Utter nonsenseHaswell is supposed to be Intel's coming ace. a low watt Haswell beating ARM by 200 percent would be a disaster for Intel, because ARM would catch up and overtake Intel in two to three years, and Intel would once again be stuck at 5-10 percent yearly improvements. Like I said before, Intel needs something very special win this game.
No one said it will be a piece of cake but that doesn't mean it isn't possible.
For example, there was always skepticism about whether or not microprocessors would be able to effectively scale to our current levels while still following Moore's Law. However, the truth remains in that history has shown us that Intel is highly capable of conquering what often times appears to be nearly insurmountable odds. Intel's massive R&D budget and their team of brilliant scientists and engineers are nothing to be scoffed at. If you were to assume that ARM and Intel's engineering teams were of equal expertise,
at a given point in time, Intel's R&D budget and fabrication infrastructure advantage alone would give them the immediate edge.
Long story short: Sure, there will be a challenge, but such challenges are expected, especially if you are Intel.
R&D budget,telented engineers and fabrication lead mean nothing and I mean nothing, if not properly used. I said somewhere in this thread give a genius dirt and he will turn it into gold. Now give that gold to a fool and he will lose it within an hour or two.
Both companies with the given circumstances have equal chances. It will depend on how well they are executed.
Intels GT3 graphics are comparrable to Nvs 650m.
Intels GT3 graphics are comparrable to Nvs 650m. Is that good or bad?
A true genius doesn't care about money. They would care about science.
According to a demo made at low settings with no AA and so on
and with a desktop CPU HSW compared to a low power CPU
powering the NVidia gfx.....
According to a demo made at low settings with no AA and so on
and with a desktop CPU HSW compared to a low power CPU
powering the NVidia gfx.....
If the GT3 graphics in their mobiles chips are even close to an 650M, that would be awesome. It would also kill the market for discrete cpus in laptops.
A 650M will run pretty much any game on the market. Obviously not with max settings, but certainly good enough for a typical laptop screen.
A really doubt that it's true though, since a 650m is 3-4x as fast as GT2 graphics.
Perhaps if I was clever enough I'd be able to create 22nm 3D finfet xtors in my basement, MacGyver style, using nothing more than dirt and a paperclip
That is a nice saying, but just because you say it doesn't make it true.
I spent many years as a process node development engineer, working on nodes spanning 0.5um to 32nm and there was absolutely no substitute for money.
There is an unmitigatable barrier to entry in this industry, you can't just give a genius dirt and have them turn out a 22nm process node complete with finfet xtors. Not even if you had 10,000 geniuses are you going to get there, as evidenced by all the foundries and IDM's who are still struggling to get a node into production at 32nm or 28nm.
Your comment basically asserts that every process node engineer out there that is working at a company which has anything less than absolute leading edge process technology must not be a genius, because all you have to do is give a genius dirt and they will make gold according to you.
The project management triangle is universally applicable to any and every industry. You can pick any 2 of the 3, but no more than 2, as your priority.
^ doesn't matter the intended schedule or scope of the project, without money the quality goes to zero. Geniuses make for poor cash, even when ground up and properly dried. (but I hear they make great chips if you don't mind them being green in color )
The days have long passed since you could rely on people making the difference. Nowadays there is no shortage of experienced talented genius engineers (process or design) but there is a shortage of R&D dollars to put them all to work unless you happen to have a job at Intel.
GloFo is about to hire a whole bunch of process development engineers as they are leaving the IBM fab club, that will be a good thing. But I cannot agree with your overly generalized statement, it is catchy and sounds great but it lacks a basis in reality when contemplated by someone who has experience in the trenches.
I think I have to explain what I meant. It won't be easy, but let's see what comes.
A good CEO will know how to raise the necessary funding, whom to employ for any given task, how to keep the employees motivated, and most importantly a vision for the future and how to achieve it.
Example: Steve jobs overtook an Apple which was on the verge of bankrupcy and look what he did with it. It was just like taking dirt and turning it into gold. It was HIM who did it and not the money available to his company, talented engineers and advanced fabrication. With him, came all the talent, money and high tech.
So, I think I explained it. wasn't as hard as I thought.
True that. Jobs (Apple), Ellison (Oracle), Grove (Intel), Jensen (Nvidia), Chang (TSMC)...all pivotal geniuses in their respective companies and the companies are proof of it.
Come on IDC you left out a good guy there from AMD . I even admired the old fart. Remember only real men have fabs.True that. Jobs (Apple), Ellison (Oracle), Grove (Intel), Jensen (Nvidia), Chang (TSMC)...all pivotal geniuses in their respective companies and the companies are proof of it.
Something something "no true Scotsman fallacy" something something...A true genius doesn't care about money. They would care about science.
True that. Jobs (Apple), Ellison (Oracle), Grove (Intel), Jensen (Nvidia), Chang (TSMC)...all pivotal geniuses in their respective companies and the companies are proof of it.
Errr, Is there no i3 there or is it just me They will lose alot of lower end sales
Haswell is supposed to be Intel's coming ace. a low watt Haswell beating ARM by 200 percent would be a disaster for Intel, because ARM would catch up and overtake Intel in two to three years, and Intel would once again be stuck at 5-10 percent yearly improvements. Like I said before, Intel needs something very special win this game.
No no no . Haswell being able to go into tablet space is whats got everyones goat. Its one one hell of an accomplishment , Soon enough arm will do battle with Silvermont the end of the arm war on intels space. The others OEMs made a mistake to challenge intel in its space . Now intel has entered there phone space. Bad move by Arm makers
One could also say the competitors are mentally challenged.
Microsoft, Microsoft/IBM, AMD, AMD, GloFo. See the trend?
Come on IDC you left out a good guy there from AMD . I even admired the old fart. Remember only real men have fabs.