8MP Team Game - The Axis team

[Crackaces]

In a Pulp Fiction fast forward glimpse of the game ... The below picture at turn 60 shows the front and I have highlighted important rail junctions. That has been my focus since turn 55. It starts with cutting the rail at Svboda or the yellow pentagon red line .. cutting that rail dramatically increased the MP cost for Soviet supply. On turn 60 the next rail line supplying the units south of the Don is cut extending supply even further. My future plan is the Yellow star dramatically cutting off supply and extending supply MP's. Every turn it costs more trucks. Maybe there is an endless supply from the United States but I am counting there is not ...

To the north the Germans cut a rail line near the Oka increasing supply MP's ...

[Telemecus]

ORIGINAL: joelmar
I have a question about the air war, in Telemecus's comments and explanations about the centralized model for the bombers groups, it comes at turn 15 coinciding with the beginnning of the industrial strategic bombings phase. But it's not clear to me if this model appeared there because you devised it then, or if it was on purpose for the strategic bombings. If it was only devised on turn 15 I'm wondering if you guys would have adopted it right at the beginning of the game if you had known then? I can't figure out reasons not to.

I hope I can help by answering some of these. If you look back in the AAR you will see Stelteck who was our centre commander for those turns complained that resources were going to other areas. Stelteck liked to complain [:)] But in private thedude3571 would complain that the south was getting all the southern allies air forces only to help. The North would get all the dive bombers and very short range bombers like the Ju86 and Do17Z. But very soon into the game the Heinkel and Ju88 bombers were already all being centralised to the centre. Once the Neva was crossed so we knew Leningrad was only turns away from falling all the shorter range bombers also came to the centre. So there was centralisation well before turn 15, although it did increase with time and indeed it became even more centralised after turn 15.

I would say there are several reasons for this

i) Rail repair: The rail repair in the south in 8MP was turns behind where it would be in other games because of the opening strategy - so there was not a good alternative position in the south to base bombers at. Many places in the south actually found their nearest railhead in the centre and not south. Also as FBDs turned from the north south and south north to meet to make a lateral connection, only the FBD heading to Moscow was still heading on a fully easterly course. So the rail network in turn 15 naturally meant that was the location on the rail network that was more easterly as well as central compared to others.

ii) Priority of the battle of Moscow. The most intense battles and warfare was occuring near Moscow up to turn 15 - so concentrating the air there also made sense.

iii) Withdrawal of a "distributed" Soviet air force - although the Soviet air force did eventually withdraw from the map entirely - they started to concentrate themselves several turns before. When you have a Soviet air force everywhere - and given our success in this air war at that time - then bombers everywhere would never run out of targets. Once there was no Soviet air force south of Voronezh, it meant any Axis air force near Crimea say would have fewer advantageous targets. So the concentration of the Luftwaffe was to some extent mirroring the concentration of an opposition.

iv) But yes the big change was the moving on to mainly strategic bombing. And it is important to stress the highly stratified nature of the targets. You could think of the bombing damage to the fighter factories at Saratov as being like winning 1,000,000 gold points, damage to the fighter factories at Gorky as being like winning 500,000 gold points and hitting some other factory as being like winning 100 gold points. I am not too sure about the numbers but this is just to stress the order of magnitude. The fighter bomber factories were not just our priority factories- they were many many times more valuable than the next set of targets. And our next level were similarly more valuable than the next and so on. Because of this it made no sense to split your best bombers between north and south and hit each with only half of them. Before the strategic bombing became the main focus the value of the targets was not so differentiated - and there were far more of them. So even if you split your force all would kept busy with targets of roughly similar value.

So in answer to the question I would say there was always some centralisation. The move to a mainly strategic bombing campaign just tipped the balance even more extremely that way.

I should say there are counter arguments to centralisation. One would be distance flown to target - as well as flying east-west your bombers are having to fly many more miles north-south on this model. So you do need the advantages to outweigh these disadvantages.

