Land based Naval Air Useless?

[Axe99]

My bad. It was late, I was tired and I guess I skimmed right over that.

No worries at all . I miss things all the time. In terms of the 'nerf', having smaller regions near the coast wouldn't nerf the long-range craft (they'd still have reduced efficiency because of the size of the zone) - it'd just make short-range aircraft and aircraft near the coast more effective. I'm not saying it's an idea without issues (creating 'doughnut' coastal air zones around islands, for example, could create potential oddness in the system), just something that would better reflect the issues of the day (even though the VLR aircraft could make the difference, the distances were so long that time on station was an issue, which is something efficiency in the current system reflects well - ie, to overcome the issues of lower efficiency, get more aircraft so that they can take turns over the distant airzones).

Apologies if I've missed something somewhere - I'm a little tenuous as to where the thread had got to.

 

[Dalwin]

No worries at all . I miss things all the time. In terms of the 'nerf', having smaller regions near the coast wouldn't nerf the long-range craft (they'd still have reduced efficiency because of the size of the zone) - it'd just make short-range aircraft and aircraft near the coast more effective. I'm not saying it's an idea without issues (creating 'doughnut' coastal air zones around islands, for example, could create potential oddness in the system), just something that would better reflect the issues of the day (even though the VLR aircraft could make the difference, the distances were so long that time on station was an issue, which is something efficiency in the current system reflects well - ie, to overcome the issues of lower efficiency, get more aircraft so that they can take turns over the distant airzones).

Apologies if I've missed something somewhere - I'm a little tenuous as to where the thread had got to.

I disagree with this concept. Yes the long range craft would still be able to perform their long range tasks, but with this concept of close in donuts the long range planes would not be able to assist with close in defense without giving up the long range tasks. They cannot be in two places at once.

I think the current system is better, which is not to say that I don't think it needs some changes, just not this proposed change.

 

[TheSpaghettiLife]

I had a naval battle off my coast recently where some land based air participated. I had a 200 ground support wing and 50 naval air wing. Of that total I got 100 ground support and only 7 naval air show up to the fight.

Worse yet they only showed up ONCE in a week long battle.

Am I wrong in thinking it was a complete waste of research time and resources to get land based naval air.?

You can use the land-based naval bombers in large volumes, that's thier main benefit. They also have higher base stats than the carrier conversions (as do all the other land-based aircraft), which allows them to fly farther and effectively cover areas where ships/targets are located. As the Soviet Union, I once destroyed around 40 Japanese ships just with one air wing of 1000 level 2 naval bombers off of the Japanese coast.

 

[C-Breeze]

I'm not saying it's an idea without issues (creating 'doughnut' coastal air zones around islands, for example, could create potential oddness in the system), just something that would better reflect the issues of the day...

I can see it from the point of view of being able to recreate certain 'historic' conditions, such as having an effective (or reasonably dense) CAP around an island stronghold. Part of me finds this appealing as it would seem to represent an added dimension to the game that doesn't exist at the moment. On the other hand, I'm uncertain as to whether such a system would prove practical.

If for the sake of argument, we did have a 'doughnut' coastal zone around an island, what would the ramifications be for the player? Could the player maintain a bid for air superiority in both zones simultaneously, (the immediate and the extended), or would he have to give up fighter cover for say active trade interdiction, (extended zone), in favor of the more localized, (immediate zone), amphibious/air defense?

In other words, the fighter strength that used to cover the entire zone is now potentially being split to cover two zones. Times this by however many provinces you control with 'doughnut' zones either surrounding or adjacent to them, (such as the Iberian coast example given by Gwydion5 in post 22), and you may run into a need for a great many more fighters or simply have weak coverage in many more zones.

While I find this discussion interesting, and think it's good food for thought, I'm not sure this wouldn't create more problems than it solves.

 

[Axe99]

I can see it from the point of view of being able to recreate certain 'historic' conditions, such as having an effective (or reasonably dense) CAP around an island stronghold. Part of me finds this appealing as it would seem to represent an added dimension to the game that doesn't exist at the moment. On the other hand, I'm uncertain as to whether such a system would prove practical.

