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.
IIRC the 12900K consumes something like 250+ watts, while the E-cores are consuming less than 50 watts. Drop a couple of P-Cores and include more E-cores and they could have dropped power while increasing MT peformance while using less power.
16 P cores at these clocks likely would need 400 watts, and a lot more die area.
The problem with the 12900K trying to compete with the 5950x is not from too many E-Cores, it's from having not enough E-Cores.
Not sure exactly where this fits in, but the 12700k is not horrible on efficiency, but the 12900x is bad due to Intel pushing the wattage and frequency to beat to 5950x in some things. If you low 12700k wattage down from stock even more, it becomes a reasonable choice over Ryzen for some use cases. (gaming)Yep, you nailed it 100%.
The thing is, the power level is fully under the user's control. If you don't like the 12900k power level, either set the tau to a short period or use a 125 W cooler. Either way and the chip will perform almost entirely at 125 W or below. Sure, you won't turbo all the time, but like you said it isn't horrible on efficiency then. The people running it over 125 W are doing so to push peak performance at the cost of efficiency.Not sure exactly where this fits in, but the 12700k is not horrible on efficiency, but the 12900x is bad due to Intel pushing the wattage and frequency to beat to 5950x in some things. If you low 12700k wattage down from stock even more, it becomes a reasonable choice over Ryzen for some use cases. (gaming)
Not sure exactly where this fits in, but the 12700k is not horrible on efficiency, but the 12900x is bad due to Intel pushing the wattage and frequency to beat to 5950x in some things. If you low 12700k wattage down from stock even more, it becomes a reasonable choice over Ryzen for some use cases. (gaming)
6+4.It's a 6+8 design.
I think the 12600K best exemplifies Intel Big-Little working.
Intel Core i5-12600K Review: 5600X Defeated
The Core i5-12600K is Intel's latest mainstream CPU and a direct competitor to AMD's popular Ryzen 5 5600X. The Alder Lake chip packs 6 P-cores and 4...www.techspot.com
It's a 6+8 design. Which is similar area to an 8+0 design, and it wins overall performance against the 8+0 Ryzen 5800x.
And it does it while using a tiny bit less power, despite Intel reportedly being on a worse process than AMD:
The 12600K is a 6+4 design. I'll be very curious to see your interpretation of 6+0 ADL-S versus 5600X. If it ends up winning in overall performance as well, will that also be proof that big.Little is working?I think the 12600K best exemplifies Intel Big-Little working.
It's a 6+8 design. Which is similar area to an 8+0 design, and it wins overall performance against the 8+0 Ryzen 5800x.
That's apples vs apples.If it ends up winning in overall performance as well, will that also be proof that big.Little is working?
The 12600K is a 6+4 design. I'll be very curious to see your interpretation of 6+0 ADL-S versus 5600X. If it ends up winning in overall performance as well, will that also be proof that big.Little is working?
6+4.
The 6+8 3rd tier mobile chip, the 12700h reportedly smokes the 5900hx by 47% in CB.Oops, yeah.
Imagine how it would smoke the 5800x if it was 6+8. It seems we will see this battle in Mobile.
Actually, for desktop, I would argue that "the problem" is that Intel's P cores are too large, use too much power, and are limited to 8 or maybe 10 cores because of the ring bus architecture.IIRC the 12900K consumes something like 250+ watts, while the E-cores are consuming less than 50 watts. Drop a couple of P-Cores and include more E-cores and they could have dropped power while increasing MT peformance while using less power.
16 P cores at these clocks likely would need 400 watts, and a lot more die area.
The problem with the 12900K trying to compete with the 5950x is not from too many E-Cores, it's from having not enough E-Cores.
And another...The 6+8 3rd tier mobile chip, the 12700h reportedly smokes the 5900hx by 47% in CB.
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
That was my point: if we want to evaluate Intel's hybrid solution, we need a proper "apples to apples" baseline for performance, so we should compare P-cores vs. equivalent number of P-cores and E-cores for ISO area.That's apples vs apples.
big.LITTLE is apples vs oranges.
Actually, for desktop, I would argue that "the problem" is that Intel's P cores are too large, use too much power, and are limited to 8 or maybe 10 cores because of the ring bus architecture.
