A new test for scenario designers?

[THH149]

I was playing a 2 hour Black Sea scenario the other day and just felt like watching the battle unfold, with Russians attacking a US manned town with a fort (Fort Apache?).

Anyway, I was "playing" Blue by not moving a unit, I could watch Red force let their attack unfold.

Curiously, my immobile Blueforce won a major victory against the AI.  Surely that doesn't bode well for a competitive match against the AI or a human player!

Should designers have a new test of their scenario where they AI needs to win a handy victory against an immobile human defense? Could BFC use that test to qualify a scenario for inclusion in a module?

Disclaimer: I did move all the US reinforcements en bloc to the center of the objective in the town.

Disclaimer: the US artillery landed as an emergency maximum barrage on the centre of the objective in the town.

Best

THH

[chuckdyke]

5 minutes ago, THH149 said:

Should designers have a new test of their scenario where they AI needs to win a handy victory against an immobile human defense?

Once you see the AI attacking just go in hide and see how they dig their own grave. Studienka in RT if you play German just organize your defense first. Because the way it starts is downright silly. I will play that one on Hotseat to do the map some justice it is not a meeting game at the start. I love the map but a 3 hour scenario and try to make triggers is TBH outside the scope of the engine in my opinion. 

[RockinHarry]

Try this one (CMBN) with the do nothing method. https://www.thefewgoodmen.com/tsd3/combat-mission-battle-for-normandy/cm-battles-for-normandy/cmbn-v4-mg-vp-you-enter-germany-introduction/

[mjkerner]

You are devious, Harry!

[chuckdyke]

You can hide inside a bunker, but you can't hide the bunker. All joking apart hiding with close cover arc works well if you defend a house. (Better use the shift key too. 

[Combatintman]

6 hours ago, THH149 said:

I was playing a 2 hour Black Sea scenario the other day and just felt like watching the battle unfold, with Russians attacking a US manned town with a fort (Fort Apache?).

Anyway, I was "playing" Blue by not moving a unit, I could watch Red force let their attack unfold.

Curiously, my immobile Blueforce won a major victory against the AI.  Surely that doesn't bode well for a competitive match against the AI or a human player!

Should designers have a new test of their scenario where they AI needs to win a handy victory against an immobile human defense? Could BFC use that test to qualify a scenario for inclusion in a module?

Disclaimer: I did move all the US reinforcements en bloc to the center of the objective in the town.

Disclaimer: the US artillery landed as an emergency maximum barrage on the centre of the objective in the town.

Best

THH

In a word - no.  So long as both sides have choices, so in this case, if the defender had setup options then such changes could result in a different outcome.  I also dispute that an H2H game would result in the same result because of the variables involved having two human players.  A human attacker will be more competitive and tactically adept than the AI attacker.  Additionally, it is difficult to script an AI force to achieve a victory of any sort against a human player.  As JonS says in the scenario design manual, the AI force should present a challenge to the human player.  Generally if I can get the AI to achieve a draw then that is good enough.  There are too many people who, when making their first scenario, want all sorts of AI tools to force an AI victory and find that either they are not available or they are not skilled enough to make the AI do exactly what they want and end up giving up totally.

[Glubokii Boy]

I really wish that we could get some more types of triggers included in the editor and not just be limited to - units in zone -

Things like:

- AI-group casualty level higher then 

- Unit killed, Unit spotted

- triggered reinforcements

etc

That would make programing a GOOD AI so much simpler...

Adding some additional trigger types ought to be possible within the CM2 game-engine. 

What is really needed though to get the AI to perform very well is multi-option/branching waypoints to allow the AI to adjust its plan to fit the current battlefield situation.

This is by far the AIs biggest shortcomming imo...Being 'locked' to ONE way forward...or non at all.

Branching/ multi-option triggers will have to be something for CM3 though i guess but some more simple trigger options would indeed be nice even in CM2.

 

 

 

[chuckdyke]

Scout find the AI enemy wait a little till he walks inside your cover arc. A human player doesn't fall for it. Editing sounds like fun see how many people you can frustrate. 

[Combatintman]

8 hours ago, chuckdyke said:

Once you see the AI attacking just go in hide and see how they dig their own grave. Studienka in RT if you play German just organize your defense first. Because the way it starts is downright silly. I will play that one on Hotseat to do the map some justice it is not a meeting game at the start. I love the map but a 3 hour scenario and try to make triggers is TBH outside the scope of the engine in my opinion. 

