01:08 <+bridge> <teero777> https://cdn.discordapp.com/attachments/293493549758939136/1495561758780428448/image-vulkan.png?ex=69e6b1ed&is=69e5606d&hm=ea13f837fea36a12b09205429b2daf38a4742f957e8e34e65769d198cdd0a18a&
01:51 <+bridge> <learath2> Surely he'll come back a reformed man and stop treating everyone like doormats
06:29 <+bridge> <manikoo> sounds like juppy
08:23 <+bridge> <Assa> where is the crying step?
08:27 <+bridge> <ryozuki> dont u also have to make ur own allocator
08:29 <+bridge> <ryozuki> https://www.ietf.org/archive/id/draft-thain-ipv8-00.html
08:29 <+bridge> <ryozuki> when ddnet supports ipv8?
08:52 <+bridge> <pilonpl> Everyone knows about ipv4 and ipv6
08:53 <+bridge> <pilonpl> What happened to ipv1 2 3 and 5?
09:13 <+bridge> <Assa> ipv5 was proposed but didn't get widespread adoption
09:27 <+bridge> <totar> I think it's possible a better protocol could overtake ipv6, I'm not sure this one is that much better tho
09:29 <+bridge> <Assa> as long as we have hardware depending on this, every stack change will be painful and slow. I believe ipv6 would not be established, if we didn't run out of address space and it's not even really "established" as it still lacks in lots of places
09:30 <+bridge> <Assa> as long as we have hardware depending on this, every stack change will be painful and slow. I believe ipv6 would not be established, if we didn't run out of address space and it's not even really "established" as it still lacks in lots of places (at least in germany, DUH)
09:30 <+bridge> <totar> the workarounds for ipv4 exhaustion became desirable features that ivp6 lacks, also you can't communicate between ipv4 and ipv6 which is the biggest issue
09:31 <+bridge> <totar> the workarounds for ipv4 exhaustion became desirable features that ipv6 lacks, also you can't communicate between ipv4 and ipv6 which is the biggest issue
09:32 <+bridge> <totar> any amount of partial transition capability would make it much easier
09:36 <+bridge> <ryozuki> did u look at the proposal of ipv8?
09:36 <+bridge> <ryozuki> i think i kinda like it
09:36 <+bridge> <ryozuki> its like ipv4 but with 4 more digits
09:36 <+bridge> <ryozuki> 1.1.1.1.n.n.n.n
09:36 <+bridge> <totar> it's good but I think 64 bits is not enough
09:37 <+bridge> <ryozuki> > IPv8 resolves address exhaustion as a consequence of its addressing architecture, not as a primary design goal. The 64-bit IPv8 address space provides 2^64 unique addresses. Each ASN holder receives 2^32 host addresses -- 4,294,967,296 addresses -- sufficient for any organisation at any scale without address exhaustion, CGNAT, or renumbering.
09:37 <+bridge> <ryozuki> > 
09:37 <+bridge> <ryozuki> > IPv4 is a proper subset of IPv8. An IPv8 address with r.r.r.r = 0.0.0.0 is an IPv4 address, processed by standard IPv4 rules. No existing device, application, or network requires modification to participate in an IPv8 network. The suite is 100% backward compatible. There is no flag day and no forced migration at any layer.
09:37 <+bridge> <ryozuki> > 
09:37 <+bridge> <ryozuki> > The global BGP8 routing table is structurally bounded at one entry per ASN. The /16 minimum injectable prefix rule prevents deaggregation. Most carriers advertise a single /8 summary route per regional ASN. The BGP4 routing table exceeded 900,000 prefixes with no architectural bound. The BGP8 routing table is bounded by ASN allocation rate -- approximately 175,000 entries today.
09:37 <+bridge> <ryozuki> the most important is backwards compatiblity
09:37 <+bridge> <ryozuki> i think it gives this a better opportunity at adoption
09:37 <+bridge> <totar> I disagree with 4 billion addresses being sufficient for any organization forever
09:38 <+bridge> <ryozuki> why
09:38 <+bridge> <ryozuki> 4 billion is half population
09:38 <+bridge> <ryozuki> and this is just for 1 org
09:39 <+bridge> <totar> ipv6 is "functionally infinite" which means you can do fancy stuff like give a single device many IPs for different purposes
09:40 <+bridge> <totar> if the whole world can exhaust 4 billion IPs then I see no reason a single org cant
09:40 <+bridge> <totar> it's not than many OOMs difference
09:43 <+bridge> <totar> also like, if you give them 4 billion ips, they will come up with ways to use them. but 4 billion is not a very big number for computers
09:43 <+bridge> <totar> so you can end up with NAT again
09:43 <+bridge> <jxsl13> the Mars will also need ips
09:46 <+bridge> <totar> if they defined in the protocol a way to extend to any number of bits then I think it would be a good enough
09:48 <+bridge> <manikoo> become the first ddnet player on mars
09:52 <+bridge> <reitw> u're too late, https://discord.com/channels/252358080522747904/293493549758939136/1494613184873762826 already asked they don't want
09:56 <+bridge> <reitw> ipv5 does exist but isn't widespread as ipv5
09:56 <+bridge> <reitw> even if yes, it was abandoned
09:56 <+bridge> <reitw> ipv5 does exist but isn't widespread-named as ipv5
09:56 <+bridge> <jxsl13> there is ipv5 twice 👀
09:57 <+bridge> <jxsl13> is that meant to be?
