Derrel
Mr. Rain Cloud
- Joined
- Jul 23, 2009
- Messages
- 48,227
- Reaction score
- 18,938
- Location
- USA
- Website
- www.pbase.com
- Can others edit my Photos
- Photos OK to edit
Juts a couple minutes ago, a forum contact sent me a PM and asked me about the Lensbaby versus Tilt/Shift lenses. Here is some of my reply:
Well, tilt/shift can be faked in post pretty well with the "minature effect", as you probably know. But the Lensbaby effect is pretty hard to really,truly fake,especially the horrific chromatic aberration the original Lensbaby delivered. Sure, blur effects can be faked, but the genuine, single-element original LB had bad chromatic aberration, plus at wider apertures it had a lot of overall, low-level of what's called "veiling glare". Bottom line: crappy optics are hard to simulate in post because the effects are applied from the highlights into the shadows, where a LB, or a diffusion filter, actually SPREADS part of the light from the highlights into the shadowed areas and mid-tones, which affects the dynamic range,and the original exposure. Re-arranging the pixels later in PS, does not do the same. Close, but not the same-same.
snip> I happen to be a Lensbaby believer. I have the Original in Canon and Nikon mounts, and the Lensbaby 2.0 in Nikon mount, and I also have the 3.0, which has been re-named the Control Freak.
I prefer the original model, with its poorer optics and significantly higher chromatic aberration. The 2.0 has better optics (2 elements, coated) and is hard to focus. My hit rate with 2.0 SUCKS. I don't like 2.0 much. The Control Freak is very slow in operation, but works repeatably,which is its claim to fame. I prefer to use the things with a 1.4x or 2.0x telephoto converter, to exaggerate the loss of optical quality.
I actually vastly prefer the poorer optics and fuzzier,more ethereal images from the original Lensbaby. The 2.0 model is hard to focus for me. COntrol Freak is slow, but the effect is totally repeatable,and will work for slow speed shots, whereas the Original and 2.0 will not. The newest one, with the simple ring system, might be the handiest model--I've not tried that one.
Well, tilt/shift can be faked in post pretty well with the "minature effect", as you probably know. But the Lensbaby effect is pretty hard to really,truly fake,especially the horrific chromatic aberration the original Lensbaby delivered. Sure, blur effects can be faked, but the genuine, single-element original LB had bad chromatic aberration, plus at wider apertures it had a lot of overall, low-level of what's called "veiling glare". Bottom line: crappy optics are hard to simulate in post because the effects are applied from the highlights into the shadows, where a LB, or a diffusion filter, actually SPREADS part of the light from the highlights into the shadowed areas and mid-tones, which affects the dynamic range,and the original exposure. Re-arranging the pixels later in PS, does not do the same. Close, but not the same-same.
snip> I happen to be a Lensbaby believer. I have the Original in Canon and Nikon mounts, and the Lensbaby 2.0 in Nikon mount, and I also have the 3.0, which has been re-named the Control Freak.
I prefer the original model, with its poorer optics and significantly higher chromatic aberration. The 2.0 has better optics (2 elements, coated) and is hard to focus. My hit rate with 2.0 SUCKS. I don't like 2.0 much. The Control Freak is very slow in operation, but works repeatably,which is its claim to fame. I prefer to use the things with a 1.4x or 2.0x telephoto converter, to exaggerate the loss of optical quality.
I actually vastly prefer the poorer optics and fuzzier,more ethereal images from the original Lensbaby. The 2.0 model is hard to focus for me. COntrol Freak is slow, but the effect is totally repeatable,and will work for slow speed shots, whereas the Original and 2.0 will not. The newest one, with the simple ring system, might be the handiest model--I've not tried that one.