|
Just reporting that in 24.1.5, and using my 65 inch 4k screen....
?
When adding a wire, I can very faintly see the extended crosshairs and cursor.
So faintly are the crosshairs that LTspice 24.1.5 is unusable with my 65 inch screen.
?
I can see the cursor as long as I don't add a wire.
|
Good point. Pixel pitch, I guess is the more proper term.
?
--
Michael Stokowski
LTspice Team
Analog Devices Inc.
|
On Tue, Mar 11, 2025 at 12:00 PM, eetech00 wrote:
I'm sometimes use a 4K 65" sony TV as monitor.
Most modern TVs use algorithms designed to smooth imperfections in the video source material.? They help reduce the appearance of "grain" and similar noise.? Such a smoothing algorithm might completely erase 1-pixel dots.
?
I'm also assuming that you have a 1:1 correspondence between your computer's video resolution, and the TV's display.? If they don't match, it has to interpolate between adjacent pixels, which might also alter their appearance.
?
Andy
?
?
|
Well, no, I meant 'somehow', because I am not
sure that it's possible to actually alter pixel size.
On 2025-03-11 17:49, mstokowski via
groups.io wrote:
On Tue, Mar 11, 2025 at 09:28 AM, John Woodgate wrote:
Yes, I brought it up, and ADI have agreed that it
needs to be fixed. I guess pixels are somehow smaller now than
when Mike E wrote the code for the dots.
I think you forgot the :^).? …somewhat smaller.
?
I do like the crosses idea.
?
mike
--
Michael Stokowski
LTspice Team
Analog Devices Inc.
--
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
|
On Tue, Mar 11, 2025 at 09:28 AM, John Woodgate wrote:
Yes, I brought it up, and ADI have agreed that it needs to be fixed. I guess pixels are somehow smaller now than when Mike E wrote the code for the dots.
I think you forgot the :^).? …somewhat smaller.
?
I do like the crosses idea.
?
mike
--
Michael Stokowski
LTspice Team
Analog Devices Inc.
|
On Tue, Mar 11, 2025 at 08:46 AM, Dennis wrote:
Yes, a click on a rectangles corner anchor does allow stretching directly, but the click target is so very small that often the user clicks on one of the sides instead and then the operation becomes a move rather than a stretch, or the click isn't recognized at all.
?
Clicking on a side of the rectangle displays two anchor circles on opposite corners of the rectangle. It seems to me that any click inside the radius of those circles should act like a click on the corner of the rectangle and begin the stretch operation.?
?
As it is now, it is too finicky to be reliable so users have started to use the lasso operation to select the corner instead of a click. Hence all the reports saying this is required.
Ah, yes. I was playing with it at quite a high magnification, where the anchors/handles are fat circles, easy to grab. Don't have to zoom very far to make it impossible to point-to-grab a handle. Ideally, the handle size would not be magnification dependent, albeit relative location is. This latter conditions is a common problem in drawing applications: can't distinguish one handle from another because locations are too close re pixels, but accepted as a necessary evil by users. This, in contrast to the LTspice problem, which you've discovered: the handle unavailable because it has shrunk, not very friendly.
?
--
Michael Stokowski
LTspice Team
Analog Devices Inc.
--
Michael Stokowski
LTspice Team
Analog Devices Inc.
|
Afraid it is automated. The script that changes the version label runs some time later in a batch process, after the file is uploaded. Lags by a day or so. You can also use Check for Updates in LTspice, to see what download is actually available.
?
--
Michael Stokowski
LTspice Team
Analog Devices Inc.
|
Yup, version 24.1.5 is the one that's there, identified as 24.1.4.
?
I already let Analog Devices know that they have to update their webpage to show what you are actually downloading.
?
Andy
?
|
On Tue, Mar 11, 2025 at 09:24 AM, Andy I wrote:
On Tue, Mar 11, 2025 at 12:16 PM, eetech00 wrote:
By the way, I'm referring to the operation in LTspice 24.1.5
Analog Devices claims the current version is 24.1.4.? They have not updated their LTspice downloads page yet.
?
?
?
Ok...well I just updated to 24.1.5
?
Here's a snippet from the log:
?
3/09/25 LTspice 24.1.5
? ? ? ?* Bug fixes
? ? ? ?* Node names can be expressions again, this time officially documented and supported
2/18/25 LTspice 24.1.4
? ? ? ?* Bug fixes and minor improvements
2/14/25 LTspice 24.1.3
? ? ? ?* Re-enabled expanded netlist functionality
? ? ? ?* Bug fixes
2/1/25 LTspice 24.1.2
? ? ? ?* Re-enabled caret operator
? ? ? ?* Duplicate .model cards are accepted if they are identical
? ? ? ?* Duplicate .func, .param, and .subckt are accepted if they come from the same location in the same file
? ? ? ?* Other bugs fixed
..........
|
Also, there doesn’t appear to have been a
24.0.15; the last one listed is 24.0.12. I can run the uploaded
.ASC with 24.0.11.
