Jump to content

benpark

Members
  • Posts

    4,730
  • Joined

  • Last visited

  • Days Won

    27

Posts posted by benpark

  1. City-type modular building textures, horizons, some Flavor Objects (S-Bahn sign, and maybe a few others) all use those tags.

    Rubble walls are a Map element, and work regardless of Mod Tags - though you could Mod Tag it as something like a stone, wall, etc.

    You will also see something like [hungarygermany] - that one does Hungary, and keeps the German-style churches. Mod Tags also work with models, so this allowed an easy swap to Hungarian terrain with a few extra textures.

  2. It is decidedly tough to pull this off in a way that looks realistic (if the player indeed sees it). Timing and the pace of the order seems the best combination to focus on. I agree with others here - It does help to divide the forces into smaller, and if possible, individual units for the AI Plan for this.

    Maybe try:

    -Give the lead tank a "Dash" order at 0:00 (let's just use this time for example)

    -Second tank a "Quick" order at 0:30

    -Third tank "Advance" at 1:00

    And so on, through the group. The timing helps avoid the units bunching up and competing with each other.

    Shorter spacing between AI orders will help them stick to the road. Take care as to about when you will have contact with the player. If the goal is to simulate an ambush, the AI can barrel through. If the goal is to have an AI that is scanning for an enemy, having the individual units fan out at some point will probably help their survivability when the enemy shows.

    The terrain on the sides of the road can help or hurt the AI's chances of the tanks staying in line, or also being able to go around if their compatriots are knocked out.

    33 minutes ago, domfluff said:

    give the crucial stuff more detailed, individual plans, but letting the AI manage the broader stuff more loosely.

    Sound advice. Most of the best working AI plans allow the AI some choices. This is one area it needs some direction and focused AI location watching, in my experience. Think like a player/historical counterpart when doing any plan - Roads are both useful in different conditions, and a beacon for an enemy that will always expect them to be used.

    Last would be to watch the Height values (hotkey "e" in Map Editor/AI Editor). That is very helpful for looking for hull-down positions, or covered approaches for the AI.

    Then, Scenario Author Mode/Turn-Based, and run and re-run the thing, and adjust, ad infinitum. Until it works, or doesn't!

  3. That would probably be the case, as the Shader file has numeric values for Red, Green, Blue colors.

    If the underlying colors change (the mod has a different set of hues), then the Movie Mode shaders will show a different set of colors than intended. The Shader interpretation remains constant, as it is just a set of coded color values.

  4. 45 minutes ago, Pete Wenman said:

    That could actually come in handy on occasion.

    Indeed! - there's always something to be mined from the "problems". That particular one (AI plan divided between two AI painted areas) could be used for splitting of a random amount of tanks to help an infantry AI Group, or something of the sort. That would need testing, but this is the sort of thing where the creative decisions can be made.

  5. Just now, Canuck21 said:

    which psychiatric institution they've sent me to

    We have a wing dedicated for that here, step right in.

    The map setup zones are a good way to direct the player's attention, or misdirect it. The ones you have might be a bit revealing as is - don't be worried to overpaint those, as players could use them - or be lead to think the attacks are coming from elsewhere.

  6. I may see what you mean - Some units start heading for one zone or another once the Go button is pushed. That is due to the Setup Zone AI areas that are painted being split between two map setup zones, so when the AI gets it's first AI order, it needs to regroup itself (into the entire AI Group assigned) to move on to the next order.

    Assign the Setup AI Plan to one zone or the other, and you should be good.

  7. 10 minutes ago, Canuck21 said:

    Ok, so am I right here in saying that you paint AI Setup Zones outside the Game Setup Zones you make when in Map Mode

    The AI painted areas must be within the map setup zone. Check that you don't have a stray yellow AI placement painted outside the map zones. I believe that can drop AI orders.

    If you want a non-setup zone AI setup, you should do hand-done placements. The engine respects those.

  8. 9 minutes ago, Canuck21 said:

    Now you stated Ben, that you never let the game deploy the units, which I understand, but you also state that you always do a "hand setup in 3D View" but for AI Plan 2 onwards, you paint in the setup (yellow) zones. That leads me to believe that all Plans start from Plan 1's position then reposition automatically for subsequent plans.

