How to: Firmware Update the AMD Radeon RX 5600 XT
MSI Radeon RX 5600 XT Gaming Z review
Sapphire Pulse Radeon RX 5600 XT OC review
ASUS STRIX Radeon RX 5600 XT TOP review
Gigabyte Radeon RX 5600 XT Gaming OC review
Arctic Liquid Freezer II 280 liquid cooler review
Promo: Windows 10 Pro licenses for under 10 USD
Team Group MP33 NVMe 512 GB SSD Review
ADATA SE800 1TB Portable SSD review
Promo: Microsoft Office 2016 for $29.82
Intel Will no Longer Disclose Multi-Core Turbo Boost Frequencies
In yet another unexpected move Intel has made is clear that it will not be sharing any details anymore on the multi-core Turbo clock frequencies of their processors.
You might already have noticed it, Intel is only listing the highest Boost frequency, and not the rest. Here’s the thing, the recent generation processors basically have three main frequencies.
- Base Baseclock
- Binned multi all-core clock turbo
- Single thread turbo
Intel from now on will only list the base and (1) and Single thread (3) turbo. As to why this is, remains uncertain, however many scenarios pop into mind. It might be a legal reason as they cannot guarantee the all core turbo on all processors.However, the longer I think about this, then an old routine kicks in .. what would be the most probable? Might it be that Intel likes that highest Turbo listed on their packaging a bit better for marketing and thus sales? I mean, it’s not unthinkable right? The guys from eteknix have a quote from Intel on this:
“[W]e’re no longer disclosing this level of detail as its proprietary to Intel. Intel only specifies processor frequencies for base and single-core Turbo in our processor marketing and technical collateral, such as ARK, and not the multi-core Turbo frequencies. We’re aligning communications to be consistent. All Turbo frequencies are opportunistic given their dependency on system configuration and workloads.”
So a Core i7 8700 is now being listed as a 4.7 GHz processor (click the link and look at the ARK info). But considering that is just one thread, it really runs 4.3 GHz on all six cores. Weird move huh?
| Processor | Cores / Threads | Base Clock | Turbo 2.0 (6c) | Turbo 3.0 (1c) | L3 | TDP | Price |
|---|---|---|---|---|---|---|---|
| Core i7 8700K | 6/12 | 3.7 GHz | 4.3 GHz | 4.7 GHz | 12 MB | 95 W | $359 / €389 |
| Core i7 8700 | 6/12 | 3.2 GHz | 4.3 GHz | 4.6 GHz | 12 MB | 65 W | $303 / €327 |
| Core i5 8600K | 6/6 | 3.6 GHz | 4.1 GHz | 4.3 GHz | 9 MB | 95 W | $257 / €273 |
| Core i5 8400 | 6/6 | 2.8 GHz | 3.8 GHz | 4.0 GHz | 9 MB | 65 W | $182 / €192 |
| Core i3 8350K | 4/4 | 4.0 GHz | NA | NA | 8 MB | 91 W | $169 / €189 |
| Core i3 8300 | 4/4 | 4.0 GHz | NA | NA | 8 MB | 65 W | - |
| Core i3 8100 | 4/4 | 3.6 GHz | NA | NA | 6 MB | 65 W | $117 / €123 |
If you look at the above table, pretty much the greyed out Turbo 2.0 info is no longer disclosed by Intel.
Intel Z370 Chipset Could Support Kaby Lake - But Intel Will Not Allow It - 09/22/2017 08:15 PM
Much has been said and spoken about Intel upcoming 6-core Coffee Lake processors. It will launch based on a LGA 1151 Socket. We’ve been able to conform that Z370 will only support Coffee La...
Intel Will Add Wi-Fi and USB 3.1 support into next-gen chipsets - 11/10/2016 06:34 PM
Intel reportedly is planning to add USB 3.1 and Wi-Fi functionality housed directly inside their motherboard chipsets, likely starting at 300-series scheduledwhich would be released at the end of 2017...
