home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.arch
- Path: sparky!uunet!wupost!spool.mu.edu!enterpoop.mit.edu!bloom-picayune.mit.edu!athena.mit.edu!solman
- From: solman@athena.mit.edu (Jason W Solinsky)
- Subject: Re: uniprocessor design ceiling
- Message-ID: <1993Jan3.215458.7960@athena.mit.edu>
- Sender: news@athena.mit.edu (News system)
- Nntp-Posting-Host: m4-035-8.mit.edu
- Organization: Massachusetts Institute of Technology
- References: <2340@sousa.tay.dec.com> <1ho4g1INNgac@nigel.msen.com>
- Date: Sun, 3 Jan 1993 21:54:58 GMT
- Lines: 46
-
- In article <1ho4g1INNgac@nigel.msen.com>, paulh@msen.com (Paul Haas) writes:
- |> Bob Supnik (supnik@human.enet.dec.com) wrote:
- |>
- |> : The ingenuity of processor designers (and semiconductor technologists) being
- |> : far from exhausted, I believe this performance growth can continue for the
- |> : rest of this decade. There are some limiting factors:
- |>
- |> : - Process technology evolution appears to be slowing somewhat, due to
- |> : lengthening development time (and cost) for new generations of semiconductor
- |> : processing equipment.
- |>
- |> Is this a long term trend?
-
- [begin sepculation]
-
- I doubt it. In the past the slowing down of process technology evolution has
- lead to process technology revolution. As the rate of improvement in
- lithography slows down, people will just start looking towards new
- technologies and begin building into the third dimension. There are certainly
- many technologies out there which are just waiting for somebody to do the
- research to make them more affordable.
-
- Additionally, now that voltage is scalling with device size, total CMOS
- performance (speed/area) is inversely proportional to the third power of
- lithography line size, not the fourth power as used to be the case. That
- means that the threshold at which it becomes more profitable to spend money
- on 3D devices than on smaller lithography has become considerably smaller.
- If cheap sub-micron fabrication is as difficult as people have been saying,
- process people will just head upwards.
-
- [end speculation :-) ]
-
- I have a question along those lines. If (as I have been led to believe) there
- exists a linewidth which is large enough for yield to be effectively 100%,
- why hasn't anybody attempted to build a very tall, large line width chip?
-
- Is the problem a difficulty in creating monocrystaline Si above the substrate?
-
- Are problems with planarization cumulative such that working with too many
- levels becomes impossible?
-
- Or is there another reason?
-
- Thanks.
-
- Jason W. Solinsky
-