Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
For chip design? Yes, I have (among other jobs as a hardware design engineer, and systems engineer - in the high-speed digital signal processing realm - at other companies). Developed full-custom very high-speed highly pipelined CMOS (effective hundreds/thousands gop/s rates), with our own tools and libraries. Resulting in a variety of multi-channel high-speed digital downconverters and up converters, digital filters, etc for direct-sample A/D-converted IF and RF spectrums on the receive side and high-speed DACs on transmit, QAM modulators, digital pre-distortion to correct power amplifier non-liearities, etc. Tape-outs went to various contract fabs as we were a small company. Devices were initially used by the military and govt agencies (they could afford the high prices - $300 for a four channel DDC), and then commercial wireless infrastructure on the BTS side, as our company grew.

SMH???

OK, how about you?
In that case you seem well qualified to clarify what you meant by rolling their own and bypassing ARM?
 
Now you’ve abandoned your whole “no, because segmentation” argument. If you think no one would buy it, there’s no point in protecting it from segment crossover. You’re really making the argument 32GB won’t be offered on rMB at all
No, I used the term 'segmentation' for all kinds of product lineup considerations. It simply looks bad if the MB has 32 GB and the MBP doesn't. And nobody buying it is adding to that as third-parties will point out how relatively pointless it is to have 32 GB of RAM in a 4.5-W CPU. I never said that Apple wouldn't add 32 GB to the MB out of fear of reducing sales of the more expensive laptops.

I repeat, there is almost no upside but some downside. Why would Apple then go for this? And I guess if Apple doesn't add 32 GB to the MB before it adds it to the MBP, you will be sure to point out that there could be all kinds of reason as why Apple does something and that in no way it supports my arguments.
 



Intel today introduced a range of new eighth-generation Core processors [PDF] appropriate for future MacBook Pro, Mac mini, and iMac models.

core-i9-coffee-lake-mac-trio.jpg

The most notable new chip is the first-ever Core i9 processor for notebooks. With six cores and 12 threads, Intel says the Core i9 is the highest-performance notebook processor it has ever designed. The H-series processor has a 2.9GHz base clock speed with a Turbo Boost frequency of up to 4.8GHz.

Given the Core i9 is a 45W chip, it is appropriate for the high-end 15-inch MacBook Pro and could be included in a refreshed version of the notebook as early as this year. Apple last updated the MacBook Pro lineup with Kaby Lake processors at WWDC in June 2017, so a Core i9 model could debut at WWDC 2018.

Of note, while the Core i9 processor allows for systems with up to 32GB of RAM, this is unlikely to apply to the next MacBook Pro, since low-power DDR4 RAM is still not supported. Back in 2016, Apple's marketing chief Phil Schiller said 32GB of standard DDR4 RAM would compromise battery life.

The eighth-generation Core processor family also includes new quad-core Core i5 and Core i7 processors with base clock speeds between 2.3GHz and 2.7GHz and integrated Iris Plus graphics. These 28W chips, part of the U-series, are suitable for future 13-inch MacBook Pro and Mac mini models.

Intel says the new Core i9, i7, and i5 processors for notebooks are based on its Coffee Lake platform and leverage its 14nm++ manufacturing process, enabling the chips to deliver up to 41 percent more frames per second in gameplay or edit 4K video up to 59 percent faster than the previous generation with the same discrete graphics, based on its internal benchmark testing.

As with Intel's Kaby Lake Refresh processors introduced last August, these new Coffee Lake chips pave the way for a quad-core 13-inch MacBook Pro should Apple choose to release one. The current lineup is limited to dual-core models.

Intel also expanded its lineup of eighth-generation Core processors for desktops today after an initial rollout last October. Two chips suitable for future 4K and 5K standard iMac models include six-core Core i5-8600 and Core i5-8500 chips with base clock speeds of 3.1GHz and 3.0GHz respectively.

The desktop lineup also includes six lower-power 35W chips with four or six cores and base clock speeds between 2.1GHz and 3.2GHz. While the current Mac mini lineup uses 28W chips, previous generations have used up to 45W chips, so the 35W processors could be suitable for future Mac mini models.

