Con Lake Con firmed [Warning: AdoredTV]

[gavinh87]

You still haven't watched the bloody video, if you had you would know Adored even said exactly that himself about the 8400, he isn't talking about the 8400 in the sense that its misleading, he is talking about the 8700/K and its misleading advertised clock rates and TDP, whether or not that's the fault of reviewers or Intel Adored is shining a light on the fact that depending on which reviewer you go to and what hardware they are using dictates how much over 'or in some cases not' the chip is performing above Intel's ratings.

The fact is different reviewers are getting vastly different performance results for the same chip, for example between 1200 points in Cinebench to 1500 points with other reviewers for the same chips, Adored is simply shedding light on that and explaining why that's happening.

Theres a deeper message in that video and you know it. The title says it all.
As for the binning comments from people, its bs. A few people here myself included have retail samples doing 5ghz+

 

[Jono8]

You still haven't watched the bloody video, if you had you would know Adored even said exactly that himself about the 8400, he isn't talking about the 8400 in the sense that its misleading, he is talking about the 8700/K and its misleading advertised clock rates and TDP, whether or not that's the fault of reviewers or Intel Adored is shining a light on the fact that depending on which reviewer you go to and what hardware they are using dictates how much over 'or in some cases not' the chip is performing above Intel's ratings.

The fact is different reviewers are getting vastly different performance results for the same chip, for example between 1200 points in Cinebench to 1500 points with other reviewers for the same chips, Adored is simply shedding light on that and explaining why that's happening.

My 8700 in an equally, nearly as cheap board (gigabyte gaming 3) does exactly the same and did so on the stock cooler (ie boosts to 4.3ghz on all cores and up to 4.6 on mixed core loads). I only changed from the stock cooler because it was noisy. Saying you need "very high end Motherboards and coolers" is just hyperbole and wrong.

That Adored video is not "simply shedding light on that and explaining why that's happening". He has entitled it "Con lake" implying that coffee lake is a con when it isnt. Intel have not conned anyone or lied about anything with regard to how the boost clocks work on Coffee-lake. If he was shedding light on why it is happening the video should just be called "Intel boost clocks explained"

 

[humbug]

Oh i'm sorry, of course if your CPU is running at 4.3Ghz on a stock 65 watt intel cooler then yes Adored must be completely wrong, just as you say.

My god how could we all have been so stupid, of course the 8700K is not a 95 watt CPU. its a 65 watt CPU even with a huge overclock, your right......

So those reviewers using less than top end components not getting anything like 4.3Ghz.... what are they doing wrong, they must be so useless.

Spoiler: Assume people are stupid

But no one is that stupid.

 

[humbug]

Oh i'm sorry, of course if your CPU is running at 4.3Ghz on a stock 65 watt intel cooler then yes Adored must be completely wrong, just as you say.

My god how could we all have been so stupid, of course the 8700K is not a 95 watt CPU. its a 65 watt CPU even with a huge overclock, your right......

So those reviewers using less than top end components not getting anything like 4.3Ghz.... what are they doing wrong, they must be so useless.

Spoiler: Assume people are stupid

But no one is that stupid.

Why do you think the 8700K does not come with Intel's stock cooler? because its not enough to cool it.
So, No.

 

[Jono8]

Oh i'm sorry, of course if your CPU is running at 4.3Ghz on its stock 65 watt cooler then yes Adored must be completely wrong, just as you say.

I have not said his findings are wrong, just that his findings demonstrate how the cpu's actually work and that it is not a "con".

My quarrel is with the title and sentiment in the video and in the OP, and then the resulting hyperbole about how Intel is conning everyone and how you need "very high end" components to have them run at their boost clocks. It is misleading .

 

[gavinh87]

Why do you think the 8700K does not come with Intel's stock cooler? because its not enough to cool it.
So, No.

Meanwhile, the 1800x and threadripper stock coolers are performing just fine, oh wait.
The guy does have some good points but the way he goes about his videos is full on baiting. There is no "con" going on here, just an angry scotsman with an agenda to slate anything that isn't AMD.
He's a charlatan, talking about his prophecies lol.

