Stranger than you CAN imagine

[Munchausen]

Originally Posted by Rockin Robbins

Actually I WAS set up for a Dick O'Kane, because the speed of those puppies really was zero. My Shooting Gallery Test Mission has them anchored out there so the only error can come from the inherent geometry of the torpedo course itself. My own submarine was also stationary.

Ah ha ... okay, I guess that accounts for the odd display on your TDC. I tested the "long shot" using your Cromwell zero-gyro setup instead ... adjusting my range so that I was 5,000 yards from the tanker when I fired. My results were as follows:

• Calculated lead angle was 21 degrees. Neither the scope nor the TBT is calibrated to one degree ... and, as distance increases, errors in just half a degree can make a bigger difference in hitting or missing your target.
• I left TDC range set to the 500 yard default. Didn't seem to make much difference.
• Set my torpedoes to run slow at a depth of 17 feet ... keel depth minus 11 feet. Perhaps if you leave them set to default, they broach, then sink. Or, if you set them to the full keel depth, they go too deep, then sink. All four of mine made it to the target.

The O'Kane method can be combined with the Cromwell method ... the above test was made at very close to a 90-degree setup. The O'Kane method gives you the optimum AOA (broadside) and the Cromwell method gives you an exact lead angle.

But, the greater the range to target, the greater the need for exact accuracy in sighting your aiming device (TBT or scope).

The biggest surprise for me: Awhile ago, someone asked why they could not set their TDC to range zero. He was all upset about it. So in my tests I tried some 8000 yard shots at the zero bearing with the TDC set to zero range. All the shots missed right. Any other range hit the target if the tropedoes got there. Anybody have any idea why that should be so?

I'm surprised the torpedoes even headed in the correct direction. Try an attack in the "auto" mode and watch what happens to the TDC when a locked target gets really close to your sub. The "torpedo track" arrows go nuts ... almost like the old computer error of dividing by zero. If, in manual mode, the TDC thinks the target is at zero range, that might be enough for it to do what it does in auto mode.

[kylania]

The conclusions we can make here are the following. Don't shoot Dick O'Kane shots from more than 2000 yards, with torpedo gyro angles much above 15º left or right. Between 345 and 15, you'll get your hits. Outside that you better put a rough range estimate into the TDC. There's plenty of slop, but the more accurate the better.

Wasn't that always the Dick O'Kane method? 345-15 shot, 90 angle but range "doesn't matter" since it's only to be used against close targets (ie, 1500y or less, preferably much less)? I'm confused how range mattering is such a surprise now?

[Rockin Robbins]

Originally Posted by Hitman

But the TDC can't automatically compensate everything. True it will take into account one side of the triangle, which is known -distance from the optics to the torpedo tube- but it still will lack the variable of distance to target in order to set the proper correction for parallax.

But the parallax between the position of the torpedo tube and the periscope isn't the whole picture. The real parallax is between the periscope and the end of the reach, plus at least part of the radius of the turn. The part caused by the reach of the torpedo is much greater than the separation between tube and periscope. Now we've lost everybody!:rotfl:

I guess what I'm trying to say is that if the torpedo took a straight line out of the tube to the target, it would be hitting targets at 3000 yards without range input. Since it runs straight out the front for the length of the reach (don't have the figure in front of me, but it's more than the length of the submarine) and then turns, the parallax is great enough to cause misses.

I'm still amazed that the devs worked all this out and built it into the simulator. Not many have noticed it or poked at it with a stick like I have. Seems like a lot of work to entertain Hitman and Rockin Robbins.:rotfl:

[Skyhawk]

Originally Posted by Rockin Robbins

Seems like a lot of work to entertain Hitman and Rockin Robbins.:rotfl:

Stranger than YOU can imagine? :rotfl:

AFWIW, I find your posts highly entertaining, informative, and to be self-revealing. We truly are not worthy.

Carry on . . .

[Arclight]

Originally Posted by Rockin Robbins

I'm still amazed that the devs worked all this out and built it into the simulator.

To be honest, I wouldn't expect anything less. You input the data to "fix" a point in space and the TDC works out how to get a torpedo to this intercept point. If the devs left that out, it would pretty much be impossible to get a torpedo on target, and they might as well have scrapped the game altogether...

[Mush Martin]

Originally Posted by Rockin Robbins

The biggest surprise for me: Awhile ago, someone asked why they could not set their TDC to range zero. He was all upset about it. So in my tests I tried some 8000 yard shots at the zero bearing with the TDC set to zero range. All the shots missed right. Any other range hit the target if the tropedoes got there. Anybody have any idea why that should be so? I think that is the only situation where the simulation yields different results than the real TDC.

Limitation of the medium I suspect.
the torpedo tells itself to arm in a
No arming zone and once it travels
beyond the set range it doesnt signal
arming.
(idiot speculation I have no data)

M

[elanaiba]

Torpedoes have a fixed arming range, depending on model. Its not related to the range you set in the TDC.

And of course, the range in the TDC is important to correct for paralax errors.

[Hitman]

But the parallax between the position of the torpedo tube and the periscope isn't the whole picture. The real parallax is between the periscope and the end of the reach, plus at least part of the radius of the turn. The part caused by the reach of the torpedo is much greater than the separation between tube and periscope. Now we've lost everybody!:rotfl:

Of course I took for granted that we were talking about the whole process until the torpedo takes the final, straight-running trajectory

I'm still amazed that the devs worked all this out and built it into the simulator. Not many have noticed it or poked at it with a stick like I have. Seems like a lot of work to entertain Hitman and Rockin Robbins.

Yeah, thus the money we paid for the game has been twice as good: we got the game and also a free puzzle to heat our brains :rotfl:

[elanaiba]

[quote=Hitman]

I'm still amazed that the devs worked all this out and built it into the simulator.

It wouldn't be a true simulator without it, would it?

We have some great programmers that like to research and do things properly, even before design asks something from them. One such example is Cristi Mihalache (one of the guys at the SH3/Microsoft booth at E3 a couple of years ago) who did the tdc code for SH3 and the impact angle calculations for torpedo duds purposes.

[Diopos]

[quote=elanaiba]

Originally Posted by Hitman

It wouldn't be a true simulator without it, would it?

We have some great programmers that like to research and do things properly, even before design asks something from them. One such example is Cristi Mihalache (one of the guys at the SH3/Microsoft booth at E3 a couple of years ago) who did the tdc code for SH3 and the impact angle calculations for torpedo duds purposes.

You almost restored my faith to professional game designers.
Almost...
For a whole minute or so...
:rotfl:

to Cristi Mihalache

Thanks for the game guys!

[Rockin Robbins]

[quote=elanaiba]

Originally Posted by Hitman

It wouldn't be a true simulator without it, would it?

We have some great programmers that like to research and do things properly, even before design asks something from them. One such example is Cristi Mihalache (one of the guys at the SH3/Microsoft booth at E3 a couple of years ago) who did the tdc code for SH3 and the impact angle calculations for torpedo duds purposes.

Actually that was my quote. I just thought it was time you devs got some demonstrated credit for all the hidden work you do. Time after time people have called bug on something that was really a carefully researched proper simulation of a little known corner of reality.

I particularly remember the screaming about the sonar working on the surface "bug" and then showing that the sonar could continue to "work efficiently on the surface" according to the Submarine Sonar Operator's Manual.

Frequency of air attacks is another area where you devs take unjustified abuse. If you play with the airplanes they'll certainly play with you.

Anyway, poking the simulator with a stick revealed the underlying complexity and allowed us to appreciate the work devs do. Too often (insert disparaging adjective) people have suggested that Ubi should just hire modders to put the next sub simulator together. There is a HUGE difference between modding and developing.

[gutted]

i see you've finally figured out what i had been talking years ago

range doesn't matter as long as the gyro is 0. the further away from 0 the more it matters due to the torpedo's initial run before turning.

thats why it's best to shoot when the gyro is close to 0, to cancel out any descrpencies in your calculated range. had the torpedos been mounted on a turret, the would be NO gyro angle.. and the range wouldn't matter.

you can put a fancy name on it and call it a cool method.. but it's just common sense.

[Rockin Robbins]

Originally Posted by gutted

i see you've finally figured out what i had been talking years ago

range doesn't matter as long as the gyro is 0. the further away from 0 the more it matters due to the torpedo's initial run before turning.

thats why it's best to shoot when the gyro is close to 0, to cancel out any descrpencies in your calculated range. had the torpedos been mounted on a turret, the would be NO gyro angle.. and the range wouldn't matter.

you can put a fancy name on it and call it a cool method.. but it's just common sense.

Well gutted, I actually understood that. I was poking the simulator with a stick to answer the question "How MUCH does range matter, what is the allowable error and how does all that play when shooting ridiculous ranges? What about Mary Lou?" It was sort of like investigating the effect of gasoline on fire. But not quite as much fun.

I had figured the error out back in the necro thread. You dredged up the thread where arronblood, you and I first got together to see if these fleet boats could do what U-Boats could. Remember, I dragged out WernerSobe's Natural Sinking Mechanics Test Mission and plugged ships from different angles there, concluding that Dick O'Kane from similar 1000 - 1500 yard distances would work with a gyro angle up to 30º? You educated me on the theory, aaronblood provided MoBo and more theory and I adopted it to fleet boats and did my PT Barnum thing without all the pretty girls and elephants. Part and parcel of Dick O'Kane was to get pretty close to zero gyro angle (although I kept it simple and never talked about that. I just said that the nearer the target was to a 90º angle when hit, the more error the method could absorb).

John P Cromwell also works from a near or exactly zero gyro, but the heart of the method is the simple rule of thumb for calculating the AoB (45-bearing from the bow). Actually we could have made a more complicated rule that would allow people to take any angle they chose. But my crime signature has always been to assemble fixed attack methods, governed by simple rules of thumb, with no external calculators or charts. I eliminate every single step that doesn't cause a miss. My "everything I needed to know about submarines I learned in kindergarten" approach is done on purpose because my goal is to get the game newbie to check that manual targeting box and KNOW they can go out there and sink ships with confidence.

Your observation was also at the root of the same conversation between me and Nisgeis, the man who suggested the 45º angle attack. He saw the heart of the technique as vector analysis to get that perfect zero gyro shot. I saw the heart of the method as the simple formula to calculate AoB that made any external calculation or remembering torpedo speeds unnecessary. I admit that I am always ready to toss out any step that isn't absolutely necessary. I admit I am willing to sacrifice precision for "close enough" if it means the procedure is easier. My focus is on the beginner and there is no limit to the foolishness I will engage in to convince them to shoot manually. I have no shame. Mr. Rogers' rotting corpse paid me a visit last night to kill me for my equally smelly imitation.:rotfl:

I know that's frustrating to you advanced shooters. I drive aaronblood crazy with my oversimplifications. But Dick O'Kane and John P Cromwell aren't for advanced shooters, even though they will still work for them. They might be a good starting point, but all kinds of refinements and variations are possible. I understand that, but pardon me if I keep it SIMPLE. You guys are already deadly. I'm just trying to enlarge the club.

[Hitman]

range doesn't matter as long as the gyro is 0. the further away from 0 the more it matters due to the torpedo's initial run before turning.

Actually YES it matters even at zero G-Angle

There will still be a difference -although minimal- in parallax due to the distance from the torpedo initial position (In its tube) to the optics. And, if you are not using an automated calculator (Like the TDC) but instead shooting based on a manually calculated lead-angle (A la WW1) you would also need to take into account the time that the torpedo takes to accelerate to its cruise speed, because you will have had the firing triangel set up with that speed, and the overall average of the run will in fact be slightly slower.

[Rockin Robbins]

Originally Posted by Hitman

Actually YES it matters even at zero G-Angle

There will still be a difference -although minimal- in parallax due to the distance from the torpedo initial position (In its tube) to the optics. And, if you are not using an automated calculator (Like the TDC) but instead shooting based on a manually calculated lead-angle (A la WW1) you would also need to take into account the time that the torpedo takes to accelerate to its cruise speed, because you will have had the firing triangel set up with that speed, and the overall average of the run will in fact be slightly slower.

And since the distance to accelerate to full speed remains constant regardless of target distance, the average speed will vary with target distance.

Fortunately this means that the effect is greater on short range shots where you have greater error tolerance than it is on long range shots. This ONE TIME geometry finally decides to work for us instead of against us.