NathanKell changed the topic of #RO to: Welcome to the discussion channel for the Realism Overhaul (meta)mod for KSP! Realism Overhaul Main Thread https://goo.gl/wH7Dzb ! RO Spreadsheet http://goo.gl/Oem3g0 ! Code of Conduct http://goo.gl/wOSv2M ! | [15:01] <soundnfury> Straight Eight Stronk (and) RP-0/1 is basically "Space Agency Spreadsheet Simulator" with a rocket-flying minigame
<awang>
As long as the information isn't repeated anywhere, there is no way for things to become inconsistent
<Pap>
soundnfury: but how do you organize things when I want to only see the lr89 variants in any other type of medium?
<Mike`>
awang, yes, yes it is
<awang>
brb
<soundnfury>
well in a database, you SELECT * FROM parts WHERE class='engine' and family='lr89'; (or whatever)
<Pap>
Oh no, I fully agree that a database is probably the ultimate solution
Raidernick has quit [Read error: Connection reset by peer]
<soundnfury>
with a script you run a function in the REPL
<Pap>
I just don't know who is going to take the time to build it.
<Mike`>
awang, well, my 1st moon sat didn't survive more than a few days with principia :D
<Pap>
It took me Many Many Many hours to get the sheet working
<soundnfury>
(though the script way means you can't _edit_ through a view like that)
<soundnfury>
Pap: I *would*, but it would mean deciphering what your sheet does in order to reimplement it
<soundnfury>
(if you could write some sort of a spec for what The Sheet currently does, that *might* be helpful? idk?)
<Mike`>
awang, err, and you can always "destroy" vessels from the tracking station?
<Pap>
OK so if I give a walk through of the sheet, it would make everything better?
<Pap>
What database format would it be put in soundnfury?
awang has quit [Ping timeout: 182 seconds]
<soundnfury>
Pap: I'
<soundnfury>
m imagining sqlite so we can check it into the repo
<Pap>
I wonder if it is something that I would want to play with in order to get some familiarity into databases (something I have been meaning to do)
<Mike`>
soundnfury, database? :s i'm not sure if that is better than the google docs sheets (it is in some ways, sure)
<Mike`>
but before you do it, ask pap and nathan if they prefer editing data there instead of the sheet :)
<soundnfury>
Mike`: well, my preferred option is that the master parts list should be a Python script
<Pap>
soundnfury: how would a python script work?
<Mike`>
soundnfury, awand and me thought about yaml/toml or maybe csv, but being editable in a plain texteditor plus being able to check it into the repo does have advantages
<Mike`>
i mean, the latter nis true for sqlite, yes, but...
<soundnfury>
Pap: it'd basically create an object for each part, associate them with families / sets of identical parts / etc., various metadata in the objects
<soundnfury>
and then there'd be a function that just consumes the list of all the objects and generates $thing_you_need (e.g. the Tree.cfg file)
<soundnfury>
Pap: did I show you EK?
<Pap>
soundnfury: I looked at the repo to EK yesterday
<Mike`>
i'd like something like this aswell, it's definitely very nice for quick edits/fixes etc. But large scale edits/rebalances etc are probably better to do in a sheet/database, i see that point.
<Pap>
How can I add say the 300 parts from Bluedog using something like that soundnfury?
<Pap>
Seriously asking as I am all for the ease of use
<Pap>
Is the main issue that people do not understand how the sheet works?
<soundnfury>
you do the same data entry job you'd do with a spreadsheet, except instead of typing into several rows of cells, you write several calls to a function sorta like my launch() function
<soundnfury>
Pap: my biggest issue is that the formulae part of the sheet still lives outside the repo, despite being essentially part of the build system
<Pap>
Who would want to edit it in a plain text editor? There are 3838 parts currently in the sheet.
<soundnfury>
but yes also not knowing what it does or how bothers me
<soundnfury>
Pap: text editors can do a lot of things.
<soundnfury>
also: can haz a link to the spreadsheet (can't find it rn) so you can explain what it does?
<Pap>
Well, I obviously don't have a lot of time to work on RO/RP-1, so I am up for whatever the people that are doing the work want to have.
<Pap>
This was the best option when I was doing all of the tech tree work, but we should be past massive changes going forward.
<Pap>
ELI5 Sqlite
<Pap>
I know that I can directly port an excel sheet into an access database. Can I then change that into a sqlite file?
