New cores (A15)

[crazy dave]

I meant the CPU cores, sorry. So far it looks just like Firestorm at 3.2ghz. Why claim new cores if you just ship old cores at higher clock?

I understood you and I agree ?. This is odd. Also, weird that all the iPhone 13 GB scores have disappeared for me so I can't go back to look at the numbers.

 

[deconstruct60]

so are you sure guys we will have just 3 segments for the silicon macs? i mean the Mn series for the macbook air, 24" imac, ipad pro etc, the middle of the road MnX for the 14" 16" macbook pro, bigger imac and the silicon for the mac pro?

You are leaving out the Mini; which could get split if there was more than one case.

MBA , MBP 13" , Mini , iMac 24"
MBP 14" , MBP 16" , Mini (?) [ doesn't look like these are all same die but can be shared segment ]
iMac "big", Mac Pro "half". [ same segment. Not same die in same package ]

So only 3 segments? i am ok with this too if it will be so but just wondering
i wonder if the mac pro will be very priced configurable with 2x Mn or 3x Mn(for the base prices of 2 possible configurations) , other step up in perf and price - config 2x MnX or 3x Mnx and so on OR will be different SoC's entirely ?!

Mac Pro Starting from $2000 - dual Mn SoC with up to 32 gb Ram, up to 4T ssd etc
$2500 - triple Mn SoC with up to 48 gb ram , 4T ssd etc

A dual die SoC perhaps ( two dies specifically designed to be paired in one package) . But two SoCs on a logic board. That is in the likelihood of some meteor coming and striking the center of Apple loop building's inner courtyard in a bullseye landing.


Those prices for a Mac Pro is in picking a wining lottery ticket range of possible. That is probably were a Mini would be. Currently if pick

Mini upper 3.2 6 core Intel , 32 GB , 512GB , 10GbE. -----> $1,999

If Apple made a "headless" iMac bigger screen, then could have.

Jade2C ( 16P+4E+64G ) , 32 GB. , 512GB. , 10GbE -----> $1,9999

presuming Apple just reused the current Mini enclosure for this. No slots. No additional internal drives. Wouldn't really be a "Mac Pro" in any sense of way it is used now.

Or possibly.

JadeFull ( 8P + 2E + 32G ) , 16 GB , 1TB , 10GbE ---> $1,9999

Mac Pro - $2999dual MnX SoC with up to 64 gb Ram, 8T ssd etc
- $3500 triple MnX SoC with up to 128 gb ram and so on

is this possible ?

8TB of SSD system at $2,999 from Apple? Again in the "alternative universe" possible. Apple charges $400/TB. They discount after 4TB , but 8TB is $2,400 alone. RAM has a very similar high mark up.

Not likely to see "triples" either. Perhaps some quads with cores turned off but not triples.

to be so config from a cheap price starting and supporting up to over $5000 based on the SoC and number of them?

The M-series processors aren't going to bring any major price drops to the iMac or Mac Pro. Pragmatically for the Mac Pro, there may be some price increases if the RAM is soldered to the SoC. End users will have to buy Apple RAM and it will probably be pretty tightly coupled to core count bought.

 

[Jorbanead]

They usually don’t announce any events until two weeks in advance. And last year’s event was end of October or into November iirc.

For in-person events it’s two weeks. Since covid, they’ve only announced them one week in advanced. Last years mac event was in November, but this was because Apple had events in September and October as well. It was the first time in apples history they held a November event, so it’s more likely they will hold an October event this year.

Your ideas of Mac Pro(s) aside, the iMac Pro definitely deserves to be on that list.

The iMac Pro was discontinued. That’s not to say they won’t bring it back, but we have no idea what the TDP would be. Surely they’re not going to use the old design if they did bring it back, so it’s likely the TDP would be much less. If they were going to do an iMac Pro, I would imagine it would look like the Pro Display XDR with a chin.

 

[Hexley]

Conjecture on how an A15 cores will behave in a M1x/M2 MacBook Pro 16"/13" 4port, iMac 27" or larger & Mac mini 4-port

Having said that if I knew I'd be stuck at home for the last 18 months I may have opted to keep my late 2017 64GB base iPhone 8 Plus rather than get a late 2019 base 64GB iPhone 11 Pro Max.

