Raptor Lake Q3 2022, Meteor Lake Q2 2023, both of which appear on time. Intel faces a do-or-die timeline against AMD.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Intel Alder Lake vs. Apple M1
- Thread starter leman
- Start date
- Sort by reaction score
Nah the laughable bit is that Mac buyers,myself included, have had to do a lot of hand waving to excuse Apples middling performance with extremely high prices for years. This paradigm change with the M1 chips brought out a lot of hilarious people strutting around here. Now that they have had their performance feather plucked from their hats we’re on to perf per watts wanking. Cute story though.
I completely agree!! This is why I have only had Mac's for short periods of time in the past. You traditionally in the past always got better hardware or features not available in Mac's or the performance difference for the cost was not worth it for me. I am OS agnostic and am at home in any OS so hardware is really important to me.
Now with these new MacBook Pro's that has all changed. There is no alternative to the MacBook Pro's in terms of overall package and performance. They just dropped the mic on all the people who Like me complained about the hardware limitations in the past. Also cost is high but again in terms of specs and performance the base model 16" is unbeatable. You can get a brand new Dell XPS 15 laptop with same ram and ssd Nvidia 3050ti and UHD its display for $2399 compared to the MacBook Pro 16" at $2499.
Dell XPS 9510 Laptop | Dell USA
Bring your ideas to life with the new XPS 15. A refined 16:10 InfinityEdge display with 100% Adobe RGB color delivers brilliance that makes your creations as vivid as your imagination.
www.dell.com
For $100 more you get so much more with Apple it is crazy. That is why I now have a MacBook Pro 16" base model and I am not looking back!! At least right now, nothing comes close to these new MacBooks and I don't think that is going to change any time soon!! So happy to see Apple finally listening to what customers want and kicking ass in the Apple silicon!!
Just curious, what happens to your workflow when your laptop dies after 2 hours(or less) and you don’t have access to a charger.When I think of my workflow, powersavings is always my 1st concern!
He’s not worth it. He’s pretending that the PPW wasn’t what people were most excited about in the M1 to sell the idea that a desktop chip besting a mobile/AIO chip by 5% while using 3-5 times more power is something that Apple users should feel bad about. It’s just toxic for the sake of being toxic. He isn’t looking for a discussion just a fight.
Yeah we’ll see what happens though (for everyone mind you) … timeliness is likely to be rare. Even Intel’s roadmaps include a “what if this all goes wrong” part, which is good in some ways but a recognition that things are going to get harder from here.
I think Intel is probably on time with Raptor Lake and Meteor Lake as of current roadmap indicated. Raptor Lake is already a filling solution for a delayed Meteor Lake (initially late 2022 bound).
I would agree that anything announced for next year is *probably* okay but anything further out than that I grow increasingly suspicious of. Nothing stops MTL from being delayed again! For example: Apple will probably release a new iPhone chip every year but I don’t expect fabrication node and uarch to change as dramatically from now on every year. The upcoming A16 may or may not be a big change but regardless the A15 is a tweaked A14 on a slightly better node. Some nice gains still, but not the big jump we’ve almost come to expect. I foresee more generations like that in our future and not just for Apple or Intel. Maybe I’m too pessimistic.
I should add that while I’m pessimistic about new fabrication nodes allowing for really big uarch changes, I do think we’re entering another era of chip innovation. It just can’t come strictly from upgrading nodes anymore. Packaging, accelerators, and innovative core designs are going to be even more important.
Last edited:
If Meteor Lake was to be delayed again, we would have known already since leaked roadmaps had Raptor Lake for at least 2 years for now.
It depends on the workload and the workflow and if I can do the same work with a small time penalty on a mobile computer that I have with me on location then I’m ahead of the game as I won’t be tied to an office desk and some slab of desktop tower that I really cannot lug around. The MacBook Pro with M1 Pro or Max is a mobile workstation. Intel doesn’t have an answer for this right now. We’ll see when the H-Series Alder Lake CPUs are released how they compare.
That doesn’t stop it from being delayed again. We’ve seen chips pushed back at this point. Ice lake Xeons come to mind. Though I think Intel’s roadmaps are touch more resilient now and have more “if this goes wrong we backport to this node”. Hopefully for Intel’s sake their future designs are more amenable to being so than Ice Lake’s was to make Rocket Lake.
I think we have been talking about consumer chips from Intel, not the server. Intel did keep a yearly cadence on their consumer lineup regardless of their process node awes and if there were delays, Intel codenames were leaked way ahead (Comet Lake and Rocket Lakes (14nm rehashes for 10nm delay), and Raptor Lake (10nm refresh for 7nm Meteor Lake delay). If Meteor Lake was to be delayed, we would have had another code name leak out after Raptor Lake already, but hasn't happened so I can say Meteor Lake is still coming out after Raptor Lake.
Yeah but if MTL is back ported to 10nm refresh that to me is still a delay. It’s supposed to be a 7nm design and the last back port didn’t go so great (and the core name was different).
Look I’m not saying it will absolutely be delayed. Everything could be fine. But two years out? I’m saying there’s still room for things to go wrong and for a new codename to be “leaked”. And this is what I’m trying to get across: The pattern is likely to change because everyone is likely to struggle. And while yes it was a server chip there is even still precedent for what I’m saying with Ice Lake. That could start to become the norm. And to me, Intel’s released, not leaked, roadmaps reflect that reality.