ORIGINAL: joelmar
I haven't read much about WitE2 but I hope for something that permits to assign needs to airbases, so even an empty air base at logistics could be fully loaded of supplies next turn or something in this line. lol... it's a mystery to me why the air management was devised the way it is, and also a wonder, because I don't understand how programmers that were so efficient in creating a game engine as complete as this could create something so weird and complicated for a part of it, it doesn't even logically model anything that could happen in real life. I mean, Kesselring could surely send an airbase and full it of supplies as he pleased, even if it was empty of air groups.

To be fair to the developers I assume the air war model was a bit of an afterthought in what is meant to be mainly a ground war game. So a lot of the consequences of what they designed such as the Kabuki dance were not anticipated by them but were an emergent property afterwards. WitE2 - which uses a similar model to WitW - tried to address this. Unfortunately I think they got it from the wrong end. Instead of getting rid of all the back office complications (airbasing, logistics) they got rid of the front office point and click to bomb model. Pointing and clicking to launch an air mission is what most people actually enjoy. I think the correct way would have been to keep the model of simply clicking to bomb or attack something with the airforce - but getting all the boring airbase movement and logistics stuff under the hood. At least my criticism is they tried to get rid of the wrong end of the air war mechanism.

In history we know the Luftwaffe had hundreds of airfields on the Eastern front - not 24. These counters are only meant to be abstractions of a far greater number of airfields distributed over a large geographical area. In WitW and WitE2 there are no counters representing airbases to move around anymore. Instead they are actually printed into the map like cities and other geographical features. Although they still have to be repaired and supplied etc. But on the supplies side I think they might have got it right. The Axis did actually have problems getting supplies from Germany to the places they needed them on the eastern front - although why and whether this is accurately modelled in the game is a good question.

[Telemecus]

ORIGINAL: joelmar
That said, I love the micromanagement and those kind of decisions are only more fun to be had! lol! It takes me forever already to play a turn. The excel spreadsheets have given me ideas about a MySQL database for managing all that data. I started work on it, but it will take sometime yet.

Join the club! Dinglir, ledo and fetmun are also known for their spreasheets. And if it is something you can share I do recommend telling EwaldvonKleist as he will tend to host them for everyone to use in the library of WitE resources. Indeed I was hoping to pool some of our efforts and co-develop some of them in future.

But databases are even more hardcore- even spreadsheeters shake in fear there. [:D] KenchiSulla is one user here who did start to develop game databases though [:'(]

[Crackaces]

Well while I await the turn from M60 I am looking at the entire front. Something just hit me .. we have just started August!

The Soviets have done an excellent job of keeping the Germans away from there main objectives. Now to use this time to think where will be the MLR (main line of resistance) for the winter? Rivers are worthless once they start to freeze. I am seeing many problems including a Soviet recapture of Moscow ...

The Soviets have so many opportunities if they can just get to winter .. my strategy is to reduce manpower through multiple means, and limit rail capacity. Hopefully, we do not suffer a catastrophe in the meantime.

[Crackaces]

Not to double post -- but to start a conversation while M60 works through the turn ..

The similarities between Stelteck 1942 and this AAR are remarkable. Stelteck was fighting 7M Soviets and the Germans here face 5.5M. But as I look at the situation I assess M60 has built more Guards and more so, it looks to me that M60 has built more SU's. I believe this has offset manpower disadvantages with firepower. I am also seeing more Soviet Guards Armor corps at this stage of the game.

One thing the Germans did differently in this AAR is to redirect Panzers to cross the Volga earlier rather than later. Turn 60 saw the 1st Panzer Army crossing the Volga. I might delay the Soviets from their victory parade in Berlin [:D]

Turn 56 when the turn comes back .. now a commercial break [8D]

[Dinglir]

ORIGINAL: Telemecus
To be fair to the developers I assume the air war model was a bit of an afterthought in what is meant to be mainly a ground war game. So a lot of the consequences of what they designed such as the Kabuki dance were not anticipated by them but were an emergent property afterwards. WitE2 - which uses a similar model to WitW - tried to address this. Unfortunately I think they got it from the wrong end. Instead of getting rid of all the back office complications (airbasing, logistics) they got rid of the front office point and click to bomb model. Pointing and clicking to launch an air mission is what most people actually enjoy. I think the correct way would have been to keep the model of simply clicking to bomb or attack something with the airforce - but getting all the boring airbase movement and logistics stuff under the hood. At least my criticism is they tried to get rid of the wrong end of the air war mechanism.