10:36 <+bridge> <bamcane_tee237> wtf
10:36 <+bridge> <bamcane_tee237> ipv8???????
10:36 <+bridge> <bamcane_tee237> omg
10:36 <+bridge> <bamcane_tee237> really nice address
10:37 <+bridge> <bamcane_tee237> Finally we could leave [::]
10:39 <+bridge> <chillerdragon> im a bit stuck debugging this boi
10:39 <+bridge> <chillerdragon> 
10:39 <+bridge> <chillerdragon> build/src/generated/protocol.h:4:10: fatal error: base/cxx.h: No such file or directory
10:39 <+bridge> <chillerdragon> https://github.com/ddnet/ddnet/blob/8611c684a4f912a4bff2443d54ca010c59880202/src/rust-bridge/base/cxx.h
10:39 <+bridge> <chillerdragon> https://github.com/ddnet/ddnet/blob/8611c684a4f912a4bff2443d54ca010c59880202/datasrc/compile.py#L51
10:40 <+bridge> <chillerdragon> how does `#include <base/cxx.h>` find `rust-bridge/base/cxx.h`
10:40 <+bridge> <chillerdragon> i don't find the magic line in CMakeLists.txt that adds the include directory
10:40 <+bridge> <chillerdragon> i guess my fork is missing it
10:43 <+bridge> <Assa> isn't it this? <https://github.com/ddnet/ddnet/blob/8611c684a4f912a4bff2443d54ca010c59880202/CMakeLists.txt#L910> dontate 10 to ddnet if I am wrong
10:43 <+bridge> <Assa> isn't it this? <https://github.com/ddnet/ddnet/blob/8611c684a4f912a4bff2443d54ca010c59880202/CMakeLists.txt#L910> donate 10 to ddnet if I am wrong
10:45 <+bridge> <Assa> rust bridge is added as a library afaict
10:45 <+bridge> <chillerdragon> but appending a target shouldnt mess with the include path should it?
10:46 <+bridge> <Assa> but adding a library should, doesn't it?
10:48 <+bridge> <chillerdragon> found it
10:48 <+bridge> <chillerdragon> https://github.com/ddnet/ddnet/blob/8611c684a4f912a4bff2443d54ca010c59880202/CMakeLists.txt#L3780-L3781
10:48 <+bridge> <chillerdragon> i am indeed missing that line in my fork oO
10:48 <+bridge> <chillerdragon> its a 4 year old line i think i lost it a while ago and never needed it until todays merge lel
10:48 <+bridge> <chillerdragon> ah never mind i looked in the wrong spot
10:48 <+bridge> <chillerdragon> i do have the line
10:48 <+bridge> <chillerdragon> oke ima investigate thanks for moral support assa
10:49 <+bridge> <chillerdragon> ye nvm i cba i will patch compile.py xd
10:49 <+bridge> <Assa> 😄 I was wrong, so you need to donate 10 now
10:50 <+bridge> <chillerdragon> -.-
11:10 <+bridge> <bamcane_tee237> huh
12:30 <+bridge> <reitw> what
12:31 <+bridge> <bamcane_tee237> Really weird to see that we release old versions
12:31 <+bridge> <reitw> probably meant to apply the security fix on older versions, if people wants to play with ti
12:31 <+bridge> <reitw> probably meant to apply the security fix on older versions, if people wants to play with it
12:54 <+bridge> <learath2> Backports of the security fix on the three last releases, yes. I don't get why people won't upgrade but at least they can be safe
13:01 <+bridge> <Assa> do teams have their own CCollision instance? I can't make sense of doors
13:05 <+bridge> <Assa> how can doors in one team be open and in another be closed huh 🤔
13:06 <+bridge> <pluskz> Switches are different for each time
13:07 <+bridge> <pluskz> Time? xd
13:07 <+bridge> <pluskz> I mean team
13:08 <+bridge> <Assa> Can you point me to the place where this is handled?
