Intel's Future Chips: News, Rumours & Reviews

[jimmysmitty]

I am not sure the node race is getting tight. It may seem that way but considering the amount of money Intel puts into R&D alone I think it is a matter of just naming it to seem more competitive than they are.

I honestly would like to see a performance version of Samsungs 14nm compared to Intels current 14nm. The best comparison for Ryzen in terms of 14nm efficiency would be using Intel low power or mobile chips where Intel focuses on efficiency and power draw more than anything else.

 

[minigaming]

I'm on 6600k, was hoping for better IPC and single core performance with the 8000 series. If it's not much better, will get a 7700k or consider ryzen considering the price will no doubt drop further on the release of these chips and make the value proposition higher.

 

[gamerk316]

Still on my 2600k, still at stock (still not sure which of the CPU and mobo lost the lottery), still see zero reason to upgrade.

 

[juanrga]

Coffe lake engineering sample doing 5GHz

You can OC it to 6Ghz using liquid nitrogen but the retail version will be stuck at poor 4.3Ghz single core and 4Ghz on all cores, what a lame cpu.

LN2 get you in the 8GHz mark. And those turbos aren't the information I have.

 

[juanrga]

In addition being Coffee Lake 14nm the IPC will be the same, despite Intel labels it 14nm++ just to catch the marketing hype. Skylake IPC was the same of Kaby Lake IPC which is the same of Coffee Lake IPC because they share the very same 14nm old archtecture.

IPC is a metric of the architecture and completely independent of the process node used. The process node used affects to other metrics such as power consumption, base and turbo clocks and overclocking capacity.

Skylake --> 14nm

Kabylake --> 14nm+ (higher clocks than 14nm)

Coffeelake --> 14nm++ (higher clocks than 14nm+)

 

[juanrga]

Intel Mesh, Does Overclocking it Fix Skylake-X?
Hardware Unboxed
Published on Jul 19, 2017
https://www.youtube.com/watch?v=mTQ6ymQIY64
3GHz Mesh overclock

As demonstrated above Hardware unboxed tested in GPU-bound and frame-limited situations. This is the reason why when overclocked the 7800X by 34% they got only 3% extra performance compared to stock: the CPU is being bottlenecked by the GPU and the settings. On such situations it doesn't matter if we push the core to 5GHz, the uncore to 3.2GHz or if we replace the CPU by one was 10x faster. The FPS will remain similar because the CPU is being bottlenecked. When the CPU is not bottlenecked, overclocking the mesh brings up to double digit percent gains compared to stock settings

If you watch the video overclocking the mesh accounts for an average of 3%.

That is what happens with GPU-bound and frame-limiting settings.

 

[juanrga]

Spoiler

I am talking about efficiency, not power consumption. If a chip consumes 50% more power than other but does 100% more work, then the chip is more efficient.

Moreover the power consumption figures that toms got for the Intel chips don't look correct (and disagree with finding with other reviews). In the first figure you can see the 6900k @3.8GHz consuming more power than the 6900k @3.2GHz. And they also had problems in the past with the 7700k:

Unfortunately, the luck of the draw worked against our German lab, and we received a sample on the lower end of the quality range.

This doesn’t affect the benchmark results at all. However, it does have an impact on the power consumption and cooling results, as well as how far this particular CPU sample can be overclocked.

More than 50% power draw by Intel and less efficient despite the disparity in fab. When you design chips you will give up certain abilities in favor of other ones. But these clearly shows the power of the Ryzen uArch in power consumption and overall efficiency for what it was designed for.

Efficiency is not measured in those graphs. We only have power consumption in one bench and performance in two different benches. That is not how we measure efficiency. Moreover, as stated a dozen of times, Prime95 works as a power virus for Intel chips, but not for RyZen chips. Here we have an example of power consumption using Excel

Here another using x264

You are using excel to set your standard, which is ridiculous! I stand by the last 2 reviews benchmarks and statements I posted, which actually are a true gauge not this.

