Stuck between a rock and a hard place

An explanation of the disappointment of Apple’s new MacBook Pro

It’s been 4 years since the last substantial update to the MacBook Pro. The exact reasons why Apple decided to neglect their pro equipment, and why they still have not updated the Mac Pro or the iMac are known only to Apple, but it seems reasonable to assume that if Mac’s were still their main profit generator that this would not have happened. The iPhone is updated like clockwork at the same time every year.

This would have been a lot more tolerable if Apple had communicated the fact that they’d start ignoring their pro line of products in advance, so people could plan around it. People could have avoided waiting literally several years to upgrade thinking that there must be one coming just around the corner.

It is adding insult to injury that before Apple became a powerhouse of a company that produces consumer electronics for information consumption, rather than information production, it was the creative pro users who were their most loyal and vocal fans. Apple would probably not exist today if it were not for this core base of users that stuck with Apple while it was the underdog for so many years.

Today, things have changed. Apple no longer needs creative types. The profit from selling just the iPhone is about the same as the entire PC industry (Mac and Windows) combined, both at about $120 billion / year.

Which is why, after a hiatus of several years, people were very interested to see what Apple would come out with. There was speculation before the release: Was the delay due to Apple transitioning from Intel processors to their own ARM processors manufactured in house? Had they been spending all this time integrating 3D XPoint (futuristic storage) technology into their laptop range? Hopes flourished in the space of uncertainty leading up to the release.

They finally did release it, and it immediately seemed underwhelming by any standard: A thinner laptop, which aside from the TouchBar which many people consider a gimmick, mostly seems to amount to a series of spec bumps. It has a slightly more modern CPU. An incrementally faster GPU. A screen with the same resolution, albeit with a wider color display and higher contrast and brightness. And most worrisome of all, it is still capped to 16GB of RAM, which has been available in their laptops for nearly a decade now. I was very disappointed myself. Amongst other things I make and sell neural nets for a living, and running this one on an image more than around 5 megapixels tends to create a 32GB swap file on my 16GB 2015 MBP, so I was really hoping for a machine with a 64GB RAM option for multimedia creation.

Predictably, there has been an onslaught on Apple’s design decisions from the masses, however most of it seems based upon a misunderstanding:

• Why is it limited to 16GB of RAM? Supporting more RAM would just require a bigger battery, so instead of making it thinner, they could have included 32GB of RAM and kept it the same width as the previous model

Although this is technically true, there is a little-known legal obstacle to this: The Federal Aviation Administration has capped the maximum allowable size of laptop batteries on flights to 100 watt-hours. That explains why Apple’s 2015 pro model contains precisely a 99.5 watt-hour battery. Although the recent MBP release only contains a 76 watt-hour battery, due to the fact that there is no low-power RAM available in greater than 16GB capacities for Intel’s latest mobile CPU it can be argued that Apple are still working within that 100 watt-hour ceiling, and that they are using the best components that they can given that ceiling.

Using the other 24 watt-hours available wouldn’t be enough to move the RAM & CPU up to the desktop counterparts in their laptop and still have reasonable battery life, so given the components they are restricted to using a 76 watt-hour battery is enough. With no such ceiling maybe they would have used very different components and created a larger machine. Since that was never an option that is a moot point.

This is also why the only laptops currently available which support >16GB RAM are huge, like this one which weighs 17 pounds (8KG). This battery capacity limitation goes a long way to explaining a lot of the problems with current laptops from all manufacturers. It’s also probably the main defining force of this iteration of the MacBook Pro. Apple were determined to have a reasonably long battery life, and everything else they did had to be designed around the limited size of the battery. They couldn’t have made it bigger or more powerful if they wanted to.

• The TouchBar is a gimmick, and they could have just made the screen itself touch sensitive

Apple said that they prototyped this and it didn’t work out, and I tend to believe them. User input is something that’s hard to get right for everybody at once, and it seems like we’re just going to have to wait and see how it works out before judging this. My perspective of it has improved since I ran the TouchBar in the Simulator in the latest version of XCode, which emulates it by putting it on the screen.

I have gotten the opportunity to see how it works in practice and it’s better than I thought it’d be after I saw it onstage. My guess is that they have not nailed the entire future of user interaction on laptops with one fell swoop with this, but whether it becomes a genuinely useful tool and the beginning of future paradigm shifting design iterations is yet to be seen.

