Largest Computer Makers When Smartphones Included - Apple dominates, nearly twice as big as nearest rival (which is, of course, Samsung not HP)

So its time to do the update for the 2011 final numbers of computer sales, when all types of computing devices are included. So I mean when we count traditional mainframes and servers, the PCs we know and love from desktops to laptops and netbooks, onto tablet PCs like the iPad, and the smartphones, as well as that lingering PDA market which is still typified by the iPod Touch. (We do not add basic 'dumbphones' nor gaming platforms or DVD players etc as these are not considered to be true computers, even as in many cases of featurephones, we could install some apps and games). So this is the new normal, when all devices that are fairly considered computers - devices that users can reprogram, in other words, that the devices have their own OS and users can install apps.

By this definition - and please readers do remember, all major PC makers have already accepted that smartphones are indeed computers, so don't bother to argue about that oldfashioned view that a smartphone would not be a proper computer, we do reach dizzying heights - the total computer market hit 950 million units sold last year. The computer market as thus defined, grew a massive 47% in just one year from 646 million units in 2010, and this growth was all driven by huge growth in smartphones and tablets, the legacy PC market was stagnant. And also just so you know, the 60 million unit tablet-market is projected by many analysts to pass 100 million in size this year, so yes, tablets are roughly one seventh the size of smartphones - yes, the smartphone market alone is 7 times bigger than the tablet market..

So lets look at this computer industry as thus defined. It splits up this way in the year 2011: 50% of all computers sold were smartphones, 37% traditional PCs (of which now a majority are laptops not desktops), 6% were tablet PCs like the iPad, 5% were PDAs like the iPod Touch, and 1% were servers. And yes, obviously four out of five computers sold today is a portable computer. No wonder Apple calls itself a mobile company (and congrats to Apple BTW for becoming most valuable company of all time and also generating the biggest profits - that was all after Apple decided to shift from the legacy PC market to the mobile market). So how does the 2011 global computer market look? Can you spell Apple? Look at this:

LARGEST COMPUTER MAKERS WHEN SMARTPHONES & TABLETS INCLUDED
Rank (was) . . Brand . . . . . . Units 2011 . . . Market Share 2011
1 (1) . . . . . . . Apple . . . . . . 195.5 M . . . . . 21%
2 (8) . . . . . . . Samsung . . . 104.9 M . . . . . 11%
3 (2) . . . . . . . Nokia  . . . . . . 77.3 M . . . . . . 8%
4 (3) . . . . . . . HP  . . . . . . . . 64.5 M . . . . . . 7%
5 (4) . . . . . . . RIM . . . . . . . . 54.5 M . . . . . . 6%
6 (7) . . . . . . . Lenovo . . . . . . 49.9 M . . . . . . 5%
7 (6) . . . . . . . Dell . . . . . . . . 46.6 M . . . . . . 5%
8 (9) . . . . . . . HTC . . . . . . . . 44.6 M . . . . . . 5%
9 (5) . . . . . . . Acer . . . . . . . . 41.3 M . . . . . . 4%
10 (-) . . . . . . Sony . . . . . . . . 31.8 M . . . . . . 3%
. . . . . . . . . . Others . . . . . . 240.1 M . . . . . 25%
TOTAL . . . . . . . . . . . . . . . . 950.0 M
Source: TomiAhonen Consulting from company and industry data, August 2012
This data and this table may be freely shared

So there you have it. Apple has grown its lead massively from last year when it took the number 1 position from Nokia. Samsung also has grown from 8th ranking to 2nd. And for the first time the Top 3 are all by manufacturers whose primary product on this chart is.. a smartphone. In fact every 'traditional' PC maker has fallen down in the rankings.

And what of HP? Just tumbling down on the rankings? Once Hewlett-Packard was clearly the biggest computer maker, not anymore. And sadly, HP had the keys to this future too, when it bought Palm but then with the new CEO carousel, they lost track of their future and today, HP falling down yet another notch on this list. And as a new service, lets also do a rough estimate of the market shares by operating system:

