...and so they haven't named this Lava Lake yet, why? It's definitely looking to be Lava Lake to me if THAT is the low TDP version. Hope the rather useless noisy cooler they include is made of something that doesn't melt easily. Aluminum might not cut it at the rate they're going.

Imagine the 10900k, thats 200w chip easy.

Gaming Chillers - Soon from Corsair.

Either a keystroke error on Intel's part or blatant mis-representation.

Is it still not globally understood by now that Intel TDP is designed at base clock, not turbo? Its the amount of cooling they require to guarantee operation without throttling the base clock. Turbo will *always* overshoot that.

looks like it's a fail bin underclocked 10900 or something, why did they think that would magically make them 35watts, who knows my 9900k stock (so 4.7 all cores stock) was around 120W if memory serves but udnerclocked to 4.1 it dropped to 60W i'm not incredibly surprised by 125w at 4.5Ghz amd 3xxx threadripper stock easily go to 280+Watts overclocked 340+ watts whatever you think is hot at Intel you have no idea what TR is like but it has 2x-3x the cores without 2-3x the watts, that's the thing

If intel wants to make a frying pan, intel makes a frying pan. Simple and Crisp

I mean... TDP is usually about thermals. Which is wasted electricity. Photons not hitting the electrons they should be hitting, but going perpendicular and leaking out as heat. Maybe only 35W ( or some typical % of that ) worth is leaking out on average, so its a cooling solution spec.

So, I guess the i9 10900K with 125TDP will top at 439 Watts. Glad I went with AMD in this one!

Peak boost isn't that much higher than 10900T. So, probably wont pull 4x more power at peak vs 10900T at peak lol

So slower, hotter, and more expensive than AMD *again*... :P

Am I the only one who thinks 125W is reasonable for all 10 cores to be running at 4.8GHz? I assume this is also a 14nm part. Sure, this isn't great. Intel does need to step up their game. But it's not that bad.

schmidtbag:

Am I the only one who thinks 125W is reasonable for all 10 cores to be running at 4.8GHz? I assume this is also a 14nm part. Sure, this isn't great. Intel does need to step up their game. But it's not that bad.

Agree. Plus if you know how Intel rates TDP's and all of the engineers making laptops, desktops etc do, then you know you can't correlate it to actual power draw. Personally I think Intel and AMD should have an actual power draw rating to end everyone discussing TDP's. It would be the layman's technical spec so to speak vs the engineers tech spec for TDP's.

JamesSneed:

Personally I think Intel and AMD should have an actual power draw rating to end everyone discussing TDP's. It would be the layman's technical spec so to speak vs the engineers tech spec for TDP's.

AMD's TDP rating come pretty close to the all-core load. The problem is Intel, who only lists the TDP for the base clock. Simply put, Intel's TDP should be completely disregarded at this point. They can simply specify a very low base clock (such as with this processor - 1.9 GHz) and get any TDP figure they want.

The way TDP is used is just wrong in every aspect. 6 Core i5, 8 core i7 and 10 core i9 all with the same 125W TDP. Sure they have different base and boost clocks, but i9 have the highest boost clock, but still the same TDP. Makes no sense at all the way TDP is used. The Thermal Design Power is the maximum amount of heat generated by a computer chip. When using real applications so how does the i3's have the same TDP when running with different all core clocks ? An i3 with an all core clock of 4.1 GHz generates just as much heat as one running 4.4 GHz ? They have the same TDP. Edit Yes I know Intel list TDP for base clocks

Yeah, there lots of confusion between TDP and actual wattage. I myself confused, so I try to stay away from such topics. I mean CPU might be able to pull 1KW and have only 15 TDP, cause it super efficient or something

These parts get a over TDP boost for a few seconds then the limiter kicks in. under sustained load the CPU will get down to the 35W number. But I agree this needs to be disclosed. Imagine a Optiplex sff office PC using this CPU, "every time I un-zip a file my PC shuts down."

I really doubt that a cooler meant for a 35W TDP is enough. Intel on TDP is simply fraudulent.

I am well aware of Intel's TDP being based on the base clock, but there really should be an industry standard that all CPU makers have to follow with TDP as it confuses everyone. I build custom rigs for people and I have to explain to people more than one would think that in order to get the best performance 5.1, 5.2, 5.3Ghz all core out of something like the 9900K you need much more cooling than can easily fit in a small form factor build. Customers look only at TDP and think all they need is something like a hyper 212 EVO and 2 case fans. AMD has a better approach to TDP as they more closely follow TDP at boost load, but even they are deceptive. CPU makers should have to follow the practice of listing TDP the same way they market their processors, ie what is the TDP at maximum boost. For Intel that means what is the TDP for your processor at maximum all core factory boost, and for AMD it would be what is the TDP for the processor with XFR maximum boost. Technically neither is "overclocked" that way and it is factory boosting. AMD is already really close to that actual TDP, but Intel would never allow it because then they would have to list their 35W components at the actual 125W+ they need for factory operation. I would love to put a 35W rated cooler on the 10900T and see exactly what happens. I have a feeling that it would either throttle to 1Ghz (or under) or it might even burn itself up. I remember the FX 9590 and how everyone made fun of its power consumption and made all the jokes about it being a fire hazard... well what goes around comes around. These 10900 series processors are going to run hot and use a lot of power. At least AMD listed the FX 9590 @ 220W, they didn't try to pass it off as a 35W, or even 125W chip.

I am having trouble with finding the same test in Sisoft Sandra 2020 downloaded from Guru3d. The interface does not look the same.