This solves the problem of not being able to set the view distance too
high, especially in pathological cases like giant trees. For
simplicity, we just freeze any atlas where allocation failed and start
allocating to a new texture and atlas, letting reference counting
destroy the old one when there are no more references to it. Because of
the spatial locality of chunk allocations, chunks allocated together
will virtually always have similar lifetimes, so the odds of this
causing significant fragmentation are very low, meaning this simple
solution should not do much worse than a much fancier one.
instead of using tokio::Threadpool for CPU bound long running tasks, use rayon, but drip feed and limit the tasks
See merge request veloren/veloren!1912
- Use Ori::{new,to_quat} and make the field private.
- Update/capitalize/add various comments.
- Implicitly drop scope guards where applicable.
- Take !Copy colliders by reference instead of cloning.
- s/cylinder_voxel_collision/box_voxel_collision/
- Unindent some physics code with a continue.
remove outdated economic simulation code
remove old values, document
add natural resources to economy
Remove NaturalResources from Place (now in Economy)
find closest site to each chunk
implement natural resources (the distance scale is wrong)
cargo fmt
working distance calculation
this collection of natural resources seem to make sense, too much Wheat though
use natural resources and controlled area to replenish goods
increase the amount of chunks controlled by one guard to 50
add new professions and goods to the list
implement multiple products per worker
remove the old code and rename the new code to the previous name
correctly determine which goods guards will give you access to
correctly estimate the amount of natural resources controlled
adapt to new server API
instrument tooltips
Now I just need to figure out how to store a (reference to) a closure
closures for tooltip content generation
pass site/cave id to the client
Add economic information to the client structure
(not yet exchanged with the server)
Send SiteId to the client, prepare messages for economy request
Make client::sites a HashMap
Specialize the Crafter into Brewer,Bladesmith and Blacksmith
working server request for economic info from within tooltip
fully operational economic tooltips
I need to fix the id clash between caves and towns though
fix overlapping ids between caves and sites
display stock amount
correctly handle invalid (cave) ids in the request
some initial balancing, turn off most info logging
less intrusive way of implementing the dynamic tool tips in map
further tooltip cleanup
further cleanup, dynamic tooltip not fully working as intended
correctly working tooltip visibility logic
cleanup, display labor value
separate economy info request in a separate translation unit
display values as well
nicer display format for economy
add last_exports and coin to the new economy
do not allocate natural resources to Dungeons (making town so much larger)
balancing attempt
print town size statistics
cargo fmt (dead code)
resource tweaks, csv debugging
output a more interesting town (and then all sites)
fix the labor value logic (now we have meaningful prices)
load professions from ron (WIP)
use assets manager in economy
loading professions works
use professions from ron file
fix Labor debug logic
count chunks per type separately
(preparing for better resource control)
better structured resource data
traders, more professions (WIP)
fix exception when starting the simulation
fix replenish function
TODO:
- use area_ratio for resource usage (chunks should be added to stock, ratio on usage?)
- fix trading
documentation clean up
fix merge artifact
Revise trader mechanic
start Coin with a reasonable default
remove the outdated economy code
preserve documentation from removed old structure
output neighboring sites (preparation)
pass list of neighbors to economy
add trade structures
trading stub
Description of purpose by zesterer on Discord
remember prices (needed for planning)
avoid growing the order vector unboundedly
into_iter doesn't clear the Vec, so clear it manually
use drain to process Vecs, avoid clone
fix the test server
implement a test stub (I need to get it faster than 30 seconds to be more useful)
enable info output in test
debug missing and extra goods
use the same logging extension as water, activate feature
update dependencies
determine good prices, good exchange goods
a working set of revisions
a cozy world which economy tests in 2s
first order planning version
fun with package version
buy according to value/priority, not missing amount
introduce min price constant, fix order priority
in depth debugging
with a correct sign the trading plans start to make sense
move the trade planning to a separate function
rename new function
reorganize code into subroutines (much cleaner)
each trading step now has its own function
cut down the number of debugging output
introduce RoadSecurity and Transportation
transport capacity bookkeeping
only plan to pay with valuable goods, you can no longer stockpile unused options
(which interestingly shows a huge impact, to be investigated)
Coin is now listed as a payment (although not used)
proper transportation estimation (although 0)
remove more left overs uncovered by viewing the code in a merge request
use the old default values, handle non-pileable stocks directly before increasing it
(as economy is based on last year's products)
don't order the missing good multiple times
also it uses coin to buy things!
