Intel Will no Longer Disclose Multi-Core Turbo Boost Frequencies

[jigger]

Its already is to a degree if you don't have a K series chip and a non Z series motherboard. A number like mine will allow all cores Turbo but not increasing the TDP limit!

Probably best to not pull the plaster off too fast. But yeah it's a bit sad and a little surprising we've hit the wall this early. I though Intel would have been able keep scaling for a while yet.

Come back 32nm all is forgiven.

 

[pastymuncher]

Hopefully review sites will still keep telling us the speeds for 2/3/4 cores etc. That's how I have found the max speed for a certain number of cores for some time now. Plenty of people who are not "in the know" are going to be caught out by this though.

 

[jigger]

Hopefully review sites will still keep telling us the speeds for 2/3/4 cores etc. That's how I have found the max speed for a certain number of cores for some time now. Plenty of people who are not "in the know" are going to be caught out by this though.

Review binned chips...

 

[smilingcrow]

@jigger Depends on what you mean by best, they most likely use the chips with low leakage characteristics to make their mobile SKUs, and the high leakage ones go to the K parts since those will reach higher frequencies.

Intel have used multiple processes at the same node size which are optimised for different qualities for years now.
So the high power desktop chips will be using a different variant so will come from different wafers.
So I doubt that chips from a wafer for desktop chips will end up being used for a laptop chip.
There may be a cross-over point as they at least used to have some high end laptop parts rated for 55W or so but that's still a long way below the desktop parts that can hit 120W TDP or more if including HEDT.

 

[Emlyn_Dewar]

Will they disclose what brand of toothpaste they use between the die and the IHS?

Love this.

 

[DragonQ]

The thing is, as soon as any new chip hits reviewers they will test what the all-core turbo is. Maybe it won't be as consistent chip-to-chip but I think it'll be pretty fixed given a custom cooler and a sufficient PSU. The numbers will be available, just not directly from Intel because they want to tout the higher number.

 

[Panos]

Hopefully review sites will still keep telling us the speeds for 2/3/4 cores etc. That's how I have found the max speed for a certain number of cores for some time now. Plenty of people who are not "in the know" are going to be caught out by this though.

Depends the site/channel.
US ones receive CPUs directly from Intel, so they could get what ever golden binned chip Intel decides, with testing instructions.
The rest eg European sites, are getting off the main market a CPU from some supplier. (thats Intel policy now).

You can see the discrepancy of the benchmarks right now between European and American channels/sites and the 8700K.
Also you can see the discrepancy on the benchmarks between some expensive Asus boards who auto overclock the CPU, and other ones like MSI or Aorus. Check the last video from Jay2Cents.... It seems the benchmarks on some reviewers like Gamers Nexus, Linus, Jay2Cents are based on the "stock clocks" which are actually auto overclocked clocks from the top of the range Asus board, making the CPU looking better.
An Aorus board, or cheaper Asus boards, without that that "auto overclock" feature, show significantly less performance on "stock clocks".

So all in light of clear Intel naming policy of course......

But all these are conspiracy theories for some in this very forum....

 

[subbytna]

How do you feel about nvidia boost 3.0?

You have a point I guess. It's similar but most graphics cards aren't advertised as running the full boost speeds, like Intel are now doing with their marketing of CPUs. Giving the implied top speed as the norm.

It's nothing to do with me, I know the difference and know that what they are claiming it's just the 1 core boost. I know how to get all cores performing the same speeds (overclocking) It's the normal person on the street, being hit by big numbers to make it sound more appealing than it really is, that I have a bit of a gripe with. Sneaky marketing, saying a truth, but making it out to be a lot more substantial than it really is.

It's like selling the FX8150 as a 8.4GHz cpu....just because 1 core hit 8.4 (albeit under liquid nitrogen cooling) it's not a true reflection of what the chip is capable of in day to day use.

As @DragonQ said, it's Intel touting the higher number to make it sound better.

 

[AndreiD]

@Panos All the big reviewers receive samples directly from the vendors, I don't understand how this is something new. Or are you telling me almost every reviewer getting their R7 1700 to 4~4.1Ghz on <1.425v is normal?

@subbytna AMD advertises their CPUs the same, max turbo not theoretical peak frequency under LN2, what are you even on about?