That was not my experience and as @George MC, whose scenario this is and is one of the most accomplished scenario designers out there, the balance of probability of your opinion being correct is pretty low.

45 minutes ago, chuckdyke said:

Scout find the AI enemy wait a little till he walks inside your cover arc. A human player doesn't fall for it. Editing sounds like fun see how many people you can frustrate. 

Editing is fun but if you set yourself the goal of frustrating people, no one will play your scenarios and will come onto the forum and rip them apart.  Why not try it first before feeding your compulsion to make random uninformed comments across multiple threads?

[chuckdyke]

11 minutes ago, Combatintman said:

the balance of probability of your opinion being correct is pretty low.

The first thing you do when you defend something is look for a plan to escape if it goes wrong. The first listening post didn't meet that criteria. I enjoyed the scenario and admire the author for the map and I don't know how he managed the AI plans for a three-hour game. I will replay it on Hotseat. I am looking at editing so what is wrong with that. I hope in time the AI will win some games. How did that platoon get out of that listening post? Platoon is too large a unit for that task. I had a scouting unit and a Panzerschreck there. Once the sound contact was good and solid, they got out. People buy the games to have fun, and sorry to disappoint you I have fun with BN, FB, SF2, RT&FR. Happy gaming. 

[Glubokii Boy]

Chuckdyke...

There are limitations within the game engine...with anything besides small scenarios it is difficult for any scenario designer to get down to the smalest detail...like having a tiny listning post in a huge, large scenario perform human like all the time...the tools are simply not there...

 

 

[chuckdyke]

3 minutes ago, Glubokii Boy said:

like having a tiny listning post in a huge, large scenario

I played for German and took the platoon out on the left-hand side of the map. All I wanted were the sound contacts so I could read the AI plan, so the Panzerschreck and scouting 3 men party was sufficient. I know large scenarios have large plans and a human player can beat it with macro-management. It plays out like that in real life too, one veteran shot a 98K in the air and thereby held up an armored company for three hours. That happened 76 years ago. We start the game on Hotseat next week I am looking forward to it. 

[Bulletpoint]

10 hours ago, THH149 said:

Anyway, I was "playing" Blue by not moving a unit, I could watch Red force let their attack unfold.

Curiously, my immobile Blueforce won a major victory against the AI.  Surely that doesn't bode well for a competitive match against the AI or a human player!

I come to the opposite conclusion. If you design a mission where the AI attacks and can win, then that is likely not very balanced for play against a human opponent.

[chuckdyke]

Play 'Drive Them Out' in FB. If you play as the German defender, it is easy and boring. However, kudos to the designer if you play the US attacker. The AI because of an earlier conversation I tested this scenario, it just has one halfhearted attack. Some games are just not designed one way or the other. As German defender you have a major victory by just staying put. To play as US attacker it is a challenge. 

Edited May 24, 2021 by chuckdyke

[Artkin]

Hey, take it to the pbem boys.... 

[StieliAlpha]

Eh, why don’t you just leave it to the guys designing new scenarios?

I would assume, their intend is to have fun with the design work and I am quite sure they spent a considerable of their free time doing so.

If we (and you) like the result: Great! And don‘t forget to say „Thanks.“

If not, shut up or make a helpful comment to the scenario designer.

As others stated, even an assault against a well placed, completely motionless Defense can be an extremely challenging task for the attacker and may be thrilling for the hard pressed defender.

Edited May 25, 2021 by StieliAlpha

[domfluff]

A multiple-hour game where you do nothing, make no decisions and wait for the screen that says "you win" sounds the opposite of thrilling to me.
 

On 5/24/2021 at 12:39 AM, THH149 said:

Should designers have a new test of their scenario where they AI needs to win a handy victory against an immobile human defense? Could BFC use that test to qualify a scenario for inclusion in a module?

I think this is worth giving some thought to, because I'm not sure the answers are entirely obvious.

The AI is always going to be worse than a competent human, all things being equal. That means that a scenario where the AI is supposed to be a real challenge will inevitably require some massaging - increasing their force ratio, improving soft factors or use of prior knowledge.  I do think it's important not to overstate that - the majority of the AI's fighting behaviour is governed by much the same systems as the player, so a good AI plan is really about co-ordination and movement - getting into a strong position, and then relying on the TacAI to handle the exciting bit.

That also means that scenarios specifically designed to be played PBEM are going to be a different breed from those which are designed to be playable from one side vs AI, which has to be different to one which should be playable from either side.

