Intel Core i3-8121U 10nm processor hits ARK database

[3dcandy]

I expect the shouty tribalistic types aren't going to bother reading a thread with i3 in the title

For now I'm sure Intel can keep tweaking their 14nm process to excellent effect, so this pointing to still having problems at 10nm I don't think is significant. Other than a dent to pride when at one point Intel were generations ahead in process compared to everyone else.

Which kinda points to them falling behind lol. It's a debate that's difficult to call. I still think they have fallen behind, not in absolute terms but relative as you said.

[DanceswithUnix]

Which kinda points to them falling behind lol. It's a debate that's difficult to call. I still think they have fallen behind, not in absolute terms but relative as you said.

Oh at the very least least their lead has dwindled to nothing, it's pretty hard to say whether Samsung's mobile centric process is better than Intel's high speed logic process. Perhaps other than pointing to Intel's lack of active foundry customers

[3dcandy]

Deffo - and TMSC seem to have a few bases covered now with the ability to choose either low power+ or FINFET designs depending on what you choose. In other news, seeing as Trump monkey has banned ZTE from using Qualcomm I wonder if they will use Exynos instead?

[DanceswithUnix]

Deffo - and TMSC seem to have a few bases covered now with the ability to choose either low power+ or FINFET designs depending on what you choose. In other news, seeing as Trump monkey has banned ZTE from using Qualcomm I wonder if they will use Exynos instead?

The possible deal breaker is if they can use Android.

Android on RISC-V would be interesting. I'm not sure anyone could stop ZTE doing the open source AOSP Android on an open CPU, but without hi-def video and play store approval it would have a limited market.

[Corky34]

Which kinda points to them falling behind lol. It's a debate that's difficult to call. I still think they have fallen behind, not in absolute terms but relative as you said.

When compared to their own yard stick, as that's the only real comparison we can make, yes they have.

I think Intel are discovering, in practical terms, what some people theorized, that eventually you hit a wall going down the faster and smaller route, physics just gets in the way.

Personally i suspect it's one of the reasons they've been so reluctant on the parallel processing (multiple cores) front.

[3dcandy]

When compared to their own yard stick, as that's the only real comparison we can make, yes they have.

I think Intel are discovering, in practical terms, what some people theorized, that eventually you hit a wall going down the faster and smaller route, physics just gets in the way.

Personally i suspect it's one of the reasons they've been so reluctant on the parallel processing (multiple cores) front.

I disagree - Intel has been able to get away with doing as little as it can for ages. People would just accept that a basic laptop needed 2c4t. Low end gamers got 4c8t. Etc. etc. until AMD came along and offered more cores for less and they suddenly came out with 6c12t in a very short timeframe. A bit like NV have been able to do, they can just sit and wait to see what AMD could do and just better it a bit

[mtyson]

The ARK database page has now been updated to add that this processor supports AVX-512 Instruction Set Extensions.

[DanceswithUnix]

The ARK database page has now been updated to add that this processor supports AVX-512 Instruction Set Extensions.

Ah, the son of Larrabee instructions. Perhaps we are supposed to do graphics in software then which is why they disabled the integrated unit

[Corky34]

I disagree - Intel has been able to get away with doing as little as it can for ages. People would just accept that a basic laptop needed 2c4t. Low end gamers got 4c8t. Etc. etc. until AMD came along and offered more cores for less and they suddenly came out with 6c12t in a very short timeframe. A bit like NV have been able to do, they can just sit and wait to see what AMD could do and just better it a bit

I'm not sure what i said that you disagree with.

I wasn't saying Intel not wanting to push the multicore aspect of CPUs was a good thing, far from it as IMO Intel wanting to hold onto their lead in serial processing (i.e single threaded workloads) has been holding us back for a long while, they've just been putting of the inevitable, they didn't like the idea of giving people more cores specifically because they knew it's not playing to their advantage.

Put it this way if software developers start making better use of parallel processing the past however many years of R&D that Intel have spent on going smaller and faster is all but wasted, if anyone with a X86 license (read AMD) can just throw more cores at the problem instead of having to spend a fortune developing small, fast transistors then Intel's lead all but vanishes, especially when you consider that to get small and fast Intel have sacrificed power efficiency making them far from ideal in performance per watt markets.

[ETR316]

Does anyone remember when the spectre and meltdown debacle started and then went into full swing? Intel stated procs would be out this year with hardware level fixes. WHERE are those AND is anyone keeping track as to WHEN? OR do we have to wait until an entirely new architecture comes out? I would think so, but that would put a hardware fix on 10th Generation and not the 9th. Does anyone have more light on that subject?