Questions tagged [gpu]

Questions related to the massively parallel computing units targeting effective computer graphics in modern machines.

Filter by
Sorted by
Tagged with
3 votes
2 answers
72 views

What is the difference between a "meshlet" and just a small mesh?

Obviously, conceptually, a meshlet is just a small mesh. But what is the reason for making the distinction then? I'm assuming it has to do with how they are implemented. In a meshlet architecture, ...
1 vote
0 answers
28 views

What motherboard spec should i look for using 4 displays

I am not sure if i am on the right forum. Let me know where to go if i'm wrong. I have an ASUS mATX motherboard H110M2 and PNY Quardro P620. Initially I put it into a mini tower which only had a 450W ...
0 votes
1 answer
79 views

Am I risking damaging my GPU?

I have written a relatively simple interactive C program using OpenGL 4.4 with a vertex shader and a fragment shader, running in the CodeBlocks IDE on Windows. I included an FPS counter basically as ...
  • 173
10 votes
6 answers
6k views

Why can't we have hardware-specific graphics APIs?

Right now, if you want to use a GPU, you have to do it through graphics APIs like DirectX or Vulkan. These APIs are meant to be hardware-independent, but more often than not, they are tied to a ...
  • 209
0 votes
0 answers
30 views

Why use buffer streaming instead of collecting data on CPU and uploading to GPU all at once

Say I have a couple of meshes that I want to reupload to GPU for each frame. I can do this by generating a single array from all the meshes on CPU and then uploading to GPU (ex with ...
0 votes
0 answers
18 views

Linux GPU selection while keeping display

I am trying to use AMD GPU, through a server Ubuntu system. Every time I execute the command export DISPLAY=:0 to switch from the integrated GPU to the AMD one, but at the same time, my display also ...
2 votes
1 answer
127 views

Dynamic array approach for GPU buffers

On CPU dynamic arrays are often used to store data of dynamic size, and often implemented by doubling the size when the array is full and copying the data over. The data structure has amortized O(1) ...
0 votes
0 answers
32 views

How to compute a GPU's theoretical GFLOPs based on its specs?

I was looking at the specs of this GPU in particular, and tried to compute its theoretical GFLOPs based on more basic specs, such as clock frequency and number of CUDA cores. My first thought was: ...
  • 111
1 vote
2 answers
789 views

What does memory bandwidth of a GPU mean exactly?

This question is motivated by looking at two numbers for different GPU's: (1) the memory size and (2) the memory bandwidth. My understanding is that memory bandwidth means, the amount of data that can ...
  • 111
0 votes
0 answers
35 views

Cg - fixed data type

I noticed that Cg has data type fixed, which is: a 12bit fixed point number Does any hardware even support that? AFAIK most GPUs don't even support 16bit floats ...
  • 596
0 votes
0 answers
96 views

Inter-Process communication in compute shaders

How do I write a code to facilitate common shared memory for communication between multiple shaders as well as processes within them in python?
1 vote
1 answer
223 views

Performance impact/implications of constant buffers / uniform buffers

Constant buffers stay constant throughout entire dispatches / draw calls, and thus their data can presumably be shared as efficiently as possible between different threads using those buffers. But I ...
  • 115
0 votes
1 answer
738 views

Compute shader workgroups execution and size [closed]

I want to clarify how are workgroups executed on various GPUs, thus I have several questions. I know that different GPU architectures work differently, but I expect an answer YES if there is single (...
0 votes
1 answer
188 views

2d gpu powered rendering engine

First of all I am not sure if this is the correct place to ask my question. If it is not please be so kind to point me in a direction where my question is relevant. A while ago I came across this ...
1 vote
1 answer
634 views

Is drawing Bezier curve with scanline algorithm possible with Vulkan?

As an information, Vulkan is a low-level abstraction layer thru modern GPU. It allows programmers to program the GPU and assemble rendering pipeline. On that pipeline, there are modules of program ...
1 vote
1 answer
108 views

Are intersections still hardware accelerated if we define a custom intersection program in Optix?

I've stumbled into Optix samples that define their own intersection programs to define how primitives intersect (or don't) with a ray, and others (mostly those only dealing with triangle meshes) that ...
  • 75
0 votes
1 answer
91 views

Is it possible to get recording device from meta data?

