8th Gen Core "Cannon Lake" Over 15% Faster Than Kaby Lake: Intel

[FordGT90Concept]

Uh... 7700K can turbo all cores to 4.4GHz? 5GHz+ is the result of 15% clock improvement not 4.83GHz.
https://forums.anandtech.com/thread...ost-cores-active-ratio.2498708/#post-38723230

The base clock is 4.2 GHz, not 4.0 GHz. Generally Intel's Turbo will not activate unless CPU load is below 100%. Handbrake isn't exactly a good benchmark for testing Turbo's behavior.

Exactly. People need to realize that the 7700K is not a high end processor. That's left up to X99 platforms.

Yet it features a high-end processor price, doesn't it?

 

[Naito]

Intel has been playing it conservatively. We all know that. If Ryzen offers any challenge, Intel will simply reply with more cores and more megahertz.

 

[Fluffmeister]

The battle of the duopolies, my heart bleeds.

 

[Aquinus]

Unless Cannon Lake is a significantly newer micro-architecture than Kaby Lake, we could expect them to come with even higher clock speeds. Will the Core i7-8700K be a 5 GHz chip?

What killed Netburst again?

 

[Rictorhell]

Assuming that Intel has not been holding CPU advancement back, intentionally, due to a lack of competition, it may be time for them to seriously consider using new materials for their CPUs or radically new methods for constructing/designing CPUs. Even Intel may be reaching the limits of what the current technology can actually accomplish. The problem is, like a lot of people AND companies, Intel probably doesn't care to make significant changes unless it is absolutely necessary.

I've been reading for a while about graphene and using light to transfer information within CPUs and this and that, but the reality is that the cost to implement these kinds of changes is probably sky high.

Intel probably has the capital and financing to pretty much do whatever they have to, but then the question of WHY pops up.

An Intel shareholder or board member will want to know why Intel should bother investing to bring these advancements and changes into reality.

If Intel can milk these incremental and pitiful performance increases for the next few years, advancing NOTHING, but still making SOME profits, while NOT heavily increasing spending, why not?

They have what most of us would probably consider to be a massive R & D budget. They have cutting edge fabs and the smartest people in the world on their payroll. They have billions of dollars. What they don't have is motivation.

People have no idea of what future computer hardware and software might be capable of, so for the most part, they are satisfied with how things are, but what if you knew that a year from now artificial intelligence could be 10x as advanced as it is now? What if you knew that a Holodeck could be an affordable reality for every home in America?

Intel doesn't HAVE to ask these questions right now, so they won't. If you're a software company or a hardware company or a computer person, start thinking about these things and asking those questions because otherwise we're going to be sitting here, 10 years from now, in the same situation.

 

[FordGT90Concept]

What killed Netburst again?

Cache misses, frequently.

 

[The Haunted]

The last time Intel released an exciting new product was in 2008 with Bloomfield, everything else until now has just been poor incremental increase in performance. I sincerely hope Amd will force them to step up their game because it's getting pretty boring. I would have never guessed that 8 years down the road, i would still be rocking the i7 920 with no reasons to throw hard earned money at Intel to "upgrade" (cpu performance wise)...

 

[ratirt]

For me Intel sucks with it's products. Got my i7 3770k and honestly i dont think of replacing it with another intel product. There's totally no point for doing that. But i really consider switching to AMD's Ryzen cpu which offers way more in terms of IPC and features and of course 8c/16t which in my opinion is awesome move from 4c/8t of my I7. For me that would be an improvement. 7700K can suck it since i'm not buying that piece of crap.

 

[efikkan]

Intel illustrated how Kaby Lake processors are on average 15 percent faster than Skylake parts, in SYSmark. While Kaby Lake has negligible IPC gains over Skylake, the newer chips are clocked significantly higher, making up Intel's performance targets. Unless Cannon Lake is a significantly newer micro-architecture than Kaby Lake, we could expect them to come with even higher clock speeds. Will the Core i7-8700K be a 5 GHz chip?
Source: VideoCardz

The IPC gains with Kaby Lake are not negligible, they are non-existent. There are nothing but minor bugfixes, so there is nothing to increase the IPC. And the advances in clock frequency and more aggressive throttling only yiels ~5-9% gains, nowhere near 15% as Intel claims.