Intel Will Launch Two LGA1150 Broadwell CPUs with 65W TDP - 03/24/2015 11:52 AM
In the second half of 2015, Intel is to launch two LGA1150 Broadwell CPUs (Intel Core I7-5775C & Core I5-5675C) based on socket 1150, both have a 65W TDP. In tital 5 models will be launched but th...
2014 was a good year for Intel with 56 Billion USD Revenue - 01/16/2015 08:09 AM
Intel issued a document sharing its fourth-quarter financial results, they had a record year. Full year revenue increased 6 percent to $55.9 billion and net profit came in at $11.7 billion. Intel sa...
Intel will pay $1.5 billion in Nvidia settlement - 01/15/2014 09:33 AM
Intel and Nvidia are burying the hatchet and put away their differences and ongoing lawsuit and have settled for a $1.5 billion, six year cross-licensing agreement between the rival chipmakers. The...
Vananovion
Member
Posts: 89
Joined: 2017-08-31
Member
Posts: 89
Joined: 2017-08-31
#5480154 Posted on: 10/10/2017 10:25 AM
I wouldn't be so sure. I think this could fall under the Unfair Commercial Practices directive, Article 7, paragraphs 1 and 2 (https://webgate.ec.europa.eu/ucp/public/index.cfm?event=public.directive.browse2&elemID=227#article-209). Looked at some of the cases and there have been fines for stuff like "disclaimer unintelligible to the average consumer" and "disclosing information on promotional medium and not the product itself" etc. Guess it all comes down to how Intel covers its ass, but even hordes of lawyers can make a mistake or an ommission and open up way for litigation. Also a court may still take an issue due to different interpretation. It's not unprecedented.
In any case, I think we can all agree that this move is only meant to mislead the average customer, no matter how legally covered Intel is.
Intel has an army of lawyers like all big tech companies with lots of patents. I'm sure they can fill the product packaging with suitable disclaimers and inform retailers to tell customers the boost frequency only applies to old single core applications and in modern multicore cases it's something totally different. After that it's the retailer burning if somebody is disappointed.
I wouldn't be so sure. I think this could fall under the Unfair Commercial Practices directive, Article 7, paragraphs 1 and 2 (https://webgate.ec.europa.eu/ucp/public/index.cfm?event=public.directive.browse2&elemID=227#article-209). Looked at some of the cases and there have been fines for stuff like "disclaimer unintelligible to the average consumer" and "disclosing information on promotional medium and not the product itself" etc. Guess it all comes down to how Intel covers its ass, but even hordes of lawyers can make a mistake or an ommission and open up way for litigation. Also a court may still take an issue due to different interpretation. It's not unprecedented.
In any case, I think we can all agree that this move is only meant to mislead the average customer, no matter how legally covered Intel is.
BLEH!
Senior Member
Posts: 5918
Joined: 2010-10-17
Senior Member
Posts: 5918
Joined: 2010-10-17
#5480155 Posted on: 10/10/2017 10:26 AM
The hell are they playing at?!
The hell are they playing at?!
BlueRay
Senior Member
Posts: 270
Joined: 2015-11-18
Senior Member
Posts: 270
Joined: 2015-11-18
#5480166 Posted on: 10/10/2017 10:52 AM
Ugh another reason to buy a K model and just set the frequency you want.
From the other side WTF Intel really? What are they trying to hide? That they can't hit big clocks on all cores for more than 1-2 secs or something?
Ugh another reason to buy a K model and just set the frequency you want.
From the other side WTF Intel really? What are they trying to hide? That they can't hit big clocks on all cores for more than 1-2 secs or something?
FlinchingNinja
Senior Member
Posts: 126
Joined: 2008-02-13
Senior Member
Posts: 126
Joined: 2008-02-13
#5480167 Posted on: 10/10/2017 10:53 AM
It's to increase yields. They just removed a test case.
It's to increase yields. They just removed a test case.
nz3777
Senior Member
Posts: 2383
Joined: 2014-01-21
Senior Member
Posts: 2383
Joined: 2014-01-21
#5480168 Posted on: 10/10/2017 11:00 AM
What is with them Lately seriously?
