Guild icon
DDraceNetwork
Development / developer
Development discussion. Logged to https://ddnet.tw/irclogs/ Connected with DDNet's IRC channel, Matrix room and GitHub repositories β€” IRC: #ddnet on Quakenet | Matrix: #ddnet-developer:matrix.org GitHub: https://github.com/ddnet
Between 2019-07-29 00:00:00Z and 2019-07-30 00:00:00Z
Avatar
@Theowyn people have been asking for a better mark for afk people since I started coding on ddrace 7-8 years ago
06:29
I think delivering sth is better then delivering nothing and asking people to git good
Avatar
@Learath2 true, I can make a pr for the emote mark if needed
Avatar
Spamming zzz?
06:34
That's hardly better and it's server sided
Avatar
not spamming
06:35
displaying the bubble permanently, using playerflag to render clientside
Avatar
Dont you think that might get crowded on maps like bit2/s3 (t0 maps where this is most needed) where there could be a lot of tees afk?
06:38
Tbh best would be if we could mark the tee itself somehow, but I couldnt come up with anything that isnt terribly ugly
Avatar
I guess I will have to look at these maps to see what you mean
06:40
I am not even saying that the mark as is right now is bad, the emote way was just a more noticeable way of displaying it
Avatar
Yes, I also considered the emote but was thinking it could be way too crowded to see which tee is actually afk
06:40
The mark also fails at that though I guess
Avatar
exactly
06:41
But other ways like changing skins to display afk are ugly
Avatar
Maybe a red "aura" around the tee?
06:42
Sounds ugly tho :(
06:42
Coloring the tee outline red?
06:42
I wonder what that'd look like
06:43
I tried colorint the eyes, looked very weird
Avatar
Can't imagine it being that bad tbh
06:44
I'm going to try the emote way later and will test it in crowded areas
Avatar
Eyes were gud imo
Avatar
@archimede67 eyes we had since the beginning, it's just not that obvious that people always wanted sth better
06:49
but even worse is that you can override it with eye emotes
06:49
tbh the eye emotes were not a good idea to implement when we were using the eyes to signify tee state
Avatar
So maybe change it?
06:53
So that the tee state overrides the tee's eye emote
Avatar
that feature already lost it's trustw
06:54
trustability
Avatar
Ah?
Avatar
people just dont trust the eye emote, as evidenced by the dozens of people failing a day trying to pick up an afk
06:55
or they just can't read the tee state from it easily, either way it's just not that good
Avatar
Yea :/
Avatar
it's very inconsistent tbh, an afk tee in freeze has different eye emotes then an afk tee out of freeze
06:57
so now you need a mental map, blink in freeze is not afk, blink out of freeze is afk
Avatar
Yea
06:57
Like not in freeze u have eyes open and in freeze eyes closed
Avatar
I'm pretty sure we'll add a consistent mark this release, so if anyone comes up with a better mark, do it before we release πŸ˜›
Avatar
For afks or for normal tee's aswell?
Avatar
Ok
Avatar
Why would we mark the non-afks, they are the minority
07:00
that'd be bad ux πŸ˜›
Avatar
xd
07:01
No i thought u wanted to change the ways its marking the non-freezed and freezed tees
Avatar
I think we'll keep the blink to keep the greybeards happy
Avatar
Ye also bcs i think a lot of ppl are used to this
07:04
I'll try to think about something for the afks but the main problem is to find something that renders well a lot of tees are close to each others
Avatar
kinda why I want it to be on the tee, rather then above or next to it
Avatar
Yea
08:16
that's the most crowded I could get rn
Avatar
Hmm that doesnt look half bad
08:23
We should open an issue with screenshots from all solutions
Avatar
did u keep the eyes?
08:23
i think this and the eyes could be something good?
Avatar
yeah kept the eyes
Avatar
as the eyes are closed
Avatar
the eyes are just open in screenshot since I am not using playerflag to display them rn
Avatar
ah k
Avatar
also these are just chatbubble position, if anyone has some suggestion on it
Avatar
Sleep bubble best
Avatar
this is nice imo
Avatar
how does the afk playerflag work? because for me it doesn't
Avatar
[quakenet] deen BOT 2019-07-29 10:09:27Z
the server needs to support it
10:09
current ddnet servers don't run that version yet
Avatar
@Theowyn the master branch has it now
Avatar
yeah I tried it with the master branch, maybe I did something wrong
Avatar
you tried it on official servers or your own?
