11:08 <+bridge> [ddnet] <Ryozuki> cool read https://eb2.co/blog/2014/04/c--14-and-sdl2-managing-resources/
12:50 <+bridge> [ddnet] <heinrich5991> interesting read, ty 👍
12:52 <+bridge> [ddnet] <Learath2> unique_ptrs are great
12:52 <+bridge> [ddnet] <Learath2> IMHO they don't need the extra layer of abstraction though
13:28 <+bridge> [ddnet] <timakro> @heinrich5991 Hey, where to I get client version from as a 0.7 server? I wnat to do what you discourage here https://github.com/teeworlds/teeworlds/pull/2430#issuecomment-578539436 ^^
13:28 <+bridge> [ddnet] <timakro> @heinrich5991 Hey, where do I get client version from as a 0.7 server? I wnat to do what you discourage here https://github.com/teeworlds/teeworlds/pull/2430#issuecomment-578539436 ^^
13:34 <+bridge> [ddnet] <heinrich5991> @Learath2 the article or `unique_ptr`?
13:35 <+bridge> [ddnet] <Learath2> well the article adds a layer of abstraction to unique_ptrs, which I don't think they need
13:36 <+bridge> [ddnet] <heinrich5991> @timakro ah, slightly misunderstood the protocol change
13:36 <+bridge> [ddnet] <timakro> yeah i guess, but i dont see how to get client version in 0.7
13:36 <+bridge> [ddnet] <heinrich5991> you can, but don't
13:37 <+bridge> [ddnet] <heinrich5991> maybe you could adjust the protocol so it uses an extra bool instead of another enum variable
13:38 <+bridge> [ddnet] <Learath2> @timakro what are you trying to do anyway? compatibility hack?
13:38 <+bridge> [ddnet] <heinrich5991> he's trying to make a race server that sends the correct message to <0.7.4 and >=0.7.5
13:39 <+bridge> [ddnet] <heinrich5991> the problem is that I slightly misunderstood the protocol change in that pull request
13:39 <+bridge> [ddnet] <timakro> adding another bool is smart
13:39 <+bridge> [ddnet] <timakro> 0.7.6 is not yet released, right
13:39 <+bridge> [ddnet] <heinrich5991> 0.7.5
13:39 <+bridge> [ddnet] <timakro> right
13:39 <+bridge> [ddnet] <heinrich5991> so better make that PR soon 😛
13:39 <+bridge> [ddnet] <Learath2> what is the extra bool for anyway?
13:40 <+bridge> [ddnet] <heinrich5991> it indicates whether it was not only personal best but also global best
13:40 <+bridge> [ddnet] <Learath2> I see
13:41 <+bridge> [ddnet] <Learath2> Do you think the SDL people would accept a patch with a new event type that'd only be used on macOS?
13:41 <+bridge> [ddnet] <heinrich5991> I'd guess so
13:41 <+bridge> [ddnet] <heinrich5991> can't you map it to an existing one?
13:42 <+bridge> [ddnet] <Learath2> the accent menu doesnt give any indication that it's started nor does it give an indication that it's ended, so I can't make it into a SDL_TEXTEDITING event without hooking into some internal stuff, which I think they'd hate more
13:43 <+bridge> [ddnet] <Learath2> I could maybe add a new field to SDL_TEXTINPUT
13:43 <+bridge> [ddnet] <heinrich5991> I mean the best way forward would be to have some sort of communication channel to them
13:43 <+bridge> [ddnet] <heinrich5991> do you have something like that?
13:44 <+bridge> [ddnet] <Learath2> I guess I could file a bug report for discussion on the subject
13:45 <+bridge> [ddnet] <heinrich5991> do you know their response times to issues? ^^
13:45 <+bridge> [ddnet] <Ryozuki> freenode has a #sdl i think
13:45 <+bridge> [ddnet] <Ryozuki> https://discordapp.com/invite/F6mPUgc
13:45 <+bridge> [ddnet] <heinrich5991> SDL Enthusiasts sounds a bit inofficial
13:45 <+bridge> [ddnet] <heinrich5991> do you know whether it is official?
