Recommend
1 
 Thumb up
 Hide
10 Posts

Fire in the Lake» Forums » General

Subject: Bots and the Pace of the Game rss

Your Tags: Add tags
Popular Tags: [View All]
Holman
United States
Philadelphia
Pennsylvania
flag msg tools
badge
Avatar
mbmbmbmbmb
I've been thinking about the difference between the bot flowcharts and more conventional "take every side" solitaire play. One thing that occurs to me is that the bots' ability to always use a full command and special activity (as opposed to a limited command) greatly speeds up the pace of the war compared to history.

For example, in the long scenario, the NVA bot spends the whole first campaign Rallying and Infiltrating in Laos, Cambodia, and the North. This means that it has usually managed to get its entire army of cubes onto the map around the time of the first Monsoon, after which it will begin significant invasions of the South. A human player wouldn't be able to amass such a force.

I know there's a good deal of abstraction in the COIN series, but (especially in the case of FITL) this rapid buildup to invasion seems to throw off the historical narrative. In the long game, it guarantees that you'll always see full-scale mechanized invasions in 1965 or 1966.

This isn't a major complaint, but it has sent me back to the "take every side" style when I play solo. Is there a way to adjust the bots for more of a sense of realistic historical progression? Is the long scenario not intended for play by the bots?

Presumably this issue could have implications for other COIN titles too. What do people think?
4 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Jesse Edelstein
United States
Portland
Oregon
flag msg tools
mbmbmbmbmb
It's a good point that bot NVA can build up cubes faster than historically happened. However this might be balanced by the bot's relative timidity in actually invading South Vietnam. I have found that NVA bot is not particularly effective at actually getting to its victory condition, because it just can't invade and hold that much population compared to a more wily human player. Because the bot isn't as good as a human at noticing weak points and opportunities, I don't think it really needs a handicap by changing its abilities but of course YMMV.

Another consideration is that one would expect the long scenario to be the most divergent from history. If the US is a bot or a human worried about the NVA, you can expect a lot of Air Strikes on the Trail as well, which will help to slow down the infiltration into the South as well. In my experience the SA-2s capability, which protects the Trail against being degraded below 2, is a lifesaver for the NVA, and it doesn't start with that capability in the long scenario.
3 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
chuck reaume
United States
Michigan
flag msg tools
designer
badge
Avatar
mbmbmbmbmb
I actually play FitL solo with a hybrid "bot-play all factions" methodology. Using the bots as more of a guide for what that faction could do, I will often supplement or modify the actual action/operation based on the current state of the game and what I, as a player, would do in that particular circumstance. It's helped me better understand the motivations and strategies of each faction as well as sped up the play of the game. I'll do this for other COIN titles as well but FitL is the one I always use this approach with.
4 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Philip Jelley
United Kingdom
Hungerford
Berkshire
flag msg tools
designer
badge
Avatar
mbmbmbmbmb
If you are playing with three bots, then you will be enacting their strategies for most of the game, rather than playing the game yourself.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Rex Stites
United States
Lawrence
Kansas
flag msg tools
Avatar
mbmbmbmbmb
Is there any data to back up the fact that the Bots actually build up larger forces quicker than a human employing the same type of strategy could?

Without going through all of the specific FitL bots, I know it's typical to limit a Bot's ability to Rally to ~3 or so spaces. In contrast, a human player doesn't have such limitations when choosing where to conduct an op, with the only limit being available resources.

I think the question is whether what you perceive as an increased acceleration of the war on the part of the bots is caused by a difference in the rules between bots and humans or based on a difference in play styles.
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Oerjan Ariander
Sweden
HUDDINGE
flag msg tools
Avatar
mbmbmbmb
In NVAbot's case, limiting the Rally to 3 spaces increases the speed of its Troop build-up, rather than slow it down. In order to Infiltrate as many Troops as possible, NVAbot wants a big stack of Guerrillas in a single space and it wants the Trail value to be high; but Rallying in many spaces at once tends to spread the Guerrillas out a few per space, and spends a lot of Resources on placing Guerrillas instead of on increasing the Trail...