All in all, Intel has potentially laid the groundwork for a high-performance, top-of-the-line 15-inch MacBook Pro, quad-core 13-inch MacBook Pro models, a long-overdue Mac mini refresh, and updated iMacs as early as this year.

Looking further ahead, Bloomberg News on Monday reported that Apple plans to design and use its own processors for Macs starting as early as 2020. Intel shares saw their biggest price drop in two years following the report.

Article Link: Intel's New Core i9 and Coffee Lake Chips Pave Way for Quad-Core 13" MacBook Pro, Mac Mini Refresh, and More

I love the optimism...
 
  • Like
Reactions: TheHurryKayne
No. But I would imagine all those Windows laptops that are using higher-power DRAM to get to 32 GB and beyond would appreciate having LPDDR4 so that they would get more than the 3 or 4 hours of battery life their Users "enjoy" now.

Thicker and heavier laptops can afford bigger batteries. Lots of folks asking for 32GB are really in the "Desktop Replacement" category. Work is done at desk at work ( and perhaps desk at home ). And they already have system's to buy that meet that criteria.

The system design tradeoff that Apple is in with their choices is that RAM capacity and Battery capacity are at odds with one another. More RAM leads to less battery. More battery leads to less RAM. They are both competing for horizontal space.

Windows system vendors can perhaps sell more 64GB laptops into a broader set of users when get to LPDDR4 support, but they are selling more than Apple is now.
 
No, I used the term 'segmentation' for all kinds of product lineup considerations. It simply looks bad if the MB has 32 GB and the MBP doesn't. And nobody buying it is adding to that as third-parties will point out how relatively pointless it is to have 32 GB of RAM in a 4.5-W CPU. I never said that Apple wouldn't add 32 GB to the MB out of fear of reducing sales of the more expensive laptops.

I repeat, there is almost no upside but some downside. Why would Apple then go for this? And I guess if Apple doesn't add 32 GB to the MB before it adds it to the MBP, you will be sure to point out that there could be all kinds of reason as why Apple does something and that in no way it supports my arguments.
Well, Apple usually makes decisions based on business reasons, not on optics. You think Apple would never introduce 32GB on rMB before MBP because it would somehow bruise their corporate ego. I couldn’t disagree more, and I don’t suppose we’ll ever know who’s right—unless 32GB is available first on rMB ;)
 
I'm in the same boat as you, but you're being foolish if you think Apple would actually accommodate our needs at the expense of 99% of the rest of their user base.

But who is to say the 16GB and lower machines wouldn't still be at whatever performance they normally would? Just adding an option shouldn't change other configurations.
 
yea I miss the early intel/apple days where Apple was getting first pick on processors. Its been sad how slow Apple has been to up date machines. Mac Pro 5 years or so and the Mac mini. For a company this size I think they can afford to throw a updated chip into an existing design. And Keep this stuff user serviceable too.

Apple doesn't respect it's consumers. There's your answer.
 
That's alright because Apple is way too busy dreaming up the next color for the iPhone X to be bothered releasing another Mac Mini, shameful. :p

Dreaming about the next color for the iPhone X and the new Animojis... Who can care less?

If they insist with the useless expensive Touchbar, limit the RAM to 16GB and not adding back the Mag-Safe (one of the best features ever), I would not touch the Macbook with a ten-foot pole.
[doublepost=1522873360][/doublepost]
So I see Apple only putting 4 cores in a TB MBP to induce people to buy touchbars. What do you think?
I think the touchbars are useless andthey are only an excuse to overcharge for a bad designed Macbook
 
  • Like
Reactions: TheHurryKayne
Well, Apple usually makes decisions based on business reasons, not on optics. You think Apple would never introduce 32GB on rMB before MBP because it would somehow bruise their corporate ego. I couldn’t disagree more, and I don’t suppose we’ll ever know who’s right—unless 32GB is available first on rMB ;)
Apple does not make decisions based on optics? Many would claim that Apple makes way too many decisions based on optics.

And having a clear and simple lineup is based on sound business reasons. This is not about Apple's ego but how Apple's products and its lineup are perceived by the public is influencing purchase decisions.

