Fritzing is somewhat hard to use

OK, I just tried to duplicate a KiCAD project I put together recently. Creating the schematic and breadboard views were not too hard. The UI does need some serious work, partitularly when disambiguating what a mouse click refers to (one needs to be overly precise when drawing wires in the schematic view – there is ambigiousness between moving parts and drawing wires).

The PCB design view is a total disaster. The circuit I designed uses three I2C breakouts from Adafruit connected to a TTGO-T1 (an ESP32 Devkit clone, available from China for $5/each). There are exactly 4 nets: SDA, SCL, +3v3, and GND. My KiCad design uses copper fill areas for the power and ground busses: the top layer for +3v3 and the bottom layer for GND. There are really only two “routed” nets: SDA and SCL. It was an excessive amout of effort to get those two runs done. Then I discovered that it does not appear to possible to do a copper fill area for power on one side (top) and a ground fill area for the other side (bottom). At that point I gave up. I have designed far more complex boards with KiCAD – Fritzing is far harder to use than KiCAD. The only thing Fritzing brings to the table is the Breadboard view. (And the fact that Adafruit has a zillion parts as Fritzing objects, but I think I will write a Tcl/Tk program to convert Fritzing fxpz files to KiCAD .lib (schematic) and .mod (footprint) files.

You are going in the wrong direction. You don’t go from an advanced EDA like KiCad to a beginner program like FZ. You start with the easy one to get EDA experience, and when you need more features you go to the complicated one. FZ is mainly for beginners to go from BB to manufactured PCB with nothing too fancy, unless you know how to work around it’s limitations. Also development has stalled, so advanced stuff might be far away if they ever get added.

I think you can do something like that, in a fudging kind-of way, but it’s not going to be as easy as something that is designed to do it.

Actually, I think FZ might not even be of much use to beginners. It is like a toy bicycle with rusted training wheels. With flat tires. At least until it gets a lot more developement… I’d just as soon just start beginners off with KiCAD.

1 Like

I’d say many beginners in here and in the wider community wouldn’t agree. I’d suggest that both adafruit and sparkfun having large repositories of Fritzing parts (but not as far as I know any kicad parts) speaks for itself, their customer base appear to like Fritizng better than kicad. I came here via egda which appeared 4 or so years ago to be dieing much like Fritzing has been, with armed camps among the developers having flame wars rather than developing and a code base in scheme where the support for the language was dieing which Fritzing doesn’t have, along with currently not many developers, to kicad which didn’t suit me to Fritzing which does. If you are willing to spend some time learning it Fritzing mostly works fine if you don’t want more than 2 layer boards (which many of our users don’t). If it doesn’t work for you by all means use one of the others, there are lots of them. It has bugs and more importantly was not finished before the research funding ran out, so isn’t complete. The change to open source has not so far been that successful, but that may be changing. Parts creation is currently a mess but is being worked on, but even at that once you are past the learning curve parts creation isn’t that bad (although it certainly could be made easier.) That will only happen if a bunch of someones step up and make it happen. We will see if that becomes true over time.

Peter

FZ is based on “copy what you see”, and there is nothing simpler for children. That’s what the BB view is all about.

Off the top of my head I think you can do split planes by copper filling the top layer and connecting the parts to the copper fill(red spots) with traces. Or copper fill the top layer with the 3.3V traces still ratsnests, and then just put the traces over the copper fill.

Ether way if you know KC you should stick with it, if you want to do advanced things in the future.
I have to use KC sometimes, but I don’t think I will ever need it so I’m not going to put much time into it.

I mean no offense and perhaps shouldn’t post this comment. But, having used Fritzing for a couple of years now and feeling somewhat of a ‘Community’ member…

In Defense of Fritzing (and the hard work gone into it) focused on the prime target audience…

It’s fair to say that most older folks, with some ‘real-world’ experience know how to accomplish their design task with the minimal of resources - we didn’t have computers in late 50’s, early 60’s.

I use Fritzing because I don’t need an Excel spreadsheet to balance my checkbook, I certainly don’t need Eagle, KiCAD … other apps to design my Circuits. I tried darn-near all of the app’s out there. I’ll stick with Fritziing. OMG! Talk about ‘Wanting to be Everything for Everyone’… Those other apps have a bazillion parts… and complicated User Interface…

For someone (Robert Heller) that referred in another post to not liking “Eye Candy” or IDE’s, suggesting a preference for apps without all the bells and whistles (eye candy), it seems a bit ironic to rail about Fritzing that gears itself as a minimal but, useful app.

For 1 or 2 layer boards, it’s a no-brainer and the ‘Eye Candy’ of Resistor color, part-shape, gizmo-parts… etc, is just that; “Eye Candy”.

Designing a circuit using only the ‘Hole part’ and ‘Copper-Image part’ is pretty much all an experienced engineer (that doesn’t like Eye-Candy) needs. Adding a graphic part shape to Silkscreen is simple, as is adding Power and Ground Planes.

