Slitherine

quote:

ORIGINAL: LoBaron

quote:

ORIGINAL: witpqs
I posted my observation a few times a number of months ago. It is my strong suspicion that performance is reasonable up to roughly pilot skill level 70. Then, as pilots gain skill above 70 the attack performance is simply too high. This applies to:

(Including Low in all variants listed below)
Naval attack (including torpedo)
ASW
Ground attack
Airfield attack
Maybe Port attack, but I have seen fewer so it is difficult to judge
Regarding the various strategic attacks (city, refinery, etc.), I have seen too few to judge.

It would be a mistake to only try to "nerf" the various ground attacks. The biggest issue seems to be to reduce performance benefit for skill levels above 70, making it more of diminishing returns.

Exactly.
I remember your posts on this topic, and it is very similar to what I mentioned a couple of times already.

When skills (and aggression or exp) values surpass a certain threshold the results get off the scale.
I do not know the exact formulas used, but I assume that a skill closing to 100 means a success roll in nearly every event,
something not in accordance with reality.

The only solution which makes sense is nerf skill gain,

quote:

or find a way to change how the game engine uses skill variables in dice rolls.

quote:


The second suggestion is, for everything I know about what can be done without tinkering with the code itself, impossible.

The first would require rolling back skill gain mechanics to the early patches of WitP AE. If some care to remember:
In the beginnings of AE there was an outcry to increase training speed, a suggestion which was finally implemented by the devs (and which
I always opposed).

Nerfing ground bombing would yield adverse results, I believe this can only be suggested if the impact is not thought through end to end.

quote:

ORIGINAL: herwin

The sea bombardments are also nuclear.

quote:

ORIGINAL: LoBaron

quote:

ORIGINAL: witpqs
I posted my observation a few times a number of months ago. It is my strong suspicion that performance is reasonable up to roughly pilot skill level 70. Then, as pilots gain skill above 70 the attack performance is simply too high. This applies to:

(Including Low in all variants listed below)
Naval attack (including torpedo)
ASW
Ground attack
Airfield attack
Maybe Port attack, but I have seen fewer so it is difficult to judge
Regarding the various strategic attacks (city, refinery, etc.), I have seen too few to judge.

It would be a mistake to only try to "nerf" the various ground attacks. The biggest issue seems to be to reduce performance benefit for skill levels above 70, making it more of diminishing returns.

Exactly.
I remember your posts on this topic, and it is very similar to what I mentioned a couple of times already.

When skills (and aggression or exp) values surpass a certain threshold the results get off the scale.
I do not know the exact formulas used, but I assume that a skill closing to 100 means a success roll in nearly every event,
something not in accordance with reality.

The only solution which makes sense is nerf skill gain, or find a way to change how the game engine uses skill variables in dice rolls.
The second suggestion is, for everything I know about what can be done without tinkering with the code itself, impossible.

The first would require rolling back skill gain mechanics to the early patches of WitP AE. If some care to remember:
In the beginnings of AE there was an outcry to increase training speed, a suggestion which was finally implemented by the devs (and which
I always opposed).

Nerfing ground bombing would yield adverse results, I believe this can only be suggested if the impact is not thought through end to end.

quote:

ORIGINAL: witpqs

quote:

ORIGINAL: herwin

The sea bombardments are also nuclear.

Harry, see PzB's post and my reply to him. He sees only wimpy sea bombardments, you see only nuclear sea bombardments, and I am seeing all three varieties (Momma Bear, Pappa Bear, and Baby Bear) - so I think sea bombardments are in the Goldilocks Zone.

quote:

ORIGINAL: witpqs

quote:

The only solution which makes sense is nerf skill gain,

This would be the wrong solution because no matter how long it takes there will still be pilots with "super powers".

quote:

quote:

or find a way to change how the game engine uses skill variables in dice rolls.

The programmers made the pilot skills have an effect (remember there were no pilots skills in WITP vanilla, they were added for AE), so they do have access to adjust the effect those skills have. I'm not suggesting it's easy, and they might not want to tackle it. But if this issue is addressed that's really the only way IMO because the other ways will create imbalances by fixing some of the issues and leaving others in place.

quote:

ORIGINAL: LoBaron

quote:

ORIGINAL: witpqs

quote:

The only solution which makes sense is nerf skill gain,

This would be the wrong solution because no matter how long it takes there will still be pilots with "super powers".

Maybe, but its probably the easiest one. Personally I prefer a solution that still is possible over a solution that is optimal, in
case the latter cannont be realised.

As far as I know Michael is (and how long still?) the only guy left working on the game (besides DaBabes). If the patch process resembles what I do at work
at least to some extent, there is access to the old pilot training model (the one with the much reduced speed) and this part can be rollbacked.

Maybe this could be flavoured with a lower cap for training (e.g. from 70 to 60, or 50).

quote:

quote:

or find a way to change how the game engine uses skill variables in dice rolls.

The programmers made the pilot skills have an effect (remember there were no pilots skills in WITP vanilla, they were added for AE), so they do have access to adjust the effect those skills have. I'm not suggesting it's easy, and they might not want to tackle it. But if this issue is addressed that's really the only way IMO because the other ways will create imbalances by fixing some of the issues and leaving others in place.

Agree, but as I said already, I´d rather have a suboptimal solution than dreaming about one which will never happen.

Edit: I´d be careful when assuming that because originally there was just one variable, and now, depending on the situation, there is one for each specific situation,
this means the team had access to the coding part of the games engine which governs the specific diec rolls.
It only means there are more pools now where the numbers are drawn from, not that anybody has access to how they are used.


That said, I doubt it will happen at all, and WitP AE is already far better than any other game of this genre, TBH I can live with the smaller glitches of the games´ engine.

quote:

ORIGINAL: LoBaron

Edit: I´d be careful when assuming that because originally there was just one variable, and now, depending on the situation, there is one for each specific situation,
this means the team had access to the coding part of the games engine which governs the specific diec rolls.
It only means there are more pools now where the numbers are drawn from, not that anybody has access to how they are used.

quote:

ORIGINAL: obvert

Why not for heavily fortified positions allow the 4E a 4,000 pound bomb to make this seem more realistic? It would also be modelled more correctly as when it hit it would do a LOT of damage, but because it's one bomb it wouldn't always hit.

quote:

ORIGINAL: PaxMondo

quote:

ORIGINAL: obvert

Why not for heavily fortified positions allow the 4E a 4,000 pound bomb to make this seem more realistic? It would also be modelled more correctly as when it hit it would do a LOT of damage, but because it's one bomb it wouldn't always hit.

And moreso as against heavily fortified positions, a 500 lb bomb should have NO effect (or a 0.001% probability of effect) and a 4000 lb bomb would be able to have effect IF it hits. That would be more accurate to the RL situation of wwii.

quote:

How about trying to increase fatigue for 4E bomb missions so they couldn't easily operate on a daily basis with all bombers participating? This would just mean it modelled the kind of planning it would take to logistically use a 100 or more 4Es daily. In one of the recent patches I believe fatigue was increased for low level attacks, and it applies differently at different altitudes for 2E and 4E bombers, so this kind of thing most not be too hard to fix.

An earlier post about different bomb loads makes sense as well. Many IJN bombers are allowed the 800 kg bomb on port attacks. Why not for heavily fortified positions allow the 4E a 4,000 pound bomb to make this seem more realistic? It would also be modelled more correctly as when it hit it would do a LOT of damage, but because it's one bomb it wouldn't always hit.

quote:

And moreso as against heavily fortified positions, a 500 lb bomb should have NO effect (or a 0.001% probability of effect) and a 4000 lb bomb would be able to have effect IF it hits. That would be more accurate to the RL situation of wwii.

quote:

How do you suggest calibrating the (non-linear) model?

quote:

ORIGINAL: obvert

quote:

How about trying to increase fatigue for 4E bomb missions so they couldn't easily operate on a daily basis with all bombers participating? This would just mean it modelled the kind of planning it would take to logistically use a 100 or more 4Es daily. In one of the recent patches I believe fatigue was increased for low level attacks, and it applies differently at different altitudes for 2E and 4E bombers, so this kind of thing most not be too hard to fix.

An earlier post about different bomb loads makes sense as well. Many IJN bombers are allowed the 800 kg bomb on port attacks. Why not for heavily fortified positions allow the 4E a 4,000 pound bomb to make this seem more realistic? It would also be modelled more correctly as when it hit it would do a LOT of damage, but because it's one bomb it wouldn't always hit.

quote:

And moreso as against heavily fortified positions, a 500 lb bomb should have NO effect (or a 0.001% probability of effect) and a 4000 lb bomb would be able to have effect IF it hits. That would be more accurate to the RL situation of wwii.

quote:

How do you suggest calibrating the (non-linear) model?

I know next to nothing about how a game like this is coded, but there must be an easy approximation for limits on damage accrued that could be used as an airfield, a unit, or a city take damage. The game already understands how if a base has a certain amount of forts then damage is limited from bombing attacks. Would it also be possible to say that if damage levels were over certain levels it would in effect raise the level of forts to make it harder to do more damage to that base? And then when the damage dropped again below those levels, the "fake' fort reading would also disappear.

So damage up to say 25% of strength would be determined as it is now. Damage from 25% to 50% would add +1 fort to the base or unit before the attack was calculated. Damage from 50% 75% might need to add +3 forts to keep a unit from becoming wiped out. And from 75% -99% damage the forts would go to the maximum 9 forts to make it next to impossible to destroy the unit or base completely.

Would something like that work?

quote:

ORIGINAL: herwin

quote:

ORIGINAL: PaxMondo

quote:

ORIGINAL: obvert

Why not for heavily fortified positions allow the 4E a 4,000 pound bomb to make this seem more realistic? It would also be modelled more correctly as when it hit it would do a LOT of damage, but because it's one bomb it wouldn't always hit.

And moreso as against heavily fortified positions, a 500 lb bomb should have NO effect (or a 0.001% probability of effect) and a 4000 lb bomb would be able to have effect IF it hits. That would be more accurate to the RL situation of wwii.

How do you suggest calibrating the (non-linear) model?

quote:

ORIGINAL: PaxMondo

quote:

ORIGINAL: herwin

quote:

ORIGINAL: PaxMondo

quote:

ORIGINAL: obvert

Why not for heavily fortified positions allow the 4E a 4,000 pound bomb to make this seem more realistic? It would also be modelled more correctly as when it hit it would do a LOT of damage, but because it's one bomb it wouldn't always hit.

And moreso as against heavily fortified positions, a 500 lb bomb should have NO effect (or a 0.001% probability of effect) and a 4000 lb bomb would be able to have effect IF it hits. That would be more accurate to the RL situation of wwii.

How do you suggest calibrating the (non-linear) model?

I would first create the boundary conditions from RL results translated into game terms.

Lower Boundary condition: Using Iwo results. We know the bombing campaign was one year long. 4E's dropping standard 500 lb package. That is lvl 6 forts.

so for 500 lb bombs:
Lvl 0 = current hit probability (don't know this, but let's just call it bl%
Lvl 6 = 0.01 * bl% (so saying, 500 lb bombs are now too effective by 100x. I think that is a starting point.) You can use an inverse function to interpolate ... or if reals are being used a log. (but i think i remember that all the math in game is integer).

For bomb size, again all I have is that old engr assessment report on the Iwo fortifications where they thought the 1000 and particularly the 2000 lb DB dropped bombs had some penetration. I might start with a square function on bomb effect as a multiplier on the above to bring in the bomb size.

The difficulty in tuning here is that for both the 2000 and 10,000, the payloads are 1 bomb per plane. The 500 lb are between 10 - 40 / plane. So you model needs to take that into acct as each bomb is checked for a hit. Right now, B-29's drop SO many bombs that with the current hit% you get a LOT of effective hits, far more than what Iwo data suggests you should. When you go to those brit bombers, instead of 25 or so 500 lb, they carry 1x10,000. Again, not knowing the modelling, but it might be simply an artifact of a linear model (IF that is what is being used and we do not know). In which case, MAYBE going to an inverse function might give a more appropriate effect (understand when I say inverse, it might take an inverse square or cube or higher level)


The after action assessment that I read years ago was that the US engrs could not find any real impact of this one year bombing campaign upon the IJ positions. The forts were simply too strong for a 500 lb bomb to have any effect. The 1000 and 2000 bombs dropped during DDay did have effect as well as some of the BB salvos. The issue of course here will be hindsight: the allies did NOT know this. They wasted considerable effort on bombing, tying up these valuable units in a wasted campaign, that once we change the model, players won't use them in this wasted roll. But this is a secondary game balance issue.

quote:

I would first create the boundary conditions from RL results translated into game terms.

Lower Boundary condition: Using Iwo results. We know the bombing campaign was one year long. 4E's dropping standard 500 lb package. That is lvl 6 forts.

so for 500 lb bombs:
Lvl 0 = current hit probability (don't know this, but let's just call it bl%
Lvl 6 = 0.01 * bl% (so saying, 500 lb bombs are now too effective by 100x. I think that is a starting point.) You can use an inverse function to interpolate ... or if reals are being used a log. (but i think i remember that all the math in game is integer).

For bomb size, again all I have is that old engr assessment report on the Iwo fortifications where they thought the 1000 and particularly the 2000 lb DB dropped bombs had some penetration. I might start with a square function on bomb effect as a multiplier on the above to bring in the bomb size.

The difficulty in tuning here is that for both the 2000 and 10,000, the payloads are 1 bomb per plane. The 500 lb are between 10 - 40 / plane. So you model needs to take that into acct as each bomb is checked for a hit. Right now, B-29's drop SO many bombs that with the current hit% you get a LOT of effective hits, far more than what Iwo data suggests you should. When you go to those brit bombers, instead of 25 or so 500 lb, they carry 1x10,000. Again, not knowing the modelling, but it might be simply an artifact of a linear model (IF that is what is being used and we do not know). In which case, MAYBE going to an inverse function might give a more appropriate effect (understand when I say inverse, it might take an inverse square or cube or higher level)


The after action assessment that I read years ago was that the US engrs could not find any real impact of this one year bombing campaign upon the IJ positions. The forts were simply too strong for a 500 lb bomb to have any effect. The 1000 and 2000 bombs dropped during DDay did have effect as well as some of the BB salvos. The issue of course here will be hindsight: the allies did NOT know this. They wasted considerable effort on bombing, tying up these valuable units in a wasted campaign, that once we change the model, players won't use them in this wasted roll. But this is a secondary game balance issue.

quote:

Maybe it also help to introduce "bombs" as a third supply item. This way players won´t drop three times the amount of bombs used by all countries together ;-).

quote:

ORIGINAL: obvert

Could we introduce some system that used forts and different bomb loads as ways to limit effectiveness against dug in units and bases?

Could fatigue be introduced as a way of limiting mission frequency?

quote:

ORIGINAL: PaxMondo

The good news is that some of us started coding back with the 4004 chip (ok, digital coding. We didn't refer to analog programing as "coding"), so we think that 20 year old code is "recent".

quote:

ORIGINAL: PaxMondo

Hey, I have no idea who that is in the mirror shaving every morning.  I mean, who is that old fa__?    But me, I'm still that 26yo with a 28" waist ....

quote:

ORIGINAL: mike scholl 1

quote:

ORIGINAL: PaxMondo

Hey, I have no idea who that is in the mirror shaving every morning.  I mean, who is that old fa__?    But me, I'm still that 26yo with a 28" waist ....

WHEREAS THEE GUY IN THE MIRROR IS 62..., WITH AN 82-INCH WAIST.

quote:

ORIGINAL: PaxMondo
From previous responses by various team members, we know that all of this is already in the model and a LOT more.

Tuning and/or model adjustments are being offered here. We have been told no new "development" is being, or will be, done; only tweaks, bug removal, and UI enhancements. Trying to keep this in the "tweak" dept or it will be put into the WitP AE II wish list which may or may not ever happen.

And we have to remember that this is +20 year old code at the heart of everything. The good news is that some of us started coding back with the 4004 chip (ok, digital coding. We didn't refer to analog programing as "coding"), so we think that 20 year old code is "recent".

quote:

ORIGINAL: Nikademus

quote:

ORIGINAL: PaxMondo
From previous responses by various team members, we know that all of this is already in the model and a LOT more.

Tuning and/or model adjustments are being offered here. We have been told no new "development" is being, or will be, done; only tweaks, bug removal, and UI enhancements. Trying to keep this in the "tweak" dept or it will be put into the WitP AE II wish list which may or may not ever happen.

And we have to remember that this is +20 year old code at the heart of everything. The good news is that some of us started coding back with the 4004 chip (ok, digital coding. We didn't refer to analog programing as "coding"), so we think that 20 year old code is "recent".

You are correct. Several key developers have also declared repeatedly that whoever tries to do "WitP II" it WONT be them. Once was enough. Gary Grigsby himself stated that he will never attempt such a beastie again.