Downgrade to a SIM-only $1-12/month text/call/data plan until vaccine passports/IDs are phased out by 2025. Then upgrade to a late 2025 iPhone with 256GB base storage

 

[deconstruct60]

I think he meant the CPU cores so it should read how is avalanche different from firestorm? Which means the other points don’t really apply

But that misses the forest for a tree. Apple sells SoCs. It is the system that the SoC is for. That is one of the major points. I didn't miss the myopic stance. I'm pointing out that the myopic stance of looking at these System on a Chip as not a System.

The CPU cores being some almost complete equivalence with the SoC in the mindset is a big source of the disconnect.

and 1) is only true *in low power mode* which most people don’t keep their phones in constantly. Otherwise, the default is that it goes faster for more power and for pretty much the same power increase you’d expect from an upclocked firestorm core on N5.

When the SoC thermal throttles the clocks will scale back also. The faster clocks run as long as thermally can get away with it. Transition over to a M-series system where someone might throw a longer term , non trivial workload at them for long periods of time.

Again trapped in the context of brief "drag race" (over in s relatively short amount of time) benchmarks as being the only driver of what is "good" or "not good".

macOS will get better "Low Power Mode" in macOS 12 (Monterey). Another issue is when have more than just two P cores, do all of them really need to only on in "full blast" for most workloads? If only have a very limited number of P cores then may have to juggle multiple 'P' workloads so need to timeslice. If the system can hand out work to more units , that timeslice need isn't quite as great. A smarter power management ( OS + SoC + etc) could cover a spectrum between "full blast" and "Low Power mode" without the user having to do tons of grunt work handling context switches.

For example is "spill" an E core workload onto a P core then don't need to run P at max power consumption to cover that. Even covering two E cores may not need full turbo clock.

 

[jdb8167]

I understood you and I agree ?. This is odd. Also, weird that all the iPhone 13 GB scores have disappeared for me so I can't go back to look at the numbers.

You can search under iphone14,2 to find the iPhone 13 pro, iphone14,3 for the iPhone pro max, iphone14,4 for the iPhone 13 mini, and iphone14,5 for the iPhone 13.

Edit: The iPad mini 6 is ipad14,2

 

[TSE]

For the love of god, does ANYONE know when these laptops are coming? Please! Anyone?!?! I am running out of patience - My Dell is making me pull my hair our and go absolutely bonkers!

 

[crazy dave]

But that misses the forest for a tree. Apple sells SoCs. It is the system that the SoC is for. That is one of the major points. I didn't miss the myopic stance. I'm pointing out that the myopic stance of looking at these System on a Chip as not a System.

The CPU cores being some almost complete equivalence with the SoC in the mindset is a big source of the disconnect.



When the SoC thermal throttles the clocks will scale back also. The faster clocks run as long as thermally can get away with it. Transition over to a M-series system where someone might throw a longer term , non trivial workload at them for long periods of time.

Again trapped in the context of brief "drag race" (over in s relatively short amount of time) benchmarks as being the only driver of what is "good" or "not good".

macOS will get better "Low Power Mode" in macOS 12 (Monterey). Another issue is when have more than just two P cores, do all of them really need to only on in "full blast" for most workloads? If only have a very limited number of P cores then may have to juggle multiple 'P' workloads so need to timeslice. If the system can hand out work to more units , that timeslice need isn't quite as great. A smarter power management ( OS + SoC + etc) could cover a spectrum between "full blast" and "Low Power mode" without the user having to do tons of grunt work handling context switches.

For example is "spill" an E core workload onto a P core then don't need to run P at max power consumption to cover that. Even covering two E cores may not need full turbo clock.

I don’t think anyone is missing that there are changes beyond the cpu core, but are commenting on how the cpu cores haven’t seemingly changed which is different from every other release and unexpected given that the cores have new names - usually indicating a new generation.

At the moment we only have 1 Chinese user’s efficiency benchmarks with a limited set of tests - especially for the cpu. So far no one seems to have run anything besides Geekbench on the cpu. We can only go by the information we have which is again limited. As far as I know we don’t have endurance tests on the avalanche cores or many tests of the E cores at all. But so far the avalanche cores appear to be little more than upclocked firestorm cores. There’s no indication that the cpu cores have anything different about them except at the designated low power mode which is not the same thing as normal thermal throttling.