10:23
official servers are not up to date atm
Avatar
I compiled master and tried it on my own
Avatar
[quakenet] deen BOT 2019-07-29 10:43:06Z
ok, that should work
Avatar
still, someone needs to fix up the AFK flag
10:49
it doesnt belong to the player flags
10:51
@Theowyn it cannot work, the DDNetPlayer doesnt have the AFK flag applied
10:53
@deen you definitely merged too fast
Avatar
@Learath2 inb4 master breaks πŸ˜›
Avatar
0bc3650 Send DDNet capability flag - Learath2
Avatar
@fokkonaut DDNetPlayer does have the afk flag applied
11:23
player.cpp L327
Avatar
@Learath2 but the AFK flag doesnt belong to playerflags?
11:24
Or does it?
Avatar
I don't know what I am doing wrong then
11:25
the afk mark seems to just work fine that means that m_Afk is true and the flag should be applied
Avatar
No it definitely needs a new flag array
11:26
But the flag is applied, its just unnecessarily in the middle of the flags int
Avatar
ah, yes
Avatar
Whats the issue @Theowyn?
11:27
It is possible that m afk is forgotten uninitialized
Avatar
no m_afk works
Avatar
@Learath2 Could you look at my prs before the release? They are very small
Avatar
but I am trying to display the "sleepbubble" the way you display the chatbubble via PLAYERFLAG_AFK
11:28
nothing happens
Avatar
@fokkonaut I'm having lunch, then ill head to a cafd tk get some studying done, i'll merge them there
Avatar
oh okay
Avatar
@Theowyn ah thats simple
11:29
Wrong playerflags int
11:29
The afk flag is in the extended player object
Avatar
@Theowyn i find it kinda meh to have something that vanilla client cant display
Avatar
@fokkonaut wrong way to think about it
Avatar
https://github.com/ddnet/ddnet/pull/1844 – says that it creates an inconsistency between scoreboard and kill messages
Since in DDrace every player is in team red, in order to show them in one scoreboard, the flag will always be blue in scoreboard and the spectator select menu, here is a simple fix I left kill mess...
Avatar
if we were to approach this way 64 players shouldn't be a thing either
11:31
@Learath2 oh thank you
Avatar
Well we did let vanilla clients see the 64 players
Avatar
@heinrich5991 yes, because kill messages would need a fix that would require more changes
Avatar
yeah that was just an example
Avatar
But back then people did care about vanilla compat
Avatar
vanilla can't display hookline either (edited)
Avatar
Tbh vanilla is irrelevant now
Avatar
I hope you guys have checked the data
Avatar
wat data
Avatar
@Ryozuki @Learath2
Avatar
we can figure out how many people connect with non-ddnet clients
Avatar
well, deen and whoever has log access can
Avatar
I can take a look if you want, we have all that teehistorian data
Avatar
@Ryozuki yes, and you can ask someone
Avatar
But id be surprised if more then 5% use it
Avatar
and
Avatar
most use the latest ddnet
Avatar
I don't think we have any way for newcomers to come into the game except vanilla
11:34
ddnet is already declining judging from races finished graphs
Avatar
is there a way to know how many newcomers we have?
11:34
tbh i dont think we got much anyway
11:34
and we doing them a favour by not playing ddnet with vanilla
Avatar
I doubt we have many newcomers aswell, and its hard to quantify
Avatar
they can't find ddnet in the first place if we don't allow them to join with vanilla
Avatar
to fix that we could release on steam
Avatar
But we could take a look at new names in novice finishes
Avatar
have our own proper visibility
11:35
it was the plan
11:35
till
Avatar
@Ryozuki yea, but before that happened, I'd be a bit slower on dropping vanilla
Avatar
something happened
Avatar
yes
Avatar
ye we can slowly drop vanilla
Avatar
Most newcomers i hear of are referred by friends
Avatar
ddnet still sends fake tunezones due to vanilla right?
Avatar
not only vanilla, also ddnet I think(?)