<Mike`>
anyway, i need sleep :/ good night.
awang has joined #RO
<soundnfury>
Pap: yeah, you'd just dump the access DB as SQL (I assume access can do that) and then 'restore' that dump into sqlite.
<Pap>
OK that is what I thought
<awang>
Mike`: idk if I'm doing something wrong, then, since my first polar satellite survived long enough to get low biome science for the entire moon
<awang>
And I don't remember seeing any significant perturbations
<Pap>
Let's wait for NK and awang to weigh in and I am for whatever is best
<awang>
Think it was a ~150km x 300km orbit?
<awang>
idk if that's too high
<soundnfury>
Pap: question, are RP-0 (column H) and RP0conf (column O) the same thing?
<awang>
Mike`: Yeah, I can delete from the tracking station, but the problem is I get like 1-2 fps in there
<awang>
So deorbiting in flight is somewhat better, since I can at least timewarp
<awang>
soundnfury: I'm sort of hesitant about a database, since git and binary blobs don't play too well together
<awang>
It mean, it works, but it isn't as convenient
Raidernick has joined #RO
<awang>
And as Mike` said, I was thinking yaml/toml and/or tsv, so the things are editable in a text editor
<awang>
Pap: adding new parts into a database might be something like pointing a script at a directory and running it
<awang>
And it would take care of parsing cfgs and inserting appropriate values if they exist
<Pap>
soundnfury: I actually do not know. My phone doesn't let me see the formulas, will have to wait until I am by a PC in a couple of hours.
<awang>
Pap: plain text editors could work if you just want to do a simple find/replace
<awang>
Maybe you know the part name, so you can ctrl-f there and fix something
<soundnfury>
okay I think I've poked around the sheet enough to figure out roughly what it does, maybe.
<Pap>
awang: if a text based idea doesn't work, would database be your vote over what we have now?
<Pap>
awang: are you saying that you want to edit the CFG directly?
<Pap>
soundnfury: it really isn't that complicated. Just a lot of nested functions to get it working.
<awang>
soundnfury: Looks like RP-0 and RP0conf differ in that the former can allow "No cost", whereas the latter is just true/false
<soundnfury>
awang: ah 'k
<awang>
Pap: Maybe? If we're going to stick a binary blob into the repo sticking with an Excel sheet might be less work
<awang>
Although you can't run queries against an excel sheet
<awang>
I think
<Pap>
Even in the old days of RP-0, the CFG were not edited. It was the Tree YAML file
<awang>
for better or worse
<awang>
Pap: not editing the cfg, editing the "master" text thing
<Pap>
Gotcha
<awang>
Open that up, ctrl-f wherever, save
<Pap>
What if I cleaned up the file and only kept the relevant information and we stay with a CSV? There is a lot of extra data on the sheet that was only needed for the changeover from old tech tree to new
<awang>
I vote TSV instead of CSV
<awang>
My main concern with TSV/CSV is that it can be awkward to edit values stuck in the middle of the file since you don't have column headers
<awang>
Not a problem if you have a spreadsheet program and can freeze them, but still
<Pap>
Tab delimited is usually only one space in a text editor, isn't it?
<Pap>
Right
<awang>
I thought tab delimited was literal tabs
<awang>
Although now that I think about it idk if any editors would replace those with spaces
<Pap>
It might be, the problem there is the descriptions. Those lines are VERY long in some circumstances
<Bornholio>
csv
<awang>
TOML is easier-ish to read, but you end up repeating key values (/column headers) over and over and over again
<Pap>
Can't you format a database query to look however you want?
<awang>
Yeah, but those descriptions are going to be long no matter the format
<awang>
I think so?
<awang>
I just think that requiring contributors to know some amount of SQL might be a bit on the steep side of the learning curve
<Pap>
You could create a query that gives a readable format that can be edited as the "master file" which then gets read back into the database, couldn't you?
<awang>
idk whether that's considered good
<Pap>
Same was true of YAML, wasn't it?
<awang>
That's sort of what I mentioned earlier
<awang>
Except the other way around
<awang>
"master file" is text, and can be read into a database to run queries against
<awang>
That's a bit more git-friendly
<awang>
Well, YAML is a bit easier to grok than SQL
<awang>
At least if you keep it simple
<Pap>
That would be the best option I think
<awang>
idk how complex the YAML file got
<awang>
Now the million dollar question: What text format?
