Are there any appealing deals for professional EDA packages these days?

[dimbmw]

Hi
I am wondering what is the best deal for a professional EDA package on the market now.
It appears that the EDA landscape is more competitive these days with a lot of free tools.
Are professional guy like Mentor (Pads) or Altium offering some attractive deals in response?
I am pretty impressed with the Fusion 360 in the other field (CAD/CAM), and hopefully something similar is coming up in the EDA ?
Thanks!

[ebclr]

Both Still charging a lot, as usual, is some cases even worse,  they didn't respond to the sea of amazing free tools

[dimbmw]

Altium is pretty cool, with its new features starting from 18, 19 and 20, it is good enough to be used on motherboards.
It does have a higher initial price, but with that price you get all modules, so if you plan to do a lot of advanced designing, it's cheaper than other competitors.

Most tools offer free trials. Try them. I found Altium easy to use, so I'm stuck with it, though I also use KiCAD for collabs since an Altium license is expensive.
The thing I like Altium the most is its intuitiveness. It's designed for "typical Windows users", and it feels intuitive. It does have a learning curve, but IMO it's pretty shallow unless you need to learn some of its most fancy features.

Also, there are lots of learning materials available. Depending on where you are, for instance, Altium China offers free seminars and tutorials every week. To be fair, Altium license is 2x as expensive in China than in US, partly to offset the loss from piracy, partly to cover the tutorials and other SNS activities.

I don't think Altium US offers free trainings, but if you ask, their FAE will give you an FTP address where you can find a lot of good stuffs.

Most of our designs are simple enough for free software. But I’d rather buy something if it cost some reasonable amount.

[dimbmw]

Both Still charging a lot, as usual, is some cases even worse,  they didn't respond to the sea of amazing free tools

Maybe I just need to sit on the fence for a little longer...

[Feynman]

We had a look at PADS Professional a couple of months ago. It looked like Mentor wanted to push the product and significantly lowered the price for it (cheaper than PADS Standard/Plus).
Altium was another tool we looked into, but it is hard to tell what you'll have to pay for it. Their are a lot of special deals (e. g. "switch now from Y and get X% off") and I even heard of people who got licenses almost for free because Altium wanted to get "in there".

[thm_w]

https://www.eevblog.com/forum/altium/what-is-the-current-price-of-altium-designer/
$4-8000

Unlikely OP would get a free license unless they are a professor at a university or something.

[nctnico]

Orcad (from Cadence) is competitively priced compared to Altium while both packages are pretty much equal in abilities. Orcad has several options; I strongly recommend to get the CIS option to make component management easier.

[OwO]

Unless you are designing PC motherboards I don't see why you can't be just as productive with gEDA or KiCad. Compared to "professional" PCB designers I can do a medium-complex design (SoC with DDR3/4 memory) in a similar amount of time assuming you want a decent layout. Sure you can wing it with autorouting, bus routing, push and shove routing etc and get it done faster, but if you want an optimized design (thus reduce cost for manufacturing) the bottleneck is never the PCB design tools.

[nctnico]

Unless you are designing PC motherboards I don't see why you can't be just as productive with gEDA or KiCad. Compared to "professional" PCB designers I can do a medium-complex design (SoC with DDR3/4 memory) in a similar amount of time assuming you want a decent layout. Sure you can wing it with autorouting, bus routing, push and shove routing etc and get it done faster, but if you want an optimized design (thus reduce cost for manufacturing) the bottleneck is never the PCB design tools.

The problem is that the free packages don't have support for real component databases. Being able to create a correct BOM with all the right part information is an important part of a CAD package to get the logistics streamlined (efficient and protected from component errors). The free packages are still at the stage where all the information for a component is in the schematic. That is a bad way of working. The schematic should only refer to a database entry where the component information is.

Besides that you might want to do some design verification. Orcad's Allegro has a similator for trace impedance and cross talk (both single trace and differential pairs). This is very usefull to check your design.

[OwO]

Do you mean vendor supplied footprint databases or BOM management? Footprint design doesn't account for much of my time and I don't remember the last time I had a footprint error, plus I have my own scripts for generating BGA footprints given parameters.

Same idea with BOM management, I have my own system for matching schematic part numbers to real part numbers. Schematics always specify generic part numbers, e.g. "0603 10V C0G 1nF", or for ICs "ADF4350" without suffixes, and every fab or prototype run is supplied with a "mappings" file that translates generics to detailed part numbers (very important because I substitute parts all the time); there is no way I want to be tied to a fixed part number specified on the schematic, and even vendor specific part numbers on the schematic are really just a placeholder for "an IC with these specs". The final BOM part numbers can and will differ from the schematic ones. This custom system is only possible because gEDA files are text files and are easy to parse. With vendor specific binary formats you are stuck with whatever system they provide. However, if you don't know any programming, then yes you are probably better off with a "ready out of the box" solution even if it isn't optimal...

OpenEMS has supported gEDA import for a while I think, the open source tools are certainly catching up...

[EEEnthusiast]

KiCad does come with an extensive library of components and footprints. It is very easy to make new component libraries as well.

[nctnico]

Do you mean vendor supplied footprint databases or BOM management? Footprint design doesn't account for much of my time and I don't remember the last time I had a footprint error, plus I have my own scripts for generating BGA footprints given parameters.

No. Component logistics management.

