3

Is Iota proof of work being done only on CPUs right now, or also GPUs?

I see people mentioning their GPU when posting about how long it takes their full node to do PoW, but I thought that required CUDA programming (for Nvidia) or OpenCL? (for AMD). I know the ternary JINN processor is supposed to be more efficient for PoW, but who knows when that is coming.

Improve this question
1
  • You can choose a WebGl implementation, which runs on the GPU.
    – Akkumulator
    Commented Feb 11, 2018 at 11:01

2 Answers 2

Reset to default
1

IOTA has some OpenCL context Library: https://github.com/iotaledger/clcontext

But I don't know what you can do with it. Maybe this is what people ware referring to.

Improve this answer
1

A team working on a spamming tool tried to implement GPU transaction spammer but they ran into big issues with OpenCL as of 20/2/2018. The library is not ready as it always crashes, so it's whether to dig into the OpenCL or wait.

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.