Edit: None of the other processors are overclocked except Ryzen. And look at that excel chart all overclocked Ryzen, which everyone knows power consumption spikes beyond normal specifications. You are using skewed charts to make your point!

I just give Excel and x264. And both contain power consumption of RyZen chips at stock settings.

Why not leave a link as a point of reference to the article so it can be judged vs. other articles.
Oh, now I see why you don't want to use the newest review from techspot. You don't want anything else brought up by that review that hurts the point you are trying to make. And by your assumptions 20nm Ryzen is doing great vs 14nm Intel in efficiency.

It is ~20nm vs ~16nm.

The SKL-X review has changed the measurement procedure, but the main conclusions are the same than in the old review.

Older review
i7-6900k: 198W
R7-1800X: 238W
Gap: 20.2%

Newest review
i7-6900k: 155W
R7-1800X: 193W
Gap: 24.5%

On both cases the '95W' chip is consuming more power than the 140W chip, specially when we consider platforms variations as dual-channel vs quad-channel.

Now on SKL-X on the new review, I am ignoring power consumption figures because many SKL-X tests used buggy BIOS launch, which affected turbo and p-state policies, and thus power consumption figures and overclocking capacities. THe own review mentions something about BIOS updates.

 

[-Fran-]

Now on SKL-X on the new review, I am ignoring power consumption figures because many SKL-X tests used buggy BIOS launch, which affected turbo and p-state policies, and thus power consumption figures and overclocking capacities. THe own review mentions something about BIOS updates.

And you haven't even made half the effort of taking into account the buggy BIOS'es and AGESA upgrades for/from AMD in articles when laying out your points that mostly favor Intel.

Cheers!

 

[goldstone77]

Intel Mesh, Does Overclocking it Fix Skylake-X?
Hardware Unboxed
Published on Jul 19, 2017
https://www.youtube.com/watch?v=mTQ6ymQIY64
3GHz Mesh overclock

As demonstrated above Hardware unboxed tested in GPU-bound and frame-limited situations. This is the reason why when overclocked the 7800X by 34% they got only 3% extra performance compared to stock: the CPU is being bottlenecked by the GPU and the settings. On such situations it doesn't matter if we push the core to 5GHz, the uncore to 3.2GHz or if we replace the CPU by one was 10x faster. The FPS will remain similar because the CPU is being bottlenecked. When the CPU is not bottlenecked, overclocking the mesh brings up to double digit percent gains compared to stock settings

If you watch the video overclocking the mesh accounts for an average of 3%.

That is what happens with GPU-bound and frame-limiting settings.

You are WRONG! Watch the video, because your comments I know you didn't. Overclocking the mesh to 3GHz caused random game crashes, and on boot failure to find the SSD. The 7700K wasn't limited in the vast majority of the games and neither was the 7800X, and those were the games he focused on. The mesh for the most part gave 3-6% gains over stock performance, but once overclocked those gains diminished to 1-3% overall. Plus system instability.

 

[goldstone77]

It is ~20nm vs ~16nm.

The SKL-X review has changed the measurement procedure, but the main conclusions are the same than in the old review.

Older review
i7-6900k: 198W
R7-1800X: 238W
Gap: 20.2%

Newest review
i7-6900k: 155W
R7-1800X: 193W
Gap: 24.5%

On both cases the '95W' chip is consuming more power than the 140W chip, specially when we consider platforms variations as dual-channel vs quad-channel.

Now on SKL-X on the new review, I am ignoring power consumption figures because many SKL-X tests used buggy BIOS launch, which affected turbo and p-state policies, and thus power consumption figures and overclocking capacities. THe own review mentions something about BIOS updates.

