Uncaught Error: argument not found
at r.lookupFunctionArgumentBitRatio (gpu-browser.min.js:14)
at WebGL2FunctionNode.L [as lookupFunctionArgumentBitRatio] (gpu-browser.min.js:14)
at WebGL2FunctionNode.astMemberExpression (gpu-browser.min.js:14)
at WebGL2FunctionNode.astGeneric (gpu-browser.min.js:14)
at WebGL2FunctionNode.astCallExpression (gpu-browser.min.js:14)
at WebGL2FunctionNode.astGeneric (gpu-browser.min.js:14)
at WebGL2FunctionNode.astExpressionStatement (gpu-browser.min.js:14)
at WebGL2FunctionNode.astGeneric (gpu-browser.min.js:14)
at WebGL2FunctionNode.astFunction (gpu-browser.min.js:14)
at WebGL2FunctionNode.astFunctionExpression (gpu-browser.min.js:14)
Seems like it should be a relatively common operation.
Expected behavior (i.e. solution)
The receiving kernel should have processed the array-from-texture successfully.
Other Comments
The application has a pipeline of image processing on the user camera on a laptop or phone, in a browser app. The pipeline starts with the HTML Video node connected to the camera, and ends with a canvas on the web page. We also wish to display the result of an intermediate step in the pipeline in a second canvas -- this requires extracting the image from the pipeline midway and writing it into a kernel attached to the second canvas.
This is my first gpu.js application, but I'm a decades-experienced software engineer in other genres. So, please insert typical noobie apologies by reference :-). Also, please let me know if I'm doing this the "hard way" somehow.
What is wrong?
Passing a 3d image ([480][640][4]) generated by texture.toArray() as an argument to another kernel generates an exception.
The HTML script block is:
The exception is:
Where does it happen?
Chrome browser with "latest" gpu.js (https://cdn.jsdelivr.net/npm/gpu.js@latest/dist/gpu-browser.min.js) on mac
How do we replicate the issue?
Jsfiddle: https://jsfiddle.net/t9Ljfevx/
How important is this (1-5)?
Seems like it should be a relatively common operation.
Expected behavior (i.e. solution)
The receiving kernel should have processed the array-from-texture successfully.
Other Comments
The application has a pipeline of image processing on the user camera on a laptop or phone, in a browser app. The pipeline starts with the HTML Video node connected to the camera, and ends with a canvas on the web page. We also wish to display the result of an intermediate step in the pipeline in a second canvas -- this requires extracting the image from the pipeline midway and writing it into a kernel attached to the second canvas.
This is my first gpu.js application, but I'm a decades-experienced software engineer in other genres. So, please insert typical noobie apologies by reference :-). Also, please let me know if I'm doing this the "hard way" somehow.