SoylentNews is people
SoylentNews
from the what-about-meltdown-and-spectre? dept.
Intel Announces 9th Gen Core CPUs: Core i9-9900K (8-Core), i7-9700K, & i5-9600K
Among many of Intel's announcements today, a key one for a lot of users will be the launch of Intel's 9th Generation Core desktop processors, offering up to 8-cores on Intel's mainstream consumer platform. These processors are drop-in compatible with current Coffee Lake and Z370 platforms, but are accompanied by a new Z390 chipset and associated motherboards as well. The highlights from this launch is the 8-core Core i9 parts, which include a 5.0 GHz turbo Core i9-9900K, rated at a 95W TDP.
[...] Leading from the top of the stack is the Core i9-9900K, Intel's new flagship mainstream processor. This part is eight full cores with hyperthreading, with a base frequency of 3.6 GHz at 95W TDP, and a turbo up to 5.0 GHz on two cores. Memory support is up to dual channel DDR4-2666. The Core i9-9900K builds upon the Core i7-8086K from the 8th Generation product line by adding two more cores, and increasing that 5.0 GHz turbo from one core to two cores. The all-core turbo is 4.7 GHz, so it will be interesting to see what the power consumption is when the processor is fully loaded. The Core i9 family will have the full 2MB of L3 cache per core.
[...] Also featuring 8-cores is the Core i7-9700K, but without the hyperthreading. This part will have a base frequency of 3.6 GHz as well for a given 95W TDP, but can turbo up to 4.9 GHz only on a single core. The i7-9700K is meant to be the direct upgrade over the Core i7-8700K, and although both chips have the same underlying Coffee Lake microarchitecture, the 9700K has two more cores and slightly better turbo performance, but less L3 cache per core at only 1.5MB per.
Intel also announced refreshed 8 to 18 core high-end desktop CPUs, and a new 28-core Xeon aimed at extreme workstation users.
Related:
Intel Teases 28 Core Chip, AMD Announces Threadripper 2 With Up to 32 Cores
AMD Threadripper 2 Available Starting on August 13
(Score: 3, Insightful) by coolgopher on Tuesday October 09 2018, @07:07AM (2 children)
Wasn't it just about conclusively proven that hyper-threading had design-level flaws that resulted in security issues?
Oh wait, I forget, that's all "operating as designed" and "Intel believes its products are the most secure in the world". I'll pass on Intel processors until they smarten up again, thanks.
(Score: 0) by Anonymous Coward on Tuesday October 09 2018, @09:42AM
Which is obviously not compatible with security of said user against third parties.
(Score: 2) by DannyB on Tuesday October 09 2018, @06:13PM
I don't want hyper threading. I want more REAL cores.
To transfer files: right-click on file, pick Copy. Unplug mouse, plug mouse into other computer. Right-click, paste.
(Score: 2) by Arik on Tuesday October 09 2018, @07:13AM (5 children)
I'd actually pay money for that. But I don't think I'd trust Intel to manufacture it.
If laughter is the best medicine, who are the best doctors?
(Score: 2, Insightful) by Anonymous Coward on Tuesday October 09 2018, @11:55AM
Are you calling the ME behind every new Intel CPU defective? For shame. Don't you trust your government?
(Score: 4, Insightful) by ledow on Tuesday October 09 2018, @12:29PM (3 children)
Try OpenRISC or even some of the ARM microcontrollers that are now FGPA-able for free.
Oops... you mean you want a top-performance, state-of-the-art, incredibly-expensive-to-manufacture, fully documented, bug-free, high-performance processor?
Cheap, fast, secure, pick any two.
(Score: 4, Insightful) by takyon on Tuesday October 09 2018, @04:12PM
I don't think you get to pick two with either Intel or RISC.
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 2) by Freeman on Tuesday October 09 2018, @04:48PM
Literally, the only thing going for Intel in that list is Fast, because they certainly aren't cheap and their security was worse by design than AMD. I've been building my personal computers with AMD parts for a decade. All four of the machines I built are still functional, only thing that's gone out on a machine or two is the Graphics Card. In every instance, I was able to build my machine cheaper, and with better than I could have bought with Intel or pre-built of either. I also, got a bit lucky on my last build. I built it between the bitcoin/etherium craze, so my RAM and GPU weren't stupidly over priced. That was about two years ago. Also, Newegg is the goods. They may have some issues, but at least they try to deliver you quality product and give you the information you need to make a decision. Amazon has no equivalent filters that allow you to sort by Best Selling items or Most Reviewed items. Also, Amazon is stupid huge, and I try to avoid them when possible. Competition is good.