My thinking on this has been to lay out the defenders first, since the defensive AI plan will generally be easier to manage, then to play through the scenario as the attacker. My intention as the attacker is to win, and also to attack in a plausible and hopefully doctrinally-correct manner. I then try to replicate my own attack using the AI plan as best I can - that way, even if the AI can't manage this efficiently, it should at least look vaguely sensible, and as if it's moving with some purpose and intent. This can then go back and forth as many times as needed to tweak and improve things.

That's often what AI programming is about, in general terms - not necessarily being intelligent, but in hiding the stupid as much as possible. If the AI loses, but doesn't do anything egregiously stupid in the process, then I think that's a reasonable outcome to aim for.


So yes, I think the answer is "no, it doesn't matter", but the reasoning behind that is a little more complex.

[THH149]

Maybe my original point got lost in translation....

In the scenario Fort Apache, the Human Player (me) won a US major victory against the AI without me (the Human player) moving a unit from the set-up location (and didn't adjust the set-up locations either).

There was no gaming the system to crush an attacking AI, or other tricks used.

In other words, there was no competent human player and they still had a major victory.

I'm only raising it 'gently' as a helpful and practical tip - to playtest scenarios a little differently - to create great tense and replayable scenarios that will draw players to the community who will play more games. A vibrant and active design community is really important too, but there's stuff to learn to make better scenarios. 

Perhaps another way to think about the issue is to consider the skill level needed to play either side in this scenario:  Beginner for the US and Expert for the Russian.

BTW, I like the scenario, well done to whoever designed it, it's got a great premise.

 

 

[domfluff]

Yes, and I do think the fundamental point sounds sensible intuitively, but that's not always possible - I'm suggesting that the AI should at least attack in a plausible manner, and whether it would actually win is a secondary concern.

If the AI attack replicates a winning line, then at least it's "trying to win".

I agree that would be an ideal, but scenarios perform multiple roles, and one which is designed for h2h play may be a lot more difficult for an AI script to win. In that situation, you're often going to limited to doing something which looks sensible, and hoping for the best.

For a scenario that is intended to play against the AI, certainly, but then you can cheat a little and give it advantages that would unbalance a human opponent.

[Combatintman]

Yep - I agree with @domfluff - there is no misunderstanding of your original point.  I agree that maybe that scenario could have included something in the description along the lines of Beginner for US and Expert for the Russian but then of course there are players who have an inflated sense of their skill level or comedians who leap straight into Iron mode the first time they play the game.  It is also important to remember that the Steam hookup with Slitherine will be bringing in new players who are unfamiliar with the whole Battlefront system and probably need a few easy scenarios that they can win without doing too much in order to get to grips with the game.  If they keep bumping up against scenarios they can't win without playing it more than a couple of times, they are unlikely to keep playing the game or buy another module/title.  Not every scenario or campaign has to be pitched at challenging genuinely experienced players.  How many threads have you seen recently along the lines of '[name of scenario] help needed' - or 'I cannot get past mission 3 in the [name of campaign] - Please help.'  Or look at the thread asking how many times people play a scenario in order to win it - the balance of opinion there seems to be I just get frustrated and move on or I had to save every turn and replay and I feel soiled.  My stuff is generally pitched at the novice to average player mainly for this reason and the fact that the main tester (me) is a middling player.  I would rather kick a scenario out of the door that can be won first time rather than making it difficult for experienced players and then having to keep coming onto the forum to explain how to win it.

Otherwise, the points that both domfluff and I raised about design philosophy, your target audience in terms of H2H, one side vs the AI or both sides vs the AI, and the constraints that you have to work with in the editor are extant and valid.

[dbsapp]

12 minutes ago, Combatintman said:

Yep - I agree with @domfluff - there is no misunderstanding of your original point.  I agree that maybe that scenario could have included something in the description along the lines of Beginner for US and Expert for the Russian but then of course there are players who have an inflated sense of their skill level or comedians who leap straight into Iron mode the first time they play the game.  It is also important to remember that the Steam hookup with Slitherine will be bringing in new players who are unfamiliar with the whole Battlefront system and probably need a few easy scenarios that they can win without doing too much in order to get to grips with the game.  If they keep bumping up against scenarios they can't win without playing it more than a couple of times, they are unlikely to keep playing the game or buy another module/title.  Not every scenario or campaign has to be pitched at challenging genuinely experienced players.  How many threads have you seen recently along the lines of '[name of scenario] help needed' - or 'I cannot get past mission 3 in the [name of campaign] - Please help.'  Or look at the thread asking how many times people play a scenario in order to win it - the balance of opinion there seems to be I just get frustrated and move on or I had to save every turn and replay and I feel soiled.  My stuff is generally pitched at the novice to average player mainly for this reason and the fact that the main tester (me) is a middling player.  I would rather kick a scenario out of the door that can be won first time rather than making it difficult for experienced players and then having to keep coming onto the forum to explain how to win it.

