Recent content by n-type

If it's any help the BCD has options to make processor groups on single socket systems. It's not ideal and a little buggy but does have it's uses when not having access to a multi socket system. Single 2683v3, running on group 1 with 12 threads Note that distribution isn't always even. Also...

Life's not fair. I'd like a 2696v3 but used prices are currently twice what I would pay, boohoo :( If you were to OC your Xeons I would say "way to go BackupProphet", not "that's not fair" :D

Some laptop single core super Pi Slowest I could find 471 - 9907. IOW that's a pretty good score.

@BlackArchon yes it can help, for instance 1620 and 1630 v3's show as not overclockable under XTU hardware results. It's okay if someone has benched with XTU but still a few CPU's not listed such as the evasive 14 core 1691 v3. @Patriot could be the board holding your CPU back, what does...

Any links to an example of V3 with RAM at 2400MT/s or higher? First time I've heard it called that, usual name is MCE (Multi Core Enhancement) Would think it should work on earlier stepping too but perhaps people are not doing the BIOS work. You can sometimes check published XTU bench...

Have you seen this one? What controls Turbo Core in Xeons? It's quite a small thread and not so popular so would be easy to miss.

Hi @gtz The sad thing is it really isn't a secret, almost for the last 2 years now. Just never took off, don't know why. A couple of validated scores that can be checked against. CPU-Z MT bench validation 21528 Geekbench4 Linux x64 MC bench 34049 I've already upset one person here, for...

The errata has nothing to do with TSX and for obvious reasons isn't listed (think about it). Ah, now I get you. You refuse to believe the truth and nothing is going to persuade you otherwise. Maybe I should have picked a user name like "heretic" :D FWIW I think you do a great job on the...

Thanks @Murilo900. Hundreds might be a bit off the mark. :) FWIW I bought mine from a seller (not ebay) indicating a little under 10,000 available at the time for a price just under $330 including DHL international shipping. Due to the "Haswell" bug I would think all Haswell CPU's can boost...

Thanks. Yes, n-type and probably that's what has happened. Just didn't know what was going on at the time.

But the V3 has the possibility to boost all cores to 3.0GHz. If the V4 can do that as well then fair enough.

That's funny because I would think in Cinebench the v3 would have the means to beat the v4 and haven't seen to the contrary yet.

I have recently purchased an E5-2683v3 and while the CPU reports the FIVR supports core fixed (override) voltage it doesn't appear to work. i.e. if I set a fixed core voltage of 1.000V the VID still appears to be dynamic. Would someone shed some light on this, is it broken on all Xeons? Cheers.

FYI. Possibly registering via the login page is broken. When done this way I don't recall seeing an "I agree" point and the membership was "waiting admin approval". Email confirmation never eventuated and next day account was gone. Using the Signup page gave nearly instant email confirmation.