If you want a Power-Plane and/or and Ground-Plane, simply make your own - as shown below…

I’ll leave it there and, again, I mean no offense to anyone…

planes.fzz (3.4 KB)

I at least am not taking offense, only pointing out that there is a large (estimated, from I think the github parts update statistics, at some 200K+ users). There are enough people using Fritzing to make pcbs via the Anseler pcb fab that the Anseler folks arranged (and possibly funded!) the new code maintainer who is restarting development. They appear to see it as a good business move. There is scheduled to be a new release (with a few changes, but mostly to test the build and distribution system, which like everything else needed an update) in a few months. Someone on Hackaday posted a (in my opinion) biased rant against Fritzing and coined the phrase “friends don’t let friends use Fritzing” which I think has done Fritzing a lot of damage. I agree with one of the first comments, that he had obviously decided Eagle was the answer and trashed everything else he reviewed rather than really tried any of them for any length of time. Closed mind, not worth responding to when someone had already done it. Fritzing works fine for lots of folks, I’m interested in making it better. It certainly needs it, and I think is worth it, the underlying code and design have lot of good design and consideration of how things should expand in the future, it just never got finished before the research funding ran out and never really caught on as open source (probably partly due to biased rants, which turned off potential developers without them even looking at the code which is admittadly difficult to read because of no documentation) which I find a pity.

Peter

‘Experiences’ are concise in their descriptions and meanings. Rusty training wheels…:face_with_raised_eyebrow:

SO from what I understand, Fritzing, ‘DOES NOT, fit your standard’, we hear that.

What can be done? Git-hub, fork repo, learn QT, contribute. What more can I say? It’s not hard for the savvy… I will admit, Like anything though, to become better, it probably needs alot more development time. BUT, does that make it unusable? Far from it actually…

Fritzing is an easy way to accommodate everyone from beginner to advance user. Learning curves are different for everyone, for every software package…(life in general)

So it’s all apples to oranges.

EVERYONE is entitled to their opinions, but you know what they say about opinions… haha everyone has one…BUT I will say an import/export to/from KiCAD feature would be A+

Thanks for giving it a go Robert!

The other way you might be able to do it is have a normal sketch, and one with 3.3V and GND reversed. Then just gerber both and swap the top layer file.

I think a big part of the problems I am having is in how the graphics is working. Either QT is a very slow library (seems unlikely) or Fritzing is not using it effiently (very possible). I am running Fritzing on a Raspberry Pi, but using my laptop as a display using a SSH X11 tunnel. (And no, it is not the SSH X11 tunnel that is slow – I do the same thing with KiCAD, the Arduino IDE, JMRI, and other X11 programs and everything else runs fast enough. The Pi does have a 100mbit Ethernet and I have a GBit switch – the network is itself quite fast enough.)

And maybe it is the version I have on the 'Pi: 0.9.0 (it was what was in the repository for Raspbian).

Another problem for me is that I don’t have a good Internet connection at home – I am on dial-up (yes, really!), because that it all that is presently available where I live, here in Western Mass. In order to do any development work with Fritzing, I will have to sneaker-net about 11meg of .deb files (for the 'Pi – my x86_64 laptop is running CentOS 6 – the libraries, etc. are too old and yes, I need to update it, but again, that is a complicated process, complete with the complication of poor Internet connectiveity). Also , I have not done any GUI programming in C++ in a very long time and have never used QT before, so there will be a pile of learning there.

This topic came up before, there is a PI version of the current release (0.9.3.b) on one of the PI repos. You would be better to run 0.9.3.b than 9.0 (which I don’t think has parts manager). It should be available here:

https://www.raspberrypi.org/forums/viewtopic.php?t=175430

I hate to think what a Qt download would be like on dial up, on ADSL it takes 8 or 9 hours.Fritzing will get slow if you are using a higher resolution that the default .1in in views. It usually takes about .001 in to really slow down my I7 but it can. Part of the problem is all the parts are svg and it needs to render all of them.

Peter

It looks like that is not going to be easy to do, unless I do a massive upgrade of the 'pi to Stretch (at least all of the libqt packages). I might be better off trying to install it on my BeagleBoneBlack… First I will try building from source and see how that goes.

OK, am going to take fruitloops (my workbench / build box Raspberry Pi) and snoopy (my BeagleBoneBlack) to library tomorrow and use the high speed Internet (all of 20mbits) to download stuff (maybe 0.9.3B on the Beagle, and the libraries to build it on the 'Pi)…

Oh man… maybe kinda off subject… But I have a pretty big project based around a BBBW. You want to run fritzing on this thing? The BBB is just painfully slooooow… I mean it works and serves its purpose. BUT ouch… The Pi is a bit better… But still ouch. Good luck at the library!

I am interested to get my hands on the newer BeagleBone AI… That seems to have a bit more promise…



It is just that BBB is running Debian 9 (Stretch), which seems to be what Fritzing 0.9.3b wants and the Pi is running Debian 8 (Jessie), which means I am limited to 0.9.0, unless I can build 0.9.3b from source.

Are you going to compile these sources on these actual devices? Say versus cross-compiling on a something that will do it in a fraction of the time…?? 24 cores plows threw a compile. BUT For a laugh, I compiled source for a different project that goes with that BBBW-head Unit (unit pictured above). It took days… Amusing to say the least…

I don’t have a 24 core machine. And my 4 core machine runs CentOS 6.

Oh man, this is going to be a full on experience. I do not envy you my man! BUT Should be fun as I am personally curious how long it will take per device. I saw some compatibility issues from BBB to Pi on my project, I to compile a version for each device.

It took me about 3 days to build a cross compiler for embed ARM. And most of a day to build a cross compiler for the ESP32.

BTW: what do I need to tell firefox NOT to grab focus when someone posts a reply. It is excessively anoying.

Next to your home button in the upper left. little i-icon with a circle. click site information for forum.fritzing.com -permissions- X allowed. should shut it up… OR selectively on the bottom of the post, Click on normal and change to mute.