Sub physics - experts needed - Page 9

Dr.Sid · 12-12-07, 03:14 AM

I wonder what kind of problem the exposed screw is. Except for noise I mean. The dramatic loss of drag could be dangerous for the machinery. On the other I think it was thought about, and maybe the screw has rpm based control loop, so it would just cut off the power.

Anyway this actually can be done in my sim already, I mean moving backwards and feeling the controls reacting in opposite manner.

XabbaRus · 12-12-07, 03:37 PM

I'd imagine that you would get a spike in RPM though what systems are there to prevent the shaft shaking to bits I don't know.

I wonder if that level of fidelity is needed though. Syaing that everything but the kitchen sink is there at the mo.

Just got your e-mail

Should be sorted the vepr_2.3ds loads in max fine

Bubblehead Nuke · 12-12-07, 07:45 PM

Quote:

Originally Posted by Dr.Sid

On the other I think it was thought about, and maybe the screw has rpm based control loop, so it would just cut off the power.

Correct in this assumption. The throttles will get a overspeed feedback that will limit the shaft RPM.

To be · 12-12-07, 07:49 PM

Quote:

Originally Posted by Dr.Sid

I wonder what kind of problem the exposed screw is. Except for noise I mean. The dramatic loss of drag could be dangerous for the machinery. On the other I think it was thought about, and maybe the screw has rpm based control loop, so it would just cut off the power.

Anyway this actually can be done in my sim already, I mean moving backwards and feeling the controls reacting in opposite manner.

I am talking about the 70deg angle though.

Dr.Sid · 12-13-07, 06:07 PM

Btw. I opened discussion about map data on general board.

http://www.subsim.com/radioroom/showthread.php?t=126877

caymanlee · 12-13-07, 11:58 PM

Quote:

Originally Posted by Dr.Sid

Btw. I opened discussion about map data on general board.

http://www.subsim.com/radioroom/showthread.php?t=126877

Dr.Sid, do you plan to set up a higher solution of the map in your sim? ---wider(even global??), more detail(say 10m??),
it's the most challenging and charming goal in my mind, haven't figure out how to start it, any tips??

Dr.Sid · 12-14-07, 05:00 AM

There are several decisions which I didn't make yet.

1) limited mission area vs. free mission area. Yes .. we all want the second. But there really is a lot of complications. Game itself will be fun even on limited area (as DW is).

2) Real data or not. With limited area we could even do without actual world data. It has little effects on actual gameplay, especially multiplayer.

3) Rounded 3d model of ground or not. Earth is round. Sim needs to reflect it. With limited mission area flat sea again would simplify many things. Corrections would have to be done in sensors like radar, sure. I think DW has flat see in fact.

4) Resolution of real world data. If we have real world, what resolution do we need ? Size is the problem. Even if we get 10m resolution maps, that would be HUGE. I guess we would not have to go under 1km, with possible interpolation, which is what DW use. This is also questions of what your hardware can render. 100m squares with 20 km visibility is quite simple to do even without any smart methods. 10m squares call for very smart methods.

At the moment, (2) seems to answer to 'real world, sure'. I have some data. (4) at the moment says 100m rendered squares based on 3km real world squares. Easy to do (tested). (3) and (1) still not decided. I'm trying different mesh division/update/mapping methods, there really is many options, many problems. On the other hand it's quite fun. Will play more and decide later.
For really rounded earth also simulation of physics will get a bit complicated (up will not always be the same direction). But that is a challenge too and to have it entirely correct is nice thing to have.

Best would be to have non-limited mesh with 10m squares, based on real data with variable resolution (better where available), with correctly round globe. I'm working on THIS. But sometimes I just ask myself 'do we really need it ?'.

Molon Labe · 12-14-07, 05:56 AM

Is it possible to have relatively low-resolution mapping for most of the world but to have greater detail near ports and coastlines, where it matters a bit more? Getting underwater canyons that are 10000 feet deep modeled in high detail isn't all that important when the subs can't get much beyond 1500.