In fact, the IPC gains from Sandy Bridge to Skylake is about ~6-15% in total, when we speak of generic calculations. Intel always touts 10-15% IPC gains between each refresh, but that has not been true since Sandy Bridge. Intel's marketing IPC claims are bloated with special features like AES acceleration, which of course have no impact on most applications. The only real changes since Sandy Bridge have been very minor, with the biggest being more vector extensions(which of course mainly helps certain applications), prefetcher improvements, and cache and memory improvements. Fundamentally the architecture of Skylake and Sandy Bridge is the same, and until Intel adds more execution ports all new revisions should be considered minor refreshes. Cannon Lake, like Kaby Lake, is not going to feature anything significant, that will at least have to wait for Icelake or later.

"+15%" for intel slides in real life means: "+0% IPC and +5% frequency" (that is a fact - not my opinion);.. and for this Cannon Lake - I would not hold my breth for another +5% frequency... so it will be precise +0% gains... well thx intel for honesty (I guess)

That is correct, minor tweaks all over. As I always say, the only reason to upgrade a Sandy Bridge or newer is to get more cores.

"Advancing Moore's law" Wasn't Intel who said Moore's law doesn't apply anymore?

Moore's law has never meant anything, and is nothing but an obscure quote. Moore's law has never said anything about performance, but simply "number of components per integrated circuit". Anyone with basic understanding of math will understand that a revolutionary new technology will have exponential growth in a limited period and will eventually flatten out. Moore's law is worthless and is not even worth a mention, especially since there are no real correlation between transistor count and performance.

I believe Intel has already hit the IPC ceilling of the architecture that originally started with Sandy Bridge. If it wasn't for the PCIe 3.0 and DMI 3.0 offered by 100 series chipsets, I'd have stayed with my Haswell.

Yes, Sandy Bridge was the last overhaul of the architecture. It increased the number of execution ports (doing instructions) from two to three, and everything since has mostly being optimizing the front-end and memory/cache, so the throughput has not increased. It is possible to break through the current "IPC ceiling", but the IPC gains are going to be minor until Intel decides to add more execution ports.

Someone needs to shake up x86 by slashing all the legacy cruft, and simplifying paging and memory types, drop puny 4k page size, replace hardware page walkers with Alpha-like PAL code. Remove self and cross modifying code hardware support. Rework floating point and slash legacy x87.

That's not accurate at all, the legacy part of x86 probably makes up less than 1% of the current transistor count. All modern x86 CPUs are implemented as a RISC architecture, and thereby attaining most of the advantages of other RISC based competitors.

If Intel were to replace x86 with ARM or something similar, almost all architectural features will still be almost the same; including the prefetcher, memory controller, the super-scalar execution ports, vector engines, special accelerated features, etc. The gain from replacing x86 with any of the competitors will probably be less than 2%, and considering all the software which needs to be optimized, it's not remotely worth it. x86 will remain until a revolutionary new architecture arrives, one which fundamentally adds/changes how we controls a CPU on microcode level.

You could argue the paging structure could use some tweaks, but none of that is blocked by the "legacy" from x86.

What killed Netburst again?

A super-long pipeline which caused huge performance penalties from branch mispredictions.

 

[Aquinus]

Cache misses, frequently.

A super-long pipeline which caused huge performance penalties from branch mispredictions.

You're both missing the point. Just like Bulldozer, if the only way you can get more performance is by pushing clocks higher and higher, you know you're going to experience the same fate as both Netburst and Bulldozer. Regardless of the "why", needing higher clocks to get more done is a warning sign.

 

[RejZoR]

And that's the thing, Intel is already hitting frequencies at which you cannot just keep increasing it further, beyond 5GHz. They are still doing it, but just barely with 100 and 200MHz gains for each new CPU release. The architecture Intel is using at the moment is not a sustainable architecture anymore. It doesn't even seem to gain anything by going from 22nm to 14nm. Only thing that they can do now is give more cores to normies who are still on quad cores in 2017. But that's about it. To make architecture sustainable again, they need to pull another "Core" architecture. A moment when they sacked Netburst and used entirely new efficient architecture which performed great at way lower clocks. This is the only way they can reset the cycle and open up a future for new CPU's.

What this means is they need to basically pull off architecture which would be running at sub 3GHz and be as or preferably more efficient than current Skylake/Kaby Lake at 4.2+ GHz. There is no other way going around this imo.

 

[ratirt]

And that's the thing, Intel is already hitting frequencies at which you cannot just keep increasing it further, beyond 5GHz. They are still doing it, but just barely with 100 and 200MHz gains for each new CPU release. The architecture Intel is using at the moment is not a sustainable architecture anymore. It doesn't even seem to gain anything by going from 22nm to 14nm. Only thing that they can do now is give more cores to normies who are still on quad cores in 2017. But that's about it. To make architecture sustainable again, they need to pull another "Core" architecture. A moment when they sacked Netburst and used entirely new efficient architecture which performed great at way lower clocks. This is the only way they can reset the cycle and open up a future for new CPU's.

