Wednesday, January 1st 2020

Core i9-10900K up to 30% Faster than i9-9900K: Intel

Intel's upcoming Core i9-10900K desktop processor is up to 30 percent faster than the Core i9-9900K according to the company, which put out a performance guidance slide that got leaked to the web. Based on the 14 nm "Comet Lake-S" silicon and built for the new LGA1200 platform (Intel 400-series chipset motherboards); the i9-10900K is a 10-core/20-thread processor that leverages increased TDP headroom of 125 W to sustain higher clock-speeds than 9th generation "Coffee Lake Refresh," while also offering a 25% increase in processing muscle over the i9-9900K, thanks to the two additional CPU cores.

In its performance guidance slide, Intel shows the i9-10900K scoring 30% more than the i9-9900K in SPECint_rate_base2006_IC16.0. There's also a 25% boost in floating-point performance, in SPECfp_rate_base2006_IC16.0, which roughly aligns with the additional core count, as both these tests are multi-threaded. Other noteworthy results include a 26% gain in Cinebench R15, and 10% in SYSMark 2014 SE. In tests that don't scale with cores, Intel appears to rely entirely on the increased clock-speeds and improved boosting algorithm to eke out performance gains in the low-to-mid single-digit percentages. Intel is introducing a new clock-speed boosting technology called Thermal Velocity Boost, which can dial up clock-speeds of the i9-10900K up to 5.30 GHz.
Sources: ITHome, Tom's Hardware
Add your own comment

143 Comments on Core i9-10900K up to 30% Faster than i9-9900K: Intel

#101
HugsNotDrugs
biffzinker
Intel should have added on four more cores for at least twelve cores instead of a measly two core bump.

Intel already looks be in second place compared to AMD's twelve, and sixteen core chips.
It's very costly to expand the size of an already-large monolithic die.
Posted on Reply
#102
Metroid
almost 30% more cores, certainly 30% faster in multicore apps, now if is 30% in single thread then I shut off my mouth, which I'm sure is not the case.
Posted on Reply
#103
ppn
Metroid
almost 30% more cores, certainly 30% faster in multicore apps, now if is 30% in single thread then I shut off my mouth, which I'm sure is not the case.
meteor lake in 2021-22. ST+30% sunny cove +50% faster than SKL. what we have to do now is wait and boycott this 14++ ridiculous refresh cycle.
Posted on Reply
#104
londiste
HugsNotDrugs
It's very costly to expand the size of an already-large monolithic die.
In this case I think adding two cores would be doable without any major issues. Running them at competitive clock speeds, however... :D
Posted on Reply
#105
trparky
londiste
In this case I think adding two cores would be doable without any major issues.
Their yields must be hideous.
Posted on Reply
#106
londiste
trparky
Their yields must be hideous.
Why? A 10-core 10900K die would be about 200mm^2. This is roughly the same size as Zen/Zen+ dies on the same type of mature manufacturing process.

When we are talking about monolithic, we are talking about many cores and large die sizes. Intel's 28-core is difficult/expensive enough as monolithic. To some degree, so is 18-core die. 10 cores and 200mm^2 or so is very much fine.
Posted on Reply
#107
efikkan
As londiste was saying, ~200mm² is not a problem. Comet Lake is likely to have the same "bottlenecks" as Coffee Lake, where the limit will be energy density. 125W (and ~185W peak) on ~200mm² is the issue, not yields. 200mm² is still a "small" chip. If we're talking >500mm², then yields become a greater concern.
Posted on Reply
#108
biffzinker
efikkan
As londiste was saying, ~200mm² is not a problem. Comet Lake is likely to have the same "bottlenecks" as Coffee Lake, where the limit will be energy density. 125W (and ~185W peak) on ~200mm² is the issue, not yields. 200mm² is still a "small" chip. If we're talking >500mm², then yields become a greater concern.
Intel's problem being stuck on 14nm++ is how many dies per wafer they can get. Bigger die means less yield per wafer.
Posted on Reply
#109
efikkan
biffzinker
Intel's problem being stuck on 14nm++ is how many dies per wafer they can get. Bigger die means less yield per wafer.
As I said; yields is not an issue for such small dies. Yields on 14nm++ is excellent. As we can see on their bigger dies, they have to sacrifice clock speeds due to heat and power long before yields become an issue.
Posted on Reply
#110
trparky
londiste
Why? A 10-core 10900K die would be about 200mm^2. This is roughly the same size as Zen/Zen+ dies on the same type of mature manufacturing process.