Avatar
I doubt anyone would let their friend use the pain thats vanilla on ddnet :D
Avatar
we should fix that for ddnet
11:37
i think u cant still predict
11:37
a tune zone
11:37
when ur outside
Avatar
vanilla players use ddnet client too
Avatar
@Ryozuki look at my pr
11:37
i at least fixed the projectiles
Avatar
I dont think we send tunezones or predict them at all
Avatar
@fokkonaut diffs can be small and the change can still be hard btw
Avatar
But it should be doable now that we have extended objects and have the tunezones in the map itself
Avatar
e.g. https://github.com/ddnet/ddnet/pull/1840, this creates a new network message and kinda implies we want to go that way for syncing settings between client and server
11:38
also, show_all has an impact on the amount of network traffic. so it's not easy to evaluate
Avatar
We do the same for showothers
Avatar
yes
Avatar
DDraceNetwork, a cooperative racing mod of Teeworlds - ddnet/ddnet
11:39
i think most of it
11:39
is coverd by ddnet network characer
11:39
now
Avatar
that doesn't mean it's good. show_others doesn't even use the new netmsg thing
Avatar
We should probably get one netobject with all of the settings packed in
11:39
That should better utilize the delta compression
Avatar
https://github.com/ddnet/ddnet/pull/1832 – it's unclear whether third-party ddnet servers enable it, and it's even on by default for these
Avatar
Yeah i was going to close 1832
Avatar
https://github.com/ddnet/ddnet/pull/1822 – this greatly increases network traffic for projectiles, even for the ddnet client AFAIUI
Different players can have different tunings for projectiles now, without them looking buggy. The projectile will reset every 4th tick. I changed the VERSION_DDNET_ANTIPING_PROJECTILE, because we n...
11:40
not so easy
Avatar
why should it?
Avatar
@heinrich5991 that one I already told him I wont merge until proper support in ddnet client exists for it
Avatar
I'm just trying to explain to @fokkonaut that PRs aren't easy because they're small
Avatar
Oh yeah true
Avatar
I know that
Avatar
Its not trivial tk evaluate these always
Avatar
@heinrich5991 about the flag colors: is it important to have the kill messages correct too?
Avatar
I appreciate your honesty there btw, I probably wouldn't have remembered that it also shows for the kill messages
11:43
hm
Avatar
The point is
Avatar
oh
11:44
the scoremessages have a flag flag, right?
11:44
so it's not really clear which flag to show
Avatar
For the kill messages to show up with flags is, that you would need to add flag support first, then enable the gamestate_flags and then even decide which tee has a flag
11:45
scoremessage?
Avatar
the kill messages
Avatar
What do you mean with a flag flag
11:45
There is a FLAG
11:45
which tells the client to render flags in scoreboard and spectator select menu
Avatar
I believe the m_ModeSpecial field contains the info whether there should be a flag on the killer and the victim
11:46
in Sv_KillMsg
Avatar
It does, correct
Avatar
but it's not clear whether it should be a red or a blue flag
Avatar
Yes, and that is because a KillMsg only has the Team of the tee providrd
Avatar
ah
Avatar
and based on that, it will render the opposite flag
Avatar
I guess it's fine then
Avatar
In ddrace it would always be a blue flag
11:48
For example
11:48
We could fake it tho
Avatar
If we go and add another var to the KillMsg struct so we know the flag
Avatar
Could you check for the flag the same way in killmessages?
11:51
The kill message updates
11:51
So it wouldnt work
Avatar
You could hack it in
11:51
When first adding the kill message to the array
Avatar
yeah exactly what I had in mind
Avatar
I'm gonna create a pr for the sleepbubble now should I make it an option of ClAfkMark?
Avatar
@Theowyn yes pls
Avatar
Wait
Avatar
and the mark should be default?
Avatar
We'll decide on that pr, then we can move around the setting to get the desired result
Avatar
@Theowyn How does it behave if one has opened the chat?
11:52
Does it react and say, no more afk then?
Avatar
good question I think chat overrides
Avatar
the chat bubble should override
Avatar
I will test that
Avatar
But if one opens the chat he should be concidered not-afk again?
Avatar
I think they are still considered afk?