<Pap>
It was kinda hard to work with. Just ugly I guess, not complicated, but difficult to maneuver and not a lot of rhyme or reason to the order
<awang>
Probably going to be true for any text file that holds this much data
<awang>
Especially if we want to combine RO stuff
<awang>
In theory, RO/RP-0 can draw from the same database
<awang>
Just pick different columns to export to get RO, RP-0, or both
<soundnfury>
fyi, I'm doing a poc prototype of the Python script approach
<Pap>
awang: You are making this sound more and more like a database
<awang>
Pap: I mean, that's sort of what it is?
lijat has quit [Ping timeout: 383 seconds]
<awang>
soundnfury: How well would the Python approach scale to the number of columns the spreadsheet has?
<awang>
Plus more if you want RO
<Pap>
soundnfury: I will be excited to see how it works
<Pap>
The sheet doesn't need to necessarily need to be as many columns as it is
<awang>
idk if I want to bet on it always being that way
<taniwha>
awang: python is pretty good with arrays
<taniwha>
(not really sure what you're asking)
<awang>
taniwha: I'm more concerned about specifying the data in the first place
<awang>
20-parameter constructors would be interesting
<awang>
Although named parameters help a lot
<awang>
But at that point you're basically replicating YAML/TOML
<awang>
Although you do bypass having to parse an external file
<awang>
brb
<taniwha>
ah
<taniwha>
do the parameters come in groups? if so, you could use classes to group those parameters and then pass those classes as parameters to the main class
<soundnfury>
yeah some of it will be through chunking
<awang>
Yeah, that would work
<awang>
soundnfury: Does Python have a SQL-ish way to searth the objects?
<awang>
Well
<awang>
filter() probably works
<taniwha>
awang: I'm pretty sure there are sql-ish modules for python
<taniwha>
I seem to remember using one for something many years ago
<awang>
Hm
<awang>
Guess I should be more surprised if there wasn't one
<taniwha>
or was that php? (I suspect the latter)
<taniwha>
still, python has been around for a LONG time
<taniwha>
there's probably a package for that ;)
<Bornholio>
pap the new guy on my team had to describe in intimate detail the nerdgasm he had touring spacex last week, SO JEALOUS.
saabstory88 has joined #RO
<soundnfury>
awang: since the data structures are (from the script's perspective) immutable, I can just have a magic indexing object instead of a simple list
<awang>
Magic indexing object?
<awang>
Hash table?
<Pap>
Bornholio that is awesome! Why does the new guy get to go?
<Bornholio>
we have redacted redacted redacted and so the boss got three of our guys a tour during a trip to solidworks world in LA
<awang>
s/redacted/████████
<Bornholio>
my 1.25m cockpit shows maxtemp skinmaxtemp of 450K ?
<awang>
Bornholio: Is that problematic?
<Bornholio>
its very burny, and since the part description says 1000k/1800k they dont' match
<awang>
Huh
<awang>
Look in the git history to see which changed most recently?
<Qboid>
awang: I added the explanation for this acronym.
<awang>
BNF?
<awang>
!acr -add:BNF Bakus-Naur Form
<Qboid>
awang: I added the explanation for this acronym.
<awang>
Dangit
<awang>
Now I don't know whether I want to spend my time learning TF or learning database stuff
<soundnfury>
awang: s/ak/ack
<Qboid>
soundnfury thinks awang meant to say: !acr -add:BNF Backus-Naur Form
<awang>
Dangit
<awang>
!acr -add:EBNF Extended Backus-Naru Form
<Qboid>
awang: I already know an explanation for EBNF! (Update it with !acr -update:EBNF Extended Backus-Naru Form)
<awang>
!acr -update:EBNF Extended Backus-Naru Form
<Qboid>
awang: I updated the explanation for EBNF
<awang>
!acr -update:EBNF Extended Backus-Naur Form
<Qboid>
awang: I updated the explanation for EBNF
<awang>
!acr -update:BNF Backus-Naur Form
<Qboid>
awang: I updated the explanation for BNF
<awang>
!acr -add:ABNF Augmented Backus-Naur Form
<Qboid>
awang: I added the explanation for this acronym.
