My Dream Mac Laptop
The Macbook X series
Target audience: mobile professionals
- 13.1" screen, resolution of 1600 x 900 (40% greater than the current 13.3" Macbook Pro)
- Smaller footprint than the Macbook Pro 13". 12.4"(W) x 1.0 - 1.3"(H) x 8.3"(D)
- Intel Core i5-540M processor (2.53GHz) with Turbo Boost up to 3.06GHz
- 256 GB SSD
- 4 GB memory
- NVIDIA GeForce GT 330M GPU (1GB VRAM)
- Optical disk drive: CD/DVD read/write
- Scratch-resistant screen, chiclet-style keyboard, large multi-touch touchpad
- SD card slot, 3 USB ports, DMI connection output, Gigabit ethernet port, DC-in, VGA output
- Made of aluminum and carbon fiber. 3.2 lb, 1 inch thin
- Unibody construction: 5-7 hours of battery life
- Offered in black and silver
Price point: $2099. Optional upgrades: faster processor, more RAM, etc.
Now, I know some of you may think that it is unreasonable to fit so much into such a light package, but Sony offers similar stats in its Vaio Z series. I've been in the market for a fully-featured ultraportable, and the Air just doesn't cut it (what with it lacking power, storage, and an optical drive). This is the laptop that I want to buy.
Comments
3 USB ports
Physically impossible. Have you ever even seen the inside of a MacBook?
HDMI connection output, VGA output
Also physically impossible. Apple isn't stupid enough to use a port that was killed in 1999. Screw VGA in every single one of its holes.
This is the laptop that I want to buy.
So buy a Sony Vaio Z.
- Made of aluminum and carbon fiber. 3.2 lb, 1 inch thin
- Unibody construction: 5-7 hours of battery life
Which one do you want? ... unibody or a combo aluminum/carbon chassis?
Making it ALL out of Carbon would be lighter... but there's a REASON you won't see that.
The Aluminum unibody also acts as a heat-sink... something Carbon Fiber can't do. So I think you'll see them sticking with aluminum for a while (unless they can get the cost down enough on that "liquid metal" acquisition they recently made
After further thought... maybe a aluminum base, and CF for the lid/screen??... would look goofy though.
The optical drive wastes too much space inside. It's handy to have but I'd rather the machine was lighter and cheaper with more ports or dual hard drives.
$2000 is a bit steep for that machine too.
These considerations are also short-term steps so fall short of a dream-like design.
I think we're going to have to pay closer attention to what's going on in the mobile sector. They've made a dual-core 2GHZ ARM chip:
http://www.theinquirer.net/inquirer/...-2ghz-arm-chip
and the dual 1.6GHz ARM uses just 2W - basically double an Atom N450 performance at 1/3 the power usage. A 2.53GHz Core i5 uses 35W. The Core i5 is about 7.5 times faster than the N450.
This means that it would take maybe a 6-core 2GHz ARM Cortex A9 to rival a dual-core i5 but it could use as little as 8W of power.
It can ramp down in a phone context and it can be docked in a 15" touch display to use as a laptop. Very light, cheap with a high quality screen. No optical drives or multitude of ports, just light-peak and/or USB 3.
Obviously lots of factors to consider such as no x86 support yet and GPU mobile performance being an order of magnitude slower but x86 isn't totally necessary if enough apps are ported over and eventually the GPUs will catch up.
My dream device is to have a mobile device so powerful that it's used for everything and you just connect it to screens. It would be backed up and encrypted as well as have remote wipe if the device is stolen. Fairly inexpensive too and always connected to a high-speed network.
I'd like to see IPS displays used in the laptops. Apple's laptop screens are fairly poor quality even after the most recent improvements.
The optical drive wastes too much space inside. It's handy to have but I'd rather the machine was lighter and cheaper with more ports or dual hard drives.
$2000 is a bit steep for that machine too.
These considerations are also short-term steps so fall short of a dream-like design.
I think we're going to have to pay closer attention to what's going on in the mobile sector. They've made a dual-core 2GHZ ARM chip:
http://www.theinquirer.net/inquirer/...-2ghz-arm-chip
and the dual 1.6GHz ARM uses just 2W - basically double an Atom N450 performance at 1/3 the power usage. A 2.53GHz Core i5 uses 35W. The Core i5 is about 7.5 times faster than the N450.
