|
The following circuit configuration was unintentional, but caused an LTspice program exit upon simulation run.
?
* Generated by LTspice 24.1.7 for Windows.
I1 0 0 1m
D1 0 0 D
V1 ADJ 0 0
.model D D
.lib (path removed)\standard.dio
.dc I1 2u 2m 100n
.backanno
.end
?
|
It doesn't crash version 24.0.11, but it's a
nonsense .ASC, isn't it? LTspice can't find I1, presumably
because it's shorted.
On 2025-04-26 18:56, eetech00 via
groups.io wrote:
The following circuit configuration was unintentional, but
caused an LTspice program exit upon simulation run.
?
* Generated by LTspice 24.1.7 for Windows.
I1 0 0 1m
D1 0 0 D
V1 ADJ 0 0
.model D D
.lib (path removed)\standard.dio
.dc I1 2u 2m 100n
.backanno
.end
?
--
Best wishes John Woodgate RAYLEIGH Essex OOO-Own Opinions Only
If something is true: * as far as we know - it's science *for
certain - it's mathematics *unquestionably - it's religion
|
LTspice used to remove both I1 and D1 as they were both shorted.? But 24.1 was a major re-write and it might not do that anymore.
?
Andy
?
|
On Sat, Apr 26, 2025 at 03:58 PM, John Woodgate wrote:
It doesn't crash version 24.0.11, but it's a nonsense .ASC, isn't it? LTspice can't find I1, presumably because it's shorted.
Yes. Like I wrote, the configuration is unintentional, but nevertheless, it shouldn't cause a program exit.
|
@eetech00: Unfortunately, you're not the author/maintainer so what you think should or should not happen may not have much effect on what actually happens. Many compilers will happily EXIT when the see garbage input. All sorts of new behaviors related to "enhanced" syntax checking have broken lots of legacy simulation and I don't see much enthusiasm for the authors/maintainers to back off on those changes.
|
Even so, an app should not quit without an
error message.
On 2025-04-27 20:23, BRUCE108 wrote:
@eetech00: Unfortunately, you're not the author/maintainer so
what you think should or should not happen may not have much
effect on what actually happens. Many compilers will happily
EXIT when the see garbage input. All sorts of new behaviors
related to "enhanced" syntax checking have broken lots of legacy
simulation and I don't see much enthusiasm for the
authors/maintainers to back off on those changes.
--
Best wishes John Woodgate RAYLEIGH Essex OOO-Own Opinions Only
If something is true: * as far as we know - it's science *for
certain - it's mathematics *unquestionably - it's religion
|
No, even when it states in throes of death: "untrapped err........"
--
Regards,
Tony
On 27/04/2025 22:06, John Woodgate via
groups.io wrote:
toggle quoted message
Show quoted text
Even
so, an app should not quit without an error message.
|
LTspice was very robust before, and rarely crashed even when errors happened.? Some of the recent changes (24.1.x) made it less robust.? Hopefully ADI can restore what was lost.
?
Andy
?
|
On 27/04/2025 23:22, Andy I via
groups.io wrote:
LTspice was very
robust before, and rarely crashed even when errors happened.? Some
of the recent changes (24.1.x) made it less robust.? Hopefully ADI
can restore what was lost.
I haven't got to 24.1.7 yet, but I've not noticed any increased
propensity to crash, so far. But then, I still have 24.0.12 and
17.1.15 running alongside 24.1.x.
I still use 24.0.12 as default for the same reasons as I have
highlighted in my posts. I don't like it when backward compatibility
is compromised. New features are great, provided nothing gets broken
in the process. Some of the things that were broken at first have
now been fixed, but IMHO the fixes should have been applied before
release. LTspice is not a disruptive start-up; it doesn't need to
break things, then fix them quick. Leave that to the others.
If LTspice gets a reputation for breaking, it will lose users.
Simple.
--
Regards,
Tony
|
I can't reproduce this. All I get when I run this netlist is an error message about the file "standard.dio" not being found.
?
Best Regards,
Mathias
?
On Sat, Apr 26, 2025 at 07:56 PM, eetech00 wrote:
The following circuit configuration was unintentional, but caused an LTspice program exit upon simulation run.
?
* Generated by LTspice 24.1.7 for Windows.
I1 0 0 1m
D1 0 0 D
V1 ADJ 0 0
.model D D
.lib (path removed)\standard.dio
.dc I1 2u 2m 100n
.backanno
.end
?
|
I'd be quite interested in any evidence that could substantiate this claim.
?