blowfish has joined #RO
Mike` has quit [Ping timeout: 207 seconds]
Mike` has joined #RO
qwertyy has joined #RO
mkalte has quit [Remote host closed the connection]
ProjectThoth has joined #RO
<ProjectThoth>
Bornholio: :P
blowfish has quit [Quit: Leaving]
BadRocketsCo has joined #RO
<BadRocketsCo>
Howdy
Moistmelon has quit [Ping timeout: 186 seconds]
Moistmelon has joined #RO
gazpachian has quit [Quit: Lost terminal]
SpecimenSpiff has joined #RO
SpecimenSpiff has quit [Client Quit]
aradapilot has quit [Remote host closed the connection]
aradapilot has joined #RO
leudaimon|ZzZz has quit [Remote host closed the connection]
leudaimon has joined #RO
<leudaimon>
o/
BadRocketsCo has quit [Quit: Bye]
<leudaimon>
Pap soundnfury I was reading the backlog and it got me thinking... would this new system soundnfury is proposing allow to have the other cfg files, that right now may be inconsistent with what is shown in the tree (namely, engine configs, RCS fuels/levels, proc avionics) to be also created from the same master file? That would be very good
<taniwha>
leudaimon: that's the impression I got
<leudaimon>
cool
<taniwha>
certainly why I suggested soundnfury use my cfgnode module
<taniwha>
(sorry, had trouble with a system upgrade)
<taniwha>
all sorted, though
<leudaimon>
yeah, just this I think justifies moving into this system then. It's a lot of double work having to track separate configs for the tree icon and the actual upgrades
<taniwha>
also, I realized soundnfury /should/ look at the version in io_object_mu because it can do ${expr} substitutions
<taniwha>
(parser.py)
<taniwha>
(used by export_mu.py)
<taniwha>
actually, I should sync them up
ProjectThoth has quit [Quit: +++out of cheese error+++]
<leudaimon>
hehe, this is kind of greek for me, but if it allows this "one file rules them all" approach I'm all for it
<taniwha>
the parsing would allow for template cfg files
<taniwha>
(mass and volume calculated by the exporter)
wb99999999 has quit [Ping timeout: 180 seconds]
<leudaimon>
cool
<taniwha>
nodes are empties in the blend
<zilti>
awang, Mike` : this is my list of mods. With these, I have ~realtime performance, so these are in my eyes guaranteed not to slow down the game: https://qbin.io/freebase-grower-44kw
Daz has quit [Ping timeout: 182 seconds]
Daz has joined #RO
Hypergolic_Skunk has joined #RO
awang has quit [Ping timeout: 207 seconds]
awang has joined #RO
Maxsimal has joined #RO
<Maxsimal>
o/
<Probus>
\o
<awang>
zilti: What are your computer specs?
<Maxsimal>
how's it goin today?
<awang>
Database normalization makes my head hurt :(
<awang>
soundnfury is looking at python-ing RO/RP-0 config files
<awang>
I'm looking into database-ing those files
<awang>
Quite a bit of conversation yesterday on migrating from the RP-0 spreadsheet to something more versionable
<taniwha>
being able to do a diff is invaluable
<Maxsimal>
Hrmm, I thought people liked the csv solution
<taniwha>
csv is good
<awang>
I think CSV/TSV are still on the table
<taniwha>
it is versionable
<awang>
Just awkward to edit in a plain text editor
<awang>
Although that is probably the case for any plain text format
<Maxsimal>
Does anyone else feel these are getting blase, now, after the heavy launch? :P
<awang>
Yeah, importing/exporting works, but can be a slight pain compared to plain text edting
<awang>
Depending on the edit you need to make, of course
<taniwha>
yeah, simple tweaks are probably best done in your favorite editor
<leudaimon>
awang most OSs will open a csv directly in a spreadsheet-like program, no?
<awang>
And that's where CSV/TSVs might stumble
<Probus>
Maxsimal, I was just thinking the same thing. This one doesn't recover, but the ones that land are still exciting.
<awang>
leudaimon: Maybe? Depends on whether you have a spreadsheet program installed, I guess
<awang>
I went for a surprisingly long time without one just because I never needed it
<awang>
Ended up getting Office for work, so I'm good, but idk if there's anyone else like that
<awang>
Or if I'm the odd one out
<leudaimon>
I see, yeah, coming from linux, most ditros include libreoffice or something similar
<Probus>
Maxsimal, We may see a fairing capture though.
<leudaimon>
and in my field (academic research in ecology focused in data analysis) you don't survive one day without a spreadsheet program
<leudaimon>
the fairing recover will be cool if they manage
<Maxsimal>
Maxsimal: Oh that will be cool, I haven't read much about how they're doing it - saw the ship with the big kerbal-looking struts to gather it though
<Maxsimal>
parachutes I guess?