Edit: yes in the Mseries the core variants may have great properties but we don’t have those yet and can only go by what we’re seeing in the phone (provided that the linked results hold true when others test it)

 

[altaic]

I believe there will be only two chips/platforms: vanilla M-series for low-power consumer Macs and a scalable professional chip for everything else. Bigger Macs will use multiple scalable chips to build more powerful systems.

I’m on the fence about 2 vs 3 Mac SoC lines.

Right now, we’ve got the M1 which targets the high end tablet and low end laptop/desktop markets. I could see Apple making a monolithic SoC that targets mid/high end laptops and mid-range desktops. Then for high end desktops, I think a multi-chiplet SoC would be a must.

OTOH, the M1 already has an interposer, so we’re already sort of out of the monolithic SoC territory. It would be sensible for Apple to go with chiplets for the rest of the lineup if there is no additional power efficiency penalty. Say, 1-4 compute chiplets, an interface (display, thunderbolt, and USB) chiplet, and RAM. That would keep the yields up while offering great configurability.

I’m leaning toward the latter because I think the next Mac SoC will not be low production. I have a sneaking suspicion that Apple will capitalize on the disastrous PS5 and XBox Series X launches and subsequent production issues, going all-in on Mac gaming.

It has been difficult in the past to get AAA titles on board, but I think that may have changed. I read a statistic a month or two ago that roughly 2/3 of the PS5s out there are sitting in scalpers’ basements— game developers have an extremely limited customer base because of that, and are financially hard hit.

Furthermore, from a customer’s perspective, what’s better: a $1500 console, or a $2000 computer that has many uses, so long as they both have the AAA titles? A lot of people need a computer these days, and a lot of people want a new console. A new console plus a computer is far more expensive than a gaming computer.

I think the scalpers have changed the equation and Apple would be wise to seize the opportunity.

Edit: I’m not basing my thoughts on Mac gaming on rumors— just the observation that the next Mac SoCs should be very capable, and that game developers may be more amenable these days.

 

[TSE]

I’m on the fence about 2 vs 3 Mac SoC lines.

Right now, we’ve got the M1 which targets the high end tablet and low end laptop/desktop markets. I could see Apple making a monolithic SoC that targets mid/high end laptops and mid-range desktops. Then for high end desktops, I think a multi-chiplet SoC would be a must.

OTOH, the M1 already has an interposer, so we’re already sort of out of the monolithic SoC territory. It would be sensible for Apple to go with chiplets for the rest of the lineup if there is no additional power efficiency penalty. Say, 1-4 compute chiplets, an interface (display, thunderbolt, and USB) chiplet, and RAM. That would keep the yields up while offering great configurability.

I’m leaning toward the latter because I think the next Mac SoC will not be low production. I have a sneaking suspicion that Apple will capitalize on the disastrous PS5 and XBox Series X launches and subsequent production issues, going all-in on Mac gaming.

It has been difficult in the past to get AAA titles on board, but I think that may have changed. I read a statistic a month or two ago that roughly 2/3 of the PS5s out there are sitting in scalpers’ basements— game developers have an extremely limited customer base because of that, and are financially hard hit.

Furthermore, from a customer’s perspective, what’s better: a $1500 console, or a $2000 computer that has many uses, so long as they both have the AAA titles? A lot of people need a computer these days, and a lot of people want a new console. A new console plus a computer is far more expensive than a gaming computer.

I think the scalpers have changed the equation, and Apple would be wise to seize the opportunity.

I respectfully disagree. Gaming does not line up with Apple's values, customer base, or markets. Apple has never and will never try to make a gaming device. And don't you dare mention the Pippin.

 

[cmaier]

I meant the CPU cores, sorry. So far it looks just like Firestorm at 3.2ghz. Why claim new cores if you just ship old cores at higher clock?

Because they didn’t

 

[altaic]

I respectfully disagree. Gaming does not line up with Apple's values, customer base, or markets. Apple has never and will never try to make a gaming device. And don't you dare mention the Pippin.

I suppose we’ll have to agree to disagree, then. I have the opposite view on all your points except the unmentionable.