There are several other factors at play here too, though. A very major one is that the FitL bots typically play a LOT more Events than human players would. The flip side of playing so many Events is that the FitL bots tend to execute fewer Operations than humans do - which is why NVAbot is so timid and so poor at exploiting COIN weaknesses: more often than not it gets distracted by some Event instead of Marching or Attacking, and by the time it finally remembers to do an Operation the US has usually destroyed most of the NVA Troops already... or played one of the many Events that remove NVA Resources, forcing poor NVAbot to Pass instead soblue

Another effect of playing fewer Operations than humans is that even though the bots get to execute more Special Activities per Operation than humans do, the actual number of Special Activities the bots execute isn't that much higher than what humans get. It is a bit higher, certainly, but not nearly as much as one might think just from looking at the bot rules.

(As a side note, the bots in the later COIN games - and also several of the updated C3i bots for the earlier games - have more elaborate instructions for how to choose between Events and Ops/Commands, specifically to allow them to execute more Commands... preferrably without needing quite as many Event Instructions as FitL has )

Regards,
Oerjan
6 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Den Ell

Nebraska
msg tools
mb
PaulWRoberts wrote:
I've been thinking about the difference between the bot flowcharts and more conventional "take every side" solitaire play. One thing that occurs to me is that the bots' ability to always use a full command and special activity (as opposed to a limited command) greatly speeds up the pace of the war compared to history.

For example, in the long scenario, the NVA bot spends the whole first campaign Rallying and Infiltrating in Laos, Cambodia, and the North. This means that it has usually managed to get its entire army of cubes onto the map around the time of the first Monsoon, after which it will begin significant invasions of the South. A human player wouldn't be able to amass such a force.

I know there's a good deal of abstraction in the COIN series, but (especially in the case of FITL) this rapid buildup to invasion seems to throw off the historical narrative. In the long game, it guarantees that you'll always see full-scale mechanized invasions in 1965 or 1966.

This isn't a major complaint, but it has sent me back to the "take every side" style when I play solo. Is there a way to adjust the bots for more of a sense of realistic historical progression? Is the long scenario not intended for play by the bots?

Presumably this issue could have implications for other COIN titles too. What do people think?


Evidently it must be a sufficient Major problem as you bring up the last issue: "Presumably this issue could have implications for other COIN titles too. What do people think?[/

First off ,you've got something wrong in your assumptions:

Quote:
For example, in the long scenario, the NVA bot spends the whole first campaign Rallying and Infiltrating in Laos, Cambodia, and the North. This means that it has usually managed to get its entire army of cubes onto the map around the time of the first Monsoon, after which it will begin significant invasions of the South. A human player wouldn't be able to amass such a force.

I know there's a good deal of abstraction in the COIN series, but (especially in the case of FITL) this rapid buildup to invasion seems to throw off the historical narrative. In the long game, it guarantees that you'll always see full-scale mechanized invasions in 1965 or 1966.


From Wikipedia:
"The Battle of Ia Drang comprised two main engagements conducted by the 1st Battalion, 7th Cavalry Regiment that took place on November 14–15, 1965 at LZ X-Ray ("eastern foot of the Chu Pong massif"[20]) and by the 2nd Battalion, 7th Cavalry Regiment on November 17 at LZ Albany farther north in the Ia Drang Valley. It was the first major battle[8] between the United States Army and the North Vietnamese Army-NVA (People's Army of Vietnam-PAVN) during the Vietnam War as part of the U.S. airmobile offensive code-named Operation Silver Bayonet I (October 23 – November 18, 1965). The battle was part of the second phase of a search-and-destroy operation code-named "Operation Long Reach"[21] that took place from October 23 to November 26 during the Pleiku Campaign.[8] The battle derives its name from the Drang River which runs through the valley west of Plei Me, where the engagement took place (Ia means "river" in the local Montagnard language)."

There's nothing wrong with the game or its design. I'm quite surprised that you didn't think of the one Major battle that did occur during the Time you questioned.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Holman
United States
Philadelphia
Pennsylvania
flag msg tools
badge
Avatar
mbmbmbmbmb
Denell wrote:

From Wikipedia:
"The Battle of Ia Drang comprised two main engagements conducted by the 1st Battalion, 7th Cavalry Regiment that took place on November 14–15, 1965 at LZ X-Ray ("eastern foot of the Chu Pong massif"[20]) and by the 2nd Battalion, 7th Cavalry Regiment on November 17 at LZ Albany farther north in the Ia Drang Valley. It was the first major battle[8] between the United States Army and the North Vietnamese Army-NVA (People's Army of Vietnam-PAVN) during the Vietnam War as part of the U.S. airmobile offensive code-named Operation Silver Bayonet I (October 23 – November 18, 1965). The battle was part of the second phase of a search-and-destroy operation code-named "Operation Long Reach"[21] that took place from October 23 to November 26 during the Pleiku Campaign.[8] The battle derives its name from the Drang River which runs through the valley west of Plei Me, where the engagement took place (Ia means "river" in the local Montagnard language)."

