Is there a PCB-CAD package on the market - with eternal licensing?

[SilverSolder]

I've been testing Diptrace and KiCad over the last couple of weeks (for non-commercial use, but I nevertheless don't want a package that isn't perpetual, just like the OP).   I've ended up liking Diptrace more out of the two,  but both packages have cool things in them and are worth a test drive.

[nigelwright7557]

I've been testing Diptrace and KiCad over the last couple of weeks (for non-commercial use, but I nevertheless don't want a package that isn't perpetual, just like the OP).   I've ended up liking Diptrace more out of the two,  but both packages have cool things in them and are worth a test drive.

Its pretty time consuming to test all the pcbcad packages.
I usually reckon at least 3 days per package so  I really get into it.
Just to find on the third day  I find a deal breaker.

Its not like choosing a colour of something, it takes some serious pondering.

[SilverSolder]

I've been testing Diptrace and KiCad over the last couple of weeks (for non-commercial use, but I nevertheless don't want a package that isn't perpetual, just like the OP).   I've ended up liking Diptrace more out of the two,  but both packages have cool things in them and are worth a test drive.

Its pretty time consuming to test all the pcbcad packages.
I usually reckon at least 3 days per package so  I really get into it.
Just to find on the third day  I find a deal breaker.

Its not like choosing a colour of something, it takes some serious pondering.

My approach was crude:  jump right in and design a small board and see how it goes.   -  Diptrace was remarkable by how rarely I had to hit the Internet looking for help.  I submitted the design to the manufacturer after about three full days invested in learning by doing, sticking to only "the minimum necessary".

For this project (syncronous motor driver), it was necessary to create three custom designed components:  the 2n6666 transistor wasn't in the library, the TDA5144 wasn't there, and the motor itself was also created as a component just for fun.  There were some complications to solve, like dealing with tree grounds and a star ground - that took most of the time (the star ground was done with the help of the zero ohm resistor in the schematic).






I found KiCad much harder work.  That made the decision for me, because - as you know - there are only so many hours in a day! 

[RFdesigner]

Have to say, things are a bit rough here at times, haha.

Frankly, I don't mind reading a bit about this pcbcad720 product. I mean we hear enough about altium and the others, why not this.

In fact, while I have been slowly been getting up to speed on kicad, I took a just took a few minutes to look at the pcbcad youtube vids. And I have to say I am impressed this one guy wrote all this. Bravo dude!

As far as user interface, sure its a bit old school (like me maybe) but its simple and didnt look too much worse than the ($100'sk+ ) cadence chip design software Ive used for 25 years.

Perhaps this one man band can have his own thread here somewhere to help us come up to speed on how to use his product. My view is we should help out the small guy when we can.

Just my opinion.

[PlainName]

My view is we should help out the small guy when we can.

I believe most of us would be happy to do that. Also, those that aren't a fan of PCBCAD720 would be happy to ignore it entirely once having experienced it.

The problem that we're seeing here is simply that the chap is forever rubbishing other products and banging on about how his is far better. You surely can't have missed the $20K given the the Kicad chaps, or the more than 7K bugs (that aren't bugs), or how not placing parts exactly how he personally prefers them in a one-off operation is a massive bug, etc.

If the guy would just shut up with the dissing and stick to promoting his stuff as it really is, hardly anyone else would be saying anything.

[Simon]

Apparently PCB CAD is no longer available so i guess he gave up in the end. The problem with one man band software is that if anything happens to the author that is the end of your software. This is the first cost overhead that any company of more than 1 person will charge you for anything. The fact that they have a system that they work to and maintain so that more than one person can work on the same project or pick up from another person. Yes it's still not possible to make this totally seamless and is something I am struggling with at work (on their behalf) being the only electrical person but most companies "get it" and to solve it have to spend money and time dealing with it but people will pay for that as they will want to know their investment is safe. That said KiCad has had it's ups and downs and I think this has been due to the individuals involved, at the moment they are in a good place as the project lead is being employed full time to work on the project by a private company. I assume they figured that for what they would spend on a commercial package that they may loose access to the moment they stop paying they could strengthen the open source project and have a software package with less strings attached.

