What approach do you take to save revisions without going through GIT ?!

Oh! Good luck and take care.

Indeed Yes…I guess you somehow got a French word put into @John_Pateman 's spell checker dictionary?

Weren’t they there to begin with? I think that normally all of those are in the project folder.

I meant the storm he suffered…

Yes, the Original PCB and Schematic are there. And, if original name has VxRx (for Version and Rev) then only need to Copy and Paste the copies into the folder (and rename as desired).

It’s just ‘Copy&Paste’ stuff as usual so, how User does it, doesn’t matter.

Point being: A single Project Folder can contain as many PCB/Schematics as desired and all will have unique names (and, if desired, unique Netlist/etc… all properly named and labeled in the Sheet’s Legend).

Yes I understand and agree. That is why I said “Indeed Yes.” But I was also commenting that John’s spell checker made a strange choice.

OK I have never experimented with having multiple Project, Schematic, and pcb files all in one folder. I assume that the project will only “see” the pcb and schematic files with the same name but due to risk of confusion I have not tried working with that.

When any of the .PRO files in the project are opened, All of the Files in the project will be listed. Can open any of them.

For Test:
• Create a New Project and do as I instructed

Simple and you can’t mess up anything (except possibly the Test …)

As Aris_Kimi states, an external repository is not required. Git works fine with local files only. Github is simply one of many providers that host repositories. It is also possible for any adept admin to host a repository on a LAN, that functionality is included in the reference implementation of git.

Incidentally the book that documents Git is free, and can be found here: Git - Book

BTW for those who haven’t heard it before, a git quip which Linus made: “I’m an egotistical bastard, and I name all my projects after myself. First ‘Linux’, now ‘git’.”

:wink:

2 Likes

@BlackCoffee, is it possible to use multisheet designs in your approach? I mean, you can have subsheet, shared accross two projects in the same dir. And sch v1_R1 can have slightly different the same subsheet than v1_R0, meaning, that one needs to create-rename subsheet inside schematic and in filesystem…?

The Best answer /advice is to setup a simple Test Project and try the various things you might want to know the answers to… That investment costs only a few minutes of your time… and the software is Free

I always suspected from context, but 54 years later… :wink:

And curse Sir Walter Raleigh.
He was such a stupid get

I think we would get interesting result if we make a table with poster age and +/- for using/not using VCS :slight_smile:

I have never tried VCS or git (I understand that git is a name of one VCS implementation, but not sure).
I don’t feel a need for any. I don’t know what it could give me.
I know for some of you it can sound like “I design PCBs without drawing schematic and I don’t feel a need for schematic.”.
I have never noticed any problem with versions.
My PCBs have the names like BR50_A and the next version is BR50_B and so on. If change is such that the software need not be changed than I name it with additional number (like BR50_B1).
I read it as Bibi (our access control system name) Reader 5 (means mifare) and 0 is for first case we used for readers. In next case it is BR51 and in next (with LCD display) is BR52 and so on.
Unique readers have 4 in place of 5. Smaller numbers were used in previous century for magnetic stripe card readers.
I don’t place at schematic any notes about what is the difference between this and previous version.
I have that in separate spreadsheet were each PCB version has only one line for it. I write there only important reasons I have done the next version.
I like to have all that information for all PCBs together as the reason for updating several PCBs happens to be the same. Having it together I protect myself against forgetting that I have changed something in another PCB and when I will be doing that one (2,3… years later) I planned to make the same change by the way.

5 Likes

Piotr I agree 217.09% +/-50% :slight_smile:

That is the part which I have been lacking.

@Piotr as you know your ESL is not perfect. I wonder whether some other ESL forum participants might have difficulty with your post.

Well, beiing ESL myself, I have no problem with his posts. @Piotr: we ESLs have to stand together against that MEGA (Make English Great Again) movement :rofl:

2 Likes

The only reason i am posting this in favor of git (or some other VCS) in a topic asking info for manual versioning handling is that, those tools were developed to help with versioning and history tracking. (and collaboration among others)

In my understanding this means that with the use of a tool like this, workflow could became more efficient.

As a guy that was really in favor of simple solutions and manual history tracking (also trying to avoid git for its obvious reasons :slight_smile: ), after getting the time needed to feel a little more comfortable using it, i can now only advice the same to the rest. It does help, it really worths it. It is not useful only to programmers, programmers may push the limits a little more though…

Honestly “Learn your way using git.” was one great piece of advice i got to get. And if i am allowed for one more thought, i believe that a lot of human power is getting lost because it only feels really difficult.

Best regards.

Git started off in the software world as the drivers of version tracking, automated workflows (especially driven by scripts), collaborative working with massive parallelism, are essential.

In the hardware world where projects have less people, sometimes only one person, working on it, a fair amount of manual steps, those drivers are weaker.

I come from the software world, where it’s risible if you have to rummage through old directories or zip files to find the difference between the code you have now with what you had 6 months ago, not to mention the huge amount of storage wastage for the files that have not changed. So it was natural for me to apply a VCS to design files too.

But it’s worth thinking how one might take advantage of VCS in future KiCads. I seem to recall mention of a visual diff where you can view changes in the schematic between versions. Automated builds and regression testing of designs could be another.

I publish my open source amateur potterings both software/firmware and hardware designs on the Internet for the unlikely possibility that someone might derive something useful and not reinvent the squeaky wheel. As mentioned it’s just one command to push my changes to the cloud so that’s a cheap price for forestalling queries of: can you share your files. They can stay anonymous, no record is kept of who clones my repository. And my stuff is backed up to the cloud. A friend of mine who contributes to a major software project is always chuffed to explain to others how he has an Arctic badge. So his code should survive apocalypse even if we don’t. :thinking:

Obviously publishing to a public repository doesn’t work for people who have proprietary designs but you can rent private VCS services or get an admin to set up one for you. I’ve set up several for myself and other engineers in my sysadmin career.

One thing I have learned is that what is easy for one person is likely to be difficult for another. I worked in one office where I was not competent at writing visual basic. But another engineer who was good with VB could not design a circuit that he had not seen, and was stymied because he had mis-rotated his coupled inductor by 90 degrees on the board. I was better with both of those…

1 Like

I don’t think they have found brain structures specialised for coding VB or placing inductors. They and you knew how to do these things well because of learning and practice.

I don’t think they have found brain structures to explain most of what makes us individual. When I was 13 I was exposed to amateur radio. I immediately decided that I wanted to become an EE and I also learned Morse code in 1-2 days. Looking at this objectively, it does not make any obvious sense. I had no idea what I was getting into! But the stuff fascinated me and in 57 years I have never regretted those decisions. I dare say that there are plenty of people who would not have given those pursuits a second glance based upon the exposure that I had. And for further reinforcement, read about studies of identical twins raised apart. (Where they grew up in separate families.) The sorts of preferences which are attributable to genes are truly uncanny. (Smoking the same brand of cigarettes for example) Here is one example:

I am quite sure that to a large degree, the way that I am (and this will be true for most of us) was largely (not completely) hard-wired. For me, it is transistors and diodes instead of lines of code.

1 Like

Maybe opportunities, time, obligations and interest have some bearing also?
Maybe you may have become a brilliant software developer had you never been exposed to amateur radio.