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
Probus has quit [Ping timeout: 198 seconds]
<awang>
Bornholio: Got logs? I might be able to take a look
<awang>
brb
awang has quit [Ping timeout: 198 seconds]
awang has joined #RO
schnobs has quit [Read error: No route to host]
Starman4308 has joined #RO
Maxsimal has quit [Quit: Web client closed]
qwertyy_ has joined #RO
qwertyy has quit [Ping timeout: 186 seconds]
qwertyy has joined #RO
<leudaimon>
anyone has any idea why procedural decoupler is working with maximum impulse even when I set it to 0 in the VAB?
<leudaimon>
(even editing persistent.sfs to set impulse percentage to 0 does not work)
stratochief has joined #RO
qwertyy_ has quit [Ping timeout: 383 seconds]
<awang>
Mike`: macOS says KSP is eating 22.35GB of RAM
<awang>
Memory pressure 30%
<awang>
Still got ~1.5GB free
<awang>
htop says KSP has ~5.5GB under RES
<awang>
(resident?)
<awang>
leudaimon: idk? It worked for me in the past
<leudaimon>
yeah, it has worked for me before
<awang>
Not sure then
<awang>
Mike`: htop also says KSP has 34 gigs virtual memory i)I
<awang>
s/i)I/o_O
schnobs has joined #RO
<Starman4308>
If it says low memory pressure, then everything is working as expected AFAIK.
<Starman4308>
Memory is a resource to be used, and if the computer doesn't have to do garbage collection, it won't.
<awang>
Yeah, I know
<awang>
I'm mostly just curious about the massively inflated number reported for KSP memory use
<taniwha>
awang: 15GB here
<taniwha>
(not using RO)
<taniwha>
RSS is 3.7GB
<taniwha>
(rss = resident set size or something like that, ie how much physical memory it is using)
<awang>
taniwha: I thought you were saying RealSolarSystem is 3.7GB, and I was about to ask how you knew how much RAM RSS was taking up
<awang>
Got a lot of part packs installed?
<taniwha>
EL, Kethane, Talisar, Tokamak
<taniwha>
might be a couple of others
<awang>
Those eat up that much RAM?
<taniwha>
nah, I think it's just how memory management works
<schnobs>
virt 10, res 7 here. Pretty small RP-1 set, neither FASA nor eye candy.
schnobs has quit [Ping timeout: 198 seconds]
ProjectThoth has quit [Ping timeout: 182 seconds]
Probus has joined #RO
saabstory88 has joined #RO
<Starman4308>
On my end. about 5.4 GB of memory used on a 16 GB Windows 10 machine.
<Starman4308>
No eyecandy: I have enough trouble keeping RP-0 stable!
<saabstory88>
5.4 is pushing the limit?
blowfish has joined #RO
<Starman4308>
I haven't the foggiest notion what's causing my crashes. I have a vague suspicion it's related to gigantic save files.
<saabstory88>
What do the logs say?
<Starman4308>
Mostly just seems to be a generic crash:
<Starman4308>
Personally, I just got accustomed to heavily modded KSP installs crashing every so often.
<Starman4308>
Though: it's curious, how quickly some of your habits change when you start playing RO/RP-0 (and I'm given to understand it'll be renamed RP-1 with the 1.31 release?). I used to just rely on built-in reaction wheels... I've barely used any in RO.
