What is the verdict QB players?

[rocketman]

I have never gotten into QBs vs AI in any CM games. Most in part to the sheer amount of content that keeps me busy. But also because the AI force selection has caught quite some flak over the years for weird selections. But I'm getting more interested as you can pick your forces and play around with stuff. And I rather not pick the enemy force since it breaks the suspense/surprise factor. So what is the verdict on FB - is force selection good most of the time? Better in defence situations? Please comment.

[A Canadian Cat]

Its kinda a relative thing.  CMFB's force selection is decent - certainly better than CMBN was when it first released.  The AI used in the QB maps is way better now than it was.  If you were going to play QBs against the AI CMFB or CMBS would be the titles to do it with.  But frankly QBs against the AI are not as good as scenarios against the AI and pale in comparison against humans.

Probably the best experience would be to attack the AI in a probe battle.  I only play QBs against the AI for testing so my opinions might be a little stale.  If someone comes along who plays that way regularly I would weigh their opinions higher than mine.

[MikeyD]

I think its a testament to the ever-improving quality of QB that so few 3rd party scenario are being constructed.

CMSF QB was rudimentary so if you wanted to fight two companies over a farmhouse you build your own scenario. And if it turned out okay you submitted it to Repository. Now if you want to fight two companies over a farmhouse you select one of the dozens of excellent QB maps with farmhouses on them, select any force mix you want in any season you want then hit GO. Sure, its going to be rougher around the edges than tailor-made AI plans to suit the specific encounter. But its 'good enough' to keep people happy, happy enough to not bother to construct make-your-own scenarios. Which is kind'a a shame actually. I always did enjoy those crazy 3rd party scenarios.

[AttorneyAtWar]

15 minutes ago, MikeyD said:

I think its a testament to the ever-improving quality of QB that so few 3rd party scenario are being constructed.

 

While its nice to look at with an optimistic view, I believe the reason we lack so many 3rd party scenarios is that there aren't too many people interested in doing them.

[Glubokii Boy]

The experience i've had with QBs vs the AI is not that the AI makes strange force selection (i have no big problems with having to select troops for it) but rather how it treats those troops during set-up and gameplay.
It seems to me that the AI is pretty much incapable of destinguishing between the different troops. It does not seem to make much difference to the AI if its a heavy machinegun-team or a mortar ammobearer-team for example.
Its simply a group of men...'Let's place these five guys here...and these 4 guys here.'

THIS ruins the fun in playing QBs Vs the AI more then anything IMO...

[Michael Emrys]

5 hours ago, RepsolCBR said:

It seems to me that the AI is pretty much incapable of destinguishing between the different troops. It does not seem to make much difference to the AI if its a heavy machinegun-team or a mortar ammobearer-team for example.

I haven't noticed anything like this myself. In fact, it seems to me that the AI is awfully canny at finding those keyhole locations for weapons teams that allow them to dominate important terrain. It does do other odd things, like expend all its mortar ammo blind firing on paths it thinks I might be taking on the chance that my troops might be there. In which case, I simply take a different path. This might inconvenience me some, but it isn't worth blowing its entire supply that it will need later for support.

Michael

[kevinkin]

 

5 hours ago, RepsolCBR said:

AI makes strange force selection

This has been a hot topic and high on many player's priority list. The developers are aware since it comes up fairly often. In the mean time, I like to pick the AI's OOB for dozens of QB battles and simply saved them for play at a later date. The OOBs I select for the AI are much less strange. My memory is OK, but tends to forget after a week which file has which enemy force structure. I may have an idea which file may have "exotic" equipment (e.g. King Tigers etc.). Not perfect. But something to try. I put together enough QBs for the entire year watching TV with the wife this January. Like eating pistachios, it's hard to stop with just one QB. 

Kevin

 

[MOS:96B2P]

2 hours ago, kevinkin said:

<Snip> I like to pick the AI's OOB for dozens of QB battles and simply saved them for play at a later date. The OOBs I select for the AI are much less strange. My memory is OK, but tends to forget after a week which file has which enemy force structure. <Snip>  

This is a great idea.  I may have to give this a try. 

[rocketman]

13 hours ago, MikeyD said:

I think its a testament to the ever-improving quality of QB that so few 3rd party scenario are being constructed.

I'd sure like to see more 3rd party scenarios (and I intend to make my own in due time) as well, but what makes QBs more appealing now is the outstanding map quality. Currently playing the La Gleize QB map in a PBEM and it is awesome looking in the snow.

[weta_nz]

First I'd like to thank the qb map makers and ai script writers. Quick battles are some of my favourite battles

i've had some really good 'infantry only' tiny - med probe type battles with me as the attacker. I generally choose my own troops and have had some surprisingly fun battles where (probably more by chance) the ai script reacts in a very realistic way counterattacking just when i'm pushing a little too much. Having said that the ai can't really follow up any localised tactical gains so normally i can reorganise and 'win' the scenario. However, i generally have a few rules i follow when playing the the qbs against the ai to make it more interesting:

1) Role play a bit, so i can't just totally change a plan

2) only area fire at enemy units with my units which have at least a ? (the exception is dense forest as it is almost impossible to advance without area fire)

3)-10% to 20% points which in small scenarios means avoiding unnecessary casualties is very important 

 

 

[kevinkin]

9 hours ago, weta_nz said:

However, i generally have a few rules i follow when playing the the qbs against the ai to make it more interesting:

Another thing to try is to set rules for self limited control over friendlies. For example, only plot those squads that have C2 contact (e.g. visual) to their platoon HQ every 3-5 turns. If they lose connection, > 3-5 turns. Depends on the battle length and your personal taste. This can make playing the defense in a QB a bit more challenging and the larger battles proceed quicker since you are self limiting many plotting activities. Re-playing community scenarios under these types of rules can be an interesting way to re-fight the ones you enjoyed in the past.