This means that it would take maybe a 6-core 2GHz ARM Cortex A9 to rival a dual-core i5 but it could use as little as 8W of power.
It can ramp down in a phone context and it can be docked in a 15" touch display to use as a laptop. Very light, cheap with a high quality screen. No optical drives or multitude of ports, just light-peak and/or USB 3.
Obviously lots of factors to consider such as no x86 support yet and GPU mobile performance being an order of magnitude slower but x86 isn't totally necessary if enough apps are ported over and eventually the GPUs will catch up.
My dream device is to have a mobile device so powerful that it's used for everything and you just connect it to screens. It would be backed up and encrypted as well as have remote wipe if the device is stolen. Fairly inexpensive too and always connected to a high-speed network.
Good post.
I'd like to see IPS displays used in the laptops. Apple's laptop screens are fairly poor quality even after the most recent improvements.
The optical drive wastes too much space inside. It's handy to have but I'd rather the machine was lighter and cheaper with more ports or dual hard drives.
Exactly the age of the optical drive in a small laptop is quickly passing.
$2000 is a bit steep for that machine too.
A bit? How about way to expensive! Laptops should get cheaper over time as higher integration chips come out and especially after AMD offers competitive processors. Laptop prices are artificially high simple because Intel charges so much for its mobile processors.
These considerations are also short-term steps so fall short of a dream-like design.
I think we're going to have to pay closer attention to what's going on in the mobile sector. They've made a dual-core 2GHZ ARM chip:
If you are interested in a true laptop there is no sense in looking at ARM. It isn't object code compatible, isn't 64 bit, doesn't virtualize well and is relatively slow. Mayby in a sub netbook, but the Apple has an OS problem.
and the dual 1.6GHz ARM uses just 2W - basically double an Atom N450 performance at 1/3 the power usage. A 2.53GHz Core i5 uses 35W. The Core i5 is about 7.5 times faster than the N450.
The key here is that AMDs Bobcat based designs are extremely low power for i86 devices, but yet deliver most of the good points of that architecture. 9 or 18 watts for a dual core i86 with a respectable GPU is nothing to sneeze at.
This means that it would take maybe a 6-core 2GHz ARM Cortex A9 to rival a dual-core i5 but it could use as little as 8W of power.
If only the math where that simple. Core performance is still an issue.
It can ramp down in a phone context and it can be docked in a 15" touch display to use as a laptop. Very light, cheap with a high quality screen. No optical drives or multitude of ports, just light-peak and/or USB 3.
Your assuming ARM can effectively ramp up to current laptop chip performance. It can't right now. Further it isn't 64 bit. ARM is simply a different class processor.
Obviously lots of factors to consider such as no x86 support yet and GPU mobile performance being an order of magnitude slower but x86 isn't totally necessary if enough apps are ported over and eventually the GPUs will catch up.
Maybe maybe not. I suspect that the reason Apple did the iOS devices the way they did (app store) is that they didn't want any confusion at all with the i86 machines. At least in part, running multiple architectures, instruction sets or what ever has never actually worked out well for Apple. Going i86 cold turky gave the company a lot of creed. Walking away from that isn't all that smart. Besides as has been pointed out i86 processors are getting smaller and cooler. AMD is actually deleting legacy functionality from some of its coming chips.
My dream device is to have a mobile device so powerful that it's used for everything and you just connect it to screens. It would be backed up and encrypted as well as have remote wipe if the device is stolen. Fairly inexpensive too and always connected to a high-speed network.
This bit with remote whip has always bothered me. If somebody is after your info they won't allow that remote wipe to take place. Pretty simple to do.
I actually think Apple is being pretty smart keeping a divide between iOS devices and laptops. It gives them a lot of flexibility for the devices while the laptops are best of the breed for that market.
If you are interested in a true laptop there is no sense in looking at ARM. It isn't object code compatible, isn't 64 bit, doesn't virtualize well and is relatively slow. Mayby in a sub netbook, but the Apple has an OS problem.