Dr.Sid · 12-14-07, 06:11 AM

Yes .. it is possible. But not so easy. Anyway it seems we will have to use several data sources anyway. All maps I found use WGS84 and meters for height, so they should go well together. I will test it very soon, since I already have etopo with 2 degree resolution and gtopo30 with 0.5 degree resolution. But gtopo does not have bathymetry ! It also is split into tiles. I have to match the maps and merge them.
With fixed grid where new maps are aligned with the grid variable resolution should be simple.
Good luck I work in GIS company for 7 years now

XabbaRus · 12-15-07, 06:46 AM

I think round earth.

XabbaRus · 12-18-07, 05:25 PM

Sid any news?

How are the models holding up? I have a modified Yankee almost ready to hit the stocks, moving tail surfaces and opening missile doors

Dr.Sid · 12-18-07, 07:27 PM

I'm fighting the maps. I'm also quite busy at the work.
Don't expect every-day updates all the time

To be · 12-18-07, 08:34 PM

Too late for that, you've spoiled us now. We will be expecting hourly updates of course.

Zachstar · 12-19-07, 05:53 AM

When dealing with rounded earth you will have to deal with the various problems of 2D to 3D So the mapping of the SRTM data or whatever to it without major problems is unlikely.

The folks at Celestia seem to have that figured out tho. Perhaps you can see how they did it?

To be honest I think a rounded earth is a waste of time. You are not going to be simulating aircraft flying high enough or other factors that desperately need a rounded earth.

The data in my links in the other topic is supposed to be laid out flat if I remember correctly.

I have a small concern here. You seem to call this sim a community subsim. Does this mean that you will fully open up development through some sort of version management? I hope you will allow others to work on the code.

Dr.Sid · 12-19-07, 06:47 AM

I don't plan to open the sources at the moment. I believe the core should be made by one man. It's going to be pretty modable and you wont need any coding form most things. Something like Orbiter, except with better modding support.

You practically need round earth when you allow limitless map. You can't map whole round Earth on plane. You can map small area, but as it gets larger you get more and more distortion.

There also might be ballistic missiles, where Earth roundness has major effects.

Anyway the roundness itself seem to bring much less problems then the free limitless map itself.

12-12-07, 03:37 PM	#122
XabbaRus Navy Seal Join Date: Sep 2001 Posts: 5,330 Downloads: 5 Uploads: 0	I'd imagine that you would get a spike in RPM though what systems are there to prevent the shaft shaking to bits I don't know. I wonder if that level of fidelity is needed though. Syaing that everything but the kitchen sink is there at the mo. Just got your e-mail Should be sorted the vepr_2.3ds loads in max fine __________________

12-14-07, 05:56 AM	#128
Molon Labe Silent Hunter Join Date: Jun 2004 Location: Along the Watchtower Posts: 3,810 Downloads: 27 Uploads: 5	Is it possible to have relatively low-resolution mapping for most of the world but to have greater detail near ports and coastlines, where it matters a bit more? Getting underwater canyons that are 10000 feet deep modeled in high detail isn't all that important when the subs can't get much beyond 1500. __________________

12-15-07, 06:46 AM	#130
XabbaRus Navy Seal Join Date: Sep 2001 Posts: 5,330 Downloads: 5 Uploads: 0	I think round earth. __________________

12-18-07, 05:25 PM	#131
XabbaRus Navy Seal Join Date: Sep 2001 Posts: 5,330 Downloads: 5 Uploads: 0	Sid any news? How are the models holding up? I have a modified Yankee almost ready to hit the stocks, moving tail surfaces and opening missile doors __________________

12-19-07, 05:53 AM	#134
Zachstar Sea Lord Join Date: Feb 2005 Location: Shreveport, Louisiana Posts: 1,956 Downloads: 13 Uploads: 0	When dealing with rounded earth you will have to deal with the various problems of 2D to 3D So the mapping of the SRTM data or whatever to it without major problems is unlikely. The folks at Celestia seem to have that figured out tho. Perhaps you can see how they did it? To be honest I think a rounded earth is a waste of time. You are not going to be simulating aircraft flying high enough or other factors that desperately need a rounded earth. The data in my links in the other topic is supposed to be laid out flat if I remember correctly. I have a small concern here. You seem to call this sim a community subsim. Does this mean that you will fully open up development through some sort of version management? I hope you will allow others to work on the code. __________________

12-12-07, 03:14 AM	#121
Dr.Sid The Old Man Join Date: May 2005 Location: Czech Republic Posts: 1,458 Downloads: 6 Uploads: 0	I wonder what kind of problem the exposed screw is. Except for noise I mean. The dramatic loss of drag could be dangerous for the machinery. On the other I think it was thought about, and maybe the screw has rpm based control loop, so it would just cut off the power. Anyway this actually can be done in my sim already, I mean moving backwards and feeling the controls reacting in opposite manner.

12-13-07, 06:07 PM	#125
Dr.Sid The Old Man Join Date: May 2005 Location: Czech Republic Posts: 1,458 Downloads: 6 Uploads: 0	Btw. I opened discussion about map data on general board. http://www.subsim.com/radioroom/showthread.php?t=126877

12-14-07, 05:00 AM	#127
Dr.Sid The Old Man Join Date: May 2005 Location: Czech Republic Posts: 1,458 Downloads: 6 Uploads: 0	There are several decisions which I didn't make yet. 1) limited mission area vs. free mission area. Yes .. we all want the second. But there really is a lot of complications. Game itself will be fun even on limited area (as DW is). 2) Real data or not. With limited area we could even do without actual world data. It has little effects on actual gameplay, especially multiplayer. 3) Rounded 3d model of ground or not. Earth is round. Sim needs to reflect it. With limited mission area flat sea again would simplify many things. Corrections would have to be done in sensors like radar, sure. I think DW has flat see in fact. 4) Resolution of real world data. If we have real world, what resolution do we need ? Size is the problem. Even if we get 10m resolution maps, that would be HUGE. I guess we would not have to go under 1km, with possible interpolation, which is what DW use. This is also questions of what your hardware can render. 100m squares with 20 km visibility is quite simple to do even without any smart methods. 10m squares call for very smart methods. At the moment, (2) seems to answer to 'real world, sure'. I have some data. (4) at the moment says 100m rendered squares based on 3km real world squares. Easy to do (tested). (3) and (1) still not decided. I'm trying different mesh division/update/mapping methods, there really is many options, many problems. On the other hand it's quite fun. Will play more and decide later. For really rounded earth also simulation of physics will get a bit complicated (up will not always be the same direction). But that is a challenge too and to have it entirely correct is nice thing to have. Best would be to have non-limited mesh with 10m squares, based on real data with variable resolution (better where available), with correctly round globe. I'm working on THIS. But sometimes I just ask myself 'do we really need it ?'.

12-14-07, 06:11 AM	#129
Dr.Sid The Old Man Join Date: May 2005 Location: Czech Republic Posts: 1,458 Downloads: 6 Uploads: 0	Yes .. it is possible. But not so easy. Anyway it seems we will have to use several data sources anyway. All maps I found use WGS84 and meters for height, so they should go well together. I will test it very soon, since I already have etopo with 2 degree resolution and gtopo30 with 0.5 degree resolution. But gtopo does not have bathymetry ! It also is split into tiles. I have to match the maps and merge them. With fixed grid where new maps are aligned with the grid variable resolution should be simple. Good luck I work in GIS company for 7 years now

12-18-07, 07:27 PM	#132
Dr.Sid The Old Man Join Date: May 2005 Location: Czech Republic Posts: 1,458 Downloads: 6 Uploads: 0	I'm fighting the maps. I'm also quite busy at the work. Don't expect every-day updates all the time

12-18-07, 08:34 PM	#133
To be Electrician's Mate Join Date: Feb 2007 Posts: 140 Downloads: 8 Uploads: 0	Too late for that, you've spoiled us now. We will be expecting hourly updates of course.

12-19-07, 06:47 AM	#135
Dr.Sid The Old Man Join Date: May 2005 Location: Czech Republic Posts: 1,458 Downloads: 6 Uploads: 0	I don't plan to open the sources at the moment. I believe the core should be made by one man. It's going to be pretty modable and you wont need any coding form most things. Something like Orbiter, except with better modding support. You practically need round earth when you allow limitless map. You can't map whole round Earth on plane. You can map small area, but as it gets larger you get more and more distortion. There also might be ballistic missiles, where Earth roundness has major effects. Anyway the roundness itself seem to bring much less problems then the free limitless map itself.