fix warnings and use the transportation from stock again
cargo fmt
prepare evaluation of trade
don't count transportation multiple times
fix merge artifact
operational trade planning
trade itself is still misleading
make clippy happy
clean up
correct labor ratio of merchants (no need to multiply with amount produced)
incomplete merchant labor_value computation
correct last commit
make economy of scale more explicit
make clippy happy (and code cleaner)
more merchant tweaks (more pop=better)
beginning of real trading code
revert the update of dependencies
remove stale comments/unused code
trading implementation complete (but untested)
something is still strange ...
fix sign in trading
another sign fix
some bugfixes and plenty of debugging code
another bug fixed, more to go
fix another invariant (rounding will lead to very small negative value)
introduce Terrain and Territory
fix merge mistakes
- remove overwritten logging setting in server-cli
- add server-cli command to load a random area for testing without a client
- make admin add/remove commands modify ingame players instead of needing to reconnect
- add spans to par_join jobs
- added test command that loads up an area of the world
- add tracy-world-server alias
- set debug directives to info for logging
only `localhost` are allowed in a release build.
when debug assertions are on, others are also allowed.
This change undoes the changes to the settings, so compared to master, there is no effect
The auth server no longer allows the protocol to be specified. we enforce `https` for the auth server, so DO NOT provide a auth url with `https://` but without.
correct is now `auth.veloren.net`
incorrect is: `https://auth.veloren.net`
lagging a bit behind on terrain. Which is fine. Block Places and Block Pickup are not handled in this stream, as they go through the standart route of event handling.
- get rid of old SysTimers for each system in favour of VSystem tracking
- move metrics generation from lib.rs to own system
- code cleanup
- remove time tracking in common::sys
Instead of keeping Runtime and manually spawn a task on `drop` this task is spawned at start and will wait to be triggered.
The `drop` methods then wait for completion, UNLESS they are in a async context, then they MUST NOT BLOCK (deadlock potential), so they defer it to the Runtime and HOPE for the runtime to exist long enough.
This get rid of the weird `block_in_place` which is only accessable with `rt-multi-threaded` and has some disadvantages.
We also wont requiere the runtime to be active all the time. Though its needed for a clean shutdown
- now last digit version is compatible 0.6.0 will connect to 0.6.1
- the TCP DATA Frames no longer contain START field, as it's not needed
- the TCP OPENSTREAM Frames will now contain the BANDWIDTH field
- MID is not Protocol internal
Update network
- update API with Bandwidth
Update veloren
- introduce better runtime and `async` things that are IO bound.
- Remove `uvth` and instead use `tokio::runtime::Runtime::spawn_blocking`
- remove futures_execute from client and server use tokio::runtime::Runtime instead
- give threads a Name
This stores the components as children of the item that contains them via the DB's `parent_container_item_id` feature, and ensures that things are loaded in a good order with breadth-first search.
Squahed fixes:
- Fix some constraint violations that occurred when swapping inventory items.
- Comment out recipes for modular weapons.
- Make update_item_at_slot_using_persistence_key and is_modular more idiomatic.
- Add changelog entry.
- Document `defer_foreign_keys` usage.
Removed weapons that were duplicated in weapons and npc_weapons from npc_weapons.
Added migration to convert npc_weapons that ended up in anyone's inventory to weapon version.
Consolidated debug boost and possess sticks into one debug_stick, and renamed the admin sword (with name cultist_purp_2h_boss-0) to admin_sword
- completly switch to Bytes, even in api. speed up TCP by fak 2
- improve benchmarks
- speed up mpsc metrics
- gracefully handle shutdown by interpreting Ok(0) as tokio::tcpstream closed now.
- fix hotloop in participants by adding `Some(n)` to fix endless handing.
- fix closing bug by closing streams after `recv_mgr` is shutdown even if now shutdown is triggered locally.
- fix prometheus
- no longer throw when a `Stream` is dropped while participant still receives a msg for it.
- fix the bandwith handling, TCP network send speed is up to 1.5GiB/s while recv is 150MiB/s
- add documentation
- tmp require rt-multi-threaded in client for tokio, to not fail cargo check
this is prob stable, i tested over 1 hour.
after that some optimisations in priomgr.
and impl. propper bandwith.
Speed is up to 2GB/s write and 150MB/s recv on a single core
sync add documentation
- better logging in network
- we now notify the send of what happened in recv in participant.
- works with veloren master servers
- works in singleplayer, using a actual mid.