Avatar
@fokkonaut the chat bubble is enough to show that they cant react
Avatar
the action of opening chat will reset the afk flag on the server
11:54
which should propagate to the client and the zzz should also disappear with that
Avatar
@Learath2 Do you want to make that KillMsg thing? I could do it on Thursday maybe
Avatar
@fokkonaut I'll knock out a couple in an hour or so
Avatar
Okay
11:55
@heinrich5991 could you merge the deci second fix,m
11:55
*?
Avatar
@fokkonaut did you check to make sure this doesnt break sth like the ghost or auto demo recording?
11:57
I think the ddrace tick is local to the hud, but I'm not sure
Avatar
It is
11:57
It shouldnt break anything, but no, I havent tested
Avatar
if it's local then it can't really break much of anything else
Avatar
m_DDRaceTick is only used for the timer
11:58
Also, it is private
11:58
Cant really break anything
Avatar
@Learath2 what is explayerflag?
Avatar
ExtendedPlayerFlag
11:59
idk I'm not that good with namind
Avatar
Belongs to DDNetPlayer
Avatar
naming*
11:59
We should have maybe named DDNetPlayer more generally aswell 😦
11:59
Maybe ExtendedPlayer
Avatar
Also the Character :D
Avatar
The character is pretty ddnet specific @fokkonaut
Avatar
@Learath2 why?
Avatar
checkpoints, freeze duration, jump count
12:01
and it's interpreted in a ddnet specific way
Avatar
So whats wrong with DDNetPlayer then?
Avatar
e.g. the freeze duration goes negative, that only has meaning in the context of ddnet
12:01
the flags in ddnetplayer are concepts that aren't specific to ddnet
12:02
AFK has no special meaning in ddnet
12:02
neither does auth level
Avatar
I think it doesnt really matter
12:02
@Learath2 I think opening chat doesnt reset Afk
12:03
It only checks for the Target, and movement
12:03
afair
Avatar
i guess chat should reset afk
Avatar
I just tested it by opening chat on dummy and moving to main and dummy still got afked
12:04
thats why in the screenshot it also shows both bubbles
Avatar
compile_commands.json is used a lot for language servers tags are generated from ctags
Avatar
vim master race
12:06
please merge
12:06
xd
Avatar
@Theowyn it should only ever show one bubble pls
Avatar
@Theowyn just go and disable the zzz emote if chat is opened
12:06
PLAYERFLAG_CHAT
Avatar
yeah working on it already
12:07
was just testing if chat prevents afk
Avatar
@Ryozuki can you order it alphabetically into the gitignore?
Avatar
@heinrich5991 i thought about doing so but i noticed gitignore is not ordered at all
12:10
9061f03 Move the afk flag up front - Learath2 29e691c Merge #1859 - bors[bot]
12:10
d -> c -> d
Avatar
nah, D -> c -> d
12:10
like in ascii
12:11
ok
Avatar
in vim, you can use V do mark the lines and then do :sort
12:11
vim best
Avatar
new vim power word acquired
12:18
Avatar
the best part is that we shouldn't include <stdio.h> or <string.h> anywhere, we have <base/system.h> that replace all of that
Avatar
Are we including them somewhere?
Avatar
src/engine/shared/masterserver.cpp src/engine/shared/uuid_manager.cpp src/engine/client/opengl_sl.cpp src/game/editor/auto_map.cpp src/game/client/components/items.cpp src/game/client/components/players.cpp src/game/server/entities/character.cpp src/game/client/prediction/entities/character.cpp
Avatar
I wonder which functions we needed, maybe some debugging leftovers
Avatar
i found this in characters.cpp
12:22
not the pred one
12:23
12:23
i always end up finding this
12:23
xd
Avatar
That looks awful ;(
12:25
that's why stoppers are still broken
Avatar
not broken
Avatar
one rotation is broken
Avatar
Is it
Avatar
as in doesn't work at all
Avatar
at every place where we have velocity change we need to recheck for stoppers
12:26
like, a hammer, a shotgun..
12:26
hook
Avatar
Really??
Avatar
fixing stoppers means breaking them
12:26
i mena
12:26
breaking maps
Avatar
Is that how we implemented those?
