Celestial Navigation Revisited
 

Returning to Silent Hunter 3 after a long absence...happy to see many of the old Kaleuns...:arrgh!:

I am interested in the old problem of celestial navigation within Silent Hunter. I am familiar with the important work done by vanjast on RealNav and the excellent Stellarium workaround advocated by don1reed. I am familiar with the old problems...with the clock, the sun, the moon, the lazy helmsman, and the tantalizing accuracy of the stars...

What I want to know is if the state of the art has been advanced since then? Have any of the old problems been solved in SH3 or SH4 or SH5?

I have a special interest in knowing this, because before I left back in 2009, I was working on my own celestial navigation mod. I focused only on understanding navigation by the stars since others seemed to be working on the sun, dead reckoning, hiding the boat on the map. It was very difficult, but I was finally able to fix a position by the stars to within 120 km...or was it 30 km :doh:...nearly all the time...and at the time that seemed pretty good to me.

I never finished my mod because it was more work than I was capable of accomplishing at the time. Anyway, I don't have my old files anymore :x but I remember what I did. And I feel like doing it again. But not if someone has already done better.

I haven't read anything really exciting recently, all I remember is getting good latitude but no longitude :O: I'm interested in another celestial navigation mod!

For SH5 you'll want to look into my Real Nav mod for my UIs mod :up: Being that SH5 let's you use Python to extend the game we can do many things that the other series only dream of.

vanjanst created a RealNavMod some time ago. It was pretty great. However it interferred with my MagUI and other mods because it changed the map so much. So i did not end up using it,

However now that ive started again and swore to myself to keep my Sh3 Modification simple im definately going to give this mod another whirl. Great for realism.

http://www.subsim.com/radioroom/show...ht=RealNav+Mod

There is the link and OG post.

Ack i forgot the damn RealNavMod baiscally only works with vanilla SH3. NYGM cant run it and i know GWX cant run it either. Lot of MENU ID xxxxx errors come up. However the game still loads after you close all the errors, but a few things are wrong and im sure its not running at full efficiency. Wish there was a way to get it to work. Its such a wonderful mod and the tutorials are great.

@TorpLos

I mentioned vanjast's RealNav in my first post. I think it was the best in-game effort at navigation. All mods risk interfering with each other because they try to modify the same game element or affect something the other mod relies on. I will try to show all my work so somebody can create their customized version that works with their soup of mods.

@postalbyke

It seems the state of the art hasn't advanced much and the old problems have not been solved. I will be solving a few of them which will improve things quite a bit.

@TheDarkWraith

:o Cool, huge, mod. SH5 does seem a different beast. I don't own SH4 or SH5 so I will be working here on SH3. I would appreciate your feedback if I find something that benefits the SH5 users. I read some of the threads, and my impression was that the star shots are performed out-of-game with Stellarium?

:rock:Good luck Bravo im excited for a good NaVmod that works well with MegaMods!

While I love the idea of getting your own coordinates I also love the ide of a navigator.Could you possibly make it so that there is no map position shown when the skies are cloudy?

@ TorpLos

"Some assembly required." I will be releasing my findings and tools in pieces so others can use as they see fit. Hopefully there will be enough success to make it worthwhile for the megamod authors to do the work for me...:har:.

@ Wolfstriked

I will do my best, but RealNav should already have that. My early work will require all the stock tools to help in testing and development. We will see where it leads...:ping:

I am proud to share with you the previously unknown voyages of U-boat U-13 "der Seeesel" captained by Herman Engel and his navigation challenged officers. These prewar voyages were to develop Navigation techniques and this first voyage was to test out the super secret SH3 clocks...

Voyage 1

Here are some observations about the two clocks in SH3. I call them "Screen Time" and "Mouse Time", but most of us think of them as Greenwich Mean Time and Local Time. There is also a "Mouse Date". There is no mention of what these really are in the manual. The game seems to think "Mouse Time" is the local time, though.

It's important to have a clear understanding of time and position if we are going to navigate. So I made a mod that would show the time zone lines and the Lat/Lon coordinates. Then I made some missions that took me to known locations so I could check things out. Let's take a look at what I am talking about...

http://img442.imageshack.us/img442/37/mission1.png

The mod uses Nautical Time Zones, which are the negative of Civil Time Zones. You can read more about it here http://en.wikipedia.org/wiki/Nautical_time. The important thing to remember is that GMT=T+TZ where T is the local time (Zone Time) in the time zone, and TZ is the nautical time zone description (a number from -12 to +12). Time Zones are 15° wide of course and TZ 00 straddles the Greenwich Meridian.

Here's what we can see already by looking at some screen shots of the clocks when I loaded the missions. The mission editor determines the Screen Time and the Mouse Date. Right off the bat it looks like Mouse Time is the Zone Time (Local Time in the zone) and that the Screen Time is the GMT. And the mouse shows us the Date as well...but which date?

First we will sit and wait a day at a fixed location and observe how the times and date change.

http://img840.imageshack.us/img840/4438/mission1a.png

Notice that the Mouse Date did not increment when the Mouse Time rolled over. If Mouse Time is the Local Time, then the Mouse Date is not the local date.

The Mouse Date does increment when the Screen Time rolls over. So the Mouse Date belongs to the same Time Zone as the Screen Time. If the Screen Time is the GMT, then the Mouse Date is the Date of TZ 00.

So the current theory is that the screen gives us the GMT and the mouse gives us the local time and the GMT date.