I disagree on that, mainly based on a game balance perspective.

With the IGOUGO system, the in turn player is basically always playing the AI. The out of turn player can set the basics for the AI every turn, but this is still very much exploitable. Basically, you can set up to protectyour airfields by setting interception "On" or you can set ground support "On" along with a decent group range on your aircraft in order to get them to the combats. If you try to do both, your settings are very exploitable.

The WitW system basically let both player set the setings for the AI, which then performs that air war on its own. That evens the playing field for the in turn and out of turn player.

That being said, my instinct tells me that the air war will be a lot more complex than it is today with many obscure details to consider. That will mean that the player with the better understanding of the system will have a huge advantage over the other player. I look forward to seeing how this will be balanced once WITE II is released.

[Dinglir]

ORIGINAL: Telemecus
Dinglir, ledo and fetmun are also known for their spreasheets.

Back when I did my first spreadsheets they were somewhat advanced. But when comparing to the stuff that has been released over the past year or more, my spreadsheets can hardly be compared favorably.

[Crackaces]

ORIGINAL: Dinglir

ORIGINAL: Telemecus
To be fair to the developers I assume the air war model was a bit of an afterthought in what is meant to be mainly a ground war game. So a lot of the consequences of what they designed such as the Kabuki dance were not anticipated by them but were an emergent property afterwards. WitE2 - which uses a similar model to WitW - tried to address this. Unfortunately I think they got it from the wrong end. Instead of getting rid of all the back office complications (airbasing, logistics) they got rid of the front office point and click to bomb model. Pointing and clicking to launch an air mission is what most people actually enjoy. I think the correct way would have been to keep the model of simply clicking to bomb or attack something with the airforce - but getting all the boring airbase movement and logistics stuff under the hood. At least my criticism is they tried to get rid of the wrong end of the air war mechanism.

I disagree on that, mainly based on a game balance perspective.

With the IGOUGO system, the in turn player is basically always playing the AI. The out of turn player can set the basics for the AI every turn, but this is still very much exploitable. Basically, you can set up to protectyour airfields by setting interception "On" or you can set ground support "On" along with a decent group range on your aircraft in order to get them to the combats. If you try to do both, your settings are very exploitable.

The WitW system basically let both player set the setings for the AI, which then performs that air war on its own. That evens the playing field for the in turn and out of turn player.

That being said, my instinct tells me that the air war will be a lot more complex than it is today with many obscure details to consider. That will mean that the player with the better understanding of the system will have a huge advantage over the other player. I look forward to seeing how this will be balanced once WITE II is released.

I might disagree that WEGO dismisses the dependence on the AI. I played WITP AE for years and the AI takes over and resolves air attacks. Sometimes the results are puzzling

[Dinglir]

ORIGINAL: Crackaces
I might disagree that WEGO dismisses the dependence on the AI. I played WITP AE for years and the AI takes over and resolves air attacks. Sometimes the results are puzzling

I am not trying to say that the players are less dependant on the AI, merely that they are equally dependant on the AI. That makes for a less exploitable system.

[Crackaces]

ORIGINAL: Dinglir

ORIGINAL: Crackaces
I might disagree that WEGO dismisses the dependence on the AI. I played WITP AE for years and the AI takes over and resolves air attacks. Sometimes the results are puzzling

I am not trying to say that the players are less dependant on the AI, merely that they are equally dependant on the AI. That makes for a less exploitable system.

The exploit in WITP AE is to put suicide single ship task forces (like small merchants ) in range of the target. The AI will send airplanes unequally toward the suicide ships. So a home rule is needed. See CanoeRebel AAR’s. Write code and someone will figure out a way to exploit it to an advantage.

My post seems argumentative ...

So to explain .. it is clear that if the designer and programmers work hard then a WEGO system eliminates exploits. But thus far -- reality and human behavior set in and the AI is exploited.

WITP AE has some interesting air combat resolution algorithms .. that I have been lambasted for pointing out (How dare I?!/!) For example, interception occurs over the target hex instead of a circles of apollonius algorithm. So a fighter group 1 hex away from the incoming attack will fly to the target hex to resolve combat .. that is exploitable ..

