Blackmagic Forum

I tested and the speed difference between:
- Magic Mask in Resolve Studio 19
- Legacy Magic Mask in Resolve Studio 20 Beta
- Magic Mask G2


As you can see, not just Magic Mask V2 is much slower than Magic Mask in Resolve 19, also Legacy Magic Mask in Resolve 20 is slower than Magic Mask in Resolve 19 which is very strange!

Please keep in mind that I am running Resolve 19 in my laptop, and running Resolve 20 in my desktop with much powerful hardware, yet MM in Resolve 19 outperformed MM in Resolve 20! Also, my laptop screen is higher resolution (3200 x 2000) than my desktop monitor (2560 x 1440) which theoretically should reduce the performance a bit in my laptop.

I know that Resolve 20 is still in early beta and I hope BMD will fix this in the final release.

Comparing using very different hardware seems… fraught. Are you sure your laptop doesn’t have a more capable GPU than your powerful desktop?

I see you provided the system specs in your video. So I looked them up.

“The RTX 4070 generally outperforms the RTX 4000 Ada in gaming scenarios, offering higher frame rates and better performance in gaming benchmarks. “

So for Resolve your laptop is the faster system.

Joe Shapiro wrote:I see you provided the system specs in your video. So I looked them up.

“The RTX 4070 generally outperforms the RTX 4000 Ada in gaming scenarios, offering higher frame rates and better performance in gaming benchmarks. “

So for Resolve your laptop is the faster system.

Well, both GPUs are Ada Gen but my laptop is using the mobile version of RTX 4070 which draws 115W. However my mini desktop is using the SFF desktop version of the RTX 4000 20GB. Although this SFF version is drawing only 70W, it should be faster in Resolve than the mobile version of RTX 4070 8GB becouse it has more CUDA cores:


And when you combine the whole hardware all together, my desktop should outperform my laptop.

As for gaming, here is comparison between both GPUs:
https://technical.city/en/video/GeForce-RTX-4070-mobile-vs-RTX-4000-SFF-Ada-Generation

Clearly the RTX 4070 Laptop did not outperforms the RTX 4000 SFF, although they came very close.

I know that my laptop is powerful but it can't be faster than my desktop as I can tell the difference from the overall responsiveness of the system and when using Resolve.

I will do a becnhmark tests for both and share the results.

Based on an unscientific bit of messing about on some very average hardware, my impression is that Magic Mask 2 is much more accurate but slower than legacy Magic Mask. Legacy Magic Mask needs more babysitting and fiddling with strokes where MM2 is happy with just a dot.

Joe Shapiro wrote:Comparing using very different hardware seems… fraught. Are you sure your laptop doesn’t have a more capable GPU than your powerful desktop?

So to make my test more scientific, I installed Resolve 20 Studio on the same machine as V19 (Dual Boot). I repeated the same test as in the video, and I am getting very similar results.

Magic Mask in Resolve 19:
- 42 FPS in Fast mode
- 33 FPS in Better mode

Magic Mask Legacy in Resolve 20:
- 32 FPS in Fast mode
- 24 FPS in Better mode

Magic Mask 2 in Resolve 20:
- 13 FPS in Fast mode
- 11 FPS in Better mode

This confirms that Magic Mask in Resolve 19 is still much faster than Magic Mask legacy in Resolve 20.

Glad you took the hardware out of the equation. Hopefully they’ll take a look now that your evidence is more compelling.