Now lets cross over the Time Zone lines and observe how the clocks change. We can also do some time-speed-distance calculations to see if we believe the clocks are accurate.

http://img835.imageshack.us/img835/53/mission1c.png

http://img542.imageshack.us/img542/8116/mission1b.png

If you squint you can see the 1° grid dots on the map. The spacing is 120 km and the Screen Time measures the same in both directions and is unaffected by the Time Zone line located midjourney. TSD gives 14.2 knots consistent with the screen shots.

The Mouse Time jumped +1 hours when crossing East and jumped -1 hours when crossing West. It is behaving like a local time should. And it matches what we observe above when the game loads the missions.

So far our best theory is the Screen Time is the GMT and the Mouse Time is the Zone Time and the Mouse Date is the Date of GMT. We can use Time-Speed-Distance and perform Dead Reckoning. Most of us know that the SH3 world is flat and 1° is 120 km and the shortest distance between two points is a line.

There seems to be nothing wrong with this theory of the game clocks...until you attempt celestial navigation.

We are going to compare some calculated and observed sunrise times. The calculated times came from some software, but you can get similar results from this website http://aa.usno.navy.mil/data/docs/RS_OneDay.php. The values are calculated for the positions shown, and are expressed as GMT and as the local Zone Time. Also shown are screen shots of the sub's clocks at each position so you can compare GMT, Zone Time, Mouse Time, and Screen Time.

The sun appears to come from the East of course, so the GMT when it arrives becomes later as your location is chosen further to the West. The calculated GMT's show that nicely in the Table. So we should also expect one of the sub's clocks to show this same trend. ALLLLLLLLAAAAAAAAARRRRRRRMMMMMM!!!

http://img339.imageshack.us/img339/6493/mission1d.png

The Screen Time was our hope for being the GMT, but it doesn't show the right trend and it gives plainly wrong answers outside of TZ 00. The Mouse time is even worse.

Now look at the calculated Zone Times for the sunrises. These Zone Times are the local times at the observing positions. The times are simply the GMT - TZ as you would expect from the equation I have plastered over every screen shot.

Looking again at the screenshots of the sub's clocks, we see that the Screen Time now appears to be the local Zone Time! What seemed so certain from before now appears to be completely wrong! How could the Screen Time be the GMT of the sea and the local Zone Time of the sky?

Let's check the sunset times to be sure...

http://img35.imageshack.us/img35/4692/mission1e.png

Ugh...same story.

Again, because the sun appears to come from the East, the GMT of a sunset becomes later as the observing position moves to the West. Again, we see this trend nicely in the calculated sunsets. And again, the screen shots of the clocks from our sub show that neither one is the GMT we seek. Instead, the Screen Time is actually the local Zone Time as before.

It is nice that we get good agreement between one of our clock's and the local sunrise and sunsets. It is nice that one of the clocks matches closely with observed celestial evens. But we need the GMT if we are to calculate the positions of stars used for celestial navigation.

Let's see if we can figure this out...

To recap, we had a perfectly good theory of how the game clocks work that is consistent with stuff happening on the sea. The Screen Time played the role of GMT that stays constant across Time Zone lines. The Mouse Time jumped around like a good local time should.

But the theory fell apart when we tried to use if for stuff happening in the sky. So we need a new theory for the sky. This time the Screen Time plays the role of Local Time because it agrees with the observations. But the Mouse Time isn't the GMT either, and appears worse than useless at first glance.

But look back at the observations of how the Mouse Time jumps when we cross a Time Zone Line. Take a good look at the Time Zone, the Mouse Hour, and the Screen Hour. There is a relationship there.

Let the Mouse Time be hhmm
Let the Screen Time be HHMM

In each instance the known TZ is equal to HH-hh. If you get a TZ outside of +/- 12 just add or subtract 24.

That means if we didn't know the TZ, we could look at the time difference between the two clocks and calculate it. And that means we could add it to the Screen Time and get the GMT necessary for celestial navigation.

Let's see what that new theory might look like for some observations in the 3 time zones. First the sunrises, and then the sunsets.

http://img215.imageshack.us/img215/9527/mission1h.png

http://img16.imageshack.us/img16/2591/mission1g.png

Let's check the observation at 0°N 8°W first.

At sunrise you see 05:39 on your Screen Clock, so you call this the local Zone Time T. You note that the TZ is HH-04 which means 05-04=+01. So you are West of the Greenwich Meridian. You add the Time and the Time Zone to get the GMT of 06:39 which compares with the calculated SR of 06:36 GMT.

At sunset you have T= 17:39 and TZ = 17-16 = +01 and GMT = 17:39 + 01 = 18:39, which compares with 18:43.

Now let's check the observation at 0°N 8°E.

At sunrise you get T= 06:35 and TZ = 06-07 = -01. This gives the GMT = 06:35 - 01 = 05:35 which compares favorably with the 05:32 of the table. At sunset we get a T=18:34 and TZ = 18-19 = -01 and GMT = 17:34 compared to 17:39.

The theory seems to be working, but only time will tell :har:

What does it all mean?

It means the SH3 Gods Must Be Crazy. It seems that the developers created one time system for the sea and one for the sky. Hopefully I'm wrong and this is all just a terrible mistake.

There is no reason not to continue using the Screen Time as the GMT for all the usual time non celestial navigation duties. In fact, there is no way that the Screen Time could really be a local Zone Time because the Screen Time doesn't change when you cross Time Zone boundaries. It isn't realistic, but it seems to work for the observations of the sky above, and might hold up to more observations by interested Kaleuns. Those who do so should report their observations here. I am interested in hearing your experiences.

I will post links to the tools I used in this first voyage after U-13 returns to port. :arrgh!:

Have you tested the compatibility with the mega mods?