White balance setting

I realise that capturing images with a consistent white balance is important, but is it better to use one over another? I’ve been using the ‘daylight’ setting on my Sony A7 which seems fine, but just wondering if I should be using a custom setting taken off my light source, or something different?

Hi @FujiLove

Your camera’s internal white balance at time of capture ultimately won’t matter since you are shooting RAW, and will be adjusting later in Lightroom (by white balance sampling off the film border).

However, I find it helpful to use a white balance that is close to the final white balance I will be using in Lightroom, so that the internal camera histogram preview will be more accurate. In my case, that’s usually a custom white balance of around 2500k, but may be different based on film, light source, etc.

Hope that helps!

Thanks Nate. Makes sense.

I’ve been shooting film-only for so long, I’d forgotten that white balance doesn’t really matter with RAW files :slight_smile:

Hi
Instead of making white balance photo per photo why don’t you make a first DSLR scan of the start of your film .
Then when you have imported all the photos into LR, you select all the photos including the “orange” one and you make a WB on this photo with the “auto sync “ activated
This is for me far more efficient and in addition you can average the “ orange” selection on a wider part
Regards
Philippe

Thanks Philippe, good suggestion.

Sorry for re-raising this one, and in full awareness that the only impact is on the preview appearance and not the final result in LR having used the picker tool to set the white balance, but is it best to set the DSLR temperature to match the light source, to have the most accurate preview? It’s not really clear to me why, in Nate’s case, 2500k works better than temp settings. I apologise for being ignorant of the fine nuances but the result is that I have this confusion!

I just want the preview of the histogram I see during capture to match somewhat close to what my final histogram will look like in LR after I sample the film border. But it is really just a preference since it is RAW and you’ll end up with the same result in either case after sampling the film border. If you prefer to have your camera white balance set to the temp of the light table, that is fine, too.

All clear, Nate, many thanks for answering.

When scanning with the Nikon ES-2, there’s often not a lot, if any, border to set the white balance to.

I’m finding that my results are a bit inconsistent and want to improve the white balance setting part of the process.

Appreciate any tips on how to best set white balance where there’s not enough border or you’re just not getting great results?

Thanks

Stephen

You really only need one frame per roll that has enough border for setting white balance… then in Lightroom, you can sync the white balance setting from that frame to the rest of the frames from that roll.

1 Like

Thanks Nate.

I adjust the shutter speed for each shot to expose to the right, which means that each shot is exposed slightly differently, so I do wonder whether applying the single sampled white balance would work in this instance, or whether it really does need a sample for each photo?

I don’t think small exposure changes would significantly change the white balance, so you should be fine… BUT, I would recommend sticking with one exposure setting for an entire roll if possible… definitely your call, but some of the features won’t work as intended if there are exposure variances between photos (for instance, “sync scene”).

As Philippe mentioned above, I usually use the 1st completely blank frame of the roll to find out the max exposure for the film base, and set the camera custom WB; and then try to use that same exposure settings for the rest of the roll.
However I do find that setting WB is tricky with a WB sampler tool. I seem to get different results depending on where I click on the frame border. Maybe it’s the film grain?
So I tried setting it based on the center of that 1st empty frame instead thinking it would be most representative of the whole roll. But I’m not sure how many pixels it averages for each reading. I’ve read on some older forums that it used to take a 5 pixel average, but that depended on zoom level because it actually sampled the preview and not the real raw image underneath. I was surprised by that.
Nate, do you happen to know what that tool does in Lightroom Classic now? Is there any way to average more pixels?
I’m also surprised that I get very different WB if I use Auto WB on that 1st clear frame instead of sampling it manually anywhere on that frame. In fact auto WB almost never works on any photo for me. It always skews everything towards some kind of warm or orange tint.
I don’t have any obvious vignetting in my negative shots as far as I can tell. And no blotches that I can see. I wonder what else it could be. Thank you in advance!
For my setup I use a Nikon D750 with a Macro lens on a tripod, with a 35mm film holder and an iPad with clear screen as light source. The iPad is about 4”/100mm behind the film frame and produces an even light as far as I can tell. It’s plugged in to keep the light level constant.

Sampling the film border or a completely unexposed frame seems to work best and should be valid for all frames on that roll.
The sample size in Lightroom is fixed at 5x5 pixels regardless of zoom ratio. It used to be adjustable (as it is in Photoshop) but that’s ‘progress’ for you.
The sampler is the best option because as you observe, Lightroom doesn’t cope well with auto white balancing colour negatives. I am not 100% sure why that is but is probably because automated white balance assumes that there is no heavy predominance of any one colour in the average photo. As there is a large amount of orange in a colour negative, the calculation made by the algorithm is inaccurate.

I’m using a blank frame to set the camera’s white balance before shooting the rest of a roll and then leaving the “as shot” camera WB in Lightroom. It seems to be producing good results, but is that same, better or worse than measuring it in LR instead?

Nikon Z7/ES-2/60mm f2.8 F mount

UPDATE: I compared the camera’s setting for a full frame custom WB to setting it in LR. Temp the same, Tint a few points different. So no diff really. More to taste and workflow preference