What this means is they need to basically pull off architecture which would be running at sub 3GHz and be as or preferably more efficient than current Skylake/Kaby Lake at 4.2+ GHz. There is no other way going around this imo.

You really think that? Well we don't know if Intel's architecture hit it's limits. Maybe intel give this ridiculous low IPC gains cause of greed not that it wont be able to push it further. Although it might be true. Cannonlake is on the way but i doubt it would bring anything more to the table. New architecture is needed for intel wonder if they have anything like that. Hopefully they are working on a new architecture.

 

[FordGT90Concept]

I doubt a much better architecture can be designed for x86 than what Intel and AMD have now. Improvements are going to have to come from scalability (dependent on software more than hardware), better materials (higher clock speeds), and more specialized instructions (addition of HEVC acceleration in Kaby Lake is a good example).

Frankly, I wonder if it is time to reinvent the x86 wheel again. It would be nice to see a flavor of x86 that can compete with ARM in terms of efficiency and simplicity. Perhaps a more modular approach to x86 is what is needed.

 

[deu]

HEDT does not have an iGPU.

No but 7600 have 7700 have all intels non 6 core+ have.

 

[efikkan]

You're both missing the point. Just like Bulldozer, if the only way you can get more performance is by pushing clocks higher and higher, you know you're going to experience the same fate as both Netburst and Bulldozer. Regardless of the "why", needing higher clocks to get more done is a warning sign.

Pushing the clocks towards the "extreme" is a sign that the maker is left with no other choice.

We do in fact need more performance, and there is no future in scaling clock frequencies. We are only get minor gains there, not because higher clocks don't work logically, but rather because of physical limitations with the production technology. That's why it's hard to scale beyond 4-5 GHz, and each 100 MHz is going to require more effort. The advances are no longer the way we experienced in the 80s and 90s, where clock frequencies keep doubling or so. Intel intended Netburst to scale to 6-8 GHz, but the energy inefficiency forced them to look for other options for performance scaling.

The "why" matters for technical discussions, because it's a clear indication of what gains we can have in the years to come. If Intel continues with no major overhaul the performance will continue to be flat. Yes they will continue to gain 100 MHz here and there, yes they will get a 0.5% performanche here and there, but none of that is going to be significant.

Even though scaling is harder now that we can't scale with clocks any more, it's still highy possible. In fact there are two ways to increase IPC looking forward; more execution ports and more vector extensions. In the future we are going to need more of both. AMD's Zen will in fact increase the superscalar abilities. While Intel has since Sandy Bridge used three execution ports (not including load/store, etc.), Zen will seemingly feature 4 ports with ALUs and 2 additional ports for FPUs. This will mean that AMD with a smaller core and less advanced front-end(prefetcher) still will have an advantage in some workloads, such as Blender and other benchmarks. Hopefully Intel will get their act together and do so too, especially adding more ALUs, since they are relatively cheap and can still help boosting IPC a lot. If we in ~10 years have a cores with 8 ALUs, some more FPUs and some improved vector engines we can get nearly twice the current IPC, of course a such increase will happen gradually.

Speaking of the mistakes of Netburst and Bulldozer. The problem for Netburst was not cache misses, the penalty of cache misses are pretty much constant, and does not scale with clock frequency. The problem were the length of the pipeline (31 stages vs. 12 for Athlon 64), which makes each branch misprediction vastly more costly. In fact Netburst got a better prefetcher than AMD, so Intel were better at guessing, but the cost of guessing wrong was so high that they lost a lot of performance. And since these CPUs are superscalar with two operations/clock, the penalty difference is actually twice as large as you might think; 31-12 = 19 clocks => 38 operations. This means that for every misprediction by both CPUs, AMD got 19 "free" clocks which Intel lost 19-38 operations.
Bulldozer did not only do the mistake of a long pipeline, but also sharing prefetcher resources and FPUs between two CPUs, which basically cripples the efficiency.

I doubt a much better architecture can be designed for x86 than what Intel and AMD have now. Improvements are going to have to come from scalability (dependent on software more than hardware), better materials (higher clock speeds), and more specialized instructions (addition of HEVC acceleration in Kaby Lake is a good example).