[scopeman]

SprintLayout6.0 as long as you don't need an autorouter and can live with 4 layers you can't beat it for the money. Updates are free, cost is minimal. no dongle. The easiest PCB CAD software I have ever used and I have used or tried nearly all of them.

I have used it for 10+ years after my dongled TangoPCB Series II+ would not work on modern operating systems.

Worth a look and a free fully functional demo (except save) is available.

https://www.electronic-software-shop.com/lng/en/electronic-software/sprint-layout-60.html?language=en

Sam
W3OHM

[Cerebus]

Out of curiosity I had a quick look at that. One huge drawback if I've understood it properly - no schematic capture, just direct PCB layout. That might be fine, even better possibly, for a simple adapter board with a few connectors on, but for anything more complex, no way. I can easily check a well laid out schematic by eye, but checking that everything that should be connected is connected without starting from a schematic - dead in the water on anything with more than 10-20 nets. Anything that you might save on the software you'll spend on re-spinning boards.

[Simon]

Out of curiosity I had a quick look at that. One huge drawback if I've understood it properly - no schematic capture, just direct PCB layout. That might be fine, even better possibly, for a simple adapter board with a few connectors on, but for anything more complex, no way. I can easily check a well laid out schematic by eye, but checking that everything that should be connected is connected without starting from a schematic - dead in the water on anything with more than 10-20 nets. Anything that you might save on the software you'll spend on re-spinning boards.

If I remember rightly I bought it once and then demanded a refund as they are two separate programs with no interaction, so just a digital drawing board. It's like my 3D cad software at work not being able to produce drawings of models and us having to then draw them whilst copying the model with no linked updating.

[scopeman]

I looked at the PCBCAD720 product and downloaded and installed the demo as well. The program reminded me of some CAD software that I used on dedicated workstations in the 80's and 90's (Intergraph?). Not at all intuitive or easy to use IMHO, but then again what would one expect for the cost of a couple of warm beers. Probably ok for entry level and someone that made a board or so a month for hobby work.

I don't mean to offend and I am sure that a lot of effort went into the PCBCAD720 program, but I could tell pretty quickly that it was not the tool for me, but as they say your mileage may vary and it may well be to your liking, but it was not my cup of tea.

In general if I can't get comfortable with CAD software in well under an hour, two at the most, I lose interest pretty quickly. This is coming from a person who suffered through learning the first DOS releases of AutoCAD where it took quite a long time to be somewhat proficient. The disadvantage then is there was no real competition so you had to learn what was available, obviously not the case today.

For PCB CAD programs I find any of them that force you to do a schematic first to be particularly painful as they don't like to work the way that I do.

The reason I say that is as an analog and RF designer I really have no need for an auto router and fancy bells and whistles that I will never use so forcing me to do the schematic in CAD first sometimes dose not work for me for several reasons.

I generally will either work the schematic out of my head or or just on a quadrille pad (my version of DaveCAD) and work the PCB layout from there. I may enter it in OrCAD (as I know how to use that tool) and if the PCB CAD tool can connect with it, I may use that feature but I would bet that in my 40+ career time I could count on one hand where I have used such a feature. In almost every case it was when I had an outside contractor or department who was doing a PCB layout under my supervision.

The last one I remember was when I used OrCAD for the schematic and the outside contractor used Mentor Graphics Expedition PCB back in 2007ish time frame.

This was a fairly complex 4 layer mixed signal PCB where I needed to contract out the layout work as I had other design work to do on the project along with a very short timeline with limited internal resources. If Sprint had been available then (I don't know if it was, as I had not heard of it) I could have used that tool internally.

Years later for fun I imported the 274-X Gerbers and the drill files from that project into Sprint and created files for a fully editable PCB in that software, so I am sure it could have been done with that tool. I did find that the 274-X Gerber import to PCB being very useful later in that project for making special test fixture boards that could easily adapt to what became a product that we produced unchanged for over 13 years, several lifetimes for an electronics design. I can thank Microchip for (still) producing the microprocessor used in that design to keeping that product version alive. We recently had to redesign it due to obsolescence of other components but that as they say, is another story.

