Looks good. The area efficiency of E-Cores (3 or 4:1, depending on the additional structures in counting) pays obvious dividends for the multithreaded performance in busts controlling on a given die space, at least. The downside of Intel's strategy in terms of the area-efficient nature of E-Cores & pushing them past their ideal energy efficiency ranges (relate to a P-Cores that is only less efficient than E-Cores at lower performances) is the harm fo energy efficiency — their graphs refer and suggest this.
One question I'm curious about is how much QoS benefits from having of these additonal cores ready for use or operating on background tasks specifically in the context of mobile solutions where throttling (for Intel anyway, and AMD) is inevitable — maybe we will still see notable responsiveness improvements reported for these units not only because of Golden Cove but the load balancing benefits, too. Windows 11 should help in this regard obviously as we've seen with desktop implementations.
That sounds like it's a big step for mobile as intel made toasters up to this point, and even a simple action did trigger warm palms. I just hope it means there will be better chance for laptops that are flexible in power usage, not as before, where it was max power or 0 and nothing in between.
It's definitely going to be interesting seeing how this pans out. I'm not particularly optimistic - I feel quite strongly that 4 powerful cores running in their efficiency sweet-spot and racing to idle is more likely to give good response times for the kind of work that gets done on ultra-portable devices. A surfeit of slower cores will work well for any tasks that can be processed in parallel, but I'm not sure that's of particularly great use to many end-users.
Sorry - I posted a reply to another comment. The questions about relative performance are totally legitimate. I hope the CPU will be available soon so that AnandTech can run some benchmark.
Still with their funny TDP I see. 15W that turbo to 55W. 28W turbo to 64W. Why not write it as 5W, as when idle, it will use lower than that. Then, their vPro. Now we have vPro enterprise, essential. Later they'll add vPro primary, secondary, home, server, etc.. Not enough fragmentation and product differentation, eh? Still the shitty old intel.
I believe I may have misjudged Alder Lake by their misguided power curve for their desktop which gave me a bit of a red herring. A keen observation is the maximum power draw (64W) at burst for these 14 core 28W is in fact no more than a 4-core Tiger Lake-U--that's right, with 10 more cores. The Alder Lake-P 28W models have two additional performance/big cores (4+2) plus 8 all-new efficiency/LITTLE cores and with all cores activate at maximum boost clocks, they are guaranteed to draw no more than 64W. Tiger Lake-U 28W models had just 4 performance/big cores and it drew 64W at boost regardless. I do not care much for Alder Lake on desktop since they are arguably threw too much power at the platform for a measly 5% gain. They could have dropped power by 30-40% and still held 95% of their multicore performance and maintained the same class-leading single-core performance (https://www.reddit.com/r/intel/comments/qwmzfm/cor... However, I think the high-power boosts on desktop may have been them doubling down on power to ensure a clear across-the-board victory especially in multicore performance which AMD has held for the most part since 2017.
I'm most interested in seeing two things: 1) Whether they actually obey that 64W limit in practice 2) What happens to performance when they bump into that limit in applications that aren't embarrassingly parallel (e.g. Cinebench).
They claimed a 15% improvement in power efficiency from the new process - there's no such thing as a free lunch, so if there's much of a improvement in performance, power isn't likely to be very different.
Hifihedgehog; interesting power observation. If Alder U stalls out of the gate the channel is flooded with Tiger U, Ice U and Comet U that today combined in WW Channel equal 18x of all Alder desktop S available and 94% of all Comet, Rocket, Alder S available, Great clearance sale pricing on all back gen Intel mobile right now U or H. mb
You can feasibly cool 9w in a slim, passive laptop for light / regular use. (see Macbook Air 15w M1) At 27w turbo for anything demanding? Not sensibly. I'd assume the turbo duration and algorithms to decide to turbo will be quite conservative, otherwise you'd really not enjoy using such a system.
Surface 7+ is passive with 28w Tiger Lake i5, but benches closer to i7 laptops with fans.
I have had one for a year, and my only complaint is that I wish it could run in 35w or 45w mode. Even if you lock it to run 3+ GHz at all times, it stays room temperature. They left something on the table, rather than handicapping it. I had to run a power virus to find where the turbo finally timed out.
If your frame of reference is something like a dual-core Skylake U series, those were hot garbage in comparison. I had one of the Y series laptops and returned it because it throttled so badly. I have a Dell XPS 13 with a 6560U, and it idles at a higher temperature than the Surface at 100% CPU. It can only briefly reach the 3 GHz boost unless it's manually undervolted. Windows antivirus and other background processes randomly coming on will peg the CPU to 100% when you're in the middle of something and instantly jump to 90C or higher.
When there's an extra core or two, it doesn't get hung up doing normal things, so it doesn't heat up. Similar premise with the AMD CPUs that have the same peak clock speeds from thin and light 15 watts to 65 watts in a desktop.
You can, but you can't cool 29W passively, so either the processor is going to throttle almost immediately in a passive design or it's never going to hit anything like its theoretical peak performance. The last Intel MacBook Air is a great example of what this looks like in practice.
Lets be fair here, the last MacBook Air had a literal credit card sized piece of slightly finned aluminum , did we really expect it not to heat up like the savanna and the surface devices are kinda an example of what could become with these chips
That's not a done deal. Intel could end up creating a successor to Jasper Lake with only Atom ("E") cores. I hope they just replace that lineup with 1+4 mobile chips, since that would give a substantial single-thread boost.
I believe Atom-only SoC's will be dedicated to enterprise. The P core's single-thread boost is too massive to throw away.
If Intel wanted to make a really low-end chip they could just make a die with a very stripped down GPU, I/O, just 1xP and 1xE core complexes, that kind of thing.
nandnandnand, Like Lakefield, Jasper N (and J is a no show) appear rejected by the supply chain. Most Atom are freebie in sales bundle and for Intel cost optimizing from producing for supply to producing for demand Atom product line is an 'avoidable cost'. Also there are no embedded Pentium or Celeron J, N or Atom x;Tremont / Jasper / Elkhart in channels; no shows. mb
The i7 1260p 4p8e is a 28w part with base clock 2.1g/1.5g; trying to get it down to 9w seems impossible. The i7 1260u 2p8e at 9w only has a 1.1g/.8g base, which is just about half the clock and half the p-cores.
The 15w version is the i7 1265u, 2p8e 1.8g/1.3g; so a 600MHz p boost and 200Mhz e boost increases the low power mode power consumption another 6w
I suspect a 4p8e i7 would be something ridiculous like 800Mhz with a boost to 1.2GHz
One of those " 9 W" U series i5 or i7 would make for a nice, light and passively cooled ultraportable. Now waiting for the pricing of those laptops or 2-in-1s. My old "typewriter" , bought in 2015, is about due for a replacement.
Microsoft will have to rewrite/change-the-detection-algorithm for Windows 11 - These 1+4 systems where only one core is at-or-over 1 GHz would fail. Which I find hard to believe Intel would release a CPU not capable of running the current OS.