There's nothing wrong with the game or its design. I'm quite surprised that you didn't think of the one Major battle that did occur during the Time you questioned.


I did recall Ia Drang. I guess I just don't picture it (in FITL terms) as 10 or so cubes coming across the border together.

Don't worry that I think the game is broken. My comment was simply that the bots might be acting on a faster timeline due to their full rather than limited commands.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Den Ell

Nebraska
msg tools
mb
NORTH VIETNAMESE DIVISIONS:

Division 325 began moving south Nov 1964 (Regiments 33, 95 and 101). It served in the Central Highlands with Regiments 95, 101 and 320 (Regiment 18 had gone to Tay Nguyen Front). The division was dissolved by late 1965, with elements going to VC Division 1 or becoming separate regiments.

Division 304 (Regiments 9, 24, 66) sent south Aug 1965; Regiment 9 initially served in Laos and Regiment 66 with Division 1 in Dec 1965. Returned north Jun 1968, but served in Laos spring 1971 and in northern South Vietnam from 1971.

Division 308 initially contributed one of three battalions, which became Regiment 320 ca. 1964. Its Regiment 88 infiltrated south in 1966 and the rest of the division Sep 1967 (Regiments 36, 88, 102).

Division 312 sent a battalion south in spring 1963 and another in 1964. Its Regiments 141 and 165 went to the B2 front in 1966, to serve as the nucleus of VC Division 7. The division went south in Sep 1967 and was regrouped after Tet back in North Vietnam, with Regiments 141, 165, and 209. It served in Laos 1969 – 71. Its regiments served with other divisions 1972, then were withdrawn back north. In spring 1975 the division was sent south again.

Left one Division off as it came into existence much later....

====
Wikipedia:

On November 11 (3 days prior to Hal Moore and his men being inserted next to the Chu Pong Massif), intelligence source revealed the disposition of the three NVA regiments: the 66th at vicinity YA9104, the 33rd at YA 940010 and the 32nd at YA 820070.[43][42] These coordinates in military terminology indicates the three regiments were located in their respective assembly areas that were each in less than one kilometer-square,[44] which constituted suitable targets for B-52 airstrike.[45] On November 12, the 3rd Brigade was given orders by General Larsen, IFFV Commander and General Knowles, 1st Air Cavalry Division Forward Headquarters Commander to prepare for "an air assault near the foot of the Chu Pongs",[46] at 13°34′11″N 107°40′54″E, 14 miles (22 km) west of Plei Me.

Moore's men weren't sent in there blind. They knew what they were up against before they sent them there, and I hardly call 3 Divisions worth of NVA in that vicinity worth less than 10 cubes on your board, would you? It was a major NVA incursion. In all total the NVA sent at one time or another 7 Divisions south. Nearly half of all those sent during the entire conflict were present in or around the Chu Pong Massif when Hal Moore was given his orders. No not broken and certainly nothing skewed, but was on a time-line that supported the need for the NVA to be where they were supposed to be, when they were supposed to have been there.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Juan Valdez
msg tools
reemer wrote:
I actually play FitL solo with a hybrid "bot-play all factions" methodology. Using the bots as more of a guide for what that faction could do, I will often supplement or modify the actual action/operation based on the current state of the game and what I, as a player, would do in that particular circumstance. It's helped me better understand the motivations and strategies of each faction as well as sped up the play of the game. I'll do this for other COIN titles as well but FitL is the one I always use this approach with.


This is what I do as well, use the bot charts as guidance.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Front Page | Welcome | Contact | Privacy Policy | Terms of Service | Advertise | Support BGG | Feeds RSS
Geekdo, BoardGameGeek, the Geekdo logo, and the BoardGameGeek logo are trademarks of BoardGameGeek, LLC.