Are long term production line underpowered?

[valisk]

So... On topic...
Should there be a bigger malus for switching to a newer model, and a smaller one for upgrading a existing one?

Yes, as a new model means a complete retooling and setting up of lines, with productivity and efficiency being low until it gets into its swing.
With a modification, such as improved guns on a Spitfire, there is minimal disruption, only one part of the line will see any change and disruption would likely be minimal.

I'd say Concentrated industry should get a very much reduced hit on variant production, but take a full hammer to the minimum on switching that line to a different model.

Right now in my opinion there is almost no reason to take concentrated industry at all given all the extra goodies dispersed gets in terms of efficiency retention and base efficiency.

I think the last time I used concentrated was as the USSR in 1.8.x and I concluded then that Flexible was better overall.

 

[Denkt]

Yes, as a new model means a complete retooling and setting up of lines, with productivity and efficiency being low until it gets into its swing.
With a modification, such as improved guns on a Spitfire, there is minimal disruption, only one part of the line will see any change and disruption would likely be minimal.

I'd say Concentrated industry should get a very much reduced hit on variant production, but take a full hammer to the minimum on switching that line to a different model.

Right now in my opinion there is almost no reason to take concentrated industry at all given all the extra goodies dispersed gets in terms of efficiency retention and base efficiency.

I think the last time I used concentrated was as the USSR in 1.8.x and I concluded then that Flexible was better overall.

Based on what, have you actually done tests looking at the production of dispersed and concentrated?

There seems to be a myth that concentrated should not switch, I showed that in a 2 year intervall between switching the non switching method only produced 10% more IC but all of it went to inferior Equipments.

 

[valisk]

Based on what, have you actually done tests looking at the production of dispersed and concentrated?

None I could show you now as I cannot load my old 1.8 games

I just recall how dissatisfied I was with the huge loss of efficiency and consequent gaps in production, this could of course be worked around by micromanaging lines, such as starting a new line for new production and ramping down the old lines, while the new one increases in productivity, which is what I found I had to do to avoid to sudden a shock to production.

I will have a look again and do some tests later, but it seemed to me that for any line that needed any change, lets say XP gain lets me upgrade the armour on my t34s I would take less of a hit if I was using dispersed. (and Germany is also less able to bomb my factories)

For producing rifles and trucks non stop, concentrated is definitely superior as long as I have no bombers to contend with. But so much of my production is usually fighters, bombers and tanks, and I am constantly upgrading them as XP (and my attention span) allows, it might well be that if I micromange the lines switching a factory at a time to the new line, I would be better off.

 

[Denkt]

Making Changes is up to the developers as they know much better than we what the game actually needs. I simply try to look at the numbers and see how resonable they are.

Now number for dispersed-flexible line, same as concentrated-streamlined test I did Before:

I did run 2 test, both using Germany with all industral Techs, concentrated-flexible line. I used 10 military factories. In the first test I produced Panzer IV from 1936 to 1938 Before switching to Panther and stopped the test in 1940. The 2 year gap represent the 2 year gap in technology between the tanks.

In the second test I produced Panzer IV from 1936 to 1940, so the only difference between the test was the switch in production.

Here are the number of tanks I had in 1940 when I ended the tests:

​

• Test with switch: 3.9k Pz IV and 3.8k Panther, total tanks 7.7k, total IC cost of the tanks 103.9k
• Test without switch: 8.8k Pz IV, total IC cost of the tanks 114.4k

So the switch cost me 10.5k IC or about 1k IC per factory. This represent about 10% more production by not switching which in turn ment I ended up with 1.1k more tanks but that is also based on the fact that Panther cost 14 IC while Panzer IV cost 13 IC. However as time goes on the gap in production will get smaller and smaller since both lines will work at 100% efficiency and if we take account for the industrial Techs the gap would have been even smaller.

So even with concentrated industry you should switch production, Assuming you are not very desperate as the lost production is quite small while the quality gained is pretty significant. Obviously you have to consider stuff like research pritority.​

Here are the number of tanks I had in 1940 when I ended the tests:
​

• Test with switch: 3.5k Pz IV and 3.8k Panther, total tanks 7.3k, total IC cost of the tanks 98.7k
• Test without switch: 7.8k Pz IV, total IC cost of the tanks 101.4k

Here the gap betwen the two test is just 3% more IC for not switching which is pretty much nothing. Obviously the numbers are lower than those seen for concentrated industry but keep in mind I did only switch once in the first test and no switch in the second test which would favor concentrated industry. The gap between the switch test is quite small and dispersed start out in a stronger position, simply falls behind in the long run if no more switches is made.

