An experimental failure

[Snarkbyte]

I took this shot of an old steam engine replica at the county fair, thinking it would be a good subject for B&W. As usual, I needed to isolate the subject from the background, and while I was playing around in Silver Efex Pro 2, I hit on the idea of using two completely different monochrome schemes as part of the subject/background separation. I brushed in the background using the Soft Sepia preset (slightly modified), and the engine using the Yellowed 2 preset. As you can see below, this really doesn't work very well, but I haven't really given up on the original idea of different monochrome schemes in the same image. I thought I would post the failure for comments, ideas, and suggestions. If you think the whole idea is just plain dumb, that's ok, just say so... it's an experiment, after all, and some weird ideas just bomb.

Still, I would like to hear your reactions and comments.

[rob marshall]

Al

I think the term 'experimental failure' is a little harsh. You only discover new things by experimenting, so when it fails to produce anything of obvious value on a specific occasion it's hardly a failure.

I often combine two presets in Silver Efex Pro. I convert the whole shot using the first preset, then select a target area, feather it in CS5, and apply another preset to it. It often works very well as it highlights a specific area of a shot.

[Snarkbyte]

I think the term 'experimental failure' is a little harsh. You only discover new things by experimenting, so when it fails to produce anything of obvious value on a specific occasion it's hardly a failure.

Thanks, Rob, it's good to hear that the basic idea isn't a complete bomb. I completely agree with what you said... first attempts are very often less than great, but I don't mind calling this one for what it is (looks like a dumb PP trick for the sake of having a PP trick, which is not at all what I'm after). Besides, sharing failures is a good way to learn; there's a good chance that someone here has played around with the same idea and has some valuable pointers. Thanks for the response... I'll keep at it, and I think this is a good shot to experiment on... it's easy to tell when things go wrong.