Windows 10 Scheduler Aware of "Lakefield" Hybrid Topologies, Benchmarked

[Chrispy_]

Apparently you're incapable of doing basic math... 10W is ~42% more power than 7W and 100% more than 5W, which are massive proportions at this level.

That's just insulting; You and I both know I've posted complex calculations on these forums before and as a chartered engineer, even if I was a bad one, basic math skills are a given.

He's simply trying to take the worst/worst scenario to exaggerate his point by looking only at the extreme options in order to falsely prop up his argument.

The article clearly states it's 7W, not 5W.
"The Core i5-L16G7 has a rated SDP (scenario driven power) rating of 7 W"

AMD's offical cTDP values for a 4800U are 10-25W, and those official figures guarantee the 1.8GHz base clock assuming adequate cooling is provided. If your 4800U doesn't achieve 1.8GHz at 10W, RMA it because it's out of spec, ergo faulty.

10W compared to 7W may still be a sizeable 42% increase, but it's not the 200% increase he's trying to make it out to be. I'd certainly wager that a 10W 4800U is more than 42% faster than this 7W Core i5-L16G7....

 

[londiste]

I've long felt a bigLITTLE approach will yield some of the easiest general improvements which could then be scaled and and intermixed a bit to suit certain consumers requirements. I still really feel FPGA's could be the best all around solution outside of combining a variety of ASIC's to really specifically maximize and prioritize a handful of the individuals use cases.

FPGA and ASIC have a different purpose. FPGA compared to ASIC is power inefficient, has lower clocks and cannot do some things (like analog circuitry). It is good for prototyping and low-volume production or very specific use cases. For most normal users, both desktop as well as enterprise, mass-manufactured ASICs are a better way to go.

AMD's offical cTDP values for a 4800U are 10-25W, and those official figures guarantee the 1.8GHz base clock assuming adequate cooling is provided. If your 4800U doesn't achieve 1.8GHz at 10W, RMA it because it's out of spec, ergo faulty.

This is incorrect. Official spec clearly states 1.8GHz and 15W. cTDP will affect clock speeds and 1.8GHz is absolutely not guaranteed at 10W.

 

[watzupken]

Finally, tests that are actually useful because they're against a competitor part designed for the same low-power scenario!

And the results are very impressive considering this is a first-gen attempt. At only ~71% of the 8CX's power (5W vs 7W) Lakefield is faster in all non-synthetic benchmarks, despite having a 3-core deficit.

In short, Windows-on-ARM just died, along with Qualcomm's ambitions for ultra-portables in the x86 space. Conversely, Microsoft now has a reason to try their hand at Windows Phone again.

This is the most exciting and most important thing to happen in the CPU industry since Ryzen.

Success or not, it depends on how fast Intel is running/ improving this. I feel it should not come as a surprise that Intel chips are faster when running Windows OS and apps, which is almost like a home ground for Intel chips anyway. However, ARM improves on their SOCs year on year, and we can already see how close they are to closing in on Intel. So I would feel it is premature to say that Windows on ARM just died.

I feel people buying a laptop with an ARM or Lakefield processor is unlikely to consider performance as a priority. Between the 2, the performance delta is not ground breaking for most users to experience during light usage, which is likely going to be the use case for this class of laptop. Rather, battery life is the main consideration which ARM based SOCs still hold an advantage.

In my opinion, another Windows phone (even if viable from a hardware standpoint) may be too late at this stage since the market is already very saturated.

 

[Chrispy_]

This is incorrect. Official spec clearly states 1.8GHz and 15W. cTDP will affect clock speeds and 1.8GHz is absolutely not guaranteed at 10W.

Rather than argue and speculate, here's a 10W 4700U exceeding its base clock and actually boosting to 2GHz on a 10W limit.

I cannot find a 4800U review yet, but on the assumption that is better-binned silicon it should be better, not worse.

As more anecdotal evidence, my ancient 1st-gen 2700U boosts above its base clock by around 150MHz even when configured down to its minimum 12W cTDP.

 

[londiste]

Cool if that is the case.
Still does not change my point about spec.

 

[Chrispy_]

Cool if that is the case.
Still does not change my point about spec.

Until I get my hands on one, I can't be sure, but based on my experience with 2000U and 3000U series, the 1.8GHz base clock may actually be the minimum P-state - as in the CPU cannot even run slower than that because the next state is "core parked".

If 1.8GHz is minimum P-state, the only time you'll see lower clocks than that is when you aren't running an all-core load and the CPU is shifting the threads around the cores to distribute heat better. The lower clocks you see will be an average between the active minimum P-state clock, and zero for a parked core. For example, in AMD uProf I can see clocks 300MHz lower on all 8 logical cores than the minimum when running just three threads, but if I change to an all-core load the clock rises to at least the base clock.