13:09 <+bridge> <_laveer> Barely any of ISPs in my country support ipv6, ipv8 will arive by year 3000 when we will have ipv67 :zzzz:
13:09 <+bridge> <chillerdragon> which refactor did i miss that caused my old ass image loader to segfault on some double free?
13:12 <+bridge> <Assa> probably mine
13:12 <+bridge> <chillerdragon> ye :D  looks like it
13:13 <+bridge> <chillerdragon> i just blamed you in
13:13 <+bridge> <chillerdragon> CImageInfo::\~CImageInfo()
13:13 <+bridge> <Assa> this one: <https://github.com/ddnet/ddnet/pull/12016>
13:13 <+bridge> <chillerdragon> thanks
13:13 <+bridge> <Assa> do you need support? I can help you with the rewrite
13:13 <+bridge> <chillerdragon> i managed to fix it by removing my call to `free()`
13:13 <+bridge> <chillerdragon> 🎉
13:13 <+bridge> <chillerdragon> looked scarier than it was :D
13:14 <+bridge> <chillerdragon> i added that free() call 7 years ago^^
13:14 <+bridge> <Assa> CImageInfo is basically a smartpointer now, ideally you'd also call "Allocate" or you run into a warning
13:14 <+bridge> <pluskz> Is not like m_vswitchers is centralised in one file but here is a function for example https://github.com/ddnet/ddnet/blob/8611c684a4f912a4bff2443d54ca010c59880202/src/game/gamecore.cpp#L732
13:14 <+bridge> <pluskz> 
13:14 <+bridge> <pluskz> There are also functions to manage ON/OFF in character.cpp
13:15 <+bridge> <chillerdragon> e wot
13:15 <+bridge> <chillerdragon> where do i allocate?
13:15 <+bridge> <chillerdragon> doesnt LoadPNG do that for me
13:15 <+bridge> <Assa> uh m_vSwitchers
13:15 <+bridge> <Assa> yes it does 🙂 it also frees the memory for you
13:15 <+bridge> <Assa> so your free was wrong in the first place xD
13:16 <+bridge> <chillerdragon> oke nice
13:16 <+bridge> <chillerdragon> thats why i didnt get a warning i guess im good now
13:16 <+bridge> <chillerdragon> hope no mem leak
13:16 <+bridge> <chillerdragon> yea i guess but i probably copy pasted it from some working code back then and it worked well for me for 7 years so it cant be that wrong xd
13:17 <+bridge> <Assa> it either does free the memory now and gives you a warning, or there is no memory to free in the first place, so you're good 🙂
13:17 <+bridge> <Assa> this is the smart part of the smart pointer 😄
13:18 <+bridge> <Assa> this is the smart part of the smart pointer 😄 because it tracks now which memory it allocates itself
13:28 <+bridge> <Assa> I can't make sense of this, doors are adding tiles in CCollision
13:29 <+bridge> <pluskz> Really? *Starts looking collision*
13:37 <+bridge> <pluskz> https://github.com/ddnet/ddnet/blob/8611c684a4f912a4bff2443d54ca010c59880202/src/game/gamecore.cpp#L192
13:41 <+bridge> <fokkonaut> @essigautomat if you need smth on doors, i can help
13:42 <+bridge> <Assa> I need to calm myself after reading that code, brb
13:43 <+bridge> <fokkonaut> its good code
13:43 <+bridge> <fokkonaut> (heinrichs implementation)
13:50 <+bridge> <Assa> I am impelenting the counter that we talked about
13:54 <+bridge> <Assa> but I think this doesn't make sense like this, not with these movement restrictions
13:58 <+bridge> <fokkonaut> Why?