I do wonder how this 1+4 core would compare to my quad-core Atom Windows tablet from 2013 running Windows 10. :-D
CharonPDX, Consider Intel tied to Microsoft at Window 11 thread directed application optimization is all about reconfiguring 'what is' the x86 standard platform. Is AMD V5x on Windows 10 now more standard than Alder on Win 11? Intel and MS reconfigured 'what is standard' leading into Zen 4 left in a design scramble to fit into the new Windows 11 'sandbox' that is an Intel Microsoft 'play ground' and this is not the first time. mb
The P-series caught me off guard, seems like they could just use H and U and have TDPs adjust as necessary (btw, U-series having turbo TDPs of 55 W is a GOOD THING, something nobody seems to get).
But the funniest part of the lineup for me is still the 6+4 i7-12650H vs. 4+8 i5-12600H.
Comet Lake U may not be the best example to cite, as compared with competing Zen 2 designs it's very hot and not especially fast. Hurry-up-to-sleep is a great strategy but they stretched it too far in an attempt to compete with AMD, who were offering more cores and better IPC.
Spunjji, Comet U was all about sitting on Renoir; on full run data Comet U production volume is 310% more than all of Renoir H/HS/U is an example of how Intel monopolizes channels financially on stuffing surplus into the channel. Today in WW channel Comet U = 127% more than Renoir mobile channel available and adding Ice U to Comet U 292% more than all Renoir mobile available. Intel did the same thing Tiger U v Cezanne H/HS/U. Also Tiger U production overage compounded in volumes on Intel relying on a quad to prove SF10 yield before moving to Tiger octa H and Sapphire Rapids 14C so there are a ton of Tiger U at 135% more than Renoir and Cezanne mobile combined available today in the WW channel and when adding Tiger H 203% more Tiger all up. mb
"U-series having turbo TDPs of 55 W is a GOOD THING" Not sure I fully agree. Being able to boost high and drop down quickly is nice, but in combination with Windows' capricious attitude to randomly pegging a CPU core at 100% and the average OEM's tendency to under-spec their cooling to provide a "thin-and-light" experience, the end-user experience is often underwhelming.
"6+4 i7-12650H vs. 4+8 i5-12600H" That one makes me chuckle too. The first is probably going to be one of the best all-around processors Intel have in their range (great for gaming with fewer E-cores sapping TDP, still enough of them to be decent for productivity) while the second is probably going to be their best-value productivity offering.
Intel may be the only company to beat the major car manufacturers at arbitrary product differentiation and the accompanying ability to sell customers more than they need.
You guys have this part backwards. "Bringing up the rear is the Core i3-1220P, which offers only two efficiency cores and eight performance cores. Thie integrated graphics are also further cut back, with another drop to 64 execution units, but still the same 28-64 W power window. This Core i3-P looks very much like what the top Core i7-U processors will be in that configuration, but the Core i7-U are much more efficient." Should say 2P + 8 E
Intel still has the Celeron N4020: 2 small (Goldmont Plus, some type of Atom) cores, no hyperthreading, 12 EU on graphics, 4MB cache. Going from that to the Celeron 7300 (1 big and 4 small cores, 48 EUs on graphics, 8 MB cache) you probably double the silicon area used (and so the price). I don't see pure "Atom" style going away from Intel - as there are some places where Atom performance is enough (low end NAS, ...), and there are some places where even Atom price is a big part of the total hardware cost.
Trenzterra, the supply chain rejected Jasper / Tremont / Elkhart Pentium, Celeron and Atom X. Only Silver N is seen in the channel and no Silver J and no embedded. Intel offers Atom for free in Core sales bundle and where Intel is now reconfiguring to producing for demand reforming the channel monopolizing tactic of producing for supply. On Intel cost optimizing Atom is an 'avoidable cost' mb
They somehow found a way to make their naming schema even worse. Well done, Intel!
My favourite is probably the i7s and i5s that are largely identical, excepting that one i7 that is significantly better than the rest. Different core counts for the same generation and same classification is something Intel really should have dropped some time ago.
I feel bad for anyone who ends up with a Celeron 7300. I don't understand what the point of disabling Turbo is, other than to punish the customer for being frugal. I guess at least that may be the one CPU that stays within TDP instead of heating itself up and then throttling like mad.
Agreed. That Celeron to the Pentium one higher in the stack is the largest relative jump in performance for adjacent parts I can ever recall in a list of CPUs.
Indeed - they've been disabling turbo for no obvious reason besides punishing cash-strapped customers for a while now, but it's rare to see them do it when the clocks are already in the toilet
If not for pricing the 9 watt U parts would be good for ipad m1 android equivalent tablets. Also now with an engineer ceo again I feel like they could engineer a competitive soc for smartphones considering qualcomms struggles.
All this binning and salvaging used to be meaningful: With the advent or multi-core chips, selling partial defects e.g. with 3 out of 4 cores still good, not only made a lot of sense, it also made me feel quite warm and comfy about those handicapped chips finding a less demanding home than mine.
But when I look at this plethora of chips and the nature of fabbing, it’s obviously quite impossible to achieve such consistent defect patterns across so many process generations, were always some cores or parts of the cache and iGPU either wind up defective or exceeding voltage threshold, which requires culling of cores, cache or GPU blocks in the name of TDP.
Significant engineering talent is obviously dedicated at producing sub-par parts from excellent chippery and I wonder what the psychological toll is on those poor bastards, who were made responsible for i3 or Pentium parts. Imagine them going to bed every night, hoping that one day they might be permitted to do i7, i9, i11 or i13! It’s just heart wrenching!
Sure, the original Pentium was a hothead and tended to unsolder itself, but it also set Intel apart from all 80486: How did it become a banner of shame? Ok, I’m drifting...
I am quite literally just fed up with Intel: too many choices, none satisfying. What I want is the ability to balance compute power versus noise and battery capacity, so just sell me a chip and levers which allow me to slide the bars between 5 and 45 Watts for a mobile and a NUC, 35 to 250 Watts on desktop, and let me adjust P1, P2 and max fan speed according to my current need at run-time.
Then I might want to order slices of P or E for my silicon pizza and that’s it!
Everything else is just torture designed to push me to i-max, which currently leads me to buy full chunks of AMD APUs or CCDs, from 15 to 150 Watts in my home lab, plenty more in the DC.
"How did it become a banner of shame" They needed to get some distance from all those years of pushing expensive Netburst chips on people, lying about how good it was, and paying OEMs not to stock competitor parts.
abufrejoval, Thank you for these astute observations on falling out on sort. Under Docket 9341 consent order disabling good parts is monitored and essentially not permitted as anticompetitive to target others with down bin parts that did not fall out of sort organically. At the same time unlike Ivy when the Intel practice of disablement at Intel became pervasive, by Comet, the cost became prohibitive on a full cores mask set design. The same is true with TSMC Zen octa high cost of disablement v. organic fall out. Intel has moved to dice area optimized masks which AMD is just beginning to do with 65/64xx Goby. The marginal cost to sort for Alder Lake Pentiums and Celeron is approximately $5 per component that have approximately a $15 all up production cost. Intel relies on statistical modeling to analyze projectable yield and the process is very sophisticated in actual use. mb
Here my naïve self had not been aware how the FTC had managed to head off the most egregious abuses, countered in gory detail in that document: I wonder if the document has been updated since, and what it should look like today.