Kevin

 

 

[weta_nz]

Kevin, that sounds like a great idea to limit one of the big advantages we have over the AI which is our god like view of the battlefield and our ability to react

Edited July 17, 2016 by weta_nz

[Sgt.Squarehead]

On ‎16‎/‎07‎/‎2016 at 0:26 PM, Michael Emrys said:

It does do other odd things, like expend all its mortar ammo blind firing on paths it thinks I might be taking on the chance that my troops might be there. In which case, I simply take a different path. This might inconvenience me some, but it isn't worth blowing its entire supply that it will need later for support.

I strongly suspect that's down to the QB map's stock AI plan rather than the game engine.

[Michael Emrys]

15 hours ago, Sgt.Squarehead said:

I strongly suspect that's down to the QB map's stock AI plan rather than the game engine.

Oh yes, I agree with you. But since that's what the game comes with...

Michael

[MikeyD]

I believe on CMFB QB maps 'opening barrage' instructions were left out. In other titles a game often starts a few rounds being dropped on likely travel routes. Which isn't really incorrect. Artillery is often use for area denial rather than straight-up targeting. There's a 'harass' artillery command made just for that. A slow drip of artillery rounds onto an intersection to keep people away from it.

[slysniper]

Some interesting concepts mentioned to make the experience better.

I personally hate the game programming so much I will not waste my time allowing it to select a force.

That is enough about that. But I must admit I have not even tried in CMFB, because I have in every other module or game and it always manages to disappoint me pretty quickly.

 

Other than that, the rest of factors it normally does ok with and can normally place units and use units in somewhat of a realistic fashion.

But it is by far the lowest option on how to play the game, other methods of playing bring a higher quality of realism,  For me, I don't have the time to waste on this form of play when there is plenty of better styles to play the game by for the amount of time I have to give it.

But to each his own.

[womble]

If you want to quickly check how the FB AI force selection suits you, you can just keep hitting "Suggest" in the player-force-pick screen. If that's still not to your taste, it might at least be a way of coming up with the core of a rational force presence that you might not have considered, for you to trim and modify to something more effective.

[Sgt.Squarehead]

Again I suspect the disappointing nature of quick battles may be down to the very basic AI plans included with QB maps.....I Haven't spent much time in the CMBS editor, but in QBs from the other titles AI units are all attached to a single group and thus must all follow the same plan.  Most maps have two or three rudimentary but functional AI plans.

Edited July 18, 2016 by Sgt.Squarehead

[Kevin2k]

I hope the QB AI force selection of both sides will be improved in the future. Initially I set both sides to a 'mix' random force, but the results are often kinda weird. Either too much infantry or very little. The germans hardly get general purpose tanks, but instead a group of Hummels, SturmTigers, FlakPanzers 38t...

Edited July 18, 2016 by Kevin2k

[womble]

2 hours ago, Sgt.Squarehead said:

Again I suspect the disappointing nature of quick battles may be down to the very basic AI plans included with QB maps.....I Haven't spent much time in the CMBS editor, but in QBs from the other titles AI units are all attached to a single group and thus must all follow the same plan.  Most maps have two or three rudimentary but functional AI plans.

I believe MarkEzra has spent a lot of time incorporating the larger number of groups in recent "official" QB AI plans in recent titles.

 

[Sgt.Squarehead]

AFAIK the game can't allocate units to those groups in a QB, it has to be done manually in the editor.....I suspect his work may be aimed more at beginning level scenario designers like myself.

Edited July 18, 2016 by Sgt.Squarehead

[MG TOW]

I prefer meeting engagements as I don't have to be concerned about AI defensive positioning. As such, the AI in CMFB is pretty good compared to some earlier titles. Just choose a longer duration for the battle as the AI is cautious and takes its time on the advance.

[kevinkin]

The QB AI system has to accommodate a huge number possibilities. So it is hard to avoid situations where the forces, terrain, battle length etc.. selected by the player are out of synch with how the QB map was designed. But the system is pretty sophisticated and unsurpassed others past and present. 

Kevin         

[womble]

21 hours ago, Sgt.Squarehead said:

AFAIK the game can't allocate units to those groups in a QB, it has to be done manually in the editor.....I suspect his work may be aimed more at beginning level scenario designers like myself.

I don't believe that's so. My testing suggests that the game allocates formations, round-robin, to the available AI groups in the plan. And recalling my testing reminds me that (at least some of MarkEzra's) QB plans have had multiple AI groups in them from at least Market Garden, and possibly more like CMBN:CWF days. It is a huge weakness, though, that the game has no inkling of what troop types are good for; in one of my tests, the force selector chose an (immobile) AA battery, and plonked it into setup zones at the back with no FoF forward towards the enemy...

[kevinkin]

My experience mirrors womble's. The QB system will allocate forces to available AI groups defined by the set-up zone designed into the QB map. I find that if you hand pick your OOB and place the forces under separate company HQs the system will tend to spread the companies into separate AI groups i.e. 3 companies into 3 AI groups. They do not have to be full companies as long as the HQ is available. I do not know if this happens each and every time. But it does a lot. Not 100% realistic having more company HQs than normal. But this will tend to utilize the available AI groups as defined by the set-up zones on the map. 

Kevin