| ID: | 5653 |
| From: | rn...@igor2.repo.hu |
| Date: | Mon, 16 May 2022 18:02:49 +0200 (CEST) |
| Subject: | [pcb-rnd] gEDA "NET" attribute and subc-subc, really tool suite plans (was: Re: RFC: multi-sheet netlist) |
| in-reply-to: | 5650 from ka...@aspodata.se |
| replies: | 5654 from rn...@igor2.repo.hu |
On Mon, 16 May 2022, karl@aspodata.se wrote: >Agreed, the NET attribute is a misfeature. I disagree; attribute based connection is a good idea, and I've included that in the cschem model too (in fact it already works in sch-rnd). Questionable idea is having such connections embedded in a generic library symbol (user's decision, tho), bad idea is having them in general purpose public libraries and terrible idea having them in the official, stock lib (developers' decision). A good feature can be misused. It's not the feature that's broken, it's just bad usage. <snip> >> The hierarchic netlist support in pcb-rnd will mean you can draw your >> amplifier channel once, save it as a subcircuit (that contains >> subcircuits), and when you are drawing the top level board, the whole >> thing is placed. > >That would be ground breaking, I would really support such a feature. It's long planned. I've already prepared some of the code back in 2016..2018 during the big data model rewrite. But then I decided to wait for sch-rnd with finishing it, because it needs proper sch-side support. So first we need a stable sch-rnd release (I expect it next year; pcb-rnd usable beta release this year). After that we finally have the chance to implement things that require cooperation between different tools of a workflow because we don't rely on unwilling 3rd party suppliers on any of the tools. That's the immediate gain. But if you zoom out a bit, it's much more! We will reach the state where we have a toolkit approach (independent tools), but without the drawback of unaligned developers and missing suite level vision/planning/leadership. This is the hard part of an EDA suite, avoiding both integration and disintegration. I think that will really make a difference on how we can approach problems. Instead of going purely tool based, we can think in workflows. "This feature needs this little addition in sch-rnd, then that little feature in pcb-rnd and this third in the transport between them, then a little tweak in camv-rnd to work properly" instead of "let's try to work everything around from within pcb-rnd." Subc-in-subc, or rather, in workflow sense hierarchical pcb design, is just a good example of that. But I have a mental list of similar stuff. Complex things like pcb-rnd drc checks generated from schematics. Simple things like an _optional_ project file that really contains the configuration and file list of your project, for sch-rnd, pcb-rnd, camv-rnd, etc. So if you want, you can use the tools together, on project basis, without having to integrate or bind them together in any way. (This project file already existed in pcb-rnd for many years now, it now exists in sch-rnd too, and it will take only a few days of work once I get there to be shared among tools, without breaking backward compatibility.) Best regards, Igor2
Reply subtree:
5653 [pcb-rnd] gEDA "NET" attribute and subc-subc, really tool suite plans (was: Re: RFC: multi-sheet netlist) from rn...@igor2.repo.hu
5654 Re: [pcb-rnd] gEDA "NET" attribute and subc-subc, really tool suite from rn...@igor2.repo.hu