Classic GUI Interface Feature


To be honest, I would prefer a more traditional GUI, similar to what was available in MicroCap 12.

A lot of people will have trouble accepting the new GUI approach you have created, and will likely not migrate from LTSpice for that reason.

How difficult would it be to offer a classic GUI interface as an option to be selected? Is this a major modification to QSpice program code or a minor tweak?

For a long time there was a program called “Classic Shell” that gave users the opportunity to make their desktop look more like a traditional Windows environment.

In a similar way, I think there are good reasons why the GUIs used in other CAD programs have not changed much over the last 20 years.


This text will be hidden

QSPICE GUI simple and very easy for study.
I personally hate GUI with hundreds buttons like MC12.

1 Like


The request was to make the Classic GUI as an option to be selected. If you like the simpler GUI as is, you would not have to engage the classic GUI.

I understand many like the buttons and are very vocal about it. I can offer my sympathy.

But entertaining two GUIs is very unappealing. I try to just look at the fundamentals, design a GUI on those, implement that and see how that GUI feels. At this point I can’t go backward in time to the toolbars.

But you have my empathy.


1 Like

What do you mean about “Classic GUI”? MC12?
There a lot of GUI that can be named “Classic”. LTspice, OrCAD (DOS, Windows)…
Most known GUI probably is MS Word. Should Mike use MS design?

Theoretically you can write own GUI and call QSPICE64.exe for simulation.

1 Like

I think the possibility of a small, customizable toolbar (perhaps just with the option of enabling the components - or the tools that the shortcut keys point to) is interesting. But I understand that the proposal is disruptive.

My big problem with toolbar buttons is a matter of fundamentals:

  1. They force the user to move the mouse all the
    way over to the button, press it and then move
    it all the way back.
  2. The user has to shift their eyes away from the
    content of the simulation.
  3. Toolbar button take up screen space that is
    better used for a bigger view of the content
    of the simulation.

When I watch a expert user of some application with lots of GUI with toolbar buttons, they are energetically swatting at the the computer and darting their eyes all over.

This view point has been reenforced by the feel I get when I use toolbar buttons vs context menu commands.


You can observe a lot just by looking.



I think the issue is more about user choice and user personal preference rather than one person deciding what is uniformally “best” or “most productive” for everyone.

It’s one of the reasons Windows gives you “six-ways-from-Sunday” for doing the exact same thing.

People then customize their desktop experience so that it works best for them.

For me, I usually turn off/remove the excessive tool bar clutter, if I don’t need it.

Windows PC’s are called “personal computers” for a reason…

I think the issue is more about user choice and user
personal preference rather than one person deciding
what is uniformally “best” or “most productive” for

Well, that is certainly the sort of sentiment that everyone would like to be able to agree on, but I’ve had life experiences that have taught me that GUI design has little to do with users’ preconceived preferences.

By the end of the 1990s, I had been writing charged particle optic simulators for some while on a large Sun cluster. I moved on from that secretive(aside from the occasional peer-reviewed article when asked for for hard core technical promotions) work to do LTspice because I had had enough contact with the original Berkeley CAD developers to know I could do a better SPICE.

But I didn’t know if I was going to be able to type it in. A decade of EMACS Ctrl- commands had done enough damage to may hands I didn’t know if I could type any more or not. Fortunately the Windows GUI for text was better than that under UNIX and didn’t need Ctrl- commands.

Ergonomics are more important than my(or anyone else’s) preferences.

The other reinforcement that ergonomics trumps preconceived user preferences was the study I did of CAD users during the design of the LTspice GUI. They had weird vertical mice or some other strange thing because CAD software had done so much damage to them. I was required to make a GUI that did less damage. The only reason LTspice had a toolbar was to give it some visual interest in a magazine placement. Everyone knew not to use it.

When it comes to GUI design, one needs to look at the fundamentals. Set preferences aside. More often than not, preferences are just baggage some some older GUI.

Yes, it goes against the flow to discard, with passion, any preconceived ideas of what a GUI should like like. And it’s not fun giving people what is good for them instead of what that want. But, ultimately people don’t know what they want until you show it to them. That is why Micro-Cap/Spectrum Soft is out of business and I am not.