Big/little seems like a very good concept for mobile and laptops, but for desktop, I still see it as a compensation for the size and power usage of the P cores, and the fact that chiplets are not ready yet. This seems obvious to me, since in order to beat the 5950x, Intel still uses more power (a lot actually).
Big.little, “P” vs. “E”, whatever, yes it is working…The 12600K is a 6+4 design. I'll be very curious to see your interpretation of 6+0 ADL-S versus 5600X. If it ends up winning in overall performance as well, will that also be proof that big.Little is working?
My 5900hx may or may not have been offering me “favors”. My desktop motherboard died suddenly today, Maybe that had something to do with it? Unless AMD pulls somethingThe 6+8 3rd tier mobile chip, the 12700h reportedly smokes the 5900hx by 47% in CB.
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
You misunderstand “P” cores then.Actually, for desktop, I would argue that "the problem" is that Intel's P cores are too large, use too much power, and are limited to 8 or maybe 10 cores because of the ring bus architecture.
Big/little seems like a very good concept for mobile and laptops, but for desktop, I still see it as a compensation for the size and power usage of the P cores, and the fact that chiplets are not ready yet. This seems obvious to me, since in order to beat the 5950x, Intel still uses more power (a lot actually).
The GPU has little to do with it, though you and I agree > 100%.I get what you're saying, but I'm not sure it's quite that simple. For example, look at the 12700K. It's an 8+4 P/E (B/L if you wish), and it's very noticeably less voracious than the 12900K stock v stock, which seems to be due to the i9 being juiced up to hit those last few hundred Mhz on the 5950X chase. It seems they weren't content to trade blows at similar power and thermals (which the i9 can do with more sane settings), but wanted a noticable lead over the Ryzen flagship, and were willing to get slightly silly to do so.
It's also a little apples oranges with Intel's insistence on cramming IGP into literally everything for consumer socket (disabled or not). It uses an amount of die space that is fairly considerable.
Huh?! Are you still talking about desktop here? Because the 12900k and all other alder lakes are one single CPU, they don't have cpu clusters, and if you call the quad-ecore-packages clusters, then the 12900k has 2 of those clusters plus the normal cores for a total of 3 clusters.Intel’s current largest issue (by far) is the inability for them to scale beyond 2 CPU clusters. If they didn’t have that limitation (i.e. Raptor Lake could have been delivered this year) Intel would have once again relegated AMD to budget builds. AMD got lucky, and Lisa Su realized that even when you are close to claiming the title, your opponent can always land a surprise blow.
As it is, we are about to see how well “new” AMD can adapt.
Time will tell.
16 P cores at these clocks likely would need 400 watts, and a lot more die area.
Not only that, it would be an MT monster even at 240W. The current split of power between 8P and 8E cores at 240W is something similar to a 4:1 ratio, meaning 21W+ for a Golden Cove core. On a 16c Golden Cove the power budget would end up at less than 15W per core, a 30% drop in power and considerable jump in efficiency.Intel wouldn't have run 16c Golden Cove @ 5 GHz or even 4.7 GHz in an MT workload. It would still be an MT monster @150W. Look at what AMD does with the 3950X and 5950X.
I'm consistently confused by your statements on this matter, do you have data on how much power the single voltage rail wastes under full load? Because last time you mentioned this together with a "50W" figure, and AFAIK those 50W represent the entire power budget of the E-core cluster.People keep crying for Intel to add more Gracemont. Until they fix the voltage rail problem it's just gonna bleed power.
I'm consistently confused by your statements on this matter,
That's an interesting notion. I would imagine that Intel has been embarrassed enough times by AMD to want to remove this thorn from its side completely. Are you saying that Intel is charitable enough to let the people at AMD keep their jobs by refraining from doing everything they can to destroy them, out of sheer goodness of heart?Also they don't want to destroy amd, if they wanted that they wouldn't have held back for the last 10 years.
So if 8P + 16E and 16P have similar MT perf at similar power consumption, 16E would be the preferred way to go as 1 Golden Cove core within cluster is much bigger than 2 Gracemont cores (closer to 4).