pzgndr

My wishlist for the AI would be for it to not make reckless attacks, specifically lunging through enemy lines and placing its units out of supply. In a few cases this is OK but generally it is suicidal and sacrificial, especially when high value tanks are used. It causes the AI to waste units unnecessarily, which makes it more of a challenge to handicap the AI with extra MPPs or extra units to compensate. IMHO the AI should more carefully consider supply. Don't push beyond current supply range! I wouldn't mind seeing the AI behave a bit more conservatively, just don't neuter it so it becomes lame. That's the risk, and Hubert continues to look for the right balance. Nothing's perfect.

Hey all, I've been updating A3R over the past several weeks. Blashy offered some good insights and I finally understand what he's been doing and what the game weaknesses are. I'm guilty of scripting the AI to counter my playing style; this doesn't work quite as well for different playing styles. Without going into too much detail, Blashy as Axis would focus on building units during 1939-41 rather than include some expensive research. For me, Italy should go for the infantry upgrades, Germany for artillery and industrial upgrades. But if you don't do that, and combined with the plunder gains, it's not too difficult for Axis to overrun Spain, Turkey and Egypt and then the rest is downhill. There are a couple of limitations which prevent better simulation of the original boardgame. For one, the force pools are fixed, which means players can build to the max without having to wait for force pool expansions in 1942 and 1944. Second, the plunder rules reward aggressive Axis behavior and provide more MPPs during the early years. Obviously, these two issues feed each other and the result is a skewed game. I have reconsidered this and rather than mess with the core economics, I've added a resource event to activate German industry at the start. By starting Bonn and Essen at zero and allowing them to build up, it reduces Germany's production by about 330 MPPs over the first 10 turns, comparable to the plunder from conquering France. I've also added some increased USA/USSR reactions to Axis DOWs on Spain and Turkey, plus added some more Axis and Allied AI unit events to toughen up the computer opponent. This may or may not be enough to resolve play balance, but should be better. I am continuing to playtest and should have an update available for cmmods.com soon after PDE v1.02 is released. If anyone would like a copy of my current beta to play, just shoot me an email. Latest changes are as follows: - fixed strength and supply scripts for possible Russian Winter Strikes Again events to occur o/a Dec 1942 - made minor correction to amphibious scripts for German 1940 Sealion invasion - made several minor adjustments to diplomacy scripts - adjusted German offensive scripts to increase Barbarossa buildup another turn - added Italian offensive script for possible invasion of Tunisia if Algeria become Free French - adjusted war_entry scripts for Finland activation, for variable entry after May 25, 1941 AND USSR is at war - added decision event #7 for Lend Lease/Murmansk Convoys and revised convoy scripts; intent now is for more Allied MPPs to be sent to USSR during mid-game and then for less MPPs to be sent after Smolensk/Kiev are reoccupied - adjusted amphibious_minor scripts to reduce chances of possible USA/UK raids into France from 10% to 5% - added transport_minor and unit events for possibily redeploying UK corps from Gibraltar and replacing it with infantry if Axis units are in Spain - fixed transport_minor scripts for UK belligerent intervention to Norway, revised amphibious_minor scripts for possible UK invasion of Norway, and added offensive scripts for possible UK attack from Bergen to Oslo - adjusted transport_minors scripts to slightly increase USA/UK reinforcements to Algeria, particularly if Axis controls Cairo - adjusted fleet scripts to improve UK sending fleets to Egypt and evacuating fleets from Red Sea - adjusted research scripts to slightly reduce some pre-war and early years research to allow more AI unit builds - added activation_1 scripts for increased USA/USSR reactions to Axis DOWs on Spain and Turkey, to help offset Axis plunder gains - added 1939 resource event script for German industry activation (Bonn and Essen) at start of WWII, to help offset German plunder gains - revised activation_2 events for German Eastern Front Garrison to allow any Axis units to satisfy requirements (Italian AI units were triggering Soviet reaction) - adjusted UK unit events for earlier arrival of Australian, Indian and South African reinforcements if Egypt is invaded - added a few more AI unit events for tougher computer opponent - upgraded AI unit events to Experience=1 for tougher computer opponent - fixed AI unit events for German DAK arriving too early

From the Wargamer interview: Could we get some more information about this rules-editor feature and other editor capabilities? Computer games generally have rules hard-coded but perhaps selectable, such as Space Empires where players can toggle a variety of rules on or off. And most games have selectable options for different gameplay. Will EOS actually allow players to create their own unique rules-sets?? THAT would be cool. :cool:

I think there is also some smoothing of the graphics at the interfaces.

