top | item 17235810

Intel demonstrates a 28-core processor running at 5GHz

163 points| mpweiher | 7 years ago |engadget.com | reply

134 comments

order
[+] patrickg_zill|7 years ago|reply
How many will they ship, though? And at what price?

Interesting how only when AMD starts shipping something competitive with their chips, does Intel roll up its sleeves and get busy on breaking new ground. There almost seems to be a correlation ...!

[+] hvidgaard|7 years ago|reply
When AMD released a competitive GPU, nVidia released a new GPU with chips that had been sitting in a warehouse for more than 6 months. Competition is necessary.
[+] fyi1183|7 years ago|reply
Exactly. Keep in mind that AMD is currently selling 32-core chips.

They don't go up to that insane frequency, but still. There's plenty of opportunity to shop around if you want really beefy systems, which is a very refreshing state of affairs compared to just two years ago.

[+] ajross|7 years ago|reply
They're already shipping them. 28 core Skylake-SP parts have been available under the Xeon Platinum branding for almost a year now. This is presumably just a consumer paint job on the existing silicon. And yes, they're very expensive: $10k+ per chip.
[+] krylon|7 years ago|reply
In Germany, we have a saying: Konkurrenz belebt das Geschäft (roughly: Competition enlivens business).

I am just afraid Intel will once more use its deep pockets and questionable business practices to crush AMD just when things were starting to get interesting again.

EDIT: Typo

[+] leovailati|7 years ago|reply
Interesting, I wonder if it has to do with the marketing side of things rather than development. They cannot develop a CPU in less than 6 months. I think it could just be that they want to have the last word, so people will talk about it for longer.
[+] api|7 years ago|reply
Did Moore's law really stagnate as much as we think or do we just have a lazy monopoly?
[+] loser777|7 years ago|reply
If we assume this design is based off of something similar to the 28-core 8180 which has a 2.5GHz base frequency and (generous, likely understated) 205W TDP, this chip could easily have a TDP north of 300W.

Those boxes on stage sure don't look air-cooled to me. The plumbing seriously resembles that of a phase-change setup, which would actually make sense here as an extremely low operating temperature would help tame the leakage power tremendously.

In summary though, it ain't gonna be cheap to buy or cool.

[+] Ardren|7 years ago|reply
Apparently Supermicro had some X299 motherboards on display that were rated for 300 watt CPUs. So it sounds like you're on the right track.

Though I don't think phase-change is really suitable for a consumer CPU. I would think it's just going to be bundled with a big integrated water-cooler (a bit like AMD's FX-9590 CPU which was a 220TDP part)

[+] zeth___|7 years ago|reply
Well it's a phase change system you can just use it to power your home steam engine.
[+] dis-sys|7 years ago|reply
I'd willing to bet that it is very different to 8180.

it is simply not practical to boost freq from 2.5 base/3.5 turbo to all cores 5Ghz.

[+] lisk1|7 years ago|reply
The title is misleading, it reached 5Ghz through overclocking and no mention of how many cores booted at 5Ghz, the nominal frequency of this part with all cores is 2.7ghz. Saw the photos but still skeptical about it like price wise and what cooling it need.
[+] dannypgh|7 years ago|reply
If it's Intel demonstrating on Intel chips, does it really count as "over"clocking? Seems like it's just clocking.
[+] dis-sys|7 years ago|reply
there is a pic in the linked article clearly showing all cores are running 5GHz.
[+] bitL|7 years ago|reply
That's cool! 32 core Threadripper coming as well? Good old competition is back and I love it! :)
[+] dis-sys|7 years ago|reply
anandtech is reporting that the chip might be for the LGA2066 platform[1] for home use. ;)

Before you get too excited, the current generation of 28-core xeon running at 2.5G costs you over $10k each [2]. I won't be too surprised if Intel sets the price of a 5G 28-core Xeon at $15-20k.

