Streamlined Turn-Based (pure TB while limiting problems?)

[Human Shield]

I've been thinking about this for a while (I realize it probably isn't that original), and more so after this great article.

This idea is best for a single player-character and probably wouldn't work for MP.

Look at these two descriptions of battles:

Battle 1 (Normal turnbased as in Fallout)

Player fights enemy A, B, C, D. Everyone has 50 hp.

Player's turn: Clicks to run - Runs for 5 steps; Clicks to shoot- shoots A for 30 damage, Clicks to shoot- shoots B for 50 damage and kill
Enemy A: Shoots player for ten damage; Shoots and misses
B: Dead
C: Shoots burst and misses; Shoots and hits for ten damage
D: Runs for 15 steps.

End of round- PC: 30 HP, A: 20, B:0, C:50, D:50

Battle 2 (Modified)
Player fights enemy A, B, C, D. Everyone has 50 hp.

Player chooses through interface to walk 5 steps, chooses to attack A (given same percentage with movement in acount), chooses to attack B.

Player runs for 5 steps, at the same time D runs for 15 steps, and C fires a burst and misses at your character while he is running.
Player turns and is hit by A for 10 damage.
Player smoothly moves out of hit and fires at A for 30 damage.
Player then dodges a shot by A and is hit mid dodge by C for 10 damage.
Player fires at B and kills him.

Player goes back to interface.
End of round- PC: 30 HP, A: 20, B:0, C:50, D:50 (all positions the same)

Now, which seems more entertaining with less downtime?

This may seem like phase-based but it is turn-based. When the player selects to move the computer already places him there, all his attacks are calulated through. The difference is the animation.

If the player selected to attack enemy D in the above, then enemy D would not move until he is hit, and the AI would probably react differently before playing out the scene. It is not phase-based because the two sides are not planning at the same time, the effects that the player sets in are constant. You don't have the problem in phase-based of an enemy moving behind a wall after you selected to attack him.

The engine would have to coded to be able to play back actions in the best order but it is doable. It is just playing out the actions differently and would be easy to toggle on and off.

The big problems I can't get around yet (feel free to add it) are line of sight and interupts.

With a pre-action selection system new items revealed with movement would be unknown to the player in the planning stage. This could be part of the game to take smaller steps for caution (extra AP into AC or whatever). In Fallout 1 enemies weren't hidden by doors or corners so it could not matter. Or if a new enemy is revealed the players can be given a choice to continue with first plan or change it, although this would mean that no enemies would move before the player is given that choice.

Interupts could be part of the game with enemy interupts calulated and play out as attacks with the animation (wouldn't let player retreat if an interupt did too much damage). Fallout 1 didn't have interupts so it could not matter. Or let the player have a choice after the enemy interupt to change plan (enemies wouldn't move until after choice).

 

[Saint_Proverbius]

I fail to see how this isn't phase based. If you're planning all your moves at the beginning of the round, then you're definitely going to have the same issues as phase based does simply because the AI can chose to move in a way that's counter productive to the set of instructions you've given it.

In your example, the player selects to move, then shoot A, then B. Then he hits the Run Battle button and everything goes forward, right? Well, in the time it takes him to move forward five steps, why couldn't A or B have moved to cover? In the time it takes for him to walk five steps and shoot A, why couldn't B have moved for cover?

Having the AI simply not move those selected to hit would end up being a hughly exploitable system. You'd end up with the player not only having direct control over his character, but a quasi-method of the player actually controlling the enemies as well, since he could fire at an enemy and prevent it from moving to cover.

 

[Psilon]

Yeah, this is phase based with simultaneous turns. See Laser Squad Nemesis for another example.

 

[Human Shield]

I am telling you guys it is pure TB just animated differently, not phase-based.

The enemy is not planing at the same time you are. Your planning phase is played out in the background by the computer and replayed after merging unrelated actions to get the same result in less time and look cooler.

In your example, the player selects to move, then shoot A, then B. Then he hits the Run Battle button and everything goes forward, right? Well, in the time it takes him to move forward five steps, why couldn't A or B have moved to cover? In the time it takes for him to walk five steps and shoot A, why couldn't B have moved for cover?

In Fallout when you move 5 steps and attack, the enemies don't move until your done.

The system merges non-conflicting events only. If an enemy runs after your turn is done in Fallout and you didn't target him then his movement is uneffected by you, why not play the animation at the same time if they don't effect each other?

Having the AI simply not move those selected to hit would end up being a hughly exploitable system. You'd end up with the player not only having direct control over his character, but a quasi-method of the player actually controlling the enemies as well, since he could fire at an enemy and prevent it from moving to cover.

Enemies have to stand still in turn-based now and wait for you to fire at them.

 

[Montez]

It seems like this system would take even longer than turn-based, as first you have to enter your moves for the round, then watch those moves be animated. Like I click on my character, drag the mouse to where I want him to move, then target A and B, then click "Run Battle" and watch it all repeat except with animation - that seems like twice as much time without any benefit except neater eye-candy.

