Alder Lake CPUs common discussion

[fevgatos]

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.

 

[RandallFlagg]

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.

The early DDR4 reviews were done mostly with DDR4-4800 and DDR4-5200, and I also bought into that line for a bit, but the data on those early takes was incomplete. In particular this line of thought comes from places like AnandTech, which I've repeatedly noted does not reflect what an enthusiast will actually use in either hardware or software - think using DDR4-2933 C20 on an Apex motherboard. It's not even reflective of what an OEM system will perform like. It then gets regurgitated everywhere.

So you might want to check TPU's DDR5-6000 vs DDR4 article. Or the video I linked to. Higher speed (6000+) XMP DDR5 is outperforming the highest end DDR4. And - (edit lower end DDR5 is being tuned to get results unobtainable on DDR4 rigs even when tuned.
None of that means that DDR4 isn't the best *deal*, but high end DDR5 is demonstrably outperforming high end DDR4 now.

 

[lexluthermiester]

The early DDR4 reviews were done mostly with DDR4-4800 and DDR4-5200

I'm talking about DDR4 3600 with low timings.

and I also bought into that line for a bit, but the data on those early takes was incomplete.

Sorry, the following was not incomplete;

DDR4 vs. DDR5 on Intel Core i9-12900K Alder Lake Review

The Intel Alder Lake platform has support for both DDR5 and DDR4 memory. We ran 38 application benchmarks and 10 games at multiple DDR4 configurations to learn what performance to expect when using DDR4 vs. DDR5 on 12th Gen, and whether there's a point at which DDR4 performance can beat the much...

www.techpowerup.com

This testing showed very clearly that currently, DDR4 has the performance advantage.

but high end DDR5 is demonstrably outperforming high end DDR4 now.

Sorry, that is just not correct.

 

[RandallFlagg]

I'm talking about DDR4 3600 with low timings.

Sorry, the following was not incomplete;

DDR4 vs. DDR5 on Intel Core i9-12900K Alder Lake Review

The Intel Alder Lake platform has support for both DDR5 and DDR4 memory. We ran 38 application benchmarks and 10 games at multiple DDR4 configurations to learn what performance to expect when using DDR4 vs. DDR5 on 12th Gen, and whether there's a point at which DDR4 performance can beat the much...

www.techpowerup.com

This testing showed very clearly that currently, DDR4 has the performance advantage.

Sorry, that is just not correct.

Are we looking at the same review? There's actually no major sub-category where DDR4 wins vs a DDR5-6000 kit on TPU's review.

 

[lexluthermiester]

Are we looking at the same review? There's actually no major sub-category where DDR4 wins vs a DDR5-6000 kit on TPU's review.

View attachment 225529

View attachment 225530

It would seem we are interpreting the results differently.

 

[oxrufiioxo]

So a 800 usd kit of ram beats a $hit 3600 kit by 1% in gaming a wins a win I guess......

 

[RandallFlagg]

IDK I just found out according to cpu-z i own all RAM.

DDR4-20500 C18 FTW baby. 26 Ghz ring too.

 

[SuperMumrik]

Tuned ddr4 4000c15 gear 1 beats tuned micron ddr5 ic's in gaming, but I highly doubt that you can beat tuned 6400c28 with any d4 config.

 

[lexluthermiester]

but I highly doubt that you can beat tuned 6400c28 with any d4 config.

Yeah try buying some of that. Not gonna happen for a year or more.

 

[SuperMumrik]

Yeah buying some of that. Not gonna happen for a year or more.

You can tweak random Hynix 4800Mhz ic's there with some luck in the binning. But ofc those are a nightmare to get atm

 

[Psychoholic]

for anyone wondering.. these are the corsair mounting screws for LGA1200 vs LGA1700
LGA1200 on the Left.

 

[phanbuey]

Tuned ddr4 4000c15 gear 1 beats tuned micron ddr5 ic's in gaming, but I highly doubt that you can beat tuned 6400c28 with any d4 config.

1. Great avatar. and 2. totally agree

 

[RandallFlagg]

More news / leaks on the 12400 :

Intel Core i5-12400 6 Core CPU Matches The AMD Ryzen 5 5600X In PugetBench Benchmarks, Little i5 To Destroy Top Ryzen 5 In Perf Per $

Benchmarks of Intel Core i5-12400 Alder Lake CPU have leaked out and show great performance per dollar against the AMD Ryzen 5 5600X.

wccftech.com

 

[tabascosauz]

for anyone wondering.. these are the corsair mounting screws for LGA1200 vs LGA1700
LGA1200 on the Left.

Which is why I was going to say that Asus including the LGA1200 holes may not work perfectly, there's also a z-height difference between LGA1200 and LGA1700 package that a lot of Asus owners seem to have glossed over.

Noctua's page mentions as well:

 

[RandallFlagg]

Alder Lake - P, H which is probably 35/45W. So far this appears to be an absolute beast for mobile.