14:00 <+bridge> <fokkonaut> https://github.com/fokkonaut/F-DDrace/blob/82ceae063a4714004b8caf9efd80ad2f4a7c4b4b/src/game/collision.cpp#L1807-L185
14:00 <+bridge> <fokkonaut> 
14:00 <+bridge> <fokkonaut> You can take a look if you need inspiration, this works pretty well
14:01 <+bridge> <fokkonaut> https://github.com/fokkonaut/F-DDrace/blob/82ceae063a4714004b8caf9efd80ad2f4a7c4b4b/src/game/collision.cpp#L1807-L185
14:01 <+bridge> <fokkonaut> https://github.com/fokkonaut/F-DDrace/blob/82ceae063a4714004b8caf9efd80ad2f4a7c4b4b/src/game/collision.cpp#L409-L432
14:01 <+bridge> <fokkonaut> 
14:01 <+bridge> <fokkonaut> You can take a look if you need inspiration, this works pretty well
14:01 <+bridge> <fokkonaut> https://github.com/fokkonaut/F-DDrace/blob/82ceae063a4714004b8caf9efd80ad2f4a7c4b4b/src/game/collision.cpp#L1807-L185
14:01 <+bridge> <fokkonaut> https://github.com/fokkonaut/F-DDrace/blob/82ceae063a4714004b8caf9efd80ad2f4a7c4b4b/src/game/collision.cpp#L409-L432
14:01 <+bridge> <fokkonaut> 
14:01 <+bridge> <fokkonaut> You can take a look if you need inspiration, this works pretty well
14:02 <+bridge> <fokkonaut> As I said previously, the index might not be required for DDNet, as only `TILE_STOPA` are set and it depends on how future proof you'd want it
14:04 <+bridge> <Assa> This would be the most future proof imo, but also the most expensive and I hate it:
14:04 <+bridge> <Assa> ```C++
14:04 <+bridge> <Assa> class CDoorTileMultiple : public CDoorTile
14:04 <+bridge> <Assa> {
14:04 <+bridge> <Assa> public:
14:04 <+bridge> <Assa>     std::vector<int> m_vNumbers;
14:04 <+bridge> <Assa> };
14:04 <+bridge> <Assa> ```
14:04 <+bridge> <Assa> I was thinking about using m_Flags as we are only using TILE_STOPA
14:05 <+bridge> <fokkonaut> I didn't notice performance issues on my server, but surely you could measure it
14:06 <+bridge> <fokkonaut> https://github.com/fokkonaut/F-DDrace/blob/82ceae063a4714004b8caf9efd80ad2f4a7c4b4b/src/game/collision.cpp#L1807-L1853
14:06 <+bridge> <fokkonaut> https://github.com/fokkonaut/F-DDrace/blob/82ceae063a4714004b8caf9efd80ad2f4a7c4b4b/src/game/collision.cpp#L409-L432
14:06 <+bridge> <fokkonaut> 
14:06 <+bridge> <fokkonaut> You can take a look if you need inspiration, this works pretty well
14:06 <+bridge> <fokkonaut> https://github.com/fokkonaut/F-DDrace/blob/82ceae063a4714004b8caf9efd80ad2f4a7c4b4b/src/game/collision.cpp#L1807-L1853
14:06 <+bridge> <fokkonaut> https://github.com/fokkonaut/F-DDrace/blob/82ceae063a4714004b8caf9efd80ad2f4a7c4b4b/src/game/collision.cpp#L409-L432
14:06 <+bridge> <fokkonaut> 
14:06 <+bridge> <fokkonaut> You can take a look if you need inspiration, this works pretty well
14:06 <+bridge> <fokkonaut> First link was broken
14:13 <+bridge> <jxsl13> grandpa
14:18 <+bridge> <Assa> you basically ended up with the same idea 👍
14:19 <+bridge> <Assa> I can't abuse m_Flags because it's not enough info if another door uses the tile if you don't know which door it is and it's state
14:21 <+bridge> <fokkonaut> I agree that collision code in general is messy, but besides some cleanups not much we can doo
14:21 <+bridge> <fokkonaut> I agree that collision code in general is messy, but besides some cleanups not much we can do
14:53 <+bridge> <Assa> I think I got it now, but old doors still need to be buggy
14:54 <+bridge> <Assa> this is part of the magic, and then you only need to iterate over m_vNumbers in the movement restrictions:
14:54 <+bridge> <Assa> ```
14:54 <+bridge> <Assa> void CCollision::SetDoorCollisionAt(float x, float y, int Type, int Number, bool OldDoor)
14:54 <+bridge> <Assa> {
14:54 <+bridge> <Assa>     if(!m_pDoor)
