Core i9-9900K Power & Thermals, Did Linus (and OC3D TV) Get it Wrong?

[mr moose]

1 hour ago, HashBrowns said:

Still over here on my i7 2600 and in heaven. Not sure when I will  be incentivized to upgrade... 9900k is insanely expensive, so is ram...my lord

 

 

 

43 minutes ago, MMKing said:

Sitting on the 4770K myself. At this point, i'm probably gonna end up skipping DDR4 altogether.

 

Ditto, I am on a i5 3550.  it still plays everything I like well enough for me.  As much as I want to build a new PC there's just not enough carrots dangling in front of me.

[leadeater]

5 minutes ago, mr moose said:

Ditto, I am on a i5 3550.  it still plays everything I like well enough for me.  As much as I want to build a new PC there's just not enough carrots dangling in front of me.

What if the carrots were cooled with chilled water? 

[mr moose]

1 minute ago, leadeater said:

What if the carrots were cooled with chilled water? 

hmmmm,  can I eat them or is it a marketing illusion?

[leadeater]

1 minute ago, mr moose said:

hmmmm,  can I eat them or is it a marketing illusion?

Only if you use them to make carrot cake, then yes you can have your cake and eat it too.

[mr moose]

1 minute ago, leadeater said:

Only if you use them to make carrot cake, then yes you can have your cake and eat it too.

sold, my new pc will be made from carrots.

[schwellmo92]

19 hours ago, leadeater said:

Oh FFS, knew I should have trusted the Intel XTU screenshot I actually saw showing 120W for the Turbo Boost Short Power Max.

 

Source: Hardware Unboxed Video

 

So maybe the 22 seconds someone else said is not correct either, or rather not Intel default. The above image is using Intel SVID so everything should be actually stock/to Intel spec.

I'm still not convinced it will actually hit 4.7GHz all-core with a 120w limit. I can see the frequency in the graph dipped (very slightly) to 4.3GHz pretty much the second the load went to 100% solid.

[Carclis]

19 hours ago, leadeater said:

Oh FFS, knew I should have trusted the Intel XTU screenshot I actually saw showing 120W for the Turbo Boost Short Power Max.

 

Source: Hardware Unboxed Video

 

So maybe the 22 seconds someone else said is not correct either, or rather not Intel default. The above image is using Intel SVID so everything should be actually stock/to Intel spec.

Did you look at the Anandtech review?

Another curious thing we learn there is 95W =/= 95W

[thorhammerz]

16 minutes ago, Carclis said:

Did you look at the Anandtech review?

Another curious thing we learn there is 95W =/= 95W

Different (c)TDP limits set by different motherboards (which has been argued on ad nauseum already).

[Carclis]

16 minutes ago, thorhammerz said:

Different (c)TDP limits set by different motherboards (which has been argued ad nauseum already).

That wasn't exactly my point. I had a discussion in a different thread where I mentioned that if Intel's TDP had any relevance it would not be the exact same value for a 6c/6t, 6c/12t, 8c/8t, 8c/16t CPU which all have very similar frequencies on the same manufacturing process. To quote Ian Cutress:

Quote

Comparing to the other two ‘95W’ processors, we can see that the Core i9-9900K pushes more power as more cores are loaded. Despite Intel officially giving all three the same TDP at 95W, and the same PL2 at 210W, there are clear differences due to the fixed turbo tables embedded in each BIOS.

It also stands to reason that Turbo behaviour renders the TDP redundant since it is default behaviour and Turbo clocks are more often than not being pushed by the CPU.

[thorhammerz]

5 hours ago, Carclis said:

It also stands to reason that Turbo behaviour renders the TDP redundant since it is default behaviour and Turbo clocks are more often than not being pushed by the CPU.

You are not wrong, but turbo clock frequencies (and power draw at anything other than base frequencies) are not guaranteed by Intel - when one buys an 9900k , what one buys is an 8C/16T chip that runs at 3.6 GHz on all cores at or under TDP, with the possibility of boosting up to 5.0 GHz "if thermal/voltage/power/etc conditions are met / the stars align" (e.g.you install a water chiller + get a z-chipset motherboard + remove motherboard PL limits + possibly modifying turbo tables via MCE or manual overclock).

 

Previous generations (7th gen and earlier) of Intel desktop CPUs did not bring attention to this technicality because often times their power draw at both base and boost were (well) within the PL1 TDP limit, and thus allowing people to adopt a rather ignorant, lax and inaccurate perception of how Intel actually defines their TDP limits. Whether one agrees with how said limits are defined (as a certain Intel-hater on this forum has crusaded ad nauseum on) is ultimately irrelevant in the face of reality.

 

Deceptive marketing that takes full advantage of general consumer ignorance? Certainly, but the company is merely out to make/fleece money at the end of the day.

[leadeater]

4 hours ago, schwellmo92 said:

I'm still not convinced it will actually hit 4.7GHz all-core with a 120w limit. I can see the frequency in the graph dipped (very slightly) to 4.3GHz pretty much the second the load went to 100% solid.

It won't, I've mentioned that before. You need 160W for that and in this case Intel has not actually published full Turbo tables with frequencies for each thread/core count, from the comments I've heard from a few of the reviews there isn't actually an Intel published all core turbo at all, only that 2 core 5.0Ghz.

[mr moose]

2 hours ago, thorhammerz said:

Deceptive marketing that takes full advantage of general consumer ignorance? Certainly, but the company is merely out to make/fleece money at the end of the day.

And on top of that Intel have to cover their arse with the most accurate TDP specs they can guarantee.  Or else when you get the worlds most valuable pc manufacturer with reputation for quality go and fuck up the cooling solution,  who does everyone blame? 

[HashBrowns]

13 hours ago, mr moose said:

 

 

Ditto, I am on a i5 3550.  it still plays everything I like well enough for me.  As much as I want to build a new PC there's just not enough carrots dangling in front of me.

Yeah, the cheapest route I was looking at would be the R5 2600 on ebay for $149 - the free 15$ ebay coupon. That would $134. Then another $124-$150 for 16gb ram, and 70-100 for a mobo.

 

That's a lot of money for me personally, but I'm not sure if that would even be worth it. Would I see a noticeable performance difference in gaming/workstation?

[mr moose]

2 minutes ago, HashBrowns said:

Yeah, the cheapest route I was looking at would be the R5 2600 on ebay for $149 - the free 15$ ebay coupon. That would $134. Then another $124-$150 for 16gb ram, and 70-100 for a mobo.

 

That's a lot of money for me personally, but I'm not sure if that would even be worth it. Would I see a noticeable performance difference in gaming/workstation?

You likely won't notice it in gaming, but if you do heavy video work you'll likely save yourself a bit of time rendering and general editing.