The future of the MacBook Pro
This is continued from the mini/iMac wishlist topic but I want to keep them separated. Can we all pretty much agree that the retina MacBook Pro is the future and that eventually the unibody MacBook Pro is going away.
So it would eventually look like this in about 2-3 years or so.
Base 13" retina: $1,199
Higher-end 13" retina: $1,499
Base 15" retina: $1,799
Higher-end 15" retina: $1,999
I still don't feel that the 15" retinas should have integrated graphics. They need discrete and people should be allowed to do good to great gaming on them as well as great professional work.
So it would eventually look like this in about 2-3 years or so.
Base 13" retina: $1,199
Higher-end 13" retina: $1,499
Base 15" retina: $1,799
Higher-end 15" retina: $1,999
I still don't feel that the 15" retinas should have integrated graphics. They need discrete and people should be allowed to do good to great gaming on them as well as great professional work.
Comments
This may be a breaking point for the retina machines and the traditional screened MBPs. If the screens continue to be expensive then a professional machine built around integrated graphics and a standard screen may be in order. The biggest advantage the MBP has right now is in secondary storage. Just the simple ability to accept disk drives, maybe even two, make the MBP a go to machine for many.
To put it another way, I still believe that the MBP sells well because some have the need for massive storage that is only cost effective today in magnetic media. At some point the retina machine will configure with enough flash to truncate the MBPs sales. It will likely be 2015 before the MBP gets axed for this reason.
Quote:
Originally Posted by wizard69
Actually I'm not too certain that the Unibody Mac Book Pro (MBP) will go away completely.
They will because thinner. Apple are also undoubtedly pissed that they still have to sell laptops with optical drives in them, so there is another reason to ditch the thicker MBP. They even gimped the iMac to make it thinner, and that's a desktop where there is no functional benefit to it being thin. There are good reasons to make a laptop computer thinner and lighter, but logical design has nothing to do with it. Ive wants thin, Jobs gave it his blessing, that's how it is.
As soon as Retina display prices drop, the standard MBP w/ODD and HDD is history.
Maybe maybe not. MBP may go, but I still see Apple having a three prong strategy with respect to laptops in the future. Laptops are 80% of Apples Mac sales, they really don't want to screw that up. Part of that means Laptops with large secondary storage capacity.
Sure but that doesn't stop the growth in storage space demand. For example I've been installing XCode for some time which you would think would be easy on disk space. It isn't though, by the time you have very thing installed you have gone through Giga Bytes of disk space. This for an IDE. Down load a couple of videos from apple that support the use of XCode and even more disk space is gone. It isn't just XCode either, install Eclipse with a lot of extras and you will chew through another GB.
I know that 512 GB or even 1TB sounds like a lot of space to many but the reality is it isn't these days. Take this from a guy that dumped huge sums of money into a 5MB hard disk for his Mac Plus all those years ago. I haven't even touched upon office productivity apps, TEX or special purpose apps. When I first got the MBP a few years ago about 90 GB was used up just by app installs before I really had any data on the machine. Eventually I dumped open office simply to save disk space.
So what I'm saying is that 512GB won't remove the need for bulk storage.
What happens when we hit the perpendicular storage limit? There's gonna be a gap there, where SSD capacities both aren't large enough nor cheap enough to make up for it. I haven't heard of any new spinning disk techniques.
The real question is which of the up and coming technologies will get industry acceptance the way flash has and thus support the R&D that flash and magnetic technology have gotten. If you had a crystal ball and some money to invest you could easily retire rich in ten years or so. Just invest in the right solution.
By the way this is one of the reasons I promote the idea of a major overhaul of the Mac Pro. The expectation
Is that the secondary storage market will change dramatically over the coming years, probably more so than it has since the days of the floppy. This means SATA will die gracefully while high performance systems move to solid state arrays connected over fast PCI Express channels. Such a move will provide some users with a rather significant boost in system performance that can be likened to the move to the first dual core systems.
