Here is an interesting article that Xnewsman sent me that he found in the New York Times. The basic premise by the author Saul Hansell was that Moore's Law will quite literally run out of space in as little as 5 years.
First, lets explain Moore's Law:
Every memory chip is made up of cells. The job of these cells is to hold electrons (memory remember is a recording and a recording is electrical and therefore containing electrons) Mr. Harari warns, “We are running out of electrons.”
No we are not eating up all the world's electrons, but there is an actual physical limit to how many electrons you can cram into each chip and therefor only a limited amount to spread between all the cells. When chips first came out, they had about a thousand electrons for each cell. With the increase in memory density each cell now is down to just a few hundred. At some point you can see where the cell density would be high enough to drop the electron count to 1 or below....and obviously you can't have a working circuit on zero electrons. The work around for this problem is quite ingenious. Older chips worked very much like computer memory, 1s and 0s or ons and offs but this method limited the size of the chip and at the time made them wildly expensive. Now the chip can actually count how many electrons are in a cell, and depending on the number it can write and read up to 16 states or for the computer geeks out there 4bits.
But at the end of the day, Mr. Harari said, it probably can double the capacity of its chips only two more times. Once the industry goes from its current 64-billion-bit chip to a 256-billion-bit chip (that’s 32 gigabytes), it will hit that brick wall.
There might be a bright spot on the horizon however. Many companies are now experimenting with stacked layers. This technology has great potential but at the present time has major drawback, like only being able to write one time.
Read complete NY Times article here
First, lets explain Moore's Law:
- Moore’s Law — the observation by Gordon Moore, the co-founder of Intel, that the capacity of semiconductors doubles roughly every two years.
- In 1990, SanDisk, shipped its first flash memory chip. Each chip stored four million bits of information. Today, the biggest chip SanDisk makes holds 64 billion bits.
Every memory chip is made up of cells. The job of these cells is to hold electrons (memory remember is a recording and a recording is electrical and therefore containing electrons) Mr. Harari warns, “We are running out of electrons.”
No we are not eating up all the world's electrons, but there is an actual physical limit to how many electrons you can cram into each chip and therefor only a limited amount to spread between all the cells. When chips first came out, they had about a thousand electrons for each cell. With the increase in memory density each cell now is down to just a few hundred. At some point you can see where the cell density would be high enough to drop the electron count to 1 or below....and obviously you can't have a working circuit on zero electrons. The work around for this problem is quite ingenious. Older chips worked very much like computer memory, 1s and 0s or ons and offs but this method limited the size of the chip and at the time made them wildly expensive. Now the chip can actually count how many electrons are in a cell, and depending on the number it can write and read up to 16 states or for the computer geeks out there 4bits.
But at the end of the day, Mr. Harari said, it probably can double the capacity of its chips only two more times. Once the industry goes from its current 64-billion-bit chip to a 256-billion-bit chip (that’s 32 gigabytes), it will hit that brick wall.
There might be a bright spot on the horizon however. Many companies are now experimenting with stacked layers. This technology has great potential but at the present time has major drawback, like only being able to write one time.
Read complete NY Times article here
No comments:
Post a Comment