And thanks for confirming that your prediction turning out to be incorrect would not be accepted by you that you were wrong and I was right.
 
A 6-core mac mini 32gb of ram.. wouldn't that be something...

I mean, really, just take the components straight out of the MBP put them in a little box and majority of people will be SUPER happy.


Good luck on that one. It would outperform the iMac Pro in some tasks where the Xeon chips aren't efficient as the consumer lineup,.
[doublepost=1522877108][/doublepost]
Man, a six core mac mini would be an instant buy...


Your best shot is buying a Macbook Pro, close the lid and hook it up to an external monitor.
 
  • Like
Reactions: TheHurryKayne
According to you, almost nobody would buy it and almost nobody needs 32 GB of RAM on a 4.5 W CPU. So I really don’t see why you’re so concerned that it could be important in influencing purchasing decisions.

But that’s just the latest reason you’re putting forth. Previously you claimed it was because Apple was afraid of “drawing even more attention to the lack of 32 GB on the 15" MBP” and rMP upstaging MBP.

Fine, I get it, you think Apple won’t offer 32GB on rMB if it’s available before MBP. For all of the various reasons you’ve put forth. I think they’ll bring it when it becomes available. I don’t think either of us will be able to change the other’s mind, so there’s really no point in either of us trying.
 
The i5 8269 and 8259 are configurable to TDP-down of 20W. That isn't 15W but is also isn't 28W either. It is closer to 15 (5) than it is to 28 ( 8) .

https://ark.intel.com/products/137980/Intel-Core-i5-8269U-Processor-6M-Cache-up-to-4_20-GHz
https://ark.intel.com/products/135935/Intel-Core-i5-8259U-Processor-6M-Cache-up-to-3_80-GHz

The eDRAM is bigger. Just bleeding off clock speed to go lower TDP may or may not be useful to most other system vendors. Intel may not offer anything different for a relatively long while (i.e, until 9th Gen). You can get into slippery slope where the performance from the previous model just pragmatically evaporates.

If the gap is 5W and can get 1-2W back from a reduction in other components ( e.g., PCH , TB, etc. ) then the gap is pretty small to justify a separate processor product SKU.




This model looks like what the MacBook Air update could have been. ( Screen is a bit more higher than a MBA would get but along the lines Apple has been going. ). I doubt Apple would pass through some huge discount for 7th gen processors.

The non touch bar MacBook Pro isn't that different from the touch bar version so there might actually be no need to reduce the clock speed of a successive product using the 28w CPU other than for product segmentation purposes. The NTB MBP is already hamstrung by only having 2 Thunderbolt 3 ports.

Rather than hunting around for a suitable 15w Iris Graphics part Apple could instead choose to dip down to the 28w i3 range i3-8109U which is 3GHz dual core, 4 four threads with Iris Graphics. It'll need some work from the marketing team to explain why 3Ghz i3 is in a lesser model than the 'slower' i5 quad core.


[doublepost=1522878891][/doublepost]
That probably isn't the point. For system designers willing to use so-DIMMs the limit is 32GB.

Apple's limit isn't just LPDDR4. It is also DRAM density. Since there are soldering flat to the logic board surface, there is only so much room for RAM chips ( fixed number of chips). To go to 32GB Apple needs denser (higher capacity ) RAM chips as well as LPDDR4. The issue is whether anyone is making those at market tolerable prices yet?

Making a RAM controller for RAM almost nobody is going to buy doesn't work.

Apple's thinnest mania is also a driver here, but Apple doesn't want to say that.

The newer DRAM controller is in Cannon Lake which is delayed. That's Intel's contribution to the problem. The PCH chipsets being used in the new 8th gen systems that are Coffee Lake are from what would have been the Cannon Lake offering. Just missing the CPU "half" which is where the DRAM controller is. Intel might have made it a priority to pull it back to Coffee Lake if most system vendors had asked to make it a priority but most likely they didn't so it wasn't. Apple's isn't Intel's only "tier 1" customer.

Phil Schiller has already said in the last 18 months for the 2016 MBP that the power draw was the main reason for not being able to offer 32Gb of RAM in a MacBook Pro. The factors here are the power draw from the additional RAM plus the logic board design that would be required to accommodate it.

