I am not sure that M4 Max for MBPs will ever exist. I think Apple will release new MBPs with M5 family in fall.
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.
LPDDR6: New Beginning of AS: 12GB Standard
- Thread starter TigeRick
- Start date
- Sort by reaction score
I am not sure that M4 Max for MBPs will ever exist. I think Apple will release new MBPs with M5 family in fall.
Do you think that new (M5) designation will be based purely on architectural changes vs. M4, or do you think they'll be switching to a new process then as well (N3P)?
Why is that? Apple would need to make the Max in order to make the Ultra. Unless you think they're going monolithic for Ultra.
I think he meant we'll be seeing the M4 Max and Ultra in the Studios in June, and then the M5 Max in the MBP's in late fall. Thus no M4 Max in the MBP's.
I'd really like to know why @leman thinks M5 will ship in the fall. Just the possible change to N3P?
N3P might not ship in volume until 2025. But assuming it does ship in 2024, and Apple uses it, I suspect Apple will continue to call those chips M4, because I think they will be simple die shrinks of the current M4. They might call it "M4+", but I doubt they'd even bother.
I doubt they'll go from M2->M5 in roughly a year. After all the M3 was just introduced last October. Too many design teams stepping on each others' heels.
So @leman thinks M3 in Oct 2023, M4 in May 2024, M4 Ultra/Extreme in June 2024, M5 Pro/Max in Fall 2024?
So 3 whole generations in 1 year and an Ultra/Extreme at WWDC. Wow.
I originally thought he meant M5 family in Oct 2025 and MBPs won't get M4 Pro/Max later this year.
I’d say it depends on whether N3P will be ready. I’d expect architectural improvement to be the central part, if they can pull the new node, even better.
I just don’t think Apple will refresh the MBP this early and I don’t think they will go that long without another Apple Silicon update. As for the desktop Macs, they could go entirely different route. They do have all these patents that describe 2.5D and 3D arrangements of multiple specialized dies on a single package. Maybe we will see this tech debut in the Mac Pro. M4 already showed that they are not afraid to spice things up, so who knows what they have in mind. I think the basic idea I have in mind that rapid move from M3 to M4 is them trying to catch up to their original schedule, and if M5 is planned for 2025 then they are not winning anything.
Or it could be that I am massively overthinking things as usual and that we will see a full M4-based family across all the products.
If we go by the 1 year cadence no matter what, then M4 should have released in Oct 2023. Then M5 in oct 2024. I think that's what you're trying to get at.
But I don't think that's going to come to fruition. I just don't see Apple releasing 3 generations in 1 year. I think we will see M4 Pro/Max later this year for MBPs - hopefully with Tandem OLEDs.
I’d expect some GPU improvements and maybe other small tweaks. Or maybe something completely different like a new packaging. N3P is secondary here I believe. They can still get a lot of mileage from current processes.
Not necessarily. For example, the team that designed the incredible tech behind the M3 GPU is unlikely to be resting on their laurels. It’s entirely possible that they’ve been working hard on new improvements for the last year. Those could be realistically ready to ship by fall. Also, we just saw the CPU division deliver not one but two micro-architectures 6 months from watch other. Who’s to say they can’t do it again?
My point is that Apple has demonstrated they are not afraid of aggressively updating their technology platform, and that they are shipping features, not timelines. If there are new features done by fall (and that is likely), I think we’ll see them.
What I’m trying to get at is that cadence is a red herring. Forget about new chip every 12/18/24 months. Think about products. What is missing? What can be done better? What do users want? What is ready to ship? If we assume that Apple is willing to spend the $$$ on new masks regularly (and it does seem so), I don’t think it’s too controversial to expect an update to some of the IP blocks.
That is a very valid objection as well. As I’ve said, I might be overthinking.
That I kind of doubt, often making larger displays is more complicated. But who knows, maybe this technology is cheaper and simpler than miniLED.
I'm curious to see what they do with the M4 Pro. Especially in the memory size and bandwidth department.
But at this point I think they are starting to save things for the redesign to convince M1/M2 Pro users to move to the new design (in theory)
Personally for me unless the M4 Pro is something really crazy in terms of efficiency when compared to the M2 Pro I won't be upgrading.
Really playing the long game here so I can get the OLED model which would be part of the MBP redesign..?
But at this point I think they are starting to save things for the redesign to convince M1/M2 Pro users to move to the new design (in theory)
Personally for me unless the M4 Pro is something really crazy in terms of efficiency when compared to the M2 Pro I won't be upgrading.
Really playing the long game here so I can get the OLED model which would be part of the MBP redesign..?
They're already doing 13". What's another 3"?
That’s a good point
It's over 50% more area ((16/13)^2). That's a lot, possibly enough to be prohibitive this year or even next, as the panels are both expensive and power-hungry.
Also, my impression is that cost does not scale linearly with area, though I know little about this.
It'd be slightly awkward for Apple to tout a new Display Engine made exclusively to power Tandem OLED in the M4 all over the press release ("display engine" repeated 5 times) and in the video and then not bring Tandem OLED to Macs when they get the M4.
Not a huge deal. If Tandem OLED isn't ready for bigger displays then it is what it is. But just slightly "wasteful" to have a display engine for Tandem OLED but no actual Tandem OLED for M4 Macs.
Can’t be helped I feel. Most of the Macs were never getting that display anyway (considering some of them are headless). IIRC Ross Young said 2027 for OLED Macs so I think we’ll be waiting a while.