 

[valisk]

Making Changes is up to the developers as they know much better than we what the game actually needs. I simply try to look at the numbers and see how resonable they are.

Of course it is

I don't expect Johan to be breathlessly awaiting my posts and demanding Podcat and Co implement my brain excretions with immediate effect, but that's no reason not to litter the forums with my own badly constructed concepts

 

[Denkt]

So no matter concentrated or dispersed, switching is probably what you should do and switching do make the gap between the two industrial path more closer in output. That concentrated should not switch is just a myth as in the long run the difference in total IC is pretty small. Dispersed have even less gap so if it make sense to switch for concentrated it make even more sense for dispersed.

The only reason you would produce obsolete Equipment is probably due to Tech situation, but the most produced Equipment should probably be the first targets for upgrades.

 

[Secret Master]

Right now in my opinion there is almost no reason to take concentrated industry at all given all the extra goodies dispersed gets in terms of efficiency retention and base efficiency.

And yet I use it successfully all the time in MP.

There are times when Dispersed makes sense, but I make extensive use of Concentrated, and it does not fail me.

My slavish devotion to researching techs in such a way that I still end up with MIC on the same piece of equipment for years at a time is probably a contributing factor. It also helps that some staples of the war don't have upgrades: trucks and support equipment.

 

[mursolini]

Right now in my opinion there is almost no reason to take concentrated industry at all given all the extra goodies dispersed gets in terms of efficiency retention and base efficiency.

I think the last time I used concentrated was as the USSR in 1.8.x and I concluded then that Flexible was better overall.

It depends on play style, but the general rule is concentrated>dispersed, unless you are MP Germany or some similar position.
Most techs come on 3-4 years upgrade cycle, but you do have tech rushing bonuses and general tech rushing that allow to extend or contract longevity of certain items.
Tanks are exception, but there are strategies that allow plenty of optimization.

War, in most cases, is carried on the back of your 1940 fighter, 1939 or 1941 medium tank, and if your are defending, 1940 AT. Most of these can be obtained a fair bit earlier than 1940, and so you will be past concentrated taking over by the time shooting starts. By the time most of these will be obsolete, you will win.

I also don`t think the issue is the amount of production shortfall, but the fact that newer models hopelessly out-class old ones. Your medium tank goes from 12 to 14 cost in 3 levels, but it`s stats will be increased 1.5-1.7 times. For fighters, it is worse, AT is far less useful unless it is modern. The rest of equipment matters far less, but the general issue is same. Far too aggressive rise of stats makes older equipment generally something you will never want to keep producing, unless you can`t afford increased resource cost.

 

[Denkt]

It depends on play style, but the general rule is concentrated>dispersed, unless you are MP Germany or some similar position.
Most techs come on 3-4 years upgrade cycle, but you do have tech rushing bonuses and general tech rushing that allow to extend or contract longevity of certain items.
Tanks are exception, but there are strategies that allow plenty of optimization.

War, in most cases, is carried on the back of your 1940 fighter, 1939 or 1941 medium tank, and if your are defending, 1940 AT. Most of these can be obtained a fair bit earlier than 1940, and so you will be past concentrated taking over by the time shooting starts. By the time most of these will be obsolete, you will win.

Tech rushing is however its own thing and don't have directly anything to do with production lines other than it can cause some very long running production lines.

I also don`t think the issue is the amount of production shortfall, but the fact that newer models hopelessly out-class old ones. Your medium tank goes from 12 to 14 cost in 3 levels, but it`s stats will be increased 1.5-1.7 times. For fighters, it is worse, AT is far less useful unless it is modern. The rest of equipment matters far less, but the general issue is same. Far too aggressive rise of stats makes older equipment generally something you will never want to keep producing, unless you can`t afford increased resource cost.

Don't think that have any issue, it encourage a more late game boom by building civilian factories early in hope of outproducing the enemy later, if there is an issue it is how early you can get these Equipment, like getting fighter 1940 early is a massive advantage if not the opponent do the exact same.

As I said before I would leave such stuff to the developers to figure out how they want the game to be.

Obviously production efficiency and variant system fail pretty much completely in this regard:

Our larger hope is that Production Lines and efficiency will offer players some interesting choices when it comes to deciding what to build. Should you go for a large number of weapons you can already churn out, or take a short term hit on production in favor of making a smaller number of higher quality ones? Sure, your new T-43 tank is better than the T-34, but is it really enough of an improvement to lose much of your Production Line's Efficiency when you switch over? Your PzIII tank may be obsolete, but perhaps instead of canceling their production entirely you could convert the Line to make Tank Destroyers or Self-Propelled Artillery on the PzIII chassis. Preserving efficiency in some of your factories could lead to a more diverse and interesting combination of units, and allow you to discover some parts of the game you might have ignored if you just constantly upgraded.