12:26
appearently
Avatar
imo they should be fixed
12:27
that one rotation
Avatar
Why not centralize velocity changes so we only have checks in one place?
12:27
@jao yeah create an issue for it pls?
Avatar
just look for it @Learath2, gamecore.cpp, laser.cpp, Character::FireWeapon, case Hammer, you will find it everyhwere
Avatar
isnt gameworld the class that checks all collisions and stuff
Avatar
@fokkonaut is hard to do on a bus
12:28
gamecore
12:28
hm
Avatar
WHAT
12:28
how
12:28
stupid
12:28
is that
12:28
We have this CCharacterCore::LimitForce(vec2
12:29
Why dont we use it
12:29
EVERYWHERE
Avatar
u cant just fix things in ddnet
12:29
u will break maps
Avatar
@Ryozuki well you can clean up
Avatar
LimitForce literally checks for the stopa collision
Avatar
We can replace that everywhere
Avatar
@fokkonaut problem would be to make sure the huge if blocks match
Avatar
Remove the whole copy-and-paste mess. A seemingly ineffective and apparently code block has been removed.
Avatar
Ouch, this code hurts
Avatar
here's your starting point for fixing stoppers
12:31
there was a bug
12:31
but if it's fixed, you can have nicer code
Avatar
@heinrich5991 any reason you didnt use limitforce as fokkonaut suggested?
Avatar
@Learath2 tbh we got duplicate code
Avatar
Remove the whole copy-and-paste mess. A seemingly ineffective and apparently code block has been removed.
Avatar
@Ryozuki just bc we have it doesnt mean its good
Avatar
just renamed it to something more sensible
Avatar
i never implied its good
12:33
but what would be a good way to fix it
Avatar
@jao can you tell me in which directions stoppers are broken?
Avatar
I dont even remember why we have charactercore and character separate tbh
Avatar
maybe I can fix up the code then
12:34
charactercore is shared between client and server
Avatar
@Learath2 i think ccharactercode is used in client
12:34
too
12:34
ye
Avatar
ddnet wrote all the wrong code then, e.g. not doing freeze in core
Avatar
If only we did the rewrite when zeta was onboard
Avatar
zeta has a working ddnet for 0.7
12:35
clean, even
12:35
but changing physics of ddrace is not accepted
Avatar
we would lose all maps
Avatar
@heinrich5991 ^
Avatar
[2:33 PM] heinrich5991: @jao can you tell me in which directions stoppers are broken?
Avatar
check the issue
Avatar
has 0.7 more security than 0.6 rn?
Avatar
yes
Avatar
can we backport it
Avatar
Lets move to 0.7 :p
Avatar
most of the issues will be gone with http masters
12:37
no need to update then
12:37
cleaning up ddnet code would be cool
12:37
and docs for newcomers
12:37
but that takes lot of time
12:37
which nobody has
12:38
i would also modernize it to c++11 but some ppl here like to use 0 as nullpointers still
12:38
monkaS
Avatar
whats the difference
Avatar
@Ryozuki I have a rule for those
Avatar
what does a nullptr do
Avatar
if I'm touching code in old files I use 0, if I'm creating a new file I use nullptr
Avatar
nullptr is a pointer-typed null pointer
Avatar
atleast the new code would be modern that way
Avatar
means what
Avatar
you can't assign it to an integer, for starteres
12:39
AFAIK
12:39
not sure
Avatar
C++11 corrects this by introducing a new keyword to serve as a distinguished null pointer constant: nullptr. It is of type nullptr_t, which is implicitly convertible and comparable to any pointer type or pointer-to-member type. It is not implicitly convertible or comparable to integral types, except for bool.