13:46 <+bridge> [ddnet] <Ryozuki> i dont think its official but maybe someone important is there
13:46 <+bridge> [ddnet] <Ryozuki> there is always the freenode #sdl
13:46 <+bridge> [ddnet] <Learath2> There is a guy who answers blazing fast to bugs, but he is the only one that quick
13:46 <+bridge> [ddnet] <Ryozuki> could that guy be https://twitter.com/icculus
13:46 <+bridge> [ddnet] <Ryozuki> ?
13:46 <+bridge> [ddnet] <Learath2> freenode #SDL doesn't have any of the people familiar with macOS internals
13:47 <+bridge> [ddnet] <Learath2> yeah icculus is the macOS guy
13:47 <+bridge> [ddnet] <Learath2> my last patch is still awaiting his feedback
13:47 <+bridge> [ddnet] <Ryozuki> icculus is like the lead dev i think
13:48 <+bridge> [ddnet] <Ryozuki> https://www.youtube.com/watch?v=MeMPCSqQ-34 :)
13:49 <+bridge> [ddnet] <Ryozuki> https://xkcd.com/1782
13:51 <+bridge> [ddnet] <Learath2> Why does it always end like this? I start out fixing a teeworlds bug, due to some mess macOS created I start fixing an SDL bug which inevitably leads to disassembling AppKit with it's 100 years worth of compatibility hacks
13:52 <+bridge> [ddnet] <Ryozuki> @Learath2 its cuz i care to much about macOS :D
13:52 <+bridge> [ddnet] <Ryozuki> you**
13:52 <+bridge> [ddnet] <Learath2> well it's the only machine I have access to right now
13:52 <+bridge> [ddnet] <Ryozuki> cant u just install linux there
13:52 <+bridge> [ddnet] <Learath2> don't have the storage for it
13:53 <+bridge> [ddnet] <Ryozuki> i mean, remove completly macos
13:53 <+bridge> [ddnet] <Learath2> I do have a small virtual machine for compiling though
13:53 <+bridge> [ddnet] <Ryozuki> :P
13:53 <+bridge> [ddnet] <Learath2> I need macOS to write iOS apps
13:53 <+bridge> [ddnet] <Ryozuki> do it through a vm then
13:53 <+bridge> [ddnet] <Ryozuki> how often do u write ios apps
13:53 <+bridge> [ddnet] <timakro> whats tw sever tickspeed? 60 hz?
13:53 <+bridge> [ddnet] <Ryozuki> 50
13:53 <+bridge> [ddnet] <Learath2> 50
13:53 <+bridge> [ddnet] <Ryozuki> iirc
13:54 <+bridge> [ddnet] <Learath2> @Ryozuki not much anymore tbh
13:54 <+bridge> [ddnet] <Ryozuki> 🤠 🔫
13:54 <+bridge> [ddnet] <Ryozuki> xd
13:54 <+bridge> [ddnet] <Ryozuki> whats this
13:54 <+bridge> [ddnet] <Ryozuki> https://cdn.discordapp.com/attachments/293493549758939136/695602096325787668/unknown.png
13:55 <+bridge> [ddnet] <Learath2> it's a watergun because american people think emojis dictate societal rules
13:55 <+bridge> [ddnet] <timakro> anybody knows how float precision works? big can the number go until i loose 0.001 precision?
13:55 <+bridge> [ddnet] <Ryozuki> xddd
13:55 <+bridge> [ddnet] <timakro> anybody knows how float precision works? how big can the number go until i loose 0.001 precision?
13:56 <+bridge> [ddnet] <Learath2> @timakro floats are more precise the closer you are to 0
13:56 <+bridge> [ddnet] <timakro> i know that much xD
13:56 <+bridge> [ddnet] <timakro> i mean how big does it have to be when the 3rd decimal becomes inaccurate
13:57 <+bridge> [ddnet] <timakro> and can i reach that with tickspeed
13:57 <+bridge> [ddnet] <timakro> of teeworlds
13:58 <+bridge> [ddnet] <timakro> wait are they? floats always confuse me
13:59 <+bridge> [ddnet] <timakro> The number of places of precision for float is the same no matter what the size of the number. Data type float can represent numbers as big as about 3.4E+38. But the precision of these large numbers will also be about 7 decimal digits.