- add `mpsc` in whole stack incl tests
- speed up internal read/write with `Bytes` crate
- use `prometheus-hyper` for metrics
- use a metrics cache
- Implementing a async non-io protocol crate
a) no tokio / no channels
b) I/O is based on abstraction Sink/Drain
c) different Protocols can have a different Drain Type
This allow MPSC to send its content without splitting up messages at all!
It allows UDP to have internal extra frames to care for security
It allows better abstraction for tests
Allows benchmarks on the mpsc variant
Custom Handshakes to allow sth like Quic protocol easily
- reduce the participant managers to 4: channel creations, send, recv and shutdown.
keeping the `mut data` in one manager removes the need for all RwLocks.
reducing complexity and parallel access problems
- more strategic participant shutdown. first send. then wait for remote side to notice recv stop, then remote side will stop send, then local side can stop recv.
- metrics are internally abstracted to fit protocol and network layer
- in this commit network/protocol tests work and network tests work someway, veloren compiles but does not work
- handshake compatible to async_std
switch to `tokio` and `async_channel` crate.
I wanted to do tokio first, but it doesnt feature Sender::close(), thus i included async_channel
Got rid of `futures` and only need `futures_core` and `futures_util`.
Tokio does not support `Stream` and `StreamExt` so for now i need to use `tokio-stream`, i think this will go in `std` in the future
Created `b2b_close_stream_opened_sender_r` as the shutdown procedure does not need a copy of a Sender, it just need to stop it.
Various adjustments, e.g. for `select!` which now requieres a `&mut` for oneshots.
Future things to do:
- Use some better signalling than oneshot<()> in some cases.
- Use a Watch for the Prio propergation (impl. it ofc)
- Use Bounded Channels in order to improve performance
- adjust tests coding
bring tests to work
* Remove tweak feature
* Remove const-tweaker
* Update tiny_http
* Update bitvec to 0.21.0
* Downgrade euc to avoid conflict with vek 0.12.0
* Require exactly vek 0.12.0
* Update all other dependencies automatically based on these changes
* Update gilrs to latest at the request of Ada Lovegirls
* Update meshing benchmarks to new criterion API
- Separate `invite` machinery from `group_manip` into it's own thing (includes renaming `group_invite` to `invite` where applicable).
- Move some invite/trade machinery to `ControlEvent`.
- Make `TradePhase` a proper enum instead of a bunch of bools.
- Make `TradeId` a proper newtype.
- Remove trades from `Trades` on accept (previously was only on decline).
- Typo fixes/misc cleanup.
- Add bullet point for trading to the changelog.
- Fix item swapping edge case
- Document more assumptions/edge cases
- fmt and clippy
- s/ServerGeneral::GroupInvite/ServerGeneral::Invite/
- Use `Client::current` in `Client::is_dead`
- Accept/decline buttons that submit the proper messages
- A phase2 screen that renders the (item, quantity) pairs as text
- More checks in the trade state machine server-side.
Eliminated extra stagger state
Responding to more comments
Move poise character state changes to character behavior system
Move poise out of JoinTuple/Data
Finish up comments (various fixes)
Adjust lvl up msg fade
Add female humanoid names WIP
Changed text formatting when skill is max level. Added message to show you have 0 skill points available.
Addressed a lot of comments.
various changes to UI
- fix skillbar offset
- remove CR indicators for group members
- add CR indicators to group member frames
- use unified CR indicator icon
Exp reward tweaks.
Fixed flamethrower range skill description.
Fixed error in converting FK to entity_id from character_id.
Giant match statements in persistence.
Tweaked roll.
Changed how skills and skill groups were assigned in the skillset builder so it's format more closely followed loadout builder.
fixed incorrect translation used on axe double strike regen skill
Changed migration so that skills and skill groups foreign key to entity_id instead of character_id
* Improved server error message for character load errors.
* Added server logging for item asset load errors during character load.
* Fixed character select error message dialog not supporting long messages.
added sandraptor glider
added sandraptor glider
stuff
working on other raptor gliders
adding file and placeholder model (copy of morpho) for monarch butterfly gliders
adding models for snow and wood raptor gliders
added a proper model for the monarch glider
adding crafting recipies for raptor gliders, and raptor feather item for the recipe. Made monarch and morpho rare drops in cave loot table
made raptors drop their feathers
fmt
- Modified energy and speed scaling of sword, axe, and hammer combo melees.
- Nerfed bow basic attack knockback. Buffed bow damage and energy regen.