If for the sake of argument, we did have a 'doughnut' coastal zone around an island, what would the ramifications be for the player? Could the player maintain a bid for air superiority in both zones simultaneously, (the immediate and the extended), or would he have to give up fighter cover for say active trade interdiction, (extended zone), in favor of the more localized, (immediate zone), amphibious/air defense?

In other words, the fighter strength that used to cover the entire zone is now potentially being split to cover two zones. Times this by however many provinces you control with 'doughnut' zones either surrounding or adjacent to them, (such as the Iberian coast example given by Gwydion5 in post 22), and you may run into a need for a great many more fighters or simply have weak coverage in many more zones.

While I find this discussion interesting, and think it's good food for thought, I'm not sure this wouldn't create more problems than it solves.

Aye, there could definitely be issues, particularly with micromanagement as aircraft change missions between zone. While it would make operations more historically plausible, on reflection (and enlightened by your posts ) I suspect it might make things a bit fiddly.

 

[Gwydion5]

While I find this discussion interesting, and think it's good food for thought, I'm not sure this wouldn't create more problems than it solves.

I agree that it could be potentially problematic or a net loss to the experience. But I think we need to realize and understand that we already deal with the drawbacks / counter arguments being expressed. For example the North Sea and Eastern North Sea are split up. If the argument is that large zones don't matter and smaller more practical zones would detract from the experience, PDS should combine them. But I have a feeling that change probably wouldn't be well received by the players since the effectiveness of air power interdicting naval power would likely drop significantly. Would it break the game combining those zones? No, but I think given a choice we would agree with having 2 different zones rather than one as being better. My point being is that there are multiple zones in the game where things were split for logical and practical reasons. The North Sea is one, Germany having 3 different air zones is another, and yet nobody is really complaining to have those zones combined. So I'm not sure making other large zones more practical would cause problems, or create community/player outcry.

At this point though in the conversation, the only way forward is with some objective data about the value of mission efficiency and how large zones are impacting that. This being a casual discussion, I think that may be a bit too much work just to be right about large zones being impractical and historically inaccurate to the simulation / representation of air power. I also have a nagging suspicion that at the heart of this discussion is the hope that PDS eventually redesigns the air power system, and thus keeping it from being improved hastens its end. Which I wouldn't mind either to be perfectly honest as I don't like how airbases on the edge of a zone have poor coverage, while air zones in the middle have great coverage. Or some zones just have really bad air base placements that you can do absolutely nothing about, etc...

 

[Gwydion5]

Also going back to topic at hand. Is it a bug that air missions only get 1 attack in naval battles or was it a design change to reduce the effectiveness of air power in naval battles? I remember release HOI4 had brutal air power effectiveness against naval, but not sure if this was a change because of that or not.

EDIT: And out of curiousity, if you could assign up to 3 zones of coverage for a mission like you can with naval, would people still be against further definition of the zones like what I'm proposing?

 

[Axe99]

Also going back to topic at hand. Is it a bug that air missions only get 1 attack in naval battles or was it a design change to reduce the effectiveness of air power in naval battles? I remember release HOI4 had brutal air power effectiveness against naval, but not sure if this was a change because of that or not.

EDIT: And out of curiousity, if you could assign up to 3 zones of coverage for a mission like you can with naval, would people still be against further definition of the zones like what I'm proposing?

The naval battle thing isn't always 1 attack - sometimes they come back as often as they 'should' - so I assume one or the other behaviour is a bug at least . My guess would be the 1 attack thing - it improved a lot. It means in my games at least that most of the time, naval air isn't too dangerous, but sometimes it smacks fleets down hard!

I like your idea of combining smaller zones with a multi-zone approach like naval .

 

[Gwydion5]

Just an FYI I did do about 5 test runs today to see what Mission Efficiency is all about. What I can say definitively is that Mission Efficiency = % of the wing that enters combat. IE If you have a wing of 100 planes with a mission efficiency of 10% only (at most) 10 planes enter combat.

