Question Alder Lake - Official Thread

[Hulk]

With the release of Alder Lake less than a week away and the "Lakes" thread having turned into a nightmare to navigate I thought it might be a good time to start a discussion thread solely for Alder Lake.

 

[Zucker2k]

There are like 50 reviews. Where to read for tests with:

• DDR4, but faster (>= 4000 MHz)
• DDR5 manual overclock (not just XMP)
• Ring/LLC frequency adjustments
• i5-12600K overclock

I don't know about all that but TPU has the 12600k review with overclocking results: https://www.techpowerup.com/review/intel-core-i5-12600k-alder-lake-12th-gen/

 

[Hitman928]

And other sites have exactly the opposite situation..

Intel Core i9-12900K, i7-12700K & i5-12600K im Test: Leistungsaufnahme, Temperaturen und OC

Intel Alder Lake im Test: Leistungsaufnahme, Temperaturen und OC / Leistungsaufnahme in Spielen / Leistungsaufnahme in Anwendungen

www-computerbase-de.translate.goog

Tha numbers the other igor put on his review seems quite off.

Computerbase has a much wider range of games tested. If you pick through their results, there are several which show similar results to Igor's but many that shown Zen 3 using less power during gaming as well. So there's some give and take depending on the game. (Edit: This shouldn't be too surprising as most games still only use 4 - 6 cores so CPUs like the 12900k and 5950x are much closer in frequency in these situations than during all core loads where the 12900k allows bonkers power numbers to sustain high frequencies on all cores. Also, a larger sample size is always preferred and this example shows why). Either way, I'm still going with, who really cares about 1 - 2 dozen watts when the whole system is consuming 500+W for gaming?

 

[blckgrffn]

Overall a very impressive jump for Intel. Not really anything earth shattering or unexpected when you look at the TGL cores but the E cores are more beneficial in general than I was expecting. Of course, power use is also more than I was expecting, Intel really wanted to try to take the all out performance crown and they didn't care who's house they burn down to do it . I would really like to see a review where PL1=125W for a 12900k, hopefully these come with time. I expect that although performance will obviously be less, the overall impression may be better when the power isn't being blown out to try and beat AMD at all costs. Intel is finally back in the game, which is exciting. I'm also excited to see how AMD can respond with Zen3d but especially what they have cooking up with Zen4.

Biggest caveat for me is the occasional scheduling issues with the hybrid setup. Hopefully they can improve that quickly as large, seemingly random, and unfixable by the user slowdowns kind of suck.

It's going to be interesting to see what you get when you try to use a ~$100 board with a 12600K in a "value" build in the same way you might put together a 5600x type build :/ With 8th/9th gen Intel you could still get a huge portion of their performance even on lesser boards. 10th & 11th gen... meh.

If you need to use a $200+ (I mean, minimum!) board to get the type of performance that actually beats a Zen 3 CPU as shown in these reviews that sorta muddies the waters for me.

I've gotten used to getting really solid board for ~$150 if I shop a little. I am talking with WiFi and cool ARGB and all that.

Gotta save the pennies for the GPUs, you know!

 

[clemsyn]

Any benchmark on the IGPU of the 12900K, 12700K and 12600K?

 

[Hitman928]

It's going to be interesting to see what you get when you try to use a ~$100 board with a 12600K in a "value" build in the same way you might put together a 5600x type build :/ With 8th/9th gen Intel you could still get a huge portion of their performance even on lesser boards. 10th & 11th gen... meh.

If you need to use a $200+ (I mean, minimum!) board to get the type of performance that actually beats a Zen 3 CPU as shown in these reviews that sorta muddies the waters for me.

I've gotten used to getting really solid board for ~$150 if I shop a little. I am talking with WiFi and cool ARGB and all that.

Gotta save the pennies for the GPUs, you know!

Yep, platform costs to get high ADL performance also kind of sucks. For those seeking the best of the best probably won't care and would spend similar amounts on a Zen 3 setup. Most people try to work within a budget though so being able to save $100 on a motherboard and possibly a decent chunk on RAM allows for a faster GPU to be purchased (if gaming is the goal) which will have a much bigger impact on your experience than the CPU. With that said, Intel should have some compelling offering in this regard with some of their lower tier SKUs. We'll have to wait and see once they are released.

 

[blckgrffn]

Any benchmark on the IGPU of the 12900K, 12700K and 12600K?

Not a huge fan of this review, but this probably says enough as its looking like a mild frequency uplift at best. Looking at 11th Gen reviews probably show you what to expect.

It's not what you get in a laptop, that's for sure.

 

[Heartbreaker]

It's going to be interesting to see what you get when you try to use a ~$100 board with a 12600K in a "value" build in the same way you might put together a 5600x type build :/ With 8th/9th gen Intel you could still get a huge portion of their performance even on lesser boards. 10th & 11th gen... meh.

