LOS to vertex

DougRim

Senior Member
Joined
Apr 23, 2012
Messages
466
Reaction score
546
Location
Ottawa
Country
llCanada
I have noticed some irregularities in how VASL handles LOS to hex vertices. Before trying to do anything about this, I want to be sure I understand the applicable ASL rules properly (or as near as can be accomplished by mere human intelligence!).

Please let me know if you think the LOS should be clear/blocked in the examples below (please note on my system red=clear/blue=blocked).

In the first example, I think the LOS is incorrect and should be blocked?

lossnip1.PNG

In the next example, I think the LOS should be clear (note that if you move the mouse slightly so that the LOS is checked in D5/C6 rather than C6/D5, then it shows as clear so something is wrong somewhere!)?
lossnip2.PNG

Finally, in the last example, I think the LOS should be clear? No reason why the hedge should block the LOS in F8.

lossnip3.PNG

Thoughts and corrections would be appreciated.
 

jrv

Forum Guru
Joined
May 25, 2005
Messages
21,998
Reaction score
6,145
Location
Teutoburger Wald
Country
llIceland
VASL traces to all three hexes that participates in a vertex, and it does not give much indication of which hex it is tracing to. It also will trace LOS to vertices where there is no game reason you should trace LOS to that vertex. In the first example the LOS is being traced to the vertex in hex K9, as indicated by the range being one. With careful mouse handling you can get it to trace to the vertex in L8 and L9. The LOS to the vertex in K9 (and L8) should be clear.

In the second example LOS is also being traced at range one to D5 or C6, you can't tell which. In this case LOS should be clear but there is no game reason to trace LOS to the vertex in either of those hexes. If the LOS were traced to the vertex in D6 (range 2), then LOS should be blocked.

Case three is similar to case 2.

In all these cases the LOS will vary depending on which hex the LOS is traced to in addition to the vertex/hexside. Whenever LOS is traced to a vertex or a hexside, it is also traced to one of the three or two hexes that abut. This will frequently affect the range as well as the LOS.

If you have the full information switched on, you can either tell which hex vasl thinks it is pointing at, or at least which hex it is not pointing at. I submitted a vasl bug suggesting that there be some visual indication of which hex the LOS was traced to, but it was denied.

JR
 

Sully

Senior Member
Silver Supporting Member
Joined
Feb 2, 2003
Messages
1,153
Reaction score
244
Location
Mpls, MN
Country
llUnited States
In the first example assume there is a building in K9. A bypassing unit would be considered to be a Level 1, so that LOS is correct.

The second example is a bug and appears to be caused by the LOS picking up pixels in D5.

The third example is a bug – in-hex hexside terrain should not block LOS to the vertices.

If you have the full information switched on, you can either tell which hex vasl thinks it is pointing at, or at least which hex it is not pointing at. I submitted a vasl bug suggesting that there be some visual indication of which hex the LOS was traced to, but it was denied.
In verbose mode if you drag the thread near a vertex the first hex (e.g. C6 in C6/D5) is the hex you are in. The second hex identified which hexside is nearest that point, as this often affects LOS. Not sure what else you are looking for.
 

jrv

Forum Guru
Joined
May 25, 2005
Messages
21,998
Reaction score
6,145
Location
Teutoburger Wald
Country
llIceland
In verbose mode if you drag the thread near a vertex the first hex (e.g. C6 in C6/D5) is the hex you are in. The second hex identified which hexside is nearest that point, as this often affects LOS. Not sure what else you are looking for.
I think a little arrow-y/triangular-y/chevron-y thing in the hex that vasl thinks it is in and pointing to the vertex might let the user know which hex is currently being considered. If you know about how vasl works, it's easy enough to get the vertex selection into the right hex by moving the mouse around. What is hard is to realize that there is a difference between tracing LOS to the vertex in one hex vs. another hex.

This is a sketch of what it might look like with a chevron and LOS traced from 1K2 center to vertex 1L1/1L2/1M2 in 1M2. The chevron (/arrow/triangle) would be in 1L1 or 1L2 if that was the hex that vasl was tracing to.

Vertex marking.jpg

I just played around with verbose mode, and it's not clear to me what the second hex in "#1/#2" indicates. If I trace to L1/L2/M2 in L2, it shows "L2/M2". Why does it say that when "L1/L2" is also a hexside at that vertex? At first I thought it recognized that M2 was a building but that turned out not to be the case. If I trace to L1/M1/M2 (the top left vertex of M2) in hex L1, it says "L1/M1", which is a road hexside. It seems to pick one of the two hexsides at the vertex and hex arbitrarily.

