Intel Will no Longer Disclose Multi-Core Turbo Boost Frequencies

sadbuttrue · 10 Oct 2017 at 15:55

So does this mean Intel can ship an 8700K only capable of 3.7ghz all-core, so long as it can do 4.7ghz single-core?

AndreiD · 10 Oct 2017 at 15:58

sadbuttrue said:
So does this mean Intel can ship an 8700K only capable of 3.7ghz all-core, so long as it can do 4.7ghz single-core?

They could already do that since technically the all core turbo isn't advertised. Realistically this changes nothing for the regular user.

Martini1991 · 10 Oct 2017 at 16:05

sadbuttrue said:
So does this mean Intel can ship an 8700K only capable of 3.7ghz all-core, so long as it can do 4.7ghz single-core?

Yes.
But it'll be advertised as such.

jigger · 10 Oct 2017 at 16:08

The best chips. Ie the best. We will never know how good they are.

jigger · 10 Oct 2017 at 16:09

sadbuttrue said:
So does this mean Intel can ship an 8700K only capable of 3.7ghz all-core, so long as it can do 4.7ghz single-core?

Pretty much.

gavinh87 · 10 Oct 2017 at 16:17

sadbuttrue said:
So does this mean Intel can ship an 8700K only capable of 3.7ghz all-core, so long as it can do 4.7ghz single-core?

Yeah, but how many K CPU's have you seen that cannot overclock to at least the single core boost speed?

Hagar · 10 Oct 2017 at 16:21

gavinh87 said:
Yeah, but how many K CPU's have you seen that cannot overclock to at least the single core boost speed?

Not many, but some.

It does point to relaxing the yield point to some extent, whether that is true or will be, I do not know.

Also not every user or rig is capable.

jigger · 10 Oct 2017 at 16:25

This is a pretty sad day to be honest. On the bright side we should see faster 30 watt Mac books.

subbytna · 10 Oct 2017 at 16:26

gavinh87 said:
Yeah, but how many K CPU's have you seen that cannot overclock to at least the single core boost speed?

That's relevant to us who are tech heads and know what's what. My point is that to the average joe on the street, he thinks he's getting a monster chip (4.7 speeds) when in fact only one core will hit that and the rest of the chip, at stock won't get even close to that. Almost, to me anyway, it seems like a sales tactic to make it sound better than what it really is.

Martini1991 · 10 Oct 2017 at 16:28

subbytna said:
That's relevant to us who are tech heads and know what's what. My point is that to the average joe on the street, he thinks he's getting a monster chip (4.7 speeds) when in fact only one core will hit that and the rest of the chip, at stock won't get even close to that. Almost, to me anyway, it seems like a sales tactic to make it sound better than what it really is.

But this isn't a new thing. This was Piledrivers advertising too.

AndreiD · 10 Oct 2017 at 16:29

@subbytna that's how Intel's Turbo has worked so far, it's the same for AMD too, you'll never get that max turbo on all cores. I don't get how this is something new to the "tech heads" on this forum.

jigger · 10 Oct 2017 at 16:32

subbytna said:
That's relevant to us who are tech heads and know what's what. My point is that to the average joe on the street, he thinks he's getting a monster chip (4.7 speeds) when in fact only one core will hit that and the rest of the chip, at stock won't get even close to that. Almost, to me anyway, it seems like a sales tactic to make it sound better than what it really is.

"All Turbo frequencies are opportunistic given their dependency on system configuration and workloads.”

That could be taken a lot of ways.

Martini1991 · 10 Oct 2017 at 16:35

In fact. This is basically going back to older Intel advertising.

It's just the same anti Intel people saying the same crap.

gavinh87 · 10 Oct 2017 at 16:36

subbytna said:
That's relevant to us who are tech heads and know what's what. My point is that to the average joe on the street, he thinks he's getting a monster chip (4.7 speeds) when in fact only one core will hit that and the rest of the chip, at stock won't get even close to that. Almost, to me anyway, it seems like a sales tactic to make it sound better than what it really is.

How do you feel about nvidia boost 3.0?

gavinh87 · 10 Oct 2017 at 16:37

Martini1991 said:
In fact. This is basically going back to older Intel advertising.

It's just the same anti Intel people saying the same crap.

I agree, mountain molehill situation here. Nothing has changed yet people find a way of making it anti consumer.

AndreiD · 10 Oct 2017 at 16:43

jigger said:
"All Turbo frequencies are opportunistic given their dependency on system configuration and workloads.”

That could be taken a lot of ways.

No, it's pretty clear cut given how their Turbo works.
Turbo is opportunistic in that your maximum Turbo frequencies are limited by several factors: power, temperature and TDP limits.

System configuration is related mostly to the temperature limit in case you don't have adequate cooling, while workloads like ones with AVX support (some video encoding for example) will reach the power/TDP limits.

jigger · 10 Oct 2017 at 16:43

This is great for consumers. Low wattage Mac books will be amazballz. 1.5Tflop APU's with 4-6 cores. Don't mind if I do.

CAT-THE-FIFTH · 10 Oct 2017 at 16:56

GPU boosting is worse,as it tends to be non-deterministic in the case of Nvidia at the upper range and for AMD in the lower range AFAIK. At least CPUs have a fixed range.

jigger · 10 Oct 2017 at 17:01

CAT-THE-FIFTH said:
GPU boosting is worse,as it tends to be non-deterministic in the case of Nvidia at the upper range and for AMD in the lower range AFAIK. At least CPUs have a fixed range.

Unless its hard limited to power a envolope. Fingers crossed Intel's next step to move the master race isnt a hard power limit.

CAT-THE-FIFTH · 10 Oct 2017 at 17:13

jigger said:
Unless its hard limited to power a envolope. Fingers crossed Intel's next step to move the master race isnt a hard power limit.

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!