Resolve "Cmd/Sync-Rework"
Closes #66 (closed)
Merge request reports
Activity
added 20 commits
-
c367ec77...c5ea5767 - 17 commits from branch
develop - 020bdcdd - [#66 (closed)] Remove depth buffer from core and take image handle vector as input for renderMesh
- e1a19e9d - [#66 (closed)] Remove unnecessary width/height from mesh drawcall
- e7e255c0 - [#66 (closed)] Merge of doom
Toggle commit list-
c367ec77...c5ea5767 - 17 commits from branch
added 4 commits
- 9caef593 - [#66 (closed)] Bundle Mesh data in struct for rendering
- fbbe5021 - [#66 (closed)] Cleaned up descriptor set usage
- 3042e8c1 - [#66 (closed)] Put push constant data into struct
- 9651c69a - [#66 (closed)] Implement support for multiple meshes
Toggle commit list
Die Meshes sollten allerdings nicht Teil vom Framework sein, oder? Meshes selbst haben je nach User-Verständnis ja Positionen, Orientierung und Relationen zu anderen Meshes. Das würde ich lieber im Szenengraph (#56 (closed)) haben.
Es sollte auch glaube ich, nicht vorausgesetzt werden, dass Index-Buffer verwendet werden, da wir nicht einmal innerhalb von GLTF einen Index-Buffer voraussetzen können:
All attribute accessors for a given primitive must have the same count. When indices property is not defined, it indicates the number of vertices to render; when indices property is defined, it indicates the upper (exclusive) bound on the index values in the indices accessor. ( https://github.com/KhronosGroup/glTF/tree/master/specification/2.0#meshes )
added 1 commit
- 369d3dcf - [#66 (closed)] Added concept of command stream and started layout transition rework
added 1 commit
- bc3ee377 - [#66 (closed)] Fixed MSVC unnoticed compiler bugs
added 2 commits
- 34979078 - [#66 (closed)] Fix first mesh project main
- e272189f - Merge branch '66-cmd-sync-rework' of...
added 2 commits
- 27a09a4c - [#66 (closed)] Fixed position of sorting vertex attributes in application code
- c9a28a7a - Merge branch '66-cmd-sync-rework' of...
added 1 commit
- 4e217306 - [#66 (closed)] Moved resizing out of event because it is not synchronized yet (see #55 (closed))
added 1 commit
- f4b06de5 - [#66 (closed)] Production ready try&catch trick
added 1 commit
- 2aa1441f - [#66 (closed)] Fixed problem with 1D depth buffer
-
Also die Resize-Crashes sind hier beide gefixt. Das eine mal mit try&catch-Blöcken, da die entsprechenden Fehler/Exceptions eigentlich bloß Result-Codes sein sollten, die man ordentlich handhaben kann (allerdings geht das in C++ nur mit try&catch, da zusätzlich zu dem Result die Exception geworfen wird).
Wenn jemand eine bessere Lösung findet, nehmen wir die gerne. Ansonsten funktioniert es so erstmal stabil und wir fangen so nur den ErrorOutOfDateKHR ab, sonst nichts damit wir keine weitere Inkonsistenz durch zu grobes Abfangen erzeugen.
added 1 commit
- d85734c7 - [#66 (closed)] Added barrier/transition function for sampled images and implemented shadow mapping as a test
mentioned in commit 84c7c24c