thalassogen
SOC-12
That's JSON. That would be easy for me as the environment I am using already supports implicit conversion of objects from and to JSON. 
T5 Web Apps
- Thread starter thalassogen
- Start date
That's JSON. That would be easy for me as the environment I am using already supports implicit conversion of objects from and to JSON.
Sorry, yes, I meant JSON. My brain is tired today.
It had occurred to me that, for a web app, JSON would be the obvious choice of return format.
Assuming we can arrive at a reasonable structure. My example above is one. There are others.
Ah, my brain.
I noticed another erratum that will affect the frequency of Gas Giants in system hexes.
Planet generation rules say roll 2D/2 - 2, ignore fractions, for number of Gas Giants (some will be converted into Ice Giants) in a system hex (with at least one star). But that means any roll of 6 or above generates at least one gas giant. The probability for 6+ on 2D is 26/36.
But this goes against the rule in EVERY edition of TRAVELLER, and p. 431 in T5, that says a gas giant exists in a hex if you roll 9 or less on 2D. That has a probability of 30/36 (5/6).
If you have been using the 2D/2 - 2 to generate gas giants within every hex in your algorithm, you will be getting less than the 5 out of 6 systems with gas-giants that have been traditionally obtained.
I'm amazed I got my book as one of the last orders but no one else noticed these probability differences. I put it in Errata and see what they think about it. Possible solution: do NOT ignore fractions but round up if there is a 1/2-result. Or do (2D-3)/2, ignore fractions so a roll of 5+ will produce at least one gas-giant. The outcome for a system (hex) will be:
0 gas giants 6/36
1 gas giant 9/36
2 gas giants 11/36
3 gas giants 7/36
4 gas giants 3/36
(again, every 2nd gas giant orbiting a distinct STAR is converted to an Ice Giant.)
So now the odds in a hex to have a gas-giant go back up to 30/36 and all is in harmony with past editions of Traveller.
thalassogen
SOC-12
OK, I'll check the upcoming errata v0.8 and will incorporate any changes.
(I am using 2d/2 -2, BTW. SectorMaker is using SystemMaker to populate the hexes.)
(I am using 2d/2 -2, BTW. SectorMaker is using SystemMaker to populate the hexes.)
Original_Carl
SOC-12
I have some questions about the .CSV file that Sector Maker produces. I recently used it to generate an "MTU" for a campaign I've started with some close friends. I'm super-super happy with the results!
1. The column "P or BG" should probably be PBG (Population, Belts, Giants) and should be a three-digit column for the system indicating population of the mainworld, number of planetoid belts, and number of gas giants (p.431, BBB). Presently, this column shows as a single number, and it has an entry for every body in a system. Did you intend something else for this column?
2. "Ring System" -- what is this? There is always an 'R' in the Satellite Orbit column when Ring System appears in the Name column and sometimes a digit appears in the Secondary Orbit column. It's too common for a Ring World (TL 27 and higher). I've been assuming that you're indicating a ring (ala Saturn and Jupiter) with satellites (moons) in it. Is this what you intended?
Thanks for producing this stuff. You're making my Referee job easy.
1. The column "P or BG" should probably be PBG (Population, Belts, Giants) and should be a three-digit column for the system indicating population of the mainworld, number of planetoid belts, and number of gas giants (p.431, BBB). Presently, this column shows as a single number, and it has an entry for every body in a system. Did you intend something else for this column?
2. "Ring System" -- what is this? There is always an 'R' in the Satellite Orbit column when Ring System appears in the Name column and sometimes a digit appears in the Secondary Orbit column. It's too common for a Ring World (TL 27 and higher). I've been assuming that you're indicating a ring (ala Saturn and Jupiter) with satellites (moons) in it. Is this what you intended?
Thanks for producing this stuff. You're making my Referee job easy.
