The MBP throttling issue isn't what everybody thinks...

[seon123]

Oh boy, this might get bad in a few months. 

https://www.overclock3d.net/news/systems/apple_fixes_2018_macbook_pro_thermal_throttling_issues_with_firmware_update/1

Quote

Following extensive performance testing under numerous workloads, we've identified that there is a missing digital key in the firmware that impacts the thermal management system and could drive clock speeds down under heavy thermal loads on the new MacBook Pro. A bug fix is included in today's macOS High Sierra 10.13.6 Supplemental Update and is recommended.

We apologize to any customer who has experienced less than optimal performance on their new systems. Customers can expect the new 15-inch MacBook Pro to be up to 70% faster, and the 13-inch MacBook Pro with Touch Bar to be up to 2X faster, as shown in the performance results on our website.

 

[ne0tic]

On 2018-07-23 at 11:14 PM, suicidalfranco said:

now let's see the applelogist try to shift the blame to intel

I don't think they will but I wouldn't be surprised.

[leadeater]

11 hours ago, Stefan Payne said:

Because you admit that the CPU uses more than other 45W TDP CPUs of Intel, you yourself admit that.

Yes, just like Intel explicitly states in their TDP guidance article. Every CPU on the planet can use more power than the stated TDP on the spec sheet, been like that since P-States and C-States came in which I believe was after Pentium 4 i.e Intel Core but may have been slightly later than that.

 

At no point ever has Intel said or even insinuated that two different 45W TDP CPUs have the same power usage potential and capabilities, that comes directly from the misunderstanding of what the TDP spec is for CPUs and the improper use of that incorrect understanding.

 

What you need to do is ignore the 45W TDP and stop trying to use it at all. It's not for you to look at and it's not a metric that manufacturers build coolers exactly too and Intel and AMD both say a cooler designed to exactly that is inadequate.

[Dabombinable]

1 hour ago, leadeater said:

Yes, just like Intel explicitly states in their TDP guidance article. Every CPU on the planet can use more power than the stated TDP on the spec sheet, been like that since P-States and C-States came in which I believe was after Pentium 4 i.e Intel Core but may have been slightly later than that.

 

At no point ever has Intel said or even insinuated that two different 45W TDP CPUs have the same power usage potential and capabilities, that comes directly from the misunderstanding of what the TDP spec is for CPUs and the improper use of that incorrect understanding.

 

What you need to do is ignore the 45W TDP and stop trying to use it at all. It's not for you to look at and it's not a metric that manufacturers build coolers exactly too and Intel and AMD both say a cooler designed to exactly that is inadequate.

Eg. The Intel stock cooler (its really weird that my i5 4440's stock cooler is better than those used with i5 8400).

1 hour ago, valdyrgramr said:

https://www.newegg.com/Product/Product.aspx?Item=N82E16834234745&cm_re=ryzen_7_laptop-_-34-234-745-_-Product Does a 1700 not count?

It's a desktop CPU not a mobile one, so no...

[HammerDrop]

1 hour ago, leadeater said:

Yes, just like Intel explicitly states in their TDP guidance article. Every CPU on the planet can use more power than the stated TDP on the spec sheet, been like that since P-States and C-States came in which I believe was after Pentium 4 i.e Intel Core but may have been slightly later than that.

 

At no point ever has Intel said or even insinuated that two different 45W TDP CPUs have the same power usage potential and capabilities, that comes directly from the misunderstanding of what the TDP spec is for CPUs and the improper use of that incorrect understanding.

 

What you need to do is ignore the 45W TDP and stop trying to use it at all. It's not for you to look at and it's not a metric that manufacturers build coolers exactly too and Intel and AMD both say a cooler designed to exactly that is inadequate.

