In D1848#76171, @elexis wrote:If we have more than one revision that is called "Fix <number>" then noone will know which one is which.
So always name them according to their content.
- Queries
- All Stories
- Search
- Advanced Search
- Transactions
- Transaction Logs
Feed All Stories
All Stories
All Stories
Apr 25 2019
Apr 25 2019
Stan retitled D1848: Consider Google Closure Compiler type expressions in JSdoc in Health.js (rP22196) from Fix rP22196 Coding Convention breach to Fix rP22196 Health.js Coding Convention breach.
wraitii added inline comments to D1840: RunSpeedMultiplier cleanup: improve variable names and reuse a duplicated function (rP22197).
Stan added a comment to D1848: Consider Google Closure Compiler type expressions in JSdoc in Health.js (rP22196).
wraitii added a comment to D1828: Health.js cleanup: add tests, add an "IsInjured" function, use GetHitpoints everywhere.
In D1828#76156, @elexis wrote:Seems unlikely that GetHitpoints returns something other than this.hitpoints, especially inside the same component.
I know the JIT can optimize, but I don't know if it can achieve this specific thing.
It's easy to change this.hitpoints if we ever need some logic change.
If we change it to use the function, then we should have a reason to (i.e. construct a realistic example where the logic would be different from return this.hitpoints).
Stan retitled D1848: Consider Google Closure Compiler type expressions in JSdoc in Health.js (rP22196) from Fix rP22196 to Fix rP22196 Coding Convention breach.
elexis added a comment to D1848: Consider Google Closure Compiler type expressions in JSdoc in Health.js (rP22196).
If we have more than one revision that is called "Fix <number>" then noone will know which one is which.
So always name them according to their content.
Vulcan added a comment to D1846: Adds a "properties"-property to resources and let mods be able to prevent resources from being bartered, traded and/or tributed..
Successful build - Chance fights ever on the side of the prudent.
Freagarach updated the diff for D1846: Adds a "properties"-property to resources and let mods be able to prevent resources from being bartered, traded and/or tributed..
Fixed tests.
In D665#76122, @elexis wrote:The expected player experience is that the ship moves to the closest point to the shoreline where the units to be picked up are.
The current behavior is that the ship moves to the correct island, but upon every garrisoning moves to a presumed better location again (which turns out to be a waste of time in many cases).
If I understood the patch correctly, the behavior of the patch is that the ship moves to the closest shoreline, which may be on an entirely different island, at least.So if I understand the patch correctly, it changes one defect for another.
But the patch might be adapted to match the expected user behavior.
One easy fix would be that the ship won't move if the units can be garrissoned with the ship not moving.
Or just never move the ship.
elexis added a comment to D1840: RunSpeedMultiplier cleanup: improve variable names and reuse a duplicated function (rP22197).
Will update with your comments, except for the "runspeedmultiplier" thingy.
Thanks for taking the time.
It's important that we are precise about the words that we use instead of hoping that reader already knows what is meant (because not every reader does, sometimes noone knows. And having precise definitions means we can test our claims against the definition of the words).
elexis added a comment to D1828: Health.js cleanup: add tests, add an "IsInjured" function, use GetHitpoints everywhere.
Seems unlikely that GetHitpoints returns something other than this.hitpoints, especially inside the same component.
I know the JIT can optimize, but I don't know if it can achieve this specific thing.
It's easy to change this.hitpoints if we ever need some logic change.
If we change it to use the function, then we should have a reason to (i.e. construct a realistic example where the logic would be different from return this.hitpoints).
elexis awarded D1847: Optimise out of frustum rendering of textures overlays a Like token.
Vulcan added a comment to D1846: Adds a "properties"-property to resources and let mods be able to prevent resources from being bartered, traded and/or tributed..
Build failure - The Moirai have given mortals hearts that can endure.
Freagarach updated the diff for D1846: Adds a "properties"-property to resources and let mods be able to prevent resources from being bartered, traded and/or tributed..
Some fixes and Petra-AI should now follow the constraints, which is difficult to test fully.
Vulcan added a comment to D1848: Consider Google Closure Compiler type expressions in JSdoc in Health.js (rP22196).
Successful build - Chance fights ever on the side of the prudent.
Stan updated the diff for D1848: Consider Google Closure Compiler type expressions in JSdoc in Health.js (rP22196).
Add missing semicolon to fix Vulkan's warning
elexis retitled D1816: Use all three color channels when loading heightmaps following rP21113 from Remove wrong assignment introduced in rP21113 to Use all three color channels when loading heightmaps following rP21113.
The new behavior is that lostwall turrets are stored in a local entity and can be restored by the wallplacement algorithm?
Vulcan added a comment to D1848: Consider Google Closure Compiler type expressions in JSdoc in Health.js (rP22196).