If you need to use a $200+ (I mean, minimum!) board to get the type of performance that actually beats a Zen 3 CPU as shown in these reviews that sorta muddies the waters for me.

I've gotten used to getting really solid board for ~$150 if I shop a little. I am talking with WiFi and cool ARGB and all that.

Gotta save the pennies for the GPUs, you know!

There is no reason a 12600K would need any kind of fancy board. It's not power hungry like a 12900K. Even a B660 Board should be enough for a 12600K.

 

[StinkyPinky]

Gaming summary from Hardware Unboxed. They noticed DDR4 vs. DDR5 goes both ways in term of performance depending on the game, but also that some games run faster on Alder Lake with Win 10 while others prefer Win11.

View attachment 52342

When it comes to memory, for now DDR4 is the winner in my book, as the results above are obtained using the most expensive DDR5 memory.

Interesting. Hardware Unboxed don't get the credit they deserve imo.

Alder Lake is nice, but not mind blowing. Hopefully AMD respond by cutting prices on the 5900x and i can justify replacing my 3900x with it.

 

[remsplease]

Psssst..

(The bot comments are starting to become glaringly obvious)

 

[blckgrffn]

There is no reason a 12600K would need any kind of fancy board. It's not power hungry like a 12900K. Even a B660 Board should be enough for a 12600K.

That'd be great if there was no compromise on the 12600K on the lower end boards. I hope we'll see that in practice, and there won't be some protection on these boards with clipped power limits that mess with things.

 

[dacostafilipe]

On the other hand I can't imagine cache being cheap or even that widely available for z3d having a real chance, at least until things go back to more normal.

The cache "tiles" will be dirt cheap as the yields at that size should be close to 100%. The additional packaging costs will be nothing compared to having to spin a new die.

 

[JoeRambo]

Intel has achieved it! Full retard mode for IMC engineers, unlocked with this otherwise great chip.


vs


Gotta be proud of themselves, beaten by AMD chip with integrated IMC on different die. An achievement for what was once top class team.
Shame on Intel to release great core and handicap it with IMC as horrible as this.

 

[Zucker2k]

Anandtech is up!

Highlights of this review

• The new P-core is faster than a Zen 3 core, and uses 55-65 W in ST
• The new E-core is faster than Skylake, and uses 11-15 W in ST
• Maximum all-core power recorded was 272 W, but usually below 241 W (even in AVX-512)
• Despite Intel saying otherwise, Alder Lake does have AVX-512 support (if you want it)!
• Overall Performance of i9-12900K is well above i9-11900K
• Performance against AMD overall is a mixed bag: win on ST, MT varies
• Performance per Watt of the P-cores still lags Zen3
• There are some fundamental Windows 10 issues (that can be solved)
• Don’t trust thermal software just yet, it says 100C but it’s not
• Linux idle power is lower than Windows idle power
• DDR5 gains shine through in specific MT tests, otherwise neutral to DDR4

 

[tamz_msc]

Intel has achieved it! Full retard mode for IMC engineers, unlocked with this otherwise great chip.

View attachment 52347
vs
View attachment 52348

Gotta be proud of themselves, beaten by AMD chip with integrated IMC on different die. An achievement for what was once top class team.
Shame on Intel to release great core and handicap it with IMC as horrible as this.

At least as per Anandtech DDR5 does actually help in memory bound multi-threaded workloads.

 

[Makaveli]

AMD made a mistake not having V-cache at Alder Lake launch.

I don't believe so. I'm on AM4 and will get it when I get it. Unless you are building a brand new pc right now ADL doesn't matter for those already on AMD platform you will get a price drop. I've seen a few people doing side grades from 3900X to ADL-S instead of just going up to Zen 3. And if you want to shell out for a whole new board memory and cpu then all the power to you. For me I no longer due incremental upgrades. You get a much bigger sense of statifaction when you sit on it for a couple years then make the jump but to each his own.

I'm not seen anything from this launch that would want me to become an gen 1 early adopter of ADL-S. I'm good for now I won't be doing a new build until like 2024 or 2025 and will see what is out at that point in time.

 

[nxre]

The E-Cores are great, even if I was expecting a bit more. Compared to the golden standard in efficiency nowadays, they consume twice the power of an M1 Firestorm.

 

[nicalandia]

Not the bloodbath you anticipated. I agree with the assessment that gaming numbers could be better with low latency DDR4, on both sides, but I'm more intrigued with ADL-S DDR4 vs DDR5 comparison.

