Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Intel Reveals Post-8th Gen. Core Architecture 10nm+ Ice Lake

anandtech.com

Post reply on HN
Print view View original

111–120 of 247 posts

Re: Intel Reveals Post-8th Gen. Core Architecture 10nm+ Ice Lake

#111

Post by luch »

turblety wrote: post #103
baq wrote: post #89

Earlier quoted context omitted.

The backdoor is super useful for large enterprises, don't hope for much.

The issue I have is companies like Google and Puri.sm have asked Intel and AMD for a blank signed blob that completely disables ME but they have refused this. It would take them literally no time at all. This raises all sorts of red flags that something dodgy is going on. If you had the chance to make a supplier who creates millions of chromebooks happy, wouldn't you take every opportunity to help them, especially if…

halvar flake (Google P0 security guy) talk a bit about it in the last Black Hat Asia : https://www.youtube.com/watch?v=JCa3PBt4r-k.

Basically he says that even Google is puny (in terms of production units) in front of Intel or Samsung, and cannot ask for custom firmwares.

Hardware security is currently a shit show because of global monopolies/"oligopolies".

Re: Intel Reveals Post-8th Gen. Core Architecture 10nm+ Ice Lake

#112

Post by hvidgaard »

Dylan16807 wrote: post #91
hvidgaard wrote: post #85

Earlier quoted context omitted.

It was simply a response to your statement about bruteforcing anything. There is some ways to factor a prime that is much faster than bruteforcing, and yes they work on a GPU. But that has nothing to do with bruteforcing, but clever math.

The statement was about the relative speed of a thousand dumb cores vs. a couple really good cores. Not about absolute speed.

The statement was stating an absolute. Algorithmic complaxity matters more than the speed of the computing hardware. Obviously, more power means you can compute more, but not everything.

Re: Intel Reveals Post-8th Gen. Core Architecture 10nm+ Ice Lake

#113

Post by imtringued »

slededit wrote: post #62
IntelMiner wrote: post #56

Earlier quoted context omitted.

"GPUs are only good at _very_ specific workloads." Let me detail you a completely specific workload citing my own personal experience over objective facts with citations Oh and let me also take a swipe at Intel again without any verifiable evidence

Primecoin is a good example. If you look at the implementation it requires generating a tightly packed bitfield for the sieve and then randomly accessing it after. Lots of synchronization required so you don't overwrite previously set bits are required and the memory accesses are random so its suboptimal for the GPU's memory subsystem. It took under a year for a GPU miner to come out. Having optimized it for Intel in…

But you don't have thousands of cores. You have a medium amount of cores (64 in the latest Vega 64 GPU) with a high amount of ALUs (and threads) per core. When the GPU executes an instruction for a thread it looks if the other threads execute the same instruction and then utilises all ALUs at once. This is great for machine learning and HPC where you often just have a large matrix or array of the same datatype that you want to process but most of the time this isn't the case.

Re: Intel Reveals Post-8th Gen. Core Architecture 10nm+ Ice Lake

#114

Post by msh »

Skunkleton wrote: post #58
msh wrote: post #53

Earlier quoted context omitted.

They have fallen a long way since Adnan ran the site...

Adnan is the guy who may or may not have killed his girlfriend in the 90's. You are thinking Anand. As in AnandTech. And even without Mr. Shimpi AnandTech is still one of the best sources for in-depth reviews of hardware.

You are right about the name, I blame autocorrect.

I think especially their phone and tablet reviews have gone down a lot. Fx no deep dive into the newer iPhones and iPads.

Re: Intel Reveals Post-8th Gen. Core Architecture 10nm+ Ice Lake

#115

Post by meric »

ido wrote: post #75
std_throwaway wrote: post #74

Earlier quoted context omitted.

Gains in instructions-per-clock start to flatten out. And that's where the gains were coming from in the last years. Some time ago a paper was posted here that showed how even if you have an infinite amount of transistors, you will still be limited in the range of 3-10 instructions-per-clock for typical programs. Clock speeds seem to have leveled and IPC will only see another gain of 50-100%. Single threaded performa…

When was the last time we saw a 50-100% performance gain in cars? airplanes? spacecraft? Was it the end of those industries? Welcome to mature technology.

> spacecraft

Economically? I think this or last year.

