AMD EPYC 9755 / 9575F / 9965 Benchmarks Show Dominating Performance

[Mark Rose]

Quite.
6980P Processor Sets Record $17,800 Flagship Price

Higher power consumption and less performance than
AMD EPYC Genoa 9684X, €4112.14 ex VAT

I actually feel sorry for Intel, these results are devastating.

Intel has advantages in areas, like more memory bandwidth, their accelerators, so on.

[HEL88]

Tonight....stock

Objectively, on October 10, for the launch of the new amd epycs, that AMD's shares fell sharply(!) by 4%, and Intel's by only 1%.

See the charts for yourself. Investors don't have your optimism .

[numacross]

BTW, I think it was rumored that Intel was working on a nonstandard MCR-DIMM-based solution. I wonder if that was ever true or maybe just an intentionally leaked red herring.

I was wondering that too, and according to ServeTheHome what Intel/Micron call "MRDIMM" is actually MCR DIMM:

Of course, we should take a second and just note that there is a big caveat here, and it is somewhat strange. JEDEC will have a MRDIMM spec, and the MRDIMMs for Xeon 6 with P cores is actually not that spec, instead it is the MCR DIMM spec. When we use the MRDIMM with Xeon 6, we really mean MCR DIMM. The concepts are largely similar, but it is unlikely when AMD supports MRDIMMs that you will take a Xeon 6 MRDIMM (really a MCR DIMM) and be able to use it in an AMD system.

As noted in the article it's also confirmed by Micron's website:

Q: Is this MRDIMM compliant with the JEDEC standard for MRDIMM?
A: No. JEDEC has not yet released the standard for MRDIMM. This is the first version of MRDIMM that supports Intel^® Xeon^® 6 processors.

[coder]

Objectively, on October 10, for the launch of the new amd epycs, that AMD's shares fell sharply(!) by 4%, and Intel's by only 1%.

See the charts for yourself. Investors don't have your optimism .

I never said anything about AMD stock. Your statement presumes each started from a roughly similar investor/market sentiment and were otherwise uniformly affected by other market dynamics, which aren't assertions I either made or would endorse. I could venture some weakly-informed statements about it, but that's really sophisticles' game, not mine. Plus, ain't you ever heard the old saying: "buy on the rumor, sell on the news"?

Ultimately, this isn't an investing forum and I'd prefer it stay that way.

[coder]

I was wondering that too, and according to ServeTheHome what Intel/Micron call "MRDIMM" is actually MCR DIMM:

Wow, that's hugely helpful! I was wondering how Intel managed to pull a coup and switch to MR-DIMM, beating AMD at their own game. Turns out they didn't and the earlier MCR-DIMM news was spot on!

Geez, what crazy disinformation: for them to be going around and calling MCR-DIMM like it's the same thing AMD/JEDEC is working on!

[mSparks]

Objectively, on October 10, for the launch of the new amd epycs, that AMD's shares fell sharply(!) by 4%, and Intel's by only 1%.

See the charts for yourself. Investors don't have your optimism .

cool, its been maybe a decade or so since Ive seen someone openly running a buy high sell low trading strategy.

Must be _that_ phase of the business cycle.

[caligula]

Intel enjoyed the performance crown for what..? Two weeks?

There are still companies and government agencies that must buy Intel even if it uses more power, is much slower and more expensive. Simply because it's Intel.

[Anux]

To bad for Intel, I thought they would have a much better chance this time but Zen 5 seems to do particularly well in server workloads.
A single socket Zen 5 matches the best Intel 2P with high end RAM. From the gains in desktop CPUs I wouldn't have guessed that.

Let's hope they can recover like AMD did after becoming fabless.

[sophisticles]

Tonight, sophisticles cried himself to sleep. He'd poured his life's savings into Intel stock, after his crack statistical analysis of its share price convinced him it was a sure bet.

It wasn't so much the sense of existential dread he experienced upon seeing this drubbing (AMD's 18.4% better performance at just 72.9% of Intel's price) and the entirely underwhelming Arrow Lake announcement, but rather the unshakable pangs of doubt that they just might be right and perhaps he is indeed a pretentious clown; not the genius day trader he fancied himself as.

Everyone has the right to be stupid once in a while, but you abuse that privilege, so allow me to explain it to you in a way that you, and those that would upvote you, can understand.

Let's just start with yesterday's closing prices for NVIDIA, Intel, AMD and Qualcomm:

Intel - $23.22 (I bought at just over $19, so a nice profit per share)
NVIDIA - $134.81
AMD - $164.18
Qualcomm - $168.52

Let's assume you have $10000 and your goal is to double that within 1 year. If you invest in a Treasury Bill, you will get a 4.24% return, far short of your goal.

So you decide to try the stock market.

Looking at closing prices since Jan 1 2023, here are the highest prices for each:

Intel - $50.76 - Dec 27 2023
NVIDIA - $135.58 - June 18 2024
AMD - $211.38 - March 7 2024
Qualcomm - $227.09 - June 18 2024

Here are the lowest prices:

Intel - $18.89 - Sept 6 2024
NVIDIA - $14.27 - Jan 5 2023
AMD - $62.33 - Jan 5 2023
Qualcomm - $103.02 - May 24 2023

Which of these do you think is fully valued, which do you think is undervalued and which do you think has the best change of doubling in value?

You don't need a Series 7 license or any software to see Intel has the most potential.

[sophisticles]

AMD's margins gotta be much better than Intel at this point.

If you look at AMD's quarterly reports you will see that it's much worse, AMD is not making anywhere near as much profit as Intel, they are intentionally under-pricing their products in order to claw market share from Intel.