14:00 <+bridge> [ddnet] <Learath2> As numbers get larger and larger you'll use less and less of the mantissa for precision
14:01 <+bridge> [ddnet] <timakro> i dont think thats how floats work
14:01 <+bridge> [ddnet] <Learath2> well I don't see any other way of them working
14:01 <+bridge> [ddnet] <Learath2> for a 32 bit float you have a sign bit, 8 bits for exponent, that leaves 23 bits for the mantissa
14:02 <+bridge> [ddnet] <Learath2> the larger the exponent, the less of the mantissa will be after the decimal point, hence smaller the precision
14:05 <+bridge> [ddnet] <timakro> from the wikipedia article it seems you are right
14:05 <+bridge> [ddnet] <Learath2> Problem with calculating 0.001 precision is that the number is stored in base2 not 10
14:06 <+bridge> [ddnet] <timakro> i dont need to know exactly
14:06 <+bridge> [ddnet] <timakro> just if its possible to reach that point or not
14:06 <+bridge> [ddnet] <timakro> with 50hz
14:06 <+bridge> [ddnet] <timakro> like in a month or so idk
14:06 <+bridge> [ddnet] <timakro> or a year
14:07 <+bridge> [ddnet] <timakro> and where does this 7 decimal places come from you can find everywhere
14:07 <+bridge> [ddnet] <timakro> 8 bits exponent, 27 bits significant
14:07 <+bridge> [ddnet] <nuborn> Think it might be something like 0.001 x 2^23
14:11 <+bridge> [ddnet] <Learath2> 7 digits is easy
14:11 <+bridge> [ddnet] <timakro> The number of digits of precision a floating point variable has depends on both the size (floats have less precision than doubles) and the particular value being stored (some values have more precision than others). Float values have between 6 and 9 digits of precision, with most float values having at least 7 significant digits.
14:11 <+bridge> [ddnet] <Learath2> 10^7 ~ 2^23, you have 23 bits for the mantissa
14:12 <+bridge> [ddnet] <Learath2> to lose 10^-3 ~ 2^-10 of precision you need to lose about 10 bits
14:12 <+bridge> [ddnet] <Learath2> so for exponents < -10 you won't have 0.001 of precision
14:13 <+bridge> [ddnet] <timakro> exponents < -10 would be a small number?
14:14 <+bridge> [ddnet] <Learath2> that doesn't sound right, I mean you need 10 bits to be after the "binary point" so 23 - 10 = 13, so numbers smaller then 2^-13
14:14 <+bridge> [ddnet] <Learath2> pretty small number
14:15 <+bridge> [ddnet] <Learath2> Sorry, I haven't worked with float internals for like 2-3 years now
14:17 <+bridge> [ddnet] <Learath2> If you use 13 bits before the point, you'll have 10 left for after the point and that's still enough for 0.001 precision, 14 bits and you no longer have enough
14:18 <+bridge> [ddnet] <timakro> but that would be 2^14 = 16384, you are saying for numbers >16384 you dont have 0.001 precision?
14:19 <+bridge> [ddnet] <timakro> sounds way off
14:20 <+bridge> [ddnet] <Patiga> messed a bit around in https://www.h-schmidt.net/FloatConverter/IEEE754.html
14:20 <+bridge> [ddnet] <redix> @heinrich5991 adding a bool for the race thing might indeed be better
14:20 <+bridge> [ddnet] <redix> However the client would drop packets which are missing the new bool. The client needs some backward compatibility for the protocol first, e.g. using a default value if a field is missing
14:20 <+bridge> [ddnet] <Patiga> got 16383
14:21 <+bridge> [ddnet] <Patiga> exponent > 140 cant represent accurately enough
14:22 <+bridge> [ddnet] <Patiga> take exponent 140 and tick all the boxes in the mantissa
14:22 <+bridge> [ddnet] <Patiga> (for 16383)
14:23 <+bridge> [ddnet] <timakro> wow this actually seems true
14:24 <+bridge> [ddnet] <timakro> and i saved the player start time in ticks as a float on my race server 😩
14:25 <+bridge> [ddnet] <Learath2> yep any bigger then 2^13 and you lose the 0.001
14:25 <+bridge> [ddnet] <timakro> i dont wanna know when i loose 0.01
14:26 <+bridge> [ddnet] <Patiga> o.o 5:27 until you lose that precision
14:26 <+bridge> [ddnet] <Patiga> (0.001)
14:26 <+bridge> [ddnet] <timakro> yeah not race time, server run time
14:26 <+bridge> [ddnet] <Patiga> ouch
14:27 <+bridge> [ddnet] <Learath2> you lose 0.001 ~= 2^-6.6 at ~2^17 I think
14:27 <+bridge> [ddnet] <Learath2> 0.01*
14:28 <+bridge> [ddnet] <Learath2> that's quite a bit more
14:28 <+bridge> [ddnet] <timakro> thats 45 minutes xD
14:28 <+bridge> [ddnet] <Patiga> looks like 131071
14:29 <+bridge> [ddnet] <Learath2> use a double?