​

​

 

[mursolini]

Tech rushing is however its own thing and don't have directly anything to do with production lines other than it can cause some very long running production lines.

Well, historically, it was half of the reason people kept old lines running, switching them out was not worth it. If there is no late-game tech to switch out to, because your late game tech has arrived pre-war, issue becomes moot.

Don't think that have any issue, it encourage a more late game boom by building civilian factories early in hope of outproducing the enemy later, if there is an issue it is how early you can get these Equipment, like getting fighter 1940 early is a massive advantage if not the opponent do the exact same.

As I said before I would leave such stuff to the developers to figure out how they want the game to be.​

I don`t think it has anything to do with early or late game boom, since it strictly doesn`t mater, what kind of equipment you produce out of your factories, early or late game, if your strategy is to outproduce in late game.

It seems to be mostly there, to open up the game, since otherwise it will be a lot harder for emergent state to triump. Imagine there is no late game tech, everything is capped at 1936.
If you start as a small country, you will be fighting against years upon years of accumulated production backlog. Even if you manage to match some major in yearly production, you`re still years behind, which, is realistic, but probably not the goal of the game. With new equipment being so much better, the years of backlog get obsolescent very fast.

It is realistically impossible to challenge 1936 naval powers, unless you are already one, for example, unless their fleet just get`s so outclassed by new designs, it is simply written off, and only past 2-3 years of dockyard balance matter.

 

[Denkt]

I don`t think it has anything to do with early or late game boom, since it strictly doesn`t mater, what kind of equipment you produce out of your factories, early or late game.

Well it matter alot if you can produce more 1944 Equipment than the enemy if the game manage to somehow reach that Point, a late game military factory would be much better than a early game military factory producing early game Equipment in absolute terms if the game reach that Point. So if one country go for an early rush but fail it would most likely lose the war because it would not keep up in the late game. However Tech rushing cause quite alot of problem here.

Well, historically, it was half of the reason people kept old lines running, switching them out was not worth it. If there is no late-game tech to switch out to, because your late game tech has arrived pre-war, issue becomes moot.

Because it is in the game pretty cheap to switch out lines, if your factories was "damaged" when you switched so you had to pay civilian industry I think people may reconsider. The damage would represent producing the machine Tools needed for the production line.

It seems to be mostly there, to open up the game, since otherwise it will be a lot harder for emergent state to triump. Imagine there is no late game tech, everything is capped at 1936.
If you start as a small country, you will be fighting against years upon years of accumulated production backlog. Even if you manage to match some major in yearly production, you`re still years behind, which, is realistic, but probably not the goal of the game. With new equipment being so much better, the years of backlog get obsolescent very fast.

The Tech system encourage an aggressive strategy of using what you have Before it become obsolete, which is a good thing in my opinion. The issue is you can later year Tech for a rather low cost with national focuses.

It is realistically impossible to challenge 1936 naval powers, unless you are already one, for example, unless their fleet just get`s so outclassed by new designs, it is simply written off, and only past 2-3 years of dockyard balance matter.

Well the early ships don't become useless but their value drops alot over time, on the other hand they have the most dockyards so they tend to have more better ships as well.​

 

[HugsAndSnuggles]

Obviously production efficiency and variant system fail pretty much completely in this regard:

To be fair, conversion allows you to turn, say, tier 1 light tanks into tier 2 light SPGs, so there are some nice uses for old equipment.

 

[Denkt]

To be fair, conversion allows you to turn, say, tier 1 light tanks into tier 2 light SPGs, so there are some nice uses for old equipment.

You would probably lose production on that trade, if you keep old light tanks in production in order to convert them.

But that said it is up to the developers to say how they want the production system to be and I won't suggest anything about Changes, in fact it is probably fine as it is right now and there are probably much more critical areas to focus on.

So... On topic...
Should there be a bigger malus for switching to a newer model, and a smaller one for upgrading a existing one?

Probably the production system should be left alone as it is likely a waste of resources to change it when there is more important stuff to work on.​

 

[mursolini]

Well it matter alot if you can produce more 1944 Equipment than the enemy if the game manage to somehow reach that Point, a late game military factory would be much better than a early game military factory producing early game Equipment in absolute terms if the game reach that Point. So if one country go for an early rush but fail it would most likely lose the war because it would not keep up in the late game. However Tech rushing cause quite alot of problem here.

