Live data from Hacker News

Intel Announces 8 Core I9-9900KS: Every Core at 5Ghz, All the Time

anandtech.com

231–240 of 277 posts

Re: Intel Announces 8 Core I9-9900KS: Every Core at 5Ghz, All the Time

#231

Earlier quoted context omitted.

"Apple’s iPhone Xs is faster than an iMac Pro on the Speedometer 2.0 JavaScript benchmark" Yes, Safari, but iPhones do compete with your average desktop processor (not the top end). https://macdailynews.com/2018/09/23/apples-iphone-xs-is-fast...

Apple's newest CPUs have hardware explicitly for accelerating Javascript, so it's not surprising they'd pull ahead there. And as you said, you're comparing the top-of-the-line iPhone to an "average" CPU.

> Apple's newest CPUs have hardware explicitly for accelerating Javascript

They really don't. A12 added a couple of instructions for floating point conversions, but contrary to claims making rounds on Twitter at the time, they were not even generated by WebKit when the benchmarks were run.

Re: Intel Announces 8 Core I9-9900KS: Every Core at 5Ghz, All the Time

#232

Should they be describing it as 8 cores 16 threads when there have been multiple security vulnerabilities that have to turn off hyperthreading to be mitigated?

This is a very good point. I hope AMD brings it up with the EU. Might be a very slow process though, but at this point it is anticompetitive behaviour. AMD could probably squeeze a fair bit more performance out of their processors if they were allowed to cut some security corners.

I hear many people continuing to say that Intel are "cutting security corners".

Are they really? I don't have an extremely deep understanding of Intel's implementation of x86 ISA, but I do know enough to say that so far we've been able to effectively mitigate almost all of these attacks with existing instructions available on the Intel CPUs. That doesn't mean that they are still not open to other variants of these attacks - but at some point you have to assume diminishing returns. Spectre is still very difficult to exploit, for example.

Perhaps this has little to do with Intel and more to do with software authors cutting corners? LFENCE and SFENCE are reasonably well documented, after all...

Re: Intel Announces 8 Core I9-9900KS: Every Core at 5Ghz, All the Time

#233

Earlier quoted context omitted.

As soon as the definition of "with all vulnerability mitigations on" stays stable long enough to put together a good review. Benchmarking a moving target is hell, and we don't have enough equipment or staff to do the around-the-clock regression testing that would be necessary to keep our benchmark database current with everything that's happened over the past 1.5 years.

People are using your benchmarks to decide what computer to buy. If they're that out of date, what should I tell them?

To use the best information available, which are estimates on what the potential impact can be, and then make their own decisions.

Re: Intel Announces 8 Core I9-9900KS: Every Core at 5Ghz, All the Time

#234
post #61

Earlier quoted context omitted.

I more and more tend to believe the rumors (started by Semiaccurate) we will not see 10nm in mass quantities and 7nm is the next. We will see very, very soon: Intel said 10nm CPUs in client systems will be on shelves for the 2019 holiday season.

But weren't there Intel 10nm chips sold for revenue in 2017? [0] Granted, it was an OEM-only part for some China specific education laptops... 0: https://www.anandtech.com/show/13405/intel-10nm-cannon-lake-...

Cannon Lake was released only because many at Intel have their bonuses tied to the process node launch. Well, they launched a 10nm CPU... so bad the GPU is disabled, performance/watt it's worse than KBR and it was only available in limited quantities.

Re: Intel Announces 8 Core I9-9900KS: Every Core at 5Ghz, All the Time

#235
post #39

I used to be Pro Anandtech and consider them one of the best Sources online for Hardware News. But the fact they have yet write a single post, big or small about Intel's Zombieload and its implication on performance worries me a bit. Then there is the "Intel" benchmarks as usual [1] on GPU. Trying to suggest the 2 CPU were both running at 25W TDP to give a "fair" comparison, without mentioning the Ice-Lake U CPU were…

Because intel strictly forbids publishing benchmarks of their processors with the "hardware vulnerability mitigations" applied. Even OEMs cannot show them to their enterprise customers. You can do your own benchmarking after buying the systems. So, no money, no real-world benchmarking.

They did for a brief period. It got reverted pretty quickly.

https://www.tomshardware.com/news/intel-cpu-microcode-benchm...