stratochief has quit [Ping timeout: 383 seconds]
Probus has quit [Quit: Leaving]
Probus has joined #RO
Hypergolic_Skunk has quit [Quit: Connection closed for inactivity]
Starman4308 has left #RO [#RO]
Mike` has quit [Ping timeout: 198 seconds]
Mike` has joined #RO
qwertyy_ has joined #RO
qwertyy has quit [Ping timeout: 198 seconds]
stratochief has joined #RO
wb99999999 has joined #RO
<wb99999999>
hey guys
stratochief has quit [Remote host closed the connection]
saabstory88 has quit [Remote host closed the connection]
ProjectThoth has joined #RO
blowfish has quit [Quit: Leaving]
Elouda has joined #RO
Senshi has joined #RO
Elouda has quit [Remote host closed the connection]
aradapilot has quit [Remote host closed the connection]
aradapilot has joined #RO
<Mike`>
awang, well, so actual ram usage is 5.5GB for you, that sounds more reasonable :D
<Mike`>
question is what macos actualyl reports there
<Mike`>
especially for diretly after game start
ProjectThoth has quit [Quit: +++out of cheese error+++]
wb99999999 has quit [Ping timeout: 180 seconds]
Shoe17 has quit [Quit: Connection closed for inactivity]
SigmaAway has joined #RO
Mike` has quit [*.net *.split]
Probus has quit [*.net *.split]
VanDisaster has quit [*.net *.split]
Kraken has quit [*.net *.split]
acharles has quit [*.net *.split]
armed_troop has quit [*.net *.split]
Ezko has quit [*.net *.split]
NCommander has quit [*.net *.split]
Agathorn has quit [*.net *.split]
Senshi has quit [*.net *.split]
leudaimon has quit [*.net *.split]
Moistmelon has quit [*.net *.split]
SirKeplan has quit [*.net *.split]
ferram4 has quit [*.net *.split]
zilti has quit [*.net *.split]
Addle has quit [*.net *.split]
egg|zzz|egg has quit [*.net *.split]
xShadowx has quit [*.net *.split]
petti has quit [*.net *.split]
NBones has quit [*.net *.split]
Snoozee has quit [*.net *.split]
taniwha has quit [*.net *.split]
acc has quit [*.net *.split]
DuoDex has quit [*.net *.split]
qwertyy_ has quit [*.net *.split]
Thomas has quit [*.net *.split]
NathanKell|AFK has quit [*.net *.split]
Sigma88 has quit [Ping timeout: 383 seconds]
BasharMilesTeg_ has quit [Ping timeout: 383 seconds]
VanDisaster has joined #RO
Mike` has joined #RO
Kraken has joined #RO
Probus has joined #RO
DuoDex has joined #RO
Ezko has joined #RO
armed_troop has joined #RO
NCommander has joined #RO
acharles has joined #RO
Agathorn has joined #RO
Hypergolic_Skunk has joined #RO
Thomas has joined #RO
SirKeplan has joined #RO
ferram4 has joined #RO
zilti has joined #RO
NBones has joined #RO
Moistmelon has joined #RO
leudaimon has joined #RO
petti has joined #RO
qwertyy_ has joined #RO
Snoozee has joined #RO
NathanKell|AFK has joined #RO
xShadowx has joined #RO
Addle has joined #RO
BasharMilesTeg has joined #RO
egg|zzz|egg has joined #RO
Senshi has joined #RO
acc has joined #RO
taniwha has joined #RO
<Probus>
!tell awang So what did you think of an Ironman mod?
<Qboid>
Probus: I'll redirect this as soon as they are around.
<taniwha>
Probus: new EL part is working nicely :)
<taniwha>
(the tiger-striped disk)
<Probus>
What's it do?
<taniwha>
it's a pad that self-destructs when you finalize the build, and the build becomes a "permanent" part of the ship (if the root node is a separator of some sort, then it's still detachable)
<Probus>
Sweet!!!
<Probus>
That's a great idea.
<taniwha>
I've been doing some testing and wound up fixing bugs in KIS, CLS and Tokomak :/
<Probus>
If you would like someone to help test, just let me know.
<taniwha>
I will, thanks
<taniwha>
have a couple of obvious things to fix, though (such as producing an error when no attach node on the root part is available)
<Starwaster>
are the source files for Porkjet's overhauled parts around anywhere?
<schnobs>
That's a big rocket. I cannot identify the engines...
<schnobs>
The dV budget is generous but oddly distributed. Can the upper two stages do restarts?