14:54 <+bridge> <Assa>         return;
14:54 <+bridge> <Assa>     int Nx = std::clamp(round_to_int(x) / 32, 0, m_Width - 1);
14:54 <+bridge> <Assa>     int Ny = std::clamp(round_to_int(y) / 32, 0, m_Height - 1);
14:54 <+bridge> <Assa> 
14:54 <+bridge> <Assa>     CDoorTileMultiple *pDoorTile = dynamic_cast<CDoorTileMultiple *>(&m_pDoor[Ny * m_Width + Nx]);
14:54 <+bridge> <Assa>     pDoorTile->m_Flags = 0;
14:54 <+bridge> <Assa>     pDoorTile->m_Number = Number;
14:54 <+bridge> <Assa>     if(!OldDoor)
14:54 <+bridge> <Assa>     {
14:54 <+bridge> <Assa>         if(Type != TILE_AIR)
14:54 <+bridge> <Assa>         {
14:54 <+bridge> <Assa>             pDoorTile->m_vNumbers.emplace(Number);
14:54 <+bridge> <Assa>             pDoorTile->m_Index = Type;
14:54 <+bridge> <Assa>         }
14:54 <+bridge> <Assa>         else
14:54 <+bridge> <Assa>         {
14:54 <+bridge> <Assa>             pDoorTile->m_vNumbers.erase(Number);
14:54 <+bridge> <Assa>             if(pDoorTile->m_vNumbers.empty())
14:54 <+bridge> <Assa>                 pDoorTile->m_Index = TILE_AIR;
14:54 <+bridge> <Assa>         }
14:54 <+bridge> <Assa>     }
14:54 <+bridge> <Assa>     else
14:54 <+bridge> <Assa>     {
14:54 <+bridge> <Assa>         if(Type != TILE_AIR)
14:54 <+bridge> <Assa>         {
14:55 <+bridge> <Assa>             pDoorTile->m_vNumbers = {Number};
14:55 <+bridge> <Assa> this is part of the magic, and then you only need to iterate over m_vNumbers in the movement restrictions:
14:55 <+bridge> <Assa> ```C++
14:55 <+bridge> <Assa> void CCollision::SetDoorCollisionAt(float x, float y, int Type, int Number, bool OldDoor)
14:55 <+bridge> <Assa> {
14:55 <+bridge> <Assa>     if(!m_pDoor)
14:55 <+bridge> <Assa>         return;
14:55 <+bridge> <Assa>     int Nx = std::clamp(round_to_int(x) / 32, 0, m_Width - 1);
14:55 <+bridge> <Assa>     int Ny = std::clamp(round_to_int(y) / 32, 0, m_Height - 1);
14:55 <+bridge> <Assa> 
14:55 <+bridge> <Assa>     CDoorTileMultiple *pDoorTile = dynamic_cast<CDoorTileMultiple *>(&m_pDoor[Ny * m_Width + Nx]);
14:55 <+bridge> <Assa>     pDoorTile->m_Flags = 0;
14:55 <+bridge> <Assa>     pDoorTile->m_Number = Number;
14:55 <+bridge> <Assa>     if(!OldDoor)
14:55 <+bridge> <Assa>     {
14:55 <+bridge> <Assa>         if(Type != TILE_AIR)
14:55 <+bridge> <Assa>         {
14:55 <+bridge> <Assa>             pDoorTile->m_vNumbers.emplace(Number);
14:55 <+bridge> <Assa>             pDoorTile->m_Index = Type;
14:55 <+bridge> <Assa>         }
14:55 <+bridge> <Assa>         else
14:55 <+bridge> <Assa>         {
14:55 <+bridge> <Assa>             pDoorTile->m_vNumbers.erase(Number);
14:55 <+bridge> <Assa>             if(pDoorTile->m_vNumbers.empty())
14:56 <+bridge> <Assa> This is of course a bit simpler, because it basically assumes TILE_STOPA
14:56 <+bridge> <Assa> I guess I should factor it fully out here
15:22 <+bridge> <fokkonaut> Does this work correctly for crossed doors of the same number? I'm assuming it can be a lot simpler than mine as doors can not be removed entirely in DDNet
16:47 <+bridge> <12944qwerty> msg.data
16:47 <+bridge> <12944qwerty> 
16:47 <+bridge> <12944qwerty> happens whenever I'm in the server and do any action. If I even move my mouse it will throw, but doesn't until then
16:47 <+bridge> <12944qwerty> https://cdn.discordapp.com/attachments/293493549758939136/1457221072272359495/supersmallhexdumplol.bin?ex=69e6f561&is=69e5a3e1&hm=aff3461064039b9bb3e5a2af7c575820ada47b878269ba355e6fa9134b1e72ed&
16:59 <+bridge> <Assa> it should work correctly, as m_vNumbers is a set
17:55 <+bridge> <fokkonaut> It does not track the usage tho
17:56 <+bridge> <fokkonaut> Is the m_Number still required here?