Agreeing with Dinglir .. I have always thought a WEGO system for WITE would demonstrate the problems of armored combat on the Eastern Front. Especially early on .. The Soviets helped create pockets by trying to intercept moving armor forces where they were last and not where they are going to be when the tanks get there .. so they missed contact and found themselves surrounded ... That does not happen in a YOUGOIGO system ..

[joelmar]

Thanks Telemecus for such a great answer, much more than I bargained for to be honest.

Your answer confirms what I had figured and was about to implement in my current campaign against the AI following what I understood of your explanations: have things centralized and keep a fluidity by moving forces north or south as the situation develops and needs arise. I also understand that the situation you were in is not common or expected, as the soviets were playing right in your hands. In the present situation in my game, I am right at the doors of Moscow and Leningrad, turn 12, so centralizing will be a big benefit for my coming offensive against Moscow, and I will shift most of the rest of the German Air Force against Leningrad. To be honest, I had a tendency to think about the Air Groups within the structure of Luftflotte and Fliegerkorps, but light has finally come thanks to your explanations: I need to see the Luftwaffe as a big pool of planes and get them were needed as the need arise... much the same way as the SU's. I would guess this way of seeing things is common to newbies who are infatuated in the PocketHub Porn and don't care much about the birds in the sky! lol!

Another mistake I made which you put under a big spotlight, is that I was thinking more about saving airmiles on each mission than maximising supplies. So I would put my airbases as forward as I could, in the 10 hexes limit, not on the railheads. But now I see the maths more clearly, the 10 hexes difference is not that much of a save compared to the expenditure on trucks and better supplies. National Reserve was also a big eye opener... also all newbie stuff I guess!

Result, I didn't make as much headway as I was expecting against a battered AI.

For the developers, I have much respect, this game is incredibly complex, and is clearly a work of art. I have read much of the thread about WITE2 before writing this comment, and through all the comments in that thread, your explanations in this thread and Dinglir and Crackaces discussion above put that even more forward. I realized levels of complexity deeper than I would have thought and people in this community incredibly knowledgeable... which is quite inspiring to me to be honest. Anyway, this reinforces my conviction that this game as so many levels of complexity that it's quite hard for a player to grasp everything, let alone for programmers to get everything right and balancing everything so that Elite, intermediary and novice player all find their pleasure. I mean... war is war and life is life. In reality, a good General will kill a bad one because of better technique and more importantly, better feel, there always will be Hannibals and Napoleons, and that is part of the game. So much depends on that. But I know a game designer has to make things fun for everyone playing the game to get them hooked.

One thing is certain, from what I understand for the Air War in the WitW style, the AI is king... which is somewhat disappointing to me also, but I guess it's OK if people like it and if it balances things out a bit, thus making the community of players stronger.

For the databases, work is started on my part, I was a bit on the slowmo partly because of what I explain below... but your comments have kicked my butt. Believe me, I know my stuff and I have many ideas and the means to make some of them happen fast. That said, I would really like to have experts input. I'm sure you guys could help me with some reporting you would have liked to have but couldn't because of the limitations of the spreadsheets.

But I have a small limitation for making this available to all. I'm a LAMP web developer. So the tools I use are not made to create .exe applications. My applications are written in PHP and have to run on a web server. I used to know other technologies that compliled code and created .exe files, but it's been more than 15 years, a lot of water has gone under bridges and it's not practical for me to learn them again only for this ludic purpose.

But sharing it to all is not impossible, as installing a free WAMP server on your computer and using it as a platform for the application and then accessing it as localhost in the browser is not rocket science for the nerdies in this community, even though it's a little more complicated than clicking on a .exe. Is it something you and others would be willing to do?

Also of course if others would want to participate, PHP has the advantage of being open source by nature and quite easy to play with and modify to anyone with a little knowledge of scripting.

[Telemecus]

I would certainly be up for the model development. Philosopically I am open source - I even use open office for spreadsheets here which brings others hilarity. I used to dual boot with Linux but practicalities meant I had to keep coming back to windows. And PHP was my next learning project.

