TI-57 Lunar Lander ?

[Gene]

From an email I received...

"I am looking for the lunar lander program for the original TI57. I already found various versions for different handheld calculators in the net, but not for the Ti57.:

Does anyone have a version for the original TI-57 ?

[Didier Lachieze]

Here is one from the French magazine "l'Ordinateur de Poche":

[Thomas Klemm]

I used this TI-57 Emulator to run the game.
That was fun. Thanks for bringing back memories.

Cheers
Thomas

[badaze]

(02-02-2019 06:31 PM)Didier Lachieze Wrote:  Here is one from the French magazine "l'Ordinateur de Poche":

I remember it very well. I still have all issues of this magazine.

[Thomas Klemm]

For a perfect landing you can use 13 litres of fuel in each step:

130: 100.5000
117:  90.4050
104:  80.3200
 91:  70.2450
 78:  60.1800
 65:  50.1250
 52:  40.0800
 39:  30.0450
 26:  20.0200
 13:  10.0050
  0:   0.0000

Cheers
Thomas

[Valentin Albillo]

(02-02-2019 08:46 PM)Thomas Klemm Wrote:  For a perfect landing you can use 13 litres of fuel in each step:

130: 100.5000
117:  90.4050
104:  80.3200
[...]
 26:  20.0200
 13:  10.0050
  0:   0.0000

Funny. I don't know in this (TI-57?) version you're using but in the HP-25 version of ~44 years ago, I quickly discovered that I could land perfectly in just one step.

Needless to say that killed my interest in the game, no improvement possible.

V.
.

[Thomas Klemm]

(02-02-2019 09:31 PM)Valentin Albillo Wrote:  I quickly discovered that I could land perfectly in just one step.

If you burn all the fuel in the first step you go up and free fall:

130:  100.5000
  0: -144.5220
  0: -128.6580
  0: -112.7780
  0:  -96.8820
  0:  -80.9700
  0:  -64.10420
  0:  -48.10980
  0:  -32.11380
  0:  -16.11620
  0:    0.11700
  0:   16.11620
  0:   32.11380
  0:   48.10980
  0:   64.10420
  0:   80.9700
  0:   96.8820
  0:  112.7780
  0:  128.6580
  0:  144.5220
  0:  160.3700
  0:  176.2020
  0:  192.0180

And then you probably die with a speed of 193.

Cheers
Thomas

PS: The program incorrectly displays heights ≥ 10000.
Instead of 16.11620 it would display it as 17.1620.
But the fix is easy: Just replace 4 by 5 in line 24.

[ijabbott]

(02-02-2019 10:32 PM)Thomas Klemm Wrote:  

(02-02-2019 09:31 PM)Valentin Albillo Wrote:  I quickly discovered that I could land perfectly in just one step.

If you burn all the fuel in the first step you go up and free fall:

So burn some of it in a different step, but still only burn in one step. (I haven't tried solving this, so don't know whether the solution exists for the particular set of parameters involved.)

[EdS2]

Hmm, if you can land successfully with a single last-minute burn, you've got a terribly over-powered engine, or equivalently the time step is too coarse. This game needs tuning!

And if you can land gently with a constant thrust all the way, you've got far too much fuel to start with!

Does the game model the decreasing remaining mass of the craft? That's a crucial part of the rocket equation. Otherwise your chemical fuel contains an unrealistic amount of energy.

[Thomas Klemm]

(02-03-2019 08:09 AM)ijabbott Wrote:  So burn some of it in a different step, but still only burn in one step.

If you free fall until the last step before you crash you can use 82 units of fuel to bring down your velocity to 0 but there are still 540 m to go.
Due to gravity you accelerate to 42 m/s on impact.
That's more than the 5 m/s which is considered a smooth landing.

Cheers
Thomas

[Thomas Klemm]

(02-03-2019 09:26 AM)EdS2 Wrote:  Does the game model the decreasing remaining mass of the craft?

From the article:

Quote:La simulation n'est pas parfaite; en particulier le programme ne tient aucun compte du fait que votre véhicule s'allège lorsqu'il consume du carburant pour freiner.

The simulation is not perfect; in particular the program ignores the fact that your vehicle gets lighter when it consumes fuel to brake.

Quote:This game needs tuning!

