Article: Coverage of IEDM 2003: Day 1
By: David Wang (dwang.delete@this.realworldtech.com), December 10, 2003 4:06 pm
Room: Moderated Discussions
Singh, S.R. (swaranNOrajSPAMsinghPLEASE@hotmail.com) on 12/10/03 wrote:
---------------------------
>David Wang (dwang@realworldtech.com) on 12/9/03 wrote:
>---------------------------
>>Does anyone care? It takes considerable energy to write these things up. The
>>material is "interesting", but perhaps only to me. If it's only interesting to
>>me, then there's no point in writing it up....
>
>Sorry, I slept from 2pm - 10pm today, and I just noticed the article. The material
>on the fibre FETs is interesting, but I'd like to see more technical material. Just
>last week, I came up with a rise/fall time model for depletion-load NMOS and CMOS,
>so I'm more interested in the actual numbers for the Ion/Ioff ratios rather than an explanation of what it is.
Right, but writing for RWT is a bit tougher because I have to account for people that have no idea what Ion/Ioff is, so by giving numbers without contexual reference would not be meaningful. It's the old "target audiance" issue. If I write at the level they are presented at IEDM, the number of people who actually understand it would be few. If I write at the level targetted for software programmers, that would take too long. In general, I try to target a somewhat sophomore level, assume that you have some hardware background, but perhaps not enough to know what Ion/Ioff is.
I do have Ion/Ioff numbers for Intel's process. That was disclosed here yesterday. I'll put those numbers in the next report.
---------------------------
>David Wang (dwang@realworldtech.com) on 12/9/03 wrote:
>---------------------------
>>Does anyone care? It takes considerable energy to write these things up. The
>>material is "interesting", but perhaps only to me. If it's only interesting to
>>me, then there's no point in writing it up....
>
>Sorry, I slept from 2pm - 10pm today, and I just noticed the article. The material
>on the fibre FETs is interesting, but I'd like to see more technical material. Just
>last week, I came up with a rise/fall time model for depletion-load NMOS and CMOS,
>so I'm more interested in the actual numbers for the Ion/Ioff ratios rather than an explanation of what it is.
Right, but writing for RWT is a bit tougher because I have to account for people that have no idea what Ion/Ioff is, so by giving numbers without contexual reference would not be meaningful. It's the old "target audiance" issue. If I write at the level they are presented at IEDM, the number of people who actually understand it would be few. If I write at the level targetted for software programmers, that would take too long. In general, I try to target a somewhat sophomore level, assume that you have some hardware background, but perhaps not enough to know what Ion/Ioff is.
I do have Ion/Ioff numbers for Intel's process. That was disclosed here yesterday. I'll put those numbers in the next report.
