For CPunks what's needed is a DIY Electron Beam Litho apparatus. Its basically an electron microscope in reverse. For relatively simple chips it could be more than adequate as its possible eliminates the need and complexity, chemistry, etc. for masks and can perform most all process functions (doping, implantation, etc.) and attain feature sizes commensurate with current foundary tech. AFAIK, its possible to build from used e-microscopes though maintaining calibration and linarity is something best learned in a working environment at someone else's expense. :) One of the main shortcomings of EBL is its low production rate since all features must be developed by "writing" them, like an old stylevector display, on the substrate. -------- Original Message -------- From: grarpamp <grarpamp@gmail.com> Apparently from: cypherpunks-bounces@cpunks.org To: "cypherpunks@cpunks.org" <cypherpunks@cpunks.org> Subject: Re: Open Fabs Date: Tue, 28 Jul 2015 22:20:41 -0400
Steve Kinney wrote: If a market is willing to pay enough to support and grow the project, it can be done. Are there potential partners and large scale consumers for "top security through total transparency" to make an open hardware project viable today?
One potential route would be to broker a deal to pool the resources of specialty hardware integrators who already have a market base for high security "solutions." The Open Office project pulled off something similar years ago, obtaining major funding and support from IBM and others who wanted Microsoft out of their hair. So, who wants a shot at defending some of their digital assets from outfits like NSA and GHCQ, badly enough to pay for it?
The first place I would start shopping this "crypto anarchist" project around would be State security services - pretty much any small to mid-sized outfit not in BRICS or FVEYE could be a potential market for auditable scrambler phones for military commanders, senior elected officials, diplomatic corps and double-nought spies. From there to high performance servers and workstations would be a natural progression.
I haven't looked at how the Black Phone folks are doing lately, but that looks like the kind of product line where open hardware might find its first viable home.
Another consideration: One needs not necessarily own the facility where the chips are made: ISO quality assurance programs already in place support client access for audit and validation. A contract that specifies the client's intrusive presence during every phase of production and handling would cost extra, but a QA process that assumes the presence of hostile actors on the shop floor is definitely possible. Such a process would also be needed at a dedicated facility: One must assume the presence of hostile actors there, too. :o)
That's basically all part of the idea. And that some serious multi philosophical combination of hardcore Stallman Ghandi Cpunk Riseup Coder Maker Opensource Auditor like motherfuckers all build, run and observe the joint from the ground up as essentially a crosschecked incorruptible thing that anyone can look at.
Todays shops are a mutable system of hierarchical employee paychecks, payoffs, closed door privacy and backroom games.
On 7/29/15, wirelesswarrior@safe-mail.net <wirelesswarrior@safe-mail.net> wrote:
For CPunks what's needed is a DIY Electron Beam Litho apparatus. Its basically an electron microscope in reverse. For relatively simple chips it could be more than adequate as its possible eliminates the need and complexity, chemistry, etc. for masks and can perform most all process functions (doping, implantation, etc.) and attain feature sizes commensurate with current foundary tech. AFAIK, its possible to build from used e-microscopes though maintaining calibration and linarity is something best learned in a working environment at someone else's expense. :) One of the main shortcomings of EBL is its low production rate since all features must be developed by "writing" them, like an old stylevector display, on the substrate.
Whereas with "photon lithography" (terminology?) a light source covers an area (of the mask?) which is then lens-focused down to the appropriate scale (eg 24nm) onto the silicon and etc physical layer? Firstly is this minimal understanding correct? Secondly is there any potential areal electron emission device (as opposed to point electron emission device) comparable to current litho tech (areal photon emission)?
----- Original Message ----- From: Zenaan Harkness <zen@freedbms.net> On 7/29/15, wirelesswarrior@safe-mail.net <wirelesswarrior@safe-mail.net> wrote:
For CPunks what's needed is a DIY Electron Beam Litho apparatus. Its basically an electron microscope in reverse. For relatively simple chips it could be more than adequate as its possible eliminates the need and complexity, chemistry, etc. for masks and can perform most all process functions (doping, implantation, etc.) and attain feature sizes commensurate with current foundary tech. AFAIK, its possible to build from used e-microscopes though maintaining calibration and linarity is something best learned in a working environment at someone else's expense. :) One of the main shortcomings of EBL is its low production rate since all features must be developed by "writing" them, like an old stylevector display, on the substrate.
Whereas with "photon lithography" (terminology?) a light source covers an area (of the mask?) which is then lens-focused down to the appropriate scale (eg 24nm) onto the silicon and etc physical layer? Firstly is this minimal understanding correct? That's the way things were about 25 or so years ago. Wafers with photoresist were exposed with machines using UV, which over the years used ever-decreasing wavelengths of UV, in equipment:436. 365nm, and 248nm, eventually reaching 193 nanometer UV. In the 1970's, entire wafer-masks (which covered the entire wafer) were used. This became impractical as feature-sizes were reduced. Step-and-repeat devices ("wafer-steppers) https://en.wikipedia.org/wiki/Stepper then allowed exposure of a much smaller portion of a wafer. (these eventually used mirrors, rather than lenses, because it is hard to process UV in a solid lens.) However, because the wavelength of light eventually became a large portion of the size of a chip feature (a line or a space) it was increasingly difficult to 'draw' the picture necessary to expose the resist on the wafer. Due to many ever-more-heroic technologies, it eventually became possible to use 293 nanometer wavelength to expose features far smaller than the wavelength itself, which would have been considered phenomenal in the 1970's. These days, EUV ("extreme ultraviolet") has been used for ever more small features. https://en.wikipedia.org/wiki/Extreme_ultraviolethttps://en.wikipedia.org/wi... EUV is strongly absorbed by air, so such exposure is typically done in a vacuum.
Secondly is there any potential areal electron emission device (as opposed to point electron emission device) comparable to current litho tech (areal photon emission)? None that I'm aware of. But EUV will take over from electron-beam: The latter is quite slow, and typically has been used mostly for making masks.
participants (3)
-
jim bell -
wirelesswarrior＠Safe-mail.net -
Zenaan Harkness