Unfortunately it’s a little bit hard to trust what you hear from these guys when it comes to power consumption. I think that because this parts go into a variety of different OEM systems, they will tend to be pushed to the higher wattage curve. They generally tend to advertise the performance so you can’t blame them for doing that can you…?
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.
The case of an Ultra chip using two M2 Pro chips
- Thread starter Rickroller
- Start date
- Sort by reaction score
A lot of PC benchmark reviewers have reported that Intel's published TDP does not reflect their power meter's measurements. It tend to go beyond that. Not surprising as Intel was stuck on 14nm from 2014-2020 when everyone else were successful to go to the next node almost always on time.
As I do not look at these reviews all that often I am unsure if it also applies to Nvidia or AMD as well.
If performance per watt and power consumption are key concerns I would focus on brands that use leading die shrinks like 3nm/4nm/5nm unless you live in a cold climate country like Iceland were waste heat from any PC is always welcome.
And they perform WAY better than Apple Silicon Mac. Not even close to compare each other. M1 Max is already close to mobile RTX 3060. Performance matters more than power by watt when you are working.
The only difference between Pro and Max is GPU cores. I dont see any point of making M2 Pro x2 which only increase the CPU cores. GPU is more important than CPU.
But laptop users prefer battery life.
Would have been nice if Apple overlocked the M1 Max on the desktop so it can suck in more power at the expense of performance per watt.
Not 100% true. Also, it already contradict the purpose of high performance chip. If you need that kind of high performance, then you wouldn't work without a charger anyway even for MBP. Even Apple Silicon Mac can not withstand for several hours if you start editing and working with pro software. At least photo, design, painting doesn't consume a lot of power but still, you do need a power charger while you are working. The battery technology is still no great so I wouldn't expect it too much especially with high end CPU/GPU for laptop.
You are correct so I should have said most laptop users usually prefer battery life. Hence marketing materials putting emphasis on that metric.
A problem of shoehorning an M1 Ultra 5m, chip into a MBP 16" would be the requirement of at least a 240W USB PD charger.
When it is placed into a Mac Studio its power consumption approaches 215W.
The top end 2023 MBP 16" M2 Max 5m, ships with a 140W USB PD charger.
The tech's there already. In the past Apple may not have enough supply of M1 Ultra SoC to satisfy demand for both the a MBP 16" and Mac Studio. This is similar to why the M1 Pro SoC never came in a 2020 Mac mini but made an appearance in the 2023 Mac mini as the M2 Pro.
Last edited:
There are a lot of workloads which requires high performance but is bursty in nature. Software development is one such case. A lot of time is spent in the editor, with CPU idling most of the time, but once in a while you need your developer tools to run as quickly as possible. There are many other examples as well. Image editing can also be done on battery with newest Macs. Video editing probably uses more power, that's true.
The real value of Apple Silicon is that it can deliver desktop-level performance on these mixed workloads while still giving you ultra-book class battery life. And that is a big thing. I can prototype software and design research posters all day long without the charger.
Most laptop users usually prefer battery life, because they have a Windows Laptop!!
Most MacBook Pro M1 MAX / M2 MAX users prefer the performance because:
-they already have 24 hours of battery life
-they can complete tasks much faster
M1 Ultra 5nm chip in MacBook Pro does not need 240W USB PD.
Let's say the battery lasts 24 hours.
Let's say the Ultra chip drains the battery within 8 hours.
This means that the battery is sufficient for an 8 hour workday without recharging.
The current M1 MAX MacBook Pro power supply offers 140 watts.
How many watts does the M1 Ultra consume? 240 watts?
In idle mode or when working or with maximum load of the entire hardware?
Even if the power supply only charges half as fast as it needs to, that means 8 hour workday + half as fast charging cable = +4 hours more battery life = 12 hour workday.
You sleep from 0-8, drive to work from 8-9, work from 9-9pm, drive home from 9-10pm, then eat something, shower, then sleep again.
I highly doubt that most people work 12 hours a day on a MacBook without a break and run it at maximum capacity all the time. But even in the worst case, the half-speed charging cable is perfectly adequate.