Im not sure if this is the right section to ask this question but i figured its worth a shot. Is it possible to take metadata from a video and use that data to identify what phone model was used to ...
  • 1
4 votes
1 answer
127 views

Why does UV tiling affect performance in volumetric ray marching?

I am building a volumetric ray marching shader in HLSL in Unreal Engine, based on Ryan Brucks’ work: https://shaderbits.com/blog/creating-volumetric-ray-marcher I am trying to add some additional ...
2 votes
2 answers
644 views

What data is passed from the CPU to the GPU each frame?

I've been reading a bit about the graphics pipeline for processing on a graphics card, and I'm interested to know in what kind of format data is passed from the CPU to the GPU when rendering 3D ...
  • 123
5 votes
1 answer
2k views

What are 'mesh clusters' / hierarchical cluster culling (with LOD?) / triangle cluster culling and how do they relate?

I've seen several references to this and related concepts now but I can't find any hard details about these systems, and how they are implemented. The earliest reference I can find is from the "...
  • 115
5 votes
1 answer
1k views

Binding vs bindless

I was trying to understand what exactly is 'binding'. I found an answer on quora on this. From this, binding basically sounds like sending data from to GPU with help from CPU. I got to know more about ...
0 votes
1 answer
62 views

How does the OS + hardware support multiple screen resolutions and multi-display?

Modern operating systems support a wide range of resolutions. Mine ranges from 800x600 to 1920x1080. They also support multi-displays. I've never tried it but I have seen other users with 2 or 3 ...
  • 103
0 votes
3 answers
105 views

FLOP - exact definition

I am interested in how much a GPU can actually calculate. When learning about rendering, it seems like a good idea as well. My Question: WHAT EXACTLY is a Floating Point Operation? I know what a ...
0 votes
1 answer
134 views

Using Software Rendering to make an External GPU

I was looking at the Raspberry Pi 4 specs and realized that running a software renderer on the pi could get some decent graphics. By using USB ports and Ethernet ports, I think I might be able to send ...
0 votes
1 answer
65 views

is utilizing multiple GPUs better than using one for all software?

In cases of Laptops, some will have two GPUs. One is built-in (e.g. Intel) and another one attachable (e.g. Nvidia). Is it better to have each of them operate different software to divide the load? or ...
2 votes
2 answers
264 views

Why do some videos look smoother and others don't on the same monitor

I have a Dell monitor with 60 Hz refresh rate and therefore games playing on it are not very smooth as compared to 120 hz. But when I open videos on YouTube of the same game on the same monitor, some ...
1 vote
0 answers
82 views

GL_OUT_OF_MEMORY Error when glDispatchCompute takes longer

I built a simple Ray Tracer which takes use of OpenGL's Compute Shader and traces ".obj" Files. The results are passed to the Host-Program via glMapBufferRange after finishing computing. ...
0 votes
1 answer
100 views

Operations in the Rendering Pipeline

I have a few short question regarding operations happening in the rendering pipeline (OpenGL, Vulkan, I'd expect for all questions to have similar answers for these two frameworks while DirectX might ...
  • 167
3 votes
2 answers
4k views

Optimizing compute shader with thread group shared memory

I'm currently learning compute shaders and I'm trying to write an optimized Game Of Life. I have a first version working that uses a Shader Storage Buffer Object. I dispatch a thread per cell I want ...
0 votes
0 answers
67 views

GPU accelerated 2D graphics library with signed distance functions

I want to create my own 2D graphics library using signed distance functions where everything is drawn on the GPU. Browsing the web I quickly found Inigo Quilez website that describes how to draw basic ...
  • 101
1 vote
0 answers
49 views

What modern C++ Unit Testing Framework do you use for highly parallel image, 3D and higher dimensional matrix processing workloads? [closed]

Most of the existing open-source libraries such as OpenCV, Tensorflow, and Pytorch use Google Test. I would like to know about strong alternative testing frameworks that thrive under a similar ...
1 vote
1 answer
78 views

Do we have OpenGL shaders write out the intermediate *.o-esque file on first compile so that we can skip compiling them in run-time every single time?