What were you expecting when they have the same IPC? The Higher Clocks and DDR5 do help but they are evenly match at IPC

 

[Bouowmx]

There are like 50 reviews. Where to read for tests with:

• DDR4, but faster (>= 4000 MHz)
• DDR5 manual overclock (not just XMP)
• Ring/LLC frequency adjustments
• i5-12600K overclock

Replying to self, I found a detailed DDR4 comparison: https://www.kitguru.net/components/cpu/luke-hill/ddr4-vs-ddr5-intel-core-i9-12900k-testing/

• 2x16GB DDR5-5200, 36-38-38-74, Gear 2 (leads in 7-Zip and 3DMark)
• 2x16GB DDR4-3600 18-22-22-42, Gear 1
• 2x16GB DDR4-3600 14-14-14-34, Gear 1
• 2x8GB DDR4-4400 19-23-23-42, Gear 2 (leads in gaming with NVIDIA RTX 3080)
• 4x8GB DDR4-4000 19-25-25-45, Gear 2

 

[tamz_msc]

PCGH.de has their review up where they tested 20 CPU demanding games at 720p or equivalent resolution. With power limits set to Intel recommended values and Intel spec memory (DDR5 4400 in gear 2 mode).

Intel Core i9-12900K & Core i5-12600K im Test: Angriff mit 10 nm und Hybrid-Architektur

Alder Lake im Benchmark-Test: Wie schnell sind die Hybrid-CPUs in Spielen? Das klären wir anhand von 20 Spiele-Benchmarks.

www.pcgameshardware.de

• 12900K 10% faster in games than 5950X.
• 12600K 2% faster in games than 5900X.

 

[Racan]

From Anandtech's review:

Alder Lake and the Golden Cove cores are able to reclaim the single-threaded performance crown from AMD and Apple. The increases over Rocket Lake come in at +18-20%, and Intel’s advantage over AMD is now at 6.4% and 16.1% depending on the suite, maybe closer than what Intel would have liked given V-cache variants of Zen3 are just a few months away.

I don't think AMD will have an insurmountable task countering Golden Cove.

 

[Hitman928]

Anandtech is up!

AVX-512 is not officially supported and the motherboard manufacturers are the ones who will have to make it available through firmware "hacking" (Intel turned it off and obscured the method to enable it but it seems at least some MB makers have figured out how to turn it back on when E cores are disabled).

 

[Hitman928]

From Anandtech's review:

Alder Lake and the Golden Cove cores are able to reclaim the single-threaded performance crown from AMD and Apple. The increases over Rocket Lake come in at +18-20%, and Intel’s advantage over AMD is now at 6.4% and 16.1% depending on the suite, maybe closer than what Intel would have liked given V-cache variants of Zen3 are just a few months away.

I don't think AMD will have an insurmountable task countering Golden Cove.

Yeah, in the Spec suite at least, the Golden Cove IPC over Zen 3 isn't much. GC should be boosting to 5.2 GHz vs. 5.05 GHz for the 5950x. When both are on the same DDR4, it shows GC with ~1.7% and ~9% IPC lead in integer and fp respectively over Zen 3. That's a little weaker than I expected. Did anyone else try to isolate IPC or ST performance?

 

[igor_kavinski]

AVX-512 is not officially supported and the motherboard manufacturers are the ones who will have to make it available through firmware "hacking" (Intel turned it off and obscured the method to enable it but it seems at least some MB makers have figured out how to turn it back on when E cores are disabled).

Damn. Man, I so hope that Intel doesn't "execute a scorched earth policy" in Ian's words and disable AVX-512 by fusing it off or issuing a new firmware.

 

[JoeRambo]

When both are on the same DD4, it shows GC with ~1.7% and ~9% IPC lead in integer and fp respectively over Zen 3.

I think due to JEDEC timings and speeds Anandtech is showing the baseline that is pretty much the best case for Zen3 vs ADL comparison. ZEN is able to buffer horrible memory latency with huge caches and somewhat faster memory controller.
While people value Anandtech results as baseline, they are only relevant for OEM systems and frankly should not even be discussed on enthusiast forums, nor those IPC values hold much relevance. Chips with 1.7% SPEC IPC advantage don't go and beat ZEN3 by tens of % all around the web.

I fully expect Z3D to have better IPC in Anandtech testing, while at same time getting soundly beaten by those monster DDR5 6000 machines. The gap once memory subsystem is unchained is just huge and shown everywhere else except than on Anandtech.

 

[Markfw]

So my take... Performance on sustained workloads, good, but a lot higher power use than 5950x. Gaming, they win by a small amount. For my use case (sustained 100% load, all cores) the power usage is a killer, so I won't be getting these. For a gamer ? its a winner. But the 12600k or 12700k....

Edit: I think the low power usage in games is due to the E cores being used primarily, as I doubt the CPU load is high, so the scheduler selects the E cores ????

Edit2: and the temps ??? I did not see what cooler, but a lot higher than Ryzen.