Hey guys. I don't normally comment in the SC forums these days, but I want to share a few thoughts. I loved SC back in the day - fast, fun and fairly historical except for North Africa and a few other things. But so many things were lacking to provide a more complete WWII gaming experience. SC2 and its WAW and PDE expansions close the gap. Perhaps too much so, but I just don't know how a game can approach a reasonable level of realism and historical accuracy without offering more unit types, more features, and more AI/event scripting capabilities? There's something to be said for the fast & fun variety, like SC and CEAW. However, when I try to consider what features of SC2-PDE I would want to give up to return to the good 'ole SC days, I'm at a loss. I played an interesting and challenging game of my Advanced Third Reich mod this past weekend to playtest yet a few more changes I've been messing with recently. I miss the old board game experience! I guess the thought to leave you all with is that with some mods to limit which new features are actually used (ie, all new unit types and roads and railroads and such do not all need to be used) and some graphics mods to simplify the display (ie 2D icons) then the game isn't so terribly complicated. Anyways, here's a toast to fast & fun. The challenge for the future of the SC series is how to take all the new innovations and then try to simplify things to get back to what's essential for great gaming.

I want to follow up on this because I have made some changes in A3R which seem to work OK. First I added a decision event for the Allied AI to check if conditions were right for reducing Lend Lease and Murmansk Convoys. Basically I'm checking if Italy is surrendered and USSR has retaken either Smolensk or Kiev. ; Status of USSR starting in January 1943 ; Used for reducing USA/UK Lend Lease/Murmansk Convoy scripts { #NAME= Status of USSR - Convoy Reduction #POPUP= #IMAGE= #SOUND= #FLAG= 1 #TYPE= 1 #AI= 2 **AI only since humans can change convoys anyway** #LEVEL= 0 #GV= 1[1,100] #LINK= 0[0] #COUNTRY_ID= 3 #TRIGGER= 100 #DISPLAY_ORDER= 0 #DECISION= 7 ; Set AI acceptance % (Allied AI will accept 100% of the time) #AI_RESPONSE= 100 #AI_RESPONSE_POPUP= #DATE= 1943/01/01 ; Set variable conditions: ; 1st Line - USA politically aligned with Allies and not surrendered AND ; 2nd Line - USSR politically aligned with Allies and not surrendered AND ; 3rd Line - Italy politically aligned with Axis and surrendered #VARIABLE_CONDITION= 3 [2] [100] [0] #VARIABLE_CONDITION= 4 [2] [100] [0] #VARIABLE_CONDITION= 6 [1] [100] [1] ; Condition position ; 1st Line - Allied unit in Smolensk ; 2nd Line - Allied unit in Kiev #CONDITION_POSITION= 120,22 [0,0] [1,1] [2] [0] #CONDITION_POSITION= 118,29 [0,0] [1,1] [2] [0] } Next I added new convoy scripts to reduce convoy #PERCENTAGE values. However I discovered I needed to do more. Unless specifically scripted, the AI will attempt to max out its convoys unless raided, and then it will temporarily reduce convoys and later max them out again. I want to allow human players to