18:01 <+bridge> <cyberfrighter> this release candidate is broken af
18:04 <+bridge> <cyberfrighter> hammering or nading in 0.7 insta crashes your game and theres no dumps
18:04 <+bridge> <cyberfrighter> https://cdn.discordapp.com/attachments/293493549758939136/1495817359640760581/2026-04-20_17-52-31.mp4?ex=69e79ff9&is=69e64e79&hm=4a0e992969eed3c25dbd01bcf5345ca29c0d44a56f031d741eebf3b38bd38816&
18:06 <+bridge> <cyberfrighter> i blame rust
18:07 <+bridge> <melon> That profile pic is peak cyberfighter
18:20 <+bridge> <jxsl13> horses are to blame for your problems.
18:21 <+bridge> <alw5> kevin is tuf
18:21 <+bridge> <alw5> 🇷🇸
18:22 <+bridge> <melon> That background pic is peak cyberfighter
18:30 <+bridge> <fokkonaut> horse denier
18:35 <+bridge> <fokkonaut> std::iterator
18:35 <+bridge> <fokkonaut> https://cdn.discordapp.com/attachments/293493549758939136/1495825267183194255/image.png?ex=69e7a756&is=69e655d6&hm=8d393580ff92cf9f619c10b719099d5ebe4fb9f099bab2ee9bd1263431fa0122&
18:46 <+bridge> <Nixus> speedrun?
18:47 <+bridge> <fokkonaut> @essigautomat Do you have an opinion on a "projectile hammer"?
18:47 <+bridge> <fokkonaut> I implemented redirecting any projectile with hammer hit, it's really fun and can be combined with tele grenade or gun too. Works solo and coop (after someone else redirected the projectile it can hit it's owner too, until he hit it at least once again).
18:47 <+bridge> <fokkonaut> 
18:47 <+bridge> <fokkonaut> I can imagine some cool parts with this, but this feature would require a new map.
18:49 <+bridge> <fokkonaut> Especially with tunes it be nice, but works well without too
19:14 <+bridge> <Assa> No I don't have an opinion about this but sounds fun
19:21 <+bridge> <bamcane_tee237> It has been a issue
19:21 <+bridge> <cyberfrighter> didnt crash for me on non release candidate
19:21 <+bridge> <bamcane_tee237> See #12069
19:21 <+bridge> <DDNet> https://github.com/ddnet/ddnet/issues/12069
19:22 <+bridge> <bamcane_tee237> You can also get same issue on non-rc
19:22 <+bridge> <bamcane_tee237> Just join a MRPG 0.7 server
20:03 <+bridge> <bamcane_tee237> Thanks for all the contributors anyway.
20:03 <+bridge> <bamcane_tee237> uwu
20:26 <+bridge> <fokkonaut> It's almost as if chillerdragon implemented 0.7 client support and then disappeared
20:27 <+bridge> <fokkonaut> without fixing edge cases. Now it's unusable imo
20:33 <+bridge> <bamcane_tee237> *though we has fixed a lot of bugs.
20:33 <+bridge> <bamcane_tee237> *though we have fixed a lot of bugs.
20:57 <+bridge> <melon> @bamcane_tee237 #6961 was reopened as a draft because people in #developer  and in the town hall thread thought closing it means it's not considered. (I.e they questioned if we'll ever get QUIC). So probably not the best example :p
20:57 <+bridge> <DDNet> https://github.com/ddnet/ddnet/pull/6961
20:57 <+bridge> <bamcane_tee237> I think we should just move it to issue instead of draft
20:57 <+bridge> <bamcane_tee237> isn't it?
21:00 <+bridge> <melon> I wouldn't close a draft just because it's been for a long time. I would close inactive PRs that aren't drafted with inactive PR authors, (i.e what Ryozuki did which led to ddnet drama nr. 9271²) ._.
21:00 <+bridge> <bamcane_tee237> Some of those have been closed so I didn't mention them in that issue.
21:01 <+bridge> <bamcane_tee237> Also we should clean the issues
21:01 <+bridge> <bamcane_tee237> like what fokko is doing now.
21:03 <+bridge> <melon> :justatest:
21:04 <+bridge> <bamcane_tee237> oh
21:04 <+bridge> <bamcane_tee237> D:
21:04 <+bridge> <bamcane_tee237> :thonk:
21:07 <+bridge> <bamcane_tee237> oh no
21:09 <+bridge> <fokkonaut> @robyt3 Sorry for closing two relevant issues. I didn't get these issues anymore.
21:10 <+bridge> <bamcane_tee237> I didn't even know that we have so many issues needed to close directly D:
21:11 <+bridge> <fokkonaut> @robyt3 sv_max_clients can't be changed ingame in DDNet anymore
21:11 <+bridge> <robyt3> The issues are still there on the client side, but you probably fixed your servers to not crash clients anymore
21:12 <+bridge> <fokkonaut> Ah, I see
21:12 <+bridge> <robyt3> But `MaxClients` and `MAX_CLIENTS` are still mixed
21:13 <+bridge> <bannzay> big dots in freeze - small dots in freeze - no dots in freeze
21:13 <+bridge> <bannzay> Could not we just came to the 'no dots' back in the ~2016 when I been asking about it so hard ? Being a tester I have not used those dots a single time.