<schnobs>
also, "solar orbit" was that a typo, did you mean the Moon instead? As the vessel's named "Luna 1", I wonder.
<zilti>
The topmost one can. The others are all Russian engines, and the best engines I have at this point in RP-0. No, "solar orbit" is correct, it's named after the Soviet "Luna 1" that ended up in solar orbit instead of impacting on the Moon
<schnobs>
OK, using stage designations as in the dV chart:
<schnobs>
3 might be a little short in TWR, no dealbreaker thhough.
<schnobs>
you need about 9-10km/s for low earth orbit
<schnobs>
stages 4-3 have ~6km/s.
<Starwaster>
zilti cannot into space?
<zilti>
According to this chart: https://i.imgur.com/SqdzxzF.png I need 12.61k delta-v to escape the earth gravity field, meaning I'm about 1k short
Probus has quit [Remote host closed the connection]
<soundnfury>
zilti: uh, Luna 1 got a gravity assist from the moon
<zilti>
Starwaster: zilti cannot into space, zilti can only into earth orbit :\
<Starwaster>
:(
<zilti>
soundnfury: Aw, of course... I forgot thinking about gravity assists
<schnobs>
stage two adds another 5km/s, that's overly generous for making earth orbit. Removing 1-2 km/s here will be beneficial, increasing lower stages dV and TWR and you should still be ablt to make orbit, easily.
<Starwaster>
I used one of those really early into my last career game, accidentally. Got me all the way into solar space for some extra science
<Starwaster>
Science!
<Starwaster>
Hell yeah Science, bitches!
<schnobs>
To the moon, another 3200m/s. You've budgeted 5000.
<soundnfury>
also, you have way more than 12.6k delta-V. were you looking at the atmo number?
<zilti>
schnobs: Are you looking at the atmospheric deltaV, or the vacuum ones?
<zilti>
Yes, I was/am looking at the atmospheric number
<schnobs>
Vacuum. Atmospheric is a decoy, disregard it.
<zilti>
Like completely?
<schnobs>
(really, one minute into the flight you're 90% in space, ISP-wise)
<zilti>
True
<schnobs>
Surface TWR is a figure to take care of, but atmospheric dV is a distraction.
<schnobs>
I gather that your last stage is uncontrolled?
<zilti>
Hmm, I see. Well I figured atmospheric deltaV to be more accurate since it took me a rocket with like 9800m/s atmospheric deltaV to get into a polar orbit at 5700km
<zilti>
Why uncontrolled?
<schnobs>
don't see any RCS or avioncs.
<schnobs>
dont's avionics anywhere, coming to think of it.
<zilti>
There are RCS though :) At the gold foil thingy
<zilti>
And procedural avionics at every stage
<Starwaster>
yeah once you're 7km up pressure is like 50% and you already have a lot more dv than ont he ground
<schnobs>
But, yeah, arguing top-down:
technicalfool has joined #RO
<schnobs>
it's no more than 3200m/s to get a lunar encounter, scale your last stage accordingly.
<schnobs>
Missing the moon will nearly automatically toss you out of the earth/moon system.
<schnobs>
I don't think you even need to pay attention to which side you miss it on.
<zilti>
Heh, alright
<Bornholio>
have 50-150dv of rcs to adjust for safety
<schnobs>
That'd be generous :)
<Bornholio>
and you can hit it from most any ol inclination even polar.
Probus has joined #RO
<zilti>
So all these launches, I used way too strong rockets to reach LEO ^^
<zilti>
Bornholio: well, for all I know the Soviets did launch over the north pole to the moon
<Bornholio>
then hit a higher PO and use slightly less in transfer
<schnobs>
zilti: Well... if you didn't notice excess dV after shutdown, they may have been right for you after all.
<zilti>
schnobs: for all I know I did rather efficient gravity turns, which makes that odd... Then again, I always had *a lot* of reserve in my topmost, small pressure-fed maneuver/deorbit stage.
