Updates to this story
Chipzilla says its revenue won't live up to its previously uttered expectations. Everyone's favourite chipmaker said it will be around $11 billion, add or take $200 million. Intel had originally claimed it would have revenues somewhere between $11.2 and $12 billion.
Evil consumers are not spending money on hardware with Intel inside are to blame as demand is weaker than expected "for consumer PCs in mature markets." It's so sad.
Despite suffering from weaker revenues, Intel says everything is going to stay just peachy, except for its gross margin which will be at 66 percent instead of 67 percent. Give or take a couple of point estimates either way.
At least the share prices are unfazed - Intel shares are up by 1.71 percent plus or minus one percent at time of print on the NYSE.
Intel's fabless mobile rival ARM so far has outperformed its competitor by not releasing a warning.
.jpg){kind=icon}
now advancement called sandy bridge, compared in print to Intel 880, see 880 as 8808 to 6800, meaning Sandy bridge, once done will be finalization of 68XX series. Sandy as in sand box. bridge to next level. that includes amd bulldozer as other half of bridge. put two together & ready for NT7 Alpha up or as O/S after8 or 8. that should be software going into deep into multicore of sandy bull. bridge dozer or as many cores as takes, first might be just repeats, 2 cores 1 modules con.roller. then ?integration, 2+1 is dual 4+1 is Multi, first 2+1 plus 2+1 mirrored, then integrated 4+1, thats lot of crossbar talk, so developes after mirror of 2+1 twice or more in repeats of fundumental 2+1. how far can crossbaring cores with controller go, 4,6,8,10,12. Shurely NOT as Far as 32 0r 50 0r80 cores. with one controler. At Some point multi core will be established & nt7, multi core O/s might standat that point can say first or very few 4 bytes of hexadecimal code be understood, beyond front or back. 13 more hexadecimals to go. Each hexadecimal adding rapidity needed to function such large beast. Multiple hexadecimal setting for broader range of possible settings.
this is based on Eye: TSVD invented hexadecimal code in 1959. in batch 82 bits per column: on or off(punch or blank), & first in corner, small area of 4 possible punched only one can be punched 1,2,3 or 4. that chose 1 of 4 preselected hexadecimal codes. hexadecimal codes to apply 82 bits in sensor for column one & revolving memory awaiting parseing into whole of columns. to 16 bytes per hexadecimal (4 x4possibilities of punched or blank)) x5( groups of 16)=80 punches per column plus front or back. front tells machine to prepare switches to be set by hexadecimal punched back turns of set up itself, then front yo run, then finally back to reset entire mechanism to null set after run of batche or idle.
Front =hexadeciaml 1 back = hexadecimal16. 14 useful sets of 16 4 bits IBM 360 = one 16 bit git, faster & quicker in very limited memory space. bytes then have unique meaning, of that 4 are picked for run, of that possible menu 1 is chosen, All batche cards in run must be ?related and that specific hexadecimal code. pre configured in machine. Or machine must be able to change setting much faster than card goes thru machine.
latter punched numbers where replaced by one corner cut, each corner 1 of 4 possible hexadeciamals posible in machine, AT THAT TIME. so IBM 360 has eye into 4 hexadecimals. 1 PER RUN & CARD, AT LEAST ONLY 1 PER CARD.
IBM is on level 14 Hexadecimal right now for developement. last of hexadecimal possibilitiee's. completeing 16 digit byte. in 2005 IBM was working on 11th hexadecimal in 1998 On 9th Hexadecimal. so see as need increased, greater numbers of hexadecimal have been mapped. Also as nm scale approached finalization, seems.
Faster machines can eat that stuff & perform more comlex task in specific time now. Migrates to public cpu(plus now gpu developement on die, as workable ?item). in fact multi core will shurely be same as batche 4 hexadecimal possiblities,more as NY # goes up. as cpu sets up or off. dual core maybe two hexadecimal & single core, semperon or lower powered atom or any single, needs No hexadecimal or one. one being one of 14 possible hexadecimals, those unique to machines needs & changeable with each file run.
So Point:conroe & two hexadecimals took 4 years to map, integrating two cores & hexadecimals more mustah be at leastanother 4, say tv 3K x 6K Pixels(HD). more multi tasks. cad cam, run small factory. No Wonder CELL Didn't push envelope as far as promised, Hasn't been invented,yet, not enough controllers, ALL Awaiting mapping. Very Versitile machine at any one point in time , ALL Cores could run One Hexadecimal in one software package & be TRUE Multicore.
So 6,SEX,HEX. Mark of Devil. STeWie.
vondrashek md Ps Most of this information is Classified:TOP SECRET.