If you had a factory earlier, and it already produced previous tier of equipment, it is easier to switch it to new models, because you have that backlog. However if you got your MIC in say 1942, it becomes a far more interesting question, should you just let it churn out 1940 fighter or switch it to 1944 fighter in 1944, because you don`t have that long backlog of previous production.

Because it is in the game pretty cheap to switch out lines, if your factories was "damaged" when you switched so you had to pay civilian industry I think people may reconsider. The damage would represent producing the machine Tools needed for the production line.

Well, you will never reconsider, unless previous tech equipment is actually somewhat competitive. There are already countries that can convert CIC to MIC easier, so harsher penalty might reduce the amount of early war factories to more CIC and mobilization, but you will not keep producing 1.5 to 2-3 times inferior equipment, it is just digging yourself into a hole.

The Tech system encourage an aggressive strategy of using what you have Before it become obsolete, which is a good thing in my opinion. The issue is you can later year Tech for a rather low cost with national focuses.

It encourages risky play.

Well the early ships don't become useless but their value drops alot over time, on the other hand they have the most dockyards so they tend to have more better ships as well.​

It is far easier to catch up, if their decades of naval building drops in value by a lot, and you only need to outmatch their current factory numbers.

Probably the production system should be left alone as it is likely a waste of resources to change it when there is more important stuff to work on.​

I kinda disagree. If they work on Barbarossa, production system and techs are probably the #1 thing they need to get right. Logistic is second.

 

[Denkt]

If you had a factory earlier, and it already produced previous tier of equipment, it is easier to switch it to new models, because you have that backlog. However if you got your MIC in say 1942, it becomes a far more interesting question, should you just let it churn out 1940 fighter or switch it to 1944 fighter in 1944, because you don`t have that long backlog of previous production.

You should Always switch to 1944 fighter right now, the small cost in production efficiency for a much better Aircrafts is a very good trade, especially in the long run.

Well, you will never reconsider, unless previous tech equipment is actually somewhat competitive. There are already countries that can convert CIC to MIC easier, so harsher penalty might reduce the amount of early war factories to more CIC and mobilization, but you will not keep producing 1.5 to 2-3 times inferior equipment, it is just digging yourself into a hole.

Yes but it make sense since most realistic production number is achieved if you encourage mostly civilan factories in the 30s, including Germany.

I kinda disagree. If they work on Barbarossa, production system and techs are probably the #1 thing they need to get right. Logistic is second.

Logistics is more important than the production system in my opinion.

 

[valisk]

Okay, as I quite like @Denkt testing posts, and realize hard data is often more reliable than my own gut instinct I decided to shut my flapping little mouth, stop filling the forums with ill considered garbage for a couple of hours and do some actual testing.

I'll try to turn it into a spreadsheet a little later, if anybody really wants the data.

To test the impact of variant production to my own satisfaction I did the following:

• I ran two games as the UK, both starting in 1936, with the AI turned off.
• I increased the number of MILs to 20 and used all on one production line.
• I cleared all production lines, and on the 1st of each year Where a new Industrial tech was available, I implemented it.
• I chose light tanks to be my testbed piece of equipment
• Every 6* months I either swapped production if a new light tank was available that year, or I created a variant and switched to it **
• I made no changes to the starting economic laws and made no different choices in the few popups that appeared
• For 5* months I produced Vickers 2, then swapped to Matilda, and then swapped production every January 1st or June 1st* I did not use PP to increase output at any time
• Where a new industrial tech was researched I waited a day for it to take effect before switching production so that it's bonuses would apply.
• I intend to, but have not yet, run a control producing only new models and not any variants.

On 1st Jan '38 production looked like:

• CI - 10.7 units per day @ 66.24% eff, total light tanks: 4654
• DI - 10.31 units per day @ 68.02% eff, total light tanks: 4633

On 1st Jan '39

• CI - 11.33 units per day @70.08% efficiency, total light tanks: 8620
• DI - 10.63 units per day @70.08% efficiency, total light tanks: 8422

On 1st Jan '41, the last day of production for Matilda variants before switching to Valentine

• CI - 13.75 units per day @ 80.08% eff, total light tanks: 18396
• DI - 12.55 units per day @ 80.08% eff, total light tanks: 17535

Switching to Valentine of crashed the CI line to 24.02% starting efficiency, the DI line started back up with 46.50% efficiency

On 1st Jan '42

• CI - 12.93 units per day @ 77.05% eff, total light tanks: 21755
• DI - 13.02 units per day @ 86.83% eff, total light tanks: 21279

On 1st Jan '43

• CI - 15.13 units per day @ 90.09% eff, total light tanks: 26806
• DI - 13.51 units per day @ 90.09% eff, total light tanks: 26131

First thing that stands out is that until 1939 tech arrives the difference is minimal but noticeable if you have large amounts of MIL on one thing, beyond that even at the same efficiency level Concentrated just produces more units per day. That's not entirely unexpected.