The sad reality is that when you look at the so called Pro computers we are seeing that the choke point these days is secondary storage. The funny thing here is that Apple demonstrated that very nicely on the low end AIRs, where the puny processors in those machines are able to mask their low performance by the fast access to secondary store. For many pro users the problem is one having enough capacity in a SSD, so many have to compromise knowing full well that they could benefit from a SSD.
So for these users that need large volumes of storage you can see demand for the traditional MBP remaining strong. If Apple where to offer tiered storage on the machines from the factory I suspect sales would remain strong for a very long time. Now offering a Fusion drive in the classic MBP is something Apple has not decided to do yet, however such an offer would make the machines very attractive for certain types of users and would certainly help maintain sales. From my perspective it makes a lot of sense to maintain a wide range of laptops due to the high percentage of Mac sales that go that way. Now Apple may not see it that way so who knows. To be honest I don't see them dropping the classics this year no matter what.
That's when we go into the
( •_•)-??-?
(??_?)
4th spatial dimension.
http://www.engadget.com/2012/03/19/seagate-hits-one-terabit-per-square-inch-compares-self-favorabl/
There's also this kind of thing:
http://www.gizmag.com/hitachi-glass-data-storage/24301/
for incorruptible storage. I always remember sci-fi films having glass pen-drive sized storage devices. That can replace tape storage and Blu-Ray.
Some of these density increasing methods are bound to help SSD too though. We can already get 64GB microSD cards in 165mm^3. The 2.5" form factor is 48895mm^3 so they should be able to get 296 x 64GB = 19TB in a 2.5" form factor, obviously minus the controller and board and other connections. Maybe half that total volume to about 10TB.
I think it's price more than anything holding it back but reliability will be a factor at lower sizes:
http://www.computerworld.com/s/article/9225064/OCZ_talks_up_2TB_4TB_SSDs_previews_Thunderbolt_enabled_drive?taxonomyId=234&pageNumber=2
The density will progress like with computing power as they go to lower fabs.
http://news.softpedia.com/news/SSD-Prices-to-Go-Low-as-0-4-dollars-per-Gigabyte-267666.shtml
"Once the NAND flash industry moves to the 10 nm manufacturing process, the density of flash cells per wafer will be more than double."
Apple can already get 768GB into their blade form factor so at 10nm, 1.5TB and the volume is 9670mm^3 so about 5 of those fit into a 2.5" form factor = 7.8TB and about $3000.
That's a little ways out yet though, next up should be 16nm:
http://www.myce.com/news/manufacturers-preparing-for-16-nm-nand-by-2013-64993/
So possibly affordable 1TB SSDs in the next year or two, 2TB in the next 5 years.
I'd like to see the 11" Air go. Although it's not quite a netbook because of the shape and performance, it's too small. The SSD prices should allow the 13" to get down to the 11" price point with 64GB. That means it would top out at $1299. They'd leave the entry 13" MBP at $1199 and 15" at $1799 and just make the Retina models replace the higher-end older MBPs, cutting the price $200-300.
I agree with this 100%.
Originally Posted by Marvin
That's when we go into the
( •_•)-??-?
(??_?)
4th spatial dimension.
???? ???? ???? ???? ????
YEAAAAAAAAAAAAAAAAAAAAAAAAAH!
Also, I'd like to see a tachyonic hard drive. It'll have files on it before I've decided to download them, but we'll have to wait until the thiotimoline hybrid version comes out before it can cache things before I want to open them.
3.5-inch hard drives… …up to 60 terabytes
????, also ????
"Once the NAND flash industry moves to the 10 nm manufacturing process, the density of flash cells per wafer will be more than double."
If the heat assisted whozawhatsis can be done properly, would SSDs run into a shrinkage wall before they reach capacities comparable to 3.5" drives?
We're almost to the really interesting part of a lot of technologies. Processors will hit a physical wall, so they'll have to go quantum, solid state storage hits a physical wall, so it'll have to adapt, and connectors… well, they're the bottleneck now that we've gotten rid of optical discs, right?
Quote:
Originally Posted by Tallest Skil
If the heat assisted whozawhatsis can be done properly, would SSDs run into a shrinkage wall before they reach capacities comparable to 3.5" drives?
