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.
It will be interesting to see which ratio of P and E cores they do for mobile. Apple went 8 and 2 for their Pro and Max cpu's but I suspect if Intel goes the same it will be pretty power hungry.
The 12900k is not power efficient when you turn off the e-cores. It seems the e-cores constrain the p-cores from guzzling power, and thereby running more efficiently (if we can call it that, in the case of the 12900k), whereas their absence signals the p-cores to use all the power budget available - resulting in huge inefficiency. Look at the slides below and pay attention to the green entry (8+8) vs the bronze entry (8+0).An observation from the compterbase review.
When limited to 88 W, a 5800x is 2% faster than a 12900k in multi-threaded performance.
View attachment 52360
If we then add in 8 E cores to the 12900k and compare that to adding in 4 or 8 Zen 3 cores on top of the 5800x
View attachment 52361
The advantage for Zen 3 grows to 8% for adding 4 Zen 3 cores or 16% for adding 8 Zen 3 cores. In other words, in this power limited scenario, you get a greater performance increase with an additional 4 Zen3 cores than adding an additional 8 E cores for ADL. It would be interesting to see if this comparison changes at all at even lower power levels.
The 12900k is not power efficient when you turn off the e-cores. It seems the e-cores constrain the p-cores from guzzling power, and thereby running more efficiently (if we can call it that, in the case of the 12900k), whereas their absence signals the p-cores to use all the power budget available - resulting in huge inefficiency. Look at the slides below and pay attention to the green entry (8+8) vs the bronze entry (8+0).
View attachment 52364
View attachment 52363
So, 8+0 consumes more power than 8+8 (6w) while incurring a 38% deficit in CB R23. That is clearly wasted power. You can see a similar pattern with your slides, where 8+0 is 25% slower than 8+8 while consuming the same power, 88w. I have a sneaky suspicion that there's over-volting going on here. I'd wager that should every deactivation of the e-cores be accompanied by some kind of voltage adjustment this pattern could be reversed, unless one wants to use AVX-512, of course.
There's also the effect of the e-cores on these results themselves, which definitely can't be ruled out.
E cores are also out of their efficiency zone I would guess at the 3.7-3.9ghz. 48w is quite a chunk of power for not a lot of throughput
The 12900k is not power efficient when you turn off the e-cores. It seems the e-cores constrain the p-cores from guzzling power, and thereby running more efficiently (if we can call it that, in the case of the 12900k), whereas their absence signals the p-cores to use all the power budget available - resulting in huge inefficiency. Look at the slides below and pay attention to the green entry (8+8) vs the bronze entry (8+0).
View attachment 52364
View attachment 52363
So, 8+0 consumes more power than 8+8 (6w) while incurring a 38% deficit in CB R23. That is clearly wasted power. You can see a similar pattern with your slides, where 8+0 is 25% slower than 8+8 while consuming the same power, 88w. I have a sneaky suspicion that there's over-volting going on here. I'd wager that should every deactivation of the e-cores be accompanied by some kind of voltage adjustment this pattern could be reversed, unless one wants to use AVX-512, of course.
There's also the effect of the e-cores on these results themselves, which definitely can't be ruled out.
Well comparing 6 year old 14nm architecture isn't really relevant. Skylake isn't a good yardstick anymore , by a long shot. It's also a performance core design with high fmax.
As for the 5950x comparison. Well firstly it's also heading into the inefficiency zone as far as frequency goes . It's also not an " efficiency core" by design. It's a one size fits all design. Still with a quite high Fmax. Secondly it would still be faster in your e.g
A better core for core MHz for MHz comparison would be something like a 5800H , which at the same power will sit around the same frequency , and significantly outperform it .
This means one of two things. Either GM is not efficient, or at 3.8Ghz it's outside it's sweetspot. I am suggesting the later.
Hazard a guess it's sweet spot is in the high 2s to low 3ghz. And post significantly better perf/watt Hopefully this can be tested properly one day
Well comparing 6 year old 14nm architecture isn't really relevant. Skylake isn't a good yardstick anymore , by a long shot. It's also a performance core design with high fmax.
As for the 5950x comparison. Well firstly it's also heading into the inefficiency zone as far as frequency goes . It's also not an " efficiency core" by design. It's a one size fits all design. Still with a quite high Fmax. Secondly it would still be faster in your e.g
A better core for core MHz for MHz comparison would be something like a 5800H , which at the same power will sit around the same frequency , and significantly outperform it .
This means one of two things. Either GM is not efficient, or at 3.8Ghz it's outside it's sweetspot. I am suggesting the later.