Otherwise, the points that both domfluff and I raised about design philosophy, your target audience in terms of H2H, one side vs the AI or both sides vs the AI, and the constraints that you have to work with in the editor are extant and valid.

Actually, the Iron mode is much more easy than "basic training".

Speaking about making new players stay in the game, I would recommend to do something with basic training skill level. I started playing on it and only my persistence and geniune love for war (I'm a pacifist) made me continue playing.

On basic training AI gives real hell with 2 minutes wait time for artillery. It's like constantly being lased from drone and suffer from innumerable artillery shells. Frankly, it's no fun. 

Playing on Iron after that feels like the breath of fresh air. 

And I'm far from having inflated sense of my skill level. I know for sure that I suck at this game

[markshot]

Are you guys saying that I could finally win a scenario?  I thought I needed to be, at least, as smart as a lab rat.  THIS IS GREAT NEWS!

 

AI - LOOKOUT - I AM COMING FOR YOU!

[domfluff]

I think the broader point is that scenario design is game design, and game design is both subjective, and always (always) harder than you think it is.

As a very recent example - Between Two Fahrbahns in CM: Cold War. That's an entertaining scenario which can be played from either side. It's not massively challenging, and the decision tree is pretty shallow, but it's been a lot of fun to knock around in, and is a well crafted thing. Well worth playing at least twice, from each side.

Is it a "bad" scenario, objectively?

There's nothing really "Cold War" about the scenario, in terms of getting across some broader conceptual or tactical point. This isn't an application of doctrinal Soviet Meeting Engagement, a demonstration of Active Defence doctrine, or a demonstration of the development of technology over the period, for example. You could replace the T-64s and M60A2s with Shermans and Panzer IVs and have essentially the same scenario. 

Does that matter? It really depends on your intent. Those concepts are clearly at the core of Cold War's release. All of those concepts are explored thoroughly in the (excellent) campaigns, so does every scenario need to do the same?

That kind of scenario isn't really what I'm looking for, personally, but it's a short, simple knock about that lets you play with some new toys. That's definitely not invalid, and regardless of whether this is objectively great or not, I know I feel like I've gotten my money's worth out of it.

I think there's a large chunk of design space there, and there's space for a lot of options. I do agree that sometimes these are communicated as well as they could be - either in relative difficulty levels or in terms of design intent, but since that's even more subjective, that's possibly pretty meaningless.


So yes, to the actual question - I think that's a logical and intuitive idea, but I also think there are some complex reasons why that might not be possible or even that important, for all situations or at all times. There's a long standing argument about difficulty in game design - there are definitely games (e.g., Cuphead, Dark Souls) which use their difficulty as a part of the appeal, but since each scenario is a game design exercise to itself, that means that you're going to find a wide range of intents and outcomes.

Now, I *do* enjoy that kind of game, and I'm more than happy to have an AI opponent (or ideally a human one) kick my arse repeatedly, especially if there's something I can learn from the experience. There's a roguelike mentality of failing forwards there which I can really get behind, but I can also appreciate that this is a niche - not everyone signs up for that. Personally, I tend to prefer steeper challenges, or even no-win situations, but I also know that that's not true for everyone else.

[chuckdyke]

The reason I play on Iron is spotting. Turn is finished and before you press the 'Big Red Button' you can see exactly who is in contact with his HQ. A stand for Administration and Communication.  last two letters of SMEAC. Something I take serious in this game. One thing I like about the US their system of abbreviations. Makes it easy to formulate actions. 

[markshot]

1 minute ago, chuckdyke said:

The reason I play on Iron is spotting. Turn is finished and before you press the 'Big Red Button' you can see exactly who is in contact with his HQ. A stand for Administration and Communication.  last two letters of SMEAC. Something I take serious in this game. One thing I like about the US their system of abbreviations. Makes it easy to formulate actions. 

That is a good point.  I have been on "Warrior" as I found seeing my own forces was a pain, but I guess "Iron" means you can determine more easily who knows what without reverse checking off of clicking on enemy units?  If your forces are in touch, then broadly speaking they have a similar picture?