Sure there are rare times when I had a digital design that had repetitive work to do, where an auto router could have helped, but for the most part copy and paste can be your friend too. I guess that is why I liked the prior TangoPCB Series II+ DOS and later the SprintLayout6.0 as I found it very easy to do manual layouts without any preliminary restrictions.

I have done quite a few very complex mixed signal/mixed technology PCB's with this software without the need for linkage between the PCB and the schematic. They all worked and most of those designs are still in production today, many for well over a decade.

Sure not having that linkage does require additional manual checking, but it does force you to become much better at layout. This is because you will inherently find yourself cross checking your layout as you go, rather than relying on a software tool to do the due diligence and then only find out that you painted yourself in a corner and left out a critical trace or component that at the last moment requires significant rework.

Perhaps as one who started out when PCB layout was done manually on a drafting board I still somewhat have an aversion to auto routers. Maybe I'll come around to one someday but I don't think for the type of work I generally do that having one is a necessity. 

The one thing that would be nice for them to implement in Sprint would at least be a netlist in/out feature. I have pressed for this, but I guess the 70K users of the program have not found it to be on their list of wanted features. However I have found the company to be very responsive to suggestions from their users, as they have implemented many of my suggestions so I do expect to see incremental enhancements in the future.

After using more complex PCB software at my many day jobs at several companies as an EE (P-CAD, OrCAD, Altium, Protel, VeriBest, etc.)
I just found that the simpler Sprint product worked very well for me and even though the last company I worked for bought Altium I found it to have a very steep learning curve (not to mention the cost). I reserved it for PCB designs that required that level of performance and I still used Sprint for boards that I needed to crank out quickly.

It was just a matter of what was needed to get the job completed at the time and also one tends to use the tools which one is most familiar at crunch time. 

It still does amaze me what one can do with low cost entry level software. IIRC Altium was 140 times the cost of the Sprint product. Altium is clearly not a tool within the financial means of most hobbyists. Maybe if they offered a low cost not for profit version they would get more traction and those who used it for hobby work would be more likely to recommend the product to their employers for professional use.

I also have tried KiCAD but as of today I am not very comfortable with it. Maybe someday I'll need to dig into it, but for anything that I have currently planned on the PCB project list I think I am covered.

Well sorry if I rambled on a bit here, but I hope that my ramblings might be useful to some of you.

All the best,

Sam
W3OHM

[SilverSolder]

I've been testing Diptrace and KiCad over the last couple of weeks (for non-commercial use, but I nevertheless don't want a package that isn't perpetual, just like the OP).   I've ended up liking Diptrace more out of the two,  but both packages have cool things in them and are worth a test drive.

Its pretty time consuming to test all the pcbcad packages.
I usually reckon at least 3 days per package so  I really get into it.
Just to find on the third day  I find a deal breaker.

Its not like choosing a colour of something, it takes some serious pondering.

My approach was crude:  jump right in and design a small board and see how it goes.   -  Diptrace was remarkable by how rarely I had to hit the Internet looking for help.  I submitted the design to the manufacturer after about three full days invested in learning by doing, sticking to only "the minimum necessary".

For this project (syncronous motor driver), it was necessary to create three custom designed components:  the 2n6666 transistor wasn't in the library, the TDA5144 wasn't there, and the motor itself was also created as a component just for fun.  There were some complications to solve, like dealing with tree grounds and a star ground - that took most of the time (the star ground was done with the help of the zero ohm resistor in the schematic).






I found KiCad much harder work.  That made the decision for me, because - as you know - there are only so many hours in a day! 

Just to follow up / close my comment:  the assembled board worked the first time, with not a single thing departing from expectations.  Of course, it is a relatively simple board in the bigger scheme of things, but that still doesn't stop it being a hole-in-one!