patch update

[WanderingHead]

Hi all,

I've been working on a patch revision. I don't want this to drag on like the last one, I want this one to come out soon (the Powers that Be willing). I'd like to actually skip lengthy closed testing and just give it out as a new beta.

Which is a little risky, but I'm trying to keep the changes targeted. So far, everything done is save compatible to the best of my knowledge. So it should be relatively easy to migrate and have the community start looking at it to prove there are no more bugs .

Done so far

• added a resource "development" feature, so that raw "undeveloped" resources can cost more supplies and take longer to bring on line. I envision a typical setting being that it takes 20 supplies and 2 turns (10 supply/turn) to bring raw resources online. This is for a few oil fields that would have required more extensive investment than the current game engine can represent (like Iraq).
• deemphasize factories in frozen regions for damaged unit tracing
• modify Russian militia mobilization, so that it won't occur if Russia attacks Germany before Germany attacks Russia.
• modify naval attacking player retreats, retreat far enough backwards through regions traversed to find one unoccupied by the enemy (rather than only moving back 1 region, occupied or not, which caused problems). A good enough fix?
• new hotkey "a", to return air units to base. This allows air to return to base without using movement points, as though they did combat. For example, you move your HB 3 MPs and it reveals FOW. Previously, it would be stuck there UNLESS it could initiate combat. Now it can return.
• fixed a problem with the "unused factories" warning (I think this problem was always there, but it is hit or miss whether it is observed ... it might have become more obvious to Jesse because of his modified map).
• VCR playback always use "non-player" speed.
• remove CAG double team of CVs, CAG priorities apply to two targeting rounds.
• improve air targeting of naval, to pick unique targets.
• improve FOW (less visibility) in VCR Playback. Enemy movement cannot be seen unless either the start or the end region is visible.
• change CAM for bombers: remove HB.
• remove island op-fire against units moving from their starting region, add CAP op-fire at sea from start region, allow air to CAP at sea after combat.

on the list - to do

• organize tech events by date

on the list - maybe

• more robust file access routines to avoid save-incompatible caused crashes ... ? it is worth doing? This would (ironically) break save compatibility, because I would need to add a new version field to the save files.
• fix season offset in the victory level calculation

unlikely due to difficulty or lack of unanimity

• change CAM for bombers: require more attacking TB than defending TB (making CAM easier or harder, depending on precise definition)
• change date displayed in View Mode after end of player turn, to reflect completed turn instead of next turn.
• change German snorkel tech even to 1942.
• make all tech events mod-able
• stacking limits for islands

Anything else important (and relatively easy and without save compatibility issues)?

[GKar]

Once again, thanks WH for your dedication to improving AWD. I'm sure it is widely appreciated, even though the need of time commitment to judge and test will usually delay the responses in the forum.

ORIGINAL: WanderingHead

• improve air targeting of naval, to pick unique targets.

I'm not 100% sure how air targeting of naval units works in all cases right now, but this is my view on the issue:

• All air units (especially CAG and non-CAG) should use the same mechanism for targeting.
• The mechanism used should be the same for attacks on the open sea and in harbors.
• Air units should use the weights described on p. 83 in the manual for targeting.
• Each air unit should determine its target independently of all others, but shouldn't "lose" their attack because the intended target has been destroyed already. In this case, it should switch to an undestroyed target if there's still one.

[WanderingHead]

ORIGINAL: GKar
I'm not 100% sure how air targeting of naval units works in all cases right now, but this is my view on the issue:

• All air units (especially CAG and non-CAG) should use the same mechanism for targeting.
• The mechanism used should be the same for attacks on the open sea and in harbors.
• Air units should use the weights described on p. 83 in the manual for targeting.
• Each air unit should determine its target independently of all others, but shouldn't "lose" their attack because the intended target has been destroyed already. In this case, it should switch to an undestroyed target if there's still one.

This won't all be possible. In all forms of combat, units pick their targets before firing. It is not possible to reselect targets after others have fired, as would be required for your 3rd bullet. I'm afraid there is no way this will change, it would be a huge overhaul of combat.

Targeting is the same in harbors or at sea.

CAGs very intentionally have a different targeting approach from other air units, focusing on the high value naval units. Whether you agree with this implementation or not, I don't think it will change much. It does make sense to allow CAGs better targeting, that aspect will not change. Personally, I think the CAGs enhanced targeting is a good concept but it is overdone.