Beyond '39 I would have expected the lower amount of efficiency lost by DI to have a greater impact, and it does but only in the short term, as the efficiency of the Concentrated line rebounds faster than that of the dispersed line because of diminishing returns as they both get closer to their cap combined with the higher overall production of CI. The CI production line in 41 took a big blow when switching to Valentine, and despite new variants every 5th and 7th month, it had mostly caught up within one year.

The end results confirm and deny my original thinking in interesting ways.

They say Valisk you are wrong about variants, the CI line can handle them as long as you are not a crazy monkey spamming variants every time you get 25xp spare, and will probably out perform DI if you batch the variants every 6+ months, and more to the point I think it looks very reasonable and balanced to do that.

They also say Valisk you are right about the heavy crashes in CI efficiency after switching pinching production, don't be a simpleton and switch to that new rifle/tank/plane if you are in the middle of a severe crisis with your equipment as you did with the USSR thinking the extra stats would be very helpful, that is dangerously stupid.

And they also validated my not entirely unobvious instinct as the USSR to build my CI factories way up North, far out of the range of bombers.

I've not controlled in any way for bombing damage, which is obviously a consideration, but in the immediate aftermath of proudly producing 52,000 tanks in my factories I can say that my first thoughts are (and I appreciate that this may already be obvious to most of you) that any one planning to play as an Arsenal of Democracy(tm) in MP should probably divvy up the production responsibilities so that they are hammering out items with no variants and for those with long runs but potential variants then make minimal changes as infrequently as possible, and either skip every other model (medium tanks looking at you) or have the countries with DI systems produce those particular types of equipment.

* I decided to split the 12 months into a 5 month and 7 month period to see if Dispersed would catch up production in the short term, and lose it vs a longer run for Concentrated in the long term, I did not at all make a mistake during the 1st year and decide to just double down on it because I thought about it for a moment when about to restart the experiment and discovered it might be useful, no sireee not me!

** I did test if adding 1 or 5 points to the variant increased the drop in production efficiency and it doesn't which is also useful information.

 

[Denkt]

There are also more stuff to think about, concentrated bonus become more significant if you have a high conscription law and suffer from production penalty while dispersed become better if you have high stability or free trade which give production output.

Also you maybe employ a strategy such as mass produing infantry Equipment in 36 to switch over to stuff like tanks and Aircrafts later which I don't know which industry type would be best suited at.

 

[balmung60]

One cause of the issue is that infantry soft attack grows enormously with tech, while in the real world the difference was really minor.

Tanks wasnt really the case, I think it is fine now for tanks.

Really? It seems a bold argument that the difference in individual firepower between a Lee-Enfield and an M2 Carbine or Stg44 was "minor". Ubiquitous semi-automatic or selective fire weapons put far more firepower in the hands of troops than bolt-actions, and intermediate cartridges like .30 Carbine and 7.92x33mm Kurz meant they could be used out to most combat ranges, rather than just close quarters like submachine guns. While machine guns were more mature, there were evolutions there, too, and had the war gone longer, we'd likely have seen older weapons like the BAR phased out by newer weapons like the WAR.

 

[Denkt]

Infantry don't provide much firepower in the first place, base value if 6 in 1936 and 12 in 1942. At the same time artillery go from 25 in 1936 to 34 in 1942 and that don't consider that artillery also get +40% modifier to its soft attack with all Techs researched while infantry 15% or 20% if motorised/mechanized.

So in absolute terms infantry don't provide much firepower, even based on combat width so you field 3 infantry or 18-36 base soft attack vs 2 artillery between 50 to 68 base soft attack and these are base numbers and generally twice soft attack have more than twice the value because how the hit system works, at minimum twice soft attack is twice as good but it can be alot better.

Hit chance of an attack is 10% or 40% if the attack division don't have a defence Point to spend on reducing the hit chance to 10%.

 

[Secret Master]

They say Valisk you are wrong about variants, the CI line can handle them as long as you are not a crazy monkey spamming variants every time you get 25xp spare, and will probably out perform DI if you batch the variants every 6+ months, and more to the point I think it looks very reasonable and balanced to do that.

To add to this point:

You should be saving XP to do big upgrades anyway.

So, instead of putting 75 XP on a plane every 3 months, put 400 XP on a plane every 6-8 months.

And if you are swimming in XP, maybe apply it to vehicles on different production lines.

In January, put 400 XP on fighters. In May, put 200 XP on CAS.