Re: Intel Announces 8 Core I9-9900KS: Every Core at 5Ghz, All the Time

#236

Earlier quoted context omitted.

explain why i should care instead of downvoting. a windows program can already just directly look at the memory of other running processes, so why do i care about sidechannel attacks outside of javascript snooping on things which i can mitigate with simple behavior changes.

>>a windows program can already just directly look at the memory of other running processes Do you run everything with administrative privileges?

Too many programs ask to run as admin already, people just click okay on the UAC prompt to make it go away now.

Re: Intel Announces 8 Core I9-9900KS: Every Core at 5Ghz, All the Time

#237

Earlier quoted context omitted.

This is a very good point. I hope AMD brings it up with the EU. Might be a very slow process though, but at this point it is anticompetitive behaviour. AMD could probably squeeze a fair bit more performance out of their processors if they were allowed to cut some security corners.

I hear many people continuing to say that Intel are "cutting security corners". Are they really? I don't have an extremely deep understanding of Intel's implementation of x86 ISA, but I do know enough to say that so far we've been able to effectively mitigate almost all of these attacks with existing instructions available on the Intel CPUs. That doesn't mean that they are still not open to other variants of these at…

Here's a register article from 2007 about page table permissions being problematic. If you look around a bit, there were a ton of security researchers who talked about the problem. It seems to have been a bit of an open secret that such a thing must exist -- they just hadn't found it yet.

https://www.theregister.co.uk/2007/06/28/core_2_duo_errata/

The scariest part is that many of the best security minds work for various intelligence agencies. They very likely have known about such things for a very long time.

Meltdown strikes me as an almost perfect vulnerability. It affects almost everyone. It is undetectable until exploited and once exploited, it immediately goes away until the next time. It's easy to keep secret. Most importantly, it's a one-way vulnerability. Keep your secure systems from running untrusted code and there's zero risk. Since this is standard protocol anyway for those systems, you don't have the risk of someone running across a code patch somewhere.

The only potential downside is that the juiciest targets also aren't running untrusted code (though most foreign affairs workers probably run untrusted code). The big point of interest here is information symmetry. In most cases, giving others secret information is bad. In this case, both the best and worse case situations work out well for the USA. If nobody else knows, they get free info. If everyone else does know, then everyone gets perfect information about everything. This favors the most powerful country. They can eliminate the unknowns (the only real danger). In contrast, knowing you are going to be crushed does nothing if you can't hide your own hand either. So, the best case is very good and the worst case is still acceptable.

Re: Intel Announces 8 Core I9-9900KS: Every Core at 5Ghz, All the Time

#238

Earlier quoted context omitted.

I didn't measure. It is not too hot. I am typically getting 50-65 C under my workloads.

I forgot to add - it's been delidded. Crazy I know.

Woah, what do you use it for?

Re: Intel Announces 8 Core I9-9900KS: Every Core at 5Ghz, All the Time

#239
post #221
post #134

Earlier quoted context omitted.

AnandTech is different. They cover μarch, and their writers clearly understand what they're talking about.

Anandtech was sold (or maybe Anand left, I don’t recall exactly) about 5-6 years ago. The depth of their technical writing isn’t what it used to be.

they're still the best in the mobile review space. though that may say more about typical mobile phone reviews than AnandTech.

Re: Intel Announces 8 Core I9-9900KS: Every Core at 5Ghz, All the Time

#240

Earlier quoted context omitted.

As soon as the definition of "with all vulnerability mitigations on" stays stable long enough to put together a good review. Benchmarking a moving target is hell, and we don't have enough equipment or staff to do the around-the-clock regression testing that would be necessary to keep our benchmark database current with everything that's happened over the past 1.5 years.

People are using your benchmarks to decide what computer to buy. If they're that out of date, what should I tell them?

End-user perceived performance is usually not affected enough to meaningfully change the ranking of products. If a chip goes from being 5% faster to 3% slower when mitigations are applied, you'll never notice that without busting out a stopwatch and digging for a reason to be disappointed. Remember, measurable performance differences aren't always noticeable performance differences, especially without a side by side comparison.

And if two competing processors are close enough in performance for these mitigations to change which one comes out on top of benchmark charts, then other factors like price, power consumption and IO capabilities are probably a much bigger deal at that point than minor CPU performance differences.

Most if not all of our benchmark suites have been updated to include at least the early Spectre/Meltdown/et al. mitigations, and new CPUs are being tested with the microcode they launch with.

Post reply on HN