Intel Core i9-9900K

[Vario]

In future reviews for the 9600K, I am interested to see how the 8600K thermals compares to the 9600K thermals at the same turbo multiplier. That would show the difference made by the thermal interface material.

 

[xorbe]

Something that hadn't exposed yet,that larger gap between base clocks and turbo clocks tend to make spikes in game.

No, this is well known. This is why you make sure your clock is stuck at max freq for gaming. Same as nvidia's prefer max performance so the clock doesn't accidentally drop during light loads, leading to crap framerates for a bit when the load increases. Good for laptop battery life, bad for performance desktops.

 

[R-T-B]

You mean (ST) performance, because if this is Intel's best "ever" then AMD's already won.

He meant what he said: IPC. AMD has not already won in that department. IPC is irrelevant to ST vs MT debates. It's measured on one core by it's nature.

 

[Prima.Vera]

So I'm guessing there is ZERO need to upgrade my god ol' 3770K CPU, especially since I am gaming on 3440 x 1440 @ 100Hz Monitor on a GTX 1080. Correct me if I am wrong, but just to see a frame increase of up to 2, 3%, I need ~1500$ in hardware upgrades (CPU+Cooler+Mobo+RAM) ; it's absolutely not worth it at all.

 

[trparky]

So I'm guessing there is ZERO need to upgrade my god ol' 3770K CPU, especially since I am gaming on 3440 x 1440 @ 100Hz Monitor on a GTX 1080. Correct me if I am wrong, but just to see a frame increase of up to 2, 3%, I need ~1500$ in hardware upgrades (CPU+Cooler+Mobo+RAM) ; it's absolutely not worth it at all.

I don't know about that, a friend of mine upgraded this older 4000-series Intel system to a new 8700K-based system and he's reporting to me amazing results. He's loving the fact that his 1080Ti is finally being used to its full potential, he never could get it to work at 100% which indicated that the processor was the bottleneck.

 

[Prima.Vera]

I don't know about that, a friend of mine upgraded this older 4000-series Intel system to a new 8700K-based system and he's reporting to me amazing results. He's loving the fact that his 1080Ti is finally being used to its full potential, he never could get it to work at 100% which indicated that the processor was the bottleneck.

If he's playing on 1080p, then yeah. But for 3440x1440p already the GPU it's the bottleneck. Just see the reviews from here or Anandtech....

 

[R0H1T]

He meant what he said: IPC. AMD has not already won in that department. IPC is irrelevant to ST vs MT debates. It's measured on one core by it's nature.

It also varies by application & task, for instance AES & SHA benchmarks is where AMD matches or beats (SHA) Intel regularly. ST is a function of IPC x clock speed, that's what I was referring to & if it weren't for Intel's super high clock speeds, there's no way Intel could price the 9900k anywhere near $500 or more, for HEDT.

 

[GoldenX]

It's not a problem of clock speed, it's their cost of producing that many great batches, and they chairmen not releasing that their products are now shi... obsolete.

 

[hat]

The 9900k is a bold move. Even as a last ditch effort of making the most of what they have (skabfee lake refresh on 14nm), though they do manage to stay on top as far as performance goes, it comes at an arrogantly high price and sucks down power like a v12 engine. It's like Intel's Fermi. They're at the end of the line now, though, and need to make some big changes.

 

[E-curbi]

I'm quoting from email the Siliconlottery.com guy.

"Got a few more 9900Ks in, it does look like they're going to clock better than 9700Ks this time.

Due to Intel hard-binning."

...end of quote.

Breaking News!

 

[HD64G]

Some benchmarks and comparisons of it with the 95TDP limit forced. Much performance loss leading to lamost 2700X levels. Intel (with the help of their board partners) cheating again as usual. Zen2 will demolish them in all market segments both in raw performance and in VFM me thinks.

 

[EarthDog]

Some benchmarks and comparisons of it with the 95TDP limit forced. Much performance loss leading to lamost 2700X levels. Intel (with the help of their board partners) cheating again as usual. Zen2 will demolish them in all market segments both in raw performance and in VFM me thinks.

https://www.anandtech.com/show/13544/why-intel-processors-draw-more-power-than-expected-tdp-turbo

 

[HD64G]

https://www.anandtech.com/show/13544/why-intel-processors-draw-more-power-than-expected-tdp-turbo

I know that BIOS settings are the cause of this controversy but in Ryzen CPU reviews none of the reviewers present results that have limits over the base BIOS settings as stock. That alone is the scandal in this case.