Intel Prepares Pre-Binned Core i9-12900KS Processors Clocked at 5.2 GHz

[ir_cow]

There are 12900K chips that hit 5.3 with ease under 1.3v - my 12600k does that as well. A binned chip won't need that much voltage to hit that clock.

I really wish they would move away from single core boost though. Almost all games now days use all your cores, so it would be nice if they boosted based on watts, not on core load.

Ocing to 5.3 and actually not crashing under load is two different things. I can boot 5.3Ghz as well, but it will lock up under stress test without 1.5v. We are talking all-core yes and not a single core turbo?

I see these same posts every generation. People set a overclock and think its stable because it doesn't crash in the game they are playing. That isn't OC stability. Give me photo proof of 5.3 All-core 5.3 @ 1.3 with Prime95.

 

[AlwaysHope]

Ocing to 5.3 and actually not crashing under load is two different things. I can boot 5.3Ghz as well, but it will lock up under stress test without 1.5v. We are talking all-core yes and not a single core turbo?

I see these same posts every generation. People set a overclock and think its stable because it doesn't crash in the game they are playing. That isn't OC stability. Give me photo proof of 5.3 All-core 5.3 @ 1.3 with Prime95.

P95 gets more complicated with AVX, & even AVX512, by default at least with rocket lake, it will down clock to offset the strain of that code. Then there is the question of time, how long to run it to prove "stability". That word stability can be subjective, is it stable enough for the end users needs? that is the ultimate question.

 

[Tartaros]

My opinion is basically that if you put a high end Intel and AMD systems next to each other with a same GPU, nobody would know the differences without an OSD/FPS counter.

And any sensible person would limit the frame rate to just a little below the screen refresh rate to avoid tearing and wasting energy on excess frames. Which, unless you are a VERY competitive CS:GO player avoiding any kind of sync tech, stretches the cases where you can tell even more.

 

[phanbuey]

Ocing to 5.3 and actually not crashing under load is two different things. I can boot 5.3Ghz as well, but it will lock up under stress test without 1.5v. We are talking all-core yes and not a single core turbo?

I see these same posts every generation. People set a overclock and think its stable because it doesn't crash in the game they are playing. That isn't OC stability. Give me photo proof of 5.3 All-core 5.3 @ 1.3 with Prime95.

I know how to oc, and yes I’ve stress tested the build with prime lol.

 

[ratirt]

I know how to oc, and yes I’ve stress tested the build with prime lol.

Your clocks must have dipped below 5Ghz for sure when stress testing. It is rather unusual what you say or you have some sort of a golden sample which I doubt.

 

[RandallFlagg]

I know how to oc, and yes I’ve stress tested the build with prime lol.

Alder Lake CPUs common discussion

For the time being, tight timing DDR4 out performs DDR5 in 99% of use-case-scenario's. That will of course change but not for another 18 to 24 months. Until then, DDR4 with tight timings is the smart choice. No. Just no.

www.techpowerup.com

 

[phanbuey]

I definitely got super lucky with this sample.

The two other 12600ks we oc’d topped out at 5.1ghz around 1.3-1.32 and the 12700k sample was an absolute dog and wouldn’t run stable at 5.0 ghz even at 1.35v so we just left it at stock.

Point is when the silicon is good these can clock like crazy, so it may not be massively different from the current one in terms of temps or wattage if they can hit those clocks at around same volts that the 4.9 ghz sample is using.

 

[chrcoluk]

extra few % performance but your pc in return also becomes a portable heater.

 

[Vayra86]

Agreed, I've been demanding it for years, but SOMEBODY has to do some research into how large a framerate difference has to be before a typical human being can actually distinguish the difference, e.g. At 142 FPS, a 5% increase would mean 149 FPS, and I highly doubt very many human beings can distinguish that difference, if at all....so it'd be nice to have an actual experiment prove it, so that when a new CPU comes out, we can say with confidence that unless it outperforms another CPU in framerates by at least 15%, it's pointless. It'd be nice to know that people bragging over a 6% average difference in framerates are just wasting their time....conclusively.

You can research perception. And we know that we all perceive things differently. So I think we already know this, and its folly trying to dumb that down to a simple number.

All things are relative, and perception improves with experience and knowledge. If you know 'what 120 fps looks and feels like' you will be more likely to notice it not running at that framerate.

The real question is, does it impact gaming, and for that, the biggest factor is not the FPS, but the frame time stability. Even a lowly 50 FPS can be perceived as perfectly smooth, in fact its sometimes even preferable because of 'cinematic experience' or to counteract motion blur.