So in effect Apple would appear to think that a laptop with 8Gb RAM has a better battery life than an identical one with 16Gb. A 32Gb laptop would be a comparative battery hog - people probably don't notice because Windows laptops don't mind needing bigger case/motherboard to use DDR4 and achieve the RAM requirement at the expense of battery life which people then blame Windows for the inefficiency.

Apple could go 32Gb right now (and could have for some time) but won't use the standard DDR4 configuration for power draw considerations.

LPDDR4 won't be coming in for Intel CPUs till Cannon Lake - the next big major architecture change.

LPDDR3 only goes up to 16Gb but that has the power profile required so Apple will have to stick with that.

It's the same reason we won't see Optane SSD in portable Macs - due to the power draw.
 
Last edited:
BSEE and MSEE with specialization in computer architecture.
30 years Silicon Valley
Amdahl computer development group in the late 1980's.
From there many companies where I developed custom processors for TCP/IP acceleration, pixel processors for H.264, AVS encode and decode, audio signal processors, ARM processor implementation for GSM cellphones, PCIe switches.
Research in cache coherence.
Papers on FPGA and ASIC unified flow and development.
Patents, and IEEE Senior Member.

I've done designs in BiCMOS, ECL and CMOS at nodes starting at 3.5 microns (I'm old) all the way down to 16nm FinFET.
I've lost count on how many chips have shipped my my stuff in them.

My area of specialization is computer architectures.
Now credential aside.

it makes no sense for them to develop a completely new instruction set with all the collateral behind that it will take.
Instruction set design alone is a monumental chore. I've done it for a custom video processor. Working with the engineer so we could have a compiler was non-trivial. I can't see them going down that path. Anyway, the other thread on moving away from Intel discusses this more and I don't want to pollute this one.
https://forums.macrumors.com/thread...om-mac-chips-starting-in-2020.2113132/page-53

We seem to have some overlap working a similar amount of time in SV (in Sunnyvale and Palo Alto). Though I only go back to 1 micron (2 metal/1 poly, and using Magic, no less - which actually gave us a competitive edge on performance and die size).

In the other thread you make the point it's not that Apple can't do it, but would it make sense. I think it does for a variety of reasons. Selling roughly 20M Macs a year certainly doesn't require a $5B-$10B foundry. And I don't think the 2020 availability means Apple has been only working on this for 2 years. I wouldn't be shocked at all if this has been ramping internally up for years. They have the resources and $. And an enviable track record with ARM under Srouji, though staying with ARM would likely not get them past Intel. I can see Apple making the investment with a new architecture in mind. That could very well explain some of the 40% R&D increase over the last few years.

Who would have thought a consumer computer company would develop a smartphone in 2007 that in a very short period of time crush Erickson, Nokia, Motorola, Sony, and others (though unbeknownst to Motorola, they helped Apple get there)?
 
Intel broke the cadence when they had yield issues with the 14nm Broadwell debacle. Apple drove a lot of innovation in those first 5-7 years of partnering with Intel, but at some point, the relationship may have soured. It could be Apple, if they drove Intel as hard as they drive some of their other partners. It may have been Intel who pushed back...who knows. I have to agree though that the lack of at least some update to the 2013 Mac Pro was just laziness because the market is so small and the lack of Mac mini updates is completely inexcusable given the strategic position the mini played in gaining switchers. However, after watching the media event last week, Apple is still trying to lead with the iPad as a solution completely at the expense of desktop and mobile computers, which tells me they still aren't getting it.
I think iPads are ideal for most skool projects. I wouldn’t read into it like that.
A desktop Mac is required for basically any serious form of content creation—so just a handful of subjects would call for desktop machines.
 
