Sub physics - experts needed

[Bubblehead Nuke]

Originally Posted by MarkShot

I just got finished reading about all the air services on a fleet boat. Next to electrical power; nothing was as critical and all encompassing as air services. In fact, it appears that air drive was favored over electrical when there was choice due to simplicity, reliability, and the low impact nature thus reducing need for maintenance.

The nice thing about air is that it is EASY to store and there is a huge amount of potential energy in it at high pressures. It takes nothing to KEEP it stored. When all else fails, air WILL flow when you open the valve. The only bad thing is you have to keep the moisture out of it, otherwise you can freeze the pipes solid.

As you stated, with reducers you only have to have one main system pressure and then step it down to the pressure that you need. We even used it to breathe in emergencies by hooking into it with what we called EAB's (Emergency Air Breathers) that had a demand regulator that you attached to your belt. We just called it 'sucking rubber'

The one BAD thing about using air is that it exhausts into the pressure hull. Over time this raises the pressure in the boat and you have to run air compressors to put it back in the tanks.

It can make for some pretty interesting atmospheric conditions in the boat as you change air densities and such. I have been in conditions where the air pressure is the same as the dead sea, yet the O2 content (partial pressure) is approaching the summit of Mount Everest. A butane cigarette lighter could NOT light due to the low O2 content

[sonar732]

Originally Posted by Bubblehead Nuke

We even used it to breathe in emergencies by hooking into it with what we called EAB's (Emergency Air Breathers) that had a demand regulator that you attached to your belt. We just called it 'sucking rubber'

Oh the memories of walking around the boat with an EAB during a drill having your thumb on the nozzle and feeling non-skid on the deck since you had the shower cap covering your face.

[Dr.Sid]

Here is little funny demo:

http://roger.questions.cz/other/sim5.zip

It shows physics model adapted to spherical earth. The sphere is made quite small so you can enjoy all the funny effects. There are also some unrealistic effects, because the sphere is so small, like virtual pitching up while moving forward, which affects many things, like depth autopilot for example (it can't hold desired depth).
Anyway all the effects now work on any position on the globe, taking changing direction of gravity into account.
There is another problem with the view orientation. Now the view is oriented by local heading. So as you approach pole (there is flag on the north pole), the view will rotate, because heading changes (even if you go straight). Also the camera is turned so 'up' is always away from the globe center, so if you try to go through the globe and to surface on the other side, the view will twist too as you pass the center.
Btw. the globe has 1000 m diameter, depth is in feet now.

There are also some minor (and some even old) features demonstrated here. GUI elements now can have textures, and I've also changed the setup so it demonstrates how GUI can adapt to different aspects of the window. There are 2 groups of control - fixed aspect and free aspect. Fixed aspect controls will be control panels which have to fit to underlying texture. Free aspect controls are textboxes and 3d viewports which can benefit from wider display and does not need to fit underlaying texture. So in this version 3d view is free (well, limited to 2:1) and the rest is fixed.
You can also use F1 and F2 to switch control group with another view, this is simply to test controls group visibility and also to test 2 viewports at once.

Here's an article you might find of some interest: http://www.geocities.com/Pentagon/Qu...8/deepdive.htm

[Dr.Sid]

Originally Posted by Albrecht Von Hesse

Here's an article you might find of some interest: http://www.geocities.com/Pentagon/Qu...8/deepdive.htm

Good reading !

[Bubblehead Nuke]

I like it Dr Sid. Is it me or did you reduce helm response times at lower speeds, or you increased it at the higher speeds? It seems different now somehow.

I take it the sphere size is easily scaleable to real world size?

[Dr.Sid]

Originally Posted by Bubblehead Nuke

I like it Dr Sid. Is it me or did you reduce helm response times at lower speeds, or you increased it at the higher speeds? It seems different now somehow.

I take it the sphere size is easily scaleable to real world size?

Hm .. I lowered dynamic stability a bit, so the sub stays more or less horizontal after emergency surface. That would affect turning too.
Sure the sphere is easy to change. Anyway this looks quite funny, I wanted to share it

Edit: Oh my .. it's easy to change the sphere .. but it's a big problem if you enlarge it to real 6378km diameter. Then you get quite on the edge of opengl precision (works in single floats). I have some ideas how to solve it, but I thought I won't need them. For a moment you play with some visuals and bang, you have to think hard again :-D

[Bubblehead Nuke]

Originally Posted by Dr.Sid

Hm .. I lowered dynamic stability a bit, so the sub stays more or less horizontal after emergency surface. That would affect turning too.
Sure the sphere is easy to change. Anyway this looks quite funny, I wanted to share it

Yes, that would explain the changes I am seeing. A 688 has a lot of longitudinal stability but it is in the roll that it is kind of unstable. I like the 'feeling' of what you did.

Is there any way for me to change these value??? or are they hardcoded into the sim? I am just curious as to the ratios and such.

Edit: Oh my .. it's easy to change the sphere .. but it's a big problem if you enlarge it to real 6378km diameter. Then you get quite on the edge of opengl precision (works in single floats). I have some ideas how to solve it, but I thought I won't need them. For a moment you play with some visuals and bang, you have to think hard again :-D

mmm.. your diameter seems a bit small, unless you were talking about radius of the sphere.

I went and just did a quick google and found this tidbit:

What is the diameter of the earth?

The diameter of the earth at the equator is 7,926.41 miles (12,756.32 kilometers).

But, if you measure the earth through the poles the diameter is a bit shorter - 7,901 miles (12,715.43 km). This the earth is a tad wider (25 miles / 41 km) than it is tall, giving it a slight bulge at the equator. This shape is known as an ellipsoid or more properly, geoid (earth-like).

--------

You have just got to love the internet eh?

Originally Posted by Dr.Sid

Good reading !

Thanks! I was hoping you (and others) might enjoy it. While it might not be giving you the info you wanted, I was hoping it might give some insight into what systems were used, how they were used, and how they performed, especially under such an extreme condition.

[Dr.Sid]

Originally Posted by Bubblehead Nuke

Is there any way for me to change these value??? or are they hardcoded into the sim? I am just curious as to the ratios and such.

At the moment everything is hardcoded. There are same basics of 'database' but it is created in the code at the moment. So please wait some more.

Originally Posted by Bubblehead Nuke

mmm.. your diameter seems a bit small, unless you were talking about radius of the sphere.

Yes .. sure .. I'm talking about radius at equator.
The 'ellipsoid' problem is known to me as I work with coordinate systems at work. It brings great complications when computing with latitude and longitude. Latitude and longitude are defined as direction of the horizontal plane - or it's normal. It's because latitude and longitude was first measured by angles of celestial bodies against horizon.
If the Earth was more or less round, the normal of the surface would have exactly same direction as the line from the surface to the center of the globe. Transforming latitude/longitude to 3d coordinates would be matter of few sin and cos.
But as the Earth is elliptical, there is difference. Latitude 30 degrees is not 30 degrees from plane of the equator, but slightly less. And the math of ellipsoids get so complicated that even the simple tasks cannot be solved exactly and numerical iterations and approximations must be used.
Anyway for the sim sphere is perfectly OK.

Btw. I almost solved the float precision problem - it's not that big after all.

[Bubblehead Nuke]

I just cut and pasted the diameter of the eath thing from a website. I hope nobody would expect you to make things real world to that extreme. I know I would not expect it.

Could you imagine the difficulty of that?? Wow that hurts thinking about it. They would probably want you to simulate the variable gravity gradients as well. The extreme ends you would have to go thru for absolute realism, and then put appoximations of ships and subs in it? That makes me howl with laughter thinking of the dichotomy of that kind of sim.

Make is a perfect sphere and you far exceed the expectations that I had about the sim.

[Dr.Sid]

You know .. the gravity gradients affects ballistic missiles ..

[Zachstar]

Just to talk a little about the database stuff.

The classic debate between modding and multiplayer. The one that quake and the half life series proved could work together.

I would love to have the ability to use mods that work well in multiplayer evironments. Say I want to take on the Rebel Hunley with my Kilo with ultracapasitors on mars that suddenly has water.. and Ice!

The best way to alllow this is to follow the example that many simulators have followed. Define ALL nonvisual aspects in a simple config file that remains under a meg and can be easily and safely sent over the connection to the other players for the session as "Custom Vessels" "Custom whatever"

I suggest XML as the choice for this. XML is very VERY easy to edit and can be used for everything from defining local weather to defining how well the fuel in the rocket launched torpedo burns.

You have a selection of quality "Base Vessels" and Base values hardwired into the core for the sake of allowing classical multiplayer oppertnities. These are obviously encryped and sealed with protection to discourage the hopefully few and far between hacking attempts.

Then you have these custom sessions that can import XML or even addon packages to simulate say a blue mars.. or a Hunley with the batteries he always wanted... Or Old Ironsides with a MK-48 launcher and 40s hydrophone equipment. A 747 that acts like a P-3.... A S-Class boat that suddenly has teeth... ETC...

Or to be more serious.. the many classes of ships and submarines that because of time and team constraints cannot be implemented in a base pack.

So to recap...

Multiplayer=Good
Modding=Good
XML=Easy modding, Easy management.
Hardcoded Earth and Vessels= Standardized engagements when wanted.

[Zachstar]

Originally Posted by Dr.Sid

You now .. the gravity gradients affects ballistic missiles ..

I will argue against having nuclear weapons allowable in this simulation later.

For now I just want to say that these things arent that difficult. Ballistic, Orbital, Rocket, etc.. equations are well published and you ought to be able to define a basic orbital model easily.

I think the most useful part about such a few days of work is to allow the simulation of info gathering spy sats. As it is simply a matter of a somewhat realistic simulation of It's orbital path... there is no need for advanced code such as drag effects or reentry.

The ballistic part may matter more here. Bullets and shells follow a ballistic trajectory just like the popcorn you tossed at the screen reading this post. When a starfish is launched and the torpedo is deployed. The torpedo needs to follow a ballistic trajectory based on it's shape while being affected by the air as drag.

If you encounter a battleship and it gives you a broadside to show you who is boss.. There needs to be a good simulation of where the shells will land.

Orbits are simply objects that fall in circles over and over again. Astronauts are "falling" the entire time they mingle about the space station.

[Dr.Sid]

As for the config files and multiplayer, sure, that how I plan it. With small exception. I really don't like XML . I have something more dense and readable, closer to normal language. It's generally name:value form with sections, references, includes and so on. We use this form of configurations where I work and it shows to be very good.

As for the ballistic missiles, it was a joke.