It's not about power draw. The assumption is that mobo manufacturers (all of them) are setting the PL2 to 210w just because they want Intel's CPU's to look better (someone actually said this in this thread). I know you are headfast stuck to 119w being the default for the 9900k. So that means motherboard manufacturers are not setting the board to default with the PL2 at 210w but are in fact setting the PL2 to 210w right off the bat. If that's the case boards where you can manually make the change and therefore can see what the PL2 is set to. I wonder how they react to other CPU's like the 8700k. It's not going to use as much power as the 9900k, but is it still going to be at 210w? Because if it's not 210w it would imply that they are picking up the default from the CPU. Which would mean that 210w isn't motherboard manufacturers illegally overclocking your CPU by default as has been suggested, but the actual default settings of the CPU by Intel.
A note for a diagram stating general ratio for the platform while noting that it may be changed to more efficient options by hardware. Does not make a set in stone standard on what the default is.
Do you see 210W on the ARK website? That 210W was a recommendation from Intel for the 9900k review. I believe Intel came up with that figure to prevent wild disparaties between the reviews. 210W is basically unlimited ACT. It goes without saying, that reviewers will also need the cooling to match.
The assumption is that mobo manufacturers (all of them) are setting the PL2 to 210w just because they want Intel's CPU's to look better
Why is this discussion about an upcoming CPU being overrun by TDP talk? Can we talk about Comet Lake-S?
Sure, got anything new to share?Why is this discussion about an upcoming CPU being overrun by TDP talk? Can we talk about Comet Lake-S?
I think the the reason if the rumor is true it's a Coffeelake-R with two extra cores, which was a Coffeelake with 2 extra cores, which was a Kaby Lake with 2 extra cores. It's obvious that Intel can't maintain its clock lead if they keep this up. 9900K really shows how thinly stretched they are because to maintain their performance and clock lead the chip needs to use 77% more than it is spec'd to. 2 more cores is going to make this even worse.Why is this discussion about an upcoming CPU being overrun by TDP talk? Can we talk about Comet Lake-S?
I think the the reason if the rumor is true it's a Coffeelake-R with two extra cores, which was a Coffeelake with 2 extra cores, which was a Kaby Lake with 2 extra cores. It's obvious that Intel can't maintain its clock lead if they keep this up. 9900K really shows how thinly stretched they are because to maintain their performance and clock lead the chip needs to use 77% more than it is spec'd to. 2 more cores is going to make this even worse.
They can just bump up the TDP/PL2 further, especially since Comet Lake-S will probably not be compatible with Z370/Z390 boards. It's really not a big deal.
Let's put this "130W TDP" cooler thing to rest:
When required to operate with a Tcase of 65C the 2015D cooler has a max TDP of 95W. It would only be able to reach a max TDP of 130W if it were allowed to function with a higher Tcase of 72-73C.
And so be it. Mine and most peoples complaints here isn't about power usuage. I mean even for AMD supporter, you get to 12 cores and you are talking about 180w. A 32c "enthusiast CPU" what was it 250w. The 9900k running at 4.7 or 4.6 or whatever value it is because Intel now hides it, at 160w is probably worth the cost of doing business for the people who don't care (most). The problem is the transparency on CPU power usage, basically a useless and untruthful TDP rating, and confusion on required cooling for the unsuspecting enthusiast. This will get worse with this chip if it exists. 10c, same process, if they can maintain anything near 4.5GHz All core turbo, it's going to suck up 200w on some peoples chips. But my guess is they will think that they are covering up their manufacturing issue and meeting some OEM requirement by calling it a what a 100W TDP CPU?
It all depends on the clocks and voltages used. You can probably even get a '95W' 10C Comet Lake if you reduce voltages and set the ACT at 4.0GHz. That is actually the optimal efficiency range for 14nm, at the low 4GHz range. 4.5GHz+ is all about outright performance, not power efficiency. There is a reason the 12C TRs are 180W and that is because they are clocked at 4.0GHz, the upper limit of efficient perfomance/watt for AMDs 12nm process.
FWIW even a 4.0GHz Comet Lake would still outperform a 9900K @ 4.7GHz in MT thoughput - 10 x 4.0 = 40 vs 8 x 4.7 = 37.6 and that is assuming no IPC uplift. I think Intel would probably set the ACT higher than that, but I doubt we'll be seeing sustained 4.7GHz all core boost clocks.
But my guess is they will think that they are covering up their manufacturing issue and meeting some OEM requirement by calling it a what a 100W TDP CPU?