<zilti>
I mean, after launch, nothing really shows me how much dV I have left.
<taniwha>
zilti: aren't you using MechJeb?
<schnobs>
^
<zilti>
taniwha: Yes, but even KER shows more info in-flight (but shows 0 dV for the currently active stage)
<taniwha>
zilti: you have not looked at MJ's custom windows :)
<zilti>
taniwha: Indeed, I only clicked through the pre-made ones
<taniwha>
also, you can edit the pre-made ones
<Bornholio>
there is probalby 120 extra values that can be added to the display windows
<taniwha>
still, KER+MJ = more power
<Bornholio>
one reason i like KER is VAB resource summary
<zilti>
Bornholio: Neat. I'll take a look at them
<taniwha>
Bornholio: EL has similar
<zilti>
I also highly prefer KER's window design. MJ has very translucent, hard-to-read plaintext windows
<Bornholio>
for me the node editor and SmartASS are too nice to do without in MJ.
<soundnfury>
Bornholio: yup, I find the resource summary very handy too
<Bornholio>
how much Commsat Payload do i have?
<schnobs>
proc avionics tends to come out rather heavy more often than not, also carries too little electricity.
<schnobs>
My pseudo-atlas tends to drain the upper stage, while the upper stage won't even last half an orbit.
<soundnfury>
schnobs: you can always add more batteries
<schnobs>
The main benefit of proc avionics is that you can make them fit, but that goes out of the window when you have to add batteries.
<schnobs>
Except proc ones, which again are heavy. Actually, the proc avionics often come out rather heavy in and of themselves.
<schnobs>
more often than not, I find that I'm better off building a interstage as service bay.
<schnobs>
also, have proc SRBs been taken out?
<Bornholio>
yes
<schnobs>
balance reasons?
<Bornholio>
yes and they are somewhat broken functionally
<schnobs>
and while I'm at it, most of my engines have no effects: no fire, no sound.
<Bornholio>
realplume?
<schnobs>
Except some (eg SXT agena) which appear to be burning all the time.
<schnobs>
now that's odd. not even installed... m)
<schnobs>
CKAN demands I install some RealPlume configs. Will these get in the way?
<Bornholio>
just remove them
Starman4308 has joined #RO
<schnobs>
Hmmm. With realplume I'm getting two MM errors related to RO_RCS_Config.
<schnobs>
I seem to have lost the aerobee and baby sergeant...
<schnobs>
also Agena, LR105 and LR79...
Shoe17 has joined #RO
<Bornholio>
RP-1 Install?
<schnobs>
yep. Just discarded, I'll start over.
<zilti>
I never even used the procedural SRBs - they seemed underpowered compared to any liquid booster I can build
<zilti>
RP-1?
<schnobs>
RP-0 for ksp131
<Bornholio>
RO Dev branch and RP0 Development branch
<zilti>
Ah
<schnobs>
last time around, I did "git clone" on all required repositories.
<schnobs>
I guess I now have to add a master before i can pull updates?
<zilti>
I guess I'm gonna wait until there's a stable release ^^ Took me long enough to get my mods together without conflicts and slowdowns for 1.2.2
<zilti>
Also, I upgraded an engine and replaced another in my Luna rocket, and suddenly, I have an additional 2k dV xD
<schnobs>
why oh why is it necessary that we D/L our mods through https?
<schnobs>
Most of my CKAN issues are certificate errors, an I wonder why we need to bother with these in the first place.
<Bornholio>
do you have CKAN Ben?
<Bornholio>
Git is the reason BTW
<schnobs>
There's ckan bruce now, released yesterday.
<zilti>
Ah, that's why my CKAN still doesn't work
qwertyy_ has quit [Ping timeout: 198 seconds]
schnobs has quit [Ping timeout: 207 seconds]
Starman4308 has quit [Ping timeout: 180 seconds]
<awang>
Mike`: Yep, looks like I'll be using htop to monitor memory usage from now on
<Qboid>
awang: Probus left a message for you in #RO [24.02.2018 12:25:49]: "So what did you think of an Ironman mod?"
