[Ktechlab-devel] Pushed fixes to github

[Alan Grimes]

I just remembered why this is...

There are two different clocks...

Well, not quite,

I made sure that there was exactly one gold-standard clock in the
system, that said, there are two completely different types of
simulation going on here.

There are the linear circuit parts that use the crappy matrix engine I
wrote for a school project years ago, it is due for an overhaul... It is
Fast Enough (tm) but still buggy.

Then there is another layer of simulation for logic elements which
implements high-level boolean algebra logic. Because it operates at a
higher level of abstraction, it is much cheaper to compute. Furthermore,
the necessity to simulate high-performance digital logic demands that it
run much faster than the analog simulations. I forgot the exact ratio,
but it is probably running about a thousand times faster than the analog
simulation you are trying to use to read the output.

Anyway, that's what's going on...


Claus Andersen wrote:
> Hi.
>
> Thanks for keeping at it.
>
> I just compiled and ran it on my virtual machine, and the bugs you
> fixed are gone here :)
>
> Regarding step freq.; can it be adjustable in the program or is it
> hardcoded? I don't need 1ms and think 20ms is fine, I can't speak for
> everyone :)
> Often I am ok with a relatively low resolution on the little circuits
> I make
>
> I can't yet create very complex circuits for testing so I hope someone
> else chimes in for testing in general.
>
> Keep up the good work :)
>
> Regards,
>
> Claus
>
> --------------------------------------------------------------
>
> umount /dev/vodka && eject /dev/stomach
>
> -Mplayer crew
>
>
> > From: zoltan.padrah at gmail.com
> > To: ktechlab-devel at lists.sourceforge.net
> > Date: Sun, 13 Sep 2015 20:10:44 +0300
> > Subject: Re: [Ktechlab-devel] Pushed fixes to github
> >
> > On Sun, 13 Sep 2015 20:01:54 +0300, Zoltan Padrah
> > <zoltan.padrah at gmail.com> wrote:
> >
> > >
> > > Hi,
> > >
> > > I have pushed several fixes to the kde4 porting branch on github. The
> > > fixes are following:
> > >
> > > -The orientation icons are broken; the graphics are scrambled, but
> > > reorienting components works.
> > > - fixed the orientation icons
> > >
> > > - Pausing sim. works (but no text hint in the main window as the old
> > > ver. had)
> > > - fixed text hint
> > >
> > > - The program uses 100% of cpu all the time.
> > > - reduced it to a reasonable level and pausing the simulation should
> > > have a noticeable effect. Here could be a discussion about simulation
> > > steps frequency (reduced it to 20ms period, and compensated in
> iteration
> > > per step count). I assume nobody is connecting ktechlab to external
> > > hardware that _needs_ ~1ms simulation step period.
> > >
> > > - ktechlab(2737)/kdeui (kdelibs) KXMLGUIClient::~KXMLGUIClient:
> > > 0xb11569a4 deleted without having been removed from the factory
> first.
> > > This will leak standalone popupmenus and could lead to crashes.
> > > - fixed
> > >
> > >
> > > The branch is hosted at github at:
> > >
> > > https://github.com/ktechlab/ktechlab-0.3/tree/port-0.3.8-kde4-v1
> > >
> > > Feel free to test and any feedback is welcome.
> > >
> > >
> > > Have fun,
> > >
> > > Zoli
> >
> > I'm also sending my current "known issues" list below; feel free to
> signal
> > other things that need fixing:
> >
> >
> >
> > - serious graphic glitches on circuit view; mostly resulting in black
> > circuit display when trying to scroll; various painting related
> warnings
> > from Qt
> >
> > - created subcircuits appear twice in the subcircuit list, and in some
> > cases they cause crashes; investigation needed
> >
> > - (from Claus) There are some other gfx glitches, but that might be
> > because I run it in a virtual machine - didn't have them on the
> previous
> > builds though.
> >
> > - Gfx error on e.g. LED in that I can't connect a wire unless I
> rotate the
> > part.
> > - some components get their connectors moved, thus (1) graphical
> issues,
> > (2) impossibility of connecting connectors results
> >
> > - Som gfx errors that leaves gaps at connection points on components.
> > (see last email with attached screen shot)
> >
> >
> > -'Scope works well, but doesn't clear 'lines' from other circuits when
> > they are closed. (Y-axis scale would be nice)
> >
> > -I think the 'scope just keeps filling data points into RAM, because
> RAM
> > usage climbs steadily until the virtual machine crashes (same as the
> kde3
> > version)
> >
> > - components have no description
> > - the context help exists; or something else is needed here?
> >
> >
> > And the issues from the TODO file:
> >
> >
> > Features needs re-implementation at Qt4 port
> > =======
> > - dbus integration for documents (used to be dcop bindings for
> documents)
> > - undo / redo interface needs to be fixed for text documents: this
> > interface is not exported by KDE4 KTextEditor
> >
> http://api.kde.org/4.10-api/kdelibs-apidocs/interfaces/ktexteditor/html/kte_port_to_kde4.html
> > - printing a text document needs fixing
> >
> >
> ------------------------------------------------------------------------------
> > _______________________________________________
> > Ktechlab-devel mailing list
> > Ktechlab-devel at lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/ktechlab-devel
>
>
> ------------------------------------------------------------------------------
>
>
> _______________________________________________
> Ktechlab-devel mailing list
> Ktechlab-devel at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/ktechlab-devel


-- 
IQ is a measure of how stupid you feel.

Powers are not rights.