 

[Jono8]

On its own yes you are correct, i would assume they know at least a little of what they are doing.
Problem is when you get folk googling these components for an idea of what to buy from there local box shifter..
I mean just look at Linus's review, if your a noob looking for information on a processor for a new box your thinking of buying then.. well the difference between what Linus says you can get and what you might actually (probably) get is ... well sizable.

This chart is shocking in how much deviance there is between each end for just one cpu (ok i know its an 8700k but still)

multiple image

Also, just to point out

http://www.guru3d.com/articles_pages/intel_core_i7_8700k_processor_review,7.html

Guru 3d scored it at 1402, not 1296.

"Note: After the initial 8700k review there have been a number of BIOS updates. We list this proc slightly higher at 1402 points, which we consider to be the final number at standard clocks and standard Intel turbo bins. This score is fairly similar accross most motherboards."

Also I assume something went wrong with expreview's run in their main review as their preview review has it at 1443 (perhaps they had an early bios issue also like Guru)

https://videocardz.com/72915/first-review-of-intel-core-i7-8700k-leaks-out

http://www.expreview.com/57166-2.html

When reading all the reviews for the 8700k, they basically all scored in the low 1400's or high 1300's give or take, without MCE on and in the low 1500's with MCE on.

I fully expected therefore around 1400 on my 8700(non k) and my last cinebench run was 1408.

People seem to be freaking out about some outlying anomalies across benches and thinking it is all a con though.....

 

[AndreiD]

The lack of multi-core boost specs is the grey area I'm referring to. It's grey because Intel now claims this is proprietary information that they choose to no longer disclose. The suspicion is that this'll harm consumers in some way, and there seems to be some validity in it, as the variation in MT CB scores imply. At the moment this type of workload is rare for consumers, so it's fortunate that they see little harm in practice. As a development though it's not necessary, it's not beyond AMD to do similar (or worse) and I don't welcome it.

Both companies have their ups and downs when it comes to what info they disclose to the public, you'd be hard pressed to find something as detailed as Intel's 8th gen datasheet on AMD's website.
The issue with giving an all core clock number would be which workload do they choose? Cinebench? The all core clock you will get in Cinebench isn't going to be the same as the all core clock you will get in something with AVX support like H264 encoding for example.
And the difference in MT CB scores is due to some outlets either using different motherboards or having MCE enabled. Intel allows motherboard makers to override TDP limits via firmware, which is why we saw Skylake-X reaching some astronomical power consumption figures in AVX loads. Some motherboards will enforce the TDP limits, others won't, others will boost the chip on all cores to the max turbo clock (MCE), that's the main variation when it comes to Intel CPUs, motherboards & their firmware.
Similar is starting to happen to AMD too, motherboard makers will include little BIOS bits like "Performance Bias" (my ASUS X370 Prime has it) that's somewhat similar to MCE and can explain some variation seen in the Ryzen CPU reviews too.

Also the narrative that Intel is somehow selling lower quality silicon doesn't make any sense, Coffee Lake is 14nm++, a mature process and Intel's yields are probably >90% and not to mention they already bin the dies quite heavily, whatever doesn't make the cut for 8700K ends up being cut down to 8700, 8600K, 8400 (and they probably bin the low leakage silicon for future mobile chips too).

 

[TrixP10]

So where is the ryzen con video then? Defend him all you like, he is a charlatan.

Exactly, where is the 'ryzen con'

Looking for an anti AMD charlatan, any ideas

 

[TrixP10]

Theres a deeper message in that video and you know it. The title says it all.
As for the binning comments from people, its bs. A few people here myself included have retail samples doing 5ghz+

Theres a deeper message in that video and you know it - yes, and it appears that you don't like it because according to you people talk bs or it is a criticism of your favorite brand.

 

[AmateurExpert]

If he was shedding light on why it is happening the video should just be called "Intel boost clocks explained"

Haha you know almost all YouTubers are into clickbait headlines - we know sensationalism is used to elicit engagement, but the points he brought up still merit discussion. I'd like to see more data on how much headroom Intel & AMD chips actually have compared to their published TDPs when operating full load, non-overclocked. There's quite a lot of variance between models, but not seen anyone tackle this in detail. Any ideas?

