Spot metering on the brightest part of any scene and opening up 2 stops doesn't seem to blow anything out. You can sometimes get away with another third of a stop (to +2.33) before the histogram hits the buffers.

I have a stofen type flash diffuser that I'll probably use. It might need a little experimentation to get it right. I've a cheapo chinese clone bowl type diffuser on order but that'll probably not arrive in time.

Andrew

If you look at the histogram, there are relatively few burnt out pixels. If a single shot had been taken and exposed so that the building looked pretty much as the above shot, there might be large burnt out blobs where the tree lights are.

I'll be interested in seeing other folk's comments about this, as I'm hoping to be part of a camera club shoot at a large local shopping mall early next week. This should include shots of lit Xmas trees, possibly both indoors and outdoors. My guess is that we will not be able to use tripods, so long exposure HDRs will not be an option. My plan, at the moment, for at least some of the shots is to expose for the lights themselves plus two or three stops - and use flash to lift the tree foliage.

Andrew

If it is just on this photo, that would be my guess too.

Andrew

If you brighten the image quite a bit, it is fairly obvious. The levels on this crop look about right on my monitor - and I can easily see the blue discoloration, below and to the left of the lowest button on the lads grey jumper.




Andrew

I think your monitor must be set too bright if you can see that easily. I have to brighten the image quite a bit to be able to see discoloration in the areas you describe.

My bet is that it's a bit of lens or filter flare, caused by shooting directly into the light from the blinds.

Andrew

Congratulations - I wish I could get away with that.

Andrew

Indeed, yes - it looks a lot better corrected.

My main concern was whether it was worth using the old Canon scanner at all. Most of my earlier attempts have lots of noise which I had assumed to be scanner related, but this scan shows that notion to be wrong (probably). If the bad scans are noisy due to film grain, there's not going to be anything that a good dedicated film scanner can do about that.

Andrew

The young girl in the ride with her mum would have been 5 or 6 years old when that photo was taken, which dates it to 1977 or 1978. Doesn't time fly!!

Andrew

I've just scanned one or two more random negs on the Canon 8400F flatbed at 3200dpi (its max resolution) and had better success with some.

I'm beginning to think that most of the "noise" might be film grain from some of the negs. More recently (from 2000 to 2007) I tended to use either Fuji Superia 400 or Fuji Reala 100. Some of these are pretty good, even on the 8400f flatbed. Earlier, I'm not at all sure what the film was - possibly even including some freebies from Bonusprint, Truprint or whoever - though I do remember using Kodak 200 mostly.

This is one of the better test scans after a bit of improvement in LR5.2 ...




and a 100% crop from that .....




I don't think that's so bad for an old cheapie flatbed. I think I'm going to have to have a real good look at samples of negs before deciding that enough are worth digitizing to make an investment in a dedicated film scanner worthwhile.

Andrew

I've been thinking of scanning negs going back over many years (not all of them ,just those that look worth the effort). I've scanned a few test negs with an old Canon 8400F flatbed scanner that I've had for many years. The results are not impressive - noise, rather than dust being the main problem - which I assume is down to the scanner.

Is a scanner of the calibre of your Plustek capable of making relatively noise free scans - or is noise still a major problem?

Cheers
Andrew