Insights into wafers, CPU's and the future.

[Rabbi]

Check this out , hope I read this right!

In fact, the lifetime of the Athlon XP processors for both 462-pin and 754-pin sockets it stretched till 2006.

http://www.xbitlabs.com/news/cpu/di...1212131528.html

[Vihsadas]

Originally posted by maggnum on 12-11-2003 at 04:00 AM
...interesting IJ.....a diamond is a matrix of carbon....

That is correct folks! I like Network Solids, they are C00L!

Anyway, here is a short blurb about nanotech and Computers!

http://www.e4engineering.com/item.as...ews&ch=e4_home

[nospamy]

Hello
Perhaps the future CPU will use Reconfigurable / WISC/
Writable Instruction Set Computing, based on FPGA or other
technology (ie StarBridge System).
Another possible future: Laser/optic computing.
An Israeli company had created TFlops CPU based on optic
CPU (as large as phone book).
Diamonds: perhaps in the future they will get cheaper diamonds
from moon / outer space.

[Dark12]

I stumbled on this article while looking for something else, as is usually the case. I found the article interesting, and thought I would share it..

http://www.cpuplanet.com/features/article.php/1499981

In the second half of 2003, Intel's "Prescott" Pentium 4 redesign will take the next step -- from 0.13-micron to 0.09-micron process technology, although it's not called that. Below a tenth of a micron (the size, Intel points out, of a typical virus), it's fair to say you've entered the realm of nanotechnology.

Newb question here..
What are they measuring in microns?
Why would they want smaller?

[Activate: AMD]

Newb question here..
What are they measuring in microns?
Why would they want smaller?

correct me if i'm wrong, but that is the measure of the transistors, and the smaller they make them the more they can pack onto a given area of silicon, which means they get more cores per wafer while at the same time having a more complex core

[jwarriner]

My grandfather, who has been an electrician (as in *NOT* the kind who inspects your circuit breaker) since the 40s, and has been a part of pioneering a lot of technology (including some of the first lasers) considers current processors to still be in the dark ages. Conceptually speaking they could be a lot better.

I think the tendency to make things smaller hurts the progression of technology, it is counter productive to making things perform better. If you could have a PC that was 100,000 times faster than your current one but was the size a buick you probably wouldn't buy it.

Anyone see where I'm going with this? Better stuff is possible, today. It's just not practical.

The new generation of CPUs will be "good" for something that wieghs an ounce and is less than two inches across. Stunning technology or performance? Not really.

[sopclod]

correct me if i'm wrong, but that is the measure of the transistors, and the smaller they make them the more they can pack onto a given area of silicon, which means they get more cores per wafer while at the same time having a more complex core

Basically, yeah. Also it reduces latencies between components, but it also introduces problems, most notable current leakage.

There is always a huge rift between research and what's actually available on the market. How many times have you read about stuff like 5 years ago that still isn't available? I keep reading about all these new technologies and ideas that are in R&D but most of that stuff will never see the light of day. The worst is universities, they work on stuff that is so cutting edge it doesn't even have an application, it's purely academic. I mean there are transistors that will run at 100+ GHz but big deal we can't use 'em.

Another problem is backwards compatability, especially with CPU's. Modern CPU's are based an an arcitecture that was invented in the 70's. I mean there's nothing revolutionary about 64-bit computing, or 128-bit or 256-bit. It's all about compatablilty, marketability and most importantly PROFITability.

What you have in your home is years behind what's going on in R&D but that's how it will always be.

[thechris]

My grandfather, who has been an electrician (as in *NOT* the kind who inspects your circuit breaker) since the 40s, and has been a part of pioneering a lot of technology (including some of the first lasers) considers current processors to still be in the dark ages. Conceptually speaking they could be a lot better.

I think the tendency to make things smaller hurts the progression of technology, it is counter productive to making things perform better. If you could have a PC that was 100,000 times faster than your current one but was the size a buick you probably wouldn't buy it.

Anyone see where I'm going with this? Better stuff is possible, today. It's just not practical.

The new generation of CPUs will be "good" for something that wieghs an ounce and is less than two inches across. Stunning technology or performance? Not really.

no, that goes against everything that has been shown in the progression of the CPU. smaller is better because you reduce capacitance. the specs that determine speed and power all improve if you can shrink the transistors. CPU design probably still is in the dark ages comared to what is possible though. why do you think that smaller is slower?

edit-- adding info: the spec measured at 0.09u is the FET's channel length. making a smaller length will make the FET faster. sometimes you want a larger width to make the FET "stronger". of course there is a tradeoff becasue wide fets are more capacitive and thus use more power.

[danny5]

Have you ever wondered how chips such as microprocessors, video processors, memories, chipsets etc are created and manufactured? In this tutorial we will explain everything you need to know about chip manufacturing process.

Read on here:
http://www.hardwaresecrets.com/article/252

[mmonnin]

Have you ever wondered how chips such as microprocessors, video processors, memories, chipsets etc are created and manufactured? In this tutorial we will explain everything you need to know about chip manufacturing process.

Read on here:
http://www.hardwaresecrets.com/article/252

Post is old I know, but thats a very bare bones intro to semiconductor manufacturing.

As far as the above post about making chips from glass...many layers filler material is essencially a type of glass. Optical inspection tools are very good for viewing not only the wafer surface but even prior level layers.

<-Semiconductor manufacture employee

[pmtilldawn]

D.N.A and atomic processors are way out there read up on them