Discussion in 'Article Discussion' started by Lizard, 22 Nov 2010.
I lol'd. Have some rep.
mobile phones will replace generic computer use but it won't replace office computers, high-end computers, gaming computers, or really any kind of specialized computer. nobody is going to want to type an essay, type out a report, do some extensive research, software development, filling out databases, filling out forms, play full-blown commercial games, etc on a phone. its impractical, even if you have a separate screen. if you have a separate keyboard and mouse, you still have to make sure the phone is compatible with software you actually need.
remember, the reason x86 is still being developed is because almost every single computer intended for direct user control uses x86, so to design something different will be incompatible. google apps are pretty good but they're not the best, and companies like microsoft are not going to let them get known too easily.
btw, ibm and sun has released products where you can just slip in more CPUs and RAM a long time ago, its just a really expensive technology. now nvidia has it with cuda servers.
If that were the case zeroOne, laptops would've obsoleted desktops by now. Sure, laptops are smaller and portable and can hook up to your TV (when available), but they really lack in other areas. Keyboard size for many is an issue. Add to that, limited graphics performance for gaming and screen size. Using a cellphone magnifies those same issues. Now you are dealing with an even smaller "keyboard", tiny screen (unless your near a TV for future phones), and even less graphics processing power than modern laptops. All three devices have their place in the future. Yes cellphones are becoming smarter and much more versatile, but they are no replacement for either a desktop or a laptop.
Intel road map.
First ARM, then AMD.
Next, the world.
In the words of "Look at my horse..."
"...pretty sure that the universe pretty much covers everything..."
I really don't know about RISC Vs x86 (pro's cons etc...) - but licensing the designs of ARM chips has obviously proven to be a fantastic business model and a strong 3rd entity in the market place is a welcome addition.
Nvidia are betting on ARM's success, given that they can't get a x86 license, they kinda have to. They have a fantastic opportunity are are getting heavily involed in RISC technology - providing a nice platform for the time when Intel takes off the gloves and engages ARM.
As for Microsoft, are they not already increasing their focus on the mobile opportunities?
Exciting times ahead I feel.
ARM has alot of strengths and with the move to cloud computing could leverage them significantly.
MS is all that is keeping intel in the game
Wonder if ARM is used in space shuttles/Rovers/stations yet?
As "Yahtzee" said
personally Microsoft dithered around too much. The Vista years really was a bad time for them in all product fronts I feel.
Depends on what you see as "Waves" I guess... Intel's been top dog for many "waves" now.
But yes, more competition is always better, and yes, execs are paid to be overly brash
It's "I think you'll find that the universe pretty much covers everything." [/fanboy]
Honestly I don't think that the Atom is any competition for ARM. That isn't derogatory, but it just isn't competitve. I think Intel shouldn't try to make it competitive. What I think they need to do is make it THE chip for low performance, extremely low power draw x86 systems. Example, NAS, HTPC, print severs, etc. Something just powerful enough to handle all of your needs of network through put, disk access and 1080P streaming and not really any more, and do it with the lowest power draw possible.
It shouldn't be trying to take on smart phones, probably not even tablets, maybe netbooks.
The world doesn't need some arms race to make the most powerful smart phone ever. In a lot of cases there is a certain minimum level of compute power required, and anything in excess is just a waste. If you need a 1 gigaflop processor, a 10 gigaflop processor is just wasteful, you want to keep to that 1 gigaflop performance threshold, and innovate to make it lower power/cheaper/better instruction set pretty much in that order.
I think that's his point, everything will eventually be an "integrated system".
You can never have enough gigaflops, but you make a valid point.
It's also valid to accept that software developers should be getting things more optimised/efficient to meet the same criteria - Crysis for example is not optimised.
As technology progresses and more integration becomes imminent, then you will need more power to handle such things.
Having an integrated system on a chip should at least be able to meet the requirement you stated above, but also have additonal power available on demand (something that can switch on and off like more cores, or ramp up the mhz based on the application) so that multi-tasking etc becomes much more useable.
lol! You do know that ARM designed chips sell by the hundreds of millions a year? The things are literally in everything, everywhere. Maybe matey should have said that Intel will never be able to displace them in the area they are good at.
"unit shipments in Q3 (our partners report royalties one quarter in arrears) increased 19% sequentially to a record 828 million units in the quarter. For the reported year, ARM partners shipped just under 3 billion units (2.9bn), up 18% on 2006 and are now at a run rate of ~9 million units per day."
....and that was in 2007.
Does no one else remember Total Annihilation - Core vs ARM
Please adjust your headline to fit with gaming lore
The OP got it wrong, ARM will totally annihilate Core
Separate names with a comma.