no ground reference – is this a joke?

[Simon]

I'm just trying to do my first schematic with diptrace. I can't find a ground symbol, is this some sort of joke? Locating parts is also "interesting". Is it just me or is this software that amateurish. The tutorial is written in poor English, my guess is Chinese.

Just a crumble from a new user.

[Simon]

okay I found it eventually. But it is hidden somewhere in discrete components or something. If I understand correctly I cannot search all of the libraries for a component. I also don't get this thing where you are also choosing the PCB footprint. Surely this is subject to change, as the design progresses and you make decisions about its physical make up later on. Not having to choose to footprint at the same time would also eliminate zillions of components that you have to wade through to find anything.

[Mechatrommer]

why dont you download all the video tutorial? and sit back relax while watching it. i have the same problem as yours about the built-in libraries. i built my own

[mukymuk]

If I understand correctly I cannot search all of the libraries for a component

You can.  Hit the binoculars on the toolbar or Objects->Find Component.  The dialog has a "All libraries" option.

Even so, it's hard as hell to find what you need.  Novarm really needs to work on how they structure their libraries so it's easier to find parts.  Stuff needs to be grouped in multiple ways primarily by function--not by manufacturer--who cares, right?  And yes, ground and power need to be really obvious--probably on the toolbar.

That said, 90% of the time I don't find what I need in their library and end up making it from scratch.  I'd be willing to pay much more for a tool that made it easy to locate parts, datasheets, and create boms that could be submitted directly to the major distributors.  That's really the biggest time sink for me.

I also don't get this thing where you are also choosing the PCB footprint.

It's possible to ignore the footprint and assign it later.  Generally, that's what I do unless I know for sure what I'll be using.

[Simon]

well footprints should not enter into it until you are doing the pcb, that way when I want a resistor I put a resistor in, not have to figure out which of the 20 footprints of resistors I may need. I opened the pcb up: I had resistors 5x the size the a DIP8 and some smaller........ not clever.

I found kicad easier although I was hoping for a nicer package this time.

[Kremmen]

I'm just trying to do my first schematic with diptrace. I can't find a ground symbol, is this some sort of joke? Locating parts is also "interesting". Is it just me or is this software that amateurish. The tutorial is written in poor English, my guess is Chinese.

Just a crumble from a new user.

The library system does need a complete rethinking, the current one is - to put it politely - suboptimal. That footprints are associated with components is both good and bad, more good though. I, and i suspect most of DT users have assembled my own libraries of most used components and footprints. I at least am very particular to the kinds of symbols i want to use ( all logic in IEC only - no shovelheads for me thanks, etc ). So you can have those but then also the schematic only symbols where you decide the footprint later. That way you can have your cake and eat it too.
Overall organization of the components is almost a joke however and definitley the bit where DT shines the least. What i would like to see is a proper database with proper metadata for the components. That would enable parametric searches and then you could actually find something.
Another gripe is that there is no practical way to select footprint classes/categories based on the specific need of the board under work. When things are easy i prefer loose land patterns with extra area and when things are tight you need smaller ones. Also different land sizes for multilayer TH components vs. single layer ones. No easy way to select for those.

BTW not Chinese, try Russian. But i for one consider it more important than nationality, that the DT team actually interacts with the user base and implement the features we need. If only they would soon see the light regarding the libraries...

[Simon]

The thing is their tutorial was of a schematic with no ground which is a bit dumb as that is probably the most used symbol. There is no power symbol library. Yes their whole library system needs completely reorganizing.

[mc]

I never worry about grounds, as I just create a 'GND' net.
My gripe about using nets, is to create a net, you first have to connect two pins, and name the net/wire, before you can select the net to connect pins to (without wire where applicable).

[Kremmen]

The thing is their tutorial was of a schematic with no ground which is a bit dumb as that is probably the most used symbol. There is no power symbol library. Yes their whole library system needs completely reorganizing.

Well, in the disc_sch library (if memory serves, i am not at DT now) you do have a bunch on power and ground net symbols. They do connect across the schematic and those are what i use. I haven't needed anything else really.

[obiwanjacobi]

Oh good, it's not just me that finds the usability of these tools (not just DIP-shit) lacking.
Man, it's like going back 25 years! Imagine if Windows looked and worked like that! (well okay, maybe that is a bad example )

I think I could do better  , but I just don't have the time  Yes I'm a programmer too 

[Simon]

hm I'll have to work at it. I can always fall back to KiCAD but it's shameful that a freeware program is better made than a commercial one

[WattSekunde]

Oh good, it's not just me that finds the usability of these tools (not just DIP-shit) lacking.
Man, it's like going back 25 years! Imagine if Windows looked and worked like that! (well okay, maybe that is a bad example )

I think I could do better  , but I just don't have the time  Yes I'm a programmer too 

I totally agree and wish I could find more MacOS engineering SW in the near future. But on the other hand I know the problem with old grown SW dinosaurs. They are extrem expensive to rebuild. You need to free a lot of manpower from other work. You could lose your customer base. Some "older" developers don't like it to give up "their" babies. The management dislike the risk? And so on.... I know it from personal experience that this is a hard fight and many extra work at home. As a freshman it's nearly impossible. ;-)

[obiwanjacobi]

Yes, but could you not make these changes gradually? If you have a sound software architecture (and they probably don't  - its very hard to maintain it on a long running project) you should be able to swap out old mistakes with new mistakes

And I even see new startup open source projects copy the same crap because its too hard to think of something original? I know, certain functionality works best a certain way. Changing that would only complicate things.

There is some new innovation, for instance KiCad module assignment after you've created the schematic. Dave doesn't like it (saw his review video) but he's is stuck (on that part) because he's used to something for so long - its very hard to change yourself. But if you think about it, the schematic has nothing to do with the packaging - I always get frustrated that I have to choose packages way-up front in the design process and then have to come back to it to do it right, because your ideas change as you're doing the design.