Intel Core i9-11900K

[oxrufiioxo]

Definitely would rather have a 10850K. Lower, but sustained clock speeds are much more useful than sudden spikes with unmanageable power consumption and heat.

I'd want it more for tweaking purposes and what it cost is irrelevant to me. When it was going for $460 it was tempting..... I already had a spare high end x570 board so it won out though.

 

[AusWolf]

I'd want it more for tweaking purposes and what it cost is irrelevant to me. When it was going for $460 it was tempting..... I already had a spare high end x570 board so it won out though.

It's not about the cost for me, either. What I mean is, what's the point of a +100 MHz Total Bull**** Boost (or whatever it's called) if the CPU drops 500 MHz after half a minute of work anyway?

 

[oxrufiioxo]

It's not about the cost for me, either. What I mean is, what's the point of a +100 MHz Total Bull**** Boost (or whatever it's called) if the CPU drops 500 MHz after half a minute of work anyway?

The difference is more pronounced once tweaked while you can get a dud with either the 10900k seem to clock 2-300Mhz all core higher with high end cooling.... Some of it is OCD the number is 8 and not 9 on the Name lol..... Even my current 5800X is a placeholder cpu for a 5950X

Total respect people opting for the 10850K such a good cpu when its under $399 especially for those with already good enough Z490 boards.

 

[ThrashZone]

Definitely would rather have a 10850K. Lower, but sustained clock speeds are much more useful than sudden spikes with unmanageable power consumption and heat.


Or just ashamed of pointlessly wasting so much money.

Hi,
Doubt that he has deep pockets besides this 11900k/... is chicken feed compared to 3090 which he also has I believe.

 

[W1zzard]

@W1zzard could you please be so kind as to provide the scene project you've used for UE4 lightmass testing and explain what source you've built?
11900K, at any frequency, shouldn't be as close to 5950X by any stretch. Tests from any other tester also don't show it to be the same.
Since I'm working with UE4 professionally I'm really, really interested in this.
How good per-core performance is compared to 5950X? How good total multicore is? Since also, UE4 scales perfectly with the cores too.
Judging by all tests of the single-core performance it seems that Intel hasn't caught up with series 5000 from AMD so even if they did, 16-cores should be 60% faster when running MT tasks?
Puget systems say 11900K is significantly behind. But they did a very strange thing of testing on a Z490 board which might have serious implications with the most advanced boost not being active and memory performance.
Could you be so kind to test Github available source compilation, 4.26.1 or 4.27.0 (dev-core branch)?
Thanks.

I'm not building UE from source, I'm building lighting, SubwaySequencer_P

Sorry for the slow response, I missed your original post

 

[Maissilapsi]

That's part of it :

View attachment 194642

Houston, we have BIOS Problems: The Rocket Lake [P]review

With the promise of up to 19% increased IPC, many Intel users – including myself – were excited to see what Rocket Lake might bring. Did Intel deliver? Keep reading.

adoredtv.com

Memory setup is critical. Computerbase.de talks about this, but the 11900K they had maxed out a 3733 Gear 1. There was about a 4% FPS difference in games with DDR4-2933 Gear 1 vs DDR4-3200 Gear 2.

So I think a lot of these reviews are giving different results due to a lack of experience of the reviewers with the new chips, how to configure their setups for them, and BIOS version issues. Just look at that CS:GO chart. That's over a 40% difference in fps from the BIOS, they concluded that something in the BIOS was greatly affecting memory latency.

This, it's very much a RAM thing too...

I just built for friends kid PC on XIII Hero and i9-11900K and kind of wondered the testing results here... Main difference was I put 16GB DDR4 5133MHz dual RAM on it. It is a fine CPU, with it's faults... and I've been AMD loyalist since a 5950X landed on the palm of my hand.

I'm not so sure are the AVX-512 things necessary though, die space is limited and there's other parts to do the vector calculations much faster, unless you want fancy benchmark numbers or something...

Comparing CPU BM scores this begins to over take my 5950X after about 8-12 threads, for it only goes 5.1GHz while this i9 5.3GHz, and beyond even...

Which is kind of wild... I've had and still do actually in closet one i5-9600K (6c/6t) on XI Gene that I got stable on 5,3MHz with 4800MHz Royals underclocked to some 4133-4333MHz. Alder Lake actually further proves Intel is taking Zen 3 seriously.


I also must add they used "Zadak Spark 240 mm AIO" to cool and 14nm phat Intel...So AIO and 2 × 120mm fans..

PS. To further carry the point home here, the cores are alright... as seen on pic here.

Carry on...