• Global community
    • Language:
      • Deutsch
      • English
      • Español
      • Français
      • Português
  • 日本語コミュニティ
    Dedicated community for Japanese speakers
  • 한국 커뮤니티
    Dedicated community for Korean speakers

GPU vs memory for AI denoise?

New Here ,
Nov 03, 2023 Nov 03, 2023

Copy link to clipboard


Hi! I have found out that both GPU and memory impact the AI denoise times - but which would be the more important factor?

Let's say I wanted to get a MacBook with an M3 Pro processor. Would it be more beneficial to go with:

a)  M3 Pro with 14 GPU cores and 36GB unified memory

b) M3 Pro with 18 GPU cores and 18GB unified memory

Which upgrade over the base M3 Pro (14c, 18GB) will provide more beneficial results?


For context, I'm not buying any machine right now. I am happy with my low-er end Air (M2, 8c, 8GB). Only time I have to wait in LR is the AI denoise - it takes 30s for 16Mp photos and 35s for 20Mp. I was wondering if I should mainly focus on more RAM or better GPU when I will be getting a new machine in a few years.


If anyone has  compared how the denoise times compare on chips that are already out, like different configurations of M2/Pro/Max/Ultra (or M1), I would really appreciate if you could share these results. Thanks







Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
no replies

Have something to add?

Join the conversation