Mac Studio M1 Ultra without a display or battery nearly uses 215W of power.
It uses 100W more than the Mac Studio M1 Max SKU.
The 2023 MBP 16" comes with 140W USB PD charger. The latest USB PD spec tops off at 240W.
You'd know this if you clicked the link I provided in my previous post.
I am not saying you cannot use a lower wattage charger with it but it would discharge at load and/or not fast charge.
Last edited:
People who advocate for Ultra in a laptop should not forget that UktraFusion connector itself adds non-trivial power overhead. In future models Apple could also use different internal networks with different power behavior depending on whether it’s a single-chip or multi-chip configuration (this is described in Apple patents).
Power difference between a Max & Ultra SKUs is 100W on the Mac Studio.
MBP 16" M2 Max uses a 140W USB PD charger.
Newest USB PD charger standard is an additional 100W that makes it 240W.
I think Apple does not enough M1 Ultra parts or the 240W USB PD charger they wanted may be too big/heavy for their design targets.
When parts are available to spec then a MBP 16" M2 Ultra SKU with a 240W USB PD charger will be released.
It’s not about power, it’s about bulk and battery. This type of multi-chip technology is impractical for laptops. And there is no way Apple will ever release a laptop that uses more than 100 watts anyway.
That's what they said about a Mac mini M2 Pro... odds are selling a MBP with a M1 Pro chip is easier to do than a Mac mini so they prioritized.
MBP 16" M2 Ultra with a 240W USB PD charger is a no brainer although it may cost a $2k premium over the $3.5k MBP 16" M1 Max.
Who is "they"? And it's not even remotely the same argument. The Mac Mini chassis can easily handle the Pro family. The 16" MBP won't even be able to physically fit an Ultra package.
Our understanding of what's a no-brainer must be very different
Who is "they"? And it's not even remotely the same argument. The Mac Mini chassis can easily handle the Pro family. The 16" MBP won't even be able to physically fit an Ultra package.
Our understanding of what's a no-brainer must be very different
They = Users of any and all Mac forums as to why no M1 Pro chip in a Mac mini or Mac Studio.
I am speaking of supply chain and economics.
M1 Pro
- limited supply of part
- MBP sells more than a Mac mini
- as there is limited supply then put the part into the Mac that sell the easiest
- when supply of part is available Apple decides to put M2 Pro chip into Mac mini last month
M1 Ultra
- released in a Mac Studio in Mar 2021
- USB PD spec for 240W charging came out in May 2022
- limited supply of part
- as there is limited supply then keep to current Mac Studio SKU
As for the M1 Ultra not fitting into the MBP 16" do you have dimensions of the space available vs the M1 Ultra SoC?
As Intel and possibly AMD offering 24-core CPU laptop chips then their incentive to reply in kind. The 240W USB PD charger spec has been available since May. $3499 MBP 16" M1 Max that uses a 140W charger. M1 Ultra requires 100W more that the 240W charger exactly covers. One could argue that Apple planned for this spec.
As the 2021 Mac Studio M1 Ultra is $2000 more expensive. If the price difference carries over between the M2 Max & M2 Ultra then the 2023 MBP 16" M2 Ultra will cost $5499.
Look at the teardowns of MacBook Pro and then the teardowns of the Ultra studio. The M1 Ultra is humongous.
Why would Apple be interested in competing agains desktop chips sold as "laptops"? The 13980HX is 30% faster in multicore than M2 series and uses around 5x times power. It's not competing in the same market segment Apple is in. You'll see these things in a few "desktop replacement" gaming laptops, and that's it. Most premium designs will use the 13700H/13900H which is slower than M2 pro.
I'm lazy to look at it. So will an Ultra in a MBP 16" come out when we switch to a 3nm M3 Ultra by late 2024 or early 2025?
Apple does have the Ultra chips so odds are there is a market.
You think they'd keep M1 Ultra performance just to make things smaller? What would be the logic of that?