I may just go back to laptop once these release since desktop GPUs cost as much as a laptop with a GPU. These scores put the mobile 12700H in the same multi-core performance area as a 3900X and 12600K, and above any Zen 3 in single core (slightly below AL desktop K series), squashes the 5980HX mobile part which is AMDs current top line Zen 3 mobile.

Intel Core i7-12700H outperforms Ryzen 9 5900HX by 47% in leaked Cinebench benchmarks - VideoCardz.com

Intel Core i7-12700H in Cinebench Notebookcheck gained access to information on the unreleased Intel Alder Lake-P laptop CPU. The website shares Cinebench R20 and R23 scores featuring unreleased Core i7-12700H Alder Lake-P CPU. This particular processor is equipped with 14 cores and 20 threads...

videocardz.com

 

[The red spirit]

More news / leaks on the 12400 :

Intel Core i5-12400 6 Core CPU Matches The AMD Ryzen 5 5600X In PugetBench Benchmarks, Little i5 To Destroy Top Ryzen 5 In Perf Per $

Benchmarks of Intel Core i5-12400 Alder Lake CPU have leaked out and show great performance per dollar against the AMD Ryzen 5 5600X.

wccftech.com

A bit of shame that it doesn't have any E cores at all and that it has base clock of just 2.5 GHz. At least it has respectable boost. BTW so far it is slower than 5600X, which is also worrying. Anyway, does not having E cores, mean that Denuvo works?

 

[lexluthermiester]

Who cares if Denuvo works? That crap needs to die, like all DRM...

 

[phanbuey]

Final 24/7 settings on the 12600K - settled in at ~5.3 ghz

Ok time to stop benching and play some games.

 

[Outback Bronze]

Final 24/7 settings on the 12600K - settled in at ~5.3 ghz

What's your temps like?

 

[phanbuey]

What's your temps like?

80c max on cores and 75c on package ring during timespy / heavy load. 60-63c during gaming. Touches 90c during cinebench r23 after about 5-6mins I have the mobo throttle it at 90.

 

[gerardfraser]

I also have settled on a daily setting on 12900K.
5400Mhz P-Core
4000Mhz E-Core
4100Mhz Ring

Forza Horizon 5 12900K 5400Mhz 4K HDR PC Gameplay CPU Voltage,Temperature OSD Shown​

 

[birdie]

Some Interesting findings from a german forum.

"Setting the TDP to 175-200W appears to the sweet, losing just under 5% performance for an efficiency gain of ~30%. At 175W, the Core i9-12900K offers 95% of stock performance while drawing 75W less power, gaining an efficiency advantage of 35%. The temperature also drops by 20 degrees, allowing for prolonged and more consistent boosts."

12900K can actually be forced to use at most 25 or even ... 2W of power. How is that even possible if it idles around 6?

 

[phanbuey]

Some Interesting findings from a german forum.

"Setting the TDP to 175-200W appears to the sweet, losing just under 5% performance for an efficiency gain of ~30%. At 175W, the Core i9-12900K offers 95% of stock performance while drawing 75W less power, gaining an efficiency advantage of 35%. The temperature also drops by 20 degrees, allowing for prolonged and more consistent boosts."

View attachment 225980

View attachment 225981

12900K can actually be forced to use at most 25 or even ... 2W of power. How is that even possible if it idles around 6?

That's crazy, CPU Z is not even an intense stress test. to hit 100C on that means you instathrottle on Cinebench/Prime.




Guys PSA - if you're running 4 dimms on alder lake turn around timings on alder lake are MASSIVE in performance impact in gear 1.

Going to play a bit more but this is with tRDRD_SG/DG/DR/DD @ 7/4/77 -- 16,445 on timespy up from ~15,900-16,000
SOTTR went from 281 to 291 on avg with higher mins.

 

[oxrufiioxo]

That's crazy, CPU Z is not even an intense stress test. to hit 100C on that means you instathrottle on Cinebench/Prime.




Guys PSA - if you're running 4 dimms on alder lake turn around timings on alder lake are MASSIVE in performance impact in gear 1.

View attachment 226080

Going to play a bit more but this is with tRDRD_SG/DG/DR/DD @ 7/4/77 -- 16,445 on timespy up from ~15,900-16,000
SOTTR went from 281 to 291 on avg with higher mins.

I think I've seen the 12900k over 360fps in SOTTR benchmark 5950X seems to cap out around 350fps. Definitely need really good dual rank ram with tight timings though.

 

[phanbuey]

I think I've seen the 12900k over 360fps in SOTTR benchmark 5950X seems to cap out around 350fps. Definitely need really good dual rank ram with tight timings though.

Yeah the 12900k has the extra cache and 2 more cores that SOTTR loves. It does go up to 400 for some areas fine so I think the cap is hardware bound. The difference between builds is so huge ~ +70 fps avg with the steam update :/ to v453.0_64 I just use it to compare to my own scores to see if my tinkering is having a positive or negative effect lol.