[TEC] Trying to figure out the hydrophones

[LukeFF]

Concerning the submarine hydrophone settings in sensors_sub_US.sim, what values do the lines Surface and RPMDetLevel control?

[LukeFF]

Hello? Anyone?

[Ducimus]

Not sure about RPM det level, surface.. ehh.. ill look at the file later. Im tired and dont want to even open a damn editor right now.

[DrBeast]

Just guess-work here, I think RPMDetLevel controls controls how noisy the enemy engine has to be for your hydrophone to pick up, i.e. works as the Noise Factor in sensors.cfg. I don't think Surface has any effect.

[LukeFF]

Originally Posted by DrBeast

Just guess-work here, I think RPMDetLevel controls controls how noisy the enemy engine has to be for your hydrophone to pick up, i.e. works as the Noise Factor in sensors.cfg. I don't think Surface has any effect.

That's what I was thinking. So, does a higher number mean the ships have to generate more noise in order for the hydrophones to pick up the sound? Currently it is set to 0.25 for the submarine passive sonar. Would setting it to, say 0.50, double the noise requirement for the hydrophones to pick up the sound?

[ref]

RPMDetLevel is the detection level as a percentage of RPMs (0 to 1, 1 is 100%), the target ship should go at for example half it's maximum speed to be detected if you set the value to 0.5

Surface is the minimum detectable surface at sensor max range in square meters, it doesn't apply for hydrophones.

Ref

[DrBeast]

Originally Posted by ref

RPMDetLevel is the detection level as a percentage of RPMs (0 to 1, 1 is 100%), the target ship should go at for example half it's maximum speed to be detected if you set the value to 0.5

Ref

Hm...interesting. However, that means that RPM as an absolute value in pretty useless, i.e. it doesn't matter if an engine's max RPM is 300 or 3000...?

[Peto]

Originally Posted by DrBeast

Hm...interesting. However, that means that RPM as an absolute value in pretty useless, i.e. it doesn't matter if an engine's max RPM is 300 or 3000...?

I think it's works as a percentage rather than an actual turn count. :hmm:

[DrBeast]

Originally Posted by Peto

I think it's works as a percentage rather than an actual turn count. :hmm:

Only one way to find out, I guess...use a Sub as a "test subject" and play around with its RPMs. Get a stock sub, race at flank speed towards a Destroyer, see at what range you're detected. Then double the RPMs of the sub (as listed in the sub's .sim) file and check to see when the DD will react.

ETA: I'd use a Sub in the test scenario to have full control of how fast it's going and also when you're detected. Provided it works the same the other way around (i.e. when you're trying to pick up noise from the enemy), we'll have pretty solid evidence on how it works.

ETA #2: To be absolutely sure it's the AI hydrophone that's detecting you, I'd also make the AI severely near-sighted and with no radar.

[LukeFF]

Originally Posted by ref

RPMDetLevel is the detection level as a percentage of RPMs (0 to 1, 1 is 100%), the target ship should go at for example half it's maximum speed to be detected if you set the value to 0.5

Surface is the minimum detectable surface at sensor max range in square meters, it doesn't apply for hydrophones.

Thanks. Me and the rest of the RFB team is testing out a new tweaked hydrophone with slightly upped RPMDetLevel values. Hopefully it's the much-sought fix to keep the surface hydrophones from being god-like in their detection capabilities.

[DrBeast]

Halfway through the test run I suggested above. Just finished tests with stock RPM settings. Will commence further testing when I get home again and report my results tomorrow.

[DrBeast]

Test run complete, as promised.

Material Components(!):
Single mission set at Midnight, Dec 1, 1945. No wind, fog or clouds. Balao class fleet boat headed north, Bungo Pete (Elite Akikaze Destroyer) headed south in a path that leads directly to the sub.
Tweaked AI_Sensors.dat and sim.cfg to give the Destroyer a super-sensitive, though relatively short-ranged hydrophone (6000m max range, min Elevation 90, Sensitivity 0.06) but virtually useless radar and eye-balls (Sensitivity set to 0.5).

Methods:
Test Run 1-3: Balao headed straight at DD, at Periscope depth and flank speed (Run 1), Ahead Standard (Run 2), and Ahead 1/3 (Run 3). Repeated each run 5 times.
Test Run 4-6: Same as 1-3, but RPM doubled in the Balao.sim file (540 vs. 270 stock).

Results:
Average detection threshold for Ahead Standard: 4014+-39 meters at stock RPM, 4220+-109 meters at x2 RPM
Average detection threshold for Ahead 1/3: 2578+-89 meters at stock RPM, 2746+-44 meters at x2 RPM
The difference is statistically not significant.

Verdict:
You could as well put 30.000 RPM, the results will be the same: it's the percentage that matters, not the absolute value. As you can see, I was detected roughly 1500 meters earlier when I was running at Ahead Standard, compared to Ahead 1/3, regardless of absolute RPM count.

Where is Ahead Flank?!?
Well, the game behaved pretty funny when I was racing at Ahead Flank. Big differences in the detection threshold, and what's weirder, I'd actually get detected earlier when I was at Ahead Standard! For comparison:
Average detection threshold for Flank: 3782+-195 meters at stock RPM, 3606+-141 meters at x2 RPM.

Now, I leave it to you guys to play with RPM Det Level