That is true right now but thinking a few years down the line, touch computing could easily replace our mouse inputs altogether and the standard OS doesn't work well without one. People are using the iOS more and more to the point where iOS users match OS X users in number and that took just 3 years to do that. What happens in another 3-5 years?
You can virtualize using a server and run a VNC app on it. You can do that right now in fact. The game streaming services will take off and improve in quality too reducing the need for Windows for games and consoles are helping there too. Once we get ubiquitous 20-100 MBit networks, this stuff will be trivial and inexpensive.
Then you don't need local virtualization and app developers are more than willing to port apps to popular platforms anyway.
When Apple switched to Intel, the reason was performance-per-watt, not raw performance, not compatibility, not virtualization. ARM beats x86 processors in performance-per-watt. x86 may catch up but I think it's similar to the GPU issue. GPUs were designed to work a certain way as was ARM. There's simply no way x86 developments come close to putting a dual-core 2GHz CPU in a phone.
http://www.theinquirer.net/inquirer/...-2ghz-arm-chip
"Nufront claims to be the first company to integrate ARM's dual core Cortex A9 chip at 2GHz. Other parts of the SoC include a multi-core graphics chip, a 64-bit DDR2/3 memory controller, a 1080p "multi-format video engine", SATA2, USB2, Ethernet and a general I/O controller.
ARM's architecture does not natively support the x86 instruction set, meaning that a bog standard copy of Microsoft Windows will not run on ARM chips. Nufront said that it is working on getting the open source operating systems Ubuntu and Android to run on its SoC."
The key here is that AMDs Bobcat based designs are extremely low power for i86 devices, but yet deliver most of the good points of that architecture. 9 or 18 watts for a dual core i86 with a respectable GPU is nothing to sneeze at.
18W is for a 1.6GHz dual-core though and they are clocking down the GPU part.
Going i86 cold turky gave the company a lot of creed. Walking away from that isn't all that smart. Besides as has been pointed out i86 processors are getting smaller and cooler.
Again this would refer to the present time though. Good decisions are not always timeless as evidenced by Apple's previous failures now successful after being applied differently at a different time. Of course, it's possible that in 3-5 years time AMD/Intel will be able to make their chips run fast and consume as little as 2W of power. It may be the case that iOS gets ported to x86 but desktop apps need rewritten anyway so it won't matter which way it goes.
Regardless of the hardware format, the ultimate goal is a mobile device that covers everything you need to do.
So buy a Sony Vaio Z.
Um, yeah, except that there's more to a computer than its internal specs. The Sony has a tiny trackpad, is very expensive for what it delivers, and loads on a massive amount of bloatware on what I believe is a poorer operating system (Windows 7) than Snow Leopard.
That's the whole reason why I wish a Mac would create a higher end ultraportable than the Macbook Air - because it would have the specs of the Sony while retaining the excellent build quality and operating system of the mac. In the end, in the absence of this "Dream Laptop," I bought a Macbook Pro 15" and maxed out the RAM. It's awesome, but I do wish it were lighter.
That is true right now but thinking a few years down the line, touch computing could easily replace our mouse inputs altogether and the standard OS doesn't work well without one. People are using the iOS more and more to the point where iOS users match OS X users in number and that took just 3 years to do that. What happens in another 3-5 years?
Remember there is a vast difference between what people expect out of a laptop versus a cell phone. Besides cell phone users far out strip computer users in general. On top of that it is not impossible to merge Touch with Mac OS/X. Personally I don't think it is worth Apples time to do so.
You can virtualize using a server and run a VNC app on it. You can do that right now in fact. The game streaming services will take off and improve in quality too reducing the need for Windows for games and consoles are helping there too. Once we get ubiquitous 20-100 MBit networks, this stuff will be trivial and inexpensive.
I will believe it when I see it. At least in the good old USA all we have is crappy networks and high prices.
Then you don't need local virtualization and app developers are more than willing to port apps to popular platforms anyway.
I seriously doubt that we will see such pervasive networks in the next few decades. Not that the concept isn't a good idea for some things but it isn't by any measure a substitute for local virtualization. As long as we have apps that only target windows we will nedd local virtualization.