JR
 

von Marwitz

Forum Guru
Joined
Nov 25, 2010
Messages
11,055
Reaction score
4,558
Location
Kraut Corner
Country
llGibraltar
I think a little arrow-y/triangular-y/chevron-y thing in the hex that vasl thinks it is in and pointing to the vertex might let the user know which hex is currently being considered.
Great idea!

von Marwitz
 

von Marwitz

Forum Guru
Joined
Nov 25, 2010
Messages
11,055
Reaction score
4,558
Location
Kraut Corner
Country
llGibraltar
And since we are talking about LOS:

Especially when overlays are in use and the LOS crosses one, I think that the current warning that pops up ("LOS tool disabled due to overlay in the vincinity" or someting to that effect) can be a problem. Depending on how the LOS is drawn, the pop-up warning can blot out the LOS for a couple of hexes almost entirely. Mostly, but not always, this can be amended by checking the LOS in the opposite direction. More commonly, some portions of terrain (say a patch of woods) is blotted out, so there might be no way to tell if the LOS beneath the pop-up warning crosses the Woods depiction (which would block LOS) or not (so LOS exists). Especially in the latter case, the pop-up warning of "crossing an overlay" does more harm than good.


von Marwitz
 

DougRim

Senior Member
Joined
Apr 23, 2012
Messages
466
Reaction score
546
Location
Ottawa
Country
llCanada
And since we are talking about LOS:

Especially when overlays are in use and the LOS crosses one, I think that the current warning that pops up ("LOS tool disabled due to overlay in the vincinity" or someting to that effect) can be a problem. Depending on how the LOS is drawn, the pop-up warning can blot out the LOS for a couple of hexes almost entirely. Mostly, but not always, this can be amended by checking the LOS in the opposite direction. More commonly, some portions of terrain (say a patch of woods) is blotted out, so there might be no way to tell if the LOS beneath the pop-up warning crosses the Woods depiction (which would block LOS) or not (so LOS exists). Especially in the latter case, the pop-up warning of "crossing an overlay" does more harm than good.


von Marwitz
Not to be defensive but . . . The previous solution was to disable LOS entirely on any map that contained an overlay. The current situation is that LOS is only disabled when the los check crosses an overlay. Thus LOS should be usable in most (but definitely not all) situations on maps with overlays.

I agree that to users it appears that LOS is disabled whenever a LOS gets near an overlay. This is because overlay files are actually rectangles and not the shape of the overlay. Once a los check crosses any part of the overlay rectangle, the LOS Engine in VASL cannot correctly determine what type of terrain it is crossing (exactly as your highlighted text suggests) and thus the LOS check cannot be completed. So what appears to the user is, as you say, anytime the LOS is in the vicinity of an overlay, the warning popup appears. Sad but necessary.

The only way I know of to fix this is to create losdata files for each overlay, in each of its possible rotational positions. And then have that data integrated into the terrain data files created for each map at runtime by VASL. Neither of these is simple and both are a lot of work. To date, we have no volunteers to take this on.

Could we move any further discussion on this point to a new thread as I am still hoping for more help on my original questions!

Thanks
 

DougRim

Senior Member
Joined
Apr 23, 2012
Messages
466
Reaction score
546
Location
Ottawa
Country
llCanada
jrv and David Sullivan: thanks for your help on this.

Do I understand it correctly that LOS to a vertex is not always the same, depending on which one of the three hexes the LOS is being drawn to? So, for example, in my first example, the LOS from J8 to the K9/L8/L9 vertex in K9 is clear but if L8 or L9 were the target hex then it would be blocked?

This is my understanding but I want to be sure that I have it right.
 

jrv

Forum Guru
Joined
May 25, 2005
Messages
21,998
Reaction score
6,145
Location
Teutoburger Wald
Country
llIceland
Do I understand it correctly that LOS to a vertex is not always the same, depending on which one of the three hexes the LOS is being drawn to? So, for example, in my first example, the LOS from J8 to the K9/L8/L9 vertex in K9 is clear but if L8 or L9 were the target hex then it would be blocked?
L8 would still be clear at the vertex. Despite looking like it is at level zero, a unit in bypass is at the base level of the hex it is in, which is level one for L8. This is per B10.1, last sentence. You are correct in observing that LOS to a vertex is not always the same, depending on which hex the unit is in. LOS is traced to a hex first, then to a point in that hex. Although LOS to a vertex seems to be tracing to the same place, because the vertex is part of different hexes the vertex may have different characteristics, including level, presence/absence of SMOKE or other inherent terrain, etc.

As another example of difference, a unit in K10 firing at K9/L8/L9 would fire at level zero (i.e. no Height Advantage) if the target were bypassing K9/L9 but at level one (with height advantage) if the target was bypassing L8, with one major exception. If the unit was attacked by first fire as it crossed the L8/L9 hexside from L9 to L8 it would not benefit from height advantage for fire through that hexside.

JR
 

DougRim

Senior Member
Joined
Apr 23, 2012
Messages
466
Reaction score
546
Location
Ottawa
Country
llCanada
L8 would still be clear at the vertex. Despite looking like it is at level zero, a unit in bypass is at the base level of the hex it is in, which is level one for L8.