Joshua 1:9 "Be strong and of a good courage; be not afraid, neither be thou dismayed: for the Lord thy God is with thee"
(Score: 1) by Arik on Thursday October 11 2018, @05:52AM
If only that were the case. It's actually pick one.
But fast is easy to sacrifice, for me at least. I learned microcomputers in the era when a single core was a given, and clock speeds were measured by the hundredth of a MHz instead of hundredth of a GHz. So in theory I should be perfectly fine with cheap and secure, in fact overjoyed.
Yeah, doesn't seem to be available. Seems like it's choose one. Granted values for 'fast' might vary.
Ideally I'm looking for something fast enough to deal with shitty 'modern' code at an acceptable speed, see. I mean, I'm not at all afraid of finding a variable and fixing it and recompiling. But rewriting the entire shitty modern OS from scratch so it will run on, say, a 486 (which by all rights is perfectly capable of handling everything most people do with their PCs today, given the proper software) is the sort of thing I could easily make my life's work and still die without finishing.
Especially because it's not like anyone is going to pay me to create sanity. Sanity is a dud, it doesn't monetize.
If laughter is the best medicine, who are the best doctors?
(Score: 2) by bradley13 on Tuesday October 09 2018, @07:46AM (13 children)
There cannot be such a thing as a "mainstream" 8-core CPU. Look at your average user - they have some application open that might make serious use of 1, or at most 2 cores. Toss in another one to handle background work by the OS, and you're done. Even then, those 2-3 cores will spend most of their time idle.
As a technical user - yesterday I had half-a-dozen background tasks running, in addition to working normally - I was maybe making half-decent use of 4 cores. Realistically, though, my background tasks were limited by disk and/or network throughput, so the CPU was definitely not the limiting factor.
tl;dr: WTF are all those cores supposed to do outside of a server? In the new Intel line-up, anything over the lowest two options will be completely wasted in any end-user device.
Everyone is somebody else's weirdo.
(Score: 4, Insightful) by takyon on Tuesday October 09 2018, @07:48AM
Damn, knew i should have kept the scare quotes.
Actually, you know what? I will write "mainstream 16-core" without scare quotes next year when AMD debuts it.
And here's your mainstream use case... uh... streaming your video game in 4K while using
DiscordFacebook Messenger![SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 4, Funny) by Arik on Tuesday October 09 2018, @07:51AM
The extra 6.5 CPUs are for the nice chaps at MI5.
You don't want the commies to win, do you?
If laughter is the best medicine, who are the best doctors?
(Score: 4, Insightful) by choose another one on Tuesday October 09 2018, @09:06AM (3 children)
> Look at your average user - they have some application open that might make serious use of 1, or at most 2 cores.
And one of those applications will be a web browser, with multiple tabs open, almost guaranteed. And because today's ad-polluted web takes so fricken long to load a page (I swear pages loaded quicker when I had a 14.4k modem), many of them will still be loading pages while the user looks at another tab and then switches back when the page has, probably, loaded.
Source - watching the kids, the wife, and jsut looking in front of me. Biggest CPU hog - chrome, biggest memory hog - chrome, tabs open - double figures, most of the damned time. God forbid chrome crashes, because reloading everything makes the machine unusable for several cups of coffee.
(Score: 0) by Anonymous Coward on Tuesday October 09 2018, @12:47PM
Thanks to the whole webapps mentality, web browsers have become virtual machines.
Never mind that CPUs are rarely utilized to capacity unless the dataset can fit inside cache.
This because reading anything from ram, never mind HDD/SSD, will have the CPU sit idle for "ages".
(Score: 0) by Anonymous Coward on Wednesday October 10 2018, @12:01AM
Sounds like you really need an SSD. Trust me, once you upgrade from an HDD, you can't go back. NAND flash is even almost reasonably priced by now.