Same idea with BOM management, I have my own system for matching schematic part numbers to real part numbers. Schematics always specify generic part numbers, e.g. "0603 10V C0G 1nF", or for ICs "ADF4350" without suffixes, and every fab or prototype run is supplied with a "mappings" file that translates generics to detailed part numbers (very important because I substitute parts all the time);

Now imagine all this work is done automatically WITH error checking... For some reason having a database-based component workflow isn't very common and what professional packages can do to make things very easy isn't well known. The worst example I have come across is when I started at one of my previous employers. They where using Orcad Capture with the CIS extension and nevertheless they had made their own tool to check the netlist against the database. When I showed them that Orcad could do exactly what that tool did (and better) there was a moment of silence. The self-made tool (which -needless to say- was a big waste of time to create) never got used again.

[thm_w]

Can you explain what you mean by "check the netlist against the database".

Checking if the parts used in the schematic/pcb match the current database information?

[nctnico]

Can you explain what you mean by "check the netlist against the database".

Checking if the parts used in the schematic/pcb match the current database information?

Yes. The tool checked whether the values and footprints matched in the database. This goes wrong if two different components have the same value. For example a whole series of connectors which have a value 'Wurth 3221' for which the symbols and footprints have a different number of pins. Because you can see the number of pins on the schematic very easely there is no need to repeat that in the component value.

[james_s]

I've been impressed with the library management improvements in KiCad lately, recently I've been able to in most cases search for the exact part I was using and not have to create it myself and assign the footprint.

I don't understand why anyone would want to buy a commercial package and get locked into that ecosystem unless it offers specific features or capabilities not available in the free package. KiCad is quite good and getting better all the time, it's free and will always be free, since it's open source you don't have to worry about them pulling the stunt that Autodesk did with Eagle and perpetually extorting your money.

[Feynman]

I don't understand why anyone would want to buy a commercial package and get locked into that ecosystem unless it offers specific features or capabilities not available in the free package. KiCad is quite good and getting better all the time, it's free and will always be free, since it's open source you don't have to worry about them pulling the stunt that Autodesk did with Eagle and perpetually extorting your money.

Well, the reason is simple: Support. It might sound odd, but I can't afford to use a free package.

If you are an expensive engineer having a problem with your tool you need to be able to pick up the phone and get an answer ASAP. When you have to grind the internet for a solution the cost of the 'free' tool skyrockets in no time. And if you think about it, the costs of many commercial tools are ridiculously low compared to what an engineer's time costs.

Does this mean that every commercial tool has good support? Of course not. But support quality should be one of the top priorities when choosing a tool.

KiCAD is a decent tool, for sure. If you are a student or hobbyist. Or to be more precise: If you have little money and much time.
And now don't get too emotional please

[OwO]

False accounting. If you assign a cost to every minute of your time, then you wouldn't be talking on this forum, and you wouldn't work on personal projects that help you learn "because it costs less off the shelf". There are fringe benefits to knowing the ins and outs of your tools, being able to modify your tools as you see fit, that don't immediately translate into dollar figures but in the long run can be worth many times more than all the time you've ever spent troubleshooting your tools added together. If you expect everything to hand-hold you you won't grow as an engineer. In this day and age I consider programming essential regardless of your engineering specialty.

I have a friend who always told me "this isn't worth my time" when dealing with anything slightly frustrating or "time wasting", including on hobby projects. Fast forward 5 years and my hourly worth has grown by several times, while his ability still stayed student level.

[nctnico]

I don't understand why anyone would want to buy a commercial package and get locked into that ecosystem unless

Then look into packages like Altium and Orcad. Afterwards you don't want to use Kicad (let alone Geda/PCB). I have given KiCad a long hard look but still decided to spend somewhere north of 5k euro on Orcad. Not for the support but because what it can do.

@Owo: your analogy isn't very good. Your friend is obviously not very motivated to improve him/herself. If I would to employ you I rather have you work on a product I can sell than on tools which I can buy off-the-shelf. Creating company proprietery tools is not without it's challenges. Other people will need to be able to use & maintain it if it ends up being part of a primary process. Basically it becomes an internal product and it is easy to sink thousands of dollars worth of hours into an internal product before it can be used by everyone, supported and maintained.

I've seen this happen before as well. At one of my former employers we needed project administration software. These packages really are a dime a dozen but nevertheless the boss wanted to develop our own package which would suit our needs. To build the package we found a very talented intern who worked on it for about a year. However parallel to that we had another project which could really use someone with similar programming skills to speed it up and my intention was to have that particular intern work on that project. Long story short: the company ended up spending several k euro on the project administration software and got a delayed project.

[james_s]

I used OrCAD briefly at a job I had years ago and I was not impressed at all, granted that was more than a decade ago. I haven't yet tried Altium but that's mostly because I haven't found a need to, maybe if I were designing complex stuff I'd see more value in that. I've been using KiCad for years, mostly for hobby projects but I did design a few low volume commercial products with it and it did what I needed it to do.

If my employer is paying for it then I'll use whatever product they pay for but if it's coming out of my pocket I'm going with the FOSS solution unless it is absolutely not suitable. I don't typically put a dollar value on my time, engineering is what I do for fun, what else am I going to use the time for? While I wouldn't call myself wealthy, I already have enough money that I'm not greatly motivated by the quest for more of it.