The Apple iPad - Anand's Analysis
by Anand Lal Shimpi on January 27, 2010 5:00 PM EST- Posted in
- Smartphones
- Mobile
About a decade ago Mike Andrawes and I kept hoping someone would come out with a device that would make surfing the web on the couch easier than it was. Mike took the notebook route. He kept buying (or stealing from me) notebooks that were cooler, thinner and lighter while still being a notebook, for the purpose of browsing the web.
I took a more extreme route. I tried ultraportables. I bought a Sony PictureBook. It had a Transmeta Crusoe processor in it, which was horribly slow but gave me the form factor and battery life I craved. I put up with a ridiculously impractical screen just to get something small to browse the web and do work on.
Matthew Witheiler, another AnandTech veteran took a different path. He embraced the tablet PC. Matt became our Tablet PC reviewer on AnandTech as he searched for the perfect device. Unfortunately, he never found it.
None of us did. Today we all went back to the tried and true device: the notebook. The iPhone came along and gave us a revolution in the smartphone space. Ultimately it and the devices that followed just complemented our notebooks - sometimes with a new level of frustration as we were now at the mercy of wireless carriers and ridiculously slow SoCs.
The smartphone revolution gave us some great devices
History likes to repeat itself, and that’s what we’ve seen happen over the past two years. The introduction of the netbook brought the journey full circle. People wanted a cheap, light, portable web surfing and light work device - the netbook did just that.
The keyboard and screen issues have been mostly solved. Performance still sucks and part of that is due to the fact that there are no good netbook OSes that are optimized for the level of performance a 1.6GHz Atom can deliver. Most OEMs ship some variant of Windows on these devices, and with less than 2GB of memory and a single-core in-order CPU, that’s just too much to be fast.
Back to ultraportables again
There’s also the issue of storage. Netbooks desperately need solid state storage, but a single 2.5” SSD is often over half the price of a netbook itself. Pair up a slow CPU with not enough memory and a really slow hard drive and it’s not a good combination.
Microsoft, Intel and Apple have all taught me one very important lesson over the past 13 years: if you’re going after a new usage model, you need new technology to tackle it. For Microsoft and Apple that meant a new UI with Media Center and the iPhone. For Intel it meant a brand new microarchitecture optimized for power efficiency. First with Banias (Pentium M/Centrino) and then with Atom.
Netbooks, and to a greater extent tablets, eReaders and smartbooks, are going after new usage models. These aren’t notebook replacements, they are a new category of device designed for a different usage model. The one thing they’ve all been missing is the perfect combination of hardware and software to deliver the whole package.
The one thing Apple prides itself on is doing just that. As one of very few one-stop hardware/software makers, it has the ability to tightly couple UI with physical design. We saw it manifest in its greatest way with the iPhone, and now Apple (or perhaps the media covering Apple) is attempting to recreate the magic with the iPad.
The final frontier?
The device doesn’t ship for another 60 days, but there’s a lot to talk about based on today’s introduction alone.
Facebook
Twitter
RSS
155 Comments
View All Comments
muretu - Wednesday, January 27, 2010 - link
Great article. I do think that iPad is a great alternative somewhere between MacBooks and the iPhone. Apple have defiantly coupled a good build and a solid O.S. together, again.Funny how it hardly has been noticed that Microsoft have been doing this for years, re-releasing new touch PC's.
The iPhone was an innovation (with the multi touch glass display). The iPad is not exactly necessary at it's current stage. I'm sure the MacBook's are fine for now, if you thought iPad was a computer replacement.
Would be interesting to see a fully operational Mac OS X Leopard running of a tablet.
But it's Apple's great O.S'. and price range which has sold their products from the past. :)
hwhacker - Wednesday, January 27, 2010 - link
Price helps sell their products? Wut?jimhsu - Wednesday, January 27, 2010 - link
Conspicuous consumption (aka the Veblen goods theory). The fact that apple has higher prices obviously suggests to the (usually uneducated) consumer that the goods in question are of a higher quality. It's worked fantastically for the LVHM conglomerate, at least.The new macbooks (in relation to the Rev A MBP) are of a more reasonable price, though.
