Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Scarlet-W vs Epic-W Helium Decision

I will post some images once i've been able to sanity check the results, i'm just wrapping my head around some colour aspects at the moment and would hate to post something that wasn't correct.

I shot my desk at night at f8. Same lens and aperture on both A7sII and EW.

The EW shot what it could see and the A7sII i shot the scene at 25600, 12800 and 6400ISO.

I debayered in RedCine and tried to find the best settings. I pushed the image in RedCine to 12800 and +1 on the flut (which i understand is a stop). So that baseline was an effective 25600. I exported that as an OpenEXR with RWG into Nuke for comparison to the XAVC UHD A7sII mp4 files. Exporting an EXR means that i retain all the camera latitude but with a baseline of 25600ISO. In Nuke i could balance the image so the waveforms were similar.

Firstly, the A7s series are low light monsters. I disagree with noise comment above, whilst the original 7s did have blotchy issues (and i remember being one of the first to point this out) the sII is much improved. It's an utterly unfair comparison, full frame massive pixels and only 4K vs much smaller pixels and no claims anywhere that it could compete at this level of light.

So no surprises that the A7sII was a lot better at 25600, it was a very usable image. However the EW was pretty impressive still. It was still sharper (comparing a downsized 8K=> UHD with UHD). At 12800 the A7sII still much clearer but the E-W closer and at 6400 i think the EW could be used quite easily.

There is no comparison between the systems in general. I love my a7sII but for what it is and the EW is a stunning camera. I can't compare to Dragon though. I have no issues with it at 3200, and 6400 at a push - knowing that some post work is needed. I'm sure as soon as i started moving the A7sII image the compression at this level of light would break down and the image become mush.

I shot 3:1 compression which helps. Trying to denoise 10:1 creates a blobby structure probably down to the compression. I think the footage needs a different denoise approach, Neat Video not so good. But i was working on single frames, temporal denoise would be very helpful. This needs more experimentation.

If you're shooting a static scene i did frame averaging as well on the EW, that also clears up noise.

I found DEB worked well on one example but created purple highlights on another. Odd.

So it's just a curiosity comparison, but the image out of the EW is just so good.

To be honest all you have to do is download an R3D and push it in RedCine to see how the noise works.

cheers
Paul
 
I will post some images once i've been able to sanity check the results, i'm just wrapping my head around some colour aspects at the moment and would hate to post something that wasn't correct.

I shot my desk at night at f8. Same lens and aperture on both A7sII and EW.

The EW shot what it could see and the A7sII i shot the scene at 25600, 12800 and 6400ISO.

I debayered in RedCine and tried to find the best settings. I pushed the image in RedCine to 12800 and +1 on the flut (which i understand is a stop). So that baseline was an effective 25600. I exported that as an OpenEXR with RWG into Nuke for comparison to the XAVC UHD A7sII mp4 files. Exporting an EXR means that i retain all the camera latitude but with a baseline of 25600ISO. In Nuke i could balance the image so the waveforms were similar.

Firstly, the A7s series are low light monsters. I disagree with noise comment above, whilst the original 7s did have blotchy issues (and i remember being one of the first to point this out) the sII is much improved. It's an utterly unfair comparison, full frame massive pixels and only 4K vs much smaller pixels and no claims anywhere that it could compete at this level of light.

So no surprises that the A7sII was a lot better at 25600, it was a very usable image. However the EW was pretty impressive still. It was still sharper (comparing a downsized 8K=> UHD with UHD). At 12800 the A7sII still much clearer but the E-W closer and at 6400 i think the EW could be used quite easily.

There is no comparison between the systems in general. I love my a7sII but for what it is and the EW is a stunning camera. I can't compare to Dragon though. I have no issues with it at 3200, and 6400 at a push - knowing that some post work is needed. I'm sure as soon as i started moving the A7sII image the compression at this level of light would break down and the image become mush.

I shot 3:1 compression which helps. Trying to denoise 10:1 creates a blobby structure probably down to the compression. I think the footage needs a different denoise approach, Neat Video not so good. But i was working on single frames, temporal denoise would be very helpful. This needs more experimentation.

If you're shooting a static scene i did frame averaging as well on the EW, that also clears up noise.

I found DEB worked well on one example but created purple highlights on another. Odd.

So it's just a curiosity comparison, but the image out of the EW is just so good.

To be honest all you have to do is download an R3D and push it in RedCine to see how the noise works.

cheers
Paul

Hey Paul! Thanks for doing that! As I read your post I was waiting for what you'd say about the 2500 - 4000 range since that's the range I'm most interested in (I really don't go higher than 4000 on the A7s). The Scarlet-W with standard olpf at 3200 is really noisy for me and definitely requires post-processing with Neat-Video or something similar, which I'd like to avoid.
 
Hey Paul! Thanks for doing that! As I read your post I was waiting for what you'd say about the 2500 - 4000 range since that's the range I'm most interested in (I really don't go higher than 4000 on the A7s). The Scarlet-W with standard olpf at 3200 is really noisy for me and definitely requires post-processing with Neat-Video or something similar, which I'd like to avoid.

I think you'll be more than happy at that range. I realised that i did my tests with the Standard OLPF so wanted to try with the Low Light one.

I'll have a go with that range later if i can.

I've spent some of the time today wrapping my head around RWG and Nuke, just making sure i'm doing the right thing.

cheers
Paul
 
Back
Top