When we are talking about monolithic, we are talking about many cores and large die sizes. Intel's 28-core is difficult/expensive enough as monolithic. To some degree, so is 18-core die. 10 cores and 200mm^2 or so is very much fine.
Yeah but I have to wonder how many of these 10-core chips can still be 10-core chips and not be down-binned to lesser models like i7s, i5s, and even worse... i3s.
Posted on Reply
#111
efikkan
trparky
Yeah but I have to wonder how many of these 10-core chips can still be 10-core chips and not be down-binned to lesser models like i7s, i5s, and even worse... i3s.
Good enough that Intel plans to use the 10-core die for only 10 and 8 core models, and 6-core die for 6 core models and down.
Posted on Reply
#112
trparky
efikkan
Good enough that Intel plans to use the 10-core die for only 10 and 8 core models
I guess we'll see based upon how much Intel prices these things. If they're priced at some ungodly high amount then we'll know how (un)successful they are at making these things.
Posted on Reply
#113
DeathtoGnomes
Renald
What a joke.
actually its not, Intel is known to do such things.
Posted on Reply
#114
Berfs1
*adds 25% more cores* (1.25x)
*adds 6% more clock speed* (1.06x)
*acts surprised that there is a 30% improvement in multithreaded is huge when 1.25x1.06 is 1.325 and with losses is ~30% gain so IPC actually is not improved*
“125W TDP.” 125W my ass.

It’s funny how intel is the one that said adding more cores is bad, yet they GLORIFY this for their 10900K. Best joke of the new decade.

efikkan
As I said; yields is not an issue for such small dies. Yields on 14nm++ is excellent. As we can see on their bigger dies, they have to sacrifice clock speeds due to heat and power long before yields become an issue.
Intel totally didn’t have any yield issues with their 9th generation.
Posted on Reply
#115
Tsukiyomi91
Rather wait for reviewers to compare this chip against the R9 3900X & possibly Ryzen 4th gen (when it comes out later this year). Hoping this year Intel is taking things seriously...
Posted on Reply
#116
Berfs1
Yo, Intel finally had the balls to come out and mention the actual power limit of the 9900K, 210W! Bravo Intel! So that means the 10900K is rated to do 250W. Ok. Consumer chips, doing 250W, no biggie. It's not like we won't have liquid nitrogen AIO coolers in 2020.
Posted on Reply
#117
biffzinker
Berfs1
So that means the 10900K is rated to do 250W. Ok.
With or without AVX2 though?
Posted on Reply
#118
Pap1er
john_
Yes, even in gaming. With an AMD you lose how much? From 0%-10% in 1080p? Up to 5% in 1440p? Up to 2% in 4K? And those differences with a 2080Ti probably.

So you go out and buy an Intel processor because reviews show you that the top Intel CPU, under the best cooling solution, on an expensive high end motherboard in a case with plenty of air flow, while paired with an RTX 2080 Ti is 5%-10% on average faster than the 3900X at 1080p? And that's while ignoring power consumption because power consumption is important only when AMD CPUs are less efficient.

What about a typical system with a mid range motherboard, mid range air cooling, probably an RTX 2070S or 5700XT at best, in a case with mediocre air flow running games at 1440p? What is the difference in games there? And can you really see it and feel it in games?


Yes they will see it and they will say "10 cores much faster than 8 cores"...... IN EVERYTHING. ;)
you nailed it !
Posted on Reply
#119
phanbuey
cant wait for my 4700x from amd :P
Posted on Reply
#120
Renald
DeathtoGnomes
actually its not, Intel is known to do such things.
I know, and that's why these numbers are ridiculous !

Very often they do it in "secret". This time it's "We are bullshiting you, it's just for shareholders"
Posted on Reply
#121
mechtech
Sysmark overall says 1.10x, but the wattage says 125W/250W over the i9s 95W/210W. Wattage % increase outdoes performance, is this cause on 14nm or pushing it too far out of the happy curve?
Posted on Reply
#122
cygnus_1
Oh boy, 33% more power usage and 25% more cores... and it ONLY gets 30% better performance??
Posted on Reply
#123
Steevo
Now 30% faster with 30% more cores 30% more power consumption and 30% more winter heating!
Posted on Reply
#124
Tsukiyomi91
funny replies aside, I wanna know does Intel's claim can be legit when it comes out? what if the 30% gain is with all the hardware mitigations & AVX-adjusted clocks thrown into the mix? Will it be at the heels of AMD's R7 & R9 when it comes to heavy workloads? Will Intel finally add more PCIe lanes for their mainstream platform? All these questions; I want it to be answered.
Posted on Reply
#125
Aerpoweron
My biggest issue with the 10th gen CPUs for LGA 1200 is, that they don't offer any advantage over the 9th gen in PCI-E lanes or PCI-E version. They have Wifi 6 included, but to use that you have to upgrade your Wifi router at least.
If they would release the 10 core CPU on LGA 1151 that would be ok. But a new platform for 2 more cores?

Come on Intel, you can do better than this.
Posted on Reply
Add your own comment