe is in Shared (C:\Program Files\National Instruments\Shared\LabVIEW Compare), so it probably got updated beyond compatibility...

[Intaris]

But surely the LV 2020 runtime doesn't touch LVCompare....? That is an IDE thing, not run-time, right?

Sorry, I don't mean to hijack the thread.

[wiebe@CARYA]

---

@Intaris wrote:

But surely the LV 2020 runtime doesn't touch LVCompare....? That is an IDE thing, not run-time, right?

---

That does make sense... Does the exe work (though the command line)?

---

@Intaris wrote:

Sorry, I don't mean to hijack the thread.

---

Indeed, maybe start a new thread...