[OT] Intel's chip-shrinking milestone

[Rubes]

Although I'm a huge nerd, I'm not all that sophisticated when it comes to computer specifics. Just wondering what some of the computer buffs here think of Intel's new milestone. Think this chip will actually ship in the next year, or will production or heat issues inevitably get in the way?

 

Intel shrinks chips to 65nm

[Cash2Burn]

Although I'm a huge nerd, I'm not all that sophisticated when it comes to computer specifics. Just wondering what some of the computer buffs here think of Intel's new milestone. Think this chip will actually ship in the next year, or will production or heat issues inevitably get in the way?

 

Intel shrinks chips to 65nm

12559[/snapback]

 

 

These are memory chips, not CPUs. This is not that big an announcment as other SRAM manufacturers lile Micron and Samsung are already working on smaller geometries.

 

Sounds like Intel is getting desperate to keep up its public image after so many missteps.

[MadBuffaloDisease]

Silicon dioxide is the wave of the future. You heard it here first.

[AKC]

These are memory chips, not CPUs.  This is not that big an announcment as other SRAM manufacturers lile Micron and Samsung are already working on smaller geometries. 

 

Sounds like Intel is getting desperate to keep up its public image after so many missteps.

12769[/snapback]

 

The 90 nm platform discussed in this old article was supposed to be incorporated into the P4 3.2 and 3.4 "Extreme" CPUs that Intel is currenty producing. The original launch for their 90 nm CPUs was December of '03, yet even the new 3.2 and 3.4 are based on the 130 nm architecture.

 

I've worked with Intel on a few "initiatives" and spent time at their Jones Farm facilities where among other things all their new CPUs are put to real use testing in varied environments. Heat is less of a concern than it is an obsession, and where you see a new architecture like the 90 nm failing to make the market in the original time estimates heat is the single most likely culprit in causing the delay. There are means to pipe heat off the CPU but they come with either an expense that doesn't fit the model for price structure and component size/case dimension or the reliability of the cooling method is unacceptable. Give 'em a little more time and I'm sure my buddy Henry Schaecterle will get your new chip on the market.

[Cash2Burn]

The 90 nm platform discussed in this old article was supposed to be incorporated into the P4 3.2 and 3.4 "Extreme" CPUs that Intel is currenty producing. The original launch for their 90 nm CPUs was December of '03, yet even the new 3.2 and 3.4 are based on the 130 nm architecture. 

 

I've worked with Intel on a few "initiatives" and spent time at their Jones Farm facilities where among other things all their new CPUs are put to real use testing in varied environments. Heat is less of a concern than it is an obsession, and where you see a new architecture like the 90 nm failing to make the market in the original time estimates heat is the single most likely culprit in causing the delay. There are means to pipe heat off the CPU but they come with either an expense that doesn't fit the model for price structure and component size/case dimension or the reliability of the cooling method is unacceptable. Give 'em a little more time and I'm sure my buddy Henry Schaecterle will get your new chip on the market.

12888[/snapback]

 

 

What do you make of the new Orion cluster computers powered by Transmeta Efficeons that were announced the other day:

 

http://biz.yahoo.com/rc/040830/tech_superc...rs_orion_1.html

 

Sounds like INTC is playing catch-up on the heat / leakage front.