Feel free to do so. The only problem I see is that all 50 steps are already used.

Cheers
Thomas

[EdS2]

(02-03-2019 11:35 AM)Thomas Klemm Wrote:  

Quote:This game needs tuning!

Feel free to do so. The only problem I see is that all 50 steps are already used.

That certainly makes it a challenge! I've a feeling I did have a lunar lander simulation for my TI-57 but I can't imagine where I would have got it. It's not in the manual. The various now-ancient Basic lunar lander programs tend to be more than 100 lines, but then some of that is user interface. I notice the 35-step Sinclair Cambridge Programmable was advertised as "It even plays a lunar landing game!" but I haven't found the program.

Edit: off topic too, but here's the HP-25 official lunar lander. Again, no modelling of mass decrease.

[Thomas Klemm]

(02-04-2019 12:17 PM)EdS2 Wrote:  here's the HP-25 official lunar lander.

(02-02-2019 09:31 PM)Valentin Albillo Wrote:  I quickly discovered that I could land perfectly in just one step.

Now I wonder if you still remember how you did that.

Cheers
Thomas

[Thomas Okken]

I'm guessing he doesn't remember, because it can't be done. If you free-fall until just before impact, you end up falling with a speed of 85 at an altitude of 27.5. Burning 90 units of fuel at that point kills your speed, but only once you're at altitude -15. Splat.

You need two burns, and I believe the optimum is to free-fall until you're falling with a speed of 80 at an altitude of 110, then burn 15 units so you're falling at a speed of 70 at an altitude of 35; then burn 75 units, killing your speed at altitude 0.

[SlideRule]

Just for giggles & grins, A Guided Genetic Algorithm for the Planning in Lunar Lander Games: a modest read.

BEST!
SlideRule

[Gamo]

Very interesting that Sinclair Enterprise Programmable also got this game
called Moon Landing Game.
I try to program this game to 12C Platinum since this calculator also got
ALG mode.
For a mean time here is the listing and information for this game.

Interesting note is "The Equations used are exact"

   
   
   


Gamo

[StephenG1CMZ]

(02-04-2019 08:50 PM)Thomas Klemm Wrote:  

(02-04-2019 12:17 PM)EdS2 Wrote:  here's the HP-25 official lunar lander.

(02-02-2019 09:31 PM)Valentin Albillo Wrote:  I quickly discovered that I could land perfectly in just one step.

Now I wonder if you still remember how you did that.

Cheers
Thomas

Having performed many unsuccessful lunar landings, I am imagining a cheat button implementing "Beam me down"

[Paul Dale]

(02-04-2019 10:42 PM)Thomas Okken Wrote:  I'm guessing he doesn't remember, because it can't be done. If you free-fall until just before impact, you end up falling with a speed of 85 at an altitude of 27.5. Burning 90 units of fuel at that point kills your speed, but only once you're at altitude -15. Splat.

Would it be possible to burn more fuel than this such that zero altitude is achieved exactly? You'll be travelling upwards but the programme might not care.


Pauli

[Thomas Okken]

(02-07-2019 04:27 AM)Paul Dale Wrote:  

(02-04-2019 10:42 PM)Thomas Okken Wrote:  I'm guessing he doesn't remember, because it can't be done. If you free-fall until just before impact, you end up falling with a speed of 85 at an altitude of 27.5. Burning 90 units of fuel at that point kills your speed, but only once you're at altitude -15. Splat.

Would it be possible to burn more fuel than this such that zero altitude is achieved exactly? You'll be travelling upwards but the programme might not care.

Yes, just wait until the last moment and then burn 120 units, i.e. everything. You end up with 30.0000, i.e. altitude 0 and moving up at 30 ft/s. The program only cares about negative altitude at the end of a time interval; it's up to the player to know that only 0.0000 means a successful landing.

[David Hayden]

(02-03-2019 09:26 AM)EdS2 Wrote:  This game needs tuning!

I wrote an article in DataFile a year or two ago about the 29C lunar lander program. It contains two bugs which I speculated were due to the limited space of its progenitor. The first bug let you drop below ground level and rise up above it without crashing if done during a single burn. I think the second bug computed your terminal velocity wrong when you ran out of fuel. Something like that.

The bugs persisted as the program was ported to later calculators with more memory.

Dave