I don't think they'll be using this i9-8950HK. Apple always uses the HQ variants not HK. You never know, but I
I don't agree. I work in IT in advertising and support about 10 of the new MBPs, and I've found them the most problematic Macs I've ever dealt with - and I've been using and supporting Macs since 1984. The unreliability of being forced to use hubs for network, monitor and wired keyboards, combined with their very temperamental keyboards and ultra-thin, delicate screens make these MBPs a very unsatisfactory user experience. Most of my users hate them. If they aren't having issues with networking or monitors, or keys that won't press due to a tiny spec of dust getting stuck underneath them, they've got cracked screens. They also seem very susceptible to water damage. A minor spill that wouldn't hurt a MacBook Air, can kill these things stone dead. I hate the darned things and am very reluctant to buy any more. For our use-case MacBook Airs - despite their old-fashioned specs - are a much better deal. They aren't sexy, but they're tough, cheap and totally reliable. The new MBPs might make a reasonable "personal" laptop, but in an office environment, they're a nightmare.

This depends how people treat these machines. My 15" from 11/2016 doesn't even have a scratch but I take good care of it and always wipe the keyboard down before I shut it down. I understand that most people don't either have the time or want to bother doing this. I use compressed air once in a while to get rid of any dust or particles. I've never had any problems with the keyboard or display as a result. I guess if people are slamming the lids then screens will crack.

My only complaint with the keyboard which is a design issue is the "sticky keys" problem around the center when the CPU/dGPU are being pushed and running hotter. I've used lots of MacBooks (pre-retina, Airs, different years of 3rd gen retinas, 13", 15" and none of them have this problem. I haven't used a 2017 15" but I'm guessing they suffer from this issue as well. That said, I still wouldn't go back to using a 2015 or earlier machine and definitely not an Air in 2018. The Airs are frustratingly slow.
[doublepost=1522886248][/doublepost]
You did not give one example to back that statement for the everyday user. TB3, while being higher performance, is not usually a superior substitute to USB. It's expensive, and most of the gadgets one plugs into a notebook do not need the speed of TB3. eGPU? Useful for a few, but who wants their Macbook straddled by that behemoth box. Most people don't need or want one. Nor do most people connect 10-bay drive arrays that might need the speed of TB3.

So far I have not seen any sign that USB is going away. Flash drives, wireless mice/keyboards (non-Apple brand), printers, scanners, iPhone/iPad cables...you name it, they are still USB-A. USB-C is trying hard to replace it, but it's still USB.

I would say that user-upgradeability ranks higher on the average users' wish-list for Macbooks than TB3.

USB-C is expensive, while Type A is not. It's really that simple. Apple was an early adopter and it's still pretty new. Go back to when USB was first introduced. It took years for it to become everywhere and high capacity flash drives to become cheap. PS/2 ports and everything that USB replaced took a while to die on PCs. The bottom line is that TB3/USB-C is here now and buying legacy tech is a waste since it's clear Type-C is the future. Apple will probably dump lightning on phones and tablets by 2020.
 
Thanks for the info, I appreciate it. Interesting about the fragile screens, I hadn’t heard complaints before.

Sounds like you’re using the dock for Ethernet/keyboard/monitor in the office and maybe charging; that’s a very typical usage and I wouldn’t expect much difficulty, but it sounds like it’s been problematic for you. Is it USB3 or Thunderbolt dock? Are they all the same model?

Thx again!
in my never ending quest to find the perfect USB-C-HDMI/USB-3/GB-Ethernet/charging hub, I've tried a few of them, and I instantly buy new ones that pop up. I'm waiting to get my hands on the new Trendnet Mini USB-C HD Docking Cube currently.

So far, none are perfect, but Satechi is probably the best, and probably the best looking too (if you care about such matters - I do). I've just bought two of their "Type-C Multi-Port Adapter 4K with Ethernet V2" to see how "V2" is improved. Needing ethernet reduces your USB-C hub choices hugely, and it's also where the most common grief is.

The main issue, apart from commodity ethernet chips of unknown origin and reliability on a Mac, seems to be the USB-C connection itself. People might complain about the cost of Apple's own adapters and cables, but at least they always feature premium plugs - many USB-C hubs don't - and any unreliability in that connection, when your monitor, keyboard, network and power are all funnelling through it, is a real pain in the ass. The slightest jiggle of your Mac or even your desk, and you lose your monitor or network.