You typically ignore things that don't help your point! Nevermind that 1st generation Ryzen is exactly that. The first generation. Intel has been optimizing the same uAch for the last ~10 years, and they have a superior process node. You pick a test that shows the greatest disparity between the two as an example which is ridiculous, because Ryzen as I stated before wasn't made to run excel it was made for multi-tasking. And when you compare the two on power consumption and performance Ryzen looks really good! And I think ASML who makes the machines for all the semiconductor foundries might have a better idea about how analyze the process nodes, and come up with an equation to do this better than random commentators on forums.

 

[YoAndy]

I'm on 6600k, was hoping for better IPC and single core performance with the 8000 series. If it's not much better, will get a 7700k or consider ryzen considering the price will no doubt drop further on the release of these chips and make the value proposition higher.

Your statement makes no sense. You want better IPC than your 6600k But if is not good enough with the 8th generation you are getting Ryzen? Ryzen IPC still lower than intel.

 

[goldstone77]

Here is just a topic for fun. Hypothetically, we are coming to the barrier on process shrink. So, when we actually hit this barrier how do we continue to increase performance. More cores more threads seems like the only answer now, and programming will have to follow suite. Anyone else have any thoughts on what the future holds?

 

[Wizard61]

Here is just a topic for fun. Hypothetically, we are coming to the barrier on process shrink. So, when we actually hit this barrier how do we continue to increase performance. More cores more threads seems like the only answer now, and programming will have to follow suite. Anyone else have any thoughts on what the future holds?

The end of silicon and the dawn of carbon nanotube based chips.

 

[goldstone77]

Here is just a topic for fun. Hypothetically, we are coming to the barrier on process shrink. So, when we actually hit this barrier how do we continue to increase performance. More cores more threads seems like the only answer now, and programming will have to follow suite. Anyone else have any thoughts on what the future holds?

The end of silicon and the dawn of carbon nanotube based chips.

Even then we will get down to the point where only a single electron can travel along a path. How do we increase performance when we hit the barrier.

 

[gamerk316]

Here is just a topic for fun. Hypothetically, we are coming to the barrier on process shrink. So, when we actually hit this barrier how do we continue to increase performance. More cores more threads seems like the only answer now, and programming will have to follow suite. Anyone else have any thoughts on what the future holds?

As I've noted for, what, almost a decade now:

Games do not scale.

What's going to happen at either 7nm or 4nm [I have concerns about 4nm being commercially viable] is we're going to hit a computing brick wall. Servers and supercomputers will still scale by virtue of having workloads that scales to more cores, but Desktops are essentially going to tap out performance wise [CPU side at least] as normal desktop tasks don't need that much computing power, and the lack of scaling in games.

There really isn't much in the immediate horizon to address this issue. Optical chips have the major downside of being power hungry beasts, quantum computing isn't anywhere close to being ready or commercially viable, silicon alternatives [Graphene, etc] aren't there yet, and so forth.

 

[goldstone77]

Here is just a topic for fun. Hypothetically, we are coming to the barrier on process shrink. So, when we actually hit this barrier how do we continue to increase performance. More cores more threads seems like the only answer now, and programming will have to follow suite. Anyone else have any thoughts on what the future holds?

As I've noted for, what, almost a decade now:

Games do not scale.

What's going to happen at either 7nm or 4nm [I have concerns about 4nm being commercially viable] is we're going to hit a computing brick wall. Servers and supercomputers will still scale by virtue of having workloads that scales to more cores, but Desktops are essentially going to tap out performance wise [CPU side at least] as normal desktop tasks don't need that much computing power, and the lack of scaling in games.

There really isn't much in the immediate horizon to address this issue. Optical chips have the major downside of being power hungry beasts, quantum computing isn't anywhere close to being ready or commercially viable, silicon alternatives [Graphene, etc] aren't there yet, and so forth.

I think it is likely we will see higher CPU counts, but with specialized cores optimized for specific tasks. And better optimizations for software to make use of these cores.

 

[TMTOWTSAC]

386 era external math coprocessors finally making a comeback?