On 2025-03-11 16:24, Andy I via
groups.io wrote:
On Tue, Mar 11, 2025 at 12:16 PM, eetech00 wrote:
By the way, I'm referring to the operation in LTspice
24.1.5
Analog Devices claims the current version is 24.1.4.? They have
not updated their LTspice downloads page yet.
?
?
--
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
|
Yes, I brought it up, and ADI have agreed that
it needs to be fixed. I guess pixels are somehow smaller now
than when Mike E wrote the code for the dots.
On 2025-03-11 16:00, eetech00 via
groups.io wrote:
I don't know if this was brought up...but I still can't see a
schematic grid.
I haven't been able to view the grid for many versions of
LTspice.
I use a light grey background but it doesn't seem to matter.
Basically, the grid switch and color does nothing.
?
I'm sometimes use a 4K 65" sony TV as monitor.
--
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
|
On Tue, Mar 11, 2025 at 12:16 PM, eetech00 wrote:
By the way, I'm referring to the operation in LTspice 24.1.5
Analog Devices claims the current version is 24.1.4.? They have not updated their LTspice downloads page yet.
?
?
|
On Tue, Mar 11, 2025 at 09:11 AM, eetech00 wrote:
On Tue, Mar 11, 2025 at 08:46 AM, Dennis wrote:
Yes, a click on a rectangles corner anchor does allow stretching directly, but the click target is so very small that often the user clicks on one of the sides instead and then the operation becomes a move rather than a stretch, or the click isn't recognized at all.
?
Clicking on a side of the rectangle displays two anchor circles on opposite corners of the rectangle. It seems to me that any click inside the radius of those circles should act like a click on the corner of the rectangle and begin the stretch operation.?
?
As it is now, it is too finicky to be reliable so users have started to use the lasso operation to select the corner instead of a click. Hence all the reports saying this is required.
?
That might be the problem.
A drag operation shouldn't turn into a move operation.
If a drag operation has been selected, a click anywhere on the edge of a rectangle should drag the edge of the rectangle and its intersecting edges..
If a drag operation has been selected, a lasso should be required to drag intersecting edges of a rectangle.
?
Just my opinion.
?
?
By the way, I'm referring to the operation in LTspice 24.1.5
?
|
On Tue, Mar 11, 2025 at 08:46 AM, Dennis wrote:
Yes, a click on a rectangles corner anchor does allow stretching directly, but the click target is so very small that often the user clicks on one of the sides instead and then the operation becomes a move rather than a stretch, or the click isn't recognized at all.
?
Clicking on a side of the rectangle displays two anchor circles on opposite corners of the rectangle. It seems to me that any click inside the radius of those circles should act like a click on the corner of the rectangle and begin the stretch operation.?
?
As it is now, it is too finicky to be reliable so users have started to use the lasso operation to select the corner instead of a click. Hence all the reports saying this is required.
?
That might be the problem.
A drag operation shouldn't turn into a move operation.
If a drag operation has been selected, a click anywhere on the edge of a rectangle should drag the edge of the rectangle and its intersecting edges..
If a drag operation has been selected, a lasso should be required to drag intersecting edges of a rectangle.
?
Just my opinion.
?
|
I don't know if this was brought up...but I still can't see a schematic grid.
I haven't been able to view the grid for many versions of LTspice.
I use a light grey background but it doesn't seem to matter.
Basically, the grid switch and color does nothing.
?
I'm sometimes use a 4K 65" sony TV as monitor.
|
On Mon, Mar 10, 2025 at 01:36 PM, mstokowski wrote:
Considered and done! Actually, no change needed, you can already do this by clicking an anchor, or node, or component—no lasso required.
?
Yes, a click on a rectangles corner anchor does allow stretching directly, but the click target is so very small that often the user clicks on one of the sides instead and then the operation becomes a move rather than a stretch, or the click isn't recognized at all.
?
Clicking on a side of the rectangle displays two anchor circles on opposite corners of the rectangle. It seems to me that any click inside the radius of those circles should act like a click on the corner of the rectangle and begin the stretch operation.?
?
As it is now, it is too finicky to be reliable so users have started to use the lasso operation to select the corner instead of a click. Hence all the reports saying this is required.
|
That's a good clarification. Thanks. Andy.