14:29 <+bridge> [ddnet] <Learath2> that gives a lot more
14:29 <+bridge> [ddnet] <timakro> maybe the error cancels out with short race times xD
14:29 <+bridge> [ddnet] <timakro> tbh i might just store ticks*10 in an integer
14:29 <+bridge> [ddnet] <redix> what about just using integer milliseconds?
14:30 <+bridge> [ddnet] <timakro> yep that
14:30 <+bridge> [ddnet] <timakro> ticks*20 i guess
14:30 <+bridge> [ddnet] <Patiga> why not just ticks?
14:31 <+bridge> [ddnet] <timakro> well the plan was to have more precision
14:31 <+bridge> [ddnet] <timakro> but i messed that up
14:31 <+bridge> [ddnet] <Patiga> ah
14:41 <+bridge> [ddnet] <timakro> @heinrich5991 @redix is right, when we just add a bool new clients won't be compatible with old servers. So tell me the secret to get client version :D can't find it in the protocol. my server code is beyond hacky anyways
14:42 <+bridge> [ddnet] <redix> i think right now there is no server which supports new map records anyway
14:43 <+bridge> [ddnet] <redix> for now just use the new personal record value
14:44 <+bridge> [ddnet] <timakro> why this client addition if you cant use it ...
14:44 <+bridge> [ddnet] <timakro> and my sever does
14:44 <+bridge> [ddnet] <timakro> i just tried to implement it
14:45 <+bridge> [ddnet] <Learath2> There are high precision timers in most OSs maybe just sample that instead of ticks
14:45 <+bridge> [ddnet] <redix> for the next version 😄 when the protocol gets backward compatibility, we could use a new bool, so the protocol might change before a new version
14:45 <+bridge> [ddnet] <timakro> oh please god stop hammering out new versions
14:46 <+bridge> [ddnet] <timakro> i just made a server that supports 0.6 and 0.7 clients
14:46 <+bridge> [ddnet] <redix> well tw release cycle is already slow
14:46 <+bridge> [ddnet] <Learath2> if only we can get some extensible protocol into teeworlds that'd be the last major update requiring this much work
14:46 <+bridge> [ddnet] <timakro> yea you can wait another 10 years :D
14:47 <+bridge> [ddnet] <redix> but in general... it would be better to rely on a stable version. i think there are already incompatibilities with the chat commands because people use the dev version =\
14:48 <+bridge> [ddnet] <Learath2> people as in fokkonaut 😛
14:48 <+bridge> [ddnet] <timakro> so tell me how to detect client version already xD
14:49 <+bridge> [ddnet] <timakro> is this black magic or am i just missing a field in the protocol?
14:49 <+bridge> [ddnet] <redix> well it was only an example 😄 demos were also broken in some clients
14:50 <+bridge> [ddnet] <redix> @timakro  `Server()->GetClientVersion(ID)`
14:50 <+bridge> [ddnet] <Learath2> @timakro the client sends it in NETMSG_INFO
14:50 <+bridge> [ddnet] <timakro> thx
14:51 <+bridge> [ddnet] <fokkonaut> The 0.7 client sends it there
14:51 <+bridge> [ddnet] <fokkonaut> what about 0.6?
14:51 <+bridge> [ddnet] <fokkonaut> i couldnt find it there
14:52 <+bridge> [ddnet] <Learath2> 0.6 vanilla clients don't send version
14:52 <+bridge> [ddnet] <onby> ddnet uses NETMSGTYPE_CL_ISDDNET
14:52 <+bridge> [ddnet] <fokkonaut> And ddnet clients only send it in cl_isddnet?
