Greetings,

$4K for a Threadripper CPU???? Wam, bam, NO thank you ma'am!

I'll stick with Intel. They don't have anything at that price, at least not that I have seen.

Have a great day! :)

Siran

That $4000 Threadripper CPU outperforms 2x $10,000 Intel Xeon CPUs (Xeon Platinum 8180)

Not sure where you get the idea that Intel doesn’t have anything at that price.

The Xeon W-3175X (a $3000 CPU) gets absolutely destroyed by this Threadripper.

The CPU is expensive. But just because you don’t have a use for it, doesn’t mean that others don’t. If you have a need for the power this CPU offers and can deal with the limitations on memory and PCIe lanes, then this is the best bang for buck CPU on the market.

Hi Ian,

I did say "at least not that I have seen."

AMD is just trying to do as Intel has done for decades. Overprice their CPUs.

Is the 3175X a 7nm chip?

Have a great day! :)

Siran

ot sure if this is the right place amongst all the motherboard/mainboard/BIOS talk but here's an interesting article for some of the discussion points:


Talking Threadripper 3990X With AMD's Robert Hallock

... Intel is woefully incapable of responding to AMD's existing 32-core Threadripper 3970X, so dropping the 64-core 3990X is akin to delivering a killing blow in HEDT: AMD says this single processor can outperform two of Intel's $10,000 Xeon 8280's in some workloads...

... AMD noticed that customers tend to jump right to the top of the stack or choose the "sweet spot" product...

... Hallock explained that the bottleneck "suddenly shifts to really weird places once you start to increase the core counts." Some areas that typically aren't a restriction, like disk I/O, can hamper performance in high core-count machines. For instance, development houses exclude project files from Windows Defender to reduce the impact of disk I/O during compile workloads.

Hallock doesn't see memory throughput as a limitation in the majority of workloads, but told us that memory capacity per core is actually one of the biggest challenges. As a result, Hallock said that compile workloads require a bare minimum of 1GB per core (64GB total), and that 2GB (128GB total) is better...

... AMD has also made architectural decisions to keep data on-chip, like an almost unthinkable 288MB of total cache, a third AGU (Address Generation Unit), and larger opcaches, among other changes, that help keep data in the caches...

... AMD's official launch date lands on February 7, 2020...



Happy fast cool crunchin'!
Martin

Omg guys AMD power the new threadrippers this is insane :O

https://www.youtube.com/watch?v=oKYY37ss3lY

https://www.youtube.com/watch?v=a8apEJ5Zt2s

Hahahahahah LOL, luv Linus' "you're a weasel and a chicken" rant about Intel.

As expected, another solid Zen 2 based product launch. Best of the binned Zen 2 chips going into the two TR3 SKU's.

If you just want to populate the existing slots on a AMD motherboard without all the mining hardware, then the AMD platform is a nice, inexpensive platform for BOINC that gives a good balance of cpu and gpu capability.

If you want to drive a boatload of gpus with mining hardware then a simple Intel platform is best because it has the PCIe lanes to accommodate. And you have to buy into the AMD HEDT ecology (Threadripper) to achieve similar with higher costs than just Ryzen.

Overclocked Ryzen 9 3950X outperforms 24-core 2970WX in Cinebench (It beats Intel's i9-9980XE, too).

In TechSpot’s own testing, both the 3900X and 3700X maxed out at 4.3 GHz on a 360mm AIO.
And finally, to the results. The 3950X simply obliterates AMD’s current 16-core monster, the Threadripper 2950X, scoring 24% higher. It falls a mere 10% behind the 24-core 2970WX. Overclocked to 4.3 GHz results are even more impressive as it can improve its score by 11%, pushing 1% past the 2790WX.

According to online results for the 18-core i9-9980XE, that chip can only achieve a ~3700 score at stock and ~4200 overclocked, both of which are about 200 points behind the 3950X when similarly configured.

So a TR 2950X would have to drop significantly in price to become competitive in the "bang for buck" arena.

Tom

Interesting speculative video over at the AdoredTV YT channel. Posits the reason why there is limited availability of 3900X parts and the delayed release of the 3950X part to November is because of demand for every good 8 core chiplet going to Epyc parts where demand is massive and more than was expected by a large margin. Absolutely no reason to sell two 8 core chiplets as a 3950X part when they could be used in a $7000 Epyc cpu.

I agree with every point made in the video with sound reasoning. So I do not expect the 3950X to be anything more than a paper launch in November. Will not have any availability on that part either. The defective dies in the 3900X part are the bottom of the barrel performing dies coming off the wafer. Everything else is going directly to Epyc. And the reason that the TR3 part is only going to be 24 cores. Again the bottom of the barrel dies are going into that cpu. And I also agree the 3950X will not sell at its original $750 MSRP. Try $999.

Another point brought up is the decision by Global Foundries to stop their 7nm ramp up. Very bad business decision in my opinion. The demand for 7nm from all sectors is outstripping supply by massive amounts.