Check again how Memory-Mapped-Files can be used for resource-saving point-exchange between data generation and visualization

[griestopf]

No description provided.

[griestopf]

Currently the points inside an octant are stored sparsely in a huge memory area which could hold all possible points inside an octant. This requires compacting the point list when it is needed by the visualization. Evaluate ideas how inside an octant the incoming points can be stored in an (unsorted) contiguous bulk-list of points which then can be passed directly (without active compacting) to the visualization. Approaches allowing this will need an additional index-array mapping the xyz-grid index to the position in the bulk list.

[ScottNorton]

Any plans to increase the amount of lights? Sorry this is the wrong place to ask. But I tried to add about 50 lights and got 8, or a freeze (and AMD responded with a driver hang) in deferred. Is that something that I could report as a bug? Trying to create a Blazor-based RPG using Voxels and I had to port to ThreeJS for now, Blazor can't handle that buffer traffic - needs it's own communication layer for WebGL

…

________________________________ From: Christoph Müller ***@***.***> Sent: Thursday, November 23, 2023 5:03 AM To: FUSEEProjectTeam/Fusee ***@***.***> Cc: Subscribed ***@***.***> Subject: Re: [FUSEEProjectTeam/Fusee] Check again how Memory-Mapped-Files can be used for resource-saving point-exchange between data generation and visualization (Issue #863) Currently the points inside an octand are stored sparsely in a huge memory area which could hold all possible points inside an octant. This requires compacting the point list when it is needed by the visualization. Evaluate ideas how inside an octant the incoming points can be stored in an (unsorted) contiguous bulk-list of points which then can be passed directly (without active compacting) to the visualization. Approaches allowing this will need an additional index-array mapping the xyz-grid index to the position in the bulk list. — Reply to this email directly, view it on GitHub<#863 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ACDKKYKEI2HF3PE2MMWR56LYF4NPJAVCNFSM6AAAAAA7XPXZ66VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMRUGEYDQMBWGM>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>