To be clear: I’m not bashing Intel here, everyone will be in the same boat. And I could very well be wrong about what’s going to happen with fabrication over the next few years. Maybe everyone will hit their roadmaps. From what I’m gathering from people in the field though: That’s going to get increasingly harder. Oh and much more expensive. ?
Last edited:
MTL is already taped-in on their own 7nm(now Intel 4) so I don't expect backporting will happen like Rocket Lake. Two years is not exactly a long time since the roadmap plan should be decided 3 years ahead of time. Which means if delay was to happen, it would have happened already, but it seems like Meteor Lake coming after Raptor Lake pretty much set in stone right now.
Starting with Meteor Lake, Intel will start doing multiple chiplets like how AMD is doing. It means only the compute node will be based on their most advanced node and due to much smaller size, it will be less prone to yield problems vs monolithic chips.
Weren’t Ice Lake Xeons supposedly being taped out but in the end were delayed multiple times? And those were on the roadmap well within three years when they got delays. Admittedly delaying an entire uarch is a bigger jump than just server chips but fabrication nodes can suffer from all sorts of delays which affect say mobile chips differently than desktop ones to pick a random example
. Again I’m not saying MTL will definitely be delayed in any way. I agree that chances are that if it’s taped out less will go wrong and that less is likely will go wrong with chiplets than monolithic designs. I agree with that.
I did edit my previous post to make clear I’m not just talking about Intel here. TSMC and Samsung are also likely to run into issues too. Supposedly TSMC already has already as 3nm is delayed. The main point is not whether any individual chip, be it MTL or the A17, will be delayed or changed into something they weren’t supposed to be. It’s that delays overall will happen and that roadmaps, leaked or not, are even less likely to reflect reality the further out they go than they were before.
Weren’t Ice Lake Xeons supposedly being taped out but in the end were delayed multiple times? And those were on the roadmap well within three years when they got delays. Admittedly delaying an entire uarch is a bigger jump than just server chips but fabrication nodes can suffer from all sorts of delays which affect say mobile chips differently than desktop ones to pick a random example
Yeah, Intel did. And we also had code names for 14nm server product fill-ins leaked out before that. (Cascade Lake and Cooper Lake). So we already knew Ice Lake was going to be delayed multiple times based on that lol.
Intel decoupled the architectures from their nodes so they won't have troubles like 14nm again. Rocket Lake was a dud but it kinda did show Intel can now move their architecture on different node in some degree. Considering Intel is also supposedly also using TSMC 3nm for their future products, looks like they got their backup plan already if somehow their own "Intel 4" node doesn't work out. Even though TSMC 3nm is delayed, it is still going to be mass produced in 2023.
This has never happened and it won't happen. IF this does happen to you, you've got bigger issues than the perf/watt ratio of your computer.
I think we’re kind of in agreement here I’m just slightly pessimistic about the outcome (and not just for Intel).
No, no I'm not. Who are you again? Why do you get to tell people what I'm saying/doing. A very vocal minority started waving the pef/watt flag. You always pop up whenever somebody has a view that doesn't go along with the heard. Why is that? Don't you have a ex chip engineer to simp for on another forum?
I'm curious about the cost of relatively higher energy computing. I costed that the 14" MBP would cost in today's money, $50 more for its shorter battery life. It was about $US1.43 I think a month extra. No big deal.
But I wonder about what the power cost differentials are for computing? While its a difficult thing to work out, on mass it should be easy. Then divided into households and I guess, computers.
I'd love to see some figures.
Hmmm ... my house runs on solar in the day time ... maybe I should buy a battery back up UPS for the desktops so they run on battery power when the sun goes down ... that twin CPU PowerPC pumps out a lot of hot air going out its back when its awake.
I wonder too whether mechanical drives use less energy than SSDs .... the SSDs run so hot, perhaps that means they use more power per GB? Is it more responsible to buy a 1 TB MBP rather than a 4 TB model, and have temporary type C drives which are mostly turned off?
But I wonder about what the power cost differentials are for computing? While its a difficult thing to work out, on mass it should be easy. Then divided into households and I guess, computers.
I'd love to see some figures.
Hmmm ... my house runs on solar in the day time ... maybe I should buy a battery back up UPS for the desktops so they run on battery power when the sun goes down ... that twin CPU PowerPC pumps out a lot of hot air going out its back when its awake.
I wonder too whether mechanical drives use less energy than SSDs .... the SSDs run so hot, perhaps that means they use more power per GB? Is it more responsible to buy a 1 TB MBP rather than a 4 TB model, and have temporary type C drives which are mostly turned off?
I suppose because it’s obvious I shouldn’t say it but saying it usually helps calm the conversation down.
Exactly what were you hoping to gain from these posts? Reasonable discussion? I’ve seen you do this before. You post inflammatory stuff and get into “internet yelling matches” with other people on here. Why? What’s in it for you? I’m just curious.
I mean if you really are looking for a conversation then hey I apologize and let’s have a conversation about why performance per watt is important for absolute performance in laptops and even desktops. But you aren’t writing like you want to have that conversation. You’re writing like you want to get into an insult fight about it.
disagree
(I thought there was a thumbs down reaction that would avoid cluttering with a reply to this, but I can’t find it)
This has been a core part of Apple’s pitch for M1 and the reason most of the technically minded here have always expected the M1 to be able to continue improving in raw performance as Apple targets different products.
