X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f X-Recipient: geda-user AT delorie DOT com X-TCPREMOTEIP: 100.0.183.69 X-Authenticated-UID: jpd AT noqsi DOT com From: John Doty Content-Type: multipart/alternative; boundary="Apple-Mail=_31EBA04C-75C8-45B6-8F2D-A80C3C87C991" Mime-Version: 1.0 (Mac OS X Mail 13.4 \(3608.120.23.2.4\)) Subject: Re: [geda-user] SAB processing patches Date: Thu, 10 Dec 2020 11:23:44 -0500 References: <20201130220505 DOT 0AE4282C54FD AT turkos DOT aspodata DOT se> <7c75ed03-456c-b408-8b50-0448f6b3a04f AT epilitimus DOT com> <1b2c64b3-6a36-c1f3-dd54-bb583c6bea17 AT epilitimus DOT com> <475f980e-fddd-60d1-9a02-a5bc5fb5805b AT epilitimus DOT com>

<25887669-D8C1-48B8-BEE7-13EEBFC4D006 AT noqsi DOT com> <7f074abb-e7ae-275c-ee49-d06fc7f2a706 AT epilitimus DOT com>

To: "Glenn (glimrick AT epilitimus DOT com) [via geda-user AT delorie DOT com]" In-Reply-To: Message-Id: X-Mailer: Apple Mail (2.3608.120.23.2.4) Reply-To: geda-user AT delorie DOT com Errors-To: nobody AT delorie DOT com X-Mailing-List: geda-user AT delorie DOT com X-Unsubscribes-To: listserv AT delorie DOT com Precedence: bulk --Apple-Mail=_31EBA04C-75C8-45B6-8F2D-A80C3C87C991 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=utf-8 > On Dec 9, 2020, at 8:01 PM, Glenn (glimrick AT epilitimus DOT com) [via = geda-user AT delorie DOT com] wrote: >=20 > Okay so I think I now understand the direction you are coming at this > from. However, I think there are use cases you are not taking into > account. For instance perhaps I need to use a verilog simulator = instead > of spice. Or maybe I am using a proprietary simulator that only uses > some non-spice netlist. Then you add the appropriate annotations and support files for *that* = approach, and use an appropriate netlister. spice-noqsi lets you draw = what you=E2=80=99re intending to build and annotate it for simulation in = SPICE. There=E2=80=99s no reason you can=E2=80=99t have annotations for = other simulators if you have a back end that understands them. > I also disagree that the component should not be in the schematic. = Some > users may be like me and prefer to see it. Others may be like you and > prefer not to. Remember, the schematics you publish to layout, manufacturing, QA, field = supports, etc. need to be generally comprehensible. Simulation = schematics are like a prototype, with extra wires tacked on and plugged = into a specialized test fixture. spice-noqsi supports test fixtures in = several ways. Test schematics with arbitrary connections may be fed to = the netlister along with selected board schematics. Unlike earlier SPICE = netlisters in gEDA, spice-noqsi can use hierarchy in a way that=E2=80=99s = compatible with board design, making modular design and simulation work = fluently. But I approve of any add-ons to either gEDA or Lepton that support other = flows. However, as a Lepton user, I do not want to see special jimcracks = and geegaws added to Lepton to support specific flows: those belong in = add-ons. spice-noqsi did not require any special code to support it. And = who knows, I might return to gEDA at some point. >=20 > Certainly if we are working on a circuit that has optional real world > sections we want to see those as part of the schematic but still be = able > to mix and match them in both the simulation and non simulation = contexts. I do that by putting my test fixtures on separate schematics. It=E2=80=99s= analogous to testing hardware on the bench.=20 >=20 > I am not advocating getting rid of anything that geda-gaf currently > offers. Except clean modularity, with specialized objectives served by optional = scripts. > I think SAB adds to it, just like spice-noqsi does. Both provide > the user with tools, and in both cases the tools are completely = optional > and up to the user to employ or not as they see fit. If you need changes to the core code, every user is using your changes = whether they choose to or not. John Doty Noqsi Aerospace, Ltd. jpd AT noqsi DOT com --Apple-Mail=_31EBA04C-75C8-45B6-8F2D-A80C3C87C991 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=utf-8

On Dec 9, 2020, at 8:01 PM, Glenn (glimrick AT epilitimus DOT com) [via geda-user AT delorie DOT com] <geda-user AT delorie DOT com> wrote:

Okay so I think I now understand = the direction you are coming at this
from. However, I think there are use cases you are not taking = into
account. For = instance perhaps I need to use a verilog simulator instead
of spice. Or maybe I am using a = proprietary simulator that only uses
some non-spice netlist.

Then you add the appropriate annotations and = support files for *that* approach, and use an appropriate netlister. = spice-noqsi lets you draw what you=E2=80=99re intending to build and = annotate it for simulation in SPICE. There=E2=80=99s no reason you = can=E2=80=99t have annotations for other simulators if you have a back = end that understands them.

I also disagree that the component should not = be in the schematic. Some
users may be like me and prefer = to see it. Others may be like you and
prefer not to.

Remember, the = schematics you publish to layout, manufacturing, QA, field supports, = etc. need to be generally comprehensible. Simulation schematics are like = a prototype, with extra wires tacked on and plugged into a specialized = test fixture. spice-noqsi supports test fixtures in several ways. Test = schematics with arbitrary connections may be fed to the netlister along = with selected board schematics. Unlike earlier SPICE netlisters in gEDA, = spice-noqsi can use hierarchy in a way that=E2=80=99s compatible with = board design, making modular design and simulation work = fluently.

But I approve of any = add-ons to either gEDA or Lepton that support other flows. However, as a = Lepton user, I do not want to see special jimcracks and geegaws added to = Lepton to support specific flows: those belong in add-ons. spice-noqsi = did not require any special code to support it. And who knows, I might = return to gEDA at some point.

Certainly if we are working on a = circuit that has optional real world
sections we want to = see those as part of the schematic but still be able
to = mix and match them in both the simulation and non simulation = contexts.

I do that = by putting my test fixtures on separate schematics. It=E2=80=99s = analogous to testing hardware on the bench.

I am not = advocating getting rid of anything that geda-gaf currently
offers.

Except clean = modularity, with specialized objectives served by optional = scripts.

= I think SAB adds to it, just like spice-noqsi does. Both provide
the user with tools, and in both cases the tools are = completely optional
and up to the user to employ or not as = they see fit.

If = you need changes to the core code, every user is using your changes = whether they choose to or not.

John Doty = Noqsi = Aerospace, Ltd.

jpd AT noqsi DOT com

= --Apple-Mail=_31EBA04C-75C8-45B6-8F2D-A80C3C87C991--