Avatar
@heinrich5991 I don't think C++ went as far to remove the cast from pointer types to intptr_t
Avatar
int X(void *foo); int X(int foo); X(0); // int X(NULL); // int X(nullptr); // void * (edited)
Avatar
ok, that's a shitty fix
12:40
one more reason to hate c++
Avatar
a.cpp:2:10: error: cannot convert β€˜std::nullptr_t’ to β€˜int’ in initialization 2 | int i = nullptr; | ^~~~~~~
Avatar
Love how they like breaking things just for the sake of correctness
Avatar
@Learath2 how is ur hate for systemd today
Avatar
I'm sure people converted pointers to ints while some of the c++ committee members were still in kindergarten
Avatar
u surte
12:42
arent comittee members supero ld ppl
12:42
old*
12:42
xdd
12:42
i always imagine old ppl with white long beard
12:42
poggers
Avatar
well if they are they are prolly the kinda people that give speeches on the merits of oop
Avatar
new breed of pseudo-intellectual knowitalls
Avatar
@Learath2 what do u think about having a button class and it accepting a lamdba function onclick
12:44
i always imagined this, rn its all logic in the same function in a loop i think
12:44
for all buttons
Avatar
depends on how you are structuring your program
Avatar
its something i wanted if i ever rewrote the ui
Avatar
ui is hard
Avatar
e.g. I usually have lambda functions that trigger actions, which modifies state, which is fed back into the ui
12:45
the flux pattern is very performant in react
12:45
and works predictably
Avatar
u mean
12:46
react-redux
12:46
ah
Avatar
Application architecture for building user interfaces
12:46
redux is an implementation of flux yeah
12:46
but it's usually overkill for small projects
Avatar
i guess
12:46
i always end up using it
12:46
its easy to manga login state
12:46
manage*
Avatar
for projects on pc I prefer the imgui pattern
12:47
mixing the logic with the ui creates very easy to follow code imho
Avatar
imgui-style has a problem of "ui lags one frame behind" IIRC
Avatar
well it can't actually be perfectly sync either way
12:49
it's either a frame behind or one frame ahead
Avatar
btw right now the menus calculate the position on each render call, this is due to it using functions like dolabel instead of what could be a label class holding info about its last calculated position
Avatar
@heinrich5991 why was your stopper rewrite reverted?
Avatar
because it had a bug
Avatar
it let people through in some direction in some map
12:50
@Learath2 why can't it be in sync?
Avatar
@Ryozuki it's performant enough in this case, for cases where it matters we did optimize them
12:51
like text containers
12:52
@heinrich5991 with the imgui pattern you render the current state, and when a change happens you only know about it the next frame
12:53
with the rmgui pattern you render the state you have in the ui code, and when a change happens the ui state changes before the application state
Avatar
With setting cl_afk_mark to 2 you will now get a sleepbubble on afk instead of the mark. Which option should be default? Also it displays chatbubble instead of sleepbubble when the playerflag is active. (I don't know which behavior is preferred) Kept sleepy eyes on afk aswell. !screenshot_2019-07-29_14-47-34
Avatar
@heinrich5991 ever debugged the stopper thing?
Avatar
no
12:54
it was just reverted, I didn't know which map it was and it was during a time where I couldn't spend time on it
Avatar
well deen reverted it but he didn't put which issue he closed in the commit description 😦
Avatar
there was no issue
12:56
it was a hotfix, people started finishing maps really quickly
Avatar
yeah found the irclogs
12:57
@Ryozuki you seem to have reported it, remember what it was?
12:57
the screenshot links seem dead
Avatar
report what?
Avatar
Some issue with the stopper fix that seems to have broken quite a lot of tile handling
Avatar
can usend the irclog
12:59
i dont remember tbh xD
Avatar
deen u saw that yesterday night server ignored front layer on a brutal map? found it
Avatar
well then its that
Avatar
apparently you killed the entire front layer @heinrich5991 not just stoppers πŸ˜„
Avatar
oh ok πŸ˜„
13:00
πŸ™‚
13:00
this slight smile is better
13:00
its weird
Avatar
I'll try to find this just looking at the diff, if I can't in 10 minutes I'll go back to studying math
13:01
ok it was this one
13:01
looks so dumb
Avatar
I find πŸ˜„ appropriate for :D but not for :)
Avatar
1bfda0f Fix deci seconds not updating in spectator/pause (fixes #1846) - fokkonaut 3a58918 Ignore some more common files - Ryozuki 8ab6e2c Merge #1848 #1860 - bors[bot]
Avatar
yeah :) should def be πŸ™‚
Avatar
yay time to fetch upstream
13:06
bors is so slow
Avatar
@heinrich5991 don't you have a tool to check for specific tiles in maps? we could check all maps then
Avatar
deen agrees with bubble so bubble we have troll
Avatar
should we have it displayed on paused tees aswell?