My personal preference would be:

* all air units (CAG and non-CAG) pick unique targets until all naval units are targeted. This would remove the CAG doubling up on CVs feature.

* CAG units select targets first, with weightings CV 16, HF 4, LF 2, TF 1, SF 1 (reducing CV weight from 40 to 16).

* non-CAG air units select unique targets next, with equal weightings on all naval units.

Only if you have more air units than naval vessels will naval units be double targeted. If a targeted unit is already destroyed before an air unit has a chance to fire, the air unit would lose its firing opportunity (unchanged from today, and this is simply not changeable).

With CAGs firing a little less on CVs, escort fleets can perform a more useful function (cannon fodder in this case). In some sense it's also a bit nice from the attackers standpoint, not to be focusing all the CAGs on a few CVs that may be overtargeted already.

[GKar]

I more or less expected that some aspects wouldn't be debatable, and thus your list sounds fair enough to me.

I just wonder if the removal of the "CAG doubling up on CVs" feature would make the game harder for Japan because it might be harder to deal pre-emptive strikes to the WAllies. Maybe CVs should be targetable as long as they haven't been targeted twice, maybe with a reduced weight for the second attack.

[Lebatron]

Brian,

I like that your adding something to support raw resources as I suggested. Thanks. You mention a 'typical' setting being 20 supplies/2 turns. That's suggests to me it's moddable? Could it be made to take 3 turns and 30 supplies for instance? And what about factories? Can I use this to place partly constructed factories on the map and have them take more time and supply than currently? And what do you do about making these raw resouces distinct from the regular ones? If they are captured and redamaged how would the game engine know to only require it to take 10 supplies the second time around? It seems to me the best approach would be to reconfigure the game engine to recognize 4 state levels instead of 2. Have level 4 take 5 supply and 1 turn to repair, level 3 take 5 supply and 1 turn to repair, and lastly the original levels 1 and 2 take 10 supply and 1 turn to repair. For a grand total of 20 supply and 3 turns to fully fix. The pic below shows there's room for two more distinct damage colors. I could do the editing and send it to you if you could find in code were it does the location picking off of this bitmap. In any case, this new feature should support both factories and resources IMO, or its going to look like your biasing your updates to address GG issues instead of general game improvement.

In clearer terms this is what I propose.
Green dot- fully repared
yellow dot- level 1 damage/5 to repair
red dot - level 2 damage/5 to repair
purple dot- level 3 damage/5 to repair, limit 1 repair
black dot- level 4 damage/5 to repair, limit 1 repair

From black to green would take 3 turns and 20 supply. Each damage level should cost 5 supply for consistency sake.

Brian would you please show in more detail what you had in mind.

[Lebatron]

Additionally, I see some new posibilities reguarding what the game engine can do when areas are taken over. Things can stay the same and if a region has a resourse damaged to level 3/purple dot, no further damage takes place when it changes hands. Or when it does change hands it could drop from level 3 to level 4 in the way yellow goes to red now. 

[WanderingHead]

ORIGINAL: Lebatron
I like that your adding something to support raw resources as I suggested.

I think this was discussed long before you suggested it. Certainly it has been on my mind forever, as long as I have been thinking about oil in the game, because it was obvious that the simple repair mechanism was inadequate for this. I always put it off because I thought it would be difficult, but it proved to be relatively easy. Graphics and code changes took a couple of hours.

As it is now, you can define the level of "undevelopment" up to 4 levels. After you complete "development", it becomes regular double damaged. Each level of development requires 5 supplies to improve, and you can mod-ably limit how many development levels can be improved per turn. Once you get to zero development levels, you cannot do any more development that turn, it will take until the next turn cycle before you can repair them as normal double-damaged. Hence, it will always take at least two turns to go from undeveloped to fully operational.

At 4 levels of undevelopment, it would cost 30 supplies to bring on line: 20 supplies to get to double-damaged, and 10 more supplies to repair. My prefered setting, at this point, is 2 levels of undevelopment, but it supports up to 4. My preferred setting for the development limit is 2 per turn, but it could be set to 1,2,3,4.