What is with them Lately seriously?
Solfaur
Senior Member
Posts: 7207
Joined: 2005-08-10
Senior Member
Posts: 7207
Joined: 2005-08-10
#5480172 Posted on: 10/10/2017 11:11 AM
So in a day and age where core number matters (so much that even Intel decided to catch up) they now only show the max clock from ONE core and brand the whole CPU as a higher frequency, clearly misleading on purpose? Sure, this is not an issue for enthusiasts, but you can bet a lot of "casual" buyers will fall for it.
So in a day and age where core number matters (so much that even Intel decided to catch up) they now only show the max clock from ONE core and brand the whole CPU as a higher frequency, clearly misleading on purpose? Sure, this is not an issue for enthusiasts, but you can bet a lot of "casual" buyers will fall for it.
nevcairiel
Senior Member
Posts: 649
Joined: 2015-05-19
Senior Member
Posts: 649
Joined: 2015-05-19
#5480175 Posted on: 10/10/2017 11:16 AM
I wouldn't be so sure. I think this could fall under the Unfair Commercial Practices directive, Article 7, paragraphs 1 and 2 (https://webgate.ec.europa.eu/ucp/public/index.cfm?event=public.directive.browse2&elemID=227#article-209). Looked at some of the cases and there have been fines for stuff like "disclaimer unintelligible to the average consumer" and "disclosing information on promotional medium and not the product itself" etc. Guess it all comes down to how Intel covers its ass, but even hordes of lawyers can make a mistake or an ommission and open up way for litigation. Also a court may still take an issue due to different interpretation. It's not unprecedented.
The boost clock was never a guarantee anyway. Otherwise, why would they bother with having a base and a boost clock? The only guaranteed clock is the base clock. Boost has always relied on the mainboard providing enough power, the cooling being adequate, and a range of other conditions - at least under Intel-specified conditions. Most mainboards come with default overclocking these days anyway, which increases the boost out of the box ("Multi-core Enhancement", which runs all cores on the single-core turbo, if thermals and power permit)
I don't care what Intel advertises their CPUs with one bit. I care about their actual performance. And that you get from reviews, not a spec sheet. Not to mention that you can overclock Intel CPUs generally quite well if you use a better cooler then their boxed cooler, at the very least defining your own all-core multiplier quite a bit above the Intel spec anyway.
I wouldn't be so sure. I think this could fall under the Unfair Commercial Practices directive, Article 7, paragraphs 1 and 2 (https://webgate.ec.europa.eu/ucp/public/index.cfm?event=public.directive.browse2&elemID=227#article-209). Looked at some of the cases and there have been fines for stuff like "disclaimer unintelligible to the average consumer" and "disclosing information on promotional medium and not the product itself" etc. Guess it all comes down to how Intel covers its ass, but even hordes of lawyers can make a mistake or an ommission and open up way for litigation. Also a court may still take an issue due to different interpretation. It's not unprecedented.
The boost clock was never a guarantee anyway. Otherwise, why would they bother with having a base and a boost clock? The only guaranteed clock is the base clock. Boost has always relied on the mainboard providing enough power, the cooling being adequate, and a range of other conditions - at least under Intel-specified conditions. Most mainboards come with default overclocking these days anyway, which increases the boost out of the box ("Multi-core Enhancement", which runs all cores on the single-core turbo, if thermals and power permit)
I don't care what Intel advertises their CPUs with one bit. I care about their actual performance. And that you get from reviews, not a spec sheet. Not to mention that you can overclock Intel CPUs generally quite well if you use a better cooler then their boxed cooler, at the very least defining your own all-core multiplier quite a bit above the Intel spec anyway.
Yttersta
Member
Posts: 23
Joined: 2012-05-23
Member
Posts: 23
Joined: 2012-05-23
#5480179 Posted on: 10/10/2017 11:40 AM
What an arse move by them!
Us knowledgeable and the enthusiasts alike will learn these details through reviews and websites anyway, but what of the mainstream who also do not overclock, what about the enterprises?
