john connor
Member
ME TOO!!!!
LnLP Forums and Resource Area Closing At The End of the Year
After careful consideration, we have made the decision to close the LnLP forums due to decreased activity, as the community has largely transitioned to other social media platforms. Going forward, all community engagement will be centralized on our Discord server. Game manuals will continue to be available in our Online Library, which now also includes our new AI Assistant. The AI Assistant is designed to answer a wide range of game rule questions and can guide you to the relevant sections of the manual. For the best experience, we encourage you to interact with the LnLP community via our Discord and Facebook pages. All support-related inquiries will be handled through our dedicated support site. LnLP Discord Server: https://discord.gg/FCj7EuqMxBSITREP
- Thread starter Dave 'Arjuna' O'Connor
- Start date
- Copy link Copy link
Grognerd
Member
I'll second that reply.
I sure have come to realize how complicated this program is by virtue of using the editors!
Hope you got it nailed down, will be some fun to play it and see the behavior differences in the AI.
I sure have come to realize how complicated this program is by virtue of using the editors!
Hope you got it nailed down, will be some fun to play it and see the behavior differences in the AI.
Dave 'Arjuna' O'Connor
Panther Games Designer
- Joined
- Jul 31, 2014
- Messages
- 3,416
- Points
- 113
- Location
- Canberra, Australia
- Website
- www.panthergames.com
Not yet, sorry. I'm wrestling with a scheduling problem at the moment. I thought I had dealt with this issue but obviously not.
Grognerd
Member
Keep going Dave, don't give up, your in the last 50 meters of the marathon!
Dave 'Arjuna' O'Connor
Panther Games Designer
- Joined
- Jul 31, 2014
- Messages
- 3,416
- Points
- 113
- Location
- Canberra, Australia
- Website
- www.panthergames.com
I thought I was in a marathon last year. No one told me it was going to turn into an epic. 
jimcarravallah
Member
The code test fix test fix test fix cycle is the bane of all software development, but I'm not sure any other method works.I thought I was in a marathon last year. No one told me it was going to turn into an epic.
Thanks for sticking with the effort!
ultradave
Member
I've run many. And my wife is an ultramarathon runner. Her advice to me when I tried one the first time (50 mile trail race in VT) was - Just keep moving forward and realize you have nothing else on your calendar today except "run"I thought I was in a marathon last year. No one told me it was going to turn into an epic.
The just keep moving forward part, at least, is appropriate.
Dave
Dave 'Arjuna' O'Connor
Panther Games Designer
- Joined
- Jul 31, 2014
- Messages
- 3,416
- Points
- 113
- Location
- Canberra, Australia
- Website
- www.panthergames.com
SITREP Friday 21 Jan 21.
Hi all,
I've made good progress this week, finally knocking on the head some of these scheduling issues bedevilling the attack. Interspersed with this bug stomping I've made mods to prevent the AI from choosing what I would call bad approach routes for the attack. I had one example where the approach route went to the right of the objective, punched through the main enemy defence line, wheeled left past the objective before wheeling left again before punching back through the enemy into friendly territory, and choosing its FUP and reserve loc. It did this because there was a river blocking a direct path from the main friendly force to this FUP.
To address this I have tweaked the modifiers for enemy controlled territory and added a cap to the total move cost of the route. Trouble is right now I determine the initial yardstick based on the frontal approach route and then modifiy its route's total tick count by 3 to produce the maxModifiedTicks as my cap. If the path finding code determines that a location's travel cost exceeds the maxModifiedTicks then that loc will be ignored and the pathfinding will have to choose a different path. But I just realised from my last run through that the frontal approach is noit factoring in map control for its shortest or direct path. So the tick count is quite low and soon exceeded for the other non-frontal approaches.
I've run out of puff now for this week. I'll get back to it on Monday.