I also suspect that some of these issues revolve around USB-C/ThunderBolt 3. Ours are all USB-C hubs, not ThunderBolt 3. I use a 12 inch MacBook myself (the best Mac Apple make these days IMO, apart from it's keyboard), which is purely USB-C, and I have far fewer issues. I moved to the MacBook prior to buying all those MBPs, precisely so I could test out the reliability of USB-C hubs, before I foisted them on my users. But it wasn't a great testing bed, because the MBPs - with their hybrid USB-C/ThunderBolt 3 ports - seem to have more issues using these hubs than the MacBook does.

For general desktop use, our usual hub is the Minix Neo C. It's a cute, well built, sleek little device, with all the ports you need, available in silver, space grey and even gold (my macBook is gold, so I needed a matching gold hub, of course). BUT, it needs a software driver for ethernet. I had trouble with my initial one, but Minix were very helpful in getting it working, and even sent me a beta of updated drivers. For some users, the Minix hub has been fine (I'm using mine right now). But others have had no end of grief with them, and I've had to change them over to others.

So I tried the Satechi Type-C Multi-port Adapter with Ethernet. Users who had had issues with the Minix, still often had issues with these. Doesn't need a driver for ethernet though.

Apart from that we've tried various mixes of adapters. The OWC USB-C Travel Dock is a nice little unit, but doesn't have ethernet. So I've used that in combination with a Belkin USB-C Ethernet adapter.

Apple's own USB-C adapters are an odd lot, and they don't do one that incorporates everything we need anyway. I don't particularly care about the premium Apple prices, but I find Apple's clumsy brutalist minimal styling of them a bit of a turn off. So I have one user who has a hideous jumble of Belikin USB-C ethernet adapter, Apple DisplayPort to USB-C adapter, and Apple Digital AV Multiport Adapter. It works, but it also looks like a junk pile.

For non-office use we all have the Satechi Aluminum Type-C Slim Multiport Adapter. It's a great little device, very nicely designed and solid as a rock (and available in gold). I think it's the pick of the USB-C adapter hubs. But it doesn't have ethernet.

And so the quest continues. As time goes on we'll inevitably move to having everyone on wireless keyboards, USB-C monitors and peripherals and for laptop users, purely wireless networking. At that point things will work much more elegantly.
 
Last edited:
Im walking around with a 12.5 - 2.4lb Quad i7.

The fact that Apple not dona quad 13 at this point just shows how far behind they are.
 
I don't think they'll be using this i9-8950HK. Apple always uses the HQ variants not HK. You never know, but I


This depends how people treat these machines. My 15" from 11/2016 doesn't even have a scratch but I take good care of it and always wipe the keyboard down before I shut it down. I understand that most people don't either have the time or want to bother doing this. I use compressed air once in a while to get rid of any dust or particles. I've never had any problems with the keyboard or display as a result. I guess if people are slamming the lids then screens will crack.

I'd have thought that too. But the three cracked screens we've had were all MBPs used by nerdish users who are very careful with their laptops. They've previously handed me back laptops after years of use, still looking nearly new.

The cracking we've had is always at the bottom of the screen, near the hinge. I think the tolerances between the base and the display panel are very tight, and the slightest misalignment there causes damage to the screen. The 13 inch MBP we just replaced the display on had developed strange cracking in the matte black strip at the bottom, near the start of the hinge - on both edges. Eventually one of those cracked areas got so severe it seemed to press into the screen, and crack it. Seemed obviously a manufacturing defect to me - the MBP was cosmetically perfect otherwise, with no sign of any mistreatment at all. but Apple disagreed, so we had to pony up to replace it.
 
I'd have thought that too. But the three cracked screens we've had were all MBPs used by nerdish users who are very careful with their laptops. They've previously handed me back laptops after years of use, still looking nearly new.

The cracking we've had is always at the bottom of the screen, near the hinge. I think the tolerances between the base and the display panel are very tight, and the slightest misalignment there causes damage to the screen. The 13 inch MBP we just replaced the display on had developed strange cracking in the matte black strip at the bottom, near the start of the hinge - on both edges. Eventually one of those cracked areas got so severe it seemed to press into the screen, and crack it. Seemed obviously a manufacturing defect to me - the MBP was cosmetically perfect otherwise, with no sign of any mistreatment at all. but Apple disagreed, so we had to pony up to replace it.