<taniwha>
Maxsimal: no rocket launch is ever blase
qwertyy has quit [Read error: -0x1: UNKNOWN ERROR CODE (0001)]
<Pap>
Maxsimal and Probus - If that is how you feel (and I am sorta the same way) then SpaceX is accomplishing exactly what they want. They want space travel to become as routine as airplane travel.
saabstory88 has joined #RO
<Pap>
OpenOffice and LibreOffice are both Open Source, right?
saabstory88 has quit [Client Quit]
<Maxsimal>
Pap: Yeah. Well I was kind of kidding - I'd still love to be at/on one and wouldn't be bored.
<Maxsimal>
Pap: And openoffice is apache licence, it's open source.
<leudaimon>
hype with launches are still high after FH, more than 150k people watching
<Maxsimal>
Was 200k - wonder where they're circularizing, since they're over 350km now and still have 3-4 minutes to go
<Maxsimal>
Wish they had an attitude telemetry indicator
<taniwha>
I wish they'd show speed in m/s
<Maxsimal>
^ that too
<leudaimon>
m/s instead of km/h wouldn't be bad
<leudaimon>
ninjad
<Probus>
Can't wait for Spacex's gigabit ethernet satellite constellation.
<leudaimon>
no info on the fairings...
VanDisaster has joined #RO
<Probus>
Parafoil deployed.
<leudaimon>
did you get it in the chatterer?
<Probus>
Twitter...
<leudaimon>
oh sure
<awang>
Does SpaceX have to catch the fairings before they hit the water?
<awang>
Guess it's pretty hard to maneuver a big boat like that
<Maxsimal>
yeah tbh I don't really get how this is supposed to work unless the parasail is very steerable - a 215 long boat isn't exactly nimble and chutes aren't particularly accurate.
<awang>
Apparently there's a parafoil?
<awang>
so I guess there's some amount of steering possible?
<Starwaster>
OMG! NyAN! NYAN EVERYWHERE
<Starwaster>
ZOMGWTFBBQ
<awang>
Is there a way to call something upon destruction without having to inherit from MonoBehaviour?
<awang>
Besides finalizers, because apparently those are tricky
<awang>
Looking into having tooling changes only update on vessel change, but the class doesn't inherit from MonoBehaviour
aradapilot has quit [Remote host closed the connection]
aradapilot has joined #RO
TaggedYa has joined #RO
Maxsimal has quit [Quit: Web client closed]
<TaggedYa>
Anyone home who knows if there is a Krash config for RP-1 Dev
<Probus>
I've been using it. I think I just followed the install spreadsheet.
<TaggedYa>
Ill look again, my google foo is failing me.
<awang>
Luckily it doesn't seem there are any changes
<TaggedYa>
I just typed the values in in game and it seems to work.
<TaggedYa>
Thanks
<awang>
np
<awang>
Make sure you use the custom DLL, too; it helps a ton with simulation costs
Probus has quit [Read error: Connection reset by peer]
Probus has joined #RO
<zilti>
awang: Ryzen R7 1800X and Radeon RX580. Oh, and 16 Gigabyte RAM.
<awang>
zilti: Oh, right. You're *that* guy :P
<awang>
I'm not sure if my game is GPU, CPU, or RAM limited
<awang>
We got quite a few of the same mods, too
<awang>
Hmmm
<awang>
Wonder if the launch clamps counting towards pad mass is intentional for RP-1
<awang>
Pap: ^?
<Pap>
No idea actually awang aren't they very very small mass?
<leudaimon>
just saw that the spectroscopy experiment from the 20in probe shows as level 1, but it's transmitable. Is that intended?
<awang>
Pap: Some of them, yeah
<awang>
I think some of them get to 0.5-1 ton, which can be significant
<awang>
I've been (ab)using the 0.1 ton ones (I think), so I haven't had too many issues
<awang>
I've been brushing right up against the 60t limit, so in theory that 0.2t would help
<Pap>
In my mind it is not intended, NK might have a different idea
<Pap>
leudaimon: I do not think that is on purpose.
<leudaimon>
(actually, that probe having so many experiments is very cheaty)
<awang>
!tell NathanKell* Are launch clamps intended to count towards pad mass limits?