I know with my spreadsheets I was trying to model the wiring that goes behind the game so that I could do what-if analysis or simulate alternatives. So for example the spreadsheet attached earlier on soviet aircraft includes all the formulas to say what will happen when factories get bomb damage and so on. From there I could forecast forward what would Soviet fighter numbers be if we changed our bombing priorities - and found it to be a winning strategy. Ultimately I would have liked to also link that to how much aircraft factories consume in supplies for their production and so on. By profession I am an Economist used to making big models on how the economy - and people - behave and try to put all the wiring and feedback loops in. So this probably betrays my reasons for going there. At the moment there are only partial models of aspects of the game, whereas I would like them to be modules that link up to the others. In my dream model the exports from the commanders report in the game would be the data inputs to a big spreadsheet/DB model - and altering some data on the sheet could tell us what other things would have changed.

In many software businesses I have known they tend to have "business analysts" who try to nail down what is the business logic of an organisation to transmit that to a programmer who turns it into code. One problem with WitE is there is still a lot of unknowns - so the big problem you would have is not so much the programming as knowing what to program. I know you would get many willing volunteers to act as your "business analysts" here including myself. But it would be the longer job. It took me one year for instance to work out exactly how bombing damage affects aircraft and AFV production (does the production calculation take place after or before repair in logisitcs phase etc.) In all the years I have had the game I have never once been able to simulate the manpower numbers though. I know, given losses and other inputs, I can approximately get the maanpower numbers for next turn - but never get it exactly right. And that bothers me as those small errors can sometimes become very big when you try a novel strategy involving it. So the biggest job is still trying to work how, exactly, the game works.

[Dinglir]

ORIGINAL: Telemecus
I know, given losses and other inputs, I can approximately get the maanpower numbers for next turn - but never get it exactly right. And that bothers me as those small errors can sometimes become very big when you try a novel strategy involving it. So the biggest job is still trying to work how, exactly, the game works.

AFAIK, manpower is produced the same way as any other production. That means that if a town/city/whatever has bewteen 1-50% damage, there is a chance it will produce, and a chance it won't. Given that, you should never be able to simulate exactly how much manpower is produced.

[Telemecus]

ORIGINAL: Dinglir

ORIGINAL: Telemecus
I know, given losses and other inputs, I can approximately get the maanpower numbers for next turn - but never get it exactly right. And that bothers me as those small errors can sometimes become very big when you try a novel strategy involving it. So the biggest job is still trying to work how, exactly, the game works.

AFAIK, manpower is produced the same way as any other production. That means that if a town/city/whatever has bewteen 1-50% damage, there is a chance it will produce, and a chance it won't. Given that, you should never be able to simulate exactly how much manpower is produced.

That accepted for the Soviet Union in 1941 - and you could only do an expected figure. But most of my attempts have been with Axis in 1941. I eliminated games where the Soviets "marauded" into Axis territory - so I was left with manpower factories that were wholely undamaged - and double checked it to be so. So there were no random elements (apart from some random rounding down/up of fractions formulas).

As far as I could tell it had something to do with how many disabled returned. And in one post about it Morvael said there was a bug to do with the different nationalities that should have been fixed but was only fixed in a later version - but it still does not work for me in this version [&:]

See this as an example tm.asp?m=4412059

There have been other occasions when I have found bugs as a result of this modelling - see for example tm.asp?m=4472786 - so if it is due to an ongoing bug it would be good to aat least know, and know which are the correct figures thegame engine works from.

[joelmar]

Great [:)] Challenging ideas indeed, creating those kind of tools would be a great achievement. But I'm afraid having this kind of report engine available will create even more layers of complexity to the game for the poor devil who wants to learn! Problem with this kind of complexity is that your opponent is beating you in a separate hidden war, but you don't even know there is war as Leonard Cohen would say!!!

Anyway, you will have to guide me in all this business logic, because you are light years ahead of me in those areas. Even your aircrafts spreadsheet I have yet to understand because I never got this point yet, my ideas come more out of frustration for not being able to see things clearly in the commander report and logistic report. There is much info in that last one also, but it's harder to import because the structure is not organised. Might be something to look upon in the future.