LARGEST OPERATING SYSTEMS OF COMPUTERS WHEN SMARTPHONES & TABLETS INCLUDED
Rank . . Brand . . . . . . Units 2011 . . . Market Share 2011
1 . . . . . Windows . . . . 291 M . . . . . . 31%
2 . . . . . Android . . . . . 218 M . . . . . . 23%
3 . . . . . iOS . . . . . . . . 196 M . . . . . . 21%
4 . . . . . Symbian . . . . . 81 M . . . . . . . 9%
5 . . . . . Blackberry  . . . 55 M . . . . . . . 6%
. . . . . . Others . . . . . . 110 M . . . . . . 12%
TOTAL . . . . . . . . . . . . 950 M
Source: TomiAhonen Consulting from company and industry data, August 2012
This data and this table may be freely shared

And yes, you can see the catastrophy that faces Microsoft, who once had over 80% of the computer market and also had 12% market share in smartphones with Windows Mobile. Microsoft could have been well poised to survive this transition into the newest computing era if it had nurtured its broad coalition to provide Windows based smartphones (which included Samsung, Sony, Motorola, LG, Dell, Lenovo etc and in 2011, Nokia too). But now on the smartphone side, Windows Phone is down to 3% market share (and falling, with Nokia the partner utterly failing with this OS) so Microsoft is stuck watching the diminishing share of the legacy computer market, where Android and iOS divvy-up the future markets of smartphones and tablets.. No wonder Microsoft threw Nokia under the bus. (And no, Windows Phone 8 will not help in smartphones, Windows 8 will be able to extend Microsoft's life in that diminishing slice which is the traditional PC market, but it will be Android and iOS who rule smartphones and tablets, and next year, Windows Phone 8 will be happy to hit the 3% market share it held in smartphones this past Q2). I have already said, that we will soon see Android outselling all Windows devices, the trend is unstoppable, inevitable.

If you arrived to this blog from the PC industry and didn't know the smartphone side is already bigger, here is the breakdown of just the smartphone sector, for the full year 2011 market shares.

About this blog, please don't bother to write if you are one of those luddites who still thinks like a caveman that the smartphone is for some reason 'not a proper computer'. We had those arguments years ago, go back to the cave you came from, I will delete all comments now in 2012 who waste the time of my readers attempting to argue that a smartphone is not a computer. We passed the point when each of the biggest PC makers, Dell, HP, Lenovo, Acer, Asus, Toshiba and Apple - have all said in public that yes, a smartphone is a computer. Don't bring that tired old argument to this blog, I won't have it anymore.

But, as I have been reporting on this phenomenon, if you want to see last year's chart, it is here. And honestly, as the big computer makers have all already agreed that a smartphone is a computer, then why can't we have some of the big analyst houses like Gartner, IDC etc who report on that data individually, give us also the comprehensive count for this whole industry? Isn't it about time?

You may freely share this info and write about it and use the chart if you want.

One plug, for those who are interested in deeper numbers on the handset industry, not just smartphones but also the 'dumbphones' parts, including market shares, operating systems, average prices, feature sets, installed bases etc etc etc, please look at the TomiAhonen Phone Book, the 2012 edition is almost ready, so the special offer to get two for one, is about to end, hurry to get your set now. See full table of contents and more info at TomiAhonen Phone Book 2012.

Comments

[John Waclawsky]

Why tear down Linux with nonsense arguments? We obviously have a Microsoft shill (Baron95 the astroturfer) continually telling us how superior windows is, now with drivers.

Using Occam's razor, the root cause of Nokia's predicament is NO ONE wants to buy a windows phone. This windows only strategy will be the death of Nokia. It is very simple. If no one buys your product you go out of business!

[URNumber6]

I don't dispute smartphones are computers but I'm far from convinced iPhone and WPx devices are truly smartphones - they are far too locked down and there's far too much missing functionality.

As another poster already pointed out even washing machines have computers in them these days but beyond helping doing the washing it's of no use to its owner.

Likewise iOS and WPx, the owner is very limited in what use their 'computer' is to them as they only have access to a little pick-and-mix of apps and content that their mummy and daddy (mummy M$ and daddy Apple) has deemed appropriate for them.

I agree with the point Huber made, I would only view Android/Symbian/MeeGo devices as smartphones/computers as they are the devices I can get to carry out my commands without requiring mummy and daddies approval.

[bjarneh]

Baron95> In consumer laptops, phones, etc with fast evolving hardware, interfaces, attachments, etc it has always and will always fall flat compared to fast movers like Apple (integrated), Android and Windows (designed for easy porting and drivers).

fast evolving interfaces and attachments? what are you on? Linux cannot move as fast as Android? Android is Linux, you can run Ubuntu on Android phones now by sharing the same Linux kernel.