14:52 <+bridge> [ddnet] <fokkonaut> yea
14:52 <+bridge> [ddnet] <Learath2> that
14:54 <+bridge> [ddnet] <Learath2> I need a way to keep track of whether we are inputting text at a global scale
14:55 <+bridge> [ddnet] <Learath2> I'm having trouble imagining how
14:56 <+bridge> [ddnet] <Learath2> (rendering happens after input update and because of imgui it's information that doesn't exist before rendering)
14:56 <+bridge> [ddnet] <Learath2> @heinrich5991 this sounds like some CS problem you might have seen before 😛
15:04 <+bridge> [ddnet] <redix> @timakro the pull request you mentioned was not included in 0.7.4. The latest dev version still uses `0x0704` as client version, so right now the client version does not tell you whether a client supports it or not =\
15:10 <+bridge> [ddnet] <fokkonaut> @timakro you can detect my F-Client, which is afaik the only client that uses the new chat system
15:11 <+bridge> [ddnet] <fokkonaut> its version is `0"0f2030`
15:11 <+bridge> [ddnet] <fokkonaut> its version is `0x0f2030`
15:14 <+bridge> [ddnet] <fokkonaut> @timakro you can set the timer precisiom serverside
15:14 <+bridge> [ddnet] <fokkonaut> if you send GAMEFLAG_RACE
15:14 <+bridge> [ddnet] <fokkonaut> Gamecontroller has a snap object, GameDataRace
15:15 <+bridge> [ddnet] <fokkonaut> https://github.com/ddnet/ddnet7/blob/a6f5861d9ab396aae0df40435a2bf120ba9f9c8f/src/game/server/gamecontroller.cpp#L400
15:17 <+bridge> [ddnet] <fokkonaut> (if that was what you were looking for earlier)
15:41 <+bridge> [ddnet] <heinrich5991> @timakro why won't it work with old server new client?
15:41 <+bridge> [ddnet] <heinrich5991> okay, I'm going to write the PR I guess
15:42 <+bridge> [ddnet] <redix> client will drop the packets because of a missing field
15:42 <+bridge> [ddnet] <heinrich5991> okay, need a default value if the field isn't present
15:42 <+bridge> [ddnet] <heinrich5991> @timakro please don't do version detection, it's evil 😦
15:43 <+bridge> [ddnet] <timakro> @heinrich5991 old server will send 1 byte, new client will try to unpack 2 byte
15:43 <+bridge> [ddnet] <heinrich5991> yes, I'll do the PR with the forward and backward-compatible message
15:43 <+bridge> [ddnet] <timakro> nice :)
15:44 <+bridge> [ddnet] <redix> @timakro you can use -1 as score. sorting in the scoreboard works then 🙂
15:45 <+bridge> [ddnet] <timakro> what where
15:45 <+bridge> [ddnet] <redix> 0.7 can display score as time. right now you use 0 if someone did not finish yet
15:46 <+bridge> [ddnet] <timakro> ok, thx
15:46 <+bridge> [ddnet] <redix> if you use -1 they will be at the bottom and the client will not display a time at all
16:19 <+bridge> [ddnet] <heinrich5991> @timakro @redix  https://github.com/teeworlds/teeworlds/pull/2536
16:24 <+bridge> [ddnet] <redix> looks good
17:18 <+bridge> [ddnet] <Learath2> That's a nice fix, how about you help me come up with a nice fix for my problem aswell so we can have good IME in teeworlds for once? 😛
18:58 <+bridge> [ddnet] <fokkonaut> https://cdn.discordapp.com/attachments/293493549758939136/695678706408226906/unknown.png
18:59 <+bridge> [ddnet] <fokkonaut> the gun damage indicator is not perfect :D. this is a straight up shot, and its not centered correctly :p
19:00 <+bridge> [ddnet] <Soreu> It looks like even the space between stars isn't same
19:00 <+bridge> [ddnet] <fokkonaut> maybe that, or the screenshot was just not nice enough
19:00 <+bridge> [ddnet] <Learath2> I don't think they are even supposed to be lined up properly
19:00 <+bridge> [ddnet] <fokkonaut> you can see though, the right side doesnt touch the block, the left side is on it a bit
19:00 <+bridge> [ddnet] <Learath2> there is randomness to it
19:00 <+bridge> [ddnet] <fokkonaut> @Learath2 i think it was ment to be lined up
19:01 <+bridge> [ddnet] <onby> are u sure there is randomness?