Hi all,
I've made good progress this week, finally knocking on the head some of these scheduling issues bedevilling the attack. Interspersed with this bug stomping I've made mods to prevent the AI from choosing what I would call bad approach routes for the attack. I had one example where the approach route went to the right of the objective, punched through the main enemy defence line, wheeled left past the objective before wheeling left again before punching back through the enemy into friendly territory, and choosing its FUP and reserve loc. It did this because there was a river blocking a direct path from the main friendly force to this FUP.
To address this I have tweaked the modifiers for enemy controlled territory and added a cap to the total move cost of the route. Trouble is right now I determine the initial yardstick based on the frontal approach route and then modifiy its route's total tick count by 3 to produce the maxModifiedTicks as my cap. If the path finding code determines that a location's travel cost exceeds the maxModifiedTicks then that loc will be ignored and the pathfinding will have to choose a different path. But I just realised from my last run through that the frontal approach is noit factoring in map control for its shortest or direct path. So the tick count is quite low and soon exceeded for the other non-frontal approaches.
I've run out of puff now for this week. I'll get back to it on Monday.
Art Hall
Member
thanks for all your hard work Dave!
song
Member
Thank you Dave for your hard works, keep it up!
Dave 'Arjuna' O'Connor
Panther Games Designer
- Joined
- Jul 31, 2014
- Messages
- 3,416
- Points
- 113
- Location
- Canberra, Australia
- Website
- www.panthergames.com
SITREP 16:30 Friday Feb 22.
Hi all,
It's been an eventful week here. Several cuts of the code now see the cap on attack pathing working. Today I successfully tested the latest version of this code and saw the initial attack by CCA in the Tutorial scenario kill off the flank approaches because of the rivers and enemy controlled areas to the flanks. It put in a frontal attack and then moved across the river at Steinbruck. It subsequently put in another attack on the north side of the crossing towards Lommersweiller. This time the flank approaches proved valid but were constrained as I hoped. It chose on of the flank attacks and I was mighty pleased with the result.
So good news on that front. I really need to test this a bit more. I will do so next week.
Hi all,
It's been an eventful week here. Several cuts of the code now see the cap on attack pathing working. Today I successfully tested the latest version of this code and saw the initial attack by CCA in the Tutorial scenario kill off the flank approaches because of the rivers and enemy controlled areas to the flanks. It put in a frontal attack and then moved across the river at Steinbruck. It subsequently put in another attack on the north side of the crossing towards Lommersweiller. This time the flank approaches proved valid but were constrained as I hoped. It chose on of the flank attacks and I was mighty pleased with the result.
So good news on that front. I really need to test this a bit more. I will do so next week.
Art Hall
Member
sounds great Dave!
nelmsm
Member
Well done Sir! So appreciative of the hard work you continue to put in on this great game
Dave 'Arjuna' O'Connor
Panther Games Designer
- Joined
- Jul 31, 2014
- Messages
- 3,416
- Points
- 113
- Location
- Canberra, Australia
- Website
- www.panthergames.com
SITREP Fri 18 Feb 22
Hi all,
With the attack code actually working, I tested the game and found that the new seniorHQ basing code has some unintended consequences. When we base the senior HQ (say a Bde HQ) it gets siphoned off from the main force group (mainFG) and assigned to its own HQ Basing task. The mainFG ends up being commanded by one of the brigade's BnHQ. This senior subordinate (seniorSub) ends up as the hub of any formation for the mainFG. It also ends up conducting the reassessments for the mainFG. This is not good. Not only is the seniorSub overloaded for command purposes, but it doesn't include any based arty when it comes to doing an attack from a Move mission. Also when launching an attack it determines that it should go to the reserveLoc and so only has two sub-attack, rather than the three Bn level sub attacks you would expect from a brigade attack.
To fix this we need to do two things. First we need to ensure that the based HQ does the reassessments and not the seniorSub. Second we need to find a way to assign the seniorSub to command its battalion and not be the hub of the formation.