Hazard a guess it's sweet spot is in the high 2s to low 3ghz. And post significantly better perf/watt Hopefully this can be tested properly one day
I think a hypothetical 24C Gracement processor with 150W power usage vs. 16C Zen3 would be similar in highly multi-threaded apps, but will lose in single-threaded apps. This feat is great compared to what Intel had before, but reaching parity one whole year after Zen 3 launched is kinda meh. And not just meh because it's a year late, but meh because a hybrid/heterogenous approach requires a new OS scheduler and software optimization. If Intel could have achieved +20% ST perf and +100% MT perf with 16 big cores in the same power budget, they likely would've gone down that road, but alas that isn't the case.Given we now KNOW Gracemont is Skylake level IPC w/o HT, how is 8 Skylake cores w/o HT at 48 watts not good? This is not speculating, this is not yesterday. Now these are facts. 8 Skylake Cores @3.7GHz without HT using 48W on the desktop is great. Do the math, 24 of them would use ~5950X power and be as or more performant in highly threaded apps that scale well with cores. The 5950X is still the performance/efficiency champ and e's alone could beat it in specific cases. Or I'm missing something? I'm always willing to learn.
Not so long ago(actually until today) the best gaming and productivity deal were on the 10850K, Many people were recommending it, but I did not see it on today's review?I think I'll stick with the cheap 10th gen I bought for now.
FYI if you order a board and something goes wrong, Amazon's holiday return period is in swing. I think it's a 90 or 120 day return period.When I upgraded my son's PC roughly a month ago, I was going to to with an 12600k and more budget friendly Z590 motherboard, but all the major retailers were out of stock (and Amazon estimating about a 30 - 45 day wait for inventory).
I think that was due to the missing iGPU or simply because it was a slightly lower binned 10900K. I bought a 10th gen for dirt cheap brand new when Zen 3 was hard to get earlier this year. I'm keeping an eye out for Zen3 3D performance, but my main concern will be Zen4 and Raptor Lake.Not so long ago(actually until today) the best gaming and productivity deal were on the 10850K, Many people were recommending it, but I did not see it on today's review?
I think a hypothetical 24C Gracement processor with 150W power usage vs. 16C Zen3 would be similar in highly multi-threaded apps, but will lose in single-threaded apps. This feat is great compared to what Intel had before, but reaching parity one whole year after Zen 3 launched is kinda meh. And not just meh because it's a year late, but meh because a hybrid/heterogenous approach requires a new OS scheduler and software optimization. If Intel could have achieved +20% ST perf and +100% MT perf with 16 big cores in the same power budget, they likely would've gone down that road, but alas that isn't the case.
Got mine.
Personally? I expected Alder Lake to compete with Zen 3, but not outperform it. After all, this is their "Zen 1" moment in my view; their first big foray into Intel's hybrid architecture, and many modern games and productivity applications will need optimizations to fully utilize Alder Lake's potential. This is all early adopter stuff, and if you could wait for Raptor Lake, where everything will be optimized by that time, I think it would be worth it.You're quite right there, @Saylick I didn't have high hopes for Alderlake myself. And even with those leaks in the weeks leading up to today's release I was left feeling underwhelmed as more info about power and heat came out. To reach parity with a 1 year old release all the while using more power, or beating a 1 year old release while using significant levels of power that make air cooling fiddly on the higher end of the spectrum feels almost like a DOA platform at this point. Even if future BIOS releases improve performance, it's still at parity or have a slight edge that make it not worth the investment, at least not at current hardware prices and relatively immature DDR5.
All eyes on AMD and what Zen3 3D will offer up. I think I'll stick with the cheap 10th gen I bought for now. I can't see myself buying a more newer Intel platform until Intel manages to deliver a wallop of a performance stride compared to AMD.
This feels more and more like historic Intel. Coming out with a half baked idea and pushing the power on it hoping to beat AMD who cruised along with a better product. Up until AMD fumbled, of course.
But it does not address the 142 watt (stock) that I run my 5950x's at, and 12900k looses in performance, wattage and heat in my use case.I think the Venn diagram of people buying flagship CPUs but fussing over pennies in electric costs are basically non-existent.
Now pure heat output as a concern for someone literally running full load 24/7, that is actually a type of person that exists.
From the link that I was quoting. I can go back and find it when I have time.Where is this quote from?
Micro Center only had DDR5 4800 CAS 40. Might be able to overclock them. Going to get better memory later.Congrats, that's gonna be fun. What memory spec did you get?
I think this proves my point, as well as the 241 watt vs 142 watt, and the temps of the 12900kI think there's a typo in their write-up. From the chart, it's clear that it's the 5900x that is 8% faster when both are at 88W. The 5950x is 16% faster than the 12900k when it is also at 88W. The problem really comes from Intel only having 8 performance cores, so trying to catch up to the 16 performance Zen 3 cores becomes a tall task and ADL has to start using larger and larger amounts of power to match the 16 cores once they are fed a bit more as well. I.e. the power curve becomes much steeper for the 12900k vs. the 5950x as higher performance levels are reached.
View attachment 52356
But it does not address the 142 watt (stock) that I run my 5950x's at, and 12900k looses in performance, wattage and heat in my use case.