Slashdot Mirror
Leaked Benchmarks Suggest Intel Will Drop Hyperthreading From Core i7 Chips (arstechnica.com)
According to leaked benchmarks found in the SiSoft Sandra database, there is an Intel Core i7-9700K processor that doesn't appear to have hyperthreading available. "This increases the core count from the current six cores in the 8th generation Coffee Lake parts to eight cores, but, even though it's an i7 chip, it doesn't appear to have hyperthreading available," reports Ars Technica. "It's base clock speed is 3.6GHz, peak turbo is 4.9GHz, and it has 12MB cache. The price is expected to be around the same $350 level as the current top-end i7s." From the report: For the chip that will sit above the i7-9700K in the product lineup, Intel is extending the use of its i9 branding, initially reserved for the X-series High-End Desktop Platform. The i9-9900K will be an eight-core, 16-thread processor. This bumps the cache up to 16MB and the peak turbo up to 5GHz -- and the price up to an expected $450. Below the i7s will be i5s with six cores and six threads and below them, i3s with four cores and four threads. Even without hyperthreading, the new i7s should be faster than old i7s. A part with eight cores is going to be faster than the four-core/eight-thread chips of a couple of generations ago and should in general also be faster than the six-core/12-thread 8th generation chips. Peak clock speeds are pushed slightly higher than they were for the 8th generation chips, too.
I've always seen them as "pseudo" cpu's, and not been all that happy with them overall. Yeah, some workloads benefit from it just fine, but others get tanked, but you'll never know because it just looks like those CPU's are flying along (according to task mangler or whatever).
Anyway, glad to see that there will be some parts out there that people can choose to buy that don't have it.
in a multi-core chip past 2-4?
I guess not.
Marketing or improved latency enough no more opportunity for HT on low core counts.
spectre-related attacks rely on predictive execution in hyperthreading.
could this be a mitigation, while providing improved performance so the new part still exceeds the outgoing part?
Will these still need to have Meltdown and Spectre patches? If Intel are just pooping out new chips with no fix for the root cause then it's kind of a moot point to talk about it's speed.
Anons need not reply. Questions end with a question mark.
Can we get more pci-e lanes on the desktop?
Not pay more or get less.
AMD FOR THE WIN!!!!!!!!!!
I know we have to disable CPU level Hyper Threading anyway. Too many false cache swaps when it's on.
How else are you going to convince suckers to buy the i9 chips. Either way, Ryzen is certainly superior.
I will always take physical cores over hyperthreading any day. Back when hyperthreading was a thing, many tested out exactly how much it benefitted. Most of the time it really wasn't anything to brag about. Considering how much the new i7 will have in more effective performance benefits. Hyperthreading does appear ready for the heap of out dated technology.
If this is how they plan to deal with the Spectre/Meltdown issues permanently, I'm ok with that.
It will degrade power performance I believe, since HT allowed a CPU's execution engine to stay busy flipping back and forth between the two instruction pipelines. Now you're just going to have full blown cores doing the same work. There's gunna be more dead time on these non-HT CPU's since when a CPU normally would be waiting in previous setups, it would just execute it's other pipeline, now it'll just wait.
And the end of the day, if I can still play my games, I really don't care what's inside that hunk of silicon, as long as my games still work as good as they always have.
So to recap the 8th gen of Intel. The i7 had the most cores at six with HT enabled. i5 was just like the i7 but with HT turned off. The i3 had HT in gen 7 so it was two cores/four threads, in 8th gen they gave it two more cores and turned off HT. So: i7=6/12, i5=6/6, i3=4/4. The i9 in gen 8 was really weird. The clock would scale down the more cores you used, it was very odd and minus the fact that the 18 core version was roughly the price of a used car, it was expensive. The price per performance with the i9 was incredibly low. A 3.4 Ghz i7 would give you a better CPU mark / $ by almost 200%, not to mention that an AMD six core FX-6300 would give you better CPU mark/$ by almost 800%. So clearly the i9 wasn't going to win you an award for price sensitive consumers.
