I9 9900k Official Reviews from Anandtech, Tomshardware. Add your own links to others !

Page 20 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

coercitiv

Diamond Member
Jan 24, 2014
7,225
16,982
136
The whole point is that you are guaranteed to get certain performance based upon the rating of your cooling, and that using better cooling may give better performance.
As a consumer that is much easier to understand than saying it'll use X amount of power, but what cooler you'll need to buy is your own business.
So where exactly in custom cooler specs do they mention this "rating of cooling" you mentioned?

If it is Intel's then they really ought to list the 9900k as a 4.2GHz base clock 95W TDP CPU.
You forgot iGPU power usage.
 
Last edited:

PotatoWithEarsOnSide

Senior member
Feb 23, 2017
664
701
106
Well, they don't, though clearly they should as that is their purpose...to dissipate heat.
Why they don't do it is probably a whole other issue, most likely related to marketing; you don't advertise that your cooling solution is just as good as that cheaper one from your competitor.
 
Last edited:

coercitiv

Diamond Member
Jan 24, 2014
7,225
16,982
136
Well, they don't, though clearly they should as that is their purpose...to dissipate heat.
Why they don't do it is probably a whole other issue, most likely related to marketing; you don't advertise that your cooling solution is just as good as that cheaper one from your competitor.
So this rating is much easier to understand than one based on electrical power usage, with the only caveat of being essentially useless, as no cooler maker will advertise this spec. You don't see the disconnect here? How can it be better and useless at the same time?
 

PotatoWithEarsOnSide

Senior member
Feb 23, 2017
664
701
106
It's only useless because no-one is forcing them to list it in an understandable way.
Sure, it's easier to sell a cooling solution by referencing fan sizes, fan speeds, exotic heatsink materials or whatever, but the word that we always see on advice forums regarding cooler choice is SHOULD. X cooler should be enough.
If you are selling a CPU without a cooler, the least that we should expect is a definitive standard of what WILL work with the CPU. Nobody wants thermal throttling.
 
  • Like
Reactions: moinmoin

TheGiant

Senior member
Jun 12, 2017
748
353
106
It's similar to 9590 except performance.
Well, it is like old American 5.0V8 engine with 3 speed transmission consuming 20liters/100km like mine bmw 540i xdrive when I am driving "a little faster"
Pretty much the same
 

dlerious

Platinum Member
Mar 4, 2004
2,068
876
136
Well, it is like old American 5.0V8 engine with 3 speed transmission consuming 20liters/100km like mine bmw 540i xdrive when I am driving "a little faster"
Pretty much the same
I had a Thunderbird with a 429SCJ, when I got a little bit of lead in my foot, it was gallons per mile. Parents sold it when I was in the military in the 70's.
 

coercitiv

Diamond Member
Jan 24, 2014
7,225
16,982
136
Nope, the TDP will always be the same when you want to keep the default AMD CPU specs.
According to AMD:
TDP (Watts) = (tCase°C - tAmbient°C)/(HSF ϴca)
HSF ϴca (°C/W): The minimum °C per Watt rating of the heatsink to achieve rated performance.

When you choose a different min spec for the heatsink and keep tCase the same, TDP rating must follow. If TDP is not directly influenced by these variables, then this is not actually the formula for TDP, it is the formula for HSF ϴca.
 

PotatoWithEarsOnSide

Senior member
Feb 23, 2017
664
701
106
That's precisely what TDP is; its a cooling metric.
That's why when you see a CPU rated 95W TDP, it stays within its thermal limit if the correct cooling solution is used. Unfortunately, coolers don't state what their TDP capacity is, which they should as I explained previously.
Otherwise engineers would just create any old power hog and say "cool it".
 

coercitiv

Diamond Member
Jan 24, 2014
7,225
16,982
136
That's precisely what TDP is; its a cooling metric.
That's why when you see a CPU rated 95W TDP, it stays within its thermal limit if the correct cooling solution is used.
And I argued it is not the correct metric to be used when advertised to consumers. If you wanted something that was more relevant and accurate than average power consumption of the CPU, they should have used the recommended heatsink thermal resistance, just like thermal paste manufacturers use thermal conductivity to describe product performance.