increase convoys to a max of 50% if desired but not the AI as the standard amount. So, I added new AI scripts for normal convoys with reduced#PERCENTAGE and #MAX_PERCENTAGE values. And script order is also important, with later scripts listed first etc. So for example, here are the UK -> USSR scripts (USA -> UK are similar): ; UK supply convoy to the USSR: { #NAME= UK Reduces Arctic Convoys To The USSR (Allied AI) #POPUP= #IMAGE= #SOUND= #FLAG= 1 #TYPE= 2 #AI= 2 #LEVEL= 0 #GV= 1[1,100] #LINK= 7[1] #COUNTRY_ID= 1 #TRANSFER_ID= 4 #COUNTRY_ID_TRIGGER= 100 #TRANSFER_ID_TRIGGER= 100 #PERCENTAGE= 25 #MAX_PERCENTAGE= 25 #SPRING_REDUCTION= 90 #SUMMER_REDUCTION= 100 #FALL_REDUCTION= 90 #WINTER_REDUCTION= 75 #SOURCE_PORT= 79,18 #DESTINATION_PORT= 123,6 #WAYPOINT= 79,10 #WAYPOINT= 95,0 #WAYPOINT= 123,5 } ; UK supply convoy to the USSR: ; Historical date September 1941 { #NAME= UK Arctic Convoys To The USSR (Allied AI) #POPUP= UK Starts Arctic Supply Convoys To The USSR #IMAGE= #SOUND= #FLAG= 1 #TYPE= 2 #AI= 2 #LEVEL= 0 #GV= 1[1,100] #LINK= 0[0] **Note AI only but NOT linked to decision event** #COUNTRY_ID= 1 #TRANSFER_ID= 4 #COUNTRY_ID_TRIGGER= 100 #TRANSFER_ID_TRIGGER= 100 #PERCENTAGE= 30 #MAX_PERCENTAGE= 30 **max percentage for AI set here** #SPRING_REDUCTION= 90 #SUMMER_REDUCTION= 100 #FALL_REDUCTION= 90 #WINTER_REDUCTION= 75 #SOURCE_PORT= 79,18 #DESTINATION_PORT= 123,6 #WAYPOINT= 79,10 #WAYPOINT= 95,0 #WAYPOINT= 123,5 } ; UK supply convoy to the USSR: ; Historical date September 1941 { #NAME= UK Arctic Convoys To The USSR #POPUP= UK Starts Arctic Supply Convoys To The USSR #IMAGE= #SOUND= #FLAG= 1 #TYPE= 2 #AI= 0 #LEVEL= 0 #GV= 1[1,100] #LINK= 0[0] #COUNTRY_ID= 1 #TRANSFER_ID= 4 #COUNTRY_ID_TRIGGER= 100 #TRANSFER_ID_TRIGGER= 100 #PERCENTAGE= 30 #MAX_PERCENTAGE= 50 #SPRING_REDUCTION= 90 #SUMMER_REDUCTION= 100 #FALL_REDUCTION= 90 #WINTER_REDUCTION= 75 #SOURCE_PORT= 79,18 #DESTINATION_PORT= 123,6 #WAYPOINT= 79,10 #WAYPOINT= 95,0 #WAYPOINT= 123,5 } As seen above, the normal convoys start at 30% and a human player could increase this to 50%. But the Allied AI convoys will remain at 30% (unless raided) and then reduce to 25% after decision event #7 triggers. I also discovered with Hubert's help that if I tried to set a default decision event (ie DE 07=0) and use that for my early AI scripts, then the later scripts with DE 07=1 would not work since the game eliminates all those failed decision scripts. He may be relooking this but for now that's how it works. I didn't realize that before, but I hadn't tried something like this before. Anyway, as a practical playtest example, I ran an AI-AI game with the above scripts. Italy didn't surrender until Oct 44 and then Smolensk was retaken in Nov 44, but I got my DE 07=1 event trigger. Before the change, USA was sending 115 MPP to UK and UK sending 92 MPP to USSR. After the change, the amounts dropped to 104 and 75 respectively. Maybe not significant enough, but it works and I can now make some further adjustments. Bottom line is that this scheme works and other developers can consider their own applications.