The dial with the new Microsoft Surface Studio certainly looks sexy and has the wow factor, but it is too soon to call it the winner yet. I will be interested to see in a year from now which input method is the most used and the most supported. Any predictions which declare either the winner at this point are too early.

• The GPU is disappointingly underpowered. They are up against things like the 1060 from Nvidia which is more than 3 times faster

This is about that battery restriction again. The 1060 draws 75 watts and the AMD Radeon which Apple uses uses draws a mere 35 watts. Given that these are the power draws under maximum loads, out of all of the criticisms of the new MBP in my opinion this one is the most valid. Apple could have included an Nvidia 1060 and under maximum load the battery would have drained a lot more quickly, but MBP’s have had on the fly GPU switching (from the integrated GPU on the processor to the discrete GPU) for years now, so they could have achieved this. But they have their reasons. I imagine when you’re working on that scale, including a GPU which makes the battery drain nearly twice as fast would result in a backlash from a certain percentage of their confused consumers.

• The storage capacity in the base models as the same as it was before

Apple’s internal storage speed is in line with the industry’s best, and the prices that they charge tend to be less than the competitors. On the contrary, Apple has stopped overcharging for its internal storage recently.

• It’s increasing the need for dongles

In the long term it’s doing the exact opposite: It’s finally setting USB-C/thunderbolt as the standard. It is pushing the industry along, like it has many times before by abandoning obsolete technologies. And thunderbolt is indeed making a lot of connection technologies obsolete, but due to the lack of support or peripherals for it it is now time for the industry to catch up. In the future we’ll be seeing USB-C everything.

With those things covered most of the design limitations due to the battery restriction for the other criticisms seem obvious. If you think about it, what would you have liked them to have included that is not affected by the power limitation?

And with all that said, personally I would by something like this oversized PC laptop monstrosity in a heart beat if it could run macOS, because I am (computational) power mad. But I do understand that if Apple were to release something like that that they would come up against far greater criticism than they are now. Their reputation is based on building reliable products that work well under normally balanced circumstances. Due to Intel’s lack of support for lower power RAM in capacities greater than 16GB in its latest laptop sized CPU’s, Apple’s arms really are tied, and it is Intel that dropped the ball in this respect, not Apple.

With peoples emotions running high over Apple’s latest release, we have missed the good news

Considering that today’s fastest SSD’s top out at about 2500 megabytes a second, and that the standard-until-now USB 3 has an upper limit of 640 megabytes per second, the new MacBook Pro’s 4 ports each with bidirectional (up and down simultaneously) 5 gigabytes per second is insane, and it is the first time that anybody has achieved this in the laptop market. The future of this is extremely exciting.

The 3rd party innovation opportunity here is huge. I predict that in the coming months and years we’ll be seeing thunderbolt connected external SSD’s with speeds that far exceed the internal SSD’s speed. I don’t see any reason why it wouldn’t be possible right now to create an obscenely fast RAID array of SATA SSD’s in a box on the end of a thunderbolt cable, and create the worlds fastest consumer storage drive by stacking up the drives. We are finally not limited by the speed of SATA at a measly half gigabyte per second, which has been the bottleneck in external I/O for many years now throughout the entire PC industry.

By having all of this bandwidth and power (100 watts over USB-C, and the MBP can output 15 watts itself on a USB-C cable out) available over these ports we’ll be seeing innovations we haven’t even thought of yet. And having power run over the same cable is just the icing on the cake, now there is no reason that thunderbolt cannot become a power connection standard, having it coming out the walls to power everything from kettles to lamps to tv’s to computers. We’ve been seeing ridiculous USB devices for years such as usb powered fans, robotic cats, etc for years, showing that the market has a willingness to create these things, and they only had 4.5 watts to work with. Now, with the advent of USB-C with its 100W available, it’s finally possible to make actually useful USB-C powered peripherals.

Even just having a standardized power supply will be a game changer. It’ll mean that we can have a battery pack standard that will power multiple devices. Imagine being able to have a USB-C charger which will power your flashlight, power your TV when you go camping, and recharge your MBP, all with the same battery with a single USB-C out cable. Part of the attraction of USB-C is that the device will draw the amount of power that it needs, so compatibility amongst all our devices can finally arrive. Pairing that with huge amounts of data what else could anybody want out of a connection standard? The industry for USB-C devices hasn’t even started yet, but if Apple continue to set industry standards like they have in the past, it’s going to be massive.

 Follow @BenedictSlaney

Vote on Hacker News