What good does this TDP measure do me, the consumer, when cooler manufacturers can't state this metric on the box? TDP is directly dependent on temperature delta, and different CPUs can have different max tCase ratings, meaning max TDP rating of a cooler can be different even for CPUs which use the same amount of electrical power, but have different tCase requirements.

This why I said AMDs definition may be good for engineers but it's impractical for consumers: it resembles power usage but it's not power usage, it resembles a cooling metric but cannot be used as such by consumers and cooler manufacturers alike. In fact consumers are better off using this rating as average power consumption and scaling their cooling accordingly.
 

moinmoin

Diamond Member
Jun 1, 2017
5,234
8,442
136
I think the reason for AMD's twisted calculation was nicely showcased by Intel's very 9900k actually: As we now know performance is not fixed to power consumption (and as such heat) but further reliant on package temperature. Above 80°C 9900k's power consumption strongly rises even while keeping the performance the same. So the heatsink thermal resistance is a significant part of the equation (and varies depending on the use case when overclocking).
 

TheGiant

Senior member
Jun 12, 2017
748
353
106
If this is the only complaint you have against that review, you may want to know this is AMDs fault, not TechSpot's.

AMD has a different way of defining TDP which I kinda despise as it relates to heatsink performance and temp delta to define a thermal dissipation rating that completely ignores electrical power usage. It may be a correct engineering solution, but this spec is one advertised to consumers and should be better anchored with this requirement. Intel's definition is far more easy to understand, consistent across their entire product stack and across previous generations (when properly implemented and obeyed).

To keep things short, an AMD CPU can have different TDP rating depending on the cooler spec used to calculate that TDP. Since both 2600X and 2700X come with boxed cooler, and since the Wraith Prism that comes with 2700X is clearly superior to Wraith Stealth.... I think you see where this is going.
Well we know all about the different TDP calculations. But that article focused pure on power. And when we focus on power, then the conclusion is system power is the same with ryzen2600X as with the 9900K while the 9900K offers very much different performance.
And the second thing is, it focuses on power with high end desktop CPU which is accidentally highest performing desktop class CPU available. And that is simply reviewing my BMW 540i xdrive with max fuel consumption artificially set to 5l/100km. Pure nonsense.
So the conlusion says nothing to the people who want to buy it and also nothing to people who don't.
Pretty much no one cares for rated TDP or temperature (if it is safe, for me max low 80s C).
Usually TR articles do have some insight in them, this one looks to me as wasted review resources.
 

coercitiv

Diamond Member
Jan 24, 2014
7,225
16,982
136
Well we know all about the different TDP calculations. But that article focused pure on power. And when we focus on power, then the conclusion is system power is the same with ryzen2600X as with the 9900K while the 9900K offers very much different performance.
Yes, and it's all good as long as we compare both products at stock, take a look at temps, power, price/ efficiency etc. We can also overclock both products and see that 9900K can scale much higher as it has a lot of frequency potential left if we relax power limits and cool it properly.

What is not ok is to review the 9900K with improper power limits and advertise the CPU as operating at stock.

And the second thing is, it focuses on power with high end desktop CPU which is accidentally highest performing desktop class CPU available. And that is simply reviewing my BMW 540i xdrive with max fuel consumption artificially set to 5l/100km. Pure nonsense.
It wouldn't be nonsense if BMW advertised the 540i xdrive as a car with a typical fuel consumption of 5l/100km. You can't have the cookie and eat it too.

So the conlusion says nothing to the people who want to buy it and also nothing to people who don't.
Pretty much no one cares for rated TDP or temperature (if it is safe, for me max low 80s C).
It says everything to the people who want to buy it, it shows them what to expect in terms of thermals and performance based on the motherboard they buy and the settings they choose. They can choose to unleash the beast with proper cooling and it also allows people who do not need sustained AVX performance to minimize cooling solutions and size of the build accordingly.

After the first round of reviews came up, people were already on these forums asking if the 9900K isn't too hot to use in their casual systems because they saw reviews in which some of the best air cooling or 240mm AIOs could barely keep the 9900K in the low 80s.