Frankly, I wonder if it is time to reinvent the x86 wheel again. It would be nice to see a flavor of x86 that can compete with ARM in terms of efficiency and simplicity. Perhaps a more modular approach to x86 is what is needed.

Like I've mentioned, there is great potential in increasing the superscalar features.
Specialized features will play a part, but only for specific needs, like AES, video acceleration, etc.

 

[R-T-B]

Probably we will see a 4.8Ghz CPU and that's all. Nothing to see here for the next 2 years...

There's already one in my specs.

 

[Nihilus]

Any chance that Cannondale will bring iGPU levels of that from Broadwell? I kind of thought that was going to be the whole point of KabyLake when I first heard of it. At least it would of had a point. Instead it was simply Skylake version 1.01 Hopefully, this is not just Skylake version 1.02

 

[dat_boi]

So what you all expect is that Intel drops the latest high-tech CPU freakin' nano-graphenetubes and no one would have to buy a new CPU(or sell any) for at least 10-20 years. This is capitalism, the goal isn't a technological advancement but profit.

Honestly, any fanboy of either brand and others who frequent sites like this needs to realize this.

 

[R-T-B]

So what you all expect is that Intel drops the latest high-tech CPU freakin' nano-graphenetubes and no one would have to buy a new CPU(or sell any) for at least 10-20 years. This is capitalism, the goal isn't a technological advancement but profit.

Honestly, any fanboy of either brand and others who frequent sites like this needs to realize this.

We do. It's precisely why everyone here (save for maybe knoxx, but he's beyond saving) wants a return to competition.

 

[FordGT90Concept]

So what you all expect is that Intel drops the latest high-tech CPU freakin' nano-graphenetubes and no one would have to buy a new CPU(or sell any) for at least 10-20 years. This is capitalism, the goal isn't a technological advancement but profit.

Honestly, any fanboy of either brand and others who frequent sites like this needs to realize this.

There's been no serious talk about commercial availability of different substrates. Intel has no doubt been experimenting with possibilities but they're no where close to being ready to roll something out in the next year or two.

 

[dat_boi]

We do. It's precisely why everyone here (save for maybe knoxx, but he's beyond saving) wants a return to competition.

At what point is it still a real competition? In a free market wouldn't Intel be able to buy out its competition already? No huge technological advancements are really desired in this industry, by companies, shareholders not even by the US government due to how big part of the economy it is.

 

[R-T-B]

At what point is it still a real competition?

When AMD produces a more competitive product, as they may be about to do (we hope)?

In a free market wouldn't Intel be able to buy out its competition already?

Yes, but we don't live in Rapture.

No huge technological advancements are really desired in this industry, by companies, shareholders not even by the US government due to how big part of the economy it is.

I don't see how that makes any sense.

 

[dat_boi]

Of course it wouldn't make sense to you, you've been a hw enthusiast for thiiiiis long. Ever wondered why others moved on?

Some red pills are really hard to swallow, mainly if they are about things you really like.

 

[FordGT90Concept]

At what point is it still a real competition?

When the decision between two products is tough because their price/performance is similar.

In a free market wouldn't Intel be able to buy out its competition already?

USA has a regulated market where it is usually illegal for businesses to buy out competitors, especially when there are few players in the market. If Intel tried to buyout AMD, the FTC would likely forbid it because AMD is the last major x86 manufacturer besides Intel.

No huge technological advancements are really desired in this industry, by companies, shareholders not even by the US government due to how big part of the economy it is.

The status quo was Moore's Law which stipulated that the number of transistors on a die doubles about every 24 months (in practice, was closer to 18 months). This rule generally applied since the first Intel microprocessors debuted until Devil's Canyon (aka Haswell Refresh). In other words, "huge technological advancements" were the norm and now they are not. Intel hasn't scrambled to fix its problems because AMD was so far behind. AMD has completely caught up and likely surpassed Intel. Everyone seems to think 10nm and 7nm is the future but no one has actually launched a product that will do it so, tech is kind of in a holding pattern right now.

 

[Prima.Vera]

For me Intel sucks with it's products. Got my i7 3770k and honestly i dont think of replacing it with another intel product. There's totally no point for doing that. But i really consider switching to AMD's Ryzen cpu which offers way more in terms of IPC and features and of course 8c/16t which in my opinion is awesome move from 4c/8t of my I7. For me that would be an improvement. 7700K can suck it since i'm not buying that piece of crap.

If you think AMD's Ryzen will be faster than your (my also) 3770K CPU, you are a very optimistic person.