Could be a balance of slightly smaller chip for improved performance per watt & power consumption. But not too small that no observable raw performance improvements could be recorded.
Looking at the iFixit teardown of the M1 Ultra Mac studio, the heatsink/fan assembly alone is probably 3-4x the thickness of the 16" MBP, and covers around 2/3 of the surface area of a 16" Pro. That seems to indicate that unless Apple was to underclock an M(x) Ultra, there is no way to fit it into a laptop without making significant sacrifices in terms of weight, size, and portability.
Yep. Even if users wanted an "Ultra" MBP, they aren't going to get it. Just like Apple is keeping iPads crippled with iPadOS because brining the power of macOS to iPad would kill the MacBook line.
The difference is there is no real competition for the iPad. They can’t even catch up to the regular iPad, let alone the Air or the Pros…Apple actually needs to punch out something powerful to pull away from the other guys. They’ve got performance on battery locked down, but maybe they’d like to be in front by the same margins as they do with the iPads.
UltraFusion is an interconnection technology with a high-speed interface that enables hardware components to be interconnected with one another. The interconnected components fundamentally work as a single hardware component.
Apple specifically designed the M1 Max chip with a silicon interposer as part of its overall chip packaging. Note that packaging represents a stage in chip fabrication that involves mounting and interconnecting integrated circuits and other components, and encapsulating a semiconductor component or an entire chip into a protective enclosure.
On the other hand, an interposer is a bridge or conduit that enables electric signals to pass through and onto another destination. The interposer used by Apple allows two M1 Max chips to connect over 10000 signals and maintain an interprocessor bandwidth of 2.5TB/s.
UltraFusion is basically an implementation of silicon interposer technology and chip packaging techniques to create an interprocessor interconnection technology. Apple implemented these using one of the advanced packaging technologies developed and deployed by TSMC called chip-on-wafer-on-substrate with silicon interposer or CoWoS-S.
Source: https://www.profolus.com/topics/apple-ultrafusion-technology-explained-what-and-how/
No M1/M2 Pro chip has UltraFusion so your case is impossible to implement.
Apple specifically designed the M1 Max chip with a silicon interposer as part of its overall chip packaging. Note that packaging represents a stage in chip fabrication that involves mounting and interconnecting integrated circuits and other components, and encapsulating a semiconductor component or an entire chip into a protective enclosure.
On the other hand, an interposer is a bridge or conduit that enables electric signals to pass through and onto another destination. The interposer used by Apple allows two M1 Max chips to connect over 10000 signals and maintain an interprocessor bandwidth of 2.5TB/s.
UltraFusion is basically an implementation of silicon interposer technology and chip packaging techniques to create an interprocessor interconnection technology. Apple implemented these using one of the advanced packaging technologies developed and deployed by TSMC called chip-on-wafer-on-substrate with silicon interposer or CoWoS-S.
Source: https://www.profolus.com/topics/apple-ultrafusion-technology-explained-what-and-how/
No M1/M2 Pro chip has UltraFusion so your case is impossible to implement.
Yup, at least M1 Max had a connector just for that but M1,2 Pro has no such thing.
Next Gen Mac Studio with M2 pro Ultra (or ultra-s) if happens It will maxout at the same as current Generation: 128GB Ram, watcht the M2 Max size and Ram placement, it may need bit more board area in a "ultra" 2P arrangement
note the wider-taller M2 Max is relative M1 Max (the photo dont include its UltraFusion edge interconnect, likely dual (top and bottom) for a 4P M2 Extreme, so the M2 Max Chip likely wont fit in M2 Ultra 2P inside the Studio chasis. likely an M2 pro - Ultra (2P) should be indicated to fit the same constrained area in the Mac Studio M2 logic board,
note the wider-taller M2 Max is relative M1 Max (the photo dont include its UltraFusion edge interconnect, likely dual (top and bottom) for a 4P M2 Extreme, so the M2 Max Chip likely wont fit in M2 Ultra 2P inside the Studio chasis. likely an M2 pro - Ultra (2P) should be indicated to fit the same constrained area in the Mac Studio M2 logic board,