I have explored ways for the latter and each is less than satisfactory or requires a lot of reworking. So I've opted to shelve this aspect for now and concentrate on having the basedHQ do the reassessments. I've started with the reassess for a Move. I've cloned the normal TaskMove::ReassessTaskSelf(), renamed it as ReassessMove() and call this for basedHQs in addition to their normal TaskDefend::ReassessTaskSelf(). I have been working my way through the code and modifying it so it uses the existing mainFG subject's location instead of the based HQ location to determine enemy threats, routes and approaches etc. I'm part way there now. But it's going to take a bit longer to finish and more to test. What I want to see is that the brigade launches an attack with three sub-attacks and with the full support of its based arty. If I can get that done I will release a new beta, knowing that there is still more to do. But at least this will get a working version of the game out to everyone.
BTW tomorrow is my birthday and I'm under starters orders not to work this weekend.
Hi all,
With the attack code actually working, I tested the game and found that the new seniorHQ basing code has some unintended consequences. When we base the senior HQ (say a Bde HQ) it gets siphoned off from the main force group (mainFG) and assigned to its own HQ Basing task. The mainFG ends up being commanded by one of the brigade's BnHQ. This senior subordinate (seniorSub) ends up as the hub of any formation for the mainFG. It also ends up conducting the reassessments for the mainFG. This is not good. Not only is the seniorSub overloaded for command purposes, but it doesn't include any based arty when it comes to doing an attack from a Move mission. Also when launching an attack it determines that it should go to the reserveLoc and so only has two sub-attack, rather than the three Bn level sub attacks you would expect from a brigade attack.
To fix this we need to do two things. First we need to ensure that the based HQ does the reassessments and not the seniorSub. Second we need to find a way to assign the seniorSub to command its battalion and not be the hub of the formation.
I have explored ways for the latter and each is less than satisfactory or requires a lot of reworking. So I've opted to shelve this aspect for now and concentrate on having the basedHQ do the reassessments. I've started with the reassess for a Move. I've cloned the normal TaskMove::ReassessTaskSelf(), renamed it as ReassessMove() and call this for basedHQs in addition to their normal TaskDefend::ReassessTaskSelf(). I have been working my way through the code and modifying it so it uses the existing mainFG subject's location instead of the based HQ location to determine enemy threats, routes and approaches etc. I'm part way there now. But it's going to take a bit longer to finish and more to test. What I want to see is that the brigade launches an attack with three sub-attacks and with the full support of its based arty. If I can get that done I will release a new beta, knowing that there is still more to do. But at least this will get a working version of the game out to everyone.
BTW tomorrow is my birthday and I'm under starters orders not to work this weekend.
song
Member
Thank you Dave for your excellent and rigorous works. And I wish you will be happy on your coming birthday.
Grognerd
Member
Happy Birthday to you Happy Birthday to you (you know the rest!) HA!
Dave 'Arjuna' O'Connor
Panther Games Designer
- Joined
- Jul 31, 2014
- Messages
- 3,416
- Points
- 113
- Location
- Canberra, Australia
- Website
- www.panthergames.com
SITREP Sat 26 Feb 2022
Hi all,
Haven't done much programming this week. We have been preoccupied with the Ukraine. My colleague and friend, Pavlo, who goes by the nic of ioncore on this forum is Ukrainian. He lives and works in Germany. He was originally from the Donbas region. He lost a brother a few years back when they managed to move their mother to Kiev. Now she is trapped there in Kiev, with fighting at its doorstep. It's a very concerning time. My thoughts and prayers are with Pavlo and his family.
Hi all,
Haven't done much programming this week. We have been preoccupied with the Ukraine. My colleague and friend, Pavlo, who goes by the nic of ioncore on this forum is Ukrainian. He lives and works in Germany. He was originally from the Donbas region. He lost a brother a few years back when they managed to move their mother to Kiev. Now she is trapped there in Kiev, with fighting at its doorstep. It's a very concerning time. My thoughts and prayers are with Pavlo and his family.
