Create account / Log in

Test builds for 3.3.0

Discussion area for the development team.

Moderators: uckelman, Tim M

Re: Test builds for 3.3.0

Postby uckelman » April 6th, 2020, 9:05 pm

ClaudioC wrote:Hello Joel,
Thanks for the clarification abt embedded java version on Windows 10 and VASL being out of scope.

Concerning the 2 issues I reprorted yesterday:
The rendenring of the texts and charts at 150% and 175% DPI scaling in now ok on TunisiaII
However I still see occasional white lines on the mapboard when moving the map around. This time I reproduced the error with right clicks to recenter the map, it also occurs like that. Again, if after it occurs I slightlyreposition the map with the up/down keys, the issue disappears. This occurred at both 150% and 175%

Bye
Claudio


I made a test build for this particular problem: http://www.vassalengine.org/~uckelman/t ... indows.exe

Does the repainting problem still happen with this build at non-integer system scale factors?
User avatar
uckelman
Site Admin
 
Posts: 8554
Joined: December 10th, 2007, 9:48 am
Location: Durham, England

Re: Test builds for 3.3.0

Postby uckelman » April 6th, 2020, 11:37 pm

I've found the problem with the offset of the location indicator from the CounterDetailViewer. Expect a fix for that on Tuesday.
User avatar
uckelman
Site Admin
 
Posts: 8554
Joined: December 10th, 2007, 9:48 am
Location: Durham, England

Re: Test builds for 3.3.0

Postby ClaudioC » April 7th, 2020, 5:30 am

Hello Joel, I tested the proposed repaint fix at both 125% and 175%. I could not reproduce the issue, only once did I see a white line for less that 1 second and that went away by some repaint action.

Issues still occurring:
Location indicator (I see you have a solution)
Absence of Movement Trails (in both Tunisia II and The Longest Day)

Kr
Claudio
User avatar
ClaudioC
 
Posts: 45
Joined: April 18th, 2018, 6:30 pm

Re: Test builds for 3.3.0

Postby uckelman » April 8th, 2020, 12:12 am

ClaudioC wrote:Hello Joel, I tested the proposed repaint fix at both 125% and 175%. I could not reproduce the issue, only once did I see a white line for less that 1 second and that went away by some repaint action.

Issues still occurring:
Location indicator (I see you have a solution)
Absence of Movement Trails (in both Tunisia II and The Longest Day)


I've just uploaded 3.3.0-svn9342, which contains the repaint fix as well adjustments to CounterDetailViewer.

I'll tackle movement trails next. Thank you for your persistence with testing builds.
User avatar
uckelman
Site Admin
 
Posts: 8554
Joined: December 10th, 2007, 9:48 am
Location: Durham, England

Re: Test builds for 3.3.0

Postby ClaudioC » April 8th, 2020, 6:18 am

All good with 9342. Only thing I found in Tunisia II is the Movement Trail. Will test other games later.
User avatar
ClaudioC
 
Posts: 45
Joined: April 18th, 2018, 6:30 pm

Re: Test builds for 3.3.0

Postby ClaudioC » April 8th, 2020, 6:29 am

I found another HDPI Scaling issue. Maybe you are already aware of it:
In Vietnam 1965/1975 version 4.3.4 (http://www.vassalengine.org/mediawiki/i ... 4.3.4.vmod)
You have a button that overlays s string of 1hex area of effects to mark a boundary. (You make them appear with the 2 buttons (red/blue squares) to the right of the 1D6 and 2D6 buttons.
At 100% these overlay appears in the correct manner. At 125% it appears somewhere else (at the 100% position I think).

I have an unpublished The Longest Day module that uses a similar technique. There too the issue appears.

The trait being used is "Area of Effect" http://www.vassalengine.org/wiki/AreaOfEffect

Kr
Claudio
User avatar
ClaudioC
 
Posts: 45
Joined: April 18th, 2018, 6:30 pm

Re: Test builds for 3.3.0

Postby ClaudioC » April 8th, 2020, 6:41 am

The Vietnam module has 2 different uses of the Area of Effect trait. One is to highlight the hexes around some counters. That one works correctly at any scaling.
The differences I see is that while both are "Fixed Radius", the one that fails has a radius of 0 (area only applies on origin hex).
econ
The other difference is that the one is set as "Always visible" (it is hidden when button pressed by a change in the Map Shading settings probably) whereas the one that works has "Always visible" unchecked.
Hope that helps.
User avatar
ClaudioC
 