Sahrin - Wednesday, January 27, 2010 - link
I'm not the biggest fan in the world of Apple. A lot of the hardware community (starting with AT and Ars) fawns all over them, which is something I simply can't understand. No Hardware company ever got this kind of affection before the "Apple reinessaince" of the last 3 years. That's why I don't like Apple. They have good hardware - sometimes. They have good software - sometimes. The only thing they do that truly differentiates themselves from their competitors all amounts to showmanship. Instead of putting a device into beta testers hands a year in advance (a la Microsoft or the console makers) Apple announces it in such a way as to manufacture the most 'buzz' (and everyone falls for it); it launches with almost zero software support (meaning it is useless) - but everyone loves it anyway. The reason usually given for the love? "The sparse environment is so modern and aesthtetically pleasing."That said, I actually was pretty stoked about the iSlate (which was what at the time I assumed it would be called). It is EXACTLY the kind of product I could see myself getting. I am not a big phone guy, I like connectivity but talking on the phone is a single-thread activity; I work in multiple instances in multiple dimensions. I really love the concept of the tablet - and have ever since Microsoft really started pushing the format.
(That's my second point, Anand. Twice in this article you cite Apple as the 'inventory' or 'first person in' to a market - you're absolutely wrong on both counts. It seems like every serious hardware guy who goes over to Apple becomes a fanboy - totally ignorant of the facts, and re-writes history to fit their vision of the world. The same happened to Jon Stokes. What's the deal?)
So, as I said, I was really, really excited for the iSlate. I don't like PMP's because the video is unworkable - and if you just want an MP3 player what's the point of the screen anyways? The most exciting parts of the ZuneHD are HD Radio and Zune Pass and portable gaming. Video isn't even on the list.
But an iSlate-like device could really change the way people use technology. I was a big fan of Earth: Final Conflict and I've always felt that the "Global" device used in that show was the future of communications. A portable screen which gives person a thin but robust computing appliance and handles all personal tech interface. You still have a desktop for serious work, but most of the utility stuff ironically gets moved to the iSlate.
So imagine my surprise when, as a non-Apple fanatic, I found myself *VERY* excited for the iSlate.
And Apple delivered the iPad.
This is *not* the device I was told to expect. The only thing it really has in common with the device I want is 3G and touch screen.
DO NOT WANT:
iPhone OS.
Apple Content Lock In.
iBook store (see above)
Lack of Flash/Silverlight support
Light ARM processor
$800 device
DO WANT:
Multi-core full-feature x86 CPU. Will settle for Tegra2 SoC.
Video acceleration
Full Desk Client OS. Preferrably Win7, so I can run all the software I want. Settle for OSX or Ubuntu in a pinch.
Same feature set of 64GB/3G for $500
I was actually very surprised to learn about the initial price point, until I heard about the 3G. It's not useless without 3G, but without 3G it isn't the same device. It's another laptop on the couch, as Anand put it so eloquently.
The Inside:
If you look at the specs, there's actually only one company that's really positioned at all to deliver the device that I'm talking about in the next 5 years. It's name starts with an A. And ends with an MD.
AMD can make their fortunes here. Designing a light SoC and tap Qualcomm/etal for radio support. Ideally a Bulldozer based solution which would support hardware backed SMT (cut down on core count) tied with 5000-series-class DX11 acceleration. Runs Win7. Access to all the content the web has to offer - which is 1 billion times better than iTunes or iBook store.
Please make this device. Apple has shown you it can be done at the right price point. Think about that last sentence, Dell/HP/MS/AMD/Intel. APPLE has shown you how something can be AFFORDABLE!
DO IT!
That said, I am still very interested in the iTab. It looks to be a great device, especially if they get the mutli-tasking problem licked in iPhoneOS4 and add in Flash/Silverlight.
Yay technology!
jasperjones - Wednesday, January 27, 2010 - link
+1Particularly, Apple vendor lock-in really scares me.
