What real schematic design software should I learn for small run manufacturing?

[Simon]

and it doesn't take the latest quadcore processor to use it.

No software should unless it is a) written by an idiot or b) doing hard core calculations, I can understand autorouting taking lots of power but anything else in any electronics CAD software should be low usage.

You're very quick to declare people idiots for being unable to achieve something you haven't attempted.

If anyone writes software in excruciatingly slow flash then they are an idiot, I don't need to try it to know, I've suffered the consequences.

[Simon]

in fact 15 years ago i used OrCAD on a 400MHz machine and it only too 5-10 minutes to complete, try running some peoples software on a 1GHz machine, if it was written with something lazy the machine is rendered useless.

[John Coloccia]

I actually really like DipTrace for basic schematic/PCB design work. It's very simple to use, and so far I'm using it for all of my products. It's really lacking in many of the features/interactions that you expect from software in a production environment, though. For example, it's very easy to make new patters/components, but then managing them, propagating changes, is complete crap. The library management is crap. Their "grid", especially as it pertains to switching between metric and imperial, is complete and utter crap. There's a lot of crap that pisses me off about it.

But all that said, I use it and really like it because the core process of creating a schematic, laying out boards, making my gerbers, etc is a joy to use. If I were doing anything moderately complex, or working in a larger environment, I probably wouldn't consider it, I'm afraid. I'll bet it would work very well for a large percentage of the work that most of us do.

[Simon]

they made a good effort, but the last version came out riding on the addition of 3D model exprt. In this version they claimed to have improved library management (where ?) added differential pair routing (don't use it) and added support for the latest gerber format (who gives a toss) the actual software that we all agree needs work on has remained untouched. I have had enough of it, I'm feeling taken for an idiot. The autorouting with copper pour is mad, if I choose that a net connected copper pour is a keep out area (because I want to preserve it as a heat or current carrier) even the tracks from the net it's connected to fail to connect to it so autorouting fails. If you allow traces through it just cuts the pour up and it was pointless. Completely mad, yet they refuse to see a problem.

I know many people claim that you should not use an autorouter but I find it very useful. My boards are not overly complex, they don't have high frequencies etc and I lay components out in a sensible way so that the autorouter can do a decent job, I will do some manual edits (because their autorouter is shit) but on the whole the autorouter can be trusted to connect all those components that are right next to each other or have an obvious path and save me hours.

[John Coloccia]

I can't argue with anything you've said. If you've read the boards over at DipTrace, I'm sure you've read my own frustrations at the garbage and their refusal to fix anything. I have absolutely zero intention of upgrading to the new version. I haven't even tried the beta. They won't see another dime until they start fixing the problems, and I am actively looking for a different solution. I can name SO many problems with DipTrace, but the fact that I keep using it sort of speaks to how crappy everything else is!

I'm going to check out KiCad again now that they're getting close to some sort of release. Eagle is just a non-starter, IMHO. I've tried it a couple of times and it's just a steaming pile of dog turd. Maybe KiCad will finally win me over.