This is something I did for USA and USSR production increases in my Advanced Third Reich mod. The alternative was to use large IT tech increases which could randomly occur too fast or too slow, either way ruining the historical economic expansions and play balance. There needs to be some variability, but not huge swings either way. I know my mod isn't for everyone, but I took a few pains to thoroughly document what I did and why. A pdf guide is available with the freebie download at cmmods.com. I'm seeing several new modders raising issues that I've wrestled with for years now. Please feel free to build upon what I've done.

I've commented on this a few times. There is indeed a skewing towards higher casualties with higher tech, and higher experience for that matter. One would think similar battles between say two L0 tanks and two L5 tanks would be comparably matched and losses would likewise be comparable. But the combat formulas do not take into account relative differences between attacker and defender tech levels. The bad news is that losses are unnaturally higher; the good news is that this skewing is applied to both sides. A couple years ago I wanted to propose some changes so I made myself a spreadsheet with the current formulas and some examples, and then tried making adjustments for "better" results. I found this to be easier said than done. The problem with relative tech values is that differences can be negative and can cause errors. I messed with it for quite a while and just could not arrive at a solution I felt comfortable recommending. I believe this is an inherent problem with attacker/defender formulas like this, rather than more simplistic odds-based combat results tables (CRTs) with modifiers. So, with the current system players get a fun game with lots of interesting tech advances but eventually at the higher levels combat starts to become an abnormally bloody affair. A work-around that I and others have used in mods is to limit tech and experience levels to about 2 or 3. This seems to keep combat results more reasonable. A similar work-around to slow down rapid tech advances is also limit how many chits can be invested, again to about 2 or 3.

The plans are checked one at a time, but multiple plans per country can all be in progress if there are sufficient units. The one criteria for same goal position allows multiple plan options but only one plan would be executed. As plans are activated, some with immediate execution and others with buildup delays, they remain in effect until the goal is achieved or the cancellation criteria is meet - or the cancel_ratio is meet, which can be difficult to determine. Good news is that there is a sense of logic behind all the madness and if you can mentally orchestrate what you want the AI to do then it's possible to get pretty close with scripting in most cases. The difference for amphib_minor and transport_minor is that these plans do not include a HQ. D-Day would be an amphib plan because you want a HQ involved. A German landing in the Baltics during Barbarossa would just be an amphib_minor plan without HQ. Etc.

Hubert can confirm the sequence, which I think goes in the order listed in the purchase script, but there are several other factors to consider. First there is generic logic for the AI to reinforce units, upgrade units and rebuild asterisked losses. For actual new purchases, the AI goes through the script and performs simple checks: 10% corps [2], if yes then buy two corps, or one, or none depending on funds; then move to next check, etc. There was some additional logic added a while ago where if the country has more than 4x its production in saved MPPs available, then the AI will cycle through the purchase scripts multiple times; there was a nasty MPP buildup issue before this. So. Consider all that and assume some percentage of MPPs that your AI country may have to spend and then assign purchase % accordingly. If the first listed units have a good chance, then its likely those units will be built before later listed units. If you want a decent statistical spread of builds, assign relatively low chances, remembering that over a year of turns you might see x number of new units built. And those are new units only you need to worry about, not rebuilding asterisked units that are destroyed in supply. Clear as mud?? I'd like to provide clearer guidance, but there are too many variables to juggle. I can offer a suggestion. Create a spreadsheet for all your purchase scripts so you can see the big picture. As you playtest and see too much of this and not enough of that, you can make adjustments accordingly. Kentucky windage and all that. Voodoo dolls, whatever.

Excellent idea! And good point on the order of scripts. I need to experiment with this in my A3R mod. One frustration has been how to reduce the Murmansk Convoy once USSR gets rolling and western Allies have a greater need for MPPs later in war. This should work.

FYI, I have made a few minor changes: I am still interested in some feedback before I post another update. I recently played a game as Allies on Expert +2 and I'm inclined to agree with Blashy that it is not as difficult as I thought it would be. So, I'll probably add a few more extra units for the AI in addition to what's already provided. I am open to other suggestions.