Baron95> There is a reason Maemo, Meego, Maltemi have failed, Bada is failing and Tizen will fail.

yes, there certainly is, it's one of Tomi's main topics in this blog. it's hard for Meego or Meltemi to succeed when they are basically cancelled before they are ever launched. Linux is by no means a beaten OS, if Android keeps going at this pace, Linux will be the dominant OS by any measure in a short period of time.

Baron95> Linux, its processes to support new devices and GPL all suck in fast moving consumer markets.

yes, look at Windows and Apple's superb hardware support on smart-phones...

Baron95> Tomi can write 1,000,000 words on Bada and Meego. It won't happen.

that maybe true, but this is in no way connected to your "arguments". you basically claim that Linux (Android) has no chance at competing against Android, not sure where to begin here...

[John Phamlore]

Tomi,
A technical question about smartphone technology: Just how dominant is Qualcomm becoming in integrating LTE with previous generation fallback support to provide multimode chipsets, and is there actually some prospect of Qualcomm some year releasing a true worldwide multimode chipset that can support almost all regions frequencies, and perhaps even the Chinese TD-LTE?

I'm just catching up on the news, but every article I read seems to have another twist. For example, seemingly months ago Samsung was allied with DoCoMo and others in an LTE alliance, then that alliance was dissolve, and now apparently some of the previous alliance but not Samsung are in alliance again.
http://www.slashgear.com/fujitsu-nec-and-docomo-take-on-qualcomm-with-new-3g4g-chip-company-01241190/
Is the above going to basically be a Japanese alliance to hold off Qualcomm mostly in Japan, or are they going to challenge Qualcomm worldwide?

And what are the other major regions going to do about Qualcomm? What are the Chinese plans? Does Europe have its own consortium plans?

[John Phamlore]

One baseband chipset maker to rule them all?

Apparently using Windows Phone 8 as the OS still requires using Qualcomm SoCs. As the Ativ S appears to first be a European phone, it's been reported to be using a Qualcomm chipset that does not need to support LTE. If Qualcomm's chipsets are now supporting TD-SCDMA, they will support the backward compatibility fallbacks for the United States, Europe, and China at least. With that base established, Qualcomm can use each die shrink to expand the number of LTE frequency bands it supports.

Nokia had the complete IP stack for the previous generation, but it does not appear to have the technology for the future, and after the settlement of the Qualcomm lawsuit, there was little indication that Nokia could have even offered the full backwards compatibility that Qualcomm now can. In this area of technology, whoever owns the past can own the future.

[Huber]

@Duke: I am no 'Microsoft astroturfer', and I don't become one when I point out that TVs are not 'real' computers despite the fact that they use basically the same technology.

I thought this is a place to discuss different points of view, not a place where you are instantly labeled depending on who you answer to.

[harry113]

@Baron95

Baron95> There is a reason Maemo, Meego, Maltemi have failed, Bada is failing and Tizen will fail.

Well, Im pretty sure that Tizen and Meego combined will be in 2013 within 3% of Windows Phone. (worldwide marketshare). Lets see and wait :)

[Fa`z]

Question regarding skew due to Infrastructure-as-a-service:
Lots of companies and individuals used to buy their own servers, but now pay for them as a service (e.g. AWS). There used to be a very low computer power utilisation as a consequence of them buying their own computers. Is part of the reason why server/desktop computers are not showing as strong due to the change in customers now paying for computing as a service rather than building own their own servers? If a server now runs 1000 virtual servers, and then sale of servers would show a big drop. Mobile phones are not divisible into virtual instances due to their personal device nature.

[Przemysław Lib]

@Baron95
It seams that we need DumbTV and SmartTV and DumbDVD and SmartDVD, DumbConsole, SmartConsole.

That is realization that *EVERY* device run by computers can be equipped with hardware that made SmartPhones possible.

So we will see more and more device categories on the "COMPUTER" side of split, but for time being there also will be "Dumb" versions too.

@Walt French
Apart from, multitasking, networking, file systems, memory managements, power savings, etc.
Even Linus Torvalds points out that focus on mainframes got head start for Linux in mobile race!

UI may be different, but underlying *infrastructure* is the same!