Not quite there though... but I have made headway, almost a breakthrough. At the moment, I am putting the finishing touches to the importation engine. Only import remaining are the locations and factories but it wont be very long to do. I will also have to add a kind of checksum at the end of importation to make sure everything is imported correctly, it would be very hard to detect missing rows otherwise and it would be misleading to work on incomplete datasets because of undetected importation errors, like data in a field longer than usual by exemple that creates an error and skips the row being imported.

Anyway, basically, you create a game in the engine and then import turn reports, the importation engin forces you to import turn 1 first and then the other sequencially. You can also easily delete all the imported game data in one click, or delete turn data individually.

I have started work on the first report, an improved OOB with all the available data and a better visual structure, like in one of the OOB spreadsheets. It has been one of my main frustrations with the commander report to have to go back and forth between screens to see everything.

A question for you guys, I have not yet exported a report from the Soviet side let alone analyse them, are they exactly the same in terms of fields structure? If it is so, then the importation works for both sides already.

I had read your post about manpower, still not certain about that part, so that is another thing I will have to work out, because some numbers seem to mismatch between the CR and the export CSV.

[Telemecus]

ORIGINAL: joelmar
some numbers seem to mismatch between the CR and the export CSV.

A common problem I have been trying to work out. Some of it is due to taking those numbers at different times during a turn e.g. before logistics phase, after logistics phase etc. So a big part is working out when exactly the numbers for the production screen and log reports are actually taken. (This does not explain all the discrepancies though).

For a similar reason I export csv at the start of the action turn (before doing anything) and end of action turn and keep both as records of the turn - that way for instance you can distinguish the fuel that was used by movement of units in your turn from what was supplied in logistics etc. It is worth remembering there is not one single csv export for the whole turn but that the figures are updated throughout your action phase.

ORIGINAL: joelmar
your opponent is beating you in a separate hidden war, but you don't even know there is war

[joelmar]

You're right that the export changes, and it's something I hadn't taken account of... [X(] I was working on the assumption of 1 report per turn. Not sure how it would be possible to make sense of many reports per turn, they would have to be sequential too, but that is beyond retrieving in the report data itself, which are all based on turn # and date of first day of the week. I think the best possible way of dealing with this would be to have a report at the beginning of each turn as basis for the general reporting, and then having a separate system to make comparison between reports of the same turn possible, but only parallel to the main system and in a case by case basis.

[Crackaces]

I had read your post about manpower, still not certain about that part, so that is another thing I will have to work out, because some numbers seem to mismatch between the CR and the export CSV.

My general take from reading more modern AAR's (Stelteck for example) is that "successful" meaning Germans are being pushed back happens around the 7M Soviet level. Coverage -- meaning Germans are generally stopped happens around 6.1M. The Soviets are at 5.5M in the field and are slowing the Germans down but cannot cover the entire front. They can counterattack but at their peril. Turn 61 sees a Soviet counterattack smashed and good size of the forces isolated. The Soviets cannot cover Stalingrad and the Caucasus while holding back 2 Panzer Armies in the center and one between Stalingrad and the center. I suspect 6.1M would be a difference maker.

I am thinking the Germans will bag another 70K returning 60K to the pools given the surrender rules on turn 62.. The Soviets are expanding at about 80K per turn given the current manpower situation? So one more turn of taking causalities, but not expanding manpower in the field. The whole causality total will be well over 100K because of units retreating over a major river. We shall see but the Soviets might be reduced to less than 5.4M.

NOTE: The Germans are not expanding either .. the combat pace including a winter offensive has hurt TO&E's .. a major reorg occurred on Turn 61. If the Soviet remains steady and accepts this attrition level -- eventually the German might break even with a lower Soviet manpower level. The typically game folds at this point so it is hard to know.

[xhoel]

Good to see the AAR is still continuing, good luck to you Aces!
PS: Can we get some screenshots of the losses so far? Both Air and Ground as well as destroyed formations I mean. Thanks in advance

[Crackaces]

TURN 56 OOB:

M60 was most gracious to take this game on. A small disaster is cleaned up turn 56 with the pocket around Rostov cleared.
This shows the OOB at Turn 56. The Soviets have 5.4M in the field and 5.3M ready to fight. And the Soviets are putting up great resistance using Guards Units