Questions tagged [cuda]
The cuda tag has no usage guidance.
22
questions
1
vote
0
answers
41
views
CUDADot (CUDA) taking longer than indicated by //AbsoluteTiming (~500x)
I'm using MMA 13.3 with a fresh install of cuda toolkit version 12.2. CUDADot is not as fast as indicated by absolute timing.
Running the commands found here
(https://reference.wolfram.com/language/...
5
votes
0
answers
152
views
Target Device->"GPU" Not Working For Machine Learning: LibraryFunction error
I'm running Wolfram Desktop version 13.2.1, and after downloading Visual Studio, Nvidia Toolkit version 12.1 and 11.2, and verifying that
...
2
votes
0
answers
64
views
SymbolicGPU documentation
I found an interesting files there
$CUDALinkPath -> ../GPUTools
here
...
1
vote
0
answers
49
views
Why CUDAImageMultiply so slow in such cases? How can I speed up it
When I'm doing Image composition task, I want to use CUDAImageMultiply to speedup my performance, however, After trying it, CUDA is very slow, I've tested the ...
4
votes
2
answers
451
views
CUDALink gives wrong results but no error on Mathematica 13.2
I installed Mathematica 13.2 under Linux Mint 21.1 (Ubuntu 22.04). After installing cuda-toolkit-11-8. I get True as answer from CUDAQ[], but if I start some calculations I get wrong results.
...
0
votes
0
answers
160
views
Integration using CUDA kernels
I have never work with CUDA in Mathematica. My task connected with the calculation of a big amount of the same integrals. I have only 8 cores processor but my PC also has video processor with large ...
1
vote
0
answers
88
views
Anti-Aliasing and Moiré Patterns in Export
I export a lot of plot expressions with Export["filename.pdf",graphicsexpr] and would like to prevent aliasing and Moiré patterns in the .pdf file. ...
1
vote
0
answers
100
views
CUDA JuliaSet Whitepaper Example with Vers. 12.3
As a newbie to using Mathematica with CUDA, I have attempted to execute an example presented in the "WOLFRAM WHITE PAPER CUDA Programming with the Wolfram Language": @ https://www.wolfram....
1
vote
1
answer
130
views
CUDAResourcesInformation Issue during CUDA setup
I am trying to set up CUDA on a Windows 10/64 bit machine with a NVIDIA GeFore GTX 1080 with Max-Q Design running Mathematica Version 12.3.
I have issued the following commands:
...
3
votes
0
answers
117
views
How to use _Complex types in CUDAFunctionLoad?
The documentation for CUDAFunctionLoad specifies various base datatypes you can use for your function's arguments, including ...
3
votes
0
answers
69
views
Unable to uninstall an older CUDAResource paclet
I am trying to uninstall an older version of CUDAResource but cannot do so. Any ideas please? Is there a manual way to do so?
...
7
votes
0
answers
189
views
Extremely Long GPU Initialization Times on Mathematica 12.2 and RTX3090 [closed]
I'm having trouble getting my new GPU system to work with Mathematica 12.2. I have two RTX3090 cards in my setup. My NVIDIA drivers are version 460.89 (the latest).
I have installed CUDAToolkit 11.0 ...
5
votes
0
answers
161
views
Is there a Mathematica CUDA ode solver?
I have a system of ODEs which I'd like to solve with many (thousands) different initial conditions. In the net there are articles that CUDA can speed up such computations, but I haven't found ...
1
vote
1
answer
62
views
CUDA Real_t data type yield wrong result
I'm working on writing CUDA code on MMA, because I found that CUDA on MMA can reach the max precision from this webpage.
When I run the code below with float type, ...
1
vote
0
answers
61
views
Discussion on computing capability of Mathematica and CUDA C [closed]
My research is computing on Legendre Polynomial, which will be done on Mathematica and optimized on GPU.
Now, My Mathematica program is completed and correct and I am working on translating it on CUDA....
1
vote
0
answers
56
views
CUDA and sparse arrays
Is there a way to use the CUDA functions (like CUDADot and so on) when we are dealing with SparseArrays? I found some answers to similar questions but they are extremely old (around five years ago) ...
3
votes
1
answer
103
views
CURAND library import error to Mathematica/Wolfram 12.1
I used Mathematica/wolfram 12.1 IDE and installed CUDA toolkit 10.2. All installed correct. I try to use library CURAND with #include with random integer generator on CUDA, but get a error:
...
3
votes
0
answers
96
views
Problems with CUDAFinancialDerivative in Mathematica 12.1
I am attempting to run some CUDA functions in MMA 12.1, specifically CUDAFinancialDerivative. As I recall, this function (and other CUDA functions) ran successfully on this machine with earlier ...
2
votes
0
answers
73
views
Using CUDALink on Ubuntu 18.04
I can not get CUDALink to work and dont know where to start. None of the CUDA functions works but there are no obvious errors. I am running Ubuntu 18.04 LTS and Mathematica 12.1.
...
4
votes
0
answers
213
views
Interactive CUDAFluidDynamics example without CUDA?
On Macbooks OpenCL works but not CUDA. I saw this fun example CUDAFluidDynamics in the CUDALink package and I wanted to play with it, but couldn't.
Perhaps someone ...
0
votes
2
answers
368
views
CUDA Trouble, MMA12 Quadro RTX 4000
Started this unfortunately as an answer in this question on GPUs under Mma12:
previous GPU question, while it should have gone into a new question. I will try to answer questions already posted on the ...
10
votes
0
answers
282
views
CUDADot[] takes ~3x time compared to Dot[]
It all started when specifying TargetDevice -> "GPU" in Classify[] yielded no speedup nor any activity in the GPU. Then I tried something simpler like ...