@Baron96
Both AMD and Intel OFFICIAL opinions are that writing opensource Linux GPU drivers can be done on the same speed as proprietary one! That is Intel have gpu drivers that are released simultaneously with proprietary, and have defined timeline for reaching feature parity (on mobile side it will be much faster due to better starting point).
AMD engineers claims that they are working on 8xxx hardware enablement from the beginning of 8xxx life cycle. While feature parity will take more time, it is doable, and AMD open source team works faster than AMD develops new GPUs.

All above prove that if anyone want to write Open Source GPU driver (GPL for kernel driver, MIT for userland MESA), it is doable.

Also it means that Intel will be first to release mobile GPU with official OpenSource driver!

PS Bada already have same market share as WinP7. LOL. It already happened. And note that Bada OS have low ambitions, and only for low end smartphones, which preclude Word Domination.

[John Waclawsky]

Are you guys serious responding to Baron95. He a Microsoft astroturfer. Don't waste you time with his nonsense. There are lots of links like:
http://www.cbsnews.com/8301-505124_162-43340158/astroturfing-antitrust-how-microsoft-is-crafting-the-grassroots-case-against-google/

http://techrights.org/wiki/index.php/AstroTurfing

[John Phamlore]

A (partial) timeline of Nokia sleeping while others acquired the IP for next generation smartphones. Observe many of the events below occur before Elop was hired by Nokia.

June 10, 2005: Nokia and Intel announce alliance on mobile WiMAX

August 11, 2005: Qualcomm buys Flarion
http://gigaom.com/2005/08/11/qualcomm-to-acquire-flarion-for-600-million/

November 14, 2007: Qualcomm announces delivery of QSD8250(TM) and QSD8650(TM) chipset products in Snapdragon platform
http://www.qualcomm.com/media/releases/2007/11/14/qualcomm-premieres-snapdragon-first-chipset-solutions-break-gigahertz-barr

April 23, 2008: Apple buys PA Semi

July 24, 2008: Nokia and Qualcomm settle their lawsuits

October 17, 2008: Nokia reveals it will make a $2.3 billion USD payment to Qualcomm as part of the earlier settlement

January 8, 2009: Nokia abandons N810 WiMAX tablet

January 20, 2009: Qualcomm buys AMD mobile graphics
http://www.qualcomm.com/media/releases/2009/01/20/qualcomm-acquires-handheld-graphics-and-multimedia-assets-amd

February 17, 2009: Nokia outsources 3G baseband to Broadcom
http://www.broadcom.com/press/release.php?id=s365393

April 27, 2009: Qualcomm and Broadcom settle their lawsuits, Qualcomm agreeing to pay $891 million over 4 years.

August 30, 2010: Intel buys Infineon's wireless unit

September 10, 2010: Nokia hires Elop

May 9, 2011: Nvidia buys Icera

August 15, 2011: Google buys Motorola Mobility

October 2011: Apple iPhone 4S revealed to have replaced Infineon chips with Qualcomm's
http://online.wsj.com/article/SB100014240529702046187045766413922809822

July 18, 2012: Intel buys assorted InterDigital wireless patents

[bjarneh]

Baron95

You are referring once again to the "slow moving server side" where the open source platform such as Linux has a chance to compete vs. the fast moving mobile + desktop world where it does not stand a chance.

This is a myth, nothing more; as a programmer I find it hard to comprehend that you think we are willing to waste our time writing drivers etc. for hardware we will never own. Why do you think Linus Torvalds only supported 1 type of hard-disk in his first release?

Hardware for modern servers etc. is insanely expensive, do you really believe that programmers go out and buy modern servers $50.000 just to see if they can get them working with their favourite operating system?

The reason Linux has succeeded on the server side is that firms understood that it's actually cheaper to help build/maintain/run this operating system, than other proprietary alternatives. There are companies helping out on the server side just as much as Google is helping Linux in the mobile space, server side Linux is by no means a non-profit.

You cannot call failed version of Linux for Linux, and successful ones something else.

[Rahul]

i am completely satisfy with your points and really apple is now in good demand as compared to apple so its nice content!

[Janne Särkelä]

@Baron95

Linux kernel has shown remarkable success in the consumer electronics market because of its scalability. You are comparing Linux to the other players as if it was a competitor. It's not. Linux is a tool and a platform.