<awang>
Maybe I should post a stack overflow question asking why
<Probus>
htop?
<awang>
Probus: An iron man mod is an interesting idea, but not sure how necessary it is, given how players tend to impose limits on themselves already
<Probus>
I give in to temptation way to often :)
<awang>
Well
<awang>
Actually
<awang>
Saving after part destruction could be interesting
<awang>
As long as there aren't any other mods creating intermediate saves
<awang>
Hmmm
<awang>
I'm not really sure, to be honest
<awang>
Bornholio: IIRC commsat payload is just the number you see when you right-click on the tank
<awang>
1 unit payload = 1 liter, I think?
stratochief has joined #RO
<Bornholio>
in how many tanks...
qwertyy has joined #RO
<Bornholio>
KER sums it up, and places it at a glance
probus_ has joined #RO
Probus has quit [Read error: Connection reset by peer]
stratochief has quit [Remote host closed the connection]
<awang>
Ah
<awang>
I've always stuck my payloads into a single tank
<awang>
So never had to deal with totalling the amount of payload
<awang>
Wonder if ship manifest could take care of that
<awang>
Later Unity versions can't really work with earlier ones, right?
BasharMilesTeg_ has joined #RO
oeuf has joined #RO
egg|zzz|egg has quit [Ping timeout: 186 seconds]
BasharMilesTeg has quit [Ping timeout: 186 seconds]
<awang>
YES
<awang>
I GOT PROFILING WORKING
<awang>
PERFORMANCE ISSUES, HERE I COME
stratochief has joined #RO
<awang>
Apparently profiling is not going to be fun :(
<blowfish>
this is putting a real damper on my motivation to mod today (and possibly ever again)
ProjectThoth has joined #RO
<Bornholio>
.sigh
<ProjectThoth>
"sigh" is not a recognized command. Type .help for a list of useful commands.
<awang>
Mike`: Had to tweak Sarbian's instructions for getting debugging/profiling set up for Visual Studio on Windows
<awang>
Mac instructions didn't work at all
<Mike`>
err okay :)
<awang>
You want to get profiling/debugging set up too?
<awang>
Looks like memory/heat profiling is possible in theory
<Bornholio>
awang so finally have ksp runningwith your dll going, lots of no TestFlight core message even though TF seems to be operating normally, which means lots of slow down with multiple engines
<awang>
Whether it actually works is something else entirely
<awang>
Bornholio: That's not good. Is it RP-0 throwing those exceptions?
<awang>
Or KRASH?
<Bornholio>
not in krash
<Bornholio>
i do have KRASH installed but have not used it
<awang>
Hm
<awang>
What's the stack trace?
<Bornholio>
i'll post a log after this mission
<awang>
Alright
<awang>
Hmmm
<awang>
Scatterer seems to be spamming NREs in the space center scene
oeuf is now known as egg
blowfish has quit [Quit: Leaving]
stratochief has quit [Remote host closed the connection]
<awang>
Well
<awang>
profiling has proven to be quite useless
<awang>
"Unaccounted - 90.5%"
<awang>
:(
<awang>
Going to try again after removing some AutoAction and updating scatterer
<awang>
Something in AutoAction spams UnityExceptions in the editor due to some method being called in a MonoBehaviour constructor instead of Awake or Start, despite nothing out of the ordinary in the source code
<awang>
Don't know why TF can't find a core, so added messages to see exactly what it's doing
<awang>
Also took out some of the other logging messages to reduce noise
<Bornholio>
k
Shoe17 has quit [Quit: Connection closed for inactivity]
<Bornholio>
Oh I officially hate Ignition failures
<ProjectThoth>
Bornholio: Sorry for leaving you hanging yesterday, I was going to ask about pencil sharpeners.