To the best of my knowledge, Renoir will follow the same rules as 1st Gen Ryzen, and if you read the cTDP whitepapers you'll see that the reason 1st-Gen's cTDP only goes down to 12W is because at this point it cannot maintain 8 active threads. I have the ability to set it lower using RyzenAdj but that just means it's going to throttle so hard it will be forced to park cores. In the context of this article discussing Lakefield's hybrid topologies, it almost doesn't matter what clockspeed the lowest active P-state is, so long as all cores are active at the stated cTDP. The whole argument is nearly a moot point anyway, because Intel's definition of TDP is very different to AMD's definition of TDP and there's no shortage of mainstream Youtubers independently demonstrating that Intel's TDP figures are even further away from real-world power consumption values than AMD's (which are quite clearly restricted and controllable via most BIOS options and/or Ryzen Master).

 

[slehmann]

I really do miss old times when review outlets were less biased and fanboyism was not even in discussion. I remember AT articles written by Anand, that were just as detailed for a via shitty CPU or the best Intel or AMD cpu.
Nowadays, people are full of hate. If Intel has had a rough time now, everyone is wishing their death. If Intel then comes and tries to create an interesting product everyone say it is shit and boring. If AMD would have come out with something similar, everyone would be praising it for how good it is. Dear people, remove the hate/love and try to see objectively what each company is doing. AMD has good products, I agree. But AMD also has bad products, very bad actually. Likewise for Intel, for Nvidia, etc. So lets learn to judge each situation separately and not introduce the same stupid hate on each product launch.
This product is a very interesting piece of design, silicon, power delivery, uArch, software, firmware, etc. It has some specific constraints that require a lot of effort, for example the very small package, like ARM SoCs. Stop being so ignorant and try to read every single piece of news with a clear mind. Intel will still launch great products in the future, don't get bogged into this 14nm+++++ hate. They didn't want to push it for so long, but making mistakes in the fab business is very costly and as we saw a multi year affair to fix. We should appreciate what AMD has done, but stop being so polarized and appreciate also what the competition does, because it might just be a good product.

Yeah, another big difference is that they are in two different leagues altogheter in terms of package sizes, power requirements, PCB minimum size, etc. Intel also can drop their i7 15W CPUs to 4.5W (which they actually do for a very long time), but Lakefield is not about beating bechmarks, it is about packaging, big-little in x86 space, die stacking, etc.

I totally agree with the "no hate" viewpoint here. We should praise competition and enjoy every great product no matter which company launches it.
But i think Intel deserves at least some of the current backlash cause of the way they treated security in their latest iterations of CPUs (Meltdown, ... the ME flaws ...).
Some people risk everything (want to be a journalist in China/Turkey/North Korea/Poland/Hungary ... at the moment?) for publishing truth and should be able to rely on their tech.

 

[Crackong]

Same price as the ARM version.

That doesn't mean it isn't i7 prices
They are both expensive.

 

[londiste]

That doesn't mean it isn't i7 prices
They are both expensive.

As far as Lakefield goes, $281:

Intel® Core™ i5-L16G7 Processor (4M Cache, up to 3.0GHz) Product Specifications

Intel® Core™ i5-L16G7 Processor (4M Cache, up to 3.0GHz) quick reference guide including specifications, features, pricing, compatibility, design documentation, ordering codes, spec codes and more.

ark.intel.com

Cheapest current i7 is 10610U at $409:

Intel® Core™ i7-10610U Processor (8M Cache, up to 4.90 GHz) Product Specifications

Intel® Core™ i7-10610U Processor (8M Cache, up to 4.90 GHz) quick reference guide including specifications, features, pricing, compatibility, design documentation, ordering codes, spec codes and more.

ark.intel.com

 

[Crackong]

As far as Lakefield goes, $281:

Intel® Core™ i5-L16G7 Processor (4M Cache, up to 3.0GHz) Product Specifications

Intel® Core™ i5-L16G7 Processor (4M Cache, up to 3.0GHz) quick reference guide including specifications, features, pricing, compatibility, design documentation, ordering codes, spec codes and more.

ark.intel.com

Cheapest current i7 is 10610U at $409:

Intel® Core™ i7-10610U Processor (8M Cache, up to 4.90 GHz) Product Specifications

Intel® Core™ i7-10610U Processor (8M Cache, up to 4.90 GHz) quick reference guide including specifications, features, pricing, compatibility, design documentation, ordering codes, spec codes and more.

ark.intel.com

And we all know the RCP on mobile chips are useless.
The price depends on the model not the CPU.

Niche product have niche prices.

 

[londiste]

RCP is normally useless in a way that OEM gets a much better deal