Successful build - Chance fights ever on the side of the prudent.
Stan updated the summary of D1848: Consider Google Closure Compiler type expressions in JSdoc in Health.js (rP22196).
elexis added a comment to rP22182: Add ConquestCritical to structures that want it, instead of removing it from….
structures that want it
(Tiny grammar improvement: Neither structures nor code wants to do something, not even the authors want to do something, but actions are performed to achieve purposes. )
(-> Add ConquestCritical to structures that should have it, instead of removing it from those that should not).
Stan updated the summary of D1848: Consider Google Closure Compiler type expressions in JSdoc in Health.js (rP22196).
elexis added inline comments to rP22188: Move tooltips on gamesetup dropdowns from xml to js, since by current gamestup….
Stan added inline comments to rP22196: Correctly handle receiving 0 damage as not receiving any damage..
elexis updated subscribers of D1739: Introduce a replacement container for EntityMap, std::map, std::unordered_map and boost::unordered_map.
@fabio would be happy to add boost references if it really improves performance?
(So the question then would be does it, and is it sufficent performance benefit or just an experiment)
Stan added inline comments to rP22196: Correctly handle receiving 0 damage as not receiving any damage..
The expected player experience is that the ship moves to the closest point to the shoreline where the units to be picked up are.
The current behavior is that the ship moves to the correct island, but upon every garrisoning moves to a presumed better location again (which turns out to be a waste of time in many cases).
If I understood the patch correctly, the behavior of the patch is that the ship moves to the closest shoreline, which may be on an entirely different island, at least.
wraitii added a comment to D1828: Health.js cleanup: add tests, add an "IsInjured" function, use GetHitpoints everywhere.
In D1828#76107, @elexis wrote:But why replace this.hitpoints with this.GetHitpoints()? Only adds performance cost without adding benefits, doesn't it?
wraitii added a comment to D1840: RunSpeedMultiplier cleanup: improve variable names and reuse a duplicated function (rP22197).
Will update with your comments, except for the "runspeedmultiplier" thingy.
elexis raised a concern with rP22196: Correctly handle receiving 0 damage as not receiving any damage..
Taking the red ink because either the coding conventions should be changed or this new Google syntax style and the practices of not abiding by conventions should change.
Stan added inline comments to D1828: Health.js cleanup: add tests, add an "IsInjured" function, use GetHitpoints everywhere.
(Could have said that you commit this unless I commit it.)
elexis added a comment to D1828: Health.js cleanup: add tests, add an "IsInjured" function, use GetHitpoints everywhere.
Test looks goodish from a superficial look (assuming that the tested values are the intended values)
elexis retitled D1840: RunSpeedMultiplier cleanup: improve variable names and reuse a duplicated function (rP22197) from rP22197 cleanup: improve variable names and reuse a duplicated function. to RunSpeedMultiplier cleanup: improve variable names and reuse a duplicated function (rP22197).
elexis added inline comments to D1840: RunSpeedMultiplier cleanup: improve variable names and reuse a duplicated function (rP22197).
In D1847#76097, @wraitii wrote:I think perf impact is negligible (it's definitely better, but probably not by much, and the main impact is likely a lower number of draw calls).
Yeah, the same thoughts. It shouldn't increase and might decrease rendering time. So it's worth to not render invisible objects.
elexis retitled D1817: Remove unused variable and fail to register component if it has no constructor [ComponentManager] from Remove not needed variable and conversion to object to Remove unused variable and conversion in the ComponentManager.
I think perf impact is negligible (it's definitely better, but probably not by much, and the main impact is likely a lower number of draw calls).
Probably a good idea to do it anyways though.
In D1772#75870, @wraitii wrote:It's not that I'm not concerned. It's that I've weighted what I estimate to be the chance of 'it fixes the issue with 0 additional work' vs the chance of 'it actually makes things worse' and I honestly believe the calculation comes out to "overall it should be committed now".
In D1847#76051, @Stan wrote:What's the perf gain like ?
vladislavbelov added inline comments to rP22225: Fix D1491 which introduced an ENSURE that should not have been there..
In D1584#76088, @elexis wrote:std::mt19937 looked good to me, as Vladislav said since it's used in the other places too.
Actually the simulation and AI seed use boost::rand48.
But still better than using custom magic numbers?
The only possible performance problem was reinitializing the seed often right?
Wouldn't it be sufficient initialize the seed upon component construction once? Or upon first soundplay?
(Though the magic numbers aren't so bad, only one line, or four including the wrapping.)
(Especially if it's some orders of magnitude faster. But perhaps there are other std or boost RNG that are also orders of magnitude faster.)
std::mt19937 looked good to me, as Vladislav said since it's used in the other places too.