When Apple switched to Intel, the reason was performance-per-watt, not raw performance, not compatibility, not virtualization.
Well if you buy that explanation, it would explain your other points here.
ARM beats x86 processors in performance-per-watt. x86 may catch up but I think it's similar to the GPU issue. GPUs were designed to work a certain way as was ARM. There's simply no way x86 developments come close to putting a dual-core 2GHz CPU in a phone.
Aren't we off track here? What I was pointing out was the slim chance ARM would end up in an Apple notebook. ARM already has dual core solutions that could go into cell phones. Right now they aren't in IPhone because Apple is still putting everything together so to speak with the OS. Even then I suspect we would see dual core first in iPad or similar device.
"Nufront claims to be the first company to integrate ARM's dual core Cortex A9 chip at 2GHz. Other parts of the SoC include a multi-core graphics chip, a 64-bit DDR2/3 memory controller, a 1080p "multi-format video engine", SATA2, USB2, Ethernet and a general I/O controller.
I'm not to sure they are the first. People have claimed A9 at 28 nm. It doesn't matter though as the performance still isn't what one expects out of a laptop.
ARM's architecture does not natively support the x86 instruction set, meaning that a bog standard copy of Microsoft Windows will not run on ARM chips. Nufront said that it is working on getting the open source operating systems Ubuntu and Android to run on its SoC."
Linux wasn't a commercial success on netbooks so I don't expect that success will gappen here either.
18W is for a 1.6GHz dual-core though and they are clocking down the GPU part.
I still don't get the problem here, AMD comes out with an excellent product and all we get is comparisons with processors it isn't designed to compete with. Think about this as a processor, GPU, Memory interface, and an I/O bus all rolled into one. If anything this is the best thing going for low power mainstream computing devices.
Again this would refer to the present time though. Good decisions are not always timeless as evidenced by Apple's previous failures now successful after being applied differently at a different time. Of course, it's possible that in 3-5 years time AMD/Intel will be able to make their chips run fast and consume as little as 2W of power. It may be the case that iOS gets ported to x86 but desktop apps need rewritten anyway so it won't matter which way it goes.
What is your obsession here with iOS? I86 has its own OS, a very capable one at that, there is absolutely no need for for iOS on i86. Further the Mac GUI can be easily extended to support Touch interfaces.
Regardless of the hardware format, the ultimate goal is a mobile device that covers everything you need to do.
That likely never happen. Your stationary computer will just become a much more powerful device.
Remember there is a vast difference between what people expect out of a laptop versus a cell phone.
They do right now but the performance will keep going up. If people are eventually buying dual 2GHz phones with 64GB+ storage, they don't need to buy a netbook, just a screen dock for their phone.
What I was pointing out was the slim chance ARM would end up in an Apple notebook.
Let's say we are 5 years down the line and Intel/AMD are on their 16nm chips readying the move to 11nm:
http://www.xbitlabs.com/news/cpu/dis...y_Roadmap.html
will they at that point be able to rival ARM for power consumption given where ARM is right now? Late 2012 will see a quad 1.2GHz ARM chip. Intel's latest Atoms (E600) are down to under 4W with integrated graphics but still just a single core 1.6GHz, although they are hyper-threaded with VT-x:
http://ark.intel.com/ProductCollecti...x?series=52490
They won't be using these in phones because that's what Moorestown is for. Here is Moorestown running World of Warcraft:
http://www.youtube.com/watch?v=HWi_Iyy-RkY
The Quake 3 demo looks fast but we'll see what it's capable of in 2011 vs the latest ARM and PowerVR chips. If they can top ARM, there's no reason to avoid switching iOS to x86.
I still don't get the problem here, AMD comes out with an excellent product and all we get is comparisons with processors it isn't designed to compete with. Think about this as a processor, GPU, Memory interface, and an I/O bus all rolled into one. If anything this is the best thing going for low power mainstream computing devices.
But which of Apple's products would it go into? They don't make a middle platform between a phone and a laptop other than the iPad and 9-18W is too much for an iPad. There isn't a single product Apple can use Bobcat in. The laptops will have to get Llano if anything, the graphics performance alone would dictate that.