We're almost to the really interesting part of a lot of technologies. Processors will hit a physical wall, so they'll have to go quantum, solid state storage hits a physical wall, so it'll have to adapt, and connectors… well, they're the bottleneck now that we've gotten rid of optical discs, right?
Many of the predicted problems there are NAND related, not necessarily related to the concept of solid state storage. NAND is just what is currently used. There is a lot of room for improvement. NAND is far from a perfect solution. In fact too many people just assume it to be more reliable than spinning drives and make the stupid mistake of neglecting backups.
You can get a 4TB SSD in a 3.5" form factor already:
http://www.anandtech.com/show/5322/oczs-4tb-35-chiron-ssd
That pretty much matches current HDDs. Once they double it again, they'll be able to fit 8TB.
We're at about $0.70/GB:
http://www.amazon.com/Samsung-Electronics-sata_6_0_gb-2-5-Inch-MZ-7TD500BW/dp/B009NHAF3I
so 4TB would cost $2800 and HDD is 1/10th of that.
Intel has plans to go to 5nm. The same will be true of GPUs. Once they reach that level, you're talking about computers that are 30-60x faster than what we have now so they'll be way beyond what most people need.
Originally Posted by Marvin
Intel has plans to go to 5nm. The same will be true of GPUs. Once they reach that level, you're talking about computers that are 30-60x faster than what we have now so they'll be way beyond what most people need.
That's true of today, and yet here we are. The constant shrinking is for the techies and businesses; the people who need that power. That everyone else can buy it is how it's made cheap, yeah?
Skymont won't be the end of the story, even for consumer stuff. I just wish they'd jump straight to it, but I've always wanted to be a time traveler…
I think at present there's still somewhere we can go to. Integrated graphics are still not great, mobile battery life isn't more than a day. Once we can get a certain realism to real-time graphics and take away any remaining bottlenecks for productivity software, I think people will be content. I think this will happen before we get all the way to 5nm too.
Take video games as an example, the upcoming consoles are supposed to be around 8x faster or more. The difference won't be all that noticeable because they still have the same storage limits. Processing power is really just one part of it and they are fixing all the other bottlenecks so the need will be left for raw processing but that can be off-loaded.
Skymont is 10nm, there would be 4 iterations after that - 2 on 7nm, 2 on 5nm. Beyond that, they'll probably struggle but overall power will be a lot by then. The CPU and GPU should work together as a single unit fairly soon.
Quote:
Originally Posted by Marvin
Skymont is 10nm, there would be 4 iterations after that - 2 on 7nm, 2 on 5nm. Beyond that, they'll probably struggle but overall power will be a lot by then. The CPU and GPU should work together as a single unit fairly soon.
They absorbed sound cards in the past. They absorbed the entire northbridge into the cpu package. This seems like a natural evolution. It's possible discrete options will hold out for a while depending on performance, but the cost would most likely be higher due to the lack of ability to sink development costs through volume sales.
I know AMD is rushing head first into heterogeneous computing and frankly seems to be dragging Intel along. We should be there in a few years. At that point hardware will begin to look dramatically different. Discrete GPUs will actually end up being the low performance choice for some workloads.
Quote:
Originally Posted by Tallest Skil
Also, I'd like to see a tachyonic hard drive. It'll have files on it before I've decided to download them, but we'll have to wait until the thiotimoline hybrid version comes out before it can cache things before I want to open them.
Except when three or more tachyon drives are mounted on the same Mac system, bad things can happen that threaten the future of humanity...
Quote:
Originally Posted by Winter
Although you have to believe that by 2015, 512 GB flash drives will be standard with an option for 1 TB or more.
I hope SSD sizes grow faster than that.
Okay 15" Retina MacBook Pro, possible scenario for when the prices drop slightly.
Higher end 15" currently sits at $2,799 and the only difference by default is a slightly faster processor and double the amount of flash storage. Perhaps either by default (doubtable) or as a CTO option, double the graphics memory. At the very least for an extra $100, the 650M should have 2 GB instead of 1 GB for the retina screen.
Thoughts?