Celestial Navigation Revisited

[11Bravo]

By manipulating the Elevation Min and taking screenshots and counting pixels, a calibration curve can be developed for the cameras. This is necessary if an accurate sextant is to be developed.








Notice that both curves are the same shape, but differ slightly in the magnitude of the correction. This seems to be a feature of how the rendering engine displays external views in SH3. Very interesting...It makes you wonder if a similar distortion exists in the horizontal direction for bearings.

That is all for today. Next up will be the actual sextants and measuring the position of some navigation stars.

Good hunting.

<Edit> Reposted calibration curves after correcting a math error. Sorry.

[reaper7]

Hi mate, can you check your email. Regarding the above, cheers.

[11Bravo]

Thanks!

[11Bravo]

I have completed the two sextants and tested one of them. The star positions in SH3 are accurate to my ability to measure them. The camera calibration curve appears to be accurate enough for our purposes.

I am showing a linear sextant here, where you have to apply the altitude correction manually, similar to a standard sextant.



The chart on the left is an altitude correction that needs to be applied to the sextant measurement Hs. It was developed from the calibration curve, and the screen altitude corrections mapped to sextant altitudes. The Elevation Max and Min of the cameras.dat file were selected in a way that tilting the camera to a hard stop at Min or Max allows you to measure sky altitudes directly. If you are tilted all the way down, you use the left side scale. If you are tilted all the way up, you use the right side scale.

Here we have tilted all the way up and put Alioth in the center of the sextant. Alioth is one of the 58 navigation stars commonly used in celestial navigation. It is easily recognized as the 3rd star in the handle of the Big Dipper constellation. Our linear sextant reads 78°30' to the nearest 10' after some squinting at SH3's low resolution interface.

You might want to zoom in on the image above. Try holding down your CTRL key and hitting + a few times. When you are done, use CTRL and 0 to return to normal. Works with many browsers.

The altitude correction factor is 41' from the table. You might notice that this sky altitude of 78° is actually a screen altitude of 36°. Tilting the camera doesn't change the screen altitude, only the sky altitude. That is why we developed the calibration curve for screen altitude.

Also shown are almanac values for the navigation stars visible from the location at the time shown by the clock. Alioth should have an altitude of 79°04' and our observed value is 79°11' after the altitude correction is applied. The difference in this case is a mere '7, or 0.117°, or 14 km in position on planet SH3. Way to go SH3!!!

[brett25]

wow pretty awesome, ive always toyed with the idea of trying real navigation is sh3 and i think i will do it now in earnest using your new tools, good job, mate

[11Bravo]

Thanks, Kaleun!

[11Bravo]

23 navigation stars later... and here are the results.

The Diff column is after the Altitude correction is applied. I have noticed that all the differences are positive. Usually an accurate sextant would give small errors distributed around zero, but ours is distributed around 13'. This qualifies as an index correction that can be subtracted from all measurements with this sextant to make it more accurate. Real sextants have index measurements arising from their mechanical construction. I'm guessing mine might be a function of my calibration system, sextant construction, and measurement bias when estimating Hs. Your mileage might vary, but I will consider using it going forward.

I have included a sample calculation that shows how to apply the Index correction if you choose to do so. Note that the average error is now less than 5' or 10 km. Well within my specification of 30 km so far. Time for something to go horribly wrong.

[11Bravo]

Where are we?

A good question to ask in a thread about navigation.

I guess the big news is that the stars are accurately modeled in SH3. It would have been nice if the developers had told us that, instead of finding out 7 years later. It would have been noticed earlier except for two things...the clocks are not easily understood and the external camera views have a distortion that introduces errors in the vertical (and perhaps horizontal) angles. Also the navigation map doesn't make locating your latitude, longitude, or time zone easy. I will continue to try to overcome these unnecessary barriers in the game.

