LOS tool and obstacle/hindrance vertices

[Andy Bagley]

Am I missing something or is there a problem with the way the VASL LOS tool deals with LOSs that run precisely through a vertex of a hex with inherent terrain?
For example, set up Board 44 on VASL and run a LOS from 44A4 to 44V5. This runs precisely through one vertex of orchard hex 44Y4 (see http://www.texas-asl.com/banzai/banzai5_2.pdf page 3 if you don't believe me), and hence should have a hindrance - which it does, on my VASL board anyway. Similarly, 44Z2 to 44Y7 also shows as having one hindrance, as it should.
Now try 44X2 to 44Y7, which has exactly the same geometry (a mirror image of the previous example), and this shows as clear - no hindrance! It's not just hex 44Y4 that causes this inconsistency. The LOS from 44CC1 to 44Y4 should be hindered by the vertex of orchard hex 44BB2 but this too shows as clear.
The same thing happens for the longer range example from David Hailey's article. 44AA3 to 44V7 correctly shows as hindered, but 44AA3 to 44T4 (an exact mirror image) shows as clear!
Can anyone explain - our better still, cure - this apparent anomaly?

 

[Andy Bagley]

PS: It's not just Board 44 either. I've found similar issues on other VASL boards.

 

[Jazz]

The cure is to not depend on the LOS engine to make the LOS call.

 

[zgrose]

Am I missing something or is there a problem with the way the VASL LOS tool deals with LOSs that run precisely through a vertex of a hex with inherent terrain?
For example, set up Board 44 on VASL and run a LOS from 44A4 to 44V5. This runs precisely through one vertex of orchard hex 44Y4 (see http://www.texas-asl.com/banzai/banzai5_2.pdf page 3 if you don't believe me), and hence should have a hindrance - which it does, on my VASL board anyway. Similarly, 44Z2 to 44Y7 also shows as having one hindrance, as it should.
Now try 44X2 to 44Y7, which has exactly the same geometry (a mirror image of the previous example), and this shows as clear - no hindrance! It's not just hex 44Y4 that causes this inconsistency. The LOS from 44CC1 to 44Y4 should be hindered by the vertex of orchard hex 44BB2 but this too shows as clear.
The same thing happens for the longer range example from David Hailey's article. 44AA3 to 44V7 correctly shows as hindered, but 44AA3 to 44T4 (an exact mirror image) shows as clear!
Can anyone explain - our better still, cure - this apparent anomaly?

BTW, helps a lot if you include images and the VASSAL+VASL versions you're using.
VASSAL 3.5.8 + VASL 6.6.2 says Hindrances: 1 on my screen.

 

[DougRim]

These inconsistencies and errors are caused by the fact that the map images are not totally consistent geometrically and therefore wonky results are produced by the application of consistent geometry to the calculation of the the LOS thread and which pixels (and therefore hexes) it passes through.

What can be done about this at this point in time? Not much. The images are what they are.

Virtually every version of VASL contains improvements and bug fixes to the los engine. But it is well short of perfect.

How players choose to use it (or not) is entirely up to them.

 

[Andy Bagley]

PS: It's not just Board 44 either. I've found similar issues on other VASL boards.

Yep, I guess you're right. Overrule LOS tool if it's obviously incorrect.

 

[Andy Bagley]

BTW, helps a lot if you include images and the VASSAL+VASL versions you're using.
VASSAL 3.5.8 + VASL 6.6.2 says Hindrances: 1 on my screen.

View attachment 19365

View attachment 19366

Thanks. I tried but could get the LOS lines to appear (technical incompetence).

 

[Andy Bagley]

These inconsistencies and errors are caused by the fact that the map images are not totally consistent geometrically and therefore wonky results are produced by the application of consistent geometry to the calculation of the the LOS thread and which pixels (and therefore hexes) it passes through.

What can be done about this at this point in time? Not much. The images are what they are.

Virtually every version of VASL contains improvements and bug fixes to the los engine. But it is well short of perfect.

How players choose to use it (or not) is entirely up to them.

Thanks Doug, this explanation is really helpful. Normally the VASL LOS tool is excellent, and I have more faith in it than a piece of cotton thread in a live game! I'm sure it's just this particular instance where the pixels don't quite line up perfectly, so I'll continue to use it but be aware of this situation. Very grateful to you.

 

[Andy Bagley]

Thanks. I tried but could get the LOS lines to appear (technical incompetence).

Couldn't

 

[Stewart]

Yep, I guess you're right. Overrule LOS tool if it's obviously incorrect.

Many players rely on the los tool as a crutch instead of actually learning the rules.
Once you learn the rules, you don't need elevation settings, etc
I understand the usefulness but like Doug said, it's not perfect.
Well knowing the rules eliminates the need for a precise computer feature as we know for a fact the shot in the example is hindered by rule