 

[Human Shield]

Maybe this would help people: when you click to move a ghost image appears in that spot.

It seems like this system would take even longer than turn-based, as first you have to enter your moves for the round, then watch those moves be animated. Like I click on my character, drag the mouse to where I want him to move, then target A and B, then click "Run Battle" and watch it all repeat except with animation - that seems like twice as much time without any benefit except neater eye-candy.

But you don't watch the enemies move around for a long-time. It is planing- watch move- planing not move- multiple enemies move- move.

It would be shorter

TB:
Click somewhere (time X)
Watch move (time Y)
Click (time X)
Watch (time Y)
Enemy moves (time Z)

In this:

Click out actions (2X)
Watch ( (2Y+Z)/2)

That got a little weird, but the enemy moves are combined into your own.

 

[Spazmo]

I am telling you guys it is pure TB just animated differently

If it's the exact same thing but "animated differently", why bother?

 

[Human Shield]

If it's the exact same thing but "animated differently", why bother?

Because it could be a more active system which is what people complain about turn-based, less down-time.

 

[Saint_Proverbius]

Enemies have to stand still in turn-based now and wait for you to fire at them.

Yeah, but the difference is they still get to move during that round. In your system, they have to wait to move until after you shoot them. Say you have ten seconds for the animation phase of your system. The play has 20 action points.

The player moves 10 units, then shoots A(5APs) then shoots B(5APs).

• 1-5 seconds: Player is moving 10 spaces. A and B are locked in place because the player has selected to shoot them, and all hits are determined based on the range Player will be at when 5 seconds is up.
  
• 5-7.5 seconds: Player shoots A. A can now move because the Player has shot him, he can now go behind the wall if he has half of his remaining APs left. B is still fixed in to place, waiting to get shot.
  
• 7.5-10 seconds: Player shoots B. A does his thing. B is now free to move with only a quarter of it's action points.

See what I'm saying? Because you have the rolls and everything computed during the player's planning phase and then locking the AI so they have to be hit based on their location, you're making a system where the AI is forced to just stand there and wait. Since the round can only last a set time frame for everything to be completed, you're allowing the player to burn the AI's APs. The enemy AI won't get to move later in the phase because the animation is in real time and would require a set timed round.

 

[Human Shield]

I never mentioned that the action phase has a limit on AP or has a set time.

All enemy actions are the same as acting by themselves but instead of locking the player in a spot and watching the enemy shoot him, the enemy moves are mixed into the player's (as long as end result is the same).

 

[Saint_Proverbius]

I never mentioned that the action phase has a limit on AP or has a set time.

All enemy actions are the same as acting by themselves but instead of locking the player in a spot and watching the enemy shoot him, the enemy moves are mixed into the player's (as long as end result is the same).

Then you'd end up with a situation where the round time is much longer than the allotted time for the player to use. Instead of the 10 second round taking 10 seconds at .5seconds per AP for the player, you'd end up with 10 seconds for the player stuff + 7.5 seconds for B to do his stuff, completing his AP usage. That wouldn't be streamlined at all.

You can't just mix in the enemy AI stuff to the animation either, since the to hit calculations are based on the range, so the AI would have to be fixed at that range when the player does his volley. If you're forcing the AI to wait there to get shot, which we've already established is what you mean when you said this happens in turn based anyway, then you're just talking about delaying his movement until after then. Basically, you're talking about a system with even more downtime than vanilla turn based.

 

[Human Shield]

round time is much longer than the allotted time for the player to use. Instead of the 10 second round taking 10 seconds at .5seconds per AP for the player

I don't understand, the player has all the time in the world to complete his turn in games like Fallout. Are you talking about gametime being effected or realtime? This system might take longer then maximum animation speed and completing turns in a flash but it would seem more entertaining.

Enemies that he isn't interfering with would be unaffected. For a large fight, 3 enemies would move at the same time with a player, 5 enemies could fire at the same time.

I don't see what is so hard to understand. Instead of enemy A shooting THEN seeing enemy B move, you see enemy A shoot AND enemy B move. Non-conflicting attacks will be merged, time watching would be less and more entertaining.

 

[chiefnewo]

The problem is, the player is affecting the enemies simply by *being* there. If the player is running around during the round,
it changes the ranges the computer was going to be firing at, leading to two situations:
The computer just uses the original range from the start of the round, leading to weird situations where the player can be shot when it appears he has run outside the range of the weapon.
or
The computer calculates a new range based on where the player is at the time the computer shoots in the action stage -- in which case you might as well make it real time from the start.
Well there is a third solution, which is to freeze everyone until after they have finished their action for that turn depending on sequence or whatever,
but it seems to me that would be very difficult to implement and you might as well go back to regular turn based.

So while it sounds like a good idea, it may be too fiddly to implement in practice without seeing any great gain.