Recently I unearthed my (original version?) Metabones Speed Booster. For those not familiar, this adapter is a revolutionary idea that takes full frame lenses and reduces the magnification and image circle to match smaller sensor sizes. They also have the added benefit of increasing the effective aperture. In my case I'm taking Canon EF mount lenses and using them on my Sony e-mount mirrorless cameras.
So one amazing feature of digital photography is the ability of software to correct lens defects. In fact profiles either built in or custom made exist for many lenses that correct for geometric distortion (pincushion or barrel distortion), chromatic aberration and vignetting. I rely on those corrections now.
And it occurred to me, how do profiles work when using the "Speed Booster"? There's glass in there, clearly there's going to be some different defects than the primary lens on its own. Does the profile for the primary lens apply at all when using the "Speed Booster"? How do image editing programs deal with the EXIF data? The exif data while using the SB (I'm going to abbreviate from here on out, ok....) gets changed to match the effective focal length and aperture. For example, when I use my Sigma Art 50mm f1.4, the EXIF data reads 35mm f1.
This is just an initial test, I hope to do more. Clearly using the manufacturer profile (embedded in the lens) isn't working right. I can imagine that somehow the correction software is thrown off by a 50mm lens reporting that it's 35mm.
Normally I'd say unfortunately I don't have any Canon mount lenses that have strong distortion to test. But of course, I'm lucky to have a wonderful stable of lenses.
The only lenses I have with serious distortion are native Sony e-mount lenses. And they were designed knowing that such flaws can now be corrected in camera on the fly and in post. Where as Canon lenses can still be put on film cameras where that's not possible.
Without regard to distortion, I found that the optical performance was pretty good in the center, OK at the edges. And AF was OK. Not
I hope to test this a little further later. But feel like I learned a little bit at least here.
(full resolution samples are here: https://urbantexture.smugmug.com/Hidden/LensbustersSamples/Metabones-Speed-Booster-and-Capture-One-profiles/n-kZcfGk/)
So one amazing feature of digital photography is the ability of software to correct lens defects. In fact profiles either built in or custom made exist for many lenses that correct for geometric distortion (pincushion or barrel distortion), chromatic aberration and vignetting. I rely on those corrections now.
And it occurred to me, how do profiles work when using the "Speed Booster"? There's glass in there, clearly there's going to be some different defects than the primary lens on its own. Does the profile for the primary lens apply at all when using the "Speed Booster"? How do image editing programs deal with the EXIF data? The exif data while using the SB (I'm going to abbreviate from here on out, ok....) gets changed to match the effective focal length and aperture. For example, when I use my Sigma Art 50mm f1.4, the EXIF data reads 35mm f1.
![]() |
Look at the straight line on the left: Not bad here applying "Generic Profile" in Capture One (using Canon Mount Sigma 50mