[FIXED B555] Seems torps miss their targets alot now.

[kaburke61]

Hey There,

Maybe this goes along with sonar tweaks, but torps (fired from helos and nimrod) miss most of the
time now, even when the sub is moving 13kts and is dead center in the acquisition cone.

I have played the scenario 3 or 4 times, and I get a hit maybe 10-20% of the time.

Even when they are lined up perfectly, dead center in the cone, the torps keep going into
circle mode, and usually later you get a "torp overshot it's target" message, even though
it didn't overshoot (it actually undershot)

In the "old" days with a fix like that, the torp usually found the target (then of course the
hit percentage kicked in).

Thanks for looking at it.

UPDATE: I just created a mission (P3 Orion vs Kilo) and it does it there too, even though
the Orion has a perfect location fix. Torp never sees the sub. Could others try this out, see what they get?
Maybe the launching platform is now dropping too early?

[Feltan]

I have observed the same issue with SH-60 and MK46 torpedoes. I haven't written it up as a defect, because I'm not sure what I am witnessing is in fact anything other than bad die rolls.

However, you can "fix" this by dropping the torpedo within the 0.5 nm range -- doing so gives you a "normal" kill percentage.

My initial non-scientific observation is that if the torpedo goes active outside of the 0.5 nm range it is likely to "go stupid" and swim in circles.

Regards,
Feltan

[mikmykWS]

Thanks Kevin. Added an extra helo and it spiraled way to many times I think. Have let the guys know.

BTW like the angry little graphic. Its not that bad[;)]

Thanks!

Mike

[kaburke61]

ORIGINAL: Feltan

I have observed the same issue with SH-60 and MK46 torpedoes. I haven't written it up as a defect, because I'm not sure what I am witnessing is in fact anything other than bad die rolls.

However, you can "fix" this by dropping the torpedo within the 0.5 nm range -- doing so gives you a "normal" kill percentage.

My initial non-scientific observation is that if the torpedo goes active outside of the 0.5 nm range it is likely to "go stupid" and swim in circles.

Regards,
Feltan

Yep, I noticed too that if you manually drop it "over" the target, it will work better. It seems that the AI
now drops too soon based on (changed?) sonar performance. This homing behavior precedes the "die rolls"
(at least the ones we see) I believe.

It looks like overall sonar detection across the board has been tweaked, as I notice less sensitive sonar bouys
(especially passive), and torp behavior. It very well could be realistic, maybe just the AI needs to be adjusted
to drop later. Either way, things have definitely gotten much more difficult in prosecuting sub-surface baddies!

[kaburke61]

ORIGINAL: mikmyk

Thanks Kevin. Added an extra helo and it spiraled way to many times I think. Have let the guys know.

BTW like the angry little graphic. Its not that bad[;)]

Thanks!

Mike

Thanks for taking a look. My Orion vs. Kilo test showed it pretty consistently. In the "old" days if the
target was in the cone, usually it would track. Now most of the time (unless VERY close) it doesn't
seem to.

Hehe, the graphic is one of your standard ones....

[mats0916@rogers.com]

ORIGINAL: kaburke61
Maybe this goes along with sonar tweaks, but torps (fired from helos and nimrod) miss most of the
time now, even when the sub is moving 13kts and is dead center in the acquisition cone.

Hello!

I can concur. I see the same thing.

[mikmykWS]

Hi guys. Was using slant range. Fix should appear in next build.

Thanks

Mike

[Dimitris]

Tested and debugged this extensively.

Put simply, you have a bad hand here. You have a crappy torp seeker to begin with (Mk46 Mod 2, 0.25-0.3nm max sonar range in perfect conditions), and on top of that you have high reverberation due to the shallow bottom depth (just 57m) and a target with a relatively small active sonar signature. When all is said and done, the actual detection range of the torp falls to just _82 meters_ . No wonder you have to almost deliver it like a bomb to have a chance.

(We slightly tweaked this in B550 so that torpedo sonars are assumed to have a lower nominal target signature to compare to, than other sonar types. This bumps the actual range in this scenario to a majestic ~250m).

This is precisely why many coastal naval forces still love ASW rockets, depth charges etc., and why higher nominal seeker range and better signal processing (to better filter out reverbs) are high priorities for littoral-oriented torps like the latest Mk48 mods.

[kaburke61]

ORIGINAL: Sunburn

Tested and debugged this extensively.

Put simply, you have a bad hand here. You have a crappy torp seeker to begin with (Mk46 Mod 2, 0.25-0.3nm max sonar range in perfect conditions), and on top of that you have high reverberation due to the shallow bottom depth (just 57m) and a target with a relatively small active sonar signature. When all is said and done, the actual detection range of the torp falls to just _82 meters_ . No wonder you have to almost deliver it like a bomb to have a chance.

(We slightly tweaked this in B550 so that torpedo sonars are assumed to have a lower nominal target signature to compare to, than other sonar types. This bumps the actual range in this scenario to a majestic ~150m).

This is precisely why many coastal naval forces still love ASW rockets, depth charges etc., and why higher nominal seeker range and better signal processing (to better filter out reverbs) are high priorities for littoral-oriented torps like the latest Mk48 mods.

OK, but did something change regarding sonar recently? The "bad hand" was always sufficient previously as I
have played this scenario (Advanced ASW Exercise) many, many times, with many versions as this is the
test scenario I use when you release a new build. In the past, AI-released torpedos would have no problem (usually
of course, not every time) homing in on their targets. As stated previously I now see torpedos unable to track
even though the target is over 50% or more deep within the targeting cone. If this is desired, due to "realistic"
tweaks, please (which I think you said you did) make sure they are releasing a little closer, because now they
mostly just waste their loadouts trying.

Thanks,

Kevin

[mikmykWS]

Best bet is to check this out next version. We can always revisit [:)]

Mike

[kaburke61]

ORIGINAL: mikmyk

Best bet is to check this out next version. We can always revisit [:)]

Mike

Will do...Thanks!

[kaburke61]

ORIGINAL: mikmyk

Best bet is to check this out next version. We can always revisit [:)]

Mike

Hey Again,

Seems to be a little better. I am getting more hits....but...

A better percentage of them now track (especially slow moving tagets),
but it seems they still launch early, it seems if they waited just a tad longer they wouldn't undershoot.
Torp sometimes can still not make it to where the "active" cone is over the sub before the torpedo starts
to circle, even though there was an exact fix on distance and speed of the sub, and it didn't make course
corrections. It is dependent on target speed.

[UPDATE] They definitely launch too soon. Here is a quick test.

If the target is moving at a decent speed, the launch is too early.
It will sometimes home, (1 out of 10 in my test scenario, and I'm not referring to hit percentage,
just homing to target) but you can see from this test, It's dropping too soon.


Thanks,

Kevin

[kaburke61]

Bump...did the updated test scenario get looked at?

I knoe it's not a showstopper, but it just that little more micromanagement that wasn't
needed previously

[mikmykWS]

No not yet. We decided to push forward with the update. We'll pick up again after a rest. Your issue wasn't forgotten.

Mike

[kaburke61]

ORIGINAL: mikmyk

No not yet. We decided to push forward with the update. We'll pick up again after a rest. You're issue wasn't forgotten.

Mike

OK...thanks for the reply.

[ComDev]

Have added a ticket on this one, suspect there may indeed be a problem here. Thanks!

[kaburke61]

ORIGINAL: emsoy

Have added a ticket on this one, suspect there may indeed be a problem here. Thanks!

Thanks for the help! I think the second attachment (Kilo vs. Orion) shows the problem better than the
first attachment I sent.