Lucky_n00bHi there, just wanted to make a small clarification here just in case this info is overlooked :)

I'm an overclocker working for small Indonesian-only OC site (oc.jagatreview.com), also sometimes I cooperate with vendors and other partners doing some overclocking tests for an upcoming pre-released products.
Both me and my friend IvanCupa do a joint LN2 session for 2990WX Extreme OC testing on the MSI X399 MEG to show some extreme OC result on the NDA Release day, I'm doing benchmarks (Cinebench R15/GPUPI for CPU 1B) and IvanCupa is doing CPU-Z clock validation.

For the CPU-Z Validation, please do note that:
- Yes, ALL 32-Core is including SMT is still enabled when booting from BIOS, but
- We searched or Best Cores using Ryzen Master, and downclocked all the other cores while doing so. So effectively it's only one core that clocked that high.
- We show this on the screenshot submitted on hwbot (core number 26 only overclocked, others downclocked)

It turned out that there's a difference of at least 150Mhz between 'best' cores and 'worst' cores for validation, and this is the best cores out of our 2990WX.
Due to LN2 and time constraint, we haven't tested doing LN2 validation for ALL-core turned to EXACT same speed. Would love to do that later though

Anyway, correct voltage was the one on MSI Command Center Lite. 1.68v. Temperature was Full pot (-184C or so)

Hopefully that clears things up

*)Sorry if bad English

Best Regards,


Alva

Thank you for the clarification :D Welcome to TPU :D
AMD never cease to amaze me of late.. (in a good way!!) I did see that core 26 was selected when doing the CPU-Z test.. How do you feel the CPU fares to the competition? Is it as good as it says? :)

HTCNot 6GHz but 5.387GHz: for 32 cores, that's very impressive, IMO.

Your not wrong :) I just wondered if there was a cherry picked CPU out there that would be able to manage 6Ghz or thereabouts :) It would be pointless for me to buy one, doesn't stop me wanting to buy one still tho lol :D

phillAMD never cease to amaze me of late.. (in a good way!!) I did see that core 26 was selected when doing the CPU-Z test.. How do you feel the CPU fares to the competition? Is it as good as it says? :)

From a perspective of an extreme LN2 overclocker to break score records, 2990WX would be strong for non-AVX heavily multithreaded 2D CPU benchmarks like Cinebench, GPUPI for CPU, and wPrime.
For 3DMark though, it's a bit of a hit and miss. Physics Score in 3DMark Fire Strike/ Time Spy Extreme Physics wasn't too good (Fire Strike Physics need to disable SMT apparently), was hoping to get higher score considering the massive core number this CPU has compared to the 7980XE.
Also for LN2 OC we need to use PCIe Gen 1 to avoid cold bug when benching lower than -100 or -120C, so multi-GPU 3DMark record was a no-go for now.

These things also eats massive amount of power during extreme OC, (I saw a 1 kW-ish draw at the wall during 5.4Ghz 1.65v Cinebench), and generate tons of thermal load. Would be a challenge to keep temperature as low as possible during bench. And benchmarking this thing at the limit would need strong VRM to say the least.

For ambient daily overclocking, at 4Ghz 1.3v I would really recommend to watch the VRM temperature during prolonged continuous use at those settings, getting a full-cover CPU-VRM monoblock would be ideal.

Lucky_n00bFrom a perspective of an extreme LN2 overclocker to break score records, 2990WX would be strong for non-AVX heavily multithreaded 2D CPU benchmarks like Cinebench, GPUPI for CPU, and wPrime.
For 3DMark though, it's a bit of a hit and miss. Physics Score in 3DMark Fire Strike/ Time Spy Extreme Physics wasn't too good (Fire Strike Physics need to disable SMT apparently), was hoping to get higher score considering the massive core number this CPU has compared to the 7980XE.
Also for LN2 OC we need to use PCIe Gen 1 to avoid cold bug when benching lower than -100 or -120C, so multi-GPU 3DMark record was a no-go for now.

