Researchers at IBM have discovered a way of getting more out of the current method used to make computer chips. They say they have been able to etch circuits on silicon wafers that are a third of the width of those produced using existing technology.
The technique could lead to smaller and higher capacity chips, and delay a switch to costlier and unproven chip-making methods.
The IBM team said they were able to “print” circuits that are 29.9 nanometres wide. This is about one-third of the width of the smallest computer circuits in mass production today.
Smaller is good. Cheaper wouldn’t be bad, either.
“Smaller is good. Cheaper wouldn’t be bad, either.”
From what I have seen, the chances of cheaper are small. Most of the cost of “expensive” chips is recouping the R&D and the building of the chip factories and machines.
Or to put it in other ways:
* the incremental cost of producing a single chip is relatively small to its price.
* once a fab plant is up and running making a chip line, the cost of producing 5000 chips vs. 1000 chips is insignificant.
So, if they can use the same existing fab lines/plants, and there wasn’t billions in R&D to get this concept, they will make LOTS of money, or they can sell it VERY cheap.
Lou,
Good point. Where I differ is that first generation products usually are more expensive. Later versions though do become more economical. The spokesman did mention that this process uses the same basic technology as current chips and will stave off moving to newer, different methods for a while.
Down the road I see bigger, faster processors using less current. That translates to less heat. Which means cooler temperatures. Cooler temps means more transistors. Which means more power.
I think Steve N. said it best last week. When he can burn a DVD in three seconds then computers are powerful enough. So I say, bring ‘em on !!!