Not entirely sure I can ignore a design constraint like tdp. You claim that people misunderstand tdp, but do you understand the following statement (now linked multiple times in this thread by some kind folks) 

 

"The processor TDP is the MAXIMUM sustained power that SHOULD BE USED for the DESIGN of the PROCESSOR THERMAL SOLUTION" (note: emphasis mine)

 

Can processors exceed the constraint? Yes (for a bit of time to boost). Can processors with the same tdp perform differently, have different potentials, can cook bacon faster? Yes, yes, and ohhhh god yes!

 

But do these processors need different thermal solutions? According to intel... no

 

the members of this forum are moving the goal posts and blurring the lines. The first line of the intel doc tells you what the tdp is. 

 

Is the tdp the measure of how much power a CPU consumes? No

 

is the tdp a measure of a CPU's potential, capabilities, overclocking heard room or any other property that makes it go vroom vroom? No

 

Is the tdp a design constraint indicating you must be this cool to run this processor according to spec (I.e. At base clock speeds?)..... yes

 

 

 

 

[leadeater]

43 minutes ago, HammerDrop said:

But do these processors need different thermal solutions? According to intel... no

According to Intel yes they do, it's covered in their TDP document as to why and they even say manufacturers must build a cooler good enough to sustain the performance of the CPU at the required frequency. The reason for that is you may want to build a product that uses the same CPU but has different performance profiles i.e a thin laptop and a thicker laptop. The thick laptop could have a larger cooling solution which allows a higher boot and/or longer boost. That decision is up to the manufacturer of the product not Intel, Intel can't tell you how you want your product to perform only give you a list of specifications of what the product can do not what it will do.

 

Even the base clock isn't a strict specification because again if the thermal solution is not good enough the CPU will down clock. Intel could not allow that of course but that would just mean laptops that hard power off from thermal protection like in the old days which is not a better situation.

[HammerDrop]

11 minutes ago, leadeater said:

According to Intel yes they do, it's covered in their TDP document as to why and they even say manufacturers must build a cooler good enough to sustain the performance of the CPU at the required frequency.

Link and source please? And if possible paragraph stating this point. Because all I have to go by is intel's documentation saying design according to tdp. And so long as tdp is the same, then the design used should be the same. The design COULD exceed the tdp, but that's not the recommendation

 

"The processor TDP is the maximum sustained power that should be used for the design of the processor thermal solution"

[leadeater]

1 hour ago, HammerDrop said:

Link and source please? And if possible paragraph stating this point. Because all I have to go by is intel's documentation saying design according to tdp. And so long as tdp is the same, then the design used should be the same. The design COULD exceed the tdp, but that's not the recommendation

 

"The processor TDP is the maximum sustained power that should be used for the design of the processor thermal solution"

All of 5.1, though I was wrong it's not in the TDP documentation but rather in the Thermal Management document which I have to say is a more applicable document when actually designing a product and the cooling solution as it pertains directly to that, not that the TDP documentation is not useful either.

 

Intel also puts a common TDP rating on an entire family of products, they use a test sample and rate it on the maximum SKU then apply down the product family. As explained in the Techquickie on TDP, https://youtu.be/yDWO177BjZY?t=75

 

This is well illustrated by the i3-8100 65W, i5-8400 65W and i5-8600 65W. There is no way these three different products are going to use the same amount of power or require the same amount of cooling but all of them will run with a 65W cooler even if that is ill-advised for the i5-8600 if you want to actually utilize the boost frequency.

 

And it's not a could exceed TDP as you emphasized it's a will exceed TDP, it's guaranteed to on the top end SKU that the TDP was rated off of at base frequency because anything above base frequency (or AVX) will be above TDP.

 

This is why I said TDP is not for the end user to look at, it's not a simple specification or designed for us in the first place and Intel has multiple documents (likely non public ones as well for OEM/ODM use) that cover how to design a thermal solution for their products. Unless you're building a cooler then it's not something you have to worry about much, look at the cooler you want to buy and check what it says it's designed for and as always recommended look at product reviews.