Are people in this thread really just trying to find anything they can to hate on Intel? If you really want to dissect hardware marketing you'll find a lot of questionable practices, even from your favorite company.

 

[jigger]

Intel have used multiple processes at the same node size which are optimised for different qualities for years now.
So the high power desktop chips will be using a different variant so will come from different wafers.
So I doubt that chips from a wafer for desktop chips will end up being used for a laptop chip.
There may be a cross-over point as they at least used to have some high end laptop parts rated for 55W or so but that's still a long way below the desktop parts that can hit 120W TDP or more if including HEDT.

No, we've been using salvaged laptop chips for a while now. Intel have been binning them accordingly. Mobile>Xeon>Desktop.

 

[AndreiD]

Oh, and this one is for @subbytna so which one is misleading marketing?

 

[jigger]

Oh, and this one is for @subbytna so which one is misleading marketing?

AMD are the saviour of the enthusiast universe and massive underdogs in this market. I think we can cut them a little slack.

 

[DragonQ]

Oh, and this one is for @subbytna so which one is misleading marketing?

Agreed, that is dumb, but that's just how OcUK is advertising them. I doubt that has anything do with AMD or Intel, since each site advertises their products differently. The website rhyming with "scam" doesn't list frequencies in their product names at all, for example (at least for these two examples that I've checked).

AMD's website doesn't even list a "headline" clock speed for the R7 1700, but on the details bit it lists both base and max turbo clocks:


Intel list the same two numbers, but also state a headline figure of "up to 4.7 GHz":

 

[jigger]

Agreed, that is dumb, but that's just how OcUK is advertising them. I doubt that has anything do with AMD or Intel, since each site advertises their products differently. The website rhyming with "scam" doesn't list frequencies in their product names at all, for example (at least for these two examples that I've checked).

AMD's website doesn't even list a "headline" clock speed for the R7 1700, but on the details bit it lists both base and max turbo clocks:


Intel list the same two numbers, but also state a headline figure of "up to 4.7 GHz":

Up to 4.7Ghz turbo. Terms and conditions apply. Products not suitable for vegans. May contain nuts and cause prolonged nose bleeds.

 

[AndreiD]

@DragonQ We're talking about retailers here, AMD almost always advertises the Turbo speed in the title of their products while Intel has the base clock, check other retailers.
It doesn't even matter much, but this is mainly a shot at the whole "boo Intel is misleading" crowd in this thread.

The whole point of argument is stupid to begin with, Turbo offers some tangible performance gains and isn't just for marketing numbers, regardless of vendor.
And this whole piece of news is taken way out of proportion by a few who don't understand how Turbo works and just wanted something to bash Intel for.

 

[jigger]

@DragonQ We're talking about retailers here, AMD almost always advertises the Turbo speed in the title of their products while Intel has the base clock, check other retailers.
It doesn't even matter much, but this is mainly a shot at the whole "boo Intel is misleading" crowd in this thread.

The whole point of argument is stupid to begin with, Turbo offers some tangible performance gains and isn't just for marketing numbers, regardless of vendor.
And this whole piece of news is taken way out of proportion by a few who don't understand how Turbo works and just wanted something to bash Intel for.

You need to check yourself a little. The situation is what it is, but you can't make the anti AMD/pro Nvidia Intel argument and play the enthusiast card at the same time. Pick one.

You don't want to see AMD do well and that's fine. However the alternative is very bleak for the enthusiast in a market that has no enthusiast interests. You'll be gaming on a dual core laptop with integrated Intel graphics. Or a Nintendo Switch.

 

[gavinh87]

I don't mind game on a dual core if it's faster than an 8 core.....

Looking at you destiny 2.

 

[jigger]

I don't mind game on a dual core if it's faster than an 8 core.....

Looking at you destiny 2.

Great then buy one. Enjoy.

 

[VKleita]

Up to 4.7Ghz turbo. Terms and conditions apply. Products not suitable for vegans. May contain nuts and cause prolonged nose bleeds.

Dammit, spat my coffee all over my desk ....

 

[subbytna]

Oh, and this one is for @subbytna so which one is misleading marketing?

That's OCUK advertising.

I'm running an INTEL CPU and have done so since Core2Quads.