But I can’t see any reference to intentionally disabled chips in there… is that perhaps in an updated document or is that simply a different number?
Anyhow, it fully explains how Intel wants to go back to feature licensing on their chips, because it completely avoids such a “do not fuse off working parts” decree, which obviously hasn’t been illegal in the mainframe world. Once they’re sure they didn’t botch the implementation (like currently the base overclock with the AL non-K parts), they’d certainly want to use that mechanism to squeeze their customers.
And to be honest, it wouldn’t be all bad, if it wasn’t then abused later. But I’m sure they’d become extremely creative, once it suits them.
Among the things I’d consider fair would be the ability to add and extract core/ECC/iGPU/clock unlocking codes from a SoC and resell them, consumer to consumer. So, if grandpop passed on his surfing machine to his granddaughter, she can turn it into a modest gaming rig buying somebody else’s unlock code off eBay. Sure, it could be tied to a generation, as long as they don’t increment those artificially to below a true technical cadence just to kick this out again.
Since these are purely digital assets, Intel shouldn’t be allowed to raise their prices through the roof three years later, just because they want to sell new chips.
Somebody better start planning a global rule book there, before it takes another ten years to pass the dozen different anti-competitive bodies to react.
On the economical side, selling octa-core chunks as a reduced performance dual-cores like IBM does for mainframes, can’t be fully repeated in PCs or consoles, I guess. With Intel having their own fab in theory their ability to switch wafers between M, X, XXXL or any other die variant on short term, give them a huge advantage vis-à-vis an AMD that has to wait in line at TMSC. But that only lasts as long as AMD/TMSC can’t cook up better post-fab/packaging stage modularity, which the 3D cache stack is hinting at, vs. Foveros etc. The CCD/IOD combo certainly stirred some engineering into action.
As a fan of good engineering I am very much looking forward to the solutions both sides with think up there.
But then come the bean-counters and that's where I am much less hopeful. With Charly warning against Pluton malware on AMD and Intel seemingly trying to leap ahead of AMD everywhere, all this doesn’t look like consumers will be served better with something they can use longer, which is the only progress I am really interested in.
I’m enlisted by the FTC to monitor Docket 9341 consent agreement since 2010; AMD, Intel, Nvidia and Via, contracted by USDOJ to recover Intel Inside price fix for buyer return.
No (hidden) Reference to “intentionally disabled chips”, FTC is amorphous in determinations.
Premeditated disablement of a good hexa to produce a quad or quad to produce a dual and for the specific purpose of flooding channel to limit competitive participation ‘organically’ falls under “no efficiency justification” and “raising cost” which is a Sherman Act Section 2 violation where Intel Inside rebate falls under Section 1 combining contracts “raising costs” as EUCC 37.990 described “unnecessary avoidable cost charge”.
FTC mandate pursuant disablement falls generally within 9341 consent agreement “Part IV A; respondent shall not invite, enter into, implement, continue, enforce any condition, policy, practice, agreement, contract, understanding, or any requirement . . .” Note, pursuant any OEM full line procurement the amount of down bin product is specified in the Intel OEM / IDM purchase agreement. Down bin is often freebie ‘sales close incentive’ Intel throws to any sales contract and specified for OEM/IDM 'supply chain' production planning.
The practice of disablement “raising cost” generally falls under Price < Cost and in this case on the cost of additional production steps associated disablement from good too bad to meet a contract supply obligation vis-à-vis what might organically fall out of sort.
Thank you for the mainframe world observation “licensing features” applied to Intel; aha!
On the AL non-K OC BIOS mod, Intel has done this before at risk production to goose sales out of the gate and then claims ignorance and no warranty responsibility; a spring board sales accelerator for certain end customer types that would not otherwise have purchases a K part.
On Intel locking out CPU upgrade I thought it prevented down grade i5 to i3 but not upgrade i5 to i7. Either way I do not believe in disabling features including Nvidia HLR on aftermarket value effect and unseen consequences into future essentially over segmentation or lock out.
EUCC traditionally leads in defining the rule book. I’m actually retained by Congress where once the FTC completes a project tends not to mind it. I do submit rule changes to Congress.
True, PC’s don’t have the IBM OLTP mark up; selling octa at reduced performance; enabling only two cores abd a lot of cache for databases. Note Epyc can be characterized and its CCX and L3 partitioned like a ‘many instances” GPU but AMD does not talk about it outside a clutch of developers.
Quad Scalable from Intel low core count deca can also be characterized (or disabled) as a quad for maximum L3 per core so it does happen in the high cost : price / margin systems market.
Intel coming out of its sabotage post Comet revitalized its mask shop at Tiger lake quad for die size optimized mask set. AMD with cash again started at GPU Beige Goby where in relation Intel AMD has to go to 3rd party mask cost.
Thanks for your thoughts.
I too look to enterprise serving customers on responsibilities in technocracy "meet the needs of customers and society as cost effectively as possible", stewardship cognizant welfare", "complimentary participation, partnership, cost and reward sharing", planning for an adequate economic return on investment", "steward in the countries of industry operation", "sustainable growth models that make economic sense", Watson Jr.
Apple, I believe, offers one chip to satisfy all of the use cases covered by this long list of chips. It is called the M1. If Intel was serious about consumer needs it would stop playing the pointless market segmentation game -- it is all sound and fury signifying nothing.
If Intel was doing right by consumers it would cover all of these use cases with two chips -- one competitive with the M1 and another non-premium part that is cheaper and more energy efficient that is suited to users who don't need the premium option.
To be fair, they have the M1, 3 variants of the M1 Pro (8/14 Core CPU/GPU, 10/14, 10/16) and 2 variants of the M1 Max (10/24, 10/32), so they have 6 chips, which is still much fewer than 20.
Yes, and the parts you mention -- M1 Pro and M1 Max -- push well into i9 territory. The higher end versions of the M1 (or, no doubt, M2) aren't targeted at thin and light laptops. Those chips are to be compared to high end Intel parts not the chips mentioned in this article. The M1 and/or M2 represent the right point of comparison for the ADL chips covered in this article.
Totally disagree. Intel's CPU is targeting for thin and light because they don't expect those CPUs to be paired with dGPU.
On the other hand, Apple's M1Pro/M1Max are paired with 16c/24c/32c GPU which are MUUUUCH faster than Intel's iGPU ... Apple take the GPU in the account when they configure their laptops.