On 2025-03-10 20:55, Andy I via
groups.io wrote:
A few days ago, John Woodgate wrote:
2. Drawing a wire with
CTRL held down doesn't allow the wire to start and end
off-grid. It's the same if I draw lines instead of wires.
I can't find a 'Snap to grid' check box to untick.
After some thought, I think that you never could use the CTRL
key, nor anything else, to make a wire not snap-to-grid.? I
think wires and components can only snap-to-grid.
?
Reading the Help instructions:
The graphical
annotations to the schematic; lines, rectangles,
circles, and arcs; snap by default to the same grid as
the used for electrical contacts of wires and pins. Hold
down the control key while positioning these to defeat
this snap.
This is the only place I
found where it talks about not snapping-to-grid, and note
that it applies ONLY to lines, rectangles,
circles, and arcs.? It does not apply to Wires.
?Note the distinction between Wires
and Lines.? Lines
(right-click < Draw > Line) are non-functional,
whereas Wires (F3, or the "Wire" toolbar button) make electrical connections. ?I think the conclusion is that holding
down CTRL never worked for drawing wires off-grid.? For
wires, CTRL only toggles whether they can be drawn at an
angle.
?
It makes sense to me that Wires could never be made off-grid,
because it could be challenging to connect anything to the
off-grid end of a wire.? The schematic editor uses integers for
location, and grid points are every 16 integer values - which
means there are 15 chances to get it wrong and miss the end of a
wire that is not on-grid.
?
You can hand-edit your .asc schematic file and move a wire's
starting or ending points, making them off-grid.? It works.?
Once in the schematic file, LTspice accepts it and leaves it
off-grid when saved again.
?
--
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
|
A few days ago, John Woodgate wrote:
2. Drawing a wire with CTRL held down doesn't allow the wire to start and end off-grid. It's the same if I draw lines instead of wires. I can't find a 'Snap to grid' check box to untick.
After some thought, I think that you never could use the CTRL key, nor anything else, to make a wire not snap-to-grid.? I think wires and components can only snap-to-grid.
?
Reading the Help instructions:
The graphical annotations to the schematic; lines, rectangles, circles, and arcs; snap by default to the same grid as the used for electrical contacts of wires and pins. Hold down the control key while positioning these to defeat this snap.
This is the only place I found where it talks about not snapping-to-grid, and note that it applies ONLY to lines, rectangles, circles, and arcs.? It does not apply to Wires. ?Note the distinction between Wires and Lines.? Lines (right-click < Draw > Line) are non-functional, whereas Wires (F3, or the "Wire" toolbar button) make electrical connections. ?I think the conclusion is that holding down CTRL never worked for drawing wires off-grid.? For wires, CTRL only toggles whether they can be drawn at an angle.
?
It makes sense to me that Wires could never be made off-grid, because it could be challenging to connect anything to the off-grid end of a wire.? The schematic editor uses integers for location, and grid points are every 16 integer values - which means there are 15 chances to get it wrong and miss the end of a wire that is not on-grid.
?
You can hand-edit your .asc schematic file and move a wire's starting or ending points, making them off-grid.? It works.? Once in the schematic file, LTspice accepts it and leaves it off-grid when saved again.
?
|
I am sorry.? In my waveform colour scheme, I used the resistor colour codes, but offset by 1.? Thus "Trace V(1)" was Black, "Trace V(2)" Brown, etc.? I had become completely accustomed to Black being the first waveform, the same way that it was the first resistor colour, even though 0 ≠ 1.? I didn't mind that the numbers were offset, as much as just having them in sequence.? Maybe I should reconsider that.
?
More important than that - to me, anyway - is getting enough contrast to be visible.
?
Andy
?
|
Yes, that's why I wrote 'near'. For colouring
traces, there is no use for '0', so it comes in useful for '10',
and the other two colours are fairly easy to generate in the
Settings.
On 2025-03-10 20:25, Donald H Locker
via groups.io wrote:
IIRC, Black represents 0 on resistors. (Colour blind, but not
_that_ colour blind!)
My favourite R was Brn-Blk-Orn
Donald.
On 3/10/25 15:17, Andy I via
groups.io wrote:
On Mon, Mar 10, 2025 at 03:01 PM, John Woodgate wrote:
I'd like the default
colours to nearly follow the resistor colour code, ...
That is essentially what I used too.? But I diverged
somewhat from the resistor colour code sequence to make them
more visible.??It is interesting that you use a light grey
background for better contrast.? I should try that.? Thanks
for the tip.
?
but with black as 10,
pink as 11 and cyan as 12, so I've set it up that way.
Does this mean you re-use black?? Or are they different shades
for 1 and 10?
?
?
--
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
|
John Woodgate