<Qboid>
awang: I'll redirect this as soon as they are around.
<awang>
leudaimon: Noooooo don't make me rebuild my Mars/Venus probes :(
<leudaimon>
^ as I was sayin
<awang>
zilti: I'm somewhat inclined to think that BackgroundProcessing would conflict with something in the RO suite
<awang>
Don't remember for sure though
<awang>
Can't find it in my logs
<awang>
Disregard me
<awang>
!tell NathanKell* Just making sure, is BackgroundProcessing compatible with the RO suite? Thought I remembered reading something about a conflict a while ago, but can't find any mention in my logs
<Qboid>
awang: I'll redirect this as soon as they are around.
<awang>
Wonder if that would explain why my early Sputnik cores seemed to last forever
<awang>
zilti: BetterCrewAssignment is supposedly not compatible with KCT
<awang>
According to the forum post
<zilti>
awang: I was somewhat wondering about BackgroundProcessing. Especially since Kerbalism already handles electric charge in the background.
<awang>
Wasn't Coherent Contracts integrated into the game?
<zilti>
BetterCrewAssignment seems to work fine so far though.
<zilti>
I'm on 1.2.2, so it's probably not integrated there yet
<awang>
zilti: Yeah, idk... I'd imagine having multiple things trying to take care of electricity wouldn't turn out well
<awang>
zilti: Maybe? I seem to remember that warning being around for a while, but I never investigated why
<awang>
You have a lot of CC packs o_O
<awang>
I thought those were pretty much all incompatible with RP-0
<awang>
Since their balance is way off
<zilti>
CC packs?
<awang>
ContractConfigurator packs
<zilti>
Ah, these. Yea, most of them give way less reward than what the rocket would cost, the discrepancy is kinda funny.
<awang>
Less?
<awang>
Last time I tried, they were giving stupid amounts more o_O
<awang>
That was quite a while ago, though
<awang>
Is the EVAParachutes mod from the Vanguard pack?
<awang>
I originally had an EVA parachutes mod, but got rid of it after finding out it made my planes cost ~100x more
<awang>
Looks like we have pretty similar installs
<zilti>
No, it's a completely separate mod. I think that guy also made a mod that goes along with it and adds an "eject from vessel" functionality.
<zilti>
Is BDB supported by RP-0 and RO now?
<awang>
Hm. I'll have to check that out
<awang>
Probably will be helpful when I screw up my X-1
<awang>
BDB is at least partially supported
<zilti>
I wish the SEP were supported... It's pretty much the only thing that gives me something to do on a surface
<zilti>
Because as much as I'm into spaceflight, what makes me want to fly to Moon, Mars and beyond in KSP is the actual surface there.
<awang>
Be the change you want to see in the world!
<awang>
But yeah, development has been slow the past few months
<zilti>
Yea, I know... The problem is, I'd be contributing to four dozen opensource projects while being without a job and without sleep, if I'd do that ^^
<awang>
Woah
<awang>
Yeah, taking care of yourself comes first
<zilti>
Jup. I mean, I can't do as much as I'd want anyway due to depression. I work, and I maintain a few small libs I use at work, and do the odd pull request if I find a bug and manage to hunt it down. I think I btw made RO/RealFuels configs for a couple Russian engines, hunting down their specs online for hours, only to find out the engines already did have RO configs ^^
<zilti>
I found some space info page that day, with all the infos on even the more obscure versions of already obscure rocket engines
<zilti>
Hmm, I'm just watching the F9 launch from today. That is a lot of boiloff they have there
TaggedYa has quit [Quit: Web client closed]
Senshi has quit [Quit: Leaving.]
<Probus>
Nyan Cats! Starwaster, I blame you!
<zilti>
Sometimes it REALLY pays off that I do excessive screenshots. I just did a rather overpowered night launch: https://i.imgur.com/xyRTw1W.png
stratochief has joined #RO
VanDisaster has quit [Ping timeout: 182 seconds]
VanDisaster has joined #RO
Kraken has joined #RO
<Probus>
Wow, that looks good zilti
VanDisaster has quit [Ping timeout: 182 seconds]
VanDisaster has joined #RO
Hypergolic_Skunk has quit [Quit: Connection closed for inactivity]
<awang>
zilti: Hey, same here
<awang>
I "work", then go home, eat, then sleep, then repeat. Sometimes I get time to look at rockets.
leudaimon has quit [Quit: Leaving]
<Mike`>
awang, well, my first moon sats orbit had a very low periapsis of 80km or so, very high periapsis.... might have been unlucky :)
<awang>
Mike`: What was your inclination?
