Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Confused-User

macrumors 6502a
Oct 14, 2014
584
625
Where exactly do you get that OLEDs are more power hungry? It's actually the opposite.
It's sometimes the opposite, see below.

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.
There are some things that I know about this, and some that I only believe with moderate confidence levels. I'll try to be clear about which.

Fact: OLEDs have a different profile than LCDs. A fully black OLED consumes tiny amounts of power as it's doing pretty much nothing. The more pixels light up, the more power it uses. With "normal" LCDs, it's the other way around - the backlight is always on, and the more pixels aren't white, the more power that uses. Of course with that advent of dimmable LED arrays, the picture get murkier, as all-black LCDs now need less power because the backlight itself is using little/no power- but when showing bright content, FALD LCDs may use more power than traditional backlit LCDs.

Believed but not known: Tandem OLEDs have various benefits, including much better longevity and less susceptibility to burn-in, but they do use more power than regular OLEDs. By how much I don't know, though my impression is that it's moderate. They may also have higher baseline consumption for an all-black state - I don't have any data on this.

TOLED and LCDs will still have very different power consumption that's content-dependent. But TOLEDs - at least the ones on the iPP - do appear to consume more than regular OLEDs.

As for the question of scaling- @Pressure, I understand about mother glass and how that's developed over time. I am still under the impression that large displays scale cost up more than linearly with area, but I do not know that for certain, nor do I have any idea what drives that cost, unless it may be the defect rate in mother glass. I can well imagine the same dynamic that plays out in silicon being applicable here (driving the cost of large chips more than linearly by area compared to small chips, due to single defects killing more total silicon when producing large chips, and leaving out the workaround of binning, which clearly doesn't help for displays).

I'd welcome further data and/or corrections.
 

crazy dave

macrumors 65816
Sep 9, 2010
1,302
1,002
It's sometimes the opposite, see below.


There are some things that I know about this, and some that I only believe with moderate confidence levels. I'll try to be clear about which.

Fact: OLEDs have a different profile than LCDs. A fully black OLED consumes tiny amounts of power as it's doing pretty much nothing. The more pixels light up, the more power it uses. With "normal" LCDs, it's the other way around - the backlight is always on, and the more pixels aren't white, the more power that uses. Of course with that advent of dimmable LED arrays, the picture get murkier, as all-black LCDs now need less power because the backlight itself is using little/no power- but when showing bright content, FALD LCDs may use more power than traditional backlit LCDs.

Believed but not known: Tandem OLEDs have various benefits, including much better longevity and less susceptibility to burn-in, but they do use more power than regular OLEDs. By how much I don't know, though my impression is that it's moderate. They may also have higher baseline consumption for an all-black state - I don't have any data on this.

TOLED and LCDs will still have very different power consumption that's content-dependent. But TOLEDs - at least the ones on the iPP - do appear to consume more than regular OLEDs.

As for the question of scaling- @Pressure, I understand about mother glass and how that's developed over time. I am still under the impression that large displays scale cost up more than linearly with area, but I do not know that for certain, nor do I have any idea what drives that cost, unless it may be the defect rate in mother glass. I can well imagine the same dynamic that plays out in silicon being applicable here (driving the cost of large chips more than linearly by area compared to small chips, due to single defects killing more total silicon when producing large chips, and leaving out the workaround of binning, which clearly doesn't help for displays).

I'd welcome further data and/or corrections.
While it's hard to know for sure, given that the screen is 0.7% bigger, the battery is 5% smaller, while the power draw from the SOC is probably similar to the M3 (better node, but higher clocks), and the quoted battery life is the same (needs to be tested), it's suggestive that the Tandem OLED display is probably at least as efficient for the average content as the previous mini-LED display, probably slightly more efficient. Obviously all a bit loosey goosey.
 

TigeRick

macrumors regular
Original poster
Oct 20, 2012
135
143
Malaysia
please read my speculation here. As I said, Apple is going to supercharge upcoming Mac with M4+ family by the end of year with 12GB LPDDR6. Current M4 SoC is only for iPad with 8GB LPDDR5X.

And yes, please use your brain to think why Apple will continue using N3B for M4 Max with more cores, NPU, Ultra-fusion connector and so on. Definitely upcoming M4 Max will break 100 billion transistors even though the amount of SLC and memory bus will be cut.