<Bornholio>
electric :P
<Bornholio>
and sandpaper paddles for drafting pencils
lamont has quit [Ping timeout: 207 seconds]
<awang>
Bornholio: How many engines do you have?
<Bornholio>
anything with more than 6 IE going past leo
<awang>
Also, for the second log you sent me
<awang>
How much flight data do you have on those engines?
<Bornholio>
had a triple ignition fail trying to circularize with high data Aj(mids)
<Bornholio>
and even one ignition fail on these missions is a full failure
lamont has joined #RO
<awang>
Right, but what about the S-3?
<awang>
TF says you have 2500 data, and I want to ensure that that is actually right
<Bornholio>
it was maxed at 2500 I think
<awang>
Alright, that's fine then
<awang>
Hmmm
<awang>
I might have an idea why
<awang>
Might just be the way probability works
<awang>
Probability of at least one failure = inverse of probability of all engines succeeding
<Bornholio>
P(T)=P1*P2*P3*P...
<awang>
So 1 - (product of all engine failure probabilities)
<awang>
Yeah
<awang>
So for the S-3s, you have a max reliability of 87%
<Bornholio>
pretty sure this is why agathorn turned it off
<awang>
So for 3 S-3s you have a ~34% chance of at least one failing
<Bornholio>
still even with that probs seem skewed
<ProjectThoth>
That seems a bit... high.
<awang>
Well
<ProjectThoth>
(I settled on the CARL Angel-5, even though that thing leaves dents in the lacquer).
<awang>
Oh, another thing I noticed
<awang>
apparently the ignition curve is maxing out at 100000
<awang>
Even though the data cap is 2500
<awang>
So your ignition chance for the S-3s is 0.8513
<awang>
So with three of them you got a nearly 40% chance of at least one failing
<awang>
So at least one bug: curves aren't scaled to data caps correctly
<awang>
At least from what it looks like in the logs
<awang>
And I don't think disabling it was intentional
<awang>
It broke as part of refactoring, I think
<awang>
Only way to know if the probabilities are right is to do some experiments, but getting a lot of samples would be incredibly tedious
<awang>
Guess the ideal way would be to find some conbination of engines with a ~50% chance of at least one failing, try to ignite a few hundred times, and compare the probabilities
<Bornholio>
I did 210 engines with 56 failures (those 2500dp S-3's)
<awang>
Yeah, then the failure rate seems a bit high
<awang>
idk what the std dev on that number is though
<awang>
Bornholio: The MM cache says the max data for S-3 is 10000
<Bornholio>
the 40 engine test rig, is at an amazing crawl
<Bornholio>
barely loads
Probus__ has quit [Quit: Leaving]
<Mike`>
awang, i thought all engines caps were 10k?
<Bornholio>
I'm going to make a chart of engine burn times and just stop using TF .sigh So tired of it making the game crawl with just a few more engines
<Bornholio>
even when they are turned off
<awang>
Mike`: Apparently it is, but Bornholio said S-3 data was maxed at 2500, so I though perhaps I was just wrong
<awang>
Bornholio: I'm working on it
<awang>
Slowly
leudaimon_ has joined #RO
<awang>
I think part of the problem is that the probabilities multiply up depressingly fast
leudaimon has quit [Ping timeout: 186 seconds]
<awang>
....I think I broke KSP
<awang>
I told it to quit over an hour ago
<Bornholio>
your test dlls are very harsh, picking engines in VAB throws a lot and lots for several seconds with each move
<Bornholio>
locks
<Bornholio>
ok 8 engine test lr43-na-3 10000dp
<awang>
Uhhh
<awang>
I was not expecting that much of a slowdown
<awang>
Really need to learn how TF is architected so I can see how to improve performance
<Bornholio>
with 8 engines i'm running at 1:10 at best
<awang>
Wow
<awang>
Probably going to have a really big log, too