And with "before", I take it you don't mean LTspiceXVII, because I had that crashing multiple times per hour on bad days. Fortunately, we now have people on the helm who deeply care about quality and robustness.
?
Best Regards,
Mathias
?
On Sun, Apr 27, 2025 at 11:22 PM, Andy I wrote:
LTspice was very robust before, and rarely crashed even when errors happened.? Some of the recent changes (24.1.x) made it less robust.? Hopefully ADI can restore what was lost.
?
Andy
?
|
On 27/04/2025 00:57, John Woodgate via
groups.io wrote:
It
doesn't crash version 24.0.11, but it's a nonsense .ASC, isn't it?
LTspice can't find I1, presumably because it's shorted.
It crashes in 24.1.4, too.
The "expanded" netlist is:
LTspice 24.1.4 for Windows
Circuit: I:\tony\Desktop\Test.cir
Start Time: Mon Apr 28 10:51:01 2025
??? --- Expanded Netlist ---
V1 ADJ 0 DC=0
.model D D()
.end
solver = Normal
Maximum thread count: 32
tnom = 27
temp = 27
method = modified trap
--
Regards,
Tony
|
I have to say that, while I did not try very
large or complicated simulations, I never had LTspiceXVII crash.
On 2025-04-28 09:46, Mathias Born via
groups.io wrote:
I'd be quite interested in any evidence that could
substantiate this claim.
?
And with "before", I take it you don't mean LTspiceXVII,
because I had that crashing multiple times per hour on bad days.
Fortunately, we now have people on the helm who deeply care
about quality and robustness.
?
Best Regards,
Mathias
?
On Sun, Apr 27, 2025 at 11:22 PM, Andy I wrote:
LTspice was very robust before, and rarely crashed even
when errors happened.? Some of the recent changes (24.1.x)
made it less robust.? Hopefully ADI can restore what was
lost.
?
Andy
?
--
Best wishes John Woodgate RAYLEIGH Essex OOO-Own Opinions Only
If something is true: * as far as we know - it's science *for
certain - it's mathematics *unquestionably - it's religion
|
Thanks! That was the missing piece.
The topology checker removes I1 and then the dc-sweep accesses that, which may or may not cause a crash. Will be fixed.
?
Best Regards,
Mathias
?
On Mon, Apr 28, 2025 at 10:54 AM, Tony Casey wrote:
On 27/04/2025 00:57, John Woodgate via groups.io wrote:
It doesn't crash version 24.0.11, but it's a nonsense .ASC, isn't it? LTspice can't find I1, presumably because it's shorted.
It crashes in 24.1.4, too.
The "expanded" netlist is:
LTspice 24.1.4 for Windows
Circuit: I:\tony\Desktop\Test.cir
Start Time: Mon Apr 28 10:51:01 2025
??? --- Expanded Netlist ---
V1 ADJ 0 DC=0
.model D D()
.end
solver = Normal
Maximum thread count: 32
tnom = 27
temp = 27
method = modified trap
--
Regards,
Tony
|
On Mon, Apr 28, 2025 at 04:41 AM, Mathias Born wrote:
I can't reproduce this. All I get when I run this netlist is an error message about the file "standard.dio" not being found.\
Perhaps you forgot to substitute your computer's "path" to the standard.dio file, in the ".lib" statement.? Obviously, eT edited that line to remove his computer's unique path.
?
I'd be quite interested in any evidence that could substantiate this claim.
?
And with "before", I take it you don't mean LTspiceXVII, ...
I did mean LTspice XVII, and LTspice III and LTspice IV.
... because I had that crashing multiple times per hour on bad days. ...
You must have had incredibly bad days or very unique schematics/netlists (or something unique about your computer), because I rarely ever ever saw LTspice III or IV or XVII crash.? Occasionally (a few times a year) it would hang and I had to manually crash it, but it did not crash itself.? If I recall correctly, that happened ONLY while it was trying to plot a particularly difficult (very 'dense') set of waveforms, but never during a simulation or editing schematics.? Now I use the word "never" which might be incorrect.? But I can truly not remember a time when LTspice III or LTspice IV or LTspice XVII crashed when simulating or preparing to simulate something.? There were ways to make simulations fail, but not to make the program itself fail and fall over.
?
I also note there are group members here who reported they could get theirs to crash, but I don't recall others being able to replicate it.? (Now there may have been exceptions where a new program release did, but it was quickly fixed, causing its memory to fall through the holes in my brain.)
?
Andy
?
|
John Woodgate
Tony Casey