@Confused-User is confused and delusional about Mac lineup. I have listed some evidence in the page before, yet he still insists I am delusional. Here is my speculation of iPhone and Mac lineup by the end of year:

  1. JEDEC going to announce LPDDR6 standard before iPhone event.
  2. iPhone 16 Pro will feature 12GB LPDDR6.
  3. By Q4 of the year, Apple will announce M4+ family (or whatever name Apple marketing feels suitable) with starting RAM of 12GB LPDDR6.
@Confused-User If you still think I am delusional, why not you write down your prediction here so that we can compare results.

I may not 100% correct, but I believe all roads lead to LPDDR6 and Apple will be the first to employ it. Thus, the secrecy behind it.
 
Last edited:

crazy dave

macrumors 65816
Sep 9, 2010
1,302
1,002
That's for Let Loose - the iPad event. I don't know if any Macs are coming to WWDC or not, but that's just saying there will be no Macs at the iPad event

  • New Macs: While the new ‌iPad Pro‌ models may contain the M4 chip, a processor that will later come to the Mac, Apple is not planning to unveil any new Mac models at the "Let loose" event.

Unless there is something else that says no Macs at WWDC?
 

TigeRick

macrumors regular
Original poster
Oct 20, 2012
135
143
Malaysia
That's for Let Loose - the iPad event. I don't know if any Macs are coming to WWDC or not, but that's just saying there will be no Macs at the iPad event



Unless there is something else that says no Macs at WWDC?
Sorry my bad but my prediction still hold. I am challenging @Confused-User with his endless questioning about LPDDR6, thus I listed down all upcoming prediction. It's his turn.
 

Confused-User

macrumors 6502a
Oct 14, 2014
584
625
And yes, please use your brain to think why Apple will continue using N3B for M4 Max with more cores, NPU, Ultra-fusion connector and so on. Definitely upcoming M4 Max will break 100 billion transistors even though the amount of SLC and memory bus will be cut.

@Confused-User is confused and delusional about Mac lineup. I have listed some evidence in the page before, yet he still insists I am delusional. Here is my speculation of iPhone and Mac lineup by the end of year:

  1. JEDEC going to announce LPDDR6 standard before iPhone event.
  2. iPhone 16 Pro will feature 12GB LPDDR6.
  3. By Q4 of the year, Apple will announce M4+ family (or whatever name Apple marketing feels suitable) with starting RAM of 12GB LPDDR6.
@Confused-User If you still think I am delusional, why not you write down your prediction here so that we can compare results.

I may not 100% correct, but I believe all roads lead to LPDDR6 and Apple will be the first to employ it. Thus, the secrecy behind it.
I don't pretend to have revelations from on high about what Apple will do. I do think some of your conclusions are low-probability. To repeat, here's what I wrote earlier, and I still think is correct:

"There is *zero* chance LPDDR6 will be available in sufficient quantities for the iPhone 16, despite Tiger's obsession with this technology here and elsewhere. It's ridiculous to even suggest it. [... M4 stuff, not relevant...]
If LPDDR6 ships at all this year, you're likely to see it in a few phones that sell in low volume, and people like Tiger will go wild about it, but then not follow up when analyses show that all that extra bandwidth nets around 2-3% better performance on typical workloads."

I also wrote:

"There is no way they're doing M4, or anything else, for N3B. That process was one-and-done, for all the reasons we've discussed previously. I will bet every one of my body parts, against a single dollar, that they will not lay out the entire processor twice, once for N3E and once for N3B. That would be the single dumbest decision made since... probably since someone won a darwin award.

"Density" is the last thing Apple cares about. They care about power and performance, usually in that order, as you can see for example with their display controllers. To the extent that they want better density (really, lower area) - for example, plausibly, with those same controllers - they'll use N3E FinFlex 2-1 or 2-2 for those transistors, where compatible with their performance targets.

Everything will stay on N3E until N3P hits."

In line with that, and things I've written elsewhere, here's a couple more specifics:

- Very likely they'll introduce high-end desktops at WWDC
- Very likely the MBPs will be introduced at WWDC or later in the Summer. But if not, in the fall, whichever month isn't iPhone month.
- Airs are a big question. Could be WWDC, but VERY unlikely. Could be later in the summer, not so likely. Could be fall, seems most likely to me.
- iMacs and Mini... no idea. Could plausibly be any time this year.

The chance of iPhones shipping with LPDDR6 remains ~0. 12GB RAM is unlikely, but I suppose possible for the Pros. I'd bet against it though.

