Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
I have an iPhone 8+ in good condition, but I’ll likely upgrade to a 14 when the 15 comes out in September or during one of the upcoming back-to-school promotions.
 
Didn't like this Apple Silicon iMac design either. I didn't mind the colors, but I also didn't see the purpose of the "chin" and preserving that aesthetic in the name of making it super thin.

The purpose of the 'chin' is to house the actual functionality that provides the features of a computer. The rest of the that design is basically something apparently handed down by the thinness politburo. You get a chin because the major theme here is to make an "iPad on a stick". Too thin to put a headphone jack on the front or back (we Apple normally likes to put it). Too thin for an Ethernet jack ( so shuffled off to the power brick). Too thin for a standard power connector. There is no room behind the screen for either decent speakers or a substantive power consuming electronics.

If look at a tear down


h3ZxqpRyhwlebxWC.large


The speakers and fans on either end are in the chin. The center shielded portion there is where the M1 , RAM , power management, SSD storage , the actual ethernet controller ( just a PHYS plug in the power brick). etc are all in the chin. Toss the chin and would toss all that also. Likely also why is constrained to just an 'plain' Mn SoC ( limited space) .

The electronics behind the screen are the USB controllers and some USB-C power management chips.


To toss the 'chin' and still have a computer, they would need to make the iMac thicker. Doubtful that will pass muster as being 'less bad' than having the historic chin.
 
  • Like
Reactions: Pinkyyy 💜🍎
From my perspective the elephant in the room regarding the A17/M3 SOC is the yield rates for the N3B lithography process.

According to reports it's still at around the 55% mark for the A17 Bionic and theirs talk for the M3 to be later moved to the N3E lithography process.

Reports when? Or just echo chamber of the same stuff. I have seen reports that N3B is moving up on yield , it is just slower than what folks would like. Additionally, the rumor that Apple is making TSMC pay for all the bad dies until later in the year, only probably makes that progress even slower. ( TSMC has very little incentive to loose money faster than they have to).

N3B has two issues besides yield as to why most folks don't want to use it. First, it is a very slow 'bake'. ( rumors around 4 months from 'blank' wafer to something useful). If the demand is variable that is going to be more difficult to control . If demand sags in 3 months and have ordered alot then stuck with inventory. Likewise if demand spikes ... it is over a whole quarter until can respond. Apple has some steady products and they also tend to throw the same SoC in to multiple products.

N3B is also more expensive. ( not completely uncoupled from the first issue. Spend longer time inside of extreme expensive fab machines ... then get charged more). There is also more long term support from design tools .


The problem with N3B is that almost nobody else wants it so TSMC is likely to drop it in 2-3 years once Apple rotates a large share of products onto a new process. The longer processing time means fab machines TSMC can't use on something else are being sucked up for a slower wafer flow. So they would be servicing less customers with the same fixed set of equipment. That isn't the business they are in. More customers is better than fewer. Yields are not really a major factor.

The major reason why Apple might switch M3 to N3E is that Apple would want to sell M3 inside of 'trickle down' products in 2-3 years from now ( e.g., iPad Air ). Or trickle down the M3 into the vintage MBA model the M1 is in now. If Apple just used M3 on N3B for 1.5 years and didn't try to prolong sales out into '25-26 they could probably use N3B. The M-series is relatively low enough volume that not going to want to do two versions ( N3B and N3E).

The M3 Pro/Max/Ultra/etc. are pretty likely one and done SoCs that get placed in just a smaller subset of products for just one gen before dropped. Doing those on N3B and then dropping for M4 ( on N3E or N3P ) wouldn't be a problem. iPad Air could just either linger on M2 for 2 year (jump to an 'old' M4) or move quicker as iPad Pro would transition to M4. But that is not normally how Apple does things ( want to keep selling 'old' SoCs in 'new' products for more than several years).



From what I understand is if they move the M3 to the N3E lithography process. Then the efficiency gains of this new chip decreases through the more simplified process.

It will be larger... this doesn't really have much to do with efficiency in the sense that term is normally used.
It is larger but would go from wafer to finished in weeks less time. So in a timely delivered to cover new demand that would be more efficient in that respect.

If trying tag "number of working dies per wafer" as 'efficient' then that is mostly likely the wrong term/notion. Costs matter. If the wafer processing costs are higher even if you get incrementally more dies out of a wafer the cost/die doesn't necessarily go down. ( if wafer processing cost increase is a higher percentage than the density improve got (to squeeze more dies onto a wafer, then not necessarily getting better efficient use of funds. )


Which could mean the M3 efficiency advances on the M2 chip will be smaller despite it being 3nm.

If talking about Power and performance uplift . There isn't much of a 'backslide' there.

N3B power
" ...
TSMC​
N3
vs
N5
N3E
vs
N5
Power
-25-30%​
-32%​
Performance
+10-15%​
+18%​

..."

Those numbers for a default ARM core that TSMC uses. Apple's implementation tend to lean much harder on SRAM/Cache so decent chance N3E is not some 5-7% win. Depends a lot on just how effectively Apple used N3B's FinFlex. It would be harder to do , but Apple has talent/resources to do harder stuff. Most other shops are looking for easier/cheaper.
 