So all that said, and this is my opinion so it's literally worth whatever value you choose to give it, I think Intel is going to reposition the line up to disable HT on all "consumer" processors and focus on just keeping HT and "pro" features in the i9. I personally think it's a back hand to Intel consumers, but I'm an AMD fanboy so full disclosure there. But yeah, I think the i3, i5, and i7 are all going to eventually be labelled as the "cheapy", "actual desktop", "gamer" CPUs in that order and the i9 is going to be viewed as "workstation" and thus the i9 isn't going to focus on price/performance balance. So, i3 will be 4/4, i5 will be 6/6, and i7 will be 8/8 with the i9 being whatever crazy numbers they throw at the chips with hopefully not any of that weird scaling core/HT/Ghz stuff.
That's just my hot take on this, open to hear what others think.
Without it you could get better "standard" chip speeds, and it made more sense for the expected workload.
I'm no CPU expert, but it's my understanding that to increase yields, Intel already disables features or entire cores that don't work right on a particular chip. If HT isn't all that useful on the new chip, and it reduces Spectre-type risks, would it not be rather easy to disable it? It can already be disabled in the BIOS, right, so disabling it doesn't require any changes to the silicon.
They have been unwilling to properly implement SMT and only went for the fake version they sold as hyperthreading to get status points with the press but not actually deliver the performance possible. Apparently they feel like the press points are of diminishing value and they no longer need to maintain the charade where it's too hard on their profit margins.
as you could have done since it was introduced decades ago.
AMD should simply go to i11.
Surely you mean "Its base clock speed is 3.6GHz" and not "It is base clock speed is 3.6GHz".
#DeleteFacebook
Intel isn't going to drop Hypthreading on all its parts, or AMD will happily kick their tail with its superior SMT implementation.
Doubtful. Increasing cores experiences diminishing returns. The difference between 8 cores and 8 cores + HT may be quite minimal except in very rare instances.
With GPUs relatively ubiquitous some (many ?) of these workloads may just shift to the GPU.
only some. the highest-end part in that line (i9-9900K) is still shipping with ht.
the decision is likely the result of not getting enough 'perfect' dies off their fabrication process and absolutely nothing to do with the known flaws in the architecture (which may or may not be partially mitigated by disabling ht)
It was a quote from the original ArsTechnica article so surely it was meant to be "It's [sic] base clock speed is 3.6GHz...". It looks like it was fixed in the original article though (or it was correct to begin with and copied incorrectly, in which case double shame on BeauHD)
You can buy a $99 dollar processor with decent single thread performance and not have to worry about games that require 4 threads.
Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
> "It's base clock speed is 3.6GHz..."
Yeah. Now call me grammar nazi. Then I'll say "Hitler". Then this thread be Goodwin'd.
Having SMT enables gives lower performance in almost every workload, only specifically parallel tasks (computational tools) or Cinebench benefits.
It is the best way to stop the page table bugs.
You may easily turn off HT in BIOS and Sandra will be none-the-wiser. This may be a test of an engineering prototype or R&D sample that is not aimed for mass production. Or it could be faked by most anybody - it is quite easy to submit fake SSS reports.
In a nutshell, we have no idea what the conditions of the "leaked" sandra report were, so this article is pure speculation.
I will not buy a processor without hyperthreading.
intel is dead long live AMD
This is their fix for melt down and Spectre. Cripple the CPU. Hyperthreading is just speculative execution. Instead of doing the right thing and creating rings they cut the branch prediction and speculative execution out of the CPU Core. They're hoping the increased CPU clock will offset the performance loss of doing this. I hate to say it, but Intel is fucked. I hope AMD is able to continue development on this technology. It is going to take a complete redesign, something Intel doesn't want to do.
From Intel, it has been confusing for a long time. Is a CPU 2 core/2 thread, 2 core/4 thread, 4 core/4 thread, 4 core/8 thread, 6 core/6 thread, 6 core/12 thread, or now, 8 core/8 thread, or 8 core/16 thread? The name alone does not really tell you much, so doing a lookup online is needed. To make it worse, the U series of chips has tended to be dual-core, even if it is branded as an i7.
In general, we have seen the i3 line cover the 2 core and up to the 4 core/4 thread mark(as of the 8th generation i3, the i3 can now be a quad-core chip). i5 has generally been either 2-core/4 thread, or 4-core/4-thread, though with the 8th generation, we are seeing 6 core/6 thread. i7 has been your 4-core/8-thread chip, or 6-core/12-thread. Once 8 core chips came into the picture, suddenly Intel added the i9, and things got really messy, because now, the 6-core/12 thread falls into the i9 range, 8 core/8-thread falls into the i7 range, and 8-core/16-thread will also be i9. This idea of "some chips get HyperThreading and some don't" is what makes it a mess. If it were a simple case of i3 is 4 core/4 thread, no Hyperthreading at all, then saying 4 and 6 core without HT being i5, i7 being 4 and 6 core with HT, and then the 8 core being i9, that would make more sense, make the version with HT have a H at the end of the name, the version without not having the H...it would make it easy to keep track of.