Thanks for the info. I guess either I've been lucky or just have to keep being careful with this damn thing. I bought AppleCare+ for it which I never do due to it being a repair nightmare for indie shops. I haven't checked if they have already started repairing these or trying to repair them. The first batches of 2016s are out of warranty for those who didn't get AppleCare. Did they charge you $99 or $299 to fix this? Do they give you back a refurb or actually repair the machines?
 
Thanks for the info. I guess either I've been lucky or just have to keep being careful with this damn thing. I bought AppleCare+ for it which I never do due to it being a repair nightmare for indie shops. I haven't checked if they have already started repairing these or trying to repair them. The first batches of 2016s are out of warranty for those who didn't get AppleCare. Did they charge you $99 or $299 to fix this? Do they give you back a refurb or actually repair the machines?

LOL. No, they charged $1000! Obviously now I get AppleCare+ for these darned things.
 
LOL. No, they charged $1000! Obviously now I get AppleCare+ for these darned things.

Wow! That's half or more than half retail depending the configuration. Really insane. Without AppleCare they're almost like totaled cars. I still haven't decided whether I should keep mine past AppleCare or sell it before the 3 years are up.
 
The system design tradeoff that Apple is in with their choices is that RAM capacity and Battery capacity are at odds with one another. More RAM leads to less battery. More battery leads to less RAM. They are both competing for horizontal space.


??? The previous MBP model had a 99 Wh battery (as it should be) and enough space for two DIMM slots. Thickness was slim enough for me.

Apple, just maximize the battery, allow us to upgrade RAM and storage, and be done with it. Two RAM slots, 2 M.2 slots. Done. Use the rest of the space to fit a more powerful GPU.
 
  • Like
Reactions: dmi
I don't care if it 'compromises battery life', I know what I'm getting into. 32GB is not a luxury anymore, it's needed for anyone who does heavy development work on the go.

90% of the time, I'm powered, anyway; the laptop just gives me the freedom to take that power anywhere.
You are probably only 5% of market so why should I have to pay for your 32Gig need go get yourself an HP.
 
...

Phil Schiller has already said in the last 18 months for the 2016 MBP that the power draw was the main reason for not being able to offer 32Gb of RAM in a MacBook Pro. The factors here are the power draw from the additional RAM plus the logic board design that would be required to accommodate it.

Phil Schiller also came out on stage with the Mac Pro and said "can't innovate my ass" and then promptly fell into a 4 year long Rip van Winkle slumber with the Mac Pro upgrades.

He also talked about how "courageous" Apple was about dumping the earphone jack when that was primarily about needed internal space for the virtual button's haptic interface.

Schiller talks alot of Sales smack and occasionally misdirection. I wouldn't bet the farm on his technological insights truthfulness. ( Jobs talked smack at times too but he had a better reality distort field. )

This is teardown of the 15" Mac Pro. the 4 chips highlighted in orange are the RAM.

CSVuDEmvqgaDMqSn.medium

Step 7 https://www.ifixit.com/Teardown/MacBook+Pro+15-Inch+Touch+Bar+Teardown/73395

They are four 4GB (32Gb) RAM chips. There aren't any 64Gb RAM LPDDR3 I could find in a quick search. In order for Apple to go to 32GB they'll need four 64Gb ( 8GB) chips. (given the relatively small amount of space allocated to RAM) I don't think anyone is looking to make those for LPDDR3. And phones are sucking more than a few of the LPDDR4 ones. How the memory dies are laid out also matters to the CPU's DRAM controller as much as LPDDR3 vs LPDDR4. The DRAM controllers have relatively fixed mappings they'll handle and that plays a constraint role. ( so yes it is a bit of a chicken vs egg thing if the D controller doesn't do denser LPDDR3 then the RAM makers won't make it. )



Apple doesn't have a "Fn Key" (touchbar less ) 15" model. Apple just fell back to the MBP 15" thickness levels of 2011-2012 they could deliver a solution. They don't want to. Those are choices they made that aren't just power.
 
Last edited:
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.