The graphics are the resource the hammer and pick, with an green check indicating maximum development already done this turn and an orange check indicating 1 or more levels (but less than max) of development done this turn. A digit indicates how many levels to go until fully developed (i..e. it reaches double damaged).

Capture of a region does not impact development level. Hence, if you develop a region then you are helping whoever captures it to produce there more quickly. Bombing does not impact level of development either.

ORIGINAL: Lebatron
That's suggests to me it's moddable? Could it be made to take 3 turns and 30 supplies for instance?

yes, modable, within the limits described above.

ORIGINAL: Lebatron
And what about factories? Can I use this to place partly constructed factories on the map and have them take more time and supply than currently?

It doesn't support factories. I always try to do things generically, which in this case would have supported all fixed infrastructure (fact/res/rail), but in this case because of the way certain things are coded and because I am trying to maintain save compatibility that would have been rather difficult. So it is resources only.

I don't think this is necessary for factories. You can already build factories. You can build them in one place then rail them around if you want to. But I think this makes a lot of sense for resources which exist all over the world but take time and effort to begin extracting.

Here's the image. The last two icons aren't actually used, I just filled the rectangle.

[Lebatron]

ORIGINAL: WanderingHead

My personal preference would be:

* all air units (CAG and non-CAG) pick unique targets until all naval units are targeted. This would remove the CAG doubling up on CVs feature.

* CAG units select targets first, with weightings CV 16, HF 4, LF 2, TF 1, SF 1 (reducing CV weight from 40 to 16).

* non-CAG air units select unique targets next, with equal weightings on all naval units.

Only if you have more air units than naval vessels will naval units be double targeted. If a targeted unit is already destroyed before an air unit has a chance to fire, the air unit would lose its firing opportunity (unchanged from today, and this is simply not changeable).

With CAGs firing a little less on CVs, escort fleets can perform a more useful function (cannon fodder in this case). In some sense it's also a bit nice from the attackers standpoint, not to be focusing all the CAGs on a few CVs that may be overtargeted already.

I think this would work well enough, however since CAG won't be doubling up on CV's, I don't think reducing the CV weight down from 40 will even be necessary. If the only change to the current system that was made was adjusting the CV weight, I would not go any lower than 30. But you're adding in a preference to pick unique targets so the effect gives CV's a bit more survivability, thus it should stay at 40 or close to it.

[WanderingHead]

here's Iraq, unmodified, 2 levels of undevelopment.

[WanderingHead]

and after doing 2 levels of development on one resource and 1 level on another. The green check indicates no more development can be done on that resource this turn. After the WA turn is completed, the resource with the green check and 0 development levels would become a regular double damaged resource. This delay is necessary for smooth integration into the system, and forces the minimum 2 turn development time.

[WanderingHead]

ORIGINAL: Lebatron
I think this would work well enough, however since CAG won't be doubling up on CV's, I don't think reducing the CV weight down from 40 will even be necessary. If the only change to the current system that was made was adjusting the CV weight, I would not go any lower than 30. But you're adding in a preference to pick unique targets so the effect gives CV's a bit more survivability, thus it should stay at 40 or close to it.

That preference stated, I wouldn't change the weighting. Nothing like this should be changed lightly. Default should be no change.

[Lebatron]

Do you mean your going to add the new unique targeting and keep the weight at 40, or did you mean by the post above that your not changing anything?

[Lebatron]

ORIGINAL: WanderingHead

ORIGINAL: Lebatron
I like that your adding something to support raw resources as I suggested.

I think this was discussed long before you suggested it.

Brian,
In an email discussion with you back in January of 07(I remember this date because a friend of mine had died and I dropped off the map for a while)when you were working on adding moddibility features, I remember suggesting to you to try an add extra damage levels to resources to give modders greater flexibility. At the time I think you said you liked the idea but didn't know if you'd get it done. You made no indication to me that you already thought of it. So to be fair, I think I can claim it as my own. And to be honest Brian do you think you would be adding this new feature right now if it wasn't for my recent post @here@ that outlines why GG is basically flawed in making the Middle East richer without making repairing more time consuming and expensive. You went full steam ahead to fix that because you knew I was 100% right. Going back one year earlier, when I was brought on to help with AWD alpha back in January of 06, two years ago, I made the suggestion to Joel to allow damaged infrastructure to be placed into scenarios. One example I gave him was that some of China's infrastructure should start damaged in the 1940 scenario. He said he would have Keith work on it given time, but it never happened till you added it. And I never read any modders asking for it before I did, so I think it's fair to say its been done because I pushed for it.