(Score: 0) by Anonymous Coward on Sunday October 14 2018, @06:01AM
Step 1. Install Firefox
Step 2. Install umatrix addon
Step 3. Enjoy faster web experience
(Score: 3, Interesting) by richtopia on Tuesday October 09 2018, @01:14PM (3 children)
Then why are mobile processors cranking up the number of cores? It feels like all phones are octacore now, and while some of it may be bloat I can't imagine the budget phones shipping with any more than required.
(Score: 4, Interesting) by bob_super on Tuesday October 09 2018, @04:19PM (2 children)
There is a battery benefit to staying in deep sleep as long as possible, then firing up the CPU to do everything that's waiting as fast as you can before going back to sleep.
Which is totally irrelevant to the fact that "my phone has more core than yours, and higher benchmarks in games you don't play because you only check FB/twit/inst/porn, so it's better" is the actual driving force for more cores.
(Score: 3, Informative) by takyon on Tuesday October 09 2018, @09:01PM (1 child)
Most phones have different cores for different purposes. For example, 4 performance cores and 4 lower power cores. Or 2 performance cores, 2-4 regular cores, and 4 lower power cores.
DynamIQ is the successor to big.LITTLE [wikipedia.org] and can allow weirder configurations [anandtech.com]:
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 3, Informative) by bob_super on Tuesday October 09 2018, @09:17PM
A good read:
https://arstechnica.com/gadgets/2018/08/p-is-for-power-how-google-tests-tracks-and-improves-android-battery-life/ [arstechnica.com]
(Score: 2) by toddestan on Wednesday October 10 2018, @02:49AM
By "mainstream" that means it fits into Intel's standard desktop socket, and not their stupidly expensive "charge more just because we can" workstation socket. On the other hand, that also means you lose features such as ECC. Though if you want ECC, I'm sure there's a Xeon somewhere in the lineup that's almost exactly the same but costs about three times as much.
(Score: 2) by exaeta on Wednesday October 10 2018, @05:04AM (1 child)
Yeah no. Stay in the dark ages if you like, but programmers *should* write multithreaded code. The reason we haven't is laziness and lack of multicore systems. As core count rises, the benefits of multithreading will increase, which will result in multithreading becoming more common. The way forward is multithreading. Since we are hitting the shrinking limit, I also announce a new law: The number of cores on a "mainstream CPU" will double approximately every 2 years.
In
The Government is a Bird
(Score: 0) by Anonymous Coward on Sunday October 14 2018, @06:04AM
but powershell multithread support is quite basic at the moment so why bother just hammer it in one
(Score: 1, Insightful) by Anonymous Coward on Tuesday October 09 2018, @08:47AM (1 child)
Just like the previous generation of disappointing CPUs, these will not be available for months, or will carry hugely inflated prices. Intel cannot even keep up with the current demand for 8000 series CPUs.
(Score: 2) by takyon on Tuesday October 09 2018, @04:14PM
I think you actually got the number of +s right, or underestimated it. Sad!
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 4, Interesting) by Anonymous Coward on Tuesday October 09 2018, @11:12AM (2 children)
Lets face it: They don't have a solution so they're ignoring the problem.
(Score: 5, Interesting) by ledow on Tuesday October 09 2018, @12:39PM (1 child)
Not really surprising, is it?
All the Intel performance came from speculative execution - you were effectively following all paths after a decision and executing all the things you would do for both paths, but then just throwing away the "not used" result.
That's been shown to be inherently flawed unless you apply the same security to your speculative executions as if you were really executing them anyway.
That's gonna be a performance hit on anything - hence all the fixes are to basically "break" speculative execution, or remove the possibility, when the chips just go back to ordinary chips with no speculative execution. This is why hyper-threading is no longer recommended or even inside most new chips.
You can make a chip, no doubt, that does the speculative execution properly. But that will require more die-space. Which means a complete redesign AND it's probably just easier at that point to make a chip without SE, but with entire extra cores, for the same price.
People were warning about SE for years because they knew something like this would happen. And it's only really a bodge unless you're literally executing all paths "for real" and the "rollback" / contention etc. of the wrong path doesn't affect the speed of the right path at any point.