 

[AmateurExpert]

The issue with giving an all core clock number would be which workload do they choose? Cinebench? The all core clock you will get in Cinebench isn't going to be the same as the all core clock you will get in something with AVX support like H264 encoding for example.

While it might not be straightforward, it's not intractable. But if that's why Intel have stopped publishing the all-core boost figures, why didn't they say so? Seems reasonable enough.

that's the main variation when it comes to Intel CPUs, motherboards & their firmware.

As large a variation is also now seen with the 8700 in cheap retail PCs vs what the 8700 does when allowed to exceed published TDP. Has it really always been this severe?

Similar is starting to happen to AMD too

Which is not good news for the consumer and also mentioned in the video.

 

[gavinh87]

Theres a deeper message in that video and you know it - yes, and it appears that you don't like it because according to you people talk bs or it is a criticism of your favorite brand.

Take a look in the other threads, I have criticised intel too.
I just don't like the way this guy come across, and its all for publicity. Remove his idealistic views from his videos and his channel would be boring.

 

[humbug]

Take a look in the other threads, I have criticised intel too.
I just don't like the way this guy come across, and its all for publicity. Remove his idealistic views from his videos and his channel would be boring.

Where? this i have to see.

 

[AndreiD]

While it might not be straightforward, it's not intractable. But if that's why Intel have stopped publishing the all-core boost figures, why didn't they say so? Seems reasonable enough.

They did in their press release, they said clocks are opportunistic: "All Turbo frequencies are opportunistic given their dependency on system configuration and workloads."
Your best bet now is to tune into reviews and just look at performance in relevant tasks, which is basically what most informed people did before.

As large a variation is also now seen with the 8700 in cheap retail PCs vs what the 8700 does when allowed to exceed published TDP. Has it really always been this severe?

Well technically the 8700 should never exceed TDP, but I think the Z370 mobos can allow that? Not 100% sure.
And I think that Aldi pre-built either had the 8700 gimped because of the poor cooling or the motherboard had the CPU on a lower TDP limit than the stock 65W, could be any number of issues, reason why building your own is almost always better than getting some random pre-built.

Adored could have made a video where he explains how EIST and Turbo Boost work on Intel chips instead of doing this click bait, the end result would be that his audience would be more informed. If he wanted something to bark at Intel there's plenty with the Meltdown/Spectre/AMT security issues.

 

[gavinh87]

Where? this i have to see.

X299 for a start

 

[humbug]

X299 for a start

Where?

 

[gavinh87]

Where?

Search for it.

 

[AmateurExpert]

clocks are opportunistic

I read that as an explanation of the Turbo Boost 2.0 clocks (which have been opportunistic for a while, not just with 8th gen - TB2.0 has been around since 2nd gen Core, though I don't know if the mechanism has changed over the generations) and not the reasoning for changing stance on all-core boost data being proprietary.

Well technically the 8700 should never exceed TDP

Not according to Intel:
"Intel® Turbo Boost Technology 2.0 allows the processor to operate at a power level that is higher than its TDP configuration and data sheet specified power for short durations to maximize performance"
but actually not that many people really know that.

either had the 8700 gimped because of the poor cooling or the motherboard had the CPU on a lower TDP limit than the stock 65W

According to Intel's published specs the 8700 isn't gimped, it's just running at around spec 65W TDP (a very short boost at PL2, then back to PL1@TDP). The cooling isn't poor - it's exactly the spec that Intel gives (65W): "The processor TDP is the maximum sustained power that should be used for design of the processor thermal solution" (from the 8th gen datasheet). The video suggests that when 8700 "stock" scores are given in the press, they are from operating well above 65W TDP or else they wouldn't be showing CB scores significantly higher than 1210.

 

[David Bisset]

That's where Intel (and reviewers) have been disingenious - if intel gave a e.g. 90w tdp for the same part it would be no issue as with that level of cooling it performs better (like in the reviews) . So arguably reviewers should test default config (cooling to match tdp) and then with better coolers. In this case even slightly better is all that's needed, though I don't know if reviews should pick budget-appropriate coolers or just 'good' #shrug#