Actually the simulation and AI seed use boost::rand48.
But still better than using custom magic numbers?
The only possible performance problem was reinitializing the seed often right?
Wouldn't it be sufficient initialize the seed upon component construction once? Or upon first soundplay?
(Though the magic numbers aren't so bad, only one line, or four including the wrapping.)
(Especially if it's some orders of magnitude faster. But perhaps there are other std or boost RNG that are also orders of magnitude faster.)
wraitii added inline comments to D1754: Pass an argument to GUI events / SendEventToAll, remove loading screen progess workaround.
elexis added a comment to D1754: Pass an argument to GUI events / SendEventToAll, remove loading screen progess workaround.
Thanks for the review!
Seems like that won't work cause of template_unit_ship_fishing.xml
lyv added a comment to D1846: Adds a "properties"-property to resources and let mods be able to prevent resources from being bartered, traded and/or tributed..
(A certain other project also has this feature implemented. Mostly irrelevant, but if this is not committed before the first release, it might be of some interest)
elexis retitled D1841: efficient -> efficiently in rP22203 from correct rP22203 to efficient -> efficiently in rP22203.
elexis added a comment to D1846: Adds a "properties"-property to resources and let mods be able to prevent resources from being bartered, traded and/or tributed..
Might want to link the other code for comparison.
Freagarach added a comment to D1846: Adds a "properties"-property to resources and let mods be able to prevent resources from being bartered, traded and/or tributed..
I'm already checking out the code there, it is quite similar but has a few things I didn't think of.
wraitii added a comment to rP22225: Fix D1491 which introduced an ENSURE that should not have been there..
It is indeed... We don't use the jpc at the moment so this doesn't crash, but it should be fixed.
Stan updated subscribers of D1846: Adds a "properties"-property to resources and let mods be able to prevent resources from being bartered, traded and/or tributed..
@Angen I know you have little time but do you have some suggestions ? :)
Itms raised a concern with rP22225: Fix D1491 which introduced an ENSURE that should not have been there..
This is removing state.jpc = m_JumpPointCache[passClass].get(); when it != m_JumpPointCache.end(), no?
Yup, that ENSURE was broken... Mind you my initial code was _also_ broken since at() would have failed here too. Apparently I didn't playtest that revision and we have no tests calling ComputeJPSPath, which we really ought to (might add some this WE).
Fix D1491 which introduced an ENSURE that should not have been there.
Not sure if related, but I got this in the game uploading loading a map:
What's the perf gain like ?
[Windows] Automated build.
wackyserious committed rP22223: New Texture: Celtic Druid.
New Texture: Celtic Druid
wackyserious committed rP22222: Retexture: Illyrian helmet (Bronze and Gold variants).
Retexture: Illyrian helmet (Bronze and Gold variants)
wackyserious committed rP22221: New Texture: Elite Greek Hoplite (Bronze cuirass, Scale and Lamellar).
New Texture: Elite Greek Hoplite (Bronze cuirass, Scale and Lamellar)
Successful build - Chance fights ever on the side of the prudent.
Successful build - Chance fights ever on the side of the prudent.
Successful build - Chance fights ever on the side of the prudent.
Make FastRand return an u32
Successful build - Chance fights ever on the side of the prudent.
Do changes according to inlines
In D1461#73338, @elexis wrote:Good.
Now the question remains how we make it visible to the players that this click-feature exists.
Usually the mouse cursor changes when one hovers a link. Underlining won't be supported by our small font base, but adding a sprite should be easily possible (resource icons in tooltips.js / setup_resources.xml as an example).
Apr 24 2019
Apr 24 2019
Itms added inline comments to D1754: Pass an argument to GUI events / SendEventToAll, remove loading screen progess workaround.
Stan added inline comments to D1837: Input.js refactoring - separate diplomatic tribute in their own file..
Successful build - Chance fights ever on the side of the prudent.
Didn't look closely at it, but we definitely shouldn't copy the general fsm definition, but instead put it somewhere where UnitAI and input.js can use it.
Use 0xFFF instead 0x7FF
replace unsigned int by u32
replace uint8_t by u8
Imarok added a comment to D1837: Input.js refactoring - separate diplomatic tribute in their own file..
In D1837#75515, @wraitii wrote:not an expert then again.
Itms added a comment to D1754: Pass an argument to GUI events / SendEventToAll, remove loading screen progess workaround.
This looks very good to me, and it works both on SM38 and SM45.
lyv added a comment to D1846: Adds a "properties"-property to resources and let mods be able to prevent resources from being bartered, traded and/or tributed..
Angen also have a patch doing this. See the mythology mod.
bb requested verification of rP22203: Communicate field diminishing returns to the player in the fields tooltip.
Correct the grammer of rP22203
Wildfire Games · Phabricator