 

[HammerDrop]

27 minutes ago, leadeater said:

All of 5.1, though I was wrong it's not in the TDP documentation but rather in the Thermal Management document which I have to say is a more applicable document when actually designing a product and the cooling solution as it pertains directly to that, not that the TDP documentation is not useful either.

 

Intel also puts a common TDP rating on an entire family of products, they use a test sample and rate it on the maximum SKU then apply down the product family. As explained in the Techquickie on TDP, https://youtu.be/yDWO177BjZY?t=75

 

This is well illustrated by the i3-8100 65W, i5-8400 65W and i5-8600 65W. There is no way these three different products are going to use the same amount of power or require the same amount of cooling but all of them will run with a 65W cooler even if that is ill-advised for the i5-8600 if you want to actually utilize the boost frequency.

 

And it's not a could exceed TDP as you emphasized it's a will exceed TDP, it's guaranteed to on the top end SKU that the TDP was rated off of at base frequency because anything above base frequency (or AVX) will be above TDP.

 

This is why I said TDP is not for the end user to look at, it's not a simple specification or designed for us in the first place and Intel has multiple documents (likely non public ones as well for OEM/ODM use) that cover how to design a thermal solution for their products. Unless you're building a cooler then it's not something you have to worry about much, look at the cooler you want to buy and check what it says it's designed for and as always recommended look at product reviews.

 

I'm sorry man but ..... you're the one confused and placing your own meaning to what tdp is. 

"The processor TDP is the maximum sustained power that should be used for the design of the processor thermal solution"

 

If the tdp for an entire family of processors is set to 65w then, according to the same doc you linked to, they can all run with the same thermal design running "base configuration at a near worst case commercially available workload"

 

and you're right, the end user shouldn't be looking at tdp. But no one here has been arguing about end user. This entire line of argument about tdp is based on trying to blame manufacturers for insufficient cooling designs for their laptops and claiming that these manufacturers are ignorant and blatantly disregarded intel's specifications for the sake of design. I'm arguing that tdp is the ONLY constraint provided by intel (as supported by the thermal doc). And is the ONLY design variable that intel claims needs to be taken into account when designing a thermal solution for their chips. If the MacBook Pro, dell xps, and other thin and light laptops cannot run the i9 at its base clock, but can run an i7 at its base clock (assuming they haven't been down rated to 35w) with the same cooling design, then the blame falls on intel for an improper spec. 

[leadeater]

45 minutes ago, HammerDrop said:

I'm sorry man but ..... you're the one confused and placing your own meaning to what tdp is. 

"The processor TDP is the maximum sustained power that should be used for the design of the processor thermal solution"

No it's not that, this is the full and actual TDP spec that Intel says it is.

 

Quote

Intel defines TDP as follows: The upper point of the thermal profile consists of the Thermal Design Power (TDP) and the associated Tcase value. Thermal Design Power (TDP) should be used for processor thermal solution design targets. TDP is not the maximum power that the processor can dissipate. TDP is measured at maximum TCASE.

 

Quote

“TDP. Thermal Design Power. The thermal design power is the maximum power a processor can draw for a thermally significant period while running commercially useful software. The constraining conditions for TDP are specified in the notes in the thermal and power tables.” Notes: - TDP is measured under the conditions of all cores operating at CPU COF, Tcase Max, and VDD at the voltage requested by the processor. TDP includes all power dissipated on-die from VDD, VDDNB, VDDIO, VLDT, VTT and VDDA. - The processor thermal solution should be designed to accommodate thermal design power (TDP) at Tcase,max.TDP is not the maximum power of the processor.

https://www.intel.com/content/dam/doc/white-paper/resources-xeon-measuring-processor-power-paper.pdf

 

This is the Intel defined TDP spec not what you just said it was and there are very important factors in that definition. To name one, the TDP is taken at Tcase Max which means the CPU is at thermal max so not boosting which is why we say it's taken a base frequency because that is actually what Intel is saying, in the most technically correct way.