If @leman's prediction of M5 in the fall is right (I doubt it, but not impossible) then I could imagine LPDDR6 in Studios or Mac Pros. Again, I wouldn't bet on it.

If you think I'm hedging on a lot of this... you're right. I don't pretend to be an oracle.
 

dgdosen

macrumors 68030
Dec 13, 2003
2,772
1,409
Seattle
Take a look at the environment:

- The M4 looks like a reasonably big upgrade (saying nothing about the pro/max/??? versions
- The M4 delivers 38 TOPS compared to 18 at most on any existing mac desktop.
- WWDC is going to be AI focused (because Apple is and doesn't want to appear to be falling behind on the subject)

That means of course they're going to introduce some hardware that will showcase the M4 and all their AI efforts on macos. It'd be a surprise if they didn't.
 

Boil

macrumors 68040
Oct 23, 2018
3,294
2,916
Stargate Command
Disclaimer - I push for a Mn Extreme SoC of some sort because proper cooling might require a larger chassis than the Mac Studio; something like, I dunno, a Cube...?

WWDC 2024
Mac Pro Cube
M4 Extreme SoC
LPDDR5X RAM
960GB w/inline ECC

Top-mounted holographic projector that enables an interactive AI-driven Steve Jobs digital assistant...!

mac pro shorty.jpg
 

mr_roboto

macrumors 6502a
Sep 30, 2020
777
1,668
I may not 100% correct, but I believe all roads lead to LPDDR6 and Apple will be the first to employ it. Thus, the secrecy behind it.
Everyone knows that memory standards move forwards. It's public knowledge that JEDEC is working on LPDDR6. Everyone knows that Apple will adopt it, that's the easiest prediction in the world. Throwing out wildly overconfident predictions about when, though? Based on misunderstandings of all the random scattered scraps of info you've found and pasted into tables? That isn't smart.

All those words you've typed and not one of them discusses the critical events that have to happen before Apple launches anything which uses LPDDR6:

1. They will need to design (or buy) a LPDDR6 memory controller IP for their SoC
2. They will need to have working LPDDR6 engineering samples to qualify against said memory controller
3. They will need to have ironclad supply contracts to secure volume production buys with deliveries beginning multiple months before product launch

I doubt all these hurdles can be cleared for LPDDR6 before the end of 2024. The standard isn't even ratified yet, so how can they call either #1 or #2 complete? And on #3... if they want to launch a product this fall, Apple needs to be working on supply chain and manufacturing RIGHT NOW.

As a real world example of how timelines tend to work: Samsung began mass production of LPDDR5 in July 2019, according to a press release on their website. Was Apple able to ship LPDDR5 in the A13 that fall? Nope, it was still LPDDR4X.

Edit: and it stayed LPDDR4X for M1. Apple didn't move to LPDDR5 until M1 Pro/Max in October 2021, about 2.25 years later.
 
Last edited:

senttoschool

macrumors 68030
Nov 2, 2017
2,577
5,339
I could see Apple adopting LPDDR6 first for their Ultra/Extreme SoCs to further differentiate them in the future. Those are the machines are need more bandwidth for AI inferencing. They have huge memory capacity due to unified memory but lack the raw bandwidth of dedicated high end GPUs.

They're also rather niche products so Apple doesn't have to secure mega supply first.
 

senttoschool

macrumors 68030
Nov 2, 2017
2,577
5,339
Makes sense to not have any new Macs at WWDC. The Mac Studio/Pro would require the M3 Ultra but Apple seems quite eager to move away from M3/N3B generation as soon as possible. Besides, it'd look silly to have the M4 in an iPad destroy the M3 Ultra in ST performance in Mac Pro. They can't announce M4 Ultra without the M4 Max. But they can't announce the M4 Max because it's far too soon since the M3 Max MBPs came out.
 

MrGunny94

macrumors 65816
Dec 3, 2016
1,117
652
Malaga, Spain
I suppose we'll see the new MacBook Pros later down in the year just like we saw with the M3 Pro/Max.