These things also eats massive amount of power during extreme OC, (I saw a 1 kW-ish draw at the wall during 5.4Ghz 1.65v Cinebench), and generate tons of thermal load. Would be a challenge to keep temperature as low as possible during bench. And benchmarking this thing at the limit would need strong VRM to say the least.

For ambient daily overclocking, at 4Ghz 1.3v I would really recommend to watch the VRM temperature during prolonged continuous use at those settings, getting a full-cover CPU-VRM monoblock would be ideal.

Stupid question: did anyone think of using an AMD RX Vega 64 (@ non sub-ambient speeds 1st, to see if there was any point in doing it with LN2)?

I'm saying this because of what's described here:



RX Vega 64 should never beat the 1080 Ti and yet, as evidenced above, not only it does but it does so by quite a margin, in some cases. For reference, the pic was taken from this article.

Dunno if this can be reproduced in 3DMark but this strongly suggests some sort of driver issues with nVidia's GPU, no?

Lucky_n00bFrom a perspective of an extreme LN2 overclocker to break score records, 2990WX would be strong for non-AVX heavily multithreaded 2D CPU benchmarks like Cinebench, GPUPI for CPU, and wPrime.
For 3DMark though, it's a bit of a hit and miss. Physics Score in 3DMark Fire Strike/ Time Spy Extreme Physics wasn't too good (Fire Strike Physics need to disable SMT apparently), was hoping to get higher score considering the massive core number this CPU has compared to the 7980XE.
Also for LN2 OC we need to use PCIe Gen 1 to avoid cold bug when benching lower than -100 or -120C, so multi-GPU 3DMark record was a no-go for now.

These things also eats massive amount of power during extreme OC, (I saw a 1 kW-ish draw at the wall during 5.4Ghz 1.65v Cinebench), and generate tons of thermal load. Would be a challenge to keep temperature as low as possible during bench. And benchmarking this thing at the limit would need strong VRM to say the least.

For ambient daily overclocking, at 4Ghz 1.3v I would really recommend to watch the VRM temperature during prolonged continuous use at those settings, getting a full-cover CPU-VRM monoblock would be ideal.

From what I'm reading and can remember from my limited benchmarking days, the 3D Mark CPU tests and such are only really touching or using about 10 cores, they aren't using all of the cores available at all, so anything past a 8/10 core CPU is a waste for it..
I was wondering if AMD was having any issues with the cold bug?? From again what I remember there were no issues as such with the FX8xxx series CPUs, none that I recall that affected the PCIe slots anyways, but I could be wrong I'm not sure :)

With regards to the amount it pulls under load, I'm really not surprised it takes a chuck load more watts to run the tests at the higher clocks.. More cores, more speed, more power.. Somehow I think these higher capacity PSU's will still be used for things exactly like this..
VRM's I do wonder about slightly because of the amount of power that they will have to put up with.. I've always gone for proper overclocking boards for myself as I'm always worried that going for something a little less than overkill might limit my overclocks in someway.. Besides, at least I'd never worry about it going bang :)

I'd guess they'd get warm but if I'm honest, I think these CPUs perform so well at their stock speeds, that if you where overclocking them, water would probably be the only real way to go?? 500w of heat etc, is a lot to get rid of.... And to get rid of well and efficiently too...

GasarakiWhat a CLICKBAIT headline. It's 5.9 on ONE CORE, not all cores.

It was a mistake. Other sites made the same, but this one hasn't been udated.

@Raevenlord Time to update the article and headline?

bugSomebody let @Liviu Cojocaru know we need his "this doesn't mean anything to me" comments in here.

Did you find the meaning of life already?!

If you did I am impressed ;)

P.S doesn't really matter anyway...

"AMD Ryzen Threadripper 2990WX Overclocked to 5.955 GHz On All Cores"

Clearly only 1 core overclocked. The level of BS on this site lately is astounding.