It depends. I'll speak from experience with my Advanced Third Reich mod. If you want any overseas activity involving transports and amphibious operations, you absolutely must script. If you want the AI to declare war or garrison/defend specific areas based on conditions, then you must script. Any production, research and diplomacy actions must be scripted. The generic AI will not do these things on its own. Most generic offensive and defensive operations are handled OK by the AI once forces are in the area. In some cases I scripted variable offensives, such as possible German emphasis in northern Russia towards Leningrad/Moscow in 1942 rather than south towards Stalingrad/Grozny, but I'm not convinced these work. I suspect the generic AI overrides these; it's hard to tell. The AI-vs-AI play feature is VERY helpful for troubleshooting. Run a game overnight, spot check key points in the game the next day, identify AI weaknesses, make corrections/additions to the scripts, and repeat. Eventually you'll come to understand what works and what doesn't, and why, and then the exercise should transform from being frustrating to fascinating. There is definitely a learning curve. Good luck!

It becomes reactivated as a Major. MPP collection starts again and they can build/research/etc. Related to your earlier questions about Free French units, those free units remain as UK units and do not revert back to French status. That would be ideal, someday.

It's all in the SC2 manual, which comes with the game and expansions. It doesn't go into excrutiating detail about the combats if I recall. Basically if unit type X is attacking unit type Y, the unit type X attack values are used against the unit type Y defense values from the CTVs. Pretty simple stuff, as Ottomops described. Should the manual provide more detail??

Use the export data feature to export any or all data from the other campaign, then in the new campaign import the data you want from the other campaign's data directory. There's info in the Help file about this. Once you import established data such as weather and combat target values, it's easy to edit rather than starting from scratch. Map data cannot be imported unless your new map is the same size as the other map, but you'll get a warning saying this and telling you what the map size should be so you can make adjustments.

The AI scripts do account for different difficulty levels, and there are scripts where extra units are available at Intermediate and/or Expert levels that are not available at Normal level. Also some events apply for either side (including human vs human play) while some events are only for Axis or Allied AI. It's all pretty flexible. Additionally, the difficulty levels for AI adjust some game percentages, particlarly for plunder, where AI gets more and you get less on higher difficulty. These handicaps can help balance a campaign somewhat but the generic AI still makes a lot of little mistakes, so giving it an experience bonus helps compensate. +1 or more is usually needed to give experienced players a challenge.

It would be nice to have a withdraw unit event script to actually remove a unit of specified type. It would also be nice to have a force pool event script to change build limits. A prime example of where this would be good is a Russian Front campaign where some German units are withdrawn in 1943 for Italy and again in 1944 for France. Right now we can't do this.

Both the new decision events and global variables offer many possibilities. If you read my AI articles at Wargamer.com, I mentioned how I used the decision events to help script US deployments to North Africa. The logic got very complicated between whether Morocco and/or Algeria was Allied/Free French or not, which determined whether US would send amphibious invasion force(s) or fast transports. Also, whether Gibraltar was Axis controlled or not determined if US would deploy to Algeria or pump everything through Morocco. This stuff drove me nuts with SC2-WAW, particularly once status chnaged; eg, amphibious invade Algeria but once liberated AI would then send transports and then too many units were in North Africa and not enough in England for D-Day, etc. In PDE using decision events, I was able to "lock in" courses of action for the AI and resolved the problems I was having. (I think.) For land loops, I used these to help the AI "teleport" units inadvertently stuck on islands. This is a bit easier than having contingency transport scripts. I agree that more innovation is needed. Modders should continue to experiment with new ideas and let Hubert and others know what works.

OK. Anybody else have some game feedback?? I can easily toughen up the AI unit scripts and add more AI units if necessary. I'd like to hear from several others before making any changes. Has anyone played hth? I'm also curious what the default balance is and whether some changes may be needed. Let me know please. If I get enough feedback, I'll try to get out another update as time permits.