To comment on the method you used to implement my suggestion I found your description hard to follow. Had to read it over and over, maybe I'm just tired from last night. And why use a new icon instead of sticking with the damage stars? A purple and black star or whatever color would blend in better.

[WanderingHead]

ORIGINAL: Lebatron
Brian,
In an email discussion with you back in January of 07(I remember this date because a friend of mine had died and I dropped off the map for a while)when you were working on adding moddibility features, I remember suggesting to you to try an add extra damage levels to resources to give modders greater flexibility. At the time I think you said you liked the idea but didn't know if you'd get it done. You made no indication to me that you already thought of it. So to be fair, I think I can claim it as my own. And to be honest Brian do you think you would be adding this new feature right now if it wasn't for my recent post @here@ that outlines why GG is basically flawed in making the Middle East richer without making repairing more time consuming and expensive. You went full steam ahead to fix that because you knew I was 100% right. Going back one year earlier, when I was brought on to help with AWD alpha back in January of 06, two years ago, I made the suggestion to Joel to allow damaged infrastructure to be placed into scenarios. One example I gave him was that some of China's infrastructure should start damaged in the 1940 scenario. He said he would have Keith work on it given time, but it never happened till you added it. And I never read any modders asking for it before I did, so I think it's fair to say its been done because I pushed for it.

To comment on the method you used to implement my suggestion I found your description hard to follow. Had to read it over and over, maybe I'm just tired from last night. And why use a new icon instead of sticking with the damage stars? A purple and black star or whatever color would blend in better.

You make it seem as though to discuss things with you one must document them. If you thought of it way back in Jan 07 then I honestly don't know who thought of it first, all I know is that it has been on my mind for as long as I can remember. I also know that just because one person says something first doesn't mean that person thought of it first.

I just get tired of hearing "my idea" everywhere. I don't really care who's idea it is. Most good ideas are synthesized from discussion and don't have a single parent anyway; standing on the shoulders of giants as it were. Read carefully and you'll see I didn't even claim this idea was mine. Can't we ditch that and just start talking about "good ideas"?

As to your post here, yes it prompted me to reconsider it. It always bothered me that it wasn't there. This is like the straw that broke the camel's back, not like "wow I never thought of that before".

I'll try another way to describe the current implementation.

There are 8 available "damage" levels:

0) undamaged, producing resources.
1) single damaged
2) double damaged
3) 0 level undeveloped
4) 1 level undeveloped
5) 2 level undeveloped
6) 3 level undeveloped
7) 4 level undeveloped

Resources can move around between 0 and 2 as we are all familiar, bombing and capture etc.

Resources never move downwards into state 3,4,5,6,7. Capture, bombing, etc never move a resource beyond the standard double damaged.

Capturing or bombing a resource that is at level 3,4,5,6,7 does not change the status of the resource, it goes neither up nor down.

A resource is automatically promoted from state 3 to state 2 (double damaged) at the end of the production phase. This imposes another 1 turn delay before it can be fully repaired (the player can _not_ expend supply to force the promotion from 3 to 2, the player must wait).

Resources may be advanced from states 4,5,6,7 at cost of 5 supply per step. The number of such steps allowed per turn is configurable, but never past state 3 ("0 levels undeveloped")


This implementation was pretty easy to glue on to the existing system, with minimal impact. I think I only had to touch 4 pieces of code: the data file read, the icon display, the repair screen display, and the repair execution. Hence high confidence it works (unlikely to introduce a bug). So I do hope you will find it acceptable as is.

Why knew icons? I thought about color coding but it doesn't convey enough information. If we have 4 levels, IMO we need to use numbers to denote the level. If there is a limit to how much advancement you can do in a turn, then you also need something to denote how much you have already done this turn (the green and orange checks), or you'll be frustrated not knowing which ones you can still advance.

We could replace the hammer and pick with a dark grey explosion to symbolize undeveloped resources, with the numbers and checks. Thinking about it now, I would change the check color coding. Green for advanced by some amount (less than maximum) this turn, red for advanced the maximum allowed amount this turn. That way the colors match the supplied/fueled icons.