Shaders get compiled during runtime every single time an app starts. I understand the reasoning for this is that there are so many GPU architectures out there that it does not make sense to ship pre-...
  • 113
2 votes
0 answers
26 views

Visualizing auras around clusters of points

For a real-time application I am trying to add glow to a few hundred points. These points are more like 2D circles of various small sizes on the screen. These points are supposed to be like balls of ...
  • 243
0 votes
0 answers
269 views

Queue from a family queue that supports presentation doesn't work [Vulkan]

I'm following this tutorial on Vulkan and I tried to see if it actually worked with different queues for drawing and presenting the image of the triangle. My GPU supports presentation with every queue ...
  • 9
0 votes
1 answer
341 views

Sending programs from CPU to GPU

How does a CPU know which piece of code should be executed on GPU (i.e how to recognize whether a code is shader or not) and how is it been sent to GPU VRAM?Is it via DMA?
0 votes
0 answers
196 views

After making any 3D model using Opengl how to export it's data in .dae file?

I'm trying to make simple 3d modeling program in c++ with Opengl. So I don't know how to transfer/convert/export or save data of 3D model in .dae file. NOTE:- 3d model is made using Opengl ...
3 votes
3 answers
546 views

Latency difference between texture copy on GPU vs CPU?

Does anyone happen to know the latency difference between copying a texture on cpu (using, say, memcpy) compared to gpu (using, say, ...
1 vote
1 answer
55 views

Performance of particle implementation

I am about to implement the renderer for a simple particle system. Every particle has to be updated every frame (position, rotation, scale) and so I was wondering which way is usually faster: ...
  • 113
4 votes
1 answer
500 views

Double buffering always necessary?

Question from the graphics newbie: What I know by now is that "usually" double buffering is used: You calculate the content of frame x+1 while frame x is shown on the screen, if you want to achieve ...
  • 277
1 vote
1 answer
144 views

Tessellation Speed in terms of interpolated points per second

For a research project, I have written a Tessellation Shader to interpolate solution points of a 3-Dimensional mathematical equation. Therefore I don't use any techniques for displacement, I merely ...
  • 11
3 votes
1 answer
884 views

Is it possible to do color conversion on decoded video without getting it from GPUmemory to CPU memory?

Decoding video will result in different pixel formats on different GPUs and drivers. I have an application which can render (using OpenGL), only the RGB8 pixel format, therefore I need to do color ...
0 votes
0 answers
193 views

Software Rendering Display Latency

I've been doing some reading as to how to reduce input delay on modern computing devices. A lot of my understanding from this comes from the VR research paper by Carmack explaining the numerous ...
7 votes
1 answer
2k views

Is vkCmdPushDescriptorSetKHR efficient?

I am new to Vulkan. Compared to DX12, the resource binding procedure seems a little bit complicated. In terms of per draw call uniform buffers, I currently have two approaches in mind, and they both ...
  • 105
2 votes
0 answers
26 views

Should I use common resource bindings across all my shaders in HLSL/D3D12?

Will unaccessed bindings consume cache or bandwidth? I want to improve cache usage by splitting the per-path information in my renderer across different resources, but I also want to keep my binding ...
2 votes
1 answer
887 views

Floating Point Operations in Line Drawing Algorithms

When it comes to line drawing, DDA is the simplest and most intuitive algorithm, the core idea being: compute the slope of the line for every increment in x, increment y by the slope However, DDA is ...
  • 121
1 vote
1 answer
263 views

How can I control memory for command allocators and command queues in D3D12?

The D3D12 programming guide suggests specific allocation strategies for command queues and allocators, but the creation methods for those objects operate directly on queue/allocator pointers without ...
2 votes
1 answer
187 views

What does GPU serialized mean?

With GPU isolation : multiple processes use the single GPU will be serialized, cause OOM easily link https://www.slideshare.net/mobile/Hadoop_Summit/deep-learning-on-yarn-running-distributed-...
4 votes
1 answer
390 views

How much does driver quality affect Vulkan and DX12 compared to OpenGL and DX11?

Considering the fact that Vulkan and DX12 delegate most of the memory management / work scheduling tasks to the application itself, how much do drivers affect performance? To put my question into ...
  • 291
2 votes
1 answer
2k views

Do I need separate Graphics and Transfer commands?

I was following this tutorial on Vulkan https://vulkan-tutorial.com/Vertex_buffers/Staging_buffer and I decided to do this bit: The buffer copy command requires a queue family that supports transfer ...
9 votes
2 answers
2k views

Why are oct trees so much more common than hash tables?

When reading papers I commonly find Oct tree implementations of geometry representations to sort the data. However whenever I think about the problem hash tables seem better overall. Hash tables have ...
  • 1,594