Avatar
@Ryozuki it's not bors that is slow, it's our build process
13:12
on windows
13:12
so more like appveyor
Avatar
in the original pr that was a topic but op said it's messy to do so
Avatar
hehe windows bad
Avatar
it's the shitty excuse for make msvc has
13:12
nmake or sth
Avatar
isnt there a faster thing than appveyor
Avatar
We could host buildbots on a server
13:15
@heinrich5991 couldn't spot the bug πŸ˜› I'm going back to math
13:21
@heinrich5991 do you happen to know a good resource to learn laplace transforms?
13:23
frozen
Avatar
@Learath2 what do you want to learn about laplace transforms?
13:24
how to do them?
Avatar
Assume I know nothing
13:25
I'd like to learn how to do them, what they are for, etc.
Avatar
do you know the fourier transform?
Avatar
well I know as much as the video @Ryozuki sent says about them
13:26
I know of the concept, but never used one
13:26
are they a prerequisite?
Avatar
no
13:26
they're similar
Avatar
" the Fourier transform of a function is a complex function of a real variable (frequency), the Laplace transform of a function is a complex function of a complex variable. "
13:27
according to wikipedia
Avatar
okay, so from a mathematical perspective, I only know one reason to do these transforms
13:27
namely to solve differential equations
13:28
the idea is that they represent some sort of bijective mapping on functions
13:28
e.g. you want to solve f'(x) - f(x) - sin(x) = 0 (don't know if that works) (edited)
13:28
and now you apply the laplace transform on both sides
13:29
and if you get a solution for that, you get a solution for the original equation
Avatar
I'm guessing that's because the transform is bijective and you can apply the inverse to get the answer
13:30
And we are just hoping the initial transform gets us an easier to solve equation?
Avatar
yes
13:30
the good thing is that the laplace transform throws out differentials
13:30
i.e. L[f'(x)] doesn't contain f'(x)
Avatar
oh that's convenient
13:31
what about L[f''(x)]?
Avatar
so we might have reason to believe that the resulting equation might be easier to solve
Avatar
it only discards one degree or is the resulting equation not differential at all?
Avatar
not differential at all
13:32
according to wikipedia, L{f'(t)} = s·L{f(t)} - f(0)
13:32
with this, you can derive a formula for L{f''(t)}
Avatar
okay, now that we have the why down, where can I learn the how? πŸ˜›
Avatar
I don't remember my course on that too well
13:34
it basically boiled down to memorizing the calculation rules of the laplace transform
13:34
IIRC
13:34
and applying them, over and over again
13:34
maybe you have some exercises from the class?
Avatar
cl_http_timeout default is too strict for people with bad internet.
Avatar
@heinrich5991 well jumping straight into exercises sound silly to me
Avatar
hm. do you have lecture notes?
13:35
(i.e. I don't have any materials to show you)
Avatar
All I know about laplace transforms are that they are the indefinite integral of the function multiplied by exp(-st) dt
13:36
Let me check if this guy posts notes
Avatar
you need to memorize some rules
13:36
e.g. the derivative rule
Avatar
only have exercises :/
13:39
I guess I'll have to resort to Salman Khan once more
Avatar
Here is my curl output: curl -o /dev/null -w "%{time_starttransfer}\n" https://info.ddnet.tw/info % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 6860 0 6860 0 0 1286 0 --:--:-- 0:00:05 --:--:-- 1467 5.333923 Also https://github.com/ddnet/ddnet/issues/1863
Avatar
why is the flag color in GameData πŸ˜›
14:21
flag carrier*
Avatar
cl_friends_ignore_clan should default to false
15:15
i mean
15:15
to true
15:15
1
Avatar
@heinrich5991 if you have a moment can you help me with that?
15:29
Idk how to simplify that sum
Avatar
:head_sideways:
15:31
sorry, need to leave soon
15:31
will get internet back in 30-45min
Avatar
Nvm forgot it was a definite integral
Avatar
hahahahahhaha
Avatar
@Gwendal 🀑
🀑 1
Avatar
ChillerDragon 2019-07-29 19:01:17Z
πŸ˜„
Avatar
I didn't get it justatest
Exported 570 message(s)