Those guys already have no idea about the importance of the architecture above the clock speed numbers, and they'll be just fall into this marketing BS.
Here is one hoping that dishonesty and tricks in marketing crap will not be met with positive outcomes in their favours... Gosh I hate corporations like this.
What an arse move by them!
Us knowledgeable and the enthusiasts alike will learn these details through reviews and websites anyway, but what of the mainstream who also do not overclock, what about the enterprises?
Those guys already have no idea about the importance of the architecture above the clock speed numbers, and they'll be just fall into this marketing BS.
Here is one hoping that dishonesty and tricks in marketing crap will not be met with positive outcomes in their favours... Gosh I hate corporations like this.
moeppel
Senior Member
Posts: 149
Joined: 2015-06-30
Senior Member
Posts: 149
Joined: 2015-06-30
#5480182 Posted on: 10/10/2017 12:09 PM
I mean this is just a very odd and strange move... it isn't like their all core turbo is even bad? so i don't get why they would do this... not to mention tech sites who review these products can just find out what they all boost up to anyway...
It is quite simple.
By stating all core boost to be 4.3 ghz they make a legal commitment. With Turbo 2.0 stating 4.3 every of their 8700(K) has to hit 4.3 ghz stable on all cores. Every 8700(K) that only hits 3.8~4.2ghz all core boost becomes a defective, non-sellable unit - or garbage, so to speak.
Ultimately Intel likely suffers from bad yields while AMD can crank it up quite easily in contrast.
Intel needs to back to the drawing board and rather quickly at that.
I mean this is just a very odd and strange move... it isn't like their all core turbo is even bad? so i don't get why they would do this... not to mention tech sites who review these products can just find out what they all boost up to anyway...
It is quite simple.
By stating all core boost to be 4.3 ghz they make a legal commitment. With Turbo 2.0 stating 4.3 every of their 8700(K) has to hit 4.3 ghz stable on all cores. Every 8700(K) that only hits 3.8~4.2ghz all core boost becomes a defective, non-sellable unit - or garbage, so to speak.
Ultimately Intel likely suffers from bad yields while AMD can crank it up quite easily in contrast.
Intel needs to back to the drawing board and rather quickly at that.
Denial
Senior Member
Posts: 12496
Joined: 2004-05-16
Senior Member
Posts: 12496
Joined: 2004-05-16
#5480183 Posted on: 10/10/2017 12:16 PM
Is this really any different than what Nvidia/AMD do with their GPUs? Nvidia guarantees a base frequency and they give you a "typical clock" frequency, but the actual boosted numbers are dynamic depending on TDP in the given application, they don't guarantee a boosted number or even state it. Like with a 980Ti, depending on your ASIC quality, there could be swings up to 10% in performance compared to someone with the exact same chip - because a lower ASIC card would hit TDP at lower frequencies.
Is this really any different than what Nvidia/AMD do with their GPUs? Nvidia guarantees a base frequency and they give you a "typical clock" frequency, but the actual boosted numbers are dynamic depending on TDP in the given application, they don't guarantee a boosted number or even state it. Like with a 980Ti, depending on your ASIC quality, there could be swings up to 10% in performance compared to someone with the exact same chip - because a lower ASIC card would hit TDP at lower frequencies.
H83
Senior Member
Posts: 2804
Joined: 2009-09-08
Senior Member
Posts: 2804
Joined: 2009-09-08
#5480184 Posted on: 10/10/2017 12:16 PM
It is quite simple.
By stating all core boost to be 4.3 ghz they make a legal commitment. With Turbo 2.0 stating 4.3 every of their 8700(K) has to hit 4.3 ghz stable on all cores. Every 8700(K) that only hits 3.8~4.2ghz all core boost becomes a defective, non-sellable unit - or garbage, so to speak.
Ultimately Intel likely suffers from bad yields while AMD can crank it up quite easily in contrast.
Intel needs to back to the drawing board and rather quickly at that.
I could be wrong but i don´t think that´s the case here. I think more than 90% of 8700K can reach the 4,3ghz on all cores and the ones that can´t, can be sold as 8600K.
For me this is just a silly marketing ploy because big numbers sell!!! So for a normal person, a CPU that reaches 4,7Ghz is much better than one that reaches 4,3Ghz. Of course those speeds are under different conditions but the truth is that most people have no idea of what single core or multi core speeds mean anyway so Intel is going to take advantage of their ignorance.
For me this is just a perfect example of sleazy marketing.
It is quite simple.
By stating all core boost to be 4.3 ghz they make a legal commitment. With Turbo 2.0 stating 4.3 every of their 8700(K) has to hit 4.3 ghz stable on all cores. Every 8700(K) that only hits 3.8~4.2ghz all core boost becomes a defective, non-sellable unit - or garbage, so to speak.
Ultimately Intel likely suffers from bad yields while AMD can crank it up quite easily in contrast.
Intel needs to back to the drawing board and rather quickly at that.
I could be wrong but i don´t think that´s the case here. I think more than 90% of 8700K can reach the 4,3ghz on all cores and the ones that can´t, can be sold as 8600K.
For me this is just a silly marketing ploy because big numbers sell!!! So for a normal person, a CPU that reaches 4,7Ghz is much better than one that reaches 4,3Ghz. Of course those speeds are under different conditions but the truth is that most people have no idea of what single core or multi core speeds mean anyway so Intel is going to take advantage of their ignorance.
For me this is just a perfect example of sleazy marketing.
moeppel
Senior Member
Posts: 149
Joined: 2015-06-30
Senior Member
Posts: 149
Joined: 2015-06-30
#5480189 Posted on: 10/10/2017 12:35 PM
I could be wrong but i don´t think that´s the case here. I think more than 90% of 8700K can reach the 4,3ghz on all cores and the ones that can´t, can be sold as 8600K.
For me this is just a silly marketing ploy because big numbers sell!!! So for a normal person, a CPU that reaches 4,7Ghz is much better than one that reaches 4,3Ghz. Of course those speeds are under different conditions but the truth is that most people have no idea of what single core or multi core speeds mean anyway so Intel is going to take advantage of their ignorance.
For me this is just a perfect example of sleazy marketing.
Going with EU rights/laws I'd consider a 8700K being advertised as "4.7ghz CPU" to be misleading/false advertising unless it specifically stated that the frequency was single core only.
Thus, in terms of marketting I personally cannot see what is to gain here for Intel, unless we're talking about the really oblivious folk
I have however not seen a box, so I don't quite know what's printed on there or what the 8700K gets advertised as as a shelf product.
I could be wrong but i don´t think that´s the case here. I think more than 90% of 8700K can reach the 4,3ghz on all cores and the ones that can´t, can be sold as 8600K.
For me this is just a silly marketing ploy because big numbers sell!!! So for a normal person, a CPU that reaches 4,7Ghz is much better than one that reaches 4,3Ghz. Of course those speeds are under different conditions but the truth is that most people have no idea of what single core or multi core speeds mean anyway so Intel is going to take advantage of their ignorance.
For me this is just a perfect example of sleazy marketing.
Going with EU rights/laws I'd consider a 8700K being advertised as "4.7ghz CPU" to be misleading/false advertising unless it specifically stated that the frequency was single core only.
Thus, in terms of marketting I personally cannot see what is to gain here for Intel, unless we're talking about the really oblivious folk
I have however not seen a box, so I don't quite know what's printed on there or what the 8700K gets advertised as as a shelf product.
Denial
Senior Member
Posts: 12496
Joined: 2004-05-16
Senior Member
Posts: 12496
Joined: 2004-05-16
#5480190 Posted on: 10/10/2017 12:36 PM
I could be wrong but i don´t think that´s the case here. I think more than 90% of 8700K can reach the 4,3ghz on all cores and the ones that can´t, can be sold as 8600K.
For me this is just a silly marketing ploy because big numbers sell!!! So for a normal person, a CPU that reaches 4,7Ghz is much better than one that reaches 4,3Ghz. Of course those speeds are under different conditions but the truth is that most people have no idea of what single core or multi core speeds mean anyway so Intel is going to take advantage of their ignorance.