21:13 <+bridge> <bannzay> Whatever. Does this change override custom entities for players ? So we need to make backups.
21:13 <+bridge> <bannzay> Since they are not frequently updated some users might lose the data
21:13 <+bridge> <bannzay> Talking about https://github.com/ddnet/ddnet/commit/3767eebead65d482a5666df145f368ef070ec2fd
21:13 <+bridge> <bannzay> This is how game looks like if you use OpenGL 3.3 with disabled "Multiple texture units" on windows. Is this an expected behaviour? (v14.5)
21:13 <+bridge> <bannzay> https://cdn.discordapp.com/attachments/293493549758939136/748213493664055386/unknown.png?ex=69e7a21e&is=69e6509e&hm=aa0a39f2d594d8f8ff1dd4aa08a6c97423aaf1980458a65ba1d80dcbe9d21bf4&
21:13 <+bridge> <bannzay> not sure, should we create an issue. But there is a thing we should know to provide users with answers
21:13 <+bridge> <bannzay> https://cdn.discordapp.com/attachments/293493549758939136/750826919372521482/unknown.png?ex=69e740cf&is=69e5ef4f&hm=00bc35adcf43e787caba91aab9017c3571c712e31e19a21ff986c62a2aa2eb2a&
21:13 <+bridge> <bamcane_tee237> Could we close #6257 now
21:13 <+bridge> <DDNet> https://github.com/ddnet/ddnet/issues/6257
21:15 <+bridge> <robyt3> Not really, there's still not enough map validation
21:15 <+bridge> <bamcane_tee237> O:
21:16 <+bridge> <robyt3> #11572 would be a start, but I tested it with all ddnet-maps and this apparently changes how 17, mostly oldschool, are loaded. Haven't looked into the individual changes yet.
21:16 <+bridge> <DDNet> https://github.com/ddnet/ddnet/pull/11572
21:17 <+bridge> <fokkonaut> ?
21:18 <+bridge> <melon> it was simply funny seeing you closing an issue and Roby visually running after you to reopen them ^^
21:18 <+bridge> <melon> it was simply funny seeing you closing an issue and Roby running after you to reopen them ^^
21:19 <+bridge> <fokkonaut> oopsie
21:37 <+bridge> <Assa> I was already about to reopen the collision calculation one :justatest:   but fix changes physics
21:42 <+bridge> <bamcane_tee237> :frozen:
21:43 <+bridge> <bamcane_tee237> I received a email from Steam
21:44 <+bridge> <bamcane_tee237> I have thought that my Store Page Review Request was been resolved. But I found that's a CS2 market email.
21:45 <+bridge> <pluskz> Are they blocking it? D:
21:48 <+bridge> <bamcane_tee237> no
21:49 <+bridge> <bamcane_tee237> It's only 3 days since I sent the request
21:49 <+bridge> <pluskz> ah
21:53 <+bridge> <Assa> @blaiszephyr can I let you do #11767 ? :owo:
21:53 <+bridge> <DDNet> https://github.com/ddnet/ddnet/issues/11767
21:55 <+bridge> <Assa> @fokkonaut https://github.com/ddnet/ddnet/issues/12031#issuecomment-4269652419
21:55 <+bridge> <Assa> 
21:55 <+bridge> <Assa> Do you have something different in mind or do you like heinrichs proposal more?