[WanderingHead]

I changed the check mark color coding to be consistent with supply and fuel.

green = have developed this turn (1 or more levels, but less than maximum)
red = have done maximum development this turn (can not do more development)

In the below image, the resource with "2" is at its start state, the green checked "1" has been advanced by 1 level this turn, the red checked "0" has been advanced by 2 (the maximum with my configuration) this turn.

Again, if you have better ideas for icons then please work them up! I just think it should convey the same amount of information.

[WanderingHead]

ORIGINAL: Lebatron
Do you mean your going to add the new unique targeting and keep the weight at 40, or did you mean by the post above that your not changing anything?

What I meant was that where someone speaks up for the status quo, we should probably go with the status quo. So given your input, my current proposal would be to leave the weighting the same, eliminate CAG double-up on CVs, and have all air units pick a unique naval target.

If I understood you, then you would be happy with this. I'd like to see more people speak up on the CAG->CV targeting though. The lack of unique targeting thing is IMO a bug, but changing the CAG targeting is changing something very intentionally put into the system. I'd rather see relatively broad consensus.

[pzgndr]

I played a couple games of AWD this past weekend. I don't play often, but when I do I usually let the computer handle the research and builds. Needless to say, there are some AI weaknesses. For example, as WA I wasn't getting CAG replacements well into 1942 and needed two of them. Airborne doesn't get built at all. Etc.

I'm not asking for an AI overhaul, but could some attention be given to the AI text file to improve things a little? I note only one editable AI file; is this correct? It seems like a one-size-fits-all script. Considering the different levels of difficulty and different combinations of human/computer control, I would expect the AI file to be more robust. Perhaps this could be enhanced in a future update.

If nothing else, could some more guidance be provided for how to edit this AI file, and some of the pros and cons for changing what's already in there? The notes are pretty cryptic.

[Marshall Art]

I am not deeply into the subject as you may be but from this more general view a couple of questions:

- If a power has to spend 20+ points on development of resources to eventually get some resource points out which power do you believe can afford that?

Axis early on: maybe but they have to capture those resources first and when they do they already have a scarce supply pool.
Allies early on: no way since they eat out of their hands and when they hit the high multiliers they do not need any more supplies as they already do not know what to do with those 100s of supplies.
So who will actually have a benefit from this?


- As far as the CAG targeting and the targeting of other air units goes: yes that is a great annoyance for the attacker, if you send 5 air units to target 2 ships and they sink one but let the other slip away. BUT - is it far from reality?

I have not found any sign in all the various air-to-sea engagements where simply all targets were erased, unless the numbers of the attacking force were too strong. Look at the carrier battles: In most cases there were some CVs hit and sunk/damaged, others were not. Air groups tried to team up on one target to increase chances of landing a sufficient number of hits to sink one target. Better to sink one than scatch three... Also look at air-to convoy battles like in the Mediterranean or N. Atlantic: rarely all target ships were destoyed, most of the times some got away. Also a cloud or pure mis-navigation might cause a couple of squadrons to abort, which might be represented by the non-attack due to loss of target.

I do not recall air squadrons chatting over their target fleet "hey Aquadron No. 1 you take on CV No. 1, Sq. 2 takes on Sq. 2, etc .... Oh wait Sq. 5, you must not attack CV No.1 because Sq. No. 1 has this privilege, take a DD instead!"...

That is why I suggest the following (if it is possible):

reduce the chance of double-teaming but still make it a common factor. How about giving each CV 40 pts, a BB gets 4 and so on, but after one ship was targeted by an air unit, for the next air unit "searching" for a target that specific odds are halfed, i.e. only 20 for that particular CV from now on. If it gets targeted a second time, it receives only 10 for the later air units targeting. Is that an option?

Also if it were possible let the strongest air units (those with highest ship attack ratings) choose first, and the others follow. SO the main focus of the attack weould remain prime targets (like carriers) but other not-so-experiences units might actually take a DD by mistake...

[WanderingHead]

ORIGINAL: Marshall Art
- If a power has to spend 20+ points on development of resources to eventually get some resource points out which power do you believe can afford that?