Intel may simply be playing the game of, "sell a crap chip in machines with an i7 brand" to improve sales of crap chips, but it will upset users when their i7 branded laptop with only two cores runs worse than an i5 with 4 cores.
I remember when Intel released its "Core 2" lineup some gamers eschewed getting the quad core ones because the dual core CPUs clocked higher, were less expensive, and most games only used one or two threads. Given how popular the slighter older i7s (4 cores, eight threads) and i5s (4 cores) have been, and how prominent they are in populating gaming PCs, it seems doubtful to me that any PC game will be incapable of running blisteringly fast on 8 physical cores. Games running on an i5 6600k/i7 7600k, with just four cores, generally go fast enough even for gamers running monitors at high refresh rates. https://www.pcgamesn.com/intel... https://www.pcgamer.com/intel-... These new CPUs will more than double their performance, and given the enormous number of customers for games with older i5s and i7s it's hard to imagine any game developer with an eye to sales who'll think that "Too bad, it's 16 threads or bust!". Non-gaming applications have been impressive in their improving on the ability to make use of more threads, but even there there are often diminishing returns, and it will be the pro's, and serious semi-pro's, who notice the savings in time by going to 16 threads (or more). But my point here goes to gamers/power users of the PC. For many of them, myself included, eight physical cores represents the opportunity for a solid upgrade. Though personally, I'd be a lot more excited if this was on 10 nanometer, it ran cooler when getting stressed, and it used less juice under all circumstances. At 10 nm it would be gold. As is, I think a lot of people will wait and see for what's on the horizon.
If you're bound by memory I/O or limited by cache then something like hyperthreading won't improve throughput and will probably make it worse. But even in that case you can have better fairness between threads and less bursty behavior at high load when doing hyperthreading.
Even in an optimal case total throughput doesn't improve very much with hyperthreading. So results are disappointing if your only performance metric is how many operations you can perform a second or how much memory bandwidth you can blast through. I don't see much value in hyperthreading for a desktop user that has relatively few ready tasks at any given instant (often in single digits and low double digits), especially now that quad core is pretty commonplace. When your ready tasks is many times your number of cores, you will need to context switch often to distribute the work fairly.
On a server where you do have lots of tasks ready to run (sometimes 100+), having 2-way or even 4-way SMT is way cheaper than adding more cores (less silicon area, less power, reduced bus infrastructure, etc). Even if an individual thread runs more slowly because of hyperthreading sharing resources of a CPU, not having to context switch as often is a net gain.
tl;dr - hyperthreading is not a panacea.
“Common sense is not so common.” — Voltaire
AMD was shit for games for a good while, but Intel got sloppy.
I went from a 2500k to a 7700k. Not happy with it; too hot, and Intel said to not OC it (what's the point of buying the K series then?).
Plus it looks like the z270 chipset (7700k) is essentially the same as the z370 chipset (8700k), but z270 doesnt support 8X00 series CPUs. (Of course this is standard for Intel, but it's still lame.)
Intel... you done goofed too many times. Your CPUs need to absolutely destroy AMD's in gaming to win my next purchase.
While I totally get that a hyperthreading core isn't as good as a real core, who cares? The system gets 20% faster with those HT cores and I'm happy to take the 20% performance advantage.
Also, if Intel takes HT out of the i7 line... I don't think I care. As long as they add in real cores to replace them it's all good. They don't even have to replace HT cores with real cores on a 1:1 basis, just add enough real cores to make up for the missing HT cores, plus some extra to continue growing the value add for tech progress and to keep some notional parity with AMD offerings.
I don't think it helps much for tech analysts to obsess too much about where their CPU performance comes from. As long as each generation is better than the last generation, what does it matter whether it is due to more/better cache designs, more cores, faster DRAM subsystems, HT, better CPU interlinks, more efficient IPC, specialized CPU subsystems, new CPU instructions, etc.?
With Dennard Scaling slowing down, we have much bigger fish to fry than whether i7 = HT.