(Warning: I'm getting into the weeds, and outside my expertise.) I don't think support for the Tandem OLED requires a lot more transistors. It needs an update, but not a big one.
The "wastefulness" is always going to be there, as each display controller will support a number of features not all used by any single display. TOLED just adds a little bit more.
I was really expecting OLED displays on the MacBooks Pro come the redesign tbh…
Hmm, actually your speculation is aligned with my theory. Your proposed M5 by the end of the year is similar to my speculated M4+. M4+ is my reference of enhancement to M4 for iPad.
FYI, upcoming Blackhawk's IPC/PPC improvement is true. It is biggest bump of ARM Cortex X series. Maybe it is due to inclusion of SME, but the threat to Apple Silicon is real. We are going to see at least four more OEMs entering WoA market next year. I believe most of them will use Cortex-X5. Samsung Exynos 2500 will feature 10 cores total with 64-bit memory bus. Apple definitely aware of upcoming threat, that's why I believe Apple will put in more cores and 12GB into M4+ to compete with bunch of ARM army.
What makes you think that ARM is capable of such large gains? They've failed annually to reach their targets, though they are improving YoY, and the X5 targets are extremely ambitious. In particular there have been rumors swirling around since Febraury that the X5 is having major problems with both performance and efficiency. But even if they hit their targets, they would be miles away from catching up to the M4.
I think the chances of the X5 catching up this year even with M3 are basically zero. They *might* beat the M2.
I’ll believe it when I see it. Cortex X4 failed to impress performance-wise and it’s nominally wider than Apple Silicon. So far we know nothing about X5 except a cryptic post on a Chinese forum. I wouldn’t be surprised if 9400 included an overclocked X5 core to achieve high benchmark scores, but in the typical Android SoC fashion that core would likely be disabled for user apps due to battery concerns.
OLED panels are cut from "mother glass" panels that are much larger and then cut down into different sizes.
As the "mother glass" increases in size so does the average size of panels, like how we quickly went from 42-inch to 55-inch and now 65-inch.
Where exactly do you get that OLEDs are more power hungry? It's actually the opposite.
IMO, those making claims about the relative power consumption of OLED vs. LCD should cite sources to back up their assertions.
OLED is self-emissive. Thus OLED is more efficient for images in which most of the pixels are dark, since only those pixels need to be activated. By contrast, LCD uses a relatively small number of large sources of light, which can be made more inherently efficient than numerous individual OLEDs for bright images. Thus when you'd need most of the OLED pixels to be on, and emitting a lot of light, LCD is more efficient.
Sources:
According to this Feb 2014 article by Peter Su, who is display specialist with Omdia, "Gaming dark mode (on-pixel ratio of 20%): Tandem OLED has the best power consumption. Multimedia mode (on-pixel ratio of 40%): Both tandem OLED and low-power LCD have better power consumption. Office work mode (on-pixel ratio of 80%): Low-power LCD has the best power consumption." [ https://omdia.tech.informa.com/om120857/display-dynamics--february-2024-comparison-of-power-consumption-between-low-power-lcd-and-oled#:~:text=Gaming dark mode (on-pixel,has the best power consumption. ]
Consistent with this, Lenovo's website says:
"Are OLED laptops more power-hungry compared to liquid crystal display (LCD) laptops?
OLED laptops do consume more power than LCD laptops, especially when displaying bright or white content. This is because each pixel in an OLED display emits its own light, while LCD displays rely on a backlight that illuminates all pixels simultaneously. However, advancements in OLED technology have significantly improved power efficiency. Many OLED laptops now offer power-saving features and optimizations that help mitigate excessive power consumption, resulting in longer battery life."
[ https://www.lenovo.com/us/en/glossary/what-is-oled-laptop/ ]
OLED is self-emissive. Thus OLED is more efficient for images in which most of the pixels are dark, since only those pixels need to be activated. By contrast, LCD uses a relatively small number of large sources of light, which can be made more inherently efficient than numerous individual OLEDs for bright images. Thus when you'd need most of the OLED pixels to be on, and emitting a lot of light, LCD is more efficient.
Sources:
According to this Feb 2014 article by Peter Su, who is display specialist with Omdia, "Gaming dark mode (on-pixel ratio of 20%): Tandem OLED has the best power consumption. Multimedia mode (on-pixel ratio of 40%): Both tandem OLED and low-power LCD have better power consumption. Office work mode (on-pixel ratio of 80%): Low-power LCD has the best power consumption." [ https://omdia.tech.informa.com/om120857/display-dynamics--february-2024-comparison-of-power-consumption-between-low-power-lcd-and-oled#:~:text=Gaming dark mode (on-pixel,has the best power consumption. ]
Consistent with this, Lenovo's website says:
"Are OLED laptops more power-hungry compared to liquid crystal display (LCD) laptops?
OLED laptops do consume more power than LCD laptops, especially when displaying bright or white content. This is because each pixel in an OLED display emits its own light, while LCD displays rely on a backlight that illuminates all pixels simultaneously. However, advancements in OLED technology have significantly improved power efficiency. Many OLED laptops now offer power-saving features and optimizations that help mitigate excessive power consumption, resulting in longer battery life."
[ https://www.lenovo.com/us/en/glossary/what-is-oled-laptop/ ]
I was under the impression that with OLED the content on the screen has a bigger influence on power consumption as compared to LCD. For example, a black background/desktop has a bigger impact on OLED power consumption as compared to LCD.
Last edited: