The DW Mod Scene - Ok, WTF?

Originally Posted by To be

I've used both RA and LWAMI - I very much enjoy the ability to control multiple platforms in RA (as well as the enhanced weapons controls) - it adds a lot to the game. LWAMI however crashes less, (though RA doesn't crash too often), and the AI behavior in LWAMI seems much better than RA. I use LWAMI unless I specifically want to control a non-standard sub, mostly due to the better AI behavior.

You can check aggression in such a way.
Make an active ping.
Also see - as quickly AI will answer to you with a torpedo shot.
LWAMI will give an answer-back torpedo shot from 6 about 9 minutes.
RA - will answer immediately - no more than 1 minute. (distance passage of a signal up to the target both back and some seconds on procedure launch of torpedos) is taken into account.

LWAMI of the doctrine - the predicted algorithms of evasion use.
Reinforce Alert - all time uses different variants is is more unpredictable.

RA - takes into account the current depth of a sub with its speed - for prevention cavitation.
LWAMI - nothing takes into account.

RA - diesel boats correctly work at snorkeling.
LWAMI - snorkeling is absent.

RA - the noise by the surface ships - depends on their speed and grows linearly.
LWAMI - the surface ships change the noise only before occurrence cavitation.
If cavitation - then noise all time identical (base noise + 20). It concerns also submarines.

RA - the PD depth for AI of sub is correctly now.
LWAMI (and also original of game) - not correct depth. It is a mistake in the formula from sonalist.

RA - shoots SUBROCs very exact.
LWAMI - AI SUBROCs practically useless.

RA - action under polar ice real.
LWAMI - the submarines pass through ice as through butter, and are not damaged.

RA - begun AI helo from AI of the ship - at first will find out a submarine, then will drop torpedos.
LWAMI - begun AI helo from AI of the ship - simply will drop torpedos in a point to which has arrived, even if a submarine there already no.

RA - the radar on submarines of the player works without a bugs.
LWAMI - the radar on 688 and Seawolfs if is lifted on depth (PD - 1 ft) not counterdetected with ESM of AI platforms.
etc. etc...

Such differences - hundred.
But I shall not list them all.

[Pillar]

But I shall not list them all.

Well yes, do that. That's the point of the thread after all.

If the concern is accuracy, well, there are developers from both mods able to comment here so maybe in the process we'll clear up some confusion.

It's not a contest. Let's figure this out.

[To be]

Originally Posted by -GrayOwl-

You can check aggression in such a way.
Make an active ping.
Also see - as quickly AI will answer to you with a torpedo shot.
LWAMI will give an answer-back torpedo shot from 6 about 9 minutes.
RA - will answer immediately - no more than 1 minute. (distance passage of a signal up to the target both back and some seconds on procedure launch of torpedos) is taken into account.

LWAMI of the doctrine - the predicted algorithms of evasion use.
Reinforce Alert - all time uses different variants is is more unpredictable.
.
.
.

Such differences - hundred.
But I shall not list them all.

The response to particular events might be great - but all I can say is that in my experience the overall employment of tactics by the AI (especially from surface vessels) in RA is significantly less challenging to defeat than the LWAMI behavior. One particular way to see this is to set up a surface fleet of (AI) warships (say, a carrier group), and place a few (AI) subs ahead of the group on barrier patrol. The ships, in my experience, do a vastly better job fighting back in LWAMI than in RA. I use both mods of course, both have their strengths. I'd be happy to send you an example replay file of the kind of test I was talking about if you'd like to see them.

Originally Posted by To be

The response to particular events might be great - but all I can say is that in my experience the overall employment of tactics by the AI (especially from surface vessels) in RA is significantly less challenging to defeat than the LWAMI behavior. One particular way to see this is to set up a surface fleet of (AI) warships (say, a carrier group), and place a few (AI) subs ahead of the group on barrier patrol. The ships, in my experience, do a vastly better job fighting back in LWAMI than in RA. I use both mods of course, both have their strengths. I'd be happy to send you an example replay file of the kind of test I was talking about if you'd like to see them.

You are mistaken.

The ships at all work in LWAMI not better.

The doctrines for the ships - in any way do not process their tactics.
The doctrines only operate the weapon - and that only partially.

You can in general remove the doctrine CIWS.txt and CIWSAttack.txt.
Despite of it - the ship will have the same behaviour and also will launch the weapon despite lacking the doctrines.

One word - it is manages from NavalSimEngine - but not from the doctrine completely.

There is one known tactics - shoot one missile against the surface ship.
After that - the ship will be sped up up to speed washout of the sensor controls.
And after that you shoot a passive torpedo. The ship will be 100 % killed.

