4 ms·

Same time schematic & PCB editing... for the first time ever... like in the world... first. time. ever.... this is a pretty big deal. In Eagle, you cannot add

Same time schematic & PCB editing... for the first time ever... like in the world... first. time. ever.... this is a pretty big deal. In Eagle, you cannot add new parts directly to the PCB (no layout duplication for panelization). Panelization is about the only place I have ever wanted that feature, as the abstract schematic is much nicer for connecting nodes. Although, I guess since I haven't tried it, I could be wrong. I'm not entirely sure why there is that limitation, but it would be a pretty nice feature. Couple other points: Full real-time design rule checking Eagle does not do real time DRC. Although, as I've yet to really have problems with a DRC (most PCB milling machines are awesome, and you don't really have anything to worry about). A system of automated footprint generators based on the IPC standards That would be way better if I could supply a PDF from the data sheet, your software looks at the dimensions and gives me the footprint back.

mindslight 14y ago
Yeah, trying to make changes to the circuit via the PCB is annoying in Eagle. It hasn't been a big issue for me, as the times when I've wanted to do it, I've wanted to do so much of it that it was worth doing that part of the layout using scripts (imagine a 4 signal layer PCB with the same functional unit duplicated 24 times, that wants to be as small as possible. adjust dimensions, autoplace, autoroute, evaluate, repeat). But I can see that kind of feature being quite handy when you have just a few copies to make. Eh? Don't know why you'd ever not worry about DRC. If the PCB fab is that good, make your features smaller :P. You also want to always make sure that you're not violating your electrically-significant constraints. And yeah, footprints are quite annoying (especially when starting out) and one of the things that I think a widely-shared online approach like Upverter will actually succeed in fixing for the common cases.
- mercuryrising 14y ago
  Did you do the scripting in Eagle? I usually just clone things a schematic building block, although that means that I have to redo the layout for each one. I haven't made any boards for manufacturing, only one off boards. I like to keep things spaced out so if I need to get in and cut/jumper I can. Soon though, I'll make a board for manufacture. Then the DRC come into play. I dislike the uncertainty in the footprints. I just sent a board out with a footprint I made, and I didn't have the parts yet to see if they would fit the dimensions. It's all a big guessing game. The Sparkfun library is pretty awesome in that regard (sharing files). I like how they say 'Production tested', keeping quality up when tons of people on Upverter are making footprints may be hard, although if you have a hundred or so 'tested' footprints you might be in the clear.
  - mindslight 14y ago
    I started off writing ULPs, but eventually moved on to using a general purpose language to generate a ULP that consisted of the actions to be performed. And to get the data out, the registered version has an option to export to such a script, too (and you can write a ULP to do the same exporting in the unregistered version). I've done mostly prototype work, but rework isn't much harder with tighter design rules. Also keep in mind a good amount of possible cutting/jumpering/airwiring can be avoided with the right use of infinite and zero ohm resistors at layout time. I unfortunately haven't touched embedded design in a while - moved on to software for more ambitious goals; unfortunately, das blinkenlights aren't so applicable to the problems of the world that are important to me.

mindslight 14y ago

Yeah, trying to make changes to the circuit via the PCB is annoying in Eagle. It hasn't been a big issue for me, as the times when I've wanted to do it, I've wanted to do so much of it that it was worth doing that part of the layout using scripts (imagine a 4 signal layer PCB with the same functional unit duplicated 24 times, that wants to be as small as possible. adjust dimensions, autoplace, autoroute, evaluate, repeat). But I can see that kind of feature being quite handy when you have just a few copies to make. Eh? Don't know why you'd ever not worry about DRC. If the PCB fab is that good, make your features smaller :P. You also want to always make sure that you're not violating your electrically-significant constraints. And yeah, footprints are quite annoying (especially when starting out) and one of the things that I think a widely-shared online approach like Upverter will actually succeed in fixing for the common cases.

mercuryrising 14y ago
Did you do the scripting in Eagle? I usually just clone things a schematic building block, although that means that I have to redo the layout for each one. I haven't made any boards for manufacturing, only one off boards. I like to keep things spaced out so if I need to get in and cut/jumper I can. Soon though, I'll make a board for manufacture. Then the DRC come into play. I dislike the uncertainty in the footprints. I just sent a board out with a footprint I made, and I didn't have the parts yet to see if they would fit the dimensions. It's all a big guessing game. The Sparkfun library is pretty awesome in that regard (sharing files). I like how they say 'Production tested', keeping quality up when tons of people on Upverter are making footprints may be hard, although if you have a hundred or so 'tested' footprints you might be in the clear.
- mindslight 14y ago
  I started off writing ULPs, but eventually moved on to using a general purpose language to generate a ULP that consisted of the actions to be performed. And to get the data out, the registered version has an option to export to such a script, too (and you can write a ULP to do the same exporting in the unregistered version). I've done mostly prototype work, but rework isn't much harder with tighter design rules. Also keep in mind a good amount of possible cutting/jumpering/airwiring can be avoided with the right use of infinite and zero ohm resistors at layout time. I unfortunately haven't touched embedded design in a while - moved on to software for more ambitious goals; unfortunately, das blinkenlights aren't so applicable to the problems of the world that are important to me.