 

[-Fran-]

Even then we will get down to the point where only a single electron can travel along a path. How do we increase performance when we hit the barrier.

Non-uniform (non-binary?) ways of addressing performance.

The whole paradigm now is based on transistors that allow or deny electricity to travel. Get away from that computing paradigm and you can find new ways to extract performance.

Case in point: Quantum computing. That is the most famous one.

I like most analog theories on how to store data that deal with multiplanar/dimensional theories. It's fun, but just a tad complicated. It all boils down to how you can make the mathematical models adapt to what you can actually manufacture.

Here is just a topic for fun. Hypothetically, we are coming to the barrier on process shrink. So, when we actually hit this barrier how do we continue to increase performance. More cores more threads seems like the only answer now, and programming will have to follow suite. Anyone else have any thoughts on what the future holds?

As I've noted for, what, almost a decade now:

Games do not scale.

What's going to happen at either 7nm or 4nm [I have concerns about 4nm being commercially viable] is we're going to hit a computing brick wall. Servers and supercomputers will still scale by virtue of having workloads that scales to more cores, but Desktops are essentially going to tap out performance wise [CPU side at least] as normal desktop tasks don't need that much computing power, and the lack of scaling in games.

There really isn't much in the immediate horizon to address this issue. Optical chips have the major downside of being power hungry beasts, quantum computing isn't anywhere close to being ready or commercially viable, silicon alternatives [Graphene, etc] aren't there yet, and so forth.

I don't think that question is related to games alone, though.

Like I said above, if we escape the binary paradigm imposed by transistors, we can extract more performance out of whatever material we choose to base manufacturing on.

Cheers!

 

[goldstone77]

386 era external math coprocessors finally making a comeback?

Well, one instance of this is Samsungs 8 core. I think things of this nature will likely continue to expand, and become more specific to run individualized tasks better than an all around core could. Like adding an ASIC processor etc.
Octa-core vs Quad-core: Conclusion
"So, what’s the difference between octa-core and quad-core in the modern smartphone business? Very little, as it turns out.

That octa-core term is more than a little misleading, as it doesn’t mean the doubling of quad-core multi-core performance that it suggests. Rather, it represents two independently operating quad-core set-ups squeezed together on one chip for the purpose of greater energy efficiency."

http://www.trustedreviews.com/opinion/octa-core-vs-quad-core-what-s-the-difference-2932452

 

[goldstone77]

Even then we will get down to the point where only a single electron can travel along a path. How do we increase performance when we hit the barrier.

Non-uniform (non-binary?) ways of addressing performance.

The whole paradigm now is based on transistors that allow or deny electricity to travel. Get away from that computing paradigm and you can find new ways to extract performance.

Case in point: Quantum computing. That is the most famous one.

I like most analog theories on how to store data that deal with multiplanar/dimensional theories. It's fun, but just a tad complicated. It all boils down to how you can make the mathematical models adapt to what you can actually manufacture.

That is actually a great point. Escape the physical boundaries! I like it!

 

[-Fran-]

That is actually a great point. Escape the physical boundaries! I like it!

Well, you never escape the physics/physical boundaries. For manufacturing purposes, you can only go sub-atomic for tangible stuff. Until you can tap into untangible stuff (electromagnetic waves and light-based ones) with tangible (for humans) purposes.

There are a *lot* of ways to tap into them, but your most annoying barrier is usually cost and current calculation prowess of current computers (modeling).

Cheers!

 

[goldstone77]

That is actually a great point. Escape the physical boundaries! I like it!

Well, you never escape the physics/physical boundaries. For manufacturing purposes, you can only go sub-atomic for tangible stuff. Until you can tap into untangible stuff (electromagnetic waves and light-based ones) with tangible (for humans) purposes.

There are a *lot* of ways to tap into them, but your most annoying barrier is usually cost and current calculation prowess of current computers (modeling).

Cheers!