Re: Intel Reveals Post-8th Gen. Core Architecture 10nm+ Ice Lake

#116

Post by vbezhenar »

sspiff wrote: post #90
turblety wrote: post #83

I wonder if this one will come with free backdoors and spyware installed, thanks to the wonderful Intel Management Engine (Intel ME) backdoor. [1][2][3] Intel (and AMT) keep pushing more and more proprietary code that can not be read, changed or removed. No one knows exactly what it does and it has built in screen and key recording. It's my advice and the advice of privacy advocates that no one should purchase or use…

I'm not sure this should be discussed in this thread. Also, I don't know of any alternative that doesn't have large unauditable blobs integrated into the chip. All ARM SoCs come with radio processors that are running a non-trivial piece of software with full access to the system memory, which is responsible for power management, boot sequence and wireless communications. It is by definition network connected. AMD has…

Check out Talos II motherboard. It's a workstation-class motherboard with dual POWER9 CPUs for $2750. It's a good price for workstation computer IMO. They claim that all their firmware is open source. Specifications are quite modern. The only problem is (kind of) exotic architecture, but many people would be able to use it with open source software.

Re: Intel Reveals Post-8th Gen. Core Architecture 10nm+ Ice Lake

#117

Post by thinkfurther »

piva00 wrote: post #87
turblety wrote: post #83

I wonder if this one will come with free backdoors and spyware installed, thanks to the wonderful Intel Management Engine (Intel ME) backdoor. [1][2][3] Intel (and AMT) keep pushing more and more proprietary code that can not be read, changed or removed. No one knows exactly what it does and it has built in screen and key recording. It's my advice and the advice of privacy advocates that no one should purchase or use…

You know the depth of how unrealistic is your advice, right?

Do you know the depths of not taking that advice and what lurks in them? Do you know that if everybody simply took it to heart, there'd be nothing unrealistic about it at all? How many months of abstinence and solidarity would be required to end these practices, or the companies if they so wish? And then that money simply shifts to ethical companies and we actually have a future. Or, we keep pretending it's all so very hard, and don't have one.

Re: Intel Reveals Post-8th Gen. Core Architecture 10nm+ Ice Lake

#118

Post by PaulKeeble »

Skunkleton wrote: post #58
msh wrote: post #53

Earlier quoted context omitted.

They have fallen a long way since Adnan ran the site...

Adnan is the guy who may or may not have killed his girlfriend in the 90's. You are thinking Anand. As in AnandTech. And even without Mr. Shimpi AnandTech is still one of the best sources for in-depth reviews of hardware.

They have fallen behind in CPU and GPU testing methodology especially around games. Their testing of databases was pretty awful (it fit in memory) and a variety of other obvious limitations to their testing I would argue they are not only worse than they were when Anand ran the show but now significantly worse than a lot of the places they compete with.

Pcper is a significantly more capable review site these days.

Re: Intel Reveals Post-8th Gen. Core Architecture 10nm+ Ice Lake

#119

Post by greggyb »

luch wrote: post #111
turblety wrote: post #103

Earlier quoted context omitted.

The issue I have is companies like Google and Puri.sm have asked Intel and AMD for a blank signed blob that completely disables ME but they have refused this. It would take them literally no time at all. This raises all sorts of red flags that something dodgy is going on. If you had the chance to make a supplier who creates millions of chromebooks happy, wouldn't you take every opportunity to help them, especially if…

halvar flake (Google P0 security guy) talk a bit about it in the last Black Hat Asia : https://www.youtube.com/watch?v=JCa3PBt4r-k . Basically he says that even Google is puny (in terms of production units) in front of Intel or Samsung, and cannot ask for custom firmwares. Hardware security is currently a shit show because of global monopolies/"oligopolies".

If Google is tiny in Intel's eyes, who isn't?

They operate a top 3 cloud service and have enormous internal data centers as well.

Re: Intel Reveals Post-8th Gen. Core Architecture 10nm+ Ice Lake

#120

Post by ksec »

bhouston wrote: post #31

What will be the clock when single threaded? Can we get above 5GHz finally?

We need something other then Silicon to have 4Ghz + at reasonable Power usage. I think Material Science is going to be the next big thing when we reach the limit of Silicon.

I really wish we could see more IPC and 10Ghz+, our single threaded performance have be stuck at the same level for far too long.

Post reply on HN