Attaching the save game if anyone wants to test. When you load this game go in to the console type "AI" to turn the AI off so you can test. The entire UK fleet excluding subs has been divided in to 2 fleets with patrol missions, 1 is in eastern north sea zone, and the other is in the Baltic Sea. I have 5x100 naval bomber operating in each sea zone, one with 100% efficiency in the north sea (It's Reliability is higher than the other group so "losses" are skewed due to accidents) the other with exactly 50% mission efficiency in the Baltic.

I'll compile the results later and post them here.

 

[C-Breeze]

For example the North Sea and Eastern North Sea are split up. If the argument is that large zones don't matter and smaller more practical zones would detract from the experience, PDS should combine them.

Can't speak for anyone else, but from my point of view, the argument is far from 'large zones don't matter.' If anything, they matter too much. For example, in one of my early play-throughs, I made a point to 'acquire' the Azores in the hope they would prove not only an excellent forward naval base, but also a great platform for aerial convoy raiding. Imagine my disappointment when I discovered the air efficiency of my tier 3 aircraft was only around 17% for the entire zone. (Don't recall the exact percentage, but it was sufficiently low to make me feel like I'd wasted the effort to establish a base there). I was so disheartened, that I've never gone back. So yeah, not a fan of large unwieldy zones.

My point being is that there are multiple zones in the game where things were split for logical and practical reasons. The North Sea is one...

As for the North Sea, I think we agree on the existing split. Your initial example (the Iberian coast) only has Portugal/Spain bordering it, whereas the North Sea has a potpourri of countries, so it makes more sense from a geographical viewpoint to split the latter but not the former. Else at the outbreak of war, the naval and/or air assets of multiple countries would all be thrown into the same behemoth zone, (a virtual blender if you will), where even the Kriegsmarine would have virtually no opportunity to establish itself without being immediately engulfed by the Royal Navy. So splitting the North Sea into two zones serves the practical purpose of creating a buffer between multiple nations whereas the Iberian coast doesn't require the same buffer. In other words, there's no 'one size fits all' formula that can or should be applied to every sea zone.

I also have a nagging suspicion that at the heart of this discussion is the hope that PDS eventually redesigns the air power system, and thus keeping it from being improved hastens its end.

Lots of talk that an overhaul of the air system is potentially a part of the v1.4 patch. If this proves true, I think we can all agree that it would be most welcome.

Which I wouldn't mind either to be perfectly honest as I don't like how airbases on the edge of a zone have poor coverage, while air zones in the middle have great coverage. Or some zones just have really bad air base placements that you can do absolutely nothing about, etc...

I wonder if they would consider adjusting the code so that range distances are calculated from the median (center) of the land zone rather than the outer edges which they are often situated in? I realize you're not always able to build an airbase near the center of a zone, (given mountains/bad terrain or just lack of roads and infrastructure), but having them purposely based near the edges seems an unnecessary nuisance for the player.

 

[Gwydion5]

Can't speak for anyone else, but from my point of view, the argument is far from 'large zones don't matter.'

I was just generalizing the arguments against more reasonable and practical air/sea zones. IF PDS is not going to revamp the air system, I do not see why we shouldn't refine the zones to address various issues or capture historical elements. An argument against that is essentially saying that the current negatives do not outweigh the potential upside / downside of the proposal. That's all I was trying to say with that short hand generalization.

Your initial example (the Iberian coast) only has Portugal/Spain bordering it, whereas the North Sea has a potpourri of countries, so it makes more sense from a geographical viewpoint to split the latter but not the former

My point about the Iberian Coast was being able to defend the coastal areas more efficiently/effectively by giving it a coastal zone. Strategically a lot of trade is going to go through these zones and I feel there is an objective difference in doing CAP for a coast or island vs interdiction in the deep blue sea. I also feel that Axis Spain or Portugal would force convoys coming through here out to the deep blue sea. I understand that is not possible right now regarding convoy routes, but I believe eventually this will change, and when that happens these large zones are still going to be poor use of air power due to coverage. Again, all based on the assumption/position of IF PDS does not revamp the system.

In other words, there's no 'one size fits all' formula that can or should be applied to every sea zone.