I guess the whole point that you seem to be missing is that the board can configure the 9900K to use more than 95W/119W without it technically being overclocking. And so if an OEM wanted to use the 9900K they would have the capability to adjust PL2/time to what their cooling can handle.
Would not be surprised if Intel just completely unrestricts K parts at stock settings and it's just the silicon lottery as to how high each core goes; assuming you can cool it of course.
Intel calls it 14nm++.No that is exactly what I am counting on and it's why I am disappointed. In the end I do have a desire for an enthusiast to also be learned Enthusiast. But that isn't true. What this does is make a 9900k OEM machine practically be a completely different processor performance wise. And people grabbing retail CPU's will have little cooling information to go on and if not careful they will be perpetually toasty. All so they could retain the power ceiling OEM's desire and hide the fact that the process isn't really advancing even though they try calling it 14nm++++++++++.
It's a joke.Intel calls it 14nm++.
There is only 14nm, 14nm+, and 14nm++.
Three gens only.
There's a marked improvement between coffeelake and this. Even configured for 95W, it still outperforms the 2700x in a majority of tasks, going by the Gamers Nexus tests, iirc.It's a joke.
Tcase is the temperature the heatsink is exposed to and directly dictates the amount of heat the cooler can dissipate. If Intel says the 9900K must have a Tcase of 65C with the reference cooler (2015D) then the TDP of the cooler for that corresponding Tcase is 95W. This is a scientifically certainty, there is no marketing way around it.Tcase max also appears to be a max temp measured during testing at the factory at the base clock, rather than a number that won't be exceeded in the real world?
The more I read, the more slightly different info I seem to run across, and the more I doubt...
It should do.
At 95w it is doing 4.2GHz on all 8 cores, versus 4.1GHz on all 8 cores of the 2700x. We already know that Intel has an IPC advantage of maybe 8%. Overall, that's a near 10% advantage.
However, it's not the advantage that you'd be wanting for its price tag. That's where 24/7 overclocking comes to the fore for the 9900k, assuming you can cool it.
The 9900k has a 10% performance advantage at 10% lower power than the 2700x; the 2700x pretty much goes 24/7 at 105w versus 95w for the 9900k.
This lead evaporates with Zen 2, assuming the leaks turn out to be true; 13-28% IPC uplift, and 40-50% more power efficient. At that point only clockspeed will determine whether the next gen Ryzen CPU has equivalent or better overall performance than the 9900k; the 9900k being overclocked to its max.
The 2700X actually doesn't hold 105W 24/7 - in fact it can draw in excess of 140W, as shown by GN: https://www.gamersnexus.net/hwrevie...w-game-streaming-cpu-benchmarks-memory/page-3
Those power draw numbers look like PB0 is enabled. (The warranty killing built in overclock). On my 2600X, it holds to a strict 95 watts period, since I haven't enabled that feature. I may at some point, just for humor. But since I'm getting 4.2 across all cores I don't really see the point. That will likely drop in the summer as it will be quite a bit warmer in this room though.The 2700X actually doesn't hold 105W 24/7 - in fact it can draw in excess of 140W, as shown by GN: https://www.gamersnexus.net/hwrevie...w-game-streaming-cpu-benchmarks-memory/page-3
If we are to hold Intel to a certain standard, particularly after the 9900K situation, then we shouldnt allow AMD a free pass on 'misleading TDPs' as well. The 2700X power draw also freely exceeds its official TDP rating. Are mobo makers at fault for that as well? Why aren't there calls for the 2700X to be limited to 105W max power draw in reviews? In reality a '140W' 2700X is 10% slower than a '95W' 9900K.
Those power draw numbers look like PB0 is enabled. (The warranty killing built in overclock). On my 2600X, it holds to a strict 95 watts period, since I haven't enabled that feature. I may at some point, just for humor. But since I'm getting 4.2 across all cores I don't really see the point. That will likely drop in the summer as it will be quite a bit warmer in this room though.
Well, I guess it depends on testing methodology as well as who you believe. Anand has the 2700X just over 106, and Toms has it a hair lower.
Interestingly, in the 9900k review on Anandtech the power draw for the 2700X was listed higher around 117, but Tom's 9900k stayed the same for the 2700X. I guess they did not retest since their methodology looks the same.
The 2700X actually doesn't hold 105W 24/7 - in fact it can draw in excess of 140W, as shown by GN: https://www.gamersnexus.net/hwrevie...w-game-streaming-cpu-benchmarks-memory/page-3