 

So if your cooler is exactly at the TDP rating the CPU will be at Tcase Max (Tj Max for non lidded) when the CPU is drawing that much power so it's impossible for it to boost long term which also means it will not use any more power than TDP under a sustained load. If you improve the cooler the CPU will boost and therefore go above the rated TDP, simple well known concept and is why Intel's says multiple times in the TDP document that TDP is not the maximum power of the processor.

 

Edit:

Which I'd like to point out that due to the MacBook Pro being unable to maintain base clocks on multiple common "commercially useful software" i.e. FCX, Premier, Resolve etc that either the cooling solution in the MacBook Pro is not a 45W cooling solution or Intel's "commercially useful software" is not that stressful of a load and does not generate that much power draw so TDPs derived from that is not that useful.

[HammerDrop]

18 hours ago, leadeater said:

No it's not that, this is the full and actual TDP spec that Intel says it is.

 

 

https://www.intel.com/content/dam/doc/white-paper/resources-xeon-measuring-processor-power-paper.pdf

 

This is the Intel defined TDP spec not what you just said it was and there are very important factors in that definition. To name one, the TDP is taken at Tcase Max which means the CPU is at thermal max so not boosting which is why we say it's taken a base frequency because that is actually what Intel is saying, in the most technically correct way.

 

So if your cooler is exactly at the TDP rating the CPU will be at Tcase Max (Tj Max for non lidded) when the CPU is drawing that much power so it's impossible for it to boost long term which also means it will not use any more power than TDP under a sustained load. If you improve the cooler the CPU will boost and therefore go above the rated TDP, simple well known concept and is why Intel's says multiple times in the TDP document that T

18 hours ago, leadeater said:

Intel's says multiple times in the TDP document that TDP is not the maximum power of the processor.

DP is not the maximum power of the processor.

 

Edit:

Which I'd like to point out that due to the MacBook Pro being unable to maintain base clocks on multiple common "commercially useful software" i.e. FCX, Premier, Resolve etc that either the cooling solution in the MacBook Pro is not a 45W cooling solution or Intel's "commercially useful software" is not that stressful of a load and does not generate that much power draw so TDPs derived from that is not that useful.

Don't really know how to make this easier so you can understand. But I'll try one more

 

tdp is a value. What you've stated and elaborated regarding tcase and tjmax etc is intel's explaination of arriving to that value. Intel then presents that value as a specification and says 

18 hours ago, leadeater said:

Thermal Design Power (TDP) should be used for processor thermal solution design targets.

However intel makes the sausage is up to them. The companies that partner with intel have to make sure it fits in the sesame seed bun. And intel provides a spec to do so. 

 

You are the one adding all this other stuff to the spec and redefining it. When intel clearly states in the original reference I mentioned and the one you just mentioned that:

18 hours ago, leadeater said:

The processor thermal solution should be designed to accommodate thermal design power (TDP)

 

 

18 hours ago, leadeater said:

Intel's says multiple times in the TDP document that TDP is not the maximum power of the processor.

You're right. But intel doesn't tell me to design a thermal solution based on maximum power of the processor, intel says:

"The processor TDP is the maximum sustained power that should be used for the design of the processor thermal solution"

 

18 hours ago, leadeater said:

Which I'd like to point out that due to the MacBook Pro being unable to maintain base clocks on multiple common "commercially useful software" i.e. FCX, Premier, Resolve etc that either the cooling solution in the MacBook Pro is not a 45W cooling solution or Intel's "commercially useful software" is not that stressful of a load and does not generate that much power draw so TDPs derived from that is not that useful.

This is only true if the i7 version is unable to maintain its clocks (but it does). 

 

Logic:

i7 is specd as 45w tdp > cooling solution based on that spec is designed to meet requirement... cooling solution seems to be good at rated(stock) speeds

 

i9 is specd as 45w tdp > same cooling solution as i7 is used... cooling solution cannot sustain chip at rated(stock) speeds

 

possible conclusion

Tdp rating is garbage and needs to be disregarded (I think you agree on this)

 

who is at fault?