    I do let the game deploy, only if I have painted the AI Zone - just not within the color zone. That's the alternative to hand-placing the units - which does not need painting (if there is, that will be followed, rather than the AI painted ones).

    Pete and I are on a similar track, once again!

  9. I think we all like to see people interested in the Editor. Ask away.

    Setup can be done by painting an AI area in yellow, or by hand placing the units individually.

    I generally use AI Plan 1 for the hand-set-up plan, and go in and place each unit in the 3D view.

    AI Plan 2 (and any after that) gets the AI plan setup, where I paint it onto the map. I never let the game engine drop forces - as you see, it is overly randomized.

    The one thing to watch is getting your AI movement orders lined up correctly with your AI Plan 1 (if using the method outlined above). Since that isn't painted (which retains the hand-placement).

    I have a Photoshop template that I do for complicated stuff like that, and load it as an Overlay when I'm doing the AI Plans. Way easier than remembering!

  10. 17 minutes ago, BFCElvis said:

    Or are Steam games somehow multiplatform when run through Steam

    They can be, but its still two OS versions "under the hood" that would need be installed and downloaded separately. They do share a Steam key, however - which is appreciated by people with both operating systems. One purchase.

    The BFC CM games that Slitherine does production for on Steam are currently Windows only (as far as the compatibility icon goes on Steam, at least).

  11. 5 minutes ago, Commanderski said:

    Did you get access to upload the modified campaign yet?

    The patch will incorporate some time extensions that should alleviate the early campaign time crunch, while still maintaining the AI's abilities and the overall plan for the battle depicted. Best to do it that way, so everyone gets it in one place.

    Shouldn't be too long on that, I'd think but Elvis will let us know when it is very very ready.

  12. It's partly based upon how a player or commander in the situation would hold a reserve back for committing when the enemy may be wearing out. Fifty Minutes into the scenario should still be time for having something interesting happen, if possible - and sometimes it takes some guys holding back to get that done - there's probably plans for them. If you wiped everyone else out, that's pretty decent.

    I've tried to keep the mid and late-game interesting - that can lead to what you are seeing if it ends early. These units are probably tasked with later times to move out, so the German player won't know where the attack really ends. There are a multiple, variable AI Plans, so that also changes attack routes and times, etc. The goal is keeping it interesting from either perspective over multiple plays.

    You will find units holding back in some scenarios if you peek behind the mothballed - curtain of Cease Fire, they should only be heading out later in the attacks in these new scenarios and campaigns - generally as reinforcements, but sometimes on other tasks. Things like assigning the same AI Group to a late group as an earlier one to act as a second assault wave also may also be the reason. Lots of reasons.

    Possibly an oversight among the many plans, and another opportunity to extend the attack and chaos? Maybe - but the AI uses as many tricks as possible (as realistically as possible, where appropriate. If there are any attackers left hanging around in bunches by scenario time-end, let me know and I'll give them something to do. The Soviet attacker is not going to mess around as it is, however.

    This is the template I've used for keeping track of everything when it gets complicated (created after the Berlin ones, sadly)----

    Slight spoiler ahead of one that absolutely has some differences with the finished product, and is missing things like time notations, etc....

    I then load this in as an Overlay while I do the AI Plans

     

    1415162250_Castlebattle2a.thumb.jpg.5856cb735901e071c3ed76eee7634106.jpg.5f75a0d64ed33fef884a931dfa2e2d50.jpg

     

    Slight spoiler about the scenario from German side....

     

    There will be some Soviet holding force left on the North river bank, for welcoming especially enterprising German players that may want the Lehrter Bahnhof back, but that doesn't necessarily seem like them.

  13. Be sure you have the two tiles alongside the tile with the bridge entrance set at the same height. Also double check that they are the same on the other side. Diagonals can be tricky, but just imagine a straight line through the three tiles (bridge and two adjacent).

    If you post a screenshot with the Elevations switched "on", that may better help decipher what may be happening.

×
×
  • Create New...