Google made the obvious choice and decided to go for the flexibility and adapt many of characteristics of the open source to spread their fork of the Linux kernel. It's because of that Android is so successful. Besides I read that Samsung is now a platinum member of the Linux foundation. http://www.linuxfoundation.org/news-media/announcements/2012/06/samsung-electronics-joins-linux-foundation-platinum-level

That's only natural, because Samsung uses Linux and its derivatives in great part of their modern products. By being a platinum member they gain access to the Board of Directors - so it's safe to say that for Samsung Linux is in the core of their business model.

Completely other point is to compare the success of different Linux based systems and projects.

[bjarneh]

Baron95> You are confusing competitive first world consumer electronics market with hobby.

Nokia certainly seems like a pretty expensive hobby with 120.000 employees..

Linux has not succeeded in the server market because "the server side moves slow enough" for the sluggish free software makers to keep up. There are huge financial interests which keep up with proprietary software makers. IBM, Oracle/Sun, RedHat are not involved in some elaborate "hobby".

What IBM, Oracle, Google etc. did for GNU/Linux on the server side, Google/Samsung/Nokia has been doing for the mobile consumer market, there is no technical argument for why this has to fail. You claim that it moves to quickly with some vague terms like interfaces and attachments, but these are hardly arguments to support your view.

Baron95> .. that open collaborative development amounts to anything in the consumer space, but it just does not.

By what metric? It almost sounds like to talk about money here, and by such measurements very few non-profits are successful. The most successful multi-media player is VLC, not in terms of profits since they give it away for free naturally, but in terms of users.

You cannot compare BSD and GPL like you are doing with Apple and Google, the licenses are both free, but copyleft is a very powerful trait of the GPL, that the BSD license lacks. What a company does for Linux will typically benefit Linux, unlike BSD projects which have no say..

[John Phamlore]

Take a look at the top mobile OSes listed in Tomi's article that have a future. There's something amazing about them hardware-wise that for some reason is never mentioned. Why is that?

1. Android
http://lwn.net/Articles/446297/
From the comments section, apparently from someone involved in the development of Android:
"Qualcomm was very supportive of us and provided documentation, technical support, etc, while we did the coreLinux port for MSM7X0X and MSM8X50. Since then they've become much more involved and are [now] working directly with the community, submitting patches on lkml, etc."

Qualcomm would have had major incentives to work with Android since earlier Apple's iPhones used Infineon chipsets.

2. Apple's iPhone 4S replaced much of Infineon with Qualcomm chips
http://www.businessweek.com/news/2011-10-14/qualcomm-displaces-intel-in-iphone-baseband-chips-ihs-says.html

3. RIM's make-or-break "first BlackBerry 10 smartphones will come equipped with Qualcomm's S4 Pro MSM8960T processor" according to
http://www.blackberryos.com/content/exclusive-first-blackberry-10-phones-have-snapdragon-s4-pro-processors-4016/.
This would complete a move for RIM from Marvell to Qualcomm:
http://www.forbes.com/sites/ericsavitz/2011/04/11/marvell-losing-rimm-processor-business-to-qualcomm/

4. Microsoft Windows Phone 8 will appear to, at least initially, continue the exclusivity that Windows Phone had had with Qualcomm SoCs:
http://www.anandtech.com/show/6030/microsoft-announces-first-windows-phone-8-hardware-partners-qualcomm-socs-inside

In a span of less than four years, Qualcomm has persuaded two of the top four phone OS makers to switch from another hardware to theirs, while keeping Microsoft's phone OS as an exclusive.

Now you understand the real reason Nokia was dead in the water after the 2008 settlement with Qualcomm and had no choice but to adopt a Qualcomm SoC.

[incognito]

@John Phamlore, you're greater Qualcomm astroturfer than Baron95 ever was for Microsoft (or at least was accused of). You are also more boring and annoying than those Chinese spammers. I wonder why Tomi keeps your comments around, it is apparent that you are talking to yourself about the amazing LTE tech and the incredible Qualcomm. Your contribution to the actual discussion: zero.

[elm70]

@leebase

Quoted: "If the sale of one device does not come at the expense of another....often enough...then they shouldn't be lumped together for analysis. Thus tablets should be lumped in with PC's, but smartphones should not."

I agree with you that there are difference between smartphone, tablet and PC (PC is also divided in netbook, laptop & desktop).
Yes each has his own "market" since each has his strong point:
Smartphone, is a computing device always in your pocket ... small and light
Tablet, is a light PC ... something comfortable in the living room
PC is a proper PC ... something designed for produce more then consume.