 

that is all I'm arguing for. But either way, you really are putting a lot on top of what tdp actually is. Tdp is a number. Doesn't matter how intel gets the number. It's like asking someone what their height is and having them tell you that they are 6ft tall because their parents are tall, have tall genes, fed them the right foods and provided the right environment to activate tall genes. All that matters is you must've 6feet tall to ride the roller coaster

 

 

 

[leadeater]

1 hour ago, HammerDrop said:

However intel makes the sausage is up to them. The companies that partner with intel have to make sure it fits in the sesame seed bun. And intel provides a spec to do so. 

 

You are the one adding all this other stuff to the spec and redefining it. When intel clearly states in the original reference I mentioned and the one you just mentioned that:

No I'm not redefining it, the spec is the spec and yes you design a cooler off that spec with the knowledge of what that spec is. The TDP being a certain value does not mean the cooler you need for the device needs to dissipate that amount of cooling.

 

You can use a cooler of that TDP but that does mean that at that power draw the CPU will be at Tcase/Tj Max exactly as the Intel specification says. The device designer may not want that, they might want to use a better cooler to allow better short term or longer term boosting.

 

Designed to accommodate doesn't mean anything other than that, it doesn't mean that you should use a cooler of that TDP spec.

 

The spec is exactly as Intel says it is, the CPUs adhear to that spec and the behavior matches it. It does not adhear to what people misunderstand it to be, Intel TDP spec is not the exact value for the appropriate cooler design, it's just a spec for a known value that is used in the design of a cooling solution. Intel CPU will use more power than the TDP spec, you will need better cooling than the rated TDP of the CPU if you want to utilize Turbo Boost.

[dalekphalm]

1 hour ago, HammerDrop said:

Don't really know how to make this easier so you can understand. But I'll try one more

 

tdp is a value. What you've stated and elaborated regarding tcase and tjmax etc is intel's explaination of arriving to that value. Intel then presents that value as a specification and says 

However intel makes the sausage is up to them. The companies that partner with intel have to make sure it fits in the sesame seed bun. And intel provides a spec to do so. 

 

You are the one adding all this other stuff to the spec and redefining it. When intel clearly states in the original reference I mentioned and the one you just mentioned that:

 

 

You're right. But intel doesn't tell me to design a thermal solution based on maximum power of the processor, intel says:

"The processor TDP is the maximum sustained power that should be used for the design of the processor thermal solution"

 

This is only true if the i7 version is unable to maintain its clocks (but it does). 

 

Logic:

i7 is specd as 45w tdp > cooling solution based on that spec is designed to meet requirement... cooling solution seems to be good at rated(stock) speeds

 

i9 is specd as 45w tdp > same cooling solution as i7 is used... cooling solution cannot sustain chip at rated(stock) speeds

 

possible conclusion

Tdp rating is garbage and needs to be disregarded (I think you agree on this)

 

who is at fault?

 

that is all I'm arguing for. But either way, you really are putting a lot on top of what tdp actually is. Tdp is a number. Doesn't matter how intel gets the number. It's like asking someone what their height is and having them tell you that they are 6ft tall because their parents are tall, have tall genes, fed them the right foods and provided the right environment to activate tall genes. All that matters is you must've 6feet tall to ride the roller coaster

 

 

 

After having read the documentation provided by @leadeater, and your counter arguments, I really have to say that you're oversimplifying it.

 

Apple (or any OEM) does not simply use the TDP as the "one size fits all" calculation. @leadeater has explained why, but you're disregarding his responses. It's okay to disagree, but I would suggest you consider his argument with more weight.

 

Intel provides the TDP, and other specification to OEM's - in addition to samples - to assist in designing a proper thermal solution. If Apple doesn't take that information and design a correct product, then the fault is not Intel's.

 

Intel is blameless here. Maybe the i9 itself isn't a great product on mobile - that's a different argument. But Apple chose that product and decided to use it. They decided to put that product into a chassis that is insufficient to suit the requirements of the i9.