They do right now but the performance will keep going up. If people are eventually buying dual 2GHz phones with 64GB+ storage, they don't need to buy a netbook, just a screen dock for their phone.
As a chronic iPhone user, I'd have to say the problem is very much the issue of usability of the small screen and the lack of a keyboard for touch typing right now. There is little difference here between docking to a screen and docking to a laptop. Besides 64+GB of storage wasn't good enough five ears ago, it won't be good enough in the future. And NO the cloud won't solve that problem.
Let's say we are 5 years down the line and Intel/AMD are on their 16nm chips readying the move to 11nm:
OK
will they at that point be able to rival ARM for power consumption given where ARM is right now? Late 2012 will see a quad 1.2GHz ARM chip. Intel's latest Atoms (E600) are down to under 4W with integrated graphics but still just a single core 1.6GHz, although they are hyper-threaded with VT-x:
They will never be able to rival ARM because ARM is uses less power on older process generations. You mis the whole point, all of the industry isn't going to switch over to ARM and throw away current investments in i86. Not for laptops anyways.
Even at that as each ARM core becomes more capable of competing with i86 processors it becomes more like those processors and thus more power hungry.
They won't be using these in phones because that's what Moorestown is for. Here is Moorestown running World of Warcraft:
The Quake 3 demo looks fast but we'll see what it's capable of in 2011 vs the latest ARM and PowerVR chips. If they can top ARM, there's no reason to avoid switching iOS to x86.
This boggles my mind who is talking about switching iOS to i86? Someone started this with the idea that Apple would switch over to ARM in their laptops. I simply can't see them doing that anytime soon.
As to this link between iOS and the ARM family that is directly related to the devices Apple wanted to build which required low power chips. Since at its heart iOS is still the same basic UNIX system that is Mac OS apple could port the OS to just about any architecture it wanted. This idea that iOS has a special link to ARM is nonsense, ARM was one of only a couple of possible contenders at the time Apple decided to get into the cell phone business.
But which of Apple's products would it go into? They don't make a middle platform between a phone and a laptop other than the iPad and 9-18W is too much for an iPad. There isn't a single product Apple can use Bobcat in. The laptops will have to get Llano if anything, the graphics performance alone would dictate that.
AIR for one as it looks like an actual upgrade to the current machine hardware. There is nothing to stop Apple from making a netbook class machine nor a tablet Mac for that matter. Or something completely new to add to the line up. People seem to forget that iPad was completely new to Apples line up only five months ago. The lack of imagination on the forums is sometimes stark.
People will probably laugh at the idea of a Apple netbook, but AIR is nothing more than a very expensive netbook. That is a light weight machine with limit capability and short battery life. The smart thing for Apple to do is to come out with an AIR revamp that makes the price reasonable.
There are probably hoots and hollers about a Mac Tablet too. A Bobcat based Fusion product would make such a product feasible. For many people wanting the power to Mac OS in a tablet such a device would be far more usable than the iPad. Not that iPad is bad, just that it is far from flexible and versatile like Mac OS is.
On top of that the Zacate is very suitable for a Mac Book class laptop, especially if it extends battery life significantly. You may say 1.6 GHz is a little slow which would be true for some uses but for many users that speed isn't as big an issue as battery life. Also it should be noted that AMD hasn't really said what the top speed of Zacate is, rather they have demoed a 1.6GHz variant. Jumping out of the gate with a 1.6 and a 2GHz Zacates would lead to many interesting design possibilities. Make a Mac book thin without the Optical drive and some other power saving moves (dropping Firewire) would give us a Mac Book with somewhere in the neighborhood of 11 to 12 hours of battery life. Many users would flock to such a device, because it cuts the cord making the laptop truly portable for the entire day.
The next argument would be that Apple could never sell such a thing. Certainly not as a Mac Book Pro and maybe not even as a top of the line Mac Book. As a long running competitor to the netbooks and other platforms a mac Book LT, lets call it, would be very marketable. As long as the buyers understood what they are getting marketing would be easy. The target markets would be students, sales reps, and other people frequently away from power sources. A market similar to iPads but requiring the keyboard of a laptop.
Dave