V4 thoughts

[lhkjacky (Jacky)]

@david.123

The v4 marketing was somewhat misleading, one of the claims made was "50% speed gain on TensorRT-compatible GPUs - ok so where is this 50% gain? What was the basis for this claim?

I am sure they are refering to Iris v2 (MQ) vs Iris v1 (LQ).

@jacob.muchow

Iris v2:

• An improved version of Iris v1, offerring more details and sharpness especially to faces.
• It uses TensorRT models for Nvidia GPUs and could result in a 50% speed-up.

I had already predicted it before the release of v4, because if you check the Program Data Model folder, you can see that most of the models except Iris v1 already using TensorRT model.

Topaz Video AI 4.0?

I guess they are refering to Iris MQ (v2) TensorRT model, which is 50% speed gain over Iris LQ (v1) onnx model on Nvidia GPU. Hopefully, they will add TensorRT model for Iris LQ (v1) as well