JR
How did I miss that L8 was at level 1? Self-smack up the side of the head!! Yes, it is clearly clear. Thanks.
 

DougRim

Senior Member
Joined
Apr 23, 2012
Messages
466
Reaction score
546
Location
Ottawa
Country
llCanada
How about this LOS? It is showing clear and I think it should be. A bypass unit at C6-C7-B6 would be at level 1 according to A4.34.
losbypass1.PNG
 

jrv

Forum Guru
Joined
May 25, 2005
Messages
21,998
Reaction score
6,145
Location
Teutoburger Wald
Country
llIceland
It would be a clear LOS except for the woods. The woods should block LOS even though the vertex is at the right level (EXC: dense jungle, where the LOS is clear for a snap shot).

JR
 

DougRim

Senior Member
Joined
Apr 23, 2012
Messages
466
Reaction score
546
Location
Ottawa
Country
llCanada
It would be a clear LOS except for the woods. The woods should block LOS even though the vertex is at the right level (EXC: dense jungle, where the LOS is clear for a snap shot).

JR
Can you elaborate as I don't understand this. Why would the woods (which are level 1 obstacle) block loss to a level 1 (as per A4.34) vertex? And why wouldn't a dense jungle (level 2) block LOS? There is something I am missing here.

B10.1: ". . . Other terrain (e.g., grain, brush, woods, building) is at the higher level throughout the entire depiction of the terrain in question for LOS purposes (but the actual Crest Line is always used for movement purposes), even if it appears to be rising from the lower level portion of the hill hex [EXC: Newer boards may depict visible Crest Lines beneath this other terrain (EX: 61F8), in which case the actual Crest Line is used to determine LOS as is the case with Inherent Terrain]3A."

Footnote B3A: 10.1 HILLS: In addition to having visible Crest Lines beneath some terrain, newer boards are much better at depicting where Crest Lines actually are by using gaps in the terrain. In many areas of these boards, most players will be able to agree on where the Crest Lines actually are underneath the other terrain of grain, brush, woods, or buildings even without visible Crest Lines. When players can so agree, we encourage them to use the actual Crest Lines to determine LOS.
 

jrv

Forum Guru
Joined
May 25, 2005
Messages
21,998
Reaction score
6,145
Location
Teutoburger Wald
Country
llIceland
Can you elaborate as I don't understand this. Why would the woods (which are level 1 obstacle) block loss to a level 1 (as per A4.34) vertex? And why wouldn't a dense jungle (level 2) block LOS? There is something I am missing here.

B10.1: ". . . Other terrain (e.g., grain, brush, woods, building) is at the higher level throughout the entire depiction of the terrain in question for LOS purposes (but the actual Crest Line is always used for movement purposes), even if it appears to be rising from the lower level portion of the hill hex [EXC: Newer boards may depict visible Crest Lines beneath this other terrain (EX: 61F8), in which case the actual Crest Line is used to determine LOS as is the case with Inherent Terrain]3A."

Footnote B3A: 10.1 HILLS: In addition to having visible Crest Lines beneath some terrain, newer boards are much better at depicting where Crest Lines actually are by using gaps in the terrain. In many areas of these boards, most players will be able to agree on where the Crest Lines actually are underneath the other terrain of grain, brush, woods, or buildings even without visible Crest Lines. When players can so agree, we encourage them to use the actual Crest Lines to determine LOS.
If you are playing footnote B3A then the LOS is clear. Footnote B3A is not official. As it says, "When players can so agree, we encourage them to use the actual Crest Lines to determine LOS" (my emphasis). Furthermore to be totally compliant with footnote B3A there has to be a crest line shown going through the woods, as is done on the new deluxe board l. Forcing the players to guess where the crest line is is not really compliant. I am not aware of another official board that shows the crest lines through anything but grain. There may be such an animal, but the one you are using for your picture is not it.

The official rule is B10.1: all non-inherent terrain is at the higher level. Even though it is not visible in the overhead view, there is a level one hill under *all* woods in the hex. If you play B10.1 as currently written, then the LOS is blocked. If you play footnote B3A and you assume all other players do too, then the LOS is clear.

JR
 

DougRim

Senior Member
Joined
Apr 23, 2012
Messages
466
Reaction score
546
Location
Ottawa
Country
llCanada
Thanks. This is clear. Since we don't have the capacity yet (as far as I know) to have VASL deal with two different rule sets, it seems to make most sense to me to code this as per the B10.1 EXC since VASL effectively draws Crest Lines beneath other terrain. All VASL maps are effectively the "newer boards" referred to in the exception. I will consult with others on this, but I am pretty sure this is how VASL functions. Does that seem sensible to you?

BTW, I am still uncertain about your original comments about dense jungle allowing a snap shot?
 
Top