Axis early on: maybe but they have to capture those resources first and when they do they already have a scarce supply pool.
Allies early on: no way since they eat out of their hands and when they hit the high multiliers they do not need any more supplies as they already do not know what to do with those 100s of supplies.

In answer to your comment ... exactly!

What was the importance of Iraq in the strategic thinking of 1940-41? It was not that the Allies could extract more oil. They had enough oil from the USA. It was that the Germans could. They did not have enough oil.

Although the great oil potential of Iraq was already known, it was barely exploited. The British were not motivated to exploit it during the war. The British poured concrete down some wells, to make it more difficult for the Germans if they should ever get there. What I would like to see is a reasonable possibility for the Germans to capture and exploit a good amount of oil. They could have, but it would take more time than the current engine allows. At the same time, in the game today if you simply use starting-as-damaged resources then the WA have every motivation to repair them, because this does not in any way help the Germans (and resources do help the WA, although not until the later game). With the mechanisms I have currently implemented, the British will not be very motivated to fire up the wells early on (the Germans might capture them and benefit from the development), and later one will have to weight the benefit against the cost (mostly because of the delay imposed in building up the resources).

In order to successfully exploit the resources, the Germans would have to get there _fast_ or give up, which is a reasonable model of the real war ... they tried to get there fast and on the cheap, but essentially gave up in 1941.

ORIGINAL: Marshall Art
- As far as the CAG targeting and the targeting of other air units goes: yes that is a great annoyance for the attacker, if you send 5 air units to target 2 ships and they sink one but let the other slip away. BUT - is it far from reality?

<snip>

I do not recall air squadrons chatting over their target fleet "hey Aquadron No. 1 you take on CV No. 1, Sq. 2 takes on Sq. 2, etc .... Oh wait Sq. 5, you must not attack CV No.1 because Sq. No. 1 has this privilege, take a DD instead!"...

Actually, I think they probably did do that to a reasonable extent, to the extent that couldn't concentrate on one vessel then reconcentrate on another (which would be "retargeting" which can't be done in the engine). Certainly as a commander I wouldn't order more units against a single vessel than I was certain could sink it.

The issue I find for the player is that if you send two air units against, say, two transports you will often (50% of the time) find that one air unit does not fire. You destroy only one transport.

But if you send in both air units but one air unit at a time, you hit both transports.

The results are counter intuitive, and extremely difficult to avoid even if you know the mechanism (habit says "throw more in and you do more damage").

The reason air units were given independently determined random targets, instead of unique targets, was because of the impact of suppression against land units. Air-to-sea is not suppression fire, so far as I can tell it was an accident that this targeting was applied for air-to-sea fire.

(FYI the code uses the name "air-to-land" to represent both air-to-land and air-to-sea components, hence it was an easy mistake to make. It is really "air-to-surface")

[Marshall Art]

Your motivation to implement the resource development I understand but if no one is going to actually use it it is not worth it IMO. You should give some more encouragement for both sides to really develop at least some of the resources.

- Although 1 year development time seems realistic, for gameplay reasons I suggest to reduce it to 2 turns each costing 5 supplies (+10 for "repairing" as usual). Thus only 2 levels instead of 4. This gives a sooner pay-back, so both Axis powers have a reasonable hope for keeping the resources until the are operational. As it is, with up to 5 turns total to get a resource fully operational, they won't do it unless the already are close to AV and then it becomes a side issue anyway.

- Actually FORCE the Allies to develop resources by making the operational ones on the map undeveloped in part, give the Allies a few supplies more in the beginning so they can start developing one or two at a time- Or if they choose to not develop them and spend the supplies elsewhere very soon (say in 1941 or when USA goes to level 2 they will face a certain shortage of war materials...

- To make the players more familiar with the concept and really make it a more common action add a few undeveloped resources to areas fully developed otherwise: e.g. add a resource to Germany (to simulate the increasing refining of fuel out of coal); add one to Romania (to account for increasing oil production); add a couple in urals and/or siberia (to account for increasing importance of that regions)&nbsp; - also maybe a few in the USA.

- I also think that if you put a factory into the production queue in the Urals, in the USA and Germany this will go inline with the gearing-up of economies. Those factories could be somewhere half-built which still leaves a significant investment to be spent but not 6 turns which lasts forever. The fixed multipliers just not provide enough choices for players to optimize production in a reasonable timeframe, based on availability or resources.