yes i was reading an article in a similar vein just now
intel must be crazy thinking ultrabooks are due for mass adoption with haswell
Sadly for Intel, with the current economic situation, ultrabooks won't be good for mass adoption until they are typically <$600. But, that's kind of not Intel's thing. They want to push their higher-priced products. For all of the threat supposedly presented by ARM, having good enough cheaper computers is more of a threat to Intel.
I'm still using my C2D, waiting for Haswell. There have been a handful of games I've put off due to that (none multiplayer), and I've occasionally wished I could have more RAM, but that's about it. A new PC seems tempting while I use one, but then I consider the money involved, and figure I can wait a little longer. My notebook, OTOH, has a new lease on life, with an SSD.
Also, with current technology, I see no point in going with full ATX, even wanting some flexibility. I've yet to see a proper SFF offering that's been suitable, but MicroATX is a no-brainer. I'm sure SFFs will get better, over time, too (FI, if you get 4 USB 2.0 for expansion, and 1 video output, it's going to have a shorter life than a cheap ATX PC, or AIO with an expansion slot or two...screw that!). If they'd cut fewer corners on ports and peripheral options, teeny SFFs could be great options for many users.
Why don't you think that win8 is worth that extra 5~10 mins of your time ?
Because I can spend that 5-10 minutes
once with Windows 7 or a good Linux distro, rather than it being 5-10 min/wk or 5-10 min/day. The choice of the word, "overlay," for Modern UI is very apt, as in
not integrated.
They can always downgrade to win7 you know !
You apparently
don't know. They can't downgrade, unless they spent extra for Pro. Then, they'll have to get a Win 7 disc, and call MS for the activation. It's only really an option for people that want Windows 7, but are stuck buying through a contracted vendor, or who want a specific proprietary computer.
The number one factor is time & money, then there is this small thing that's performance hit cause opengl still lags directx on that front & by a wide margin last I checked !
When was that? nV gimped OpenGL in Windows on newer drivers for Fermi Geforce hardware, and probably Kepler, now (they've been facing the, "problem," of gaming cards being used for content creation, you see...), and that's pretty much been it, TMK. OpenGL as a standard lags behind, but performance should be, and generally has been, similar, with equal support (like from AMD, or on Linux).
Can we name and count the killer apps that drove pc sales? I can start with Winamp and Crysis, i'm pretty sure there are more involving "eras" in pc evolution from the 90's, Napster was one.
Visicalc
Doom
IE & Netscape, after the web became
the web (many people didn't have enough RAM, couldn't use a new modem, etc. etc.).
Mechwarrior 2
Quake
Starcraft (this is
back in the day, OK?)
Quake 2/Unreal
Dragon (very few people kept with it, I know, but you could get practically real-time dictation with a P3/Athlon ~1GHz, and that was an amazing gimmick for most non-gamers I knew, at the time)
Flash programs, and software build with RAD bloatware
"Web 2.0"
Havok (HL2 was the flag-bearer)
Crysis (seriously...it was a good game, but we all had to get it as a tech demo, too)
Lately, BF3, though I am consistently puzzled by that, myself.
Conversely people don't upgrade their hardware just for the "latest OS", ask yourself this ~ when was the last time you upgraded your PC for the latest MS offerings ?
NT 4 Workstation. 95 was just too unreliable.
It's always been for an application, aside from that one case, and the new OS had to come along for the ride, for either newer APIs, newer hardware support, or both (FI, now you need WDDM--can't have DX10+ w/o it--and soon you'll need WDDM and x86-64).
For businesses, Win2k w/ AD was the last time any of them actually considered upgrading for core OS features. It's typically the secondary benefits of a new OS that matter, beyond applications. Having an old MS OS comes with its problems, including security, so upgrading 7 either has been done, or is in the pipe, today (yes, some businesses are basically going to be transitioning right until security updates stop coming). But then, they aren't upgrading the hardware, either. The two may happen around the same time, but the PCs are generally configured and bought entirely separate from OS concerns (buy, wipe, load most recent image, give it a new SID, and send it to its cube).
Wintel is on the verge of cracking at the seems.
Wintel has been practically dead for years, now. Both companies have enough inertia, though, that they can benefit quickly enough for shareholders when they do things really well, but customers are stuck with long durations for migrating
away from something, so it often takes several years for their bad decisions to bite them hard.
You can't go from .NET(or C++ w/ COM)+SQL Server+AD security to pure application-level security, with a cross-platform app, in a year, though you might be able to develop your first version inside the Windows stack, in that time frame. But, today, it's already very common to not make your programs deeply rely on MS' tech, for just that reason. FI, create an application with its own sound security, then bypass the login stage with a domain user check, if they're connecting on the domain. Doing that, or making your DB easy to port to MySQL or PostgreSQL, or making your app Mono-compatible, or doing it a portable language, etc., etc., are all much easier to do at the start, than later, and has become fairly common. Windows could still be easier for your whole business, but you only end up minimally locked in. When they negatively change server licensing, you can then fire up a CentOS dev box, and get to porting, if you planned accordingly (awfully common). Likewise, adding support for inherently insecure non-Windows devices would be easier, too, since you could implement just about any software security layer on top of what you have, internally, as long as you didn't design it just against your domain.
Windows or x86 lock-in is just one of many trade-offs, today, and not one that anyone in a business that needs customers to be happy with their services takes lightly, anymore. But, the existing software bases keep dragging people along, as long as they can. If you've gone to build anything new within the last 5 years, or maybe longer, and it wasn't a PC game, Wintel either has been dead to you, or you will wish you had taken the advice to treat it that way.
x86 matters as much as it does today, FI, because Intel and AMD have done a better job at providing overall value than have cheap SoC implementers, and have, by and large, created much more open HW platforms. Long-term, hardware is getting cheaper, and SoC vendors are gradually starting to get with the program (IE, you can let end-users have documents, or can be told to f*** off, when Linux isn't working well enough on your hardware--we have some middle grounds, for now, but the end-game is for everyone to have similar openness to what we've become accustomed to on x86).
Facebook
Twitter