Gurman said Air and Mac mini will only be refreshed next year
 

Mac_fan75

macrumors member
Jun 1, 2023
44
53
Makes sense to not have any new Macs at WWDC. The Mac Studio/Pro would require the M3 Ultra but Apple seems quite eager to move away from M3/N3B generation as soon as possible. Besides, it'd look silly to have the M4 in an iPad destroy the M3 Ultra in ST performance in Mac Pro. They can't announce M4 Ultra without the M4 Max. But they can't announce the M4 Max because it's far too soon since the M3 Max MBPs came out.
They perfectly could for the Studio's as they still on M2 old stuff (is anyone buying it now?), but yeah Laptops first unfortunately.
It would be a complete joke if they give the Studio the M4 Max next year what Gurman is saying. I am still hoping next month but they might want it to combine it with the next MacOS for new AI stuff and show off?
 

senttoschool

macrumors 68030
Nov 2, 2017
2,577
5,339
They perfectly could for the Studio's as they still on M2 old stuff (is anyone buying it now?), but yeah Laptops first unfortunately.
I just don't see it. In order to update the Studio to Ultra, you need the M4 Max first. You can't make an Ultra without 2 Max chips. Unless you think they will unveil an M3 Ultra. That'd be weird. I suppose if Apple were to update the Mac Studio/Pro, they're more likely to use an M3 Ultra than an M4 Ultra. Neither are likely.
 

crazy dave

macrumors 65816
Sep 9, 2010
1,302
1,002
I just don't see it. In order to update the Studio to Ultra, you need the M4 Max first. You can't make an Ultra without 2 Max chips. Unless you think they will unveil an M3 Ultra. That'd be weird. I suppose if Apple were to update the Mac Studio/Pro, they're more likely to use an M3 Ultra than an M4 Ultra. Neither are likely.
That is unless they’re switching things up and M4 Ultra is its own monolithic die and an M4 Extreme is two Ultras. ;) I’m not saying that’s what will happen but certainly Gurman is back to predicting a special Mac Pro chip and that is one way to achieve that. Not the only one of course. Plus that prediction is still that we won’t see those chips until next year. Still though as @mr_roboto pointed out Apple would have to move ahistorically fast to even do a mid-generation memory upgrade next year. Then again they moved ahistorically fast releasing the M4 this quickly so who knows? Indeed not just for memory but for chips, we could see anything from the M5 coming out this fall to M4 products still releasing next summer. To me, the latter is more likely than the former, but I ain’t discounting either. Obviously when we see what does and doesn’t get announced at WWDC we’ll have a lot more information about the rate of M4 product releases.
 
  • Like
Reactions: senttoschool

MrGunny94

macrumors 65816
Dec 3, 2016
1,117
652
Malaga, Spain
That is unless they’re switching things up and M4 Ultra is its own monolithic die and an M4 Extreme is two Ultras. ;) I’m not saying that’s what will happen but certainly Gurman is back to predicting a special Mac Pro chip and that is one way to achieve that. Not the only one of course. Plus that prediction is still that we won’t see those chips until next year. Still though as @mr_roboto pointed out Apple would have to move ahistorically fast to even do a mid-generation memory upgrade next year. Then again they moved ahistorically fast releasing the M4 this quickly so who knows? Indeed not just for memory but for chips, we could see anything from the M5 coming out this fall to M4 products still releasing next summer. To me, the latter is more likely than the former, but I ain’t discounting either. Obviously when we see what does and doesn’t get announced at WWDC we’ll have a lot more information about the rate of M4 product releases.
Wouldn't they wait for 2025 for the Ultra update? Gurman mentioned the M4U is a 2025 product as well on his latest weekly newsletter.

I mean they never introduced an M chip with an iPad before.. so yeah.. Anything can happen really
 

PaulD-UK

macrumors 6502a
Oct 23, 2009
576
268
😂 So Apple Sales took Gurman to one side with their straightest
we-don't-wan't-to-Osborne-Effect-our-M2-Studio/Mini-sales-do-we look,
and told him "...the M4U is a 2025 product, please leak that in your latest weekly newsletter..."
To quote Mandy Rice-Davies: "Well they would say that wouldn't they..."

We shall see 😁
 

Boil

macrumors 68040
Oct 23, 2018
3,294
2,916
Stargate Command
WWDC 2024 is all about AI...
M4-series of SoCs are supposed to be all about AI...
Gotta have something more than a M4 iPad Pro for the developers to get all about AI...
M4 Max & (monolithic) M4 Ultra Mac Studios will be all about AI...
M4-series Mac Studios @ WWDC 2024...
Because it's all about AI...
AI...!
 