Remember this CPU allows someone with almost no OC experience to go from 100W to 200W power consumption based purely on motherboard choice and choosing YES at the MCE prompt in BIOS (for some even enabling XMP can be enough). You choose a badass looking MSI board, press Yes when BIOS asks you if you want more performance, and your cooling requirements go from air to water.
 
Last edited:

TheGiant

Senior member
Jun 12, 2017
748
353
106
It wouldn't be nonsense if BMW advertised the 540i xdrive as a car with a typical fuel consumption of 5l/100km. You can't have the cookie and eat it too.
.
Yes they can (if they create their own methodics like the Intel's TDP which is accurate but by common sense it is unusable, well not unusable but complete nonsense). It is like buying the 540i but advertising its 5l fuel consumption with 2 of 6 cylinders active without turbo on the engine and without the 4x4 mode, only rear and with steady 80km/h on flat road with 1 70kg person inside and 25C road temperature and wind 3 km/h from behind...OMG basic clocks
What I am saying to to me as a buyer of 90k EUR car, I am buying top model I am not interested in such results.
It tells me nothing. I do really care about fuel consumption because of refueling period. That's it (call me ecoterrorist).
The unlimited power of 9900K tells me nothing, because like with the car they can setup the engine to consume 25l or more/100km and gain a little more than with 10l/100 (driving with 8.1, the engine is really good).
My scope of the review:
  1. opinion
    1. TDP - Intel is misleading but doesn't lie (how nice they are), AMD overcomplicated the definition with good purpose but the random person doesn't understand it
    2. How many MB manufacturers set which mode as default (non direct statement, that is Intel intention to do so)
  2. Possible configurations of power (not TDP)
    1. how does the 8C 9900K and 8C 2700X perform with the same power consumption set to 65W,95W,130W (or the max of the 2700X), 150W (the max of reasonable air cooling IMO), unlimited
    2. coolers/motherboard examples for the power levels and temps- ambient is 22C (pretty normal) - this is important for the buyer decision- can I get better than 2700X without burning the room or investing in water cooling?
    3. performance/power curve and if they measure also sound levels that would be nice (and with 2080Ti, so if the CPU cooler really matter in terms of loudness)
    4. min fps performance/power dependency
    5. just for info- if they can deactivate 4C to make that 7700K with moar L3, what is the impact
  3. I wouldn't bother commenting the price, it is not important with 9900K segment of buyers
 
Last edited:

coercitiv

Diamond Member
Jan 24, 2014
7,225
16,982
136
Intel should make a customized implementation of Windows Power Plans to replace standard MS presets:
  • Power Saver renamed to "Air Cooling" 95W TDP
  • Balanced renamed to "AIO Water" 140W TDP
  • High Performance renamed to "Custom Loop" 250A
There you go, stock power settings fixed, enjoy the fastest mainstream CPU on planet Earth.
 
  • Like
Reactions: lightmanek

TheGiant

Senior member
Jun 12, 2017
748
353
106
Intel should make a customized implementation of Windows Power Plans to replace standard MS presets:
  • Power Saver renamed to "Air Cooling" 95W TDP
  • Balanced renamed to "AIO Water" 140W TDP
  • High Performance renamed to "Custom Loop" 250A
There you go, stock power settings fixed, enjoy the fastest mainstream CPU on planet Earth.
Renamed
  • You are doing it wrong 95W TDP
  • Someone knows what he is buying 140W TDP
  • My pink Ferrari is awful 250W
 
  • Like
Reactions: lightmanek

JoeRambo

Golden Member
Jun 13, 2013
1,814
2,105
136
Honest question*: is TDP any problem at all if user is not doing any rendering and DC computing? This back and forth about TDP in enthusiast forum is a bit like reading proceedings of "Bees against honey" association. As long as people have selection of chips, mbs, coolers and BIOS settings there should be no problem with TDP at all for enthusiast chips like 9900K or 2700x stock or oced.

*I ran HWInfo on my oced 8700K, doing what I do and gaming for like 4 days without reboot and it registered a max of 90W Package Power. Far cry from those Cinebench or god forbid Prime95 wattages.
 

Pandamonia

Senior member
Jun 13, 2013
433
49
91
Total system power at 8 cores 5ghz is only what is important. Tdp is BS.

I saw a system power that looked reasonable even at 5ghz but can't remember where