But the border line between these 3 is fading ... and in a sense you can see a continuous line with fading/merging extreme ... that unified smartphone to PC, passing via tablet.

There are some almost 5" smartphone (for me is an aberration as a phone, but other think different) ... so what is the difference with a small tablet ?
Some smartphone have DLNA out ... so you can plug to a monitor, use a bluetooth keyboard and mouse ... and you may have the same PC experience ("nobody" is doing this now, but it can happen in the future)

So ... the line between smartphone-tablet-PC is very clear ... Apple probably is the first that understood the importance of it ... maybe also microsoft, but it has never success on implement a convergence experience so far (they hope with WP8 and W8 to start with)

Anyhow, there is no doubt that this is going to happen ... a convergence of OS-Software between smartphone-table-pc ... that explain also why microsoft is investing a lot in the smartphone segment ... giving "away" 1B $ a year for support Nokia in the Mobile OS war.

Maybe now a smrtphone can't replace a desktop PC ... but ... the possibility is almost available today ... and in a near future it could be the reality.

Comparing bicycle to cars ... think that the engine of a today smartphone is more powerfully the the engine of a Desk PC of few years ago ... so the today bicycles have more horse power the the cars of yesterday ....

Tchuss

e_lm_70

ps: Saying that ... they article of Tomi, putting together PC, Tablet and Smartphone it make totally sense :-)

pps: If tomorrow Apple will come up with iOS in a TV ... then this will also enter in the game ... my Samsung TV bought years ago, has linux inside, I can plug in a USB keyboard and USB mouse, I can run and install new programs ... get a linux prompt via telnet ... so far ... is only a PC for hackers ... but if Apple will enter the game, it will cause awareness and tons of copy cat ... and Android TV could become the next "revolution" ....

[John Phamlore]

Compare and contrast Nokia around 2008 with another company, Marvell, that has been making a massive effort into developing a LTE capable baseband chipset. And note Marvell is not assured of its place in the future, but what they are doing is absolutely essential to staying in the baseband chipset competition.

http://blogs.barrons.com/techtraderdaily/2010/07/06/nokia-sells-wireless-modem-unit-to-renesas-for-200-million/
"Nokia (NOK) this morning said it agreed to sell its wireless modem business to Japan-based Renesas Electronics for about $200 million ... The business being sold includes technologies for LTE, HSPA and GSM standards, including related patents."

Considering the outcry on say this blog about Nokia selling off patents, where's the story about this sale before Elop was hired.

While Nokia, an enormous business that was still making large profit, was selling off its IP, observe Marvell would have around the same time started to have become a little nervous because much of its mobile wireless business was based on wins it had selling to RIM, a partner that was a danger to jump to another supplier.

http://www.eetimes.com/electronics-news/4376402/Marvell-aims-to-be-China-IC-leader

Instead of selling IP and exiting businesses, Marvel started from scratch pursuing the future Chinese LTE market, hiring 800 engineers in China: "... we decided that we can do better if we do it ourselves." Therefore it would not have been too late for Nokia invest in 2008. The price would have been partnering with the Chinese. "The team here did everything from conceptualization of TD-SCDMA to its architectural development and production. I describe 99.9 percent of these 800 people as engaged in R&D."

At least Marvell will still be in the LTE baseband chipset game late 2012 "But at Marvell, we do. Our coming LTE solution will be universal – not only applicable for China Mobile but for the worldwide market." Nokia? Having sold off their LTE IP before Elop was even hired for a $200 million pittance, they must dance to other companies tunes.

[lost]

NUMBERS instead of VALUE = FAIL

This comparison fails since author insists on counting "units", and count one average smartphone (100$?) same as one average mainframe (1,000,000$?)

It is ok to compare number of units shipped when comparing same class of devies (smartphones to smartphones, or PCs to PCs), in other word when comparing devices where each unit has similar value.

I really like most of Tomi articles, but some of them try to artificially push value of mobiles too hard - and this is one of them.

If you really want to be objective, give us list of computer makers by revenue. Or even better, give us average value of each of 6 classes of devices (smartphones, tablets, notebooks, desktops, servers, mainframes), and calculate real TOP 10 list based on unit numbers and those averages, thus removing 'who has better sales' from influencing list, but still keeping 'how much each device is valued by customers' as influence in your list.