If your current generator is "No. of Gas Giants in a system = 2D/2 -2 (ignore fractions)," it is generating FEWER systems with gas giants than in past editions of TRAVELLER. The probability works out to 26 out of 36 systems with gas giants, instead of the previous 30 out of 36 systems (or 5 out of 6, based on a roll of 9- on 2D6). It is creating a slight scarcity of gas-giants which may cause refuelling problems and a restriction of jump-routes.
But I will wait to see if this point is addressed in the Errata.
Last edited:
thalassogen
SOC-12
Hi Original_Carl,
I am glad you like the app!
1. Every hex as a mostly blank first line - this is the line of the primary star and it also contains part of the UWP pertaining to the system as a whole. Therefore it contains the BG info of the UWP. If you open the CSV with Excel, leading zeros will not be rendered and sometimes only one digit is seen in the first line of a hex, i.e. there are no belts and the indicated number of gas giants. The underlying CSV contains the leading zero, you can verify by opening the file in a text editor.
On all other lines (that indicate planets and not stars, gas/ice giants, or brown dwarfs - where the column is empty) this column indicates the P - the population multiple of the corresponding world.
2. A ring system is just that: a system of rings (like Saturns). I chose R as the orbit indicator as rings are normally very close to their planet and as I understood the rules they do not get a real satellite orbit letter of their own and I wanted them to have their own line instead of just being an additional descriptor to their world.
A digit in the secondary orbit column indicates that the ring orbits a world that itself orbits a companion star at the indicated (secondary) orbit which revolves around the primary at the indicated (primary) orbit.
Thalassogen - thank you for your wonderful apps. Like others, I also like the aesthetics.
I wonder if I am doing something wrong with your sector generator, though.
When I click to generate a Traveller Map, more than half the worlds go missing.
Might it be something to do with the .sec file generated?
I wonder if I am doing something wrong with your sector generator, though.
When I click to generate a Traveller Map, more than half the worlds go missing.
Might it be something to do with the .sec file generated?
Thalassogen - excellent UWP output with your sector maker.
I was thinking that it "should" look more like traditional .SEC files, but I changed my mind -- I think it's FINE as is.
Perhaps, though, you could print a simple "key" at the bottom of the page?
Hex Worldname UWP Codes {I} (Econ) [Soc] Nobility Bases TAS PBG ? Allegiance Stellar-data-to-end-of-line
There's only one field I couldn't figure out -- the character after the PBG.
On additional bit: how do we tell which companion stars are present? In the past, I've added a dash to "close" stars, and added a plus sign to "far" stars. So:
K0 V -M0 V +LD ; M0 V is close; LD is far
K0 V M0 V +LD ; M0 V is near; LD is far
K0 V -M0 V LD ; M0 V is close; LD is near
etc.
I was thinking that it "should" look more like traditional .SEC files, but I changed my mind -- I think it's FINE as is.
Perhaps, though, you could print a simple "key" at the bottom of the page?
Code:
0218 Alpha B7549AA-B Co Hi Lk Sa Tu {+4} (B89+2) [7D4E] BEf NS G 831 C Im K0 V M0 V LDThere's only one field I couldn't figure out -- the character after the PBG.
On additional bit: how do we tell which companion stars are present? In the past, I've added a dash to "close" stars, and added a plus sign to "far" stars. So:
K0 V -M0 V +LD ; M0 V is close; LD is far
K0 V M0 V +LD ; M0 V is near; LD is far
K0 V -M0 V LD ; M0 V is close; LD is near
etc.
Last edited:
Something else for you to idly think about:
Joshua (travellermap.com) and I have been playing around with two pseudo-random number generators from Bob Jenkins' website (http://www.burtleburtle.net/bob/rand/index.html); in particular, ISAAC (which is well-known) or, better, his "small noncryptographic PRNG", which has the ultimate convenience of being VERY small. I've implemented his small PRNG in Perl, JavaScript, Objective C, and Java, modified to use plain old 'long' ints where available (or just 'int' in Java) instead of unsigned versions, all with the same results.
