This is response to a comment made in the replies on another topic.

I chose to use square grids to speed the game. (For me this being a solo game). The experience for this post mainly comes from ECW wargaming.

From what I found on various blogs (but mainly regarding 'Portable Wargaming') the terrain features fill the grid in which they occur. So if a unit moves from a grid in the open to a grid covered by trees then it has 100% moved into a wood: there is no 'almost'. Although it looks strange hills also cover the full grid and again there is no 'probably' about being on a hill.

Selection of what terrain appears on the battlefield tends to be done by a random method. I have divided my 10x10 grid board into four 3x3 grid corners with a central 4x4 zone which then gives 9 zones. I have used two different methods to randomly select the terrain features in each grid zone:
Method A] I have a list of terrain options numbered 1 to 12 and a D12. Roll for each zone.
Method B] I have generated a stack of terrain cards (some with more than 1 feature). These are randomly dealt onto the grid zones. (Plus a spare to swap out any 'howlers').

'Fiddle factors'
1] Sometimes it does not work: e.g. hills vs. rivers. Re-roll the D12/ use the spare card.
2] Neither of my methods specifies the exact location of the terrain feature within the zone. I will (very un-randomly!) move the terrain pieces with in the zone to give a degree of 'flow' into the 'landscape'. (Example: farm in one zone, orchard in a bordering zone & so move both so they share a grid side).
3] the relative position of elements in adjacent zones can achieve the transition of an open are - good for cavalry - or introduce a pinch point which will constrain cavalry maneuver.

Method C - previously used:
>   Select a terrain card (these only having one terrain feature listed).
>   Roll a D10 twice. Using the same datum corner the first throw is 'across' and the second is 'up' and place the terrain feature centrally on that grid. (It may be a large hill which takes up 15 grid spaces).
>   Repeat the process for 10 terrain cards.
The downside? It's almost too random. I have used this method and got the majority of the 10 terrain features in 2 of the 9 zones.

Currently I use Method B. A key factor is it takes less time to set up the game compared with the other two methods.

I am interested in what terrain selecting methods other gamers use.

[Apologies if this post is teaching grandmother's to suck eggs, etc]

I'm not sure if you've seen Bob's terrain generating ideas or not, but worth checking them out:

http://wargamingmiscellany.blogspot.com/2019/10/my-terrain-generator-version-2-feedback.html

For solo games I just set up terrain that looks right (to me) for the area the game is set in.

I number the table edges, roll a D4 and the defender sets up on that edge.

Works for me.

I tend to use a few different methods

1) what looks good
2) I sometimes roll a couple of d10 to give an overall idea of the amount of cover or hills
3) use a set of tables my mate has come up with - these have different values for different basic terrain type (coastal, hills, forest etc). They are skewed to put terrain around the edges to leave space to fight
4) the map from a scenario - even if we don't use the scenario

We simply divide the available terrain pieces into two fairly even piles.   Then each side takes it in turn
to place a piece.   The key is that neither side knows which table edge they will deploy on which keeps
us relatively honest.

The alternative is for the game 'host' to set out the terrain and for the visiting player to choose a table side.


Phil

Thanks for all the replies and ideas. (Apologies for the slow response: family, Christmas, business & charity 'challenges' & I dropped from 2 or 3 solo games a week to 3 in total for December).