How to select one of the overlapping objects in gEDA PCB tool?

[zzsczz]

I want to select a part on board and see the detail information of the part.
When the mouse cursor is on the  origin point of U1, I use info->generate object report to see what is there.
I get a fragment of trace  on top layer. See the select_trace.png

I can get U1's info by  hide signal layers or  change to the empty layer that there is  nothing  at the U1's origin point  ,then generate object report.
see the select_u2.png.


Is there any methord to select one of the overlapping objects ?


the pcb is  zynq_som_2.pcb of  https://github.com/gabriel-tenma-white/sdr5

[Doctorandus_P]

I waited a few days on purpose before responding, but it seems unlikely you would get much of an answer to your question. gEDA has received very little development over 10 or so years. Even their website: www.geda-project.org has not worked of the last 3 days.

I also had a look at: https://en.wikipedia.org/wiki/GEDA and apparently they did release a stable version in 2020 and that surprised me.

I used gEDA myself for a short while some 8 to 10 years ago. It's user interface was archaic but "usable", but after drawing a few schematics I wanted to make a PCB and the user interface was completely different. It had the look and feel of two completely separately developed programs that got glued together at some point in time. I was not happy about that and decided to spend some serious amount of time on reviewing PCB design programs. I had a list of around 8 programs, and the 3rd or 4th I reviewed was KiCad. KiCad was quite clunky back then (for example library management was broken) but it also had a lot of things I liked. Back then it had a very good "Getting started with KiCad" guide, and with that I designed my first PCB in a single afternoon after installing KiCad. KiCad also ticked all the other important boxes. It's completely FOSS (which is important to me), and it was being actively developed. Unfortunately far to many FOSS projects fail to gain a big enough community to become successful projects. KiCad however is flourishing. Development speed is increasing, and it is now at a quite astonishing height for a FOSS project. Bugs get fixed quickly, lots of new features get added annually (yearly mayor version cycle) It has a big user community, an active forum where people get lots of help when needed.

All these things combined resulted in aborting the review process I was doing. I just started using KiCad and did not even look at the other programs on my list. And I am very happy with the progress KiCad has made over the years. KiCad has all I want (and more) for a PCB design program.

KiCad is not a magic bullet. It can't do everything (regarding PCB design), and some people are better off with other PCB design software. But if you are using gEDA now, then you are almost certainly working on a hobby level for PCB design, and don't want / need the "advanced" things.

And even if you don't like KiCad for some reason, unless gEDA has made huge progress in the last 7 years (which seems unlikely, although wikipedia does give a hint in that direction) then almost any EDA suite is likely to be better then gEDA.

Also, if you started with gEDA because of this zynq project, there are several (5+) open source/hardware zynq projects for KiCad too. For example:

https://github.com/julianfl0w/zynqPCB
https://github.com/ciaa/Hardware/tree/master/PCB/ACC/CIAA_ACC

And Zynq is quite popular, so you there are probably projects for other EDA suites around too.

And last. KiCad can also create a PCB project from back-importing a set of gerber files. The process is not perfect (for example footprint information is simply not present in gerber files) and therefore some significant rework is needed after the conversion, but the conversion is good enough to be an advantage.

[nctnico]

AFAIK gEDA (schematics) and PCB (board design) where never developed as 1 package. And frankly it doesn't have to be. I've used several combinations of schematics and PCB packages over the decades.
It might be helpful to know which version of PCB the project is for. IIRC some people have forked PCB into PCB-new several years ago and have done quite a bit of work on it.

[zzsczz]

Thanks for providing the information.

The projects that you provided are great, but they are too complicated for me.

The  sdr5  project is described in https://www.eevblog.com/forum/fpga/4-layer-board-for-zynq-and-ddr3/

The autor of sdr5  said  " full 32 bits DDR3 on a 4 layer board while having >70% read/write eye at 1066MT/s".


I  have  convert the sdr5  project( ddr3 's singnals   occupy two layers ) to gerber files  and  imported 2 of  the gerber files  to  eagle. see "Screenshot 2024-02-01 225616.png" in  Attach.
1 import gerber files to eagle's layout editor.
2 create sch  ,add component  and change to layout editor.
3 put the added footprint in the exact same position on gerber track( I check the component's name in gEDA PCB and   calculate the position in eagle ).
4 rename the gerber track's name to the net name in sch .
5 add via and name it according to net name.
6 remove redundant tack.

It's simple  physical work  .   I can clear one to three net in an hour . The whole work may last a week or two. May some  ULPs can speed up the process.




The process is described in the thread:
https://www.eevblog.com/forum/kicad/convert-pcb-from-geda-to-kicad-format/

I tried kicad7. kicad7 can import gerber files but can not save layout.  I guess the reason is that new version kicad  default settings force synchronization form sch to pcb.


The process that eagle imports gerber files(version 7.5 or up is supported.
) is described  in





I  will  study gEDA in this year ,   start from pcb-rnd. Kicad can import footprint form gEDA  format. The sch can be recreated for learning purpose . May a netlist format can be choosed for saving the sch recreating process.

[zzsczz]

I use gEDA 1.8.2-6 and pcb 4.0.2-4  in ubuntu  18.04  WSL.

I have exported the sdr5 project to gerber files .

I  try the latest pcb-rnd 。 pcb-rnd can select element by name