To compare Apples to Apples, M1Pro/M1Max CPU has comparable power draw (albeit even lower power consumption since they don't have "peak power draw" figures. Their CPU simply max out at 30W (40W CPU SoC total draw including RAM,MB,etc..).
This is simply on par with Intel's U/P series "thin and light" CPU's power numbers to be honest.
On ARM, Apple has 2 laptops, 1 of which comes in 3 sizes. And they have one mini computer that uses laptop parts. Of course they're going to have significantly less CPU SKUs than Intel, who makes a wide range of chips for everything thing from embedded devices, to thin and lights, to standard laptops, desktop replacements, desktops, workstations, servers, etc.
Of course they're going to have significantly more SKUs not only for different types of devices, but for different price points within these types.
If it made sense for Intel to offer only 2 chips to meet all client computing needs and budgets, they would. But it doesn't at all.
Well, the M1 is for thin and lights and no chip mentioned in this article is for embedded devices, desktop replacements, desktops, workstations or servers. Your comments on those points are irrelevant. The ADL chips covered in this article are for thin and light laptops. And, that huge profusion of chips is there for the sake of Intel's margins, nothing more. There is no benefit for users in any of it.
How exactly to users lose out by choosing the M1 over all of these ADL parts? (Please note, I have conceded that in terms of genuine consumer needs a more energy efficient and lower power option would be useful. Perhaps, I'm wrong, maybe even two non-premium options might be needed.)
Personally, I wouldn't equate Intel's profit margins with user needs. If that was valid Intel oughtn't to be experiencing any problems today but it is one of the few major semiconductor companies that isn't growing.
"workstations, servers" The servers and workstations have their own processors, not included there. Usually they have a bit more memory bandwidth for the High Performance Computer (one more fast memory channel) or several more channels of slower memory for servers.
…AMD is able to run a Zen 3 core at 3.5GHz at less than 3W, so it’s really hard to imagine a situation where any number of P cores which are probably ~3W themselves combined with a large number of E-cores which are substantially but not magically more efficient is doing much better in power terms,
Please add a share button in all these articles. I would like to send a copy of some of these excellent review articles to my friends. For example, the links should be sent in WhatsApp, Facebook, Twitter etc.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
91 Comments
Back to Article
16KGranule - Wednesday, January 19, 2022 - link
Looks good. The area efficiency of E-Cores (3 or 4:1, depending on the additional structures in counting) pays obvious dividends for the multithreaded performance in busts controlling on a given die space, at least. The downside of Intel's strategy in terms of the area-efficient nature of E-Cores & pushing them past their ideal energy efficiency ranges (relate to a P-Cores that is only less efficient than E-Cores at lower performances) is the harm fo energy efficiency — their graphs refer and suggest this.One question I'm curious about is how much QoS benefits from having of these additonal cores ready for use or operating on background tasks specifically in the context of mobile solutions where throttling (for Intel anyway, and AMD) is inevitable — maybe we will still see notable responsiveness improvements reported for these units not only because of Golden Cove but the load balancing benefits, too. Windows 11 should help in this regard obviously as we've seen with desktop implementations.
16KGranule - Wednesday, January 19, 2022 - link
bursts*refer to and suggest this*
deil - Wednesday, January 19, 2022 - link
That sounds like it's a big step for mobile as intel made toasters up to this point, and even a simple action did trigger warm palms.I just hope it means there will be better chance for laptops that are flexible in power usage, not as before, where it was max power or 0 and nothing in between.
Spunjji - Thursday, January 20, 2022 - link
It's definitely going to be interesting seeing how this pans out. I'm not particularly optimistic - I feel quite strongly that 4 powerful cores running in their efficiency sweet-spot and racing to idle is more likely to give good response times for the kind of work that gets done on ultra-portable devices. A surfeit of slower cores will work well for any tasks that can be processed in parallel, but I'm not sure that's of particularly great use to many end-users.meacupla - Wednesday, January 19, 2022 - link
Just when you thought AMD's mobile ryzen 6000 series product stack had too many numbers, intel does this.BedfordTim - Thursday, January 20, 2022 - link
I was thinking the same thing. It used to be so simple.Kangal - Thursday, January 20, 2022 - link
I'm curious as to how they compare:(10W) AMD r3-5400u -vs- Intel i7-1260U (9W)
(15W) AMD r5-6600u -vs- Intel i7-1265U (15W)
(15W+)AMD r7-6800u -vs- Intel i7-1270P (28W)
(35W) AMD r7-6980hs -vs- Intel i7-1280P (+28W)
...which has the higher ?
- single-core performance (probably Intel)
- multi-threaded performance (probably equal)
- sustained/thermal performance (probably AMD)
- graphics performance (probably AMD)
yankeeDDL - Friday, January 21, 2022 - link
Sorry - I posted a reply to another comment.The questions about relative performance are totally legitimate. I hope the CPU will be available soon so that AnandTech can run some benchmark.
t.s - Monday, February 7, 2022 - link
remember that intel 9 and 25w peak at 55Wt.s - Monday, February 7, 2022 - link
*15wByte - Tuesday, February 22, 2022 - link
graphics is practically a PS5/XBS under there, not contest.Samus - Friday, January 21, 2022 - link
At least AMD hasn't gone full ridiculous like Intel did with the Celeron products. I mean 1.1GHz base clock with no turbo?Seriously, 1GHz x86 processors in 2022.
olde94 - Friday, January 21, 2022 - link
it's within the windows 11 "atleast 1ghz" requirement i guess?GeoffreyA - Friday, January 21, 2022 - link
Lucky it's not the post-1 GHz Pentium III!yankeeDDL - Friday, January 21, 2022 - link
And at 9W!I could understand if the Celeron burns <1W...
yankeeDDL - Friday, January 21, 2022 - link
And at 9W!I could understand if the Celeron burns <1W...
GeoffreyA - Friday, January 21, 2022 - link
Well, it's almost nostalgic, taking us back to the time when 1 GHz was still a barrier.t.s - Wednesday, January 19, 2022 - link
Still with their funny TDP I see. 15W that turbo to 55W. 28W turbo to 64W. Why not write it as 5W, as when idle, it will use lower than that. Then, their vPro. Now we have vPro enterprise, essential. Later they'll add vPro primary, secondary, home, server, etc.. Not enough fragmentation and product differentation, eh? Still the shitty old intel.prashplus - Wednesday, January 19, 2022 - link
lol trueHifihedgehog - Thursday, January 20, 2022 - link
vPro: When you have more editions than a Microsoft Windows OS. 🤣Hifihedgehog - Thursday, January 20, 2022 - link
I believe I may have misjudged Alder Lake by their misguided power curve for their desktop which gave me a bit of a red herring. A keen observation is the maximum power draw (64W) at burst for these 14 core 28W is in fact no more than a 4-core Tiger Lake-U--that's right, with 10 more cores. The Alder Lake-P 28W models have two additional performance/big cores (4+2) plus 8 all-new efficiency/LITTLE cores and with all cores activate at maximum boost clocks, they are guaranteed to draw no more than 64W. Tiger Lake-U 28W models had just 4 performance/big cores and it drew 64W at boost regardless. I do not care much for Alder Lake on desktop since they are arguably threw too much power at the platform for a measly 5% gain. They could have dropped power by 30-40% and still held 95% of their multicore performance and maintained the same class-leading single-core performance (https://www.reddit.com/r/intel/comments/qwmzfm/cor... However, I think the high-power boosts on desktop may have been them doubling down on power to ensure a clear across-the-board victory especially in multicore performance which AMD has held for the most part since 2017.Spunjji - Thursday, January 20, 2022 - link
I'm most interested in seeing two things:1) Whether they actually obey that 64W limit in practice
2) What happens to performance when they bump into that limit in applications that aren't embarrassingly parallel (e.g. Cinebench).