Also note that I’m not predicting Apple will make a game console. I’m predicting Apple will get AAA titles for their very capable computers. And thus said computers would not have low production quantities, so chiplet-based SoCs would be favorable.

 

[cmaier]

I suppose we’ll have to agree to disagree, then. I have the opposite view on all your points except the unmentionable.

Also note that I’m not predicting Apple will make a game console. I’m predicting Apple will get AAA titles for their very capable computers. And thus said computers would not have low production quantities, so chiplet-based SoCs would be favorable.

The best computer games used to only be on Apple ][. Apple gaming ftw…

(I agree with you, btw)

 

[TSE]

Apple will never have a full
Library of AAA games. Maybe one every couple years like they have since the beginning of time. Keep this quoted and check in on this in a couple years because there’s no way Apple is going to have a developed game library!

 

[altaic]

Apple will never have a full
Library of AAA games. Maybe one every couple years like they have since the beginning of time. Keep this quoted and check in on this in a couple years because there’s no way Apple is going to have a developed game library!

No need. Apple already has a “developed” game library, and that’s not going away in a couple years. If the new SoCs are as good as the expectations, AAA titles will follow. I’d hope that Apple would help game studios along, but who knows.

 

[Serban55]

8TB of SSD system at $2,999 from Apple?

didnt said that you get 8T at that price...maybe you misunderstood , others got it right. We will wait and see, no more than 1 year

 

[Serban55]

For the love of god, does ANYONE know when these laptops are coming? Please! Anyone?!?! I am running out of patience - My Dell is making me pull my hair our and go absolutely bonkers!

this year, maybe...but high hopes for October to get at least some announcement

 

[Serban55]

No need. Apple already has a “developed” game library, and that’s not going away in a couple years. If the new SoCs are as good as the expectations, AAA titles will follow. I’d hope that Apple would help game studios along, but who knows.

Agree, with a better gpu, the developers will follow and see the potential, not all of course, but most of, its just a matter of time, from what ive heard some developers already went native m1

 

[leman]

Because they didn’t

I think there is still a chance that A15 contains support for some newer technologies (e.g. SVE), which are not yet exposed to the developer because *politics*, but yeah, the evidence so far looks bleak.

I respectfully disagree. Gaming does not line up with Apple's values, customer base, or markets. Apple has never and will never try to make a gaming device. And don't you dare mention the Pippin.

I agree that Apple won't be making a gaming device any time soon. But then again, they spend the last few years designing and building some first-grade gaming APIs and gaming hardware, so gaming seems to be a part of their overall strategy. I have little doubt that the upcoming prosumer Macs are going to be great gaming machines, rivaling mid+high-end gaming laptops.

The customer base and market issues you mention are real however. And they will not go away over night. My hope is that the culture will change gradually. There are great games being released for Mac and as people start recognizing that gaming on Macs is not only possible but enjoyable, more games will appear and the negative sentiments will slowly erode.

 

[altaic]

The customer base and market issues you mention are real however. And they will not go away over night. My hope is that the culture will change gradually. There are great games being released for Mac and as people start recognizing that gaming on Macs is not only possible but enjoyable, more games will appear and the negative sentiments will slowly erode.

Leman, I always appreciate your insight and analysis, so I’m curious if you’d weigh in. The crux of what I was getting at earlier is that the gaming industry has had hard times recently despite pent up demand, so Apple’s position is different compared to previous dealings with developers/studios, therefore I think an overnight industry epiphany is quite possible. Apple should have significant negotiation advantage vs the console industry, and they’re poised to fulfill a need, so I expect them to do so.

To be more on topic, I expect a flexible line of chiplet-based SoCs because of a wider market.

Full disclosure: I am not the kind of person who thinks the next thing will solve my needs or wants (gaming being neither of those), and regarding gaming I have previously been very dubious of Macs suddenly becoming a gaming mecca— yet I am an investor. But now I think the equation has changed. The upcoming SoCs may fill a vacuum.

 

[leman]

Leman, I always appreciate your insight and analysis. I’d like to add that the crux of what I was getting at earlier is that the gaming industry has had hard times recently despite pent up demand, so it is different than previous dealings with developers/studios, and an overnight industry epiphany is quite possible. Apple should have significant negotiation advantage vs the console industry. They’re poised to fulfill a need, so I expect them to do so.