Reducing the limitation placed by heat is a step in the right direction. And it will be interesting to see how they harness the power of quantum entanglement for distribution of information.

 

[gamerk316]

That is actually a great point. Escape the physical boundaries! I like it!

Well, you never escape the physics/physical boundaries. For manufacturing purposes, you can only go sub-atomic for tangible stuff. Until you can tap into untangible stuff (electromagnetic waves and light-based ones) with tangible (for humans) purposes.

There are a *lot* of ways to tap into them, but your most annoying barrier is usually cost and current calculation prowess of current computers (modeling).

Cheers!

Reducing the limitation placed by heat is a step in the right direction. And it will be interesting to see how they harness the power of quantum entanglement for distribution of information.

The problem with Quantum Computers I foresee is that while a certain subset of tasks are orders of magnitude faster to complete, the remaining tasks won't see any benefit. In that case, per-core performance and clocks will drive performance. And both those are likely to be SIGNIFICANTLY lower with quantum based CPUs.

I'll also say it again: 99.9% of desktop users are quite fine with an i3 class, if not a Pentium class CPU. I basically see 4-core + SMT becoming standard (with some 2+2 CPUs at the low end for budget/OEM builds), with everyone else being pushed more toward low-end server class CPUs. I don't see these high core count desktop CPUs having enough sales to justify production; there just aren't enough people out there who benefit from the performance of those amount of CPU cores in a desktop environment.

 

[Yarvolino]

Coffe lake engineering sample doing 5GHz

You can OC it to 6Ghz using liquid nitrogen but the retail version will be stuck at poor 4.3Ghz single core and 4Ghz on all cores, what a lame cpu.

Are you trying to say that they will not be able to be overclocked? That is a patently false statement.

No, I wasn't trying to say that.

I speak from the point of view of those ones who want to buy a CPU without OC them, yes they do exist. Such persons, me included, would like to have the fastest CPU for gaming when changing rig every lot of years and they choose the quickest game CPU therefore I7 7700K.

We are sad to know that Coffee Lake will be worst in gaming than a year old i7 7700k.

I do not want to overclock CPUs, I hope you won't blame me for that

And you are making that assumption on what? There are no retail versions of the chips yet just a few engineering samples that could or could not be like the final retail chip. So you are just speculating at this point.

Statistics say that Engineering Samples are always toned down in terms of Ghz by roughly 10%.

Unless this is the very first case in history that an ES is toned down by a 25-30%, the final numbers are quite closely predictable as f.i. just happened for Skylake and Kaby Lake when ESs were roughly 10% toned down.

In addition being Coffee Lake 14nm the IPC will be the same, despite Intel labels it 14nm++ just to catch the marketing hype. Skylake IPC was the same of Kaby Lake IPC which is the same of Coffee Lake IPC because they share the very same 14nm old archtecture.

You are one of a few rare people ho spends premium on a K unlocked processor that doesn't want to overclock it. And then blames the company. Is like asking Ferrari to sell you a Minivan faster than a formula one car. Those processors are like formula one cars, you have to tweak them

You made my day mate.

Cause I am not an OClocker, in your opinion, I can't buy a K CPU and complain it is slower in gaming than a year older CPU. Funny.

I am entitled to buy a K CPU even if I don't overclock it, that's because the 8700 without K doesn't have the same frequencies of the K version, therefore your statement pertains to those snob elitarian guys which OC CPU and think the non oveerclockers ones have to be wiped out from earth.

Wrong.

I have fiddled with OC a while in my life, I am just too old too get fun out of it anymore.

Intel is producing IMHO another Skylake X fiasco if to have the most FPS out of a game you are forced to buy an year old CPU called 7700k.

 

[Averletum]

Interesting informations

https://hardforum.com/threads/coffee-lake-lga-1151-6c-12t-launching-in-august-several-k-models-planned.1930226/page-16

https://www.overclock3d.net/news/cpu_mainboard/leaked_intel_i7_8700k_specifications/1