Actually I disagree with that. From a design perspective, zones should always be given some consideration to the lowest and highest possible coverage with/without doctrines and tech. If you do that, then everything else will work itself out or be more balanced since zones are being split up and divided which increases their (potential) usability in a game. It opens up more options as I described with Eastern Micronesia where there could be adjacent zones next to the coastal areas that have say 50% coverage in 1940, and 80% coverage by 1944. Maybe part of the problem is that I'm not fully explaining myself. I give examples, but never paint the complete picture because I expect you guys to build the same framework and end goal that I am. So here is another picture.

Now it doesn't HAVE to be these numbers, I'm just giving an example of how a framework / baseline on the design and implementation of air/sea zones would work using my Iberian Coast and your Azores as an example. But let's say the above numbers represent stock 1940 naval bomber coverage from Azores and Western Spain/Portugal. If Western Spain / Azores is Axis, you give the player the ability to actually use these places strategically. You could put decent air power here and try to create some risk/threat to allied shipping. This opens up possibilities for players, because if you continue to balance the zones around some framework like the above, what do you think the rest of the map will look like? Does Brazil's strategic value go up for a player looking to close off the rest of the Atlantic here? What about the Carribean Islands? Things start to change quite a bit when you start dividing things up in to practical zones / coverage.

Just consider what enforceable zones like the above means to convoys and ships. Think about what a player could do with that, versus what they can do now. I just don't see the downside or feel it's greater than current negatives.

Lots of talk that an overhaul of the air system is potentially a part of the v1.4 patch. If this proves true, I think we can all agree that it would be most welcome.

That will be an interesting Dev Diary to read if that happens.

I wonder if they would consider adjusting the code so that range distances are calculated from the median (center) of the land zone rather than the outer edges which they are often situated in? I realize you're not always able to build an airbase near the center of a zone, (given mountains/bad terrain or just lack of roads and infrastructure), but having them purposely based near the edges seems an unnecessary nuisance for the player.

Land based zones are going to be tricky because of proximity due to states and thus some sort of balance / compromise of airbases sharing zones will have to be made. It sucks but the best you can hope for are adjacent air zones giving more purpose to airbases in a shared air zone that are on the edge.

 

[Gwydion5]

So for whatever reason, 4 of the 5 test runs I did yesterday produce the same results over a 3 month period. Exactly the same results in fact. So the seed / RNG system looks to have a bug as that should not happen. With that in mind I will need to do more testing and see if I can't identify the issue and come up with a work around.

 

[Gwydion5]

So I created a bug report due to the lack of variability in the results of running multiple tests.

https://forum.paradoxplaza.com/foru...tion-issues-with-loading-saved-games.1011854/

There is the spreadsheet / data I was compiling. But because of this assumed bug we only have 3 data samples instead of the intended 5 or 10 that I would like.

 

[AegonTheUnready]

Just an FYI I did do about 5 test runs today to see what Mission Efficiency is all about. What I can say definitively is that Mission Efficiency = % of the wing that enters combat. IE If you have a wing of 100 planes with a mission efficiency of 10% only (at most) 10 planes enter combat.

Hmm, in my recent experience that isn't entirely accurate.

Playing as Brazil, I had set in the northeast corner of Brazil (Cape Verde sea zone) a NAV interdiction force. 1000 fighter II, 600 naval II. Radar II, and Decrypt II/Encrypt I b/c these are supposed to positively affect naval combat.

Over the course of 1 game year I intercepted many Allied fleets - maybe several dozen times. Allied subs were almost always attacked by only a handful of planes with predictable abysmal results. Allied surface ships usually attracted a few dozen, but never more than a bit over 100. Results amounted to a few percentage points in these cases.

There was one exception to the above - Allied CV. On four different occassions Allied CV wandered into Cape Verde (unescorted!). Each time they were attacked by hundreds of my NAV - 599 in one attack - and sent to the bottom.

Mission efficieny is part of the equation, but not the whole of it. There is something wonky in the land naval air combat. Just my opinion of course.

 

[Gwydion5]

Mission efficieny is part of the equation, but not the whole of it. There is something wonky in the land naval air combat. Just my opinion of course.