21:56 <+bridge> <scarletsexo> bro
21:56 <+bridge> <scarletsexo> https://cdn.discordapp.com/attachments/293493549758939136/1495875778854912061/1.jpg?ex=69e7d661&is=69e684e1&hm=a69772dd0c9feb1b5f16c6f037a8efc67aef442ee6c7729e7582c54fe7eb77e4&
21:56 <+bridge> <scarletsexo> https://cdn.discordapp.com/attachments/293493549758939136/1495875787180736512/2.jpg?ex=69e7d663&is=69e684e3&hm=8a4e98777b6cc95e1727875ec4afb512081d4eaf3ce7b8ad17e987db06393577&
21:56 <+bridge> <scarletsexo> https://cdn.discordapp.com/attachments/293493549758939136/1495875787784720394/3.jpg?ex=69e7d663&is=69e684e3&hm=3c77f7c07313f17d330647b4f726f6b2dea531472260213593c3c58c4bcca33e&
21:56 <+bridge> <scarletsexo> https://cdn.discordapp.com/attachments/293493549758939136/1495875788690423808/4.jpg?ex=69e7d664&is=69e684e4&hm=2e912b1ea414efd1519a4e9be21224a4d61084881014e3a490458b7c83624453&
22:01 <+bridge> <fokkonaut> It's good but my concern is if something else will be changed based on that indicator, mods might want to not send it at all. Similar to how I do not use `PLAYERFLAG_BOT` for 0.7: https://github.com/teeworlds/teeworlds/pull/2093/changes#diff-597779b4eb51af9adfedd04b8a235afff01091a4ba741daa604cb5cbeda4e3daR1457
22:02 <+bridge> <fokkonaut> will comment
22:13 <+bridge> <Assa> so I checked all my issues and closed all that are completed or I deem not wanted, which ended up closing 1 issue and will open 1 extra PR :justatest: there is one other issue I am thinking about closing as not planned
22:15 <+bridge> <Assa> so I checked all my issues and closed all that are completed or I deem not wanted, which ended up closing 1 issue and will open 1 extra PR :justatest: there is one other issue I am thinking about closing as not planned which is <https://github.com/ddnet/ddnet/issues/10196> . Another candidate would be the ramdomize button - but for this one I'll need to force other maintainers to map 1 million random twinkling stars without loosing their mind.
22:27 <+bridge> <byfox> ```
22:27 <+bridge> <byfox> #[derive(Clone, Default)]
22:27 <+bridge> <byfox> pub struct RawSnap {
22:27 <+bridge> <byfox>     offsets: BTreeMap<i32, ops::Range<u32>>,
22:27 <+bridge> <byfox>     buf: Vec<i32>,
22:27 <+bridge> <byfox> }
22:27 <+bridge> <byfox> ```
22:27 <+bridge> <byfox> Why is it i32, it should be u32?...
22:29 <+bridge> <Assa> CachyOs is crashing on custom build :justatest: any ideas?
22:29 <+bridge> <Assa> https://cdn.discordapp.com/attachments/293493549758939136/1495884152954945678/Bildschirmfoto_20260420_222839.png?ex=69e7de2e&is=69e68cae&hm=418aff302f5997f2580ecb195bd1df3d198904472f89a13c9ba24107e4085a8e&
22:29 <+bridge> <Assa> steam version is working (but probably uses proton in the bg)
22:32 <+bridge> <Assa> I also get this message between the logs: `Authorization required, but no authorization protocol specified`
22:38 <+bridge> <Assa> `2026-04-20 22:33:45 I engine: operating system version: Linux 7.0.0-1-cachyos (x86_64, #1 SMP PREEMPT Fri, 17 Apr 2026 11:49:42 +0000); "CachyOS"`
22:38 <+bridge> <Assa> 
22:38 <+bridge> <Assa> Uhh :justatest:
22:40 <+bridge> <Assa> @robyt3 is this a me issue or do we get a problem in the future? I wonder if ubuntu26 will have the same issue
22:40 <+bridge> <robyt3> I'm on Windows. Haven't tested on Ubuntu recently
22:42 <+bridge> <Assa> this is what the steam client reports as OS funnily enough: `2026-04-20 22:41:12 I engine: operating system version: Linux 7.0.0-1-cachyos (x86_64, #1 SMP PREEMPT Fri, 17 Apr 2026 11:49:42 +0000); "Steam Runtime 3 (sniper)"`
22:42 <+bridge> <melon> what kind of custom build are we talking?
22:42 <+bridge> <Assa> checking out current master - building - running
22:42 <+bridge> <melon> let me check
22:42 <+bridge> <Assa> this might be a case of bad-nvidia-driver (as usual)
22:43 <+bridge> <melon> Debug or Release?
22:43 <+bridge> <Assa> release
22:48 <+bridge> <melon> works for me :p
22:48 <+bridge> <melon> tested both local build and the NixOS package
22:52 <+bridge> <Assa> are you on the latest kernel (as in my logs?)
22:53 <+bridge> <Assa> I am trying a clean build
22:55 <+bridge> <Assa> uhh <https://www.neowin.net/news/cachyos-newest-release-drops-x11-for-wayland-in-live-isos/>
22:56 <+bridge> <Assa> hmm, idk if this is related
23:10 <+bridge> <Assa> I restartet the terminal and it now works again oO
23:11 <+bridge> <Assa> even vulkan runs again, which didn't before for some reason
23:11 <+bridge> <Assa> idk, weird magic happening