The core OpenGL alternative to вЂњbindlessвЂќ is вЂњbinding lessвЂќ by using large buffers and array textures and manually managing the sub-allocations. Due to state inheritance and the ability to stitch sequences from arbitrary buffer addresses, it can actually be faster than VulkanвЂ™s secondary command buffers.Д«indless Rendering the NV_command_list builds on top of previous extensions that improved the rendering hot-loop performance by using native GPU addresses, and therefore avoiding per-object handle lookups and validation. Due to its binary nature it is the fastest way to record commands (can also be used threaded) and provides two alternate render modes, either as pre-compiled object, or interpreted data from buffer objects. The draw-indirect like buffers encode tokens for typical binding state (UBO, VBO, EBO) and draw-calls as well as minor state changes such as front-face or viewport dimensions. We encourage you to have a look at the GTC presentation as well as samples on github. Just like in Vulkan, it is possible to re-use those command-buffers as well. NV_command_list is the latest extension and comes closest to VulkanвЂ™s command buffers and provides very fast ways to record and submit commands that are most common in the rendering hot-loop. NVIDIA provides various extensions to make the hot-loop faster The buffer content does not have to be filled every frame, but can be re-used and even manipulated by the GPU directly, for example for level of detail or culling purposes. The MDI buffer can also be filled from different threads using persistent mapped memory, whose pointers are passed to threads that donвЂ™t need an OpenGL context.

To make most use of this feature use the previous encouraged larger buffers for geometry data, as it will allow drawing many geometries at once via MDI. Compared to instancing we can draw any sub-range of vertex/index-buffers. ARB_multi_draw_indirect (MDI), OpenGL 4.3 core, allows accelerating draw-call submission.However, there is still some other ways to speed up rendering in OpenGL. In theory OpenGLвЂ™s display lists allow re-use, in practice they are only beneficial for a limited functionality. It is not as flexible as the Vulkan texture memory management, but it can reduce run-time texture creation costs.Д®ven in a single-threaded scenario VulkanвЂ™s command buffers provide efficient ways to render data: recording a command-buffer is faster than traditional OpenGL commands and command buffers can be re-used. Similar to using larger buffers and sub-allocating from those, a developer could use texture arrays for popular dimensions and re-use individual layers. The developer can also use it as view on a sub-resource, for example individual texture layer or mipmap in an array texture. ARB_texture_view (core 4.3) introduced the concept of casting texture formats to different types as long as they have the same texel size.

glTextureStorage is the better alternative for the driver. The classic glTexImage never вЂњknewвЂќ how many mip-maps would actually be specified, often causing lazy allocation.

ARB_texture_storage (core 4.2) provides immutable textures whose definition is complete up-front.

We also encourage the use of few but rather large buffers as suggested in the memory management blog post.

This would allow an identical workflow to Vulkan. Ideally the developer uses persistent mapped buffers for all buffers that he expects to for reading or writing access on the CPU. ARB_buffer_storage (core 4.4) should be preferred over the classic glBufferData, as it gives better usage hints than glBufferData.Some of these functions may also be available under EXT_direct_state_access, the primary difference between ARB and EXT is that ARB requires the use of glCreateвЂќResourceвЂќ rather than working from glGenвЂќResourceвЂќ object handles. In this case the вЂњindirectionвЂќ by passing the object handle, may cause additional work.

вЂњBind to EditвЂќ can still be beneficial in the hot loop, for example when the same buffer is updated very often by glBufferSubData, or glUniforms for the active program are changed. glBindBuffer would most likely only be used with the ELEMENT_ARRAY, DRAW_INDIRECT in the rendering loop.glBindBuffersBase / glBindBuffersRange for UNIFORM, SHADER_STORAGE, ATOMIC_COUNTER, TRANSFORM_FEEDBACK.Within the rendering hot loop the new set of binding calls is encouraged, as they only affect rendering state: Using DSA makes object manipulation cleaner as it doesnвЂ™t affect the binding state used for rendering, therefore is also middleware friendly. ARB_direct_state_access (DSA), OpenGL 4.5 core, allows direct manipulation of objects (textures, buffers, fbosвЂ¦) rather than the classic вЂњbind to editвЂќ.Vulkan follows an object-oriented design and we always manipulate objects directly with their handles.

Opengl 4.4 texture units