19:01 <+bridge> [ddnet] <Learath2> well i don't remember, take a look at the code, someone would have to mess up really bad for this to be this bad
19:02 <+bridge> [ddnet] <fokkonaut> doesnt look like its random
19:02 <+bridge> [ddnet] <fokkonaut> https://cdn.discordapp.com/attachments/293493549758939136/695679531650121838/unknown.png
19:02 <+bridge> [ddnet] <fokkonaut> these are multiple shots now
19:02 <+bridge> [ddnet] <fokkonaut> (i just shot like 5 time in a row or so)
19:03 <+bridge> [ddnet] <fokkonaut> i guess its because the amount is 10, and thats even.
19:03 <+bridge> [ddnet] <fokkonaut> looks like adding an 11th star on the right would balance it out again
19:04 <+bridge> [ddnet] <fokkonaut> i remember, i had this issue for my spread gun, when it was even i needed to shift all projectile directions, so its center
19:04 <+bridge> [ddnet] <fokkonaut> centered again*
19:11 <+bridge> [ddnet] <Learath2> weird, the server sends the angle but the client seems to ignore it, how does the client decide what side the stars should be displayed on?
19:13 <+bridge> [ddnet] <onby> doesnt the server send the stars too
19:32 <+bridge> [ddnet] <louis> help posting in the ddnet forum doesnt work
19:32 <+bridge> [ddnet] <louis> gives http error 500
19:35 <+bridge> [ddnet] <Ryozuki> what r u doing? applying for mod?
20:13 <+bridge> [ddnet] <fokkonaut> @Learath2 the server sends 10 objects
20:13 <+bridge> [ddnet] <fokkonaut> 10 stars
20:13 <+bridge> [ddnet] <fokkonaut> in 0.7 its different, but for 0.6
20:13 <+bridge> [ddnet] <Learath2> oh one per location
20:13 <+bridge> [ddnet] <fokkonaut> in 0.7 the angle is ignored
20:14 <+bridge> [ddnet] <Learath2> wait 0.7 still sends all the stars one by one?
20:14 <+bridge> [ddnet] <fokkonaut> no
20:14 <+bridge> [ddnet] <fokkonaut> in 0.7 this is not possible anymore
20:14 <+bridge> [ddnet] <fokkonaut> thats why i only send 1 star for ddnet7
20:14 <+bridge> [ddnet] <Learath2> and how does the client know which way the stars go?
20:15 <+bridge> [ddnet] <fokkonaut> it calculates based on whether there is another one next to it i guess, because if i keep shooting, it will perform a circle
20:15 <+bridge> [ddnet] <fokkonaut> but thats not the problem right now :D
20:15 <+bridge> [ddnet] <fokkonaut> we are still here for the 0.6 thing
20:15 <+bridge> [ddnet] <onby> NETEVENTTYPE_DAMAGEIND
20:15 <+bridge> [ddnet] <onby> in 0.6 each star is added to the snapshot with this type
20:16 <+bridge> [ddnet] <fokkonaut> in 0.6 the server loops through the amount of stars we want, does its math and then sends the angle with the object to the client
20:53 <+bridge> [ddnet] <timakro> @heinrich5991 What do i need to send now after your fix?
20:55 <+bridge> [ddnet] <timakro> Hmm the new clients will accept both msgs with 2 booleans and with 1 bool
20:55 <+bridge> [ddnet] <timakro> right?
20:57 <+bridge> [ddnet] <timakro> If I send just the short one I can't get the map record message. If I send just the long one old clients don't get the message. If I send both new clients get two messages.
20:58 <+bridge> [ddnet] <Learath2> @timakro old clients will just ignore the extra bool
21:02 <+bridge> [ddnet] <timakro> old clients cant be changed
21:02 <+bridge> [ddnet] <timakro> oh they will right now
21:04 <+bridge> [ddnet] <Learath2> yep