Thanks for the reply. There are probably more variables to be considered, hell some of those variables may be completely hidden. Based on your post I decided to double check some of the saved game results I have of my tests, and sometimes the full wing is engaged in the Baltic Sea with only 50% mission efficiency. Here is the breakdown.

Test Game 1 E-North Sea (100% Mission Efficiency)
3 engagements with 100 naval bombers
2 engagements with 70 naval bombers (Poor weather reduces mission efficiency in this case it is like that a storm for -30% mission efficiency was responsible)

Test Game 1 Baltic Sea (50% Mission Efficiency)
5 engagements with 50 naval bombers
2 engagements with 45 naval bombers
1 engagement with 99 naval bombers

Test Game 2 E-North Sea (100% Mission Efficiency)
6 engagements with 100 naval bombers

Test Game 2 Baltic Sea (50% Mission Efficiency)
4 engagements with 50 naval bombers
1 engagement with 45 naval bombers

Test Game 6 E-North Sea (100% Mission Efficiency)
4 engagements with 100 naval bombers
1 engagement with 99 naval bombers

Test Game 6 Baltic Sea (50% Mission Efficiency)
2 engagements with 99 naval bombers
4 engagements with 50 naval bombers
2 engagements with 35 naval bombers

There might be a bug or some RNG critical success that allows more than the average 50 or less to participate in the Baltic Sea. You are more than welcome to download the saves and do your own testing, or double check my results. But I feel pretty confident in my conclusion that mission efficiency is primarily responsible for the number of aircraft that can engage in a combat in the scenario I created which has 2 rather large British fleets.

 

[Dan1109]

@Gwydion5

For pacific you need to use TAC´s. Tac´s participate in naval combats as long they have the ground support button enabled.
Of course they need 2 ships fighting each other to join.

There is a reason in history why both USA and JAP used more heavy fighters and bombers in those regions instead of using the more light ones.

A 1940 TAC if i remember right have a range of 1700 to 2500km Well above the range of a NAV 1944. (can´t acess the game right now, the wiki is very wrong on ranges of air units unfortunally).

Also some regions are just not fit to have a good amount of air superiority, at least unless you heavily invest in heavy fighters and tac´s.

Historically, it was STRATs that were used. B-17s and then the B-24 was a huge maritime hit with the navy, so much the navy developed a specific anti-naval/sub version of it.

 

[Gwydion5]

Ran two more tests (Test 12 and 13) this morning. Fortunately, (or unfortunately depending on how you look at it) I did not shut down my computer, rather I put it in sleep mode.

Does anyone else find it a little concerning that Air Wings on Naval Strike missions against Fleets on Patrol missions are producing the exact same results over a three month period?

EDIT: So I ran 2 more tests from a complete shutdown and produced the same results. I then deleted naval strike mission in the E-North Sea zone and put it back in and that reseeded the results for both air missions.

It's really odd behavior when you think about it. Because clearly the game is differentiating between 1st load of a save game, and current session reload of that same game. Oh well. At least now I can produce different results. Yay! I also probably should've figured this out sooner, I guess we all have our slow days.

 

[Dalwin]

Ran two more tests (Test 12 and 13) this morning. Fortunately, (or unfortunately depending on how you look at it) I did not shut down my computer, rather I put it in sleep mode.

View attachment 255543

Does anyone else find it a little concerning that Air Wings on Naval Strike missions against Fleets on Patrol missions are producing the exact same results over a three month period?

EDIT: So I ran 2 more tests from a complete shutdown and produced the same results. I then deleted naval strike mission in the E-North Sea zone and put it back in and that reseeded the results for both air missions.

It's really odd behavior when you think about it. Because clearly the game is differentiating between 1st load of a save game, and current session reload of that same game. Oh well. At least now I can produce different results. Yay! I also probably should've figured this out sooner, I guess we all have our slow days.

It has already been acknowledged and I thought was common knowledge that the RNG seed is not itself entirely random. This is often intentional to avoid save scumming in games but also is more efficient from a programming perspective than a truly random generator would be.

