Ultra Thin Notebooks Climbing While Desktop Sales Decline; Tablets Slowing

They are only make us thinking as they want. I mean sellers as they. I love my desktop pc and i love to change it as i want. So i don't think i'll love a laptop like my desktop :)

for desktop they mean pre built computers righ

Portable laptops (or tablets) that easily double as your desktop, especially with a good Docking Station are a great idea.

People got Tablets but then realized they still wanted keyboards. Hence the tablets that added keyboards (uh, isn't that a laptop?). I guess if the keyboard is removable it's technically a tablet + keyboard..

Anyway, for non-gamers we finally have really light laptop/tablets with Windows 8.1 (desktop/Metro as desired) that can be both portable AND serve your desktop.

13.28 million units were sold during the quarter whereas 14.62 million units were sold in Q3 2013, which is down 19 percent.

"19 percent"?! That seems huge. Something didn't seem right though.
I calculated the values and I only saw a drop of 9.166%! The "1" in "19" may be a typo but in my opinion it is one of the most misleading examples of errors. This should really not be allowed. :-(

Calculation: difference of old value and new value / old value; (14.62 - 13.28) / 14.62 = 0.09166

I don't believe the ultrabook comment. I'll believe that when I see it, as they are just far too overpriced compared to what you get in a loaded regular laptop. There are not that many pure road warriors out there that require this type of overpriced LIGHT laptop. Most of us are bang for buck and an extra pound or two means nothing. The price of awesome laptop perf can be had far below the ultrabook neighborhood. Intel's sales have sucked for a reason. PRICE/Value.

Tablets have slowed because we're waiting for 20nm that is right around the block. This close to something that important will put a lot of people off for purchases. I mean if you like K1's perf, at 20nm EVERY low-end chip will have that (of course the top will be M1 or whatever maxwell 20nm is named). Today low-end stuff can't play games at the ridiculous resolutions of most tablets (or even 1080p), but 20nm will allow pretty good gaming tablets from a lot of socs. A $200 unit will run like NV's shield today which is clearly a leader now. So why buy unless you absolutely have to for some reason? Wait 4-6 months and you'd likely double your gpu or more (bringing low-end to K1). On the cpu side they are all pretty close, but the gpu has huge differences from top to bottom. Now that users are keying more on gaming, they are aware the low stuff doesn't cut it. Since tablets are usually subsidized (unlike phones on contract) you spend a little time making sure you're buying a BETTER model. On contract people still seem to think that phone is free so not as picky, though it's stupid as that 2yrs apple contract costs you $1800-2400 ($75-100 month usually).

RE: lower pc sales, you have to realize they were selling 380mil worldwide, so a drop to 260 is a large haircut (mainly due to arm, 21% of notebooks when to chromebooks). Arm will steal more as 20nm, 64bit+androidL makes it into desktops (low stuff first) and more capable notebooks. I really don't understand Intel's stock price rise knowing this and the damage already done to notebooks and their profits (~13B 2011, ~11B 2012, $9.5B TTM, dropping yearly now). How do they think ultrabooks will increase massively when the world has proven they'll take good enough chromebooks instead and they haven't even made it into a 500w PC like box desktop yet. When ultrabooks don't also come with an ULTRA price, maybe they'll increase massively.

I wouldn't be surprised by 20% desktops being arm by next xmas (2015 xmas, not this year) as 20nm is all over IF someone puts them in two boxes. One with soc only and another with an NV discrete gpu for really good gaming. So I'm saying competition for the $300 dell/hp box on low-end and good competition for a $500-1000 PC (with your choice of Nv gpus in the whole range). What you save in the box comes at the expense of windows apps, which for many won't matter one bit as chromebooks show. They want to merely browse, email, consume vids/music and game. You don't need windows for any of these 5 things. NV will surely put out both with Denver and beyond, with NVlink replacing buses from Intel/AMD and PCIE (hypertransport & infiniband can be replaced with NVlink along with PCIE as NV has already noted with pascal talks). The whole point of unified memory is to share it with their own cpu and gpu. So your cpu/gpu will talk over NVlink with 3d stacked memory. That's going to be a competitive PC for the 4 tasks mentioned and can come with a tri-boot of 3 OS's that are free. Linux, SteamOS and AndroidL giving the user a TON of software already out there and more coming every day. SteamOS will be ported to ARM shortly either by NV or Valve or both working together. Nvidia's goal is to steal Intel cpus sales and keep selling you gpus on either side. You grow your revenue by putting YOUR cpus in boxes to replace Intel/AMD completely and do it with FREE operating systems. At the same time you kill DX/Wintel more and push OpenGL which devs will love due to portability speed. Intel has 20x the income of NV. They want some of it and will get it via a WINTEL-LESS pc. ;)

Also that cpu/nvlink/stacked memory combo will replace AMD/Intel in their top500/green500 supercomputers. NV will feed their gpus with their own cpus. Owning NV stock is a no-brainer, and I'm not even talking cars etc. They'll either explode on their own or be a takeover target by google/ms/Intel who could all afford the ~$23-25B tag (more? Market cap ~9.5B with all the patents maybe far more). MS makes that almost a year. Google 1/2 of it, so only 2yrs income to own great stuff for them (socs for all their devices to expand their power in cheap devices for longer, vertical integration just like apple/samsung/qcom, the more you can integrate on your own the better). MS can do the same now that they own Nokia it would be wise to buy NV. Not sure if Intel can get it done without making Jen Hsun CEO as he has stated he'd want. Too much hate for that between them and their engineers on both sides probably but definitely Intel's best move to stop losing $4B a year on mobile. Put out a 14nm K1 next year and you'll do some damage to Qcom's train from then on, alone keep losing 4B+ which at some point will piss off shareholders.

I think a lot of people are just using their cell phone or tablet as a computer instead of using a computer. So including smartphones should be part of the survey. Some people also use their tablets as phones using Skype also. This can blur the lines of what is a computer and what is a cell phone.