That being said, having more choices is always good. Can't wait to see how AMD is going to respond to this using their Epyc line of chips.

[1] https://www.anandtech.com/show/12893/intels-28core-5-ghz-cpu... [2] https://ark.intel.com/products/120496/Intel-Xeon-Platinum-81...

[+] drewg123|7 years ago|reply
Is this a single 28-core CPU? The only other 28 core chips that Intel has are the Xeon platinum 8176 and 8180, which cost $9-$13K and don't clock nearly as high.

Or is this like AMD's Threadripper any EPYC, and they are duct taping together multiple CPUs in a single socket? Eg, 2 x 14 cores CPUS connected with QPI or EMIB ? 4 x 7 cores? I wonder if AMD patented this technique..

[+] maaark|7 years ago|reply
>duct taping together multiple CPUs

IIRC, according to Intel's slides, AMD used glue.

[+] innocenat|7 years ago|reply
You know that intel has been stacking CPU before, right? Pentium D, etc. It's not like AMD is the inventer of thus technique.
[+] ihsw2|7 years ago|reply
It is very much likely as you guessed, an EMIB-driven CCX design. Comparable Intel CPUs retail for north of $10K but they're nowhere near 5GHz.
[+] nrclark|7 years ago|reply
I wonder how much cache that CPU has, and how many memory channels? 28 cores don't do you much good if half of them are always blocked waiting for instructions.
[+] xtreme|7 years ago|reply
24/26 core CPUs have been in the market for quite a while. The caches generally look like this: each core has 32KB instruction + 32KB data cache (L1) and 1MB L2 cache. All cores share the 30-40MB L3 cache and the 4/6 memory channels.

In my experience, cores getting blocked for instructions is not commonplace. They can block waiting for data if the computation is memory-bound.

Edit: These are the specs for the [Skylake 8180](https://ark.intel.com/products/120496/Intel-Xeon-Platinum-81...). These chips are nowhere close to 5GHz though.

[+] farseer|7 years ago|reply
It would be infuriating if this doesn't come with ECC RAM support.
[+] qwerty456127|7 years ago|reply
It's infuriating ECC RAM doesn't come in everything.
[+] Aardwolf|7 years ago|reply
I wonder if there isn't some way to manufacture ECC RAM that doesn't need CPU support? It could act towards the CPU/motherboard as if it's non-ECC RAM but still contain 9 bits per byte inside.

The RAM chip itself would do any computation if needed instead of the CPU, and if it needs to give interrupts to warn about error conditions, it could output it as an electric signal you can hook up to anything you want yourself (including to the CPU in some way, like via USB or something lower level, or otherwise a warning light or beep or so)

[+] swarnie_|7 years ago|reply
Serious question, Who buys these chips?

There can't be many people in the world who will pay the 10k-20k price tag. How many people have the pull with their employers to justify it?

The production run must be insanely low or i just don't understand the marketplace around high end chips.

[+] IronWolve|7 years ago|reply
They really need to release a cpu for the FCLGA2066 platform that can beat the 8700K/9700K in games. People are waiting for the 9700K to upgrade, and avoiding intels high end because they dont have the a higher end cpu for gamer enthusiasts.
[+] ksec|7 years ago|reply
I am thinking could this be fitted inside the iMac Pro, or the normal iMac. Or is it for the new Mac Pro.

How did we manage to break 28 Core all running at 5Ghz barrier when we were previously hold up even 8 Core running at 4.4Ghz at the same time.

[+] majidazimi|7 years ago|reply
Atom/Electron Devs: Challenge Accepted...
[+] ramijames|7 years ago|reply
Will these be able to be used in multi-cpu systems? Will you be able to build a 56 core computer?
[+] outside1234|7 years ago|reply
Can we get a processor that can handle 32GB of LPDDR first?
[+] olliej|7 years ago|reply
Wow a floating video over the content with no dismiss button? Wut? :-/

Seriously why have the video float? If I’ve scrolled down it’s probably to read - like wtf is the intended user interaction goal there?