With RA - such will not allow.

You can put the test mission on a forum RedRodgers as attachment archive.

[To be]

Originally Posted by -GrayOwl-

You are mistaken.

The ships at all work in LWAMI not better.

The doctrines for the ships - in any way do not process their tactics.
The doctrines only operate the weapon - and that only partially.

You can in general remove the doctrine CIWS.txt and CIWSAttack.txt.
Despite of it - the ship will have the same behaviour and also will launch the weapon despite lacking the doctrines.

One word - it is manages from NavalSimEngine - but not from the doctrine completely.

There is one known tactics - shoot one missile against the surface ship.
After that - the ship will be sped up up to speed washout of the sensor controls.
And after that you shoot a passive torpedo. The ship will be 100 % killed.

With RA - such will not allow.

You can put the test mission on a forum RedRodgers as attachment archive.

I am not mistaken about my own experiences, and I encourage you to try setting up a mission of the type I described to see for yourself the behavior I described. Any improvement to both, or either, mod would be very welcome to me. However, you appear to be more interested in ignoring any criticism blindly, rather than taking it constructively. No mod will ever be perfect, so outright rejecting any comments about ways it could be improved isn't very productive. Additionally your comments about LWAMI being "absolutely wrong" are not very respectful to the developers of LWAMI - whereas they have shown nothing but respect for the work of the RA team. I doubt that you would appreciate the same sort of 'feedback' given to your own work.

[goldorak]

Originally Posted by To be

No mod will ever be perfect, so outright rejecting any comments about ways it could be improved isn't very productive. Additionally your comments about LWAMI being "absolutely wrong" are not very respectful to the developers of LWAMI - whereas they have shown nothing but respect for the work of the RA team. I doubt that you would appreciate the same sort of 'feedback' given to your own work.

You are right, no mod will be the perfect mod. This doesn't mean that all mods are created equal. As incredibile as the work on Lwami is and has been over the past several years, there is a point beyond which they simply cannot/willnot enhance the game. Lwami still carries a lot of bugs that are hardcoded in the game engine. They were present in DW 1.0 and are still present in DW 1.04 + lwami 3.10. No amount of modding the database and doctrines will fix these bugs. If people accept this situation then all is good.
But we now have a mod that tries to fix those hardcoded bugs. And this is a good thing too. From this point of view Lwami is "wrong" is the sense that it still relies on buggy behaviour from the navalsimengine.
The problem wouldn't exist in the first place if SCS had done their job and released a functionning navalsimengine. That unfortunately was not the case.

[To be]

Originally Posted by goldorak

You are right, no mod will be the perfect mod. This doesn't mean that all mods are created equal. As incredibile as the work on Lwami is and has been over the past several years, there is a point beyond which they simply cannot/willnot enhance the game. Lwami still carries a lot of bugs that are hardcoded in the game engine. They were present in DW 1.0 and are still present in DW 1.04 + lwami 3.10. No amount of modding the database and doctrines will fix these bugs. If people accept this situation then all is good.
But we now have a mod that tries to fix those hardcoded bugs. And this is a good thing too. From this point of view Lwami is "wrong" is the sense that it still relies on buggy behaviour from the navalsimengine.
The problem wouldn't exist in the first place if SCS had done their job and released a functionning navalsimengine. That unfortunately was not the case.

I very much agree. RA has taken DW mods to a whole new level, and I enjoy it very much. (Except I'm not going to blame SCS - they created a game where there is little market and even less hope of a profit. I'm thankful DW exists - just as no mod is perfect no game is perfect, I'm happy to have what we have.)

Originally Posted by To be

I am not mistaken about my own experiences, and I encourage you to try setting up a mission of the type I described to see for yourself the behavior I described. Any improvement to both, or either, mod would be very welcome to me. However, you appear to be more interested in ignoring any criticism blindly, rather than taking it constructively. No mod will ever be perfect, so outright rejecting any comments about ways it could be improved isn't very productive. Additionally your comments about LWAMI being "absolutely wrong" are not very respectful to the developers of LWAMI - whereas they have shown nothing but respect for the work of the RA team. I doubt that you would appreciate the same sort of 'feedback' given to your own work.

Here asked - what distinctions between these mods.

Partially I have answered.

If I shall begin list other distinctions (I can direct to name them bugs, default comes from SCS game version) - then you again will say that I intentionally " lower downwards " LWAMI.