They claimed a 15% improvement in power efficiency from the new process - there's no such thing as a free lunch, so if there's much of a improvement in performance, power isn't likely to be very different.
Mike Bruzzone - Friday, January 21, 2022 - link
Hifihedgehog; interesting power observation. If Alder U stalls out of the gate the channel is flooded with Tiger U, Ice U and Comet U that today combined in WW Channel equal 18x of all Alder desktop S available and 94% of all Comet, Rocket, Alder S available, Great clearance sale pricing on all back gen Intel mobile right now U or H. mbSpunjji - Thursday, January 20, 2022 - link
Bit of a piss-take, isn't it?olde94 - Friday, January 21, 2022 - link
wait wait wait? 15W can peak at 55W?how much can AMD peak at?
wr3zzz - Wednesday, January 19, 2022 - link
Can you cool 9W passively? Don't see the point if it can't.Jorgp2 - Wednesday, January 19, 2022 - link
Yes, but you'll have a Hotspot unless you do a really good job of spreading heat around.It will be uncomfortable if you touch it directly.
BushLin - Wednesday, January 19, 2022 - link
You can feasibly cool 9w in a slim, passive laptop for light / regular use. (see Macbook Air 15w M1)At 27w turbo for anything demanding? Not sensibly. I'd assume the turbo duration and algorithms to decide to turbo will be quite conservative, otherwise you'd really not enjoy using such a system.
brantron - Thursday, January 20, 2022 - link
Surface 7+ is passive with 28w Tiger Lake i5, but benches closer to i7 laptops with fans.I have had one for a year, and my only complaint is that I wish it could run in 35w or 45w mode. Even if you lock it to run 3+ GHz at all times, it stays room temperature. They left something on the table, rather than handicapping it. I had to run a power virus to find where the turbo finally timed out.
If your frame of reference is something like a dual-core Skylake U series, those were hot garbage in comparison. I had one of the Y series laptops and returned it because it throttled so badly. I have a Dell XPS 13 with a 6560U, and it idles at a higher temperature than the Surface at 100% CPU. It can only briefly reach the 3 GHz boost unless it's manually undervolted. Windows antivirus and other background processes randomly coming on will peg the CPU to 100% when you're in the middle of something and instantly jump to 90C or higher.
When there's an extra core or two, it doesn't get hung up doing normal things, so it doesn't heat up. Similar premise with the AMD CPUs that have the same peak clock speeds from thin and light 15 watts to 65 watts in a desktop.
BushLin - Thursday, January 20, 2022 - link
The Surface 7+ has a fan btwbrantron - Thursday, January 20, 2022 - link
"Surface 7+ is passive with 28w Tiger Lake i5"Bruh, it's the first thing I said...
Spunjji - Thursday, January 20, 2022 - link
You can, but you can't cool 29W passively, so either the processor is going to throttle almost immediately in a passive design or it's never going to hit anything like its theoretical peak performance. The last Intel MacBook Air is a great example of what this looks like in practice.CanadianX86USER - Friday, January 21, 2022 - link
Lets be fair here, the last MacBook Air had a literal credit card sized piece of slightly finned aluminum , did we really expect it not to heat up like the savanna and the surface devices are kinda an example of what could become with these chipspoohbear - Wednesday, January 19, 2022 - link
Great to see that even the lowest end of CPUs will have 5 cores moving forward.nandnandnand - Wednesday, January 19, 2022 - link
That's not a done deal. Intel could end up creating a successor to Jasper Lake with only Atom ("E") cores. I hope they just replace that lineup with 1+4 mobile chips, since that would give a substantial single-thread boost.Wereweeb - Thursday, January 20, 2022 - link
I believe Atom-only SoC's will be dedicated to enterprise. The P core's single-thread boost is too massive to throw away.If Intel wanted to make a really low-end chip they could just make a die with a very stripped down GPU, I/O, just 1xP and 1xE core complexes, that kind of thing.
I doubt that makes much economic sense.
Mike Bruzzone - Friday, January 21, 2022 - link
nandnandnand, Like Lakefield, Jasper N (and J is a no show) appear rejected by the supply chain. Most Atom are freebie in sales bundle and for Intel cost optimizing from producing for supply to producing for demand Atom product line is an 'avoidable cost'. Also there are no embedded Pentium or Celeron J, N or Atom x;Tremont / Jasper / Elkhart in channels; no shows. mbSourav Das - Wednesday, January 19, 2022 - link
We will see how this performs in real life. My guess is that these CPUs gonna still heat upyeeeeman - Wednesday, January 19, 2022 - link
Would have liked to see a 4P + 8E variant in the 15W space.michael2k - Thursday, January 20, 2022 - link
You mean you wish Intel had lied about the power consumption? Or that they just cut frequency in half?nandnandnand - Friday, January 21, 2022 - link
Well, it's a different die.What you could do is take your 4P + 8E and try to undeclock it yourself.
michael2k - Friday, January 21, 2022 - link
The i7 1260p 4p8e is a 28w part with base clock 2.1g/1.5g; trying to get it down to 9w seems impossible. The i7 1260u 2p8e at 9w only has a 1.1g/.8g base, which is just about half the clock and half the p-cores.The 15w version is the i7 1265u, 2p8e 1.8g/1.3g; so a 600MHz p boost and 200Mhz e boost increases the low power mode power consumption another 6w
I suspect a 4p8e i7 would be something ridiculous like 800Mhz with a boost to 1.2GHz
vlad42 - Wednesday, January 19, 2022 - link
"Bringing up the rear is the Core i3-1220P, which offers only two efficiency cores and eight performance cores."This seems to be backwards. From Intel's supplied tables, it should be two performance cores and eight efficiency cores.
repoman27 - Wednesday, January 19, 2022 - link
"It’s also worth noting that the chipset is also on die..."That should be on package, not on die.