  • Like
Reactions: Pinkyyy 💜🍎
The almost certain outcome here is a 30.5" or something similar. That would make it a big jump from 24" iMac, but not cannibalize Studio Display focused buyers

Studio Display buyers are less price sensitive to tolerate spending at least $1,600 on just a monitor. Much of clamor for the 'keep 27" iMac is really more so clamor to keep the $1,799 price tag ( or less). If get a Studio Display that has an ergonmic stand then have already have blown threw the old iMac 27" entry price norm.

the iMac at 32" is likely also abandoning that old entry price point. Apple isn't looking for the old historic run rates on large screen iMac sales. The Mini/Mini Pro/Mac Studio are going to have the same performance so they can't artificially herd as many folks into buying an iMac anymore. When Apple can't inflate the unit volume artificially high then the unit cost on the screen/panel isn't going to go down as much. So that whole 'bargin hunter' (get a computer for 'free') approach is likely going to get dropped.
 
Studio Display buyers are less price sensitive to tolerate spending at least $1,600 on just a monitor. Much of clamor for the 'keep 27" iMac is really more so clamor to keep the $1,799 price tag ( or less). If get a Studio Display that has an ergonmic stand then have already have blown threw the old iMac 27" entry price norm.

the iMac at 32" is likely also abandoning that old entry price point. Apple isn't looking for the old historic run rates on large screen iMac sales. The Mini/Mini Pro/Mac Studio are going to have the same performance so they can't artificially herd as many folks into buying an iMac anymore. When Apple can't inflate the unit volume artificially high then the unit cost on the screen/panel isn't going to go down as much. So that whole 'bargin hunter' (get a computer for 'free') approach is likely going to get dropped.

A base Mac mini (8 GB/256), base Studio Display, base keyboard and base Magic Mouse costs $2,420+ on the Apple website. Most people saying they want a 27" iMac say they want a minimum of 16GB RAM / 512 GB RAM. I would see Apple offering that as a minimum configuration for any larger screened iMac. This Machine is starting at $2,499 at best in an M2/M3, and likely $2,999 or more in an M2 Pro / M3 Pro.
 
I prefer the versatility of the mini and studio , particularly as two of my iMacs packed up due to the display not the internal stuff breaking and my current late 2015 is showing very odd display issues recently .. Fingers crossed 🤞
 
  • Like
Reactions: Lioness~
I prefer the versatility of the mini and studio , particularly as two of my iMacs packed up due to the display not the internal stuff breaking and my current late 2015 is showing very odd display issues recently .. Fingers crossed 🤞
8 years is a good run... for just about anything. cars, GFs, fridges, TVs, etc, etc. =)
 
  • Like
Reactions: George Dawes
A base Mac mini (8 GB/256), base Studio Display, base keyboard and base Magic Mouse costs $2,420+ on the Apple website.
Stop including the Studio Display in any serious price comparison. Anyone who cares about price is not going to consider it when far better monitors can be had for $499. Apple simply can't have that monitor and a 27" iMac in their lineup at the same time. The only reason anyone ever bought iMacs was because the intel Mac minis had terrible graphics and were too slow.

42" 4k

32" 4k

32" 4k 144 hz
 
Last edited:
"when far better monitors can be had for $499."
we are gonna have to disagree on this point. you saw my setup, 5K between 2 4K, there is diff.

Of course, goes without saying that all users have different needs. single monitors kill me.

Dual screens for over 20+ years. tjhis is 2002/3. G4933 (or maybe dual 867l) + Sony GFDW900 (must have been $1400 in 2002 money..) @ 1920 x 1080 + some 17" viewsonic design series 17". Had the dual monitors long before this even. Prob started with my PTP or PCP (powercenters) in mid 90s. And a 9600/350 with a Proformance III 8 MEGABYTE GPU. I owned a small design firm, my designer had some of the first Dell flat screens -- 2001FP $750? ( https://www.cnet.com/reviews/dell-ultrasharp-2001fp-review/ 1600 x 1200) and later 2405FPs (a "gigantic" LCD 1920 x 1080 https://www.cnet.com/reviews/dell-ultrasharp-2405fpw-review/)

Ive been there, and done that, many times over.

8.jpg
 
Last edited:
single monitors kill me.
What if you had a 42" monitor and drew a line down the middle? Then it would be dual 28" portrait displays with 2160x1920 resolution each. (Maybe a bit less because the line would cover up some pixels.)

I suppose I do have dual screens if you count my iPad. I hardly ever take it off it's keyboard, except to draw.
 
Last edited:
  • Like
Reactions: Pinkyyy 💜🍎
Apps dont work like that. I know there are apps (maybe windows only) that will split the screen with an invisible line -- but you stuck with 2 *vertical* shapes.

I have plenty of pixels. middle screen adobe apps, right screen web and finder, left screen email, font management, music, text edit, Grammarly, etc.
 
For the cost of a Pro Display you could have four 42" displays around you and one on the roof, and you could just live inside your displays.
4 "4K" displays. And my neck would hurt. Im glad you love your 42" screen. I have clients that use the same. Not for me. 20+ years has got me here. =)

I could have 4 Yugos for the price of an nice used BMW X5, but I wouldnt do it either.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.