By hashing a unique sector identifier string using MD5 (implemented everywhere) with Jenkins' small PRNG, we can programmatically generate the same output data given the same sector ID, regardless of which of our applications are being run (as long as both apps generate data in the same order as the rulebook). So for example, if someone enters "Thalassogen Sector A12" as the unique sector ID into my code, it will generate exactly the same sector as when said sector ID is entered into your code.
Proof of concept (but not T5 conformant): http://eaglestone.pocketempires.com...ilized&density=Standard&Generate=Submit+Query
Joshua (travellermap.com) and I have been playing around with two pseudo-random number generators from Bob Jenkins' website (http://www.burtleburtle.net/bob/rand/index.html); in particular, ISAAC (which is well-known) or, better, his "small noncryptographic PRNG", which has the ultimate convenience of being VERY small. I've implemented his small PRNG in Perl, JavaScript, Objective C, and Java, modified to use plain old 'long' ints where available (or just 'int' in Java) instead of unsigned versions, all with the same results.
By hashing a unique sector identifier string using MD5 (implemented everywhere) with Jenkins' small PRNG, we can programmatically generate the same output data given the same sector ID, regardless of which of our applications are being run (as long as both apps generate data in the same order as the rulebook). So for example, if someone enters "Thalassogen Sector A12" as the unique sector ID into my code, it will generate exactly the same sector as when said sector ID is entered into your code.
Proof of concept (but not T5 conformant): http://eaglestone.pocketempires.com...ilized&density=Standard&Generate=Submit+Query
Last edited:
thalassogen
SOC-12
This one is new, but you are right, data gets lost. I will look into it. Maybe the .SEC file format used by TravellerMap changed?
thalassogen
SOC-12
I will change the UWP output from simple paragraphs to table lines and put the "key" into the table header.
That's the number of worlds besides the main world, gas giants and planetoid belts. That's W as defined on page 431. Oops, there I see that W is the sum of all this - a BUG to fix.
Maybe we just add the orbit number in curly braces or something like this?
Even better.
thalassogen
SOC-12
That's interesting. Only problem is that I did not code the rolls exactly the same way (order and rolling convention) as the rulebook (e.g. page 31 Even Distribution: I don't use 2D6 for this but rather 1D9 and 1D10-1 respectively) . And it would require some lifting to change it all.
But if such repeatability is wished for I would gladly consume your implementation as a web service.
thalassogen
SOC-12
This one is complicated... the "old" Poster.aspx POST and the new /api/poster POST both render the same "reduced" sector. I guess I have to use the new T5-specific sector data format to get complete results. This will be a bit of work and I guess it will take me some time to get this running correctly. Sorry for the inconvenience. I will post here when things are back to normal with TravellerMap rendering.
UPDATE: Done! The sector and subsector data are now formatted as a table and the rows are clickable showing the system details...
That's the number of worlds besides the main world, gas giants and planetoid belts. That's W as defined on page 431. Oops, there I see that W is the sum of all this - a BUG to fix.
Ah, no, the table output is correct, I guess. Only the CSV output is showing a lower number (only including the other worlds).
The poster api needs to guess the data format. If it guesses wrong you might see odd results. Tab-delimited is going to be the most reliable. Feel free to email me to try and get this figured out.
Ah, found it. The legacy SEC parser doesn't handle "-" as a placeholder in the bases field. I've never encountered files in the wild formatted that way. Supporting that would be possible, but let's try and move to a more rigorous format like the http://travellermap.com/doc/fileformats.html#t5tab and http://travellermap.com/doc/fileformats.html#t5col rather than slapping on more patches.
If you just output a space when there's no base, everything will be fine.
If you just output a space when there's no base, everything will be fine.
thalassogen
SOC-12
Great tip, thanks for your support!
I just updated the Web Apps and its working fine now.
Switching to the new formats is a job to be done soon.
Are you joking? I am sitting here benefiting from your efforts for free - no inconvenience, quite the contrary.
The dash is how Marc's Second Survey spreadsheet represents "no bases".
