Matching MapLibre's Igor Hillshade with GDAL

[diehl]

I've got a map with a terrain layer where I'm rendering the hillshade using MapLibre's Igor algorithm with the following settings.

{
  "id": "Terrain RGB",
  "type": "hillshade",
  "source": "terrain-rgb-v2",
  "minzoom": 0,
  "maxzoom": 22,
  "layout": {"visibility": "visible"},
  "paint": {
    "hillshade-shadow-color": "hsl(0, 0%, 0%)",
    "hillshade-exaggeration": 0.35,
    "hillshade-method": "igor"
  }
}

I'm trying to precompute the hillshade with GDAL so I can make the map more responsive and so far, I've not been able to match MapLibre's output with GDAL. I'm curious if anyone has any insight about how to do this? I've not gone so far as to compare the algorithms yet. I love what I'm seeing in terms of the result in MapLibre. GDAL's Igor output is not as visually striking.

[HarelM]

Can you please share more information in order for someone else to be able to reproduce what you see?
CC: @NathanMOlson

[diehl]

Will do @HarelM. I'm realizing I need to do a few more things to make sure I've got a true apples-to-apples comparison here. The terrain layer I'm currently rendering uses the publicly available Terrarium-encoded Terrain RGB data on AWS. What I'm pre-computing with GDAL is leveraging the DEM data from the Copernicus GLO-30 dataset. So I need to visualize the Copernicus DEM with client-side rendering of the hillshade and check how that compares to the GDAL result. I'll circle back once I've done that.