hemedans - Wednesday, January 19, 2022 - link
Am so Happy for those i3, pentium and Celeron, they Benefit a lot, imagine sub $200/300 laptop with 6 or 8 thread.Spunjji - Thursday, January 20, 2022 - link
"6-thread" Celeron with no turbo and 4 Skylake-IPC cores running at 7/900Mhz is going to run like a wounded dognandnandnand - Friday, January 21, 2022 - link
If Jasper Lake with only E-cores is decent enough for basic tasks, throwing a P-core and better E-cores in is even better.Jorgp2 - Wednesday, January 19, 2022 - link
>It’s also worth noting that the chipset is also on die, but offers more functionality than the 9 W chips.You mean on package?
eastcoast_pete - Wednesday, January 19, 2022 - link
One of those " 9 W" U series i5 or i7 would make for a nice, light and passively cooled ultraportable. Now waiting for the pricing of those laptops or 2-in-1s. My old "typewriter" , bought in 2015, is about due for a replacement.CharonPDX - Wednesday, January 19, 2022 - link
Microsoft will have to rewrite/change-the-detection-algorithm for Windows 11 - These 1+4 systems where only one core is at-or-over 1 GHz would fail. Which I find hard to believe Intel would release a CPU not capable of running the current OS.I do wonder how this 1+4 core would compare to my quad-core Atom Windows tablet from 2013 running Windows 10. :-D
Mike Bruzzone - Friday, January 21, 2022 - link
CharonPDX, Consider Intel tied to Microsoft at Window 11 thread directed application optimization is all about reconfiguring 'what is' the x86 standard platform. Is AMD V5x on Windows 10 now more standard than Alder on Win 11? Intel and MS reconfigured 'what is standard' leading into Zen 4 left in a design scramble to fit into the new Windows 11 'sandbox' that is an Intel Microsoft 'play ground' and this is not the first time. mbvelanapontinha - Wednesday, January 19, 2022 - link
Once upon a time, Intel lineup was Pentium II at 350, 400 and 450 MHz, and Celeron 266, 300 and 333.A grand total of 6 processors for desktop.
Compare that with today's insanity.
CharonPDX - Wednesday, January 19, 2022 - link
They still sold Pentium MMX and 300 + 333 MHz P2 at the same time. They didn't just discontinue *ALL* "not current gen" CPUs.nandnandnand - Wednesday, January 19, 2022 - link
The P-series caught me off guard, seems like they could just use H and U and have TDPs adjust as necessary (btw, U-series having turbo TDPs of 55 W is a GOOD THING, something nobody seems to get).But the funniest part of the lineup for me is still the 6+4 i7-12650H vs. 4+8 i5-12600H.
Fulljack - Thursday, January 20, 2022 - link
good how? explain please.brantron - Thursday, January 20, 2022 - link
Is waiting on things good?Spoiler alert: you've probably used a U series laptop with a higher power limit. Comet Lake U laptops went up to 100 watts.
Spunjji - Thursday, January 20, 2022 - link
Comet Lake U may not be the best example to cite, as compared with competing Zen 2 designs it's very hot and not especially fast. Hurry-up-to-sleep is a great strategy but they stretched it too far in an attempt to compete with AMD, who were offering more cores and better IPC.Mike Bruzzone - Friday, January 21, 2022 - link
Spunjji, Comet U was all about sitting on Renoir; on full run data Comet U production volume is 310% more than all of Renoir H/HS/U is an example of how Intel monopolizes channels financially on stuffing surplus into the channel. Today in WW channel Comet U = 127% more than Renoir mobile channel available and adding Ice U to Comet U 292% more than all Renoir mobile available. Intel did the same thing Tiger U v Cezanne H/HS/U. Also Tiger U production overage compounded in volumes on Intel relying on a quad to prove SF10 yield before moving to Tiger octa H and Sapphire Rapids 14C so there are a ton of Tiger U at 135% more than Renoir and Cezanne mobile combined available today in the WW channel and when adding Tiger H 203% more Tiger all up. mbSpunjji - Thursday, January 20, 2022 - link
"U-series having turbo TDPs of 55 W is a GOOD THING"Not sure I fully agree. Being able to boost high and drop down quickly is nice, but in combination with Windows' capricious attitude to randomly pegging a CPU core at 100% and the average OEM's tendency to under-spec their cooling to provide a "thin-and-light" experience, the end-user experience is often underwhelming.
"6+4 i7-12650H vs. 4+8 i5-12600H"
That one makes me chuckle too. The first is probably going to be one of the best all-around processors Intel have in their range (great for gaming with fewer E-cores sapping TDP, still enough of them to be decent for productivity) while the second is probably going to be their best-value productivity offering.
Spunjji - Thursday, January 20, 2022 - link
Intel may be the only company to beat the major car manufacturers at arbitrary product differentiation and the accompanying ability to sell customers more than they need.HideOut - Wednesday, January 19, 2022 - link
You guys have this part backwards. "Bringing up the rear is the Core i3-1220P, which offers only two efficiency cores and eight performance cores. Thie integrated graphics are also further cut back, with another drop to 64 execution units, but still the same 28-64 W power window. This Core i3-P looks very much like what the top Core i7-U processors will be in that configuration, but the Core i7-U are much more efficient." Should say 2P + 8 Etrenzterra - Wednesday, January 19, 2022 - link
I hope this means the pure Atom-based Pentiums and Celerons are gone for good?nandnandnand - Wednesday, January 19, 2022 - link
It's unknown, but I hope so too. Get at least 1 big core into every mini PC.Calin - Thursday, January 20, 2022 - link
Intel still has the Celeron N4020: 2 small (Goldmont Plus, some type of Atom) cores, no hyperthreading, 12 EU on graphics, 4MB cache.Going from that to the Celeron 7300 (1 big and 4 small cores, 48 EUs on graphics, 8 MB cache) you probably double the silicon area used (and so the price).
I don't see pure "Atom" style going away from Intel - as there are some places where Atom performance is enough (low end NAS, ...), and there are some places where even Atom price is a big part of the total hardware cost.
Mike Bruzzone - Friday, January 21, 2022 - link
Trenzterra, the supply chain rejected Jasper / Tremont / Elkhart Pentium, Celeron and Atom X. Only Silver N is seen in the channel and no Silver J and no embedded. Intel offers Atom for free in Core sales bundle and where Intel is now reconfiguring to producing for demand reforming the channel monopolizing tactic of producing for supply. On Intel cost optimizing Atom is an 'avoidable cost' mbET - Thursday, January 20, 2022 - link
Those 9W 1+4 cores 48 EU Pentium and Celeron chips could mean quite decent entry level laptops, even capable of light gaming.Spunjji - Thursday, January 20, 2022 - link
No boost on the Celeron will get ugly fast, but the Pentium will be a big step up from older Atom designs.Spunjji - Thursday, January 20, 2022 - link
They somehow found a way to make their naming schema even worse. Well done, Intel!My favourite is probably the i7s and i5s that are largely identical, excepting that one i7 that is significantly better than the rest. Different core counts for the same generation and same classification is something Intel really should have dropped some time ago.