To be more on topic, I expect a flexible line of chiplet-based SoCs because of a wider market.

I am not the kind of person who thinks the next thing will solve my needs or wants (gaming being neither of those), and regarding gaming I have previously been very dubious of Macs suddenly becoming a gaming mecca. But now I think the equation has changed. The upcoming SoCs may fill a vacuum.

I doubt it's going to be that simple or quick though. For example, it seems that Unreal Engine is all but dead at this point — Epic is clearly using it as a blackmail material and Apple is not going to dignify that.

Personally, I would hate if Apple simply buys off some studios to then pump bunch of out soulless titles with zero artistic value, like what they did with Apple TV. Consoles are alder bad enough, that's not the kind of gaming I am looking for. In the meantime, most really interesting games, made by passionate studios, that have been released or announced this year appear to have a well-optimized Mac port. All I can say is that I feel confident in ditching Windows for my gaming needs already today. Games I like to play are available on macOS, and for the very rare exception I have GeForce Now.

 

[altaic]

I doubt it's going to be that simple or quick though. For example, it seems that Unreal Engine is all but dead at this point — Epic is clearly using it as a blackmail material and Apple is not going to dignify that.

Personally, I would hate if Apple simply buys off some studios to then pump bunch of out soulless titles with zero artistic value, like what they did with Apple TV. Consoles are alder bad enough, that's not the kind of gaming I am looking for. In the meantime, most really interesting games, made by passionate studios, that have been released or announced this year appear to have a well-optimized Mac port. All I can say is that I feel confident in ditching Windows for my gaming needs already today. Games I like to play are available on macOS, and for the very rare exception I have GeForce Now.

Thank you for your perspective— I share a similar attitude toward gaming (also @cmaier, do you remember Bard’s Tale? Might and Magic was shortly thereafter which taught me how to reverse engineer so I could cheat— never underestimate the power of a kid facing crippling boredom).

Decoupling personal preference though, I think the AAA crowd may have an epiphany given the current+forecasted circumstances. I wasn’t suggesting that Apple helping studios along would mean buying them — I very much hope not — more, providing assistance porting and developing interfaces&tooling to make porting easier. Vulkan is more than half (or at least 1/aleph) of the way there, IMHO.

 

[leman]

Vulkan is more than half (or at least 1/aleph) of the way there, IMHO.

I am actually worried about Vulcan‘s future. Paradoxically enough, progress of Proton makes it easy for the devs to just stick to DX12 and windows and get Linux support for free, and I expect that Vulkan in the mobile space will be superseded by WebGPU for all practical needs. Where would it leave Vulkan? As an API GPU vendors have to implement so that DX emulation works on Linux?

 

[altaic]

I am actually worried about Vulcan‘s future. Paradoxically enough, progress of Proton makes it easy for the devs to just stick to DX12 and windows and get Linux support for free, and I expect that Vulkan in the mobile space will be superseded by WebGPU for all practical needs. Where would it leave Vulkan? As an API GPU vendors have to implement so that DX emulation works on Linux?

Thank you for your thoughts. My thinking is that it’s sort of cycled back to the wild west for graphics APIs (and multitudinous APIs in general). It makes me think about the javascript problem for language theorists— just apply more people and let the situation evolve. Turns out evolution can be pretty ugly. There was a ray of hope when Nvidia pseudo-endorsed Vulkan, but, yeah, academics and industry stanchions haven’t seized the opportunity there, so my hopes of some kind of unification seem misplaced.

 

[Romain_H]

I am actually worried about Vulcan‘s future. Paradoxically enough, progress of Proton makes it easy for the devs to just stick to DX12 and windows and get Linux support for free, and I expect that Vulkan in the mobile space will be superseded by WebGPU for all practical needs. Where would it leave Vulkan? As an API GPU vendors have to implement so that DX emulation works on Linux?

Sorry if I am asking the obvious: is it know how Proton works? Apparently it is proprietary - how would the situation change if they released it as open source? It would probably make Vulcan obsolete; however, Vulcan and Proton are kind of redundant - its just the licenses are different, so Proton is not publicly available.

Or is there another issue I am unaware of?