Posts: 45
Joined: April 18th, 2018, 6:30 pm

Re: Test builds for 3.3.0

Postby uckelman » April 8th, 2020, 10:08 pm

3.3.0-svn9343 should fix the problem with Movement Trails.
User avatar
uckelman
Site Admin
 
Posts: 8554
Joined: December 10th, 2007, 9:48 am
Location: Durham, England

Re: Test builds for 3.3.0

Postby marktb1961 » April 9th, 2020, 1:48 pm

3.3.0-svn9343 on MacBook Pro (with Retina) MacOS Catalina 10.15.4

Reporting against bugs that I logged earlier:

- movement trails now working - tested on C&C Napoleonics 3.42

- deck getting deleted in set up on C&C Napoleonics 3.42 - still a problem.

- zoom bug when using the "saved map as PNG file" - still a problem

- crashes in C&C Medieval - not re-tested.


Regards,

Mark
User avatar
marktb1961
 
Posts: 67
Joined: October 4th, 2018, 2:36 pm
Location: Liverpool, England

Re: Test builds for 3.3.0

Postby bdgza » April 9th, 2020, 9:02 pm

I ran the latest 3.3.0-svn9343 on my iMac 5K (Retina), screen set to Scaled resolution. The module shows correctly

Image

Then I dragged this window over to my external FullHD (non-Retina) monitor. This image is at the same zoom level as the first image:

Image

As you can see the board is clipped at the bottom, the first time I zoomed out from 100% to 44% to view the entire board the clipped part was black until I resized the panel to repaint it.
bdgza
 
Posts: 310
Joined: February 26th, 2010, 10:51 am

Re: Test builds for 3.3.0

Postby bdgza » April 9th, 2020, 9:33 pm

Not only is the image drawn at the wrong size, the position calculations are also all off. I can not click on anything as it's not where it's being drawn.
bdgza
 
Posts: 310
Joined: February 26th, 2010, 10:51 am

Re: Test builds for 3.3.0

Postby bdgza » April 9th, 2020, 9:51 pm

3.3.0-svn9343 on my iMac 5K (Retina), screen set to Scaled resolution.

Selection rectangle drawn is completely wrong. In this image I am dragging from the small cross (left top) to about where the big cross is (right bottom). The selection when I let go is correct, despite the rectangle visual.

Image
bdgza
 
Posts: 310
Joined: February 26th, 2010, 10:51 am

Re: Test builds for 3.3.0

Postby uckelman » April 9th, 2020, 10:58 pm

Thus spake bdgza:
> I ran the latest 3.3.0-svn9343 on my iMac 5K (Retina), screen set to
> Scaled resolution. The module shows correctly
>
> [1]
>
> Then I dragged this window over to my external FullHD (non-Retina)
> monitor. This image is at the same zoom level as the first image:

That's new to me. I had no idea it was possible to have different OS
scale factors _at once_.

I'm not surprised this doesn't work, as it's not a situation I had
envisoned as something one could do. That's a pretty significant
setback.

--
J.
User avatar
uckelman
Site Admin
 
Posts: 8554
Joined: December 10th, 2007, 9:48 am
Location: Durham, England

Re: Test builds for 3.3.0

Postby uckelman » April 10th, 2020, 12:15 am

Thus spake marktb1961:
>
> - zoom bug when using the "saved map as PNG file" - still a problem

3.3.0-svn9344 should fix this.

--
J.
User avatar
uckelman
Site Admin
 
Posts: 8554
Joined: December 10th, 2007, 9:48 am
Location: Durham, England

Re: Test builds for 3.3.0

Postby bdgza » April 10th, 2020, 8:02 am

uckelman wrote:Thus spake bdgza:
> I ran the latest 3.3.0-svn9343 on my iMac 5K (Retina), screen set to
> Scaled resolution. The module shows correctly
>
> [1]
>
> Then I dragged this window over to my external FullHD (non-Retina)
> monitor. This image is at the same zoom level as the first image:

That's new to me. I had no idea it was possible to have different OS
scale factors _at once_.

I'm not surprised this doesn't work, as it's not a situation I had
envisoned as something one could do. That's a pretty significant
setback.

--
J.


The only silver lining I can provide is that macOS doesn't allow window spanning across multiple displays at once. It's either on the one (and clipped on the side), or on the other (and clipped on the opposite side). So a single window can't be at multiple scale factors at once.
bdgza
 
Posts: 310
Joined: February 26th, 2010, 10:51 am

PreviousNext

Return to Developers

Who is online

Users browsing this forum: No registered users and 1 guest