[dd149]

GrayOwl, I think that is the US it could be a legal concern to openly modify dlll/exe, so please understand Molon Labe and others. They are in no way responsible for the shortcomings of SCS, but have tried to improve DW while avoiding legal trouble, which can really problem in the US.
On the other hand I believe that everybody likes what you are doing in the RA team, as many of original bugs are now corrected by your hard work, with others still being under work.
May I suggest that we could all join forces in developing the next stage. You and RA are certainly the one for the hardcoding, but why not making use of modding by others too? We know that it is not easy and would involve some communication/documentation issues but it would be worth trying.

[Pillar]

Well, does anyone feel enlightened?

I don't really.

Anyways, GreyOwl I had a look with DWEdit and noticed in RA you have passive sonars giving range, course and speed data (like Radar might.) Why is this?

[goldorak]

Originally Posted by -GrayOwl-

You can check aggression in such a way.

[cut]

RA - shoots SUBROCs very exact.
LWAMI - AI SUBROCs practically useless.

[cut]

And THIS is one hell of a big difference. AI subs with subrocs are very very deadly, we in Betasom learned the lesson.
Lada/Kilo AI subs getting within no more than 15-20 minutes deadly accurate solutions on a playable Type 212 some 10 nm + out at slow speed. It was a shock to some of our players just how reveresed the situation was with respect to AT3 or even Lwami. It makes mixing AI/playable units that much more challenging.

Such differences - hundred.
But I shall not list them all.

I think this is a mistake although I understand why you don't want to take that route. Good software comes with good documentation, and although the included weapons manual and sonar profiles are very good, a good deal of the behaviour of the mod has to be learned on the field by playing and playing. And we always discover new things. This of course takes time, and certain players accustumed to just having a changelog with all the differences with respect to DW default may be put off. And some even think the mod works badly when its just the correct behaviour because its not documented at all.
But, if you stick with it, and learn by trial and error you'll see just how big a quantum leap RA/DWX really is over the other mods.

[Hawk66]

Originally Posted by goldorak

And THIS is one hell of a big difference. AI subs with subrocs are very very deadly, we in Betasom learned the lesson.
Lada/Kilo AI subs getting within no more than 15-20 minutes deadly accurate solutions on a playable Type 212 some 10 nm + out at slow speed.

But is that realistic that a Kilo can detect a Type 212 at slow speed 10 nm away with an accurate solution?

[goldorak]

Originally Posted by Hawk66

But is that realistic that a Kilo can detect a Type 212 at slow speed 10 nm away with an accurate solution?

Well the Lada has a towed array so its acceptable.
As for the Kilo, while it may not have a towed array, its sonar is not completely useless. You can't track targets at very long range, but 10 nm is not out of the question.
Having an accurate solution is a matter of having the correct number of bearing lines + speed information. If the target mantains fixed speed, the autotma pinpoints the solution much faster than if you changed speed and or/course every so often.
20 minutes gives you 9 bearing lines, plus speed info is more than enough to pin down a very good solution.

In any case some things have to take into account playability yes even in a simulation. And before people start knocking down RA, Lwami in the last versions (although it was finally corrected) went completedly overboard in making the SSK practically invisibile.
So much so that Molon Labe in Lwami 3.10 reajusted the values for something more balanced.

You have to balance realism with playability.
Those that think otherwise would be better served by enlisting in a real navy and use the professional simulators.
Thats real work.

[Molon Labe]

Originally Posted by goldorak

In any case some things have to take into account playability yes even in a simulation. And before people start knocking down RA, Lwami in the last versions (although it was finally corrected) went completedly overboard in making the SSK practically invisibile.
So much so that Molon Labe in Lwami 3.10 reajusted the values for something more balanced.

You have to balance realism with playability.
Those that think otherwise would be better served by enlisting in a real navy and use the professional simulators.
Thats real work.

We did move a few submarine active source levels up in 3.10, but it wasn't done for balance. It was done because I accidentally moved the low end too far down in 3.09 and acquisition ranges ended up shorter than we ever wanted them--unrealistically short, at least in my estimation.

There are a number of spots where balance is accounted for in the mod, though, so your point remains valid (as long as Batman isn't riding an elephant, anyways).

ML - The conversation goes concerning noise of a submarine, instead of about her answer-back active echo.

Your mistake that you use the formula Ludger's which is more- less correct for surface ships (will not come yet cavitation - then noise all time not increased).
The formula uses LINEAR gain noise.

However - Submarine Gain noise, uses other formula - LOGARITHMIC!.

Your tables of comparison of noise of submarines - absolutely wrong.

Your tables are suitable only for the surface ships - from some share of a mistake (however on some speeds a share of a mistake makes 3 units(!) from current real noise).

The submarines generate absolutely other noise.