[leadeater]

11 minutes ago, dalekphalm said:

Intel is blameless here. Maybe the i9 itself isn't a great product on mobile - that's a different argument. But Apple chose that product and decided to use it. They decided to put that product into a chassis that is insufficient to suit the requirements of the i9.

I happen to think the i9 is a rather dumb mobile SKU, though it does make a bit of sense in high end gaming laptops with a GTX 1080 for example. Intel's TDP spec was defined in 2011 when CPU boosting wasn't so significantly different from the base and peak power wasn't order of magnitude higher. Now we're in 2018 and have CPUs almost doubling their frequency when boosting and doubling the peak power draw, recipe for problems.

 

I would agree that it is probably a good time for Intel to revisit the TDP spec, they tried a while ago with SDP but that is even more problematic and I don't thnik they use it any more.

[HammerDrop]

24 minutes ago, leadeater said:

The spec is exactly as Intel says it is, the CPUs adhear to that spec and the behavior matches it. It does not adhear to what people misunderstand it to be, Intel TDP spec is not the exact value for the appropriate cooler design, it's just a spec for a known value that is used in the design of a cooling solution. Intel CPU will use more power than the TDP spec, you will need better cooling than the rated TDP of the CPU if you want to utilize Turbo Boost.

So, if two processors have the same tdp spec, can I, according to intel's spec, use the same cooler for both, and expect them to perform at their stock/base clocks? Or whatever they deem is "commercially available heavy load?"

[leadeater]

2 minutes ago, HammerDrop said:

So, if two processors have the same tdp spec, can I, according to intel's spec, use the same cooler for both, and expect them to perform at their stock/base clocks? Or whatever they deem is "commercially available heavy load?"

Should do yea, wouldn't be a very useful spec if it that wasn't case (even though I think it's not that useful anyway). Most of the followup articles show the i9 in the MacBook Pro running at base now, or just ever so slightly below depending on application (those I saw first hence my comment about the cooler being below TDP spec).

 

Not knowing what Intel actually uses software wise to get the TDP I have to say since most examples post update show it running at base and 90+ degrees that the cooler is pretty much at or slightly above the TDP spec.

[HammerDrop]

26 minutes ago, dalekphalm said:

Intel is blameless here

 

26 minutes ago, dalekphalm said:

If Apple doesn't take that information and design a correct product, then the fault is not Intel's.

 

Im arguing apple took intels' specifications and followed them. And ended up with the thermal issues

 

27 minutes ago, dalekphalm said:

Apple (or any OEM) does not simply use the TDP as the "one size fits all" calculation. @leadeater has explained why, but you're disregarding his responses. It's okay to disagree, but I would suggest you consider his argument with more weight.

I am not disregarding his points, I am evaluating them and deciding what is pertinent to the case at hand. Most Of the information in addition to the definition of tdp that lead eater's provided changes absolutely nothing regarding the spec and how it should be treated by engineers and designers. Adding in how intel arrives to the spec does nothing to differentiate between a 45w tdp i7 and a 45w tdp i9. 

 

In, fact I would argue that out of anyone here I've actually taken the time to read his sources and comprehend them. I'd say you haven't. Because well I'll quote it again:

"The processor TDP is the maximum sustained power that should be used for the design of the processor thermal solution"

 

and yes you can design a cooler that is beyond the spec, but the spec is an agreement. If you build it, then processor x will perform according to spec. If you build it better, processor x MAY perform better. But I can only guarantee that if you build it based on what I tell you, then you will get the performance I am writing on paper. This is the agreement between every device and component manufacturer/supplier to consumer/designer etc...

 

 

[HammerDrop]

4 minutes ago, leadeater said:

Should do yea, wouldn't be a very useful spec if it that wasn't case (even though I think it's not that useful anyway). Most of the followup articles show the i9 in the MacBook Pro running at base now, or just ever so slightly below depending on application (those I saw first hence my comment about the cooler being below TDP spec).