[+] tjoff|7 years ago|reply
How is that extremely intrusive tracking-management popup even attempting to adhere to GDPR?

For one it is opt out, which isn't allowed. It also claims that you need to accept to continue using the site, which isn't allowed, then it's countless steps required to opt out (if you even can, I got lost and accidentally allowed everything a few clicks in).

This is probably worse than just pretending GDPR didn't exist. This is a deliberate and disrespectful fuck you to every single visitor.

[+] victorNicollet|7 years ago|reply
For me it was opt-in (all trackers were unchecked when I opened it) from both of my computers.

It is also not illegal to require visitors to clarify to which data processing they consent to (GDPR only says that if you require consent for processing that is not necessary to provide the service, then that does not count as consent for GDPR purposes).

[+] setquk|7 years ago|reply
It stopped me visiting the site.

Which is good becase that saved me a chunk of my life on Engadget who clearly have no respect for the readers.

[+] RealDinosaur|7 years ago|reply
Engadget... What the hell is with your GDPR consent.

1) Manage Options

2) (Body font sized link) Manage

3) (Body font sized link) Show

4) Scroll down

5-100) Click each toggle individually (select all button is currently broke) (It is intentionally ambiguous which is the 'correct' selection)

101-150) Do the same on other page

151) Click done

152) Click OK (hoping it's saved, it doesn't feel like it has)

This is much harder than it is to opt-in. I'm pretty sure this against the spirit of the GDPR.

Even if it isn't against the GDPR. It's a dick move.

[+] kzrdude|7 years ago|reply
GDPR rules in practice seem to turn into the "this site uses cookies" notice but on mega steroids. Privacy control is great, but it's just not interesting to have a checkbox or agreement click through on every new site visited.
[+] oelmekki|7 years ago|reply
Same here. I went up to the partner list, didn't get what the toggle buttons were supposed to mean (should I toggle them on or off?), closed the page. The title (on HN) will have to do, good job to them.
[+] dvfjsdhgfv|7 years ago|reply
For most sites just disabling JS works perfectly, but not for Engadet. I just had a look at their source and found out the actual post ("18-core i9 CPU is still impressive, but that's last year's news") starts at the line 1086! Over one thousand of lines of useless crap before I can read the actual text. Incredible.
[+] jug|7 years ago|reply
> I'm pretty sure this against the spirit of the GDPR.

I'm pretty sure almost every reaction and e-mail that I have got due to GDPR is against the spirit. I had never thought it would be allowed, much less massively adopted, to "opt in" 100% of your existing userbase by fleshing out your existing Terms of Service and that's that. I thought it would require a new contract/agreement with the user to be signed/clicked "yes" to. I swear by far most have just somehow pulled me into their new "more privacy aware" system.

[+] RealDinosaur|7 years ago|reply
And text cause you know...

Intel's 18-core i9 CPU is still impressive, but that's last year's news. Today at Computex, Intel SVP Gregory Bryant demoed a 28-core processor running at 5GHz. It's the first time we've seen a single socket desktop CPU cram in that many cores, and it's certainly powerful. In Cinebench, it achieved a score of 7,334 on stage, something that's only bested by multi-processor systems. Better yet, Bryant says it's not just a concept, as Intel plans to sell the 28-core CPU by the end of the year. Time to start saving.

[+] megaman22|7 years ago|reply
I mean, we all knew this was going to happen, right? There's no way this was actually going to turn out to be a helpful thing, at least on this class of sites.
[+] Asooka|7 years ago|reply
You can bypass all that by disabling javascript for everything except engadget.com. But the article is entirely contained in the headline.
[+] JBiserkov|7 years ago|reply
From the photo:

# of cores: 28

# of threads: 56

Hyper Threading FTW!

[+] stephengillie|7 years ago|reply
Doesn't the success of hyperthreading depend on the ineffectiveness of branch prediction?