<awang>
My satellite was in a polar orbit, which is pretty close to the frozen inclination of 87 degrees
<awang>
idk how close you have to be, though
<awang>
egg|zzz|egg: ^?
<Mike`>
err, my inclination was probably more equatorial than polar, but i don't know exactly
<zilti>
Ooh, so it's not possible to go completely polar?
<awang>
Mike`: Apparently the Moon's frozen orbits are at 27, 50, 76, and 86 degrees
<awang>
zilti: It's possible, just not as stable
<awang>
I think
<awang>
I made it through at least a month
<Mike`>
okay, good to know :)
<Mike`>
anyway, need to sleep again, nighty.
<awang>
\o
<zilti>
337 tons of rocket to propel a 2-ton payload into space with 9935m/s delta-v...
leudaimon has joined #RO
<leudaimon>
anybody knows who has written the earth observation sat contract?
<leudaimon>
it has a bug... whenever you launch a satellite with the contract active, it activates a counter of 24h for completion of the contract and then fails
<awang>
leudaimon: The Earth Observation Imaging Satellite?
<leudaimon>
yep
<awang>
git says Pap wrote it
<awang>
Actually, Pap is the only one who has touched it
<Pap>
Github lies
<Pap>
Pap doesn't make mistskes
<leudaimon>
hahaha
<soundnfury>
kparts now has code to generate TREE-Parts.cfg, TREE-Engines.cfg, ECM-Engines.cfg and identicalParts.cfg
<soundnfury>
I'm getting there :)
<awang>
soundnfury: You're so fast o_O
<awang>
I haven't even started the actual-database prototype
<awang>
At this rate, I probably won't need to :P
<soundnfury>
awang: I'm fast because Python :D
<soundnfury>
faster than a greased snek
<awang>
2fast2quick
<awang>
Now that I think about it...
<awang>
databases are going to be awkward
<awang>
Like an Engine table probably should have its configs listed somewhere
<awang>
Can't store lists
<awang>
So now we end up with a bajillion tables to store configs for each engine
<soundnfury>
awang: normalisation is your friend :P
<soundnfury>
you'd have a foreign key in the configs table that references the pk of the engines table
<soundnfury>
then all the configs for all the engines go in the configs table
<awang>
soundnfury: Uh
<awang>
Wait
<awang>
I think I see what you mean
<awang>
Oh right
<awang>
Derp
<awang>
One engine can have multiple configs, but any one config will only be for a single engine
<awang>
Or at least I hope
<soundnfury>
yeah, if not then you have a many-to-many binding table
<soundnfury>
(engines with pk 'ei', configs with pk 'ci', and a table with ei and ci fk columns)
<soundnfury>
also, now have ECM-Parts.cfg too
<soundnfury>
I'm not sure, but I *think* that's all the generated cfgs covered.
<awang>
Jeez
<awang>
Oh, so you need a third table for a many-to-many?
<awang>
...Makes sense
<awang>
The third table has the cartesian product of engine/config types?
<awang>
engine/config foreign keys only exist if it's valid?
<soundnfury>
yeah
<soundnfury>
each row is a "this engine has this config" relation
<awang>
Ow
<awang>
That seems like it'd be a big table
<soundnfury>
nah, because most (engine, config) pairs aren't in it
<soundnfury>
like, there's no (A4, AJ10-118F) row
<awang>
Sparse table?
<awang>
Oh, so just a missing key
<awang>
...Right
<awang>
I knew that
<soundnfury>
xD
<awang>
Are you using taniwha's cfg parsing stuff and generating the scripts from there?
<awang>
Sort of reverse of what the scripts should eventually do
<soundnfury>
no, I'm not
<soundnfury>
data entry is manual
<soundnfury>
because it wants to be written in a sane & structured way
<awang>
Seems like you're a fast typer then :P
<soundnfury>
nope, I've only got a tiny test dataset
<soundnfury>
it's that I'm generating more cfgs for that tiny dataset
<awang>
Oh
<awang>
Right
<awang>
And here I thought that you were almost done with everything
<soundnfury>
lolno
<soundnfury>
but data entry can come later and I don't have to do it all myself
<soundnfury>
(we can parcel it out, by the glory of merges)