 

Not knowing what Intel actually uses software wise to get the TDP I have to say since most examples post update show it running at base and 90+ degrees that the cooler is pretty much at or slightly above the TDP spec.

Exactly!!!

So intel screwed the pooch with a bad spec

 

 

cool case closed. Returning to apple. Btw reports for the i9 from dave2d still show that it's under base clocks with the fix. Waiting for more tests but yeah the 45w tdp for the i9 seems like a bad spec.

[Arika]

6 minutes ago, HammerDrop said:

"The processor TDP is the maximum sustained power that should be used for the design of the processor thermal solution"

you keep saying this as if it's some big "gotcha" sentence. but you're missing the point of it "maximum sustained power" this means the base clock. Turbo is only a temporary boost.

 

Quote

Intel® Turbo Boost Technology 2.01 accelerates processor and graphics performance for peak loads, automatically allowing processor cores to run faster than the rated operating frequency if they’re operating below power, current, and temperature specification limits. Whether the processor enters into Intel® Turbo Boost Technology 2.0 and the amount of time the processor spends in that state depends on the workload and operating environment.

So why would a temporary boost be defined under the maximum sustained power

[leadeater]

30 minutes ago, HammerDrop said:

Exactly!!!

So intel screwed the pooch with a bad spec

I'm not sure how you got to that? If most tests show the i9 running at base clocks now then wouldn't that mean the spec is correct?

 

Edit:

Riding the line between 2.8 and 3.0, obviously not the same testing conditions Intel uses both ambient temp and software but looks to me like it's now able to run at base and is below Tj Max.

[HammerDrop]

20 minutes ago, Arika S said:

keep saying this as if it's some big "gotcha" sentence. but you're missing the point of it "maximum sustained power" this means the base clock. Turbo is only a temporary boost.

I've only ever cared about base clocks. And the basis of my arguments is that the i9 cannot run at its base clock

 

21 minutes ago, Arika S said:

So why would a temporary boost be defined under the maximum sustained power

Please show me where this is stated and where turbo boost (unless you are referring to another kind of boost) is part of the tdp spec

[Arika]

1 minute ago, HammerDrop said:

I've only ever cared about base clocks. And the basis of my arguments is that the i9 cannot run at its base clock

which is the fault of apple screwing up with VRM power delivery. no other laptop with the same i9 has trouble keeping base clocks

 

Quote

Please show me where this is stated and where turbo boost is part of the tdp spec

Thats not what I'm saying at all. I'm saying turbo boost is NOT part of the TDP spec

[HammerDrop]

Just now, leadeater said:

I'm not sure how you got to that? If most tests show the i9 running at base clocks now then wouldn't that mean the spec is correct?

Actually your right. I was wrong. With the fix it's now at 3ghz so it was a 45w tdp after all. 

[HammerDrop]

5 minutes ago, Arika S said:

which is the fault of apple screwing up with VRM power delivery. no other laptop with the same i9 has trouble keeping base clocks

 

Thats not what I'm saying at all. I'm saying turbo boost is NOT part of the TDP spec

Actually as leadeater just corrected me the i9 is now running at base clocks without issues after the software fix. Tells me apple designed the Vrms and cooling to spec. 

 

and yup, turbo boost is not part of the spec. Intel guarantees base clocks with the specified tdp. Turbo boost is not guaranteed at all and intel does not take it into account as part of their spec. Perfectly good since they don't promise any specific turbo boost performance for a given thermal spec 

[Arika]

1 minute ago, HammerDrop said:

Actually as leadeater just corrected me the i9 is now running at base clocks without issues after the software fix. Tells me apple designed the Vrms and cooling to spec. 

yes, Apple had to release a software fix because they stuffed up. It altered how power is delivered to the CPU which was off spec to begin with causing the throttling. once apple fixed their power issue the i9 is now running withing the TDP spec like it should.

Which means? Apple's fault, not Intel's