- Corrected xp values for theropods. Removed unnecessary function for xp increase per level.
- Guards no longer flee at low health.
- Buffed sword spin.
- Nerfed axe spin and sword dash moderately.
- Dash now takes a little time to reach full speed. Sword M2 and hammer M1 damage values tweaked.
Before we had accessed velocities in a nested loop O(n²).
Now we copy it only once out of the ECS system and store it in a tmp Vec.
As we no longer need to hold a mut and imutable reference, we can iter of `&mut velocities` again in the outer loop.
Also improved many calculations called in the loop to make the check if 2 entities are to far apart as easy as possible
In order to keep the performance we made it Internal Mutability and use a `Mutex` per Stream, till `Stream.send` is no longer `&mut self`.
The old solution didn't rely on this, but needed multiple Components instead which zest didn't liked
- fix wording in error msg
- find better name for structs
- unify errors and cleanup code with `(|| {foo?; Some(())})()` pattern
- fix the negative PlayersOnline, it was caused by having a gracefull shutdown AND a timeout error. we now unregister the client when he issues TERMINATE
player buffs animation
more testing debuffs
sorting and display limit fix
overhead buffs
fix
WIP buff removal function
fmt
Update buffs.rs
Now with compiling: WIP group UI buffs
positioning
Update group.rs
Update group.rs
Small optimizations.
Fixed positioning of buffs in group panel. Broke buff tooltips in group panel.
buff frame visuals
added setting for displaying buffs at minimap
So first off all we had to update the toolchain, i think everything in september is okay, but we got with this current version.
Then we had to update several dependencies, which broke:
- need a specific fix of winit, i think we want to get rid of this with iced, hopefully, because its buggy as hell. update wayland client to 0.27
- use a updated version of glutin which has wayland-client 0.27 and no longer the broke version 0.23
- update conrod to use modern copypasta 0.7
- use `packed_simd_2` instead of `packed_simd` as the owner of the create abandoned the project.
- adjust all the coding to work with the newer glutin and winit version. that also includes fixing a macro in one of the dependencies that did some crazy conversion from 1 event type to another event type.
It was called `convert_event`
- make a `simd` feature which is default and introduce conditional compiling.
AS I HAVE NO IDEA OF SIMD AND THE CODE. AND I DIDN'T INTRODUCE THE ERROR IN THE FIRST PLACE, WE PANIC FOR NON SIMD CASE FOR NOW. BUT IT WORKS FOR TESTS.
- tarpaulin doesnt support no-default-features. so we have to `sed` them away. semms fair.
His reason to reqeust that is, that there might not be a perfect disctinction in the future.
Now we need to send ServerGeneral over streams and do additional checking at various places to verify that not the wrong variant is send.
- Instread we have a dedicated thread that will async wait for new participants to connect and then notify the main thread
- registry no longer sends a view distance with it.
- remove ClientMsg::Command again as it's unused
```
//This is a helper structure, containing all possible data send over
pub enum ClientMsg {
Initial(ClientType),
General(ClientGeneralMsg),
InGame(ClientInGameMsg),
NotInGame(ClientNotInGameMsg),
Register(ClientRegisterMsg),
Ping(PingMsg)
}
```
Rather than having a single Stream to handle ALL data, seperate into multiple streams:
- Ping Stream, for seperate PINGS
- Register Stream, only used till the client is registered, then no longer used!
- General Stream, used for msg that can occur always
- NotInGame Stream, used for everything NOT ingame, e.g. Character Screen
- InGame Stream, used for all GAME data, players, terrain, entities, etc...
This version does compile, and gets the client registered (with auth too) but doesnt get to the char screen yet.
This fixes also the ignoring messages problem we had, as we are not sending data to the register stream!
This fixes also the problem that the server had to sleep for the Stream Creation, as the Server is now creating the streams and client has to sleep.
3x - 5x depending on terrain. We can do a lot better but this is a good
start.
Also, added chunk group count to metrics. This correlates with memory
usage specifically by chunk voxel data in a much more direct way than
chonk or chunk count do, so this should provide extra useful information
(especially for our average overhead per chonk / chunk).
In the process, also try to address a few edge cases related to block
detection, such as adding back previously solid sprites and removing
filters that may be vestiges of earlier logic.
`stable-0.7.0 (<hash>-<datetime>)` for release versions.
And
`nightly-<date> (<hash>)` for nightly and master versions
Reason is, many players only give information that they are running `0.x.0` but are not giving us the information which day, or commit they are running. So we should make master builds less confusing.