You want an Office app? Shell out $30 for iWork. I doubt Apple will allow anything akin to an OpenOffice port in the AppStore for the iPad.
jimhsu - Wednesday, January 27, 2010 - link
Agreed. I question whether your request (x86 modern CPU, "discrete" graphics, 64GB of flash, touchscreen, 3G modem, long battery life) is doable for $500 though. Such a device will likely be $800, and hype/demand/shipping/tax will probably drive that closer to $1000.Sahrin - Wednesday, January 27, 2010 - link
I agree that the x86 would push the cost up a bit, but the other side of it is that the manufacturing cost for x86 will be lower than it will be for ARM (because the 2 x86 manufacturers are ahead of any foundry). It will take some agressive work, but I think this is exactly the kind of thing bulldozer would be great at.The alternative is, of course, to port Win7 to ARM. MS needs to get on this right away. It would kill AMD, of course, (like overnight), but it would also really rejuvinate the PC hardware market. HP and Dell have been terrible stewards - all the innovation comes out of Taiwan.
If you can make a T&L laptop with Core2Duo for $600, then I really believe you can make a tab for about the same amount of money (smaller screen (always the most expensive part of a portable), fewer raw materials, no drive, smaller board, simpler board (no peripherial traces because it would have to be a SoC).
I agree that the price is probably north of $500, but I think it's less than $700 (maybe 699).
The main reason I ask for x86 is because I can take my Win7 desktop applications and (With a full x86 multicore) run them on the tablet piece of cake.
I would pay $1,000 for this because it would be useful to me, but cost is probably about $699.
MS could really hit a home run here, too, if they did some work and tied all the marketplaces (Zune, WinMo, XBL) together and used the Tablet as the "ambassador" device. The *BEST* part is that Apple took the STUPIDEST name for a tablet, meaning we can get something like "Slate" or "Canvas."
jimhsu - Thursday, January 28, 2010 - link
Our of those two possibilities, I would say x86 is the most likely one. Porting windows to ARM would be ... nasty. That said application compatibility with (real) OS X/Windows is sorely needed in this formfactor.The reason I quoted the given price is here: http://www.isuppli.com/News/Pages/iPhone-3G-S-Carr...">http://www.isuppli.com/News/Pages/iPhon...ing-Cost...
THE most expensive component in the iPhone is the screen+touch interface. The touch module is in fact almost the same price as the screen.
Sahrin - Thursday, January 28, 2010 - link
I'd say the change from x86 to ARM won't be as bad as you think - after all, they ported NT to IA-64 (and I don't have anything but a vague sense to back this up) which I think it a much bigger change than x86 to ARM. Totally different programming models, whereas x86 and ARM are fundamentally the same in the way they think about processing data, just different in the way the way they handle specific instructions.I agree with you, but the Bill of Materials you linked just enhances my point, I think. There is nothing the Slate needs that the iPhone has - so basically you scale the cost of the display 400%. Your BoM also doesn't include the cost of the battery (at least I don't see it).
By basically doubling the price of all components but RAM (x8 to get 2GB), Display (x4 for combined touch+display, which is probably a little strong), CPU ($50) and leaving the radios all the same (don't need a different radio) you get $397. Add in a baterry for...I don't know $50 you get $450. At $500 that's a gross margin of 10% (obviously nowhere near good enough, it needs to be way more than that - 40-50%), which gets you to $625 (including battery at 40% profit margin). Sell it for a handsom $74 profit per unit at $699, and you're doing quite well.
It's worth pointing out that my projections are significantly strong in hardware (I believe), but that puts it in my target range. $649 would obviously be better, and I really believe $599 is the sweet spot for such a device...but. Yeah.
I'm *really* excited for tablets, and if MS/PC OEM's don't have something by next winter I may just have to partake (especially if there is OSX or a much more fully-featured iOS4 available).
We'll see.
Ananke - Wednesday, January 27, 2010 - link
I believe the company that will make the most successful tablet device is Asus, followed by MSI, Acer and Lenovo. It is just a matter of time, and understanding to use attractive and durable case. Use metal or fiber, not cheap plastic; use Android or Win 7 with x.264 decoding, SDXC, USBs and HDMI out. Very simple mix to create the ultimate product. Just please do it :).