The best random seed to my knowledge is one that is based on a capture from the system clock itself (since down to the microsecond this is a very random number and essentially impossible to get to repeat itself consistently). The tradeoff usually comes if you try to apply this method more than once at the beginning of a session. I also know this method is not at all in common usage.

The consequence from a HOI perspective is that if you repeat the same situation from a save multiple times you should almost always get the same result. The likeliest reason to not get the same result is to have done some step in a slightly different order or with different timing. If your test scenario allows it, try letting a variable number of days elapse before the firing of each trial. Even one or two days should have enough other events requiring randomization that your subsequent trials will not just use the same series of random numbers.

 

[Gwydion5]

It has already been acknowledged and I thought was common knowledge that the RNG seed is not itself entirely random. This is often intentional to avoid save scumming in games but also is more efficient from a programming perspective than a truly random generator would be.

Great. Now I have to do more forum searching to see the history / observations about seeding. On the scrumming part, if that is the intent it does a pretty bad job at it as you can change the seed for all air missions by simply changing one air mission. In addition the behavior demonstrated shows a 1st session load being consistent with other 1st session loads (other than test1 which was the original new game start.) and current session loads being consistent with other current session reloads. So I'm thinking something is not working as it should regarding the seeding. It might also be a bug that gives SteelVolt some insight to other behavior when loading from a save game.

Now with that said, it's really hard to say how big of a deal this might be, perhaps any change to any air mission by any nation will create a new seed for all other air missions. So this really only applies because I'm loading the game and shutting the AI off right away to do my tests. Plus now that I know how to work around it, it's not a big deal. I may play around with it some more just to explore what affects the seed(s). Perhaps tagging to a different nation causes a new seed. Perhaps changing a naval mission causes a new seed for air missions too. etc...

The best random seed to my knowledge is one that is based on a capture from the system clock itself

Which is what I was expecting and kind of shocked it didn't happen. I also haven't seen this before in other tests, but the only test where I did multiple passes of the same configuration was the fighters vs cas test. Hmmm I might have to go look at that again and see why that might be different. My (old) naval bombers vs ship AA test was just 1 pass with a different configuration of the wings each time.

The consequence from a HOI perspective is that if you repeat the same situation from a save multiple times you should almost always get the same result.

I always assumed there was a RNG check for detection/encounters for the zones. But it appears that information is dependent on the seed and is not independently generated per day when the wing(s) are active in the zone.

 

[Dalwin]

Perhaps I was not sufficiently clear. The save scumming comment was meant for games other than HOI and was a comment on seeding methods in general.

To the best of my knowledge the seed in HOI is generated only once at the beginning of the first session of a particular game. This is why repeatedly running a test from a save will tend to create repetitive results.

This topic has come up a few times previously in other threads though I cannot recall enough specifics to point you to a particular thread or a particular dev to stalk for a stray comment. (I am always quick to point a finger in SteelVolt's direction though I think in this particular instance it is the bailiwick of one of the feature programmers instead of the AI guy).

One such thread involved an aspect of naval combat and how many ships were lost on each side. The poster had run a series of tests (10 IIRC) from a save taken at the beginning of a substantial naval engagement and gotten identical results on each trial. He incorrectly assumed there was no real randomization instead of seeing the subtleties of RNG seeding methods. This is also the thread for which I vaguely recall an official comment. This was at least 3-4 months ago and possibly more than 6. Sorry for the vagueness.

My comment on an ideal method was to have the game reseed itself from the clock at least once at the beginning of each session which allows resumes from a save for testing purposes to have more variable results.

(EDIT: rereading your chart again, perhaps what it tells us is this. The RNG seed is determined once at the start of a game session, i.e. loading into HOI after it being shut down. That string of preset but unpredictable numbers (technically not random at all) is continued rather than reset if you don't close out the game. Continuing from the middle of the string, instead of reusing the same portion of it, is as good as getting a new string.

Even this does not seem entirely consistent with your results unless I am misreading part of your chart. Are you saying that if you close the game completely and then reopen the save from scratch that you do or do not get the identical results as on the previous iteration of the test?)