I feel bad for anyone who ends up with a Celeron 7300. I don't understand what the point of disabling Turbo is, other than to punish the customer for being frugal. I guess at least that may be the one CPU that stays within TDP instead of heating itself up and then throttling like mad.
GeoffreyA - Thursday, January 20, 2022 - link
~1 GHz in the '20s ain't going to be pretty.Sunrise089 - Thursday, January 20, 2022 - link
Agreed. That Celeron to the Pentium one higher in the stack is the largest relative jump in performance for adjacent parts I can ever recall in a list of CPUs.Spunjji - Friday, January 21, 2022 - link
Indeed - they've been disabling turbo for no obvious reason besides punishing cash-strapped customers for a while now, but it's rare to see them do it when the clocks are already in the toiletnewblar - Thursday, January 20, 2022 - link
If not for pricing the 9 watt U parts would be good for ipad m1 android equivalent tablets. Also now with an engineer ceo again I feel like they could engineer a competitive soc for smartphones considering qualcomms struggles.abufrejoval - Thursday, January 20, 2022 - link
All this binning and salvaging used to be meaningful: With the advent or multi-core chips, selling partial defects e.g. with 3 out of 4 cores still good, not only made a lot of sense, it also made me feel quite warm and comfy about those handicapped chips finding a less demanding home than mine.But when I look at this plethora of chips and the nature of fabbing, it’s obviously quite impossible to achieve such consistent defect patterns across so many process generations, were always some cores or parts of the cache and iGPU either wind up defective or exceeding voltage threshold, which requires culling of cores, cache or GPU blocks in the name of TDP.
Significant engineering talent is obviously dedicated at producing sub-par parts from excellent chippery and I wonder what the psychological toll is on those poor bastards, who were made responsible for i3 or Pentium parts. Imagine them going to bed every night, hoping that one day they might be permitted to do i7, i9, i11 or i13! It’s just heart wrenching!
Sure, the original Pentium was a hothead and tended to unsolder itself, but it also set Intel apart from all 80486: How did it become a banner of shame? Ok, I’m drifting...
I am quite literally just fed up with Intel: too many choices, none satisfying. What I want is the ability to balance compute power versus noise and battery capacity, so just sell me a chip and levers which allow me to slide the bars between 5 and 45 Watts for a mobile and a NUC, 35 to 250 Watts on desktop, and let me adjust P1, P2 and max fan speed according to my current need at run-time.
Then I might want to order slices of P or E for my silicon pizza and that’s it!
Everything else is just torture designed to push me to i-max, which currently leads me to buy full chunks of AMD APUs or CCDs, from 15 to 150 Watts in my home lab, plenty more in the DC.
Spunjji - Friday, January 21, 2022 - link
"How did it become a banner of shame"They needed to get some distance from all those years of pushing expensive Netburst chips on people, lying about how good it was, and paying OEMs not to stock competitor parts.
Mike Bruzzone - Friday, January 21, 2022 - link
abufrejoval, Thank you for these astute observations on falling out on sort. Under Docket 9341 consent order disabling good parts is monitored and essentially not permitted as anticompetitive to target others with down bin parts that did not fall out of sort organically. At the same time unlike Ivy when the Intel practice of disablement at Intel became pervasive, by Comet, the cost became prohibitive on a full cores mask set design. The same is true with TSMC Zen octa high cost of disablement v. organic fall out. Intel has moved to dice area optimized masks which AMD is just beginning to do with 65/64xx Goby. The marginal cost to sort for Alder Lake Pentiums and Celeron is approximately $5 per component that have approximately a $15 all up production cost. Intel relies on statistical modeling to analyze projectable yield and the process is very sophisticated in actual use. mbabufrejoval - Saturday, January 22, 2022 - link
Thanks Mike for the reference, (https://www.ftc.gov/sites/default/files/documents/... which I've just read through.Here my naïve self had not been aware how the FTC had managed to head off the most egregious abuses, countered in gory detail in that document: I wonder if the document has been updated since, and what it should look like today.
But I can’t see any reference to intentionally disabled chips in there… is that perhaps in an updated document or is that simply a different number?
Anyhow, it fully explains how Intel wants to go back to feature licensing on their chips, because it completely avoids such a “do not fuse off working parts” decree, which obviously hasn’t been illegal in the mainframe world. Once they’re sure they didn’t botch the implementation (like currently the base overclock with the AL non-K parts), they’d certainly want to use that mechanism to squeeze their customers.
And to be honest, it wouldn’t be all bad, if it wasn’t then abused later. But I’m sure they’d become extremely creative, once it suits them.
Among the things I’d consider fair would be the ability to add and extract core/ECC/iGPU/clock unlocking codes from a SoC and resell them, consumer to consumer. So, if grandpop passed on his surfing machine to his granddaughter, she can turn it into a modest gaming rig buying somebody else’s unlock code off eBay. Sure, it could be tied to a generation, as long as they don’t increment those artificially to below a true technical cadence just to kick this out again.
Since these are purely digital assets, Intel shouldn’t be allowed to raise their prices through the roof three years later, just because they want to sell new chips.
Somebody better start planning a global rule book there, before it takes another ten years to pass the dozen different anti-competitive bodies to react.
On the economical side, selling octa-core chunks as a reduced performance dual-cores like IBM does for mainframes, can’t be fully repeated in PCs or consoles, I guess. With Intel having their own fab in theory their ability to switch wafers between M, X, XXXL or any other die variant on short term, give them a huge advantage vis-à-vis an AMD that has to wait in line at TMSC. But that only lasts as long as AMD/TMSC can’t cook up better post-fab/packaging stage modularity, which the 3D cache stack is hinting at, vs. Foveros etc. The CCD/IOD combo certainly stirred some engineering into action.
As a fan of good engineering I am very much looking forward to the solutions both sides with think up there.
But then come the bean-counters and that's where I am much less hopeful. With Charly warning against Pluton malware on AMD and Intel seemingly trying to leap ahead of AMD everywhere, all this doesn’t look like consumers will be served better with something they can use longer, which is the only progress I am really interested in.
Mike Bruzzone - Monday, January 24, 2022 - link
I’m enlisted by the FTC to monitor Docket 9341 consent agreement since 2010; AMD, Intel, Nvidia and Via, contracted by USDOJ to recover Intel Inside price fix for buyer return.No (hidden) Reference to “intentionally disabled chips”, FTC is amorphous in determinations.
Premeditated disablement of a good hexa to produce a quad or quad to produce a dual and for the specific purpose of flooding channel to limit competitive participation ‘organically’ falls under “no efficiency justification” and “raising cost” which is a Sherman Act Section 2 violation where Intel Inside rebate falls under Section 1 combining contracts “raising costs” as EUCC 37.990 described “unnecessary avoidable cost charge”.