For me this is just a perfect example of sleazy marketing.
Eh, i think it's more than that.
We're at the point again where we are rubbing up against the frequency wall and there are going to be certain apps where Intel isn't going to be able to boost all it's cores to 4.3 or whatever arbitrary 2.0 boost number they set, given a specified TDP. So you have OEM's building machines, validating the machines at specific TDPs and then if you go and say "well all our processors hit 4.3" and then there are apps where they can't without exceeding it - isn't that just as misleading? So the alternative would be set Turbo 2.0 boost frequencies to 4Ghz or something lower - but then every app where it could go higher, you'd be losing performance.
I think the solution for Intel to smooth things over is to do what Nvidia did and educate people on the issue (Tom Peterson talked about it on PC Per podcast/Anandtech) and perhaps list a "typical boost clock" or whatever Nvidia calls it. Like I remember when GPU Boost came out people were super against it, saying it's misleading, but if you think about it, it actually kind of makes sense to do it that way. There are going to be apps that hit the processors TDP instantly because they are utilizing AVX or whatever and then there are going to be apps that can boost the processor to 4.4/4.5, even if it's only for a couple seconds before dropping, because they aren't hitting TDP limit. And like how do you market that number? Outside of "this processor can boost between 4 and 4.8" or something, which is basically what they are doing anyway.
I could be wrong but i don´t think that´s the case here. I think more than 90% of 8700K can reach the 4,3ghz on all cores and the ones that can´t, can be sold as 8600K.
For me this is just a silly marketing ploy because big numbers sell!!! So for a normal person, a CPU that reaches 4,7Ghz is much better than one that reaches 4,3Ghz. Of course those speeds are under different conditions but the truth is that most people have no idea of what single core or multi core speeds mean anyway so Intel is going to take advantage of their ignorance.
For me this is just a perfect example of sleazy marketing.
Eh, i think it's more than that.
We're at the point again where we are rubbing up against the frequency wall and there are going to be certain apps where Intel isn't going to be able to boost all it's cores to 4.3 or whatever arbitrary 2.0 boost number they set, given a specified TDP. So you have OEM's building machines, validating the machines at specific TDPs and then if you go and say "well all our processors hit 4.3" and then there are apps where they can't without exceeding it - isn't that just as misleading? So the alternative would be set Turbo 2.0 boost frequencies to 4Ghz or something lower - but then every app where it could go higher, you'd be losing performance.
I think the solution for Intel to smooth things over is to do what Nvidia did and educate people on the issue (Tom Peterson talked about it on PC Per podcast/Anandtech) and perhaps list a "typical boost clock" or whatever Nvidia calls it. Like I remember when GPU Boost came out people were super against it, saying it's misleading, but if you think about it, it actually kind of makes sense to do it that way. There are going to be apps that hit the processors TDP instantly because they are utilizing AVX or whatever and then there are going to be apps that can boost the processor to 4.4/4.5, even if it's only for a couple seconds before dropping, because they aren't hitting TDP limit. And like how do you market that number? Outside of "this processor can boost between 4 and 4.8" or something, which is basically what they are doing anyway.
MaCk0y
Senior Member
Posts: 602
Joined: 2015-12-30
Senior Member
Posts: 602
Joined: 2015-12-30
#5480191 Posted on: 10/10/2017 12:42 PM
They do state it's single core frequency on their website. Cleverly hidden within the ? next to Max Turbo Frequency.
They do state it's single core frequency on their website. Cleverly hidden within the ? next to Max Turbo Frequency.
Click here to post a comment for this news story on the message forum.
Senior Member
Posts: 9030
Joined: 2004-05-10
It doesnt matter. Reviews will reveal the actual performance. If Intel doesnt want to lose sales to AMD, they will have to make sure their buyers are not shafted due to unexpected shortcomings in performance. If its to cover inconsistencies in boost clocks, they still have to ensure their products are good enough in the end.