I want to share some comments I already provided to Blashy. The generic AI is not "brilliant" and needs a handicap to offset mistakes most seasoned players would not make. IMHO, it needs the combat experience bonus moreso than the economic help. So while I recommend Intermediate +1 settings, I also find the AI needs more than +1.0 to be challenging. For me. +1.5 is pretty good. Newbies just getting used to the game/mod may be challenged on +1 or less, but experienced players should crank it up. I have not tried Expert +2 but suspect that may be practically impossible to beat. I'm interested in any game reports to prove otherwise. Those familiar with the original Third Reich board game should recall that yes, Axis should completely max out its force pools by 1942. There are no more units to build (ie, realistic manpower and other resource limitations). Axis would then hit a high water mark (ie, Stalingrad, El Alamein) where it is strained to hold what it has taken, and the time delay to move replacements to the front usually puts Axis on the strategic defensive as Allies gear up and start pushing back. If this is not happening, then Allied AI needs another boost to be a challenging computer opponent. And vice versa for Axis AI if playing as Allies. If you make the AI challenging enough, your unit losses should prevent you from accumulating too many excess MPPs and taking advantage of inherent AI limitations and weaknesses. And then AI performance in the later years should prove to be much better. I mean, if you want to take advantage of the AI and cripple it in the early war then obviously it's not going to play well for a full game. I've already provided both Axis and Allied AIs with additional free units here and there as handicaps. I suppose I could consider more to make the Intermediate +1 setting more "balanced" but I'd rather reserve the higher settings up to Expert +2 as the natural limits. If Expert +2 still isn't tough enough for some of you, then I could certainly make a few more changes. I did note one minor "bug" in my alternate Russian Winter Strikes Again strength and supply scripts. These should happen o/a Dec 1942 and not Dec 1941. If anyone spots anything else odd or not working very well, please let me know and I'll consider a v1.03a hotfix. This is about it for me for a while. I've already announced to Hubert and others that tomorrow I'm returning to the US Army staff to manage its nuclear reactor program. Between the additional work and commute, my gaming time is going to take a hit. I have enjoyed the gaming these past several years and will remain active in some limited capacity. I'll still be around.

OK, my PDE v1.01 updates are now posted at http://www.cmmods.com/ Here are the version notes: Mostly a lot of little bug fixes and tweaks not fully described in the version notes. While I did make some additions and revisions, most things were working pretty good so I'm not messing with them. I may make another update at some point, but this may be it for a while.

I've made what should be final tweaks. If my ongoing AI-AI game plays out without any surprises, I'll post my PDE v1.01 mod updates tomorrow. One thing I noticed is the AI improvements Hubert made with v1.01 have resulted in slightly faster runs AND the optimizations have also improved the AI's air target selection routines. For example, where the German AI was sluggish against France because it wasn't using air effectively, it is noticeably better now. Armuss, there are still the default 3D icons. I like the original Third Reich counters that I customized, but you don't have to use them. All this stuff can be further customized using MS Paint or similar bitmap editors. I don't plan to do much more myself, but you and others are free to use other 2D or 3D icons and terrain sprites.

I'll offer one shameless plug for SC2-PDE. If you're interested in the classic Third Reich game, my Advanced Third Reich mod is updated to PDE. IMHO it plays much better with the new decision events, global variables and AI improvements; I do not intend to go back to WAW. AFAIK, Hubert is still considering updating the classic SC2 campaigns to PDE. If and when that happens, PDE would definitely be the way to go.

I prefer the older plain interface myself. For my A3R mod, I provided some guidance in the readme file: 3a. For fancier graphics with Axis and Allied symbols, copy the entire customized \Interface directory from the \_1946 Case Neptune directory into the \_Advanced Third Reich directory. 3b. For plainer graphics without the Axis and Allied symbols, copy the entire customized \Interface directory from the \_1974 Invasion of Cyprus directory into the \_Advanced Third Reich directory. So, you could either 1) save the default PDE interface directory someplace and copy what you want into the default directory, or 2) customize the campaigns you play and copy what you want into that campaign's directory. The nice thing is there are options to choose from.