FTC mandate pursuant disablement falls generally within 9341 consent agreement “Part IV A; respondent shall not invite, enter into, implement, continue, enforce any condition, policy, practice, agreement, contract, understanding, or any requirement . . .” Note, pursuant any OEM full line procurement the amount of down bin product is specified in the Intel OEM / IDM purchase agreement. Down bin is often freebie ‘sales close incentive’ Intel throws to any sales contract and specified for OEM/IDM 'supply chain' production planning.
The practice of disablement “raising cost” generally falls under Price < Cost and in this case on the cost of additional production steps associated disablement from good too bad to meet a contract supply obligation vis-à-vis what might organically fall out of sort.
Thank you for the mainframe world observation “licensing features” applied to Intel; aha!
On the AL non-K OC BIOS mod, Intel has done this before at risk production to goose sales out of the gate and then claims ignorance and no warranty responsibility; a spring board sales accelerator for certain end customer types that would not otherwise have purchases a K part.
On Intel locking out CPU upgrade I thought it prevented down grade i5 to i3 but not upgrade i5 to i7. Either way I do not believe in disabling features including Nvidia HLR on aftermarket value effect and unseen consequences into future essentially over segmentation or lock out.
EUCC traditionally leads in defining the rule book. I’m actually retained by Congress where once the FTC completes a project tends not to mind it. I do submit rule changes to Congress.
True, PC’s don’t have the IBM OLTP mark up; selling octa at reduced performance; enabling only two cores abd a lot of cache for databases. Note Epyc can be characterized and its CCX and L3 partitioned like a ‘many instances” GPU but AMD does not talk about it outside a clutch of developers.
Quad Scalable from Intel low core count deca can also be characterized (or disabled) as a quad for maximum L3 per core so it does happen in the high cost : price / margin systems market.
Intel coming out of its sabotage post Comet revitalized its mask shop at Tiger lake quad for die size optimized mask set. AMD with cash again started at GPU Beige Goby where in relation Intel AMD has to go to 3rd party mask cost.
Thanks for your thoughts.
I too look to enterprise serving customers on responsibilities in technocracy "meet the needs of customers and society as cost effectively as possible", stewardship cognizant welfare", "complimentary participation, partnership, cost and reward sharing", planning for an adequate economic return on investment", "steward in the countries of industry operation", "sustainable growth models that make economic sense", Watson Jr.
mb
ChrisGX - Saturday, January 22, 2022 - link
...too many choices, none satisfying.Yes, that could easily be the epitaph on Intel's tombstone.
ChrisGX - Thursday, January 20, 2022 - link
Apple, I believe, offers one chip to satisfy all of the use cases covered by this long list of chips. It is called the M1. If Intel was serious about consumer needs it would stop playing the pointless market segmentation game -- it is all sound and fury signifying nothing.If Intel was doing right by consumers it would cover all of these use cases with two chips -- one competitive with the M1 and another non-premium part that is cheaper and more energy efficient that is suited to users who don't need the premium option.
KPOM - Thursday, January 20, 2022 - link
To be fair, they have the M1, 3 variants of the M1 Pro (8/14 Core CPU/GPU, 10/14, 10/16) and 2 variants of the M1 Max (10/24, 10/32), so they have 6 chips, which is still much fewer than 20.ChrisGX - Thursday, January 20, 2022 - link
Yes, and the parts you mention -- M1 Pro and M1 Max -- push well into i9 territory. The higher end versions of the M1 (or, no doubt, M2) aren't targeted at thin and light laptops. Those chips are to be compared to high end Intel parts not the chips mentioned in this article. The M1 and/or M2 represent the right point of comparison for the ADL chips covered in this article.web2dot0 - Thursday, February 10, 2022 - link
Totally disagree. Intel's CPU is targeting for thin and light because they don't expect those CPUs to be paired with dGPU.On the other hand, Apple's M1Pro/M1Max are paired with 16c/24c/32c GPU which are MUUUUCH faster than Intel's iGPU ... Apple take the GPU in the account when they configure their laptops.
To compare Apples to Apples, M1Pro/M1Max CPU has comparable power draw (albeit even lower power consumption since they don't have "peak power draw" figures. Their CPU simply max out at 30W (40W CPU SoC total draw including RAM,MB,etc..).
This is simply on par with Intel's U/P series "thin and light" CPU's power numbers to be honest.
kwohlt - Thursday, January 20, 2022 - link
On ARM, Apple has 2 laptops, 1 of which comes in 3 sizes. And they have one mini computer that uses laptop parts. Of course they're going to have significantly less CPU SKUs than Intel, who makes a wide range of chips for everything thing from embedded devices, to thin and lights, to standard laptops, desktop replacements, desktops, workstations, servers, etc.Of course they're going to have significantly more SKUs not only for different types of devices, but for different price points within these types.
If it made sense for Intel to offer only 2 chips to meet all client computing needs and budgets, they would. But it doesn't at all.
ChrisGX - Thursday, January 20, 2022 - link
Well, the M1 is for thin and lights and no chip mentioned in this article is for embedded devices, desktop replacements, desktops, workstations or servers. Your comments on those points are irrelevant. The ADL chips covered in this article are for thin and light laptops. And, that huge profusion of chips is there for the sake of Intel's margins, nothing more. There is no benefit for users in any of it.How exactly to users lose out by choosing the M1 over all of these ADL parts? (Please note, I have conceded that in terms of genuine consumer needs a more energy efficient and lower power option would be useful. Perhaps, I'm wrong, maybe even two non-premium options might be needed.)
Personally, I wouldn't equate Intel's profit margins with user needs. If that was valid Intel oughtn't to be experiencing any problems today but it is one of the few major semiconductor companies that isn't growing.
Calin - Friday, January 21, 2022 - link
"workstations, servers"The servers and workstations have their own processors, not included there. Usually they have a bit more memory bandwidth for the High Performance Computer (one more fast memory channel) or several more channels of slower memory for servers.
Spunjji - Friday, January 21, 2022 - link
Yup. Two chips and perhaps three bins of each chip for good yields and some price distinctions - fastest, not-as-fast, fast-enough.Sahrin - Thursday, January 20, 2022 - link
…AMD is able to run a Zen 3 core at 3.5GHz at less than 3W, so it’s really hard to imagine a situation where any number of P cores which are probably ~3W themselves combined with a large number of E-cores which are substantially but not magically more efficient is doing much better in power terms,iphonebestgamephone - Friday, January 21, 2022 - link
Which zen 3 core is this?Da Kat - Sunday, January 23, 2022 - link
>Bringing up the rear is the Core i3-1220P, which offers only two efficiency cores and eight performance cores.I BELIEVE that should be the other way around, 2 performance, 8 efficiency..
ptr2ptr - Wednesday, February 2, 2022 - link
Im new here trying to understand. Could someone explain why Intel makes so many variants of its processors even within i7, or i5 or i3 ?Dr VHR Krishnan - Thursday, February 17, 2022 - link
Please add a share button in all these articles.I would like to send a copy of some of these excellent review articles to my friends.
For example, the links should be sent in WhatsApp, Facebook, Twitter etc.