I had no clue that biometric / health issues would
intrude into this topic of discussion.

At the risk of appearing “obstinate” in the face of all
argument, let me offer the following response.

If the argument against a GUI interface with tool bars is HEALTH DRIVEN, then the CAD software companies that offer products with extensive tool bar menus should all be held liable in a class action lawsuit for creating “injurious products”.

Clearly, some health warning should have been posted as to “the dangers of using tool bar menus”.

Please understand: I am 66 years old. I have been using
Windows and CAD software for a very long time.

However, I do not have issues with:

1)  Carpal Tunnel Syndrome
2)  Damaged joints
3)  Damaged cartiledge
4)  Nerve damage
5)  Arthritic pain
6)  Any other medical conditions impacting 
      finger, hand, wrist, elbow, or other joint 
      or nerve conditions limiting my ability to 
      use a computer.

It could be argued that I have been “lucky”.

However I would also attribute my lack of “damage”
to proper nutritional support, allowing my body to repair itself from the normal wear and tear issues associated with living over many years while doing computer work.

So, for me, arguing that tool bar menus should be eliminated on “Health Grounds” is a non-starter.

I make my own health decisions when it comes to using a computer. I take responsibility for my own health, rather than put the blame on software or computer hardware manufacturers.

Again: At issue here is personal choice and autonomy in making GUI setup decisions rather than having them pre-made by someone else who feels they “know better”.

If a software manufacturer needs to display a health
warning statement as to the risks involved in engaging “tool bar features”, then so be it.


For me the one thing that is the most inconvenient is when I have to enter expressions in the waveform viewer. If I could just click one node, drag, press + over another node, and this would add the nodes in the waveform viewer, this would save me a lot of time. Same with + - * / buttons. This would even eliminate the need to implement some things on the schematic reducing circuit nodes and complications, loop gain probes etc. What if I could drag a trace name onto another and hit * to multiply them in the waveform viewer? This would save numerous clicks, no need to open editor dialogs to do something simple.

My other problems with Qspice are the click points for wires and components are so small it takes too much concentration to work with a schematic. Having to use the shift key to select things has always been confusing to me in any program, and the logic for what gets selected and dragged isn’t as useful as in LTspice where I could freely rearrange parts of the schematic fairly quickly.

In the end I appreciate an interface that minimizes the number of clicks and eye movements, but doesn’t force me to use keyboard shortcuts (I forget the ones I’m not constantly using, or if I start using another CAD program).

You can click on node 1 and then click on ALT + click on node 2. Then you have a differential between two nodes.
Best regards

I know about that, but I keep missing the wires, I have to be pixel perfect.

Have you consider .plot directive, which can help to save many clicks by controlling what to plot in waveform viewer after simulation is run. But for .plot in effect, waveform viewer must be closed before simulation is run.

1 Like

I see, that is useful. But it only helps if I restart the simulation or if I already know what I’m looking for. As an example, in a given circuit I need to find the response of a given stage. I also have a specific set of wires that are considered the input and another specific set that is considered the output of that stage. These can change freely depending on what it is I need to know. I cannot simply define all my plots ahead of time, this would give me very little insight.

If we’re talking about baggage from old interfaces, I would say the process of constructing expressions in the waveform editor leads to many bad habits, the main one being that most people avoid it at all costs. Freely constructing waveform expressions opens many doors to gaining insight into a circuit.

Hello, I just love the new context menu approach, it’s great. Speaking of the contextual approach, why does the context help popup text change to the entered values when entering values. For example, when you edit the .tran you get your values and your values again in the context help, which is ironically not that helpful at that point. In my opinion, the context help should show the field labels so you know which one you are editing, not the entered values that are already on screen.

It gives you the hint of the things you need to type in and not hints on things you already typed in. If you want to revisit the hints, Ctrl-X the you part you want to revisit and then Ctrl-V to put your input back in.

This is pretty common behavior for a modern IDE. The QSPICE behavior is based on the way Microsoft Visual Studio works.


1 Like