Pretty much, the more "fancy tricks" you use to make your stuff go faster, the harder it bites further down the road. And without the fancy tricks and pretending, Intel chips are precisely THIS fast, with THIS many cores, and no fancy compiler/processor tricks take place.
The solution can only be a complete redesign of all chips (which they won't have got close to doing), which will result in a larger/hotter chip to do the same job. Or they remove all the fancy junk, at the same cost, and give you slightly more cores at the same speed.
It'll be 5-10 years before Intel chips progress again from a new design. At least. And even other manufacturer's aren't immune, and all because they played fast-and-loose with security when speculatively executing instructions.
(Score: 2) by RamiK on Wednesday October 10 2018, @07:01AM
Designing a chip without speculative execution that performs even half as good for general purpose loads is a hundred billion dollars problem that baffled Intel and HP with the Itanium and remains unsolved. Disabling hyperthreading in favor of more cores only address a very specific speculative execution vulnerability and only to some extent. The other fixes require recompilation and have a significant performance hit.
Staying OoO, Intel/AMD/ARM are more likely to redesign their microarchitectures following SafeSpec [arxiv.org]. But it's still needs researching and is years away.
Regardless, right now, they really don't have a solution that doesn't hit you with huge performance overheads.
compiling...
(Score: 2) by realDonaldTrump on Tuesday October 09 2018, @01:19PM (1 child)
As you know, I have a running war with the Fake News Media. They are among the most dishonest human beings on Earth. And they sort of made it sound like I had a feud with the Intel community. And I just want to let you know, it's exactly the opposite -- exactly. I absolutely love what our Intel guys are doing with cyber. The new cyber that's coming out. The Teraflop cyber. We love Teraflop. And the kids -- the boys -- love it. Great Christmas gift for the boys. Santa's going to be bringing a lot of Teraflops this year. Merry Christmas!!!!
(Score: 0) by Anonymous Coward on Sunday October 14 2018, @06:07AM
It's good to see such a distinguished person as yourself into cybering.
If more people cybered the world would be a better place.
Mainframe has had teraflops for years. You should look into it!
(Score: 2, Informative) by doke on Tuesday October 09 2018, @04:49PM (2 children)
Intel is using NDA embargoes to block third party reviews of their new chip until the 19th. However, they commissioned and published their own review. To no one's surprise, that review has been shown to be highly biased. In my opinion, these actions show Intel's ethics can't be trusted.
https://www.pcgamesn.com/intel-core-i9-9900k-vs-amd-ryzen-7-2700x-gaming-benchmarks [pcgamesn.com]
https://www.techspot.com/article/1722-misleading-core-i9-9900k-benchmarks/ [techspot.com]
https://www.patreon.com/posts/21950120 [patreon.com]
(Score: 4, Insightful) by Azuma Hazuki on Tuesday October 09 2018, @04:57PM (1 child)
Intel hasn't been trutworthy since 15+ years ago when they were...making offers OEMs couldn't refuse...not to use AMD products.
I am "that girl" your mother warned you about...
(Score: 0) by Anonymous Coward on Friday October 12 2018, @10:48AM
The reason we have so few motherboard chipsets today is directly attributable to their handling of the Pentium launch in the early 90s and then finished with the PPro launch. Super Socket 7 was the community retaliation against that, but it only lasted until AMD bought the DEC assets and migrated to Alpha's EV6 bus and compatible chipsets, leaving those chipset manufacturers without technology they could license, design, and produce. As a result many chipset companies died out, the few that didn't find themselves with many technical glitches since they didn't have access to the same level of detail Intel/AMD had internally, and slowly the pool of companies producing them dwindled until only AMD, Intel, and Via remained, with some small showings by ALi/Nvidia(ULi?) before both gave up on their own chipset designs from the AM2/LGA775 era.
Now chipsets and processors are innately tied together and replacement boards/cpus are single source, rather than multiple source like in the heyday of the 386 to k6-3 (Intel abandoning Socket 7 for Socket 8, Slot 1, and then Socket 370, despite all three using the same bus, which was incompatible with the Socket 7 bus, but intercompatible with adapters, like the PPro Overdrive chips, and Socket 370 to Slot 1 adapters prove.)