(Another reason is that I didn't have the courage to report my incomplete results of several years ago. I have resolved now to report my results and my ideas as I go forward to atone for my poor judgement on this matter in the past)

It took a lot of testing to find this. And further testing is needed to find the limits of this method. I have no doubt there will be some serious limitations to this method. My immediate plan is to qualify the flak gun sextant in the same way I showed above. After that I will demonstrate the method of celestial navigation to yield an error triangle and measured position. (I showed one earlier in the thread) Then I will fix up the previously released mini mods and release what I have so far as another download. That will conclude the second voyage of U-13 "der Seeesel".

After that, I want to tackle the sun. And then the moon. And continue to test the nautical chart, nautical clock, the sextants, time speed distance measurements, study some morse code, learn the enigma machine, and if I find some time, I might even attempt to sink a ship.

If anyone is interested in hundreds of loading test missions, making detailed measurements, and reporting the results privately, feel free to PM me.

[reaper7]

Brilliant, the Devs created a proper star system. Well done on proving this.
Not that we know that, I shall move forward with coding the clocks.
Did you get to test my external code for time zones?

[11Bravo]

In process, I got your email. Thanks again for all the hard work.

[Wolfstriked]

Thank You 11Bravo

[reaper7]

Hi 11bravo, just a quick question on the star positions. Are they appearing at the correct positions for the actual date, years etc or are they offset and a conversion is required to there positions to calculate where they should be from almac data?

Even if offset does the sky model in sh3 reflect real world position and dates?
I guess what I'm asking is would the stars in the sky on a certain date and position in the globe look the same to the player as it would to a person if they were looking at the sky at that time in real-life, that would be amazing for sh3 if it were the case.


EDIT
Well some good news I've now got the code in place and working that automatically changes the Screen Clock to the Correct GMT time.
That is the BZT is now taken from the Time:

Stock SH3:
GMT = T+TZ+BTZ

Coded SH3:
GMT = T+TZ

Unfortunately I haven't found a way yet to change the Mouse-over Local Time as it is not a Numerical Value like most values but is in fact the Text entry from the en_menu.txt ;2703=|Local time: %02d:%02d

where %02d:%02d updates the text values.

Will see if I can do a workaround, if not the Modded Map that 11Bravo made can be used to see the timezone for localtime.

[vanjast]

Originally Posted by reaper7

Hi 11bravo, just a quick question on the star positions. Are they appearing at the correct positions for the actual date, years etc or are they offset and a conversion is required to there positions to calculate where they should be from almac data?

Even if offset does the sky model in sh3 reflect real world position and dates?
I guess what I'm asking is would the stars in the sky on a certain date and position in the globe look the same to the player as it would to a person if they were looking at the sky at that time in real-life, that would be amazing for sh3 if it were the case.

From what I remember the stars were accurate 'enough'... To which date I'm not sure, but this shouldn't be a problem as you use the almanac for that year.

The optics of the cameras were difficult to work with but it looks like it was a SIN (COS) function of sorts. If you play with the camera magnifications you'll fook the POV angles. I looked at trying to extract and display the elevation angle of the scopes, but this didn't happen. If you can do this you'll should get good accuracy.

[postalbyke]

This is totally sweet! I usually get too wrapped up in playing the game to "seriously" mod :P
I'm still working on completing my first mod lol

I did some work on celnav a while ago, and my sky always seemed to show up at GMT=local (I haven't followed what all has been said about the clocks, I hope that Reaper has "fixed" that little oversight)
I shot all my elevations using the observation periscope after discovering that I couldn't get my MaGui sextant to work (I think my screen buggers the angularity somehow...)(or it's just that pesky 5:4 v 4:3 problem)

Either way, I'm really stoked about your inventiveness and investment of time on this project!

[reaper7]

Originally Posted by vanjast

From what I remember the stars were accurate 'enough'... To which date I'm not sure, but this shouldn't be a problem as you use the almanac for that year.

Good, good so the stars are pretty accurate for a certain date/year.
Cheers.

Originally Posted by postalbyke

I did some work on celnav a while ago, and my sky always seemed to show up at GMT=local (I haven't followed what all has been said about the clocks, I hope that Reaper has "fixed" that little oversight)

Hi mate, I can send you a test file since your fimilar with the GMT=GMT+local time issue in sh3. You can test to see if my fix is working as intended.
Cheers mate.