Confused-User

macrumors 6502a
Oct 14, 2014
584
625
I just don't see it. In order to update the Studio to Ultra, you need the M4 Max first. You can't make an Ultra without 2 Max chips. Unless you think they will unveil an M3 Ultra. That'd be weird. I suppose if Apple were to update the Mac Studio/Pro, they're more likely to use an M3 Ultra than an M4 Ultra. Neither are likely.
This ignores the easiest and most obvious option.

Release M4 Max (and maybe Ultra) in Studio *first*. Everything else (MBPs, Minis, iMacs, Airs) can come later, or some of it (MBPs, would be my guess, but maybe Minis too/instead) at the same time. The Mac Pro would be updated at the same time as the Studio, though I doubt it will get the rumored "Extreme". Not to rule that out though - after the M4 iPad, the universe of possibilities seems notably bigger than before.
 

senttoschool

macrumors 68030
Nov 2, 2017
2,577
5,339
This ignores the easiest and most obvious option.

Release M4 Max (and maybe Ultra) in Studio *first*. Everything else (MBPs, Minis, iMacs, Airs) can come later, or some of it (MBPs, would be my guess, but maybe Minis too/instead) at the same time. The Mac Pro would be updated at the same time as the Studio, though I doubt it will get the rumored "Extreme". Not to rule that out though - after the M4 iPad, the universe of possibilities seems notably bigger than before.
You're right. I did miss that. However, if they're unveiling the Max, they'd likely want to unveil the Pro as well. It'd make logical sense because unveiling the Max months before the Pro would make any Pro announcement look silly later on. "Today, we're excited to announce the M4 Pro that is much slower than the M4 Max we announced 5 months ago". If they unveil the Pro, the only logical device to put it in is the Mac Mini.

If they unveil the M4 Pro for Mac Mini and M4 Max/Ultra for the Studio/Mac Pro, it'll tank M3 Pro/Max MBP sales until those laptops get an update.

Perhaps Apple wants to go that route. I'd love to see it but I don't think it will happen.
 
Last edited:

senttoschool

macrumors 68030
Nov 2, 2017
2,577
5,339
That is unless they’re switching things up and M4 Ultra is its own monolithic die and an M4 Extreme is two Ultras. ;) I’m not saying that’s what will happen but certainly Gurman is back to predicting a special Mac Pro chip and that is one way to achieve that. Not the only one of course. Plus that prediction is still that we won’t see those chips until next year. Still though as @mr_roboto pointed out Apple would have to move ahistorically fast to even do a mid-generation memory upgrade next year. Then again they moved ahistorically fast releasing the M4 this quickly so who knows? Indeed not just for memory but for chips, we could see anything from the M5 coming out this fall to M4 products still releasing next summer. To me, the latter is more likely than the former, but I ain’t discounting either. Obviously when we see what does and doesn’t get announced at WWDC we’ll have a lot more information about the rate of M4 product releases.
Regarding the bolded part, I think it's unlikely. I could be wrong but the M3 Max die size is 600 - 700mm². TSMC's reticle limit is 800mm². In other words, you can't make an Ultra that is twice the size of M3 Max on a single die.
 

SBeardsl

macrumors member
Aug 9, 2007
54
12
Regarding the bolded part, I think it's unlikely. I could be wrong but the M3 Max die size is 600 - 700mm². TSMC's reticle limit is 800mm². In other words, you can't make an Ultra that is twice the size of M3 Max on a single die.

It was surprisingly hard to get a definitive size for the M2 Max die size. I guess Apple stopped giving official size info after the M1 family The best I could do was this article Why Intel and AMD don't make chips like the M2 Max and M2 Ultra:

"At an estimated die size of 550mm2 (assuming Apple's side-by-side comparison above is to scale as no actual measurements seem to exist), the M2 Max is super big, and the M2 Ultra is the largest consumer chip ever made at over 1,000mm2"

I believe the answer to how could Apple make such a huge chip is the "Ultra Fusion Connector" on the M1 and M2 chips. It connects two separate chips together (the manufacturing/installation must be mind-bogglingly precise) to create what is in effect a single larger chip.

Part of the "evidence" for the idea that the M3 Ultra would be a single die was the absence of a Fusion Connector on the M3 Max die. Perhaps Apple, already knowing that they would not be producing an M3 Max, just eliminated the connector on the M3 Max leaving the possibility that it could return on the M4 Max allowing for the creation of an M4 Ultra out of two M4 Max chips.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.