a6300 - slog2 vs slog3 - Does slog3 provide any benefit specifically on the a6300?

J

JoeThePro

Well-known member
From clips Ive seen (dont own the camera... yet), it doesnt appear that the slog3 implementation on the a6300 actually provides any more dynamic range than slog2. The shadows are a bit more lifted, but the highlights look like the hod the same amount of data. Can anyone confirm this?
 
Last edited:
This is the best a6300 DR test I've seen. It seems slog3 doesn't have any advantage in the highlights, and may have an advantage in the shadows but it's not clear with this kind of test. And of course slog3 is going to have banding issues relatively often, so you have to be sure about the advantages if you're going to use it.

It would be nice if somebody did a good DR test, slog2 vs slog3 (vs a7S or something like that?): set your scene, and shoot a big exposure ramp with both settings (e.g. expose for f/22, shoot at every stop between f/2.8 and f/22, then add six stops of ND, and shoot again at every stop from f/2.8 to f/22). Then see when highlights and shadows become unrecoverable in each case.
 
Slog2 and Slog3 have the same DR. Slog3 curve has room for upto 16 stops (future proofing?) which is why it clips at 93. The data distribution is far different with Slog2 allocating more data into the highlights and Slog3 having more data below.

If you expose properly and drop and de-log with the official sony rec-709 lut for each they should look pretty much the same in most circumstances. For me with the Fs7 I use Slog3 because I can use the Sgamut3.cine colour space which I prefer over the larger Sgamut3 and originaly sony Sgamut. I'm not the most technical cinematographer around but I also would image that because Slog3 is a slightly flatter curve like Cineon it is a little more forgiving with exposure. My perception could be completely wrong. Both are great curves and if I was shooting something where highlight data was key I would probably switch to slog2. In practice most will not notice much difference and may enjoy the benefits of sgamut3.cine for easier colour correcting. I'm not sure how that works on the A7sII/A6300/FS5 and because I'm using an Fs7 the 10bit gives me no banding issues so that isn't a concern which it might be on the 8bit cameras. Anytime you are spreading that much of data into 8bits its a risk.
 
The thing is: the theory works on the FS7, but not so much in the stills cameras, where we often get results that don't make a lot of sense (bad sgamut color on a7S, for example). It would be nice if someone checked...
 
You must log in or register to reply here.
Back
Top