Everything else is just epeen and marketing. The height of FPS is largely irrelevant above 50~60. And let's not forget input latency. Cool story, 60 FPS Vsynced... but not with 20ms delay on input. Another thing you see in games is that the simulation (game logic) is actually interleaved and runs at half (or fraction of) the FPS you get on screen. Space Engineers, for example.

 

[Chrispy_]

Alder Lake's performance is solid, but the extreme power consumption and heat dissipation requirements quickly make it unviable for the climate I live in.

Alder Lake is still competitive with Zen3 in terms of perf/Watt. @W1zzard's power scaling article shows that the 12900K hamstrung by a 125W PL2 limit is still getting most of the performance on offer.

The 12900K as a 24-thread solution capped at 125W is a very good match for the 5900X as a 24-thread solution (nominally 105W, max 142W). The reason the 12900K looks stupid in power consumption is because Intel literally doubled the consumption to try and catch the 5950X in multi-threaded workloads. Emphasis on try. At 125W the 12900K is great on power efficiency - better than its thread-count equal, the 5900X and not far off what a 5950X can achieve.

If you're a gamer then there's no need to even bother with higher PL2 on a 12900K or even a the i7 for that matter. 125W will get you 99.x percent of the performance for well under half the power consumption. You're just throwing silly motherboard costs, heat, noise, and power wastage at the last 0.5% with a 250W stock power budget on the 12900K, de-restricted it'll throw another 100W under the bus at least and you'll see no benefits that couldn't also be chalked up to measurement margins-of-error.

extra few % performance but your pc in return also becomes a portable heater.

that's what cryptomining rigs are for, silly

The height of FPS is largely irrelevant above 50~60. And let's not forget input latency. Cool story, 60 FPS Vsynced... but not with 20ms delay on input. Another thing you see in games is that the simulation (game logic) is actually interleaved and runs at half (or fraction of) the FPS you get on screen. Space Engineers, for example.

I have a laptop with a 100Hz display but no VRR and a puny 4700U's Vega7 so I'm quite frequently targeting 50fps with vsync on and let me tell you with 100% certainty that 50fps feels like ass these days. 60Hz is an absolute minimum for something to look even vaguely smooth at speed, but like you say - framerate and input latency stability are more important at those lower framerates, so 50fps is a safe target for an integrated graphics option as it stands a fighting chance of never dropping a frame like that.

It varies quite alarmingly on the how fast the content is moving and how much contrast there is. Sometimes 50fps can look great and smooth, other times 60fps can be very noticeably inadequate for the illusion of motion. On my high-refresh displays I can clearly see the difference between 90Hz and 120Hz, I'd genuinely struggle to spot the difference between 120, 144Hz, and 165Hz in a blind test so I'm guessing my point of diminishing returns is somewhere north of 90fps.

 

[Stag]

Guessing these will be rated 90 or higher bin.Maybe a ring boost.

 

[Dr. Dro]

Alder Lake is still competitive with Zen3 in terms of perf/Watt. @W1zzard's power scaling article shows that the 12900K hamstrung by a 125W PL2 limit is still getting most of the performance on offer.

The 12900K as a 24-thread solution capped at 125W is a very good match for the 5900X as a 24-thread solution (nominally 105W, max 142W). The reason the 12900K looks stupid in power consumption is because Intel literally doubled the consumption to try and catch the 5950X in multi-threaded workloads. Emphasis on try. At 125W the 12900K is great on power efficiency - better than its thread-count equal, the 5900X and not far off what a 5950X can achieve.

If you're a gamer then there's no need to even bother with higher PL2 on a 12900K or even a the i7 for that matter. 125W will get you 99.x percent of the performance for well under half the power consumption. You're just throwing silly motherboard costs, heat, noise, and power wastage at the last 0.5% with a 250W stock power budget on the 12900K, de-restricted it'll throw another 100W under the bus at least and you'll see no benefits that couldn't also be chalked up to measurement margins-of-error.

I don't disagree, it's mostly that the simple fact that your logic is reasonable that makes this product (an eventual 12900KS) entirely unreasonable

To be entirely fair? Even the regular K processors released so far are not a must-have this time around, since B660 and locked processors will sustain PL1=PL2 for an indefinite amount of time, an i9-12900 processor seems like the one I would personally pick amongst these.

 

[cadaveca]

I'm still impressed they push so much power through such a small piece of glass.