Intel Core i5-12600K

[WhoDecidedThat]

@W1zzard What do you think about using a separate PSU when doing CPU tests to see how much only the CPU is consuming?

LTT did it in this video but I'm not sure if it's an effective method.

Do it! This sounds awesome!

 

[phanbuey]

12600K and MSI mobo on the way.,.. cant wait to tune some Gear 1 ram settings and run some benches.

I have a disease...

 

[The red spirit]

It's nice to see Team Blue back on top of the mountain again but tbh I'm looking forward to the locked cpu's w/B660 boards and DDR4.

Only in power consumption it's on top of mountain.

 

[Why_Me]

Only in power consumption it's on top of mountain.

Obviously you haven't looked at the benchmarks.

 

[The red spirit]

Obviously you haven't looked at the benchmarks.

I have, horrendous power consumption with tiny edge over competition. It's literally the same as overclocking FX 9590 and saying that it's faster than i7 2600K, while ignoring horrendous power consumption and heat output.

 

[Why_Me]

I have, horrendous power consumption with tiny edge over competition. It's literally the same as overclocking FX 9590 and saying that it's faster than i7 2600K, while ignoring horrendous power consumption and heat output.

Thankfully I live in a country with electricity. With that said Alder Lake is at the top of the mountain atm and anybody who says otherwise either didn't look at the benches or is straight up lying.

 

[B-Real]

"Intel Core i5-12600K 47% Faster Than Ryzen 5 5600X in Leaked CPU-Z Benchmark"

Real life circumstances: 16% faster with 4 extra cores.

 

[The red spirit]

Thankfully I live in a country with electricity. With that said Alder Lake is at the top of the mountain atm and anybody who says otherwise either didn't look at the benches or is straight up lying.

Benches say: STAY AWAY, IT'S GONNA BURN DOWN YOUR HOUSE.

 

[B-Real]

Thankfully I live in a country with electricity. With that said Alder Lake is at the top of the mountain atm and anybody who says otherwise either didn't look at the benches or is straight up lying.

Yes. It is an undoubtedly faster than a 1 year old CPU. With 4 extra cores... What an achievement, really!

 

[Why_Me]

Yes. It is an undoubtedly faster than a 1 year old CPU. With 4 extra cores... What an achievement, really!

What line-up of AMD cpu's was Alder Lake supposed to be compared to?

 

[R-T-B]

I will also have to check the support on Linux.

Thread Director presently has zero linux support.

 

[HenrySomeone]

I have, horrendous power consumption with tiny edge over competition. It's literally the same as overclocking FX 9590 and saying that it's faster than i7 2600K, while ignoring horrendous power consumption and heat output.

Yeah....no. You'd burn through your motherboard OC-ing it and it still wouldn't match a perhaps slightly OCed 2600k at stock voltage and going the other way, downclocking it to Sandy Bridge level of power use, you got about q6600 level of performance. Alder Lake on the other hand, while indeed consuming a fair bit of juice while under all-core loads, can outdo even the (more expensive) 5950x and when downtuned a bit, will still healthily beat 5900x at the same power or match it at about 5800x's, so no comparison at all. Oh, and just to drive the point from the first sentence home:
MSI AMD 970 Krait with a 9590 on fire - YouTube

 

[B-Real]

What line-up of AMD cpu's was Alder Lake supposed to be compared to?

Intel was preparing for this new architecture like crazy. What they achieved is out of the 3 models, only 1 is able to beat its Zen 3 rival. The other 2 are 4-5% slower than the Zen 3 models because they have the same amount of cores. The i5 has 67% more cores and can outpower the 5600X by 16%. And this is with DDR5 memory compared to DDR4, so it's possible their will be a bigger loss for i7 and i9 and a smaller win for i5. Not to mention that all 3 Intel models have worse efficiency than the Zen 3 models. And Zen 4 is yet to come. I have no idea why you are that optimistic. And BTW, your "With that said Alder Lake is at the top of the mountain atm" is basically NOT TRUE. Regarding CPU power, the 5950X beats the i9 while the 5900X beats the i7. So in terms what AL is at the top of the mountain? Because it's 7% faster in FHD with an RTX 3080 which no one will use in real life circumstances?

 

[Turmania]

Isn't this wonderfull, now it's your turn AMD, consumers wins. . This is what I want to see, they outdoing each release...

 

[HenrySomeone]

"Intel Core i5-12600K 47% Faster Than Ryzen 5 5600X in Leaked CPU-Z Benchmark"

Real life circumstances: 16% faster with 4 extra cores.

You dirty little liar! Real life circumstances - 58% faster and that's in AMD's favorite benchmark of them all - Cinebench multi thread!

 

[B-Real]

You dirty little liar! Real life circumstances - 58% faster and that's in AMD's favorite benchmark of them all - Cinebench multi thread!

What I see is an overall 16% lead over the 5600X with all the results averaged. What do You see?

and when downtuned a bit, will still healthily beat 5900x at the same power or match it at about 5800x's

This is really funny tbh. Hope you was writing the same regarding Vega 56.

 

[HenrySomeone]

The leaked CPU-z benchmark (that you of course quoted) was multi-thread and in that regard it actually outdoes it by another 10% and if you can't understand (or rather pretend you don't because it doesn't suit your red agenda) why an aggregate of all tests can never be so high (not even 5950x is 50% ahead of 5600x), we can have no further discussion.
And oh, Vega was crap that couldn't even compete against 1080, never mind the 1080Ti, both of which came out before (way before in the case of the former). You could tune the 56 to about 1070 power and performance, but that one was much cheaper, so what was the point in that? Either way, the market had it's say and the fact that they never achieved any penetration is saying enough.

 

[The red spirit]

Yeah....no. You'd burn through your motherboard OC-ing it and it still wouldn't match a perhaps slightly OCed 2600k at stock voltage and going the other way, downclocking it to Sandy Bridge level of power use, you got about q6600 level of performance. Alder Lake on the other hand, while indeed consuming a fair bit of juice while under all-core loads, can outdo even the (more expensive) 5950x and when downtuned a bit, will still healthily beat 5900x at the same power or match it at about 5800x's, so no comparison at all. Oh, and just to drive the point from the first sentence home:
MSI AMD 970 Krait with a 9590 on fire - YouTube

It wouldn't beat any Ryzen at power normalized benches. It manages to get 10% advantage over Ryzen with over 300 watts. Ryzen's turbo (for X parts) is a little bit over 100 watts. Sure power usage and clock speed scaling is non linear (it's exponential), but still, i9 would (and i5) would need a lot of detuning and as result, performance would suffer.

 

[HenrySomeone]

It (handily!) beats 5950x! It will easily outdo the 5900x at the same power!

 

[regs]

Would be also interesting to see Gracemont cores performance in comparison to Skylake - single thread, multi thread and single thread per same clock. Is it possible to switch off P-cores instead to use E-cores only, so they can be tested?

 

[W1zzard]

Is it possible to switch off P-cores instead to use E-cores only, so they can be tested?

The official answer is "at least 1 p core must be active". Working around that is on my list for next week

 

[The red spirit]

It (handily!) beats 5950x! It will easily outdo the 5900x at the same power!

That's just 2000 point difference. IPC of Alder lake sucks. You will need 1GHz reduction at minimum to start looking at similar power consumption to 5950X and that will mean performance way worse than 5950X.

 

[HenrySomeone]

Read again what I said, carefully this time.

 

[The red spirit]

Read again what I said, carefully this time.

I read everything properly last time. I simply doubt what you say, that's all.

 

[Why_Me]

Intel was preparing for this new architecture like crazy. What they achieved is out of the 3 models, only 1 is able to beat its Zen 3 rival. The other 2 are 4-5% slower than the Zen 3 models because they have the same amount of cores. The i5 has 67% more cores and can outpower the 5600X by 16%. And this is with DDR5 memory compared to DDR4, so it's possible their will be a bigger loss for i7 and i9 and a smaller win for i5. Not to mention that all 3 Intel models have worse efficiency than the Zen 3 models. And Zen 4 is yet to come. I have no idea why you are that optimistic. And BTW, your "With that said Alder Lake is at the top of the mountain atm" is basically NOT TRUE. Regarding CPU power, the 5950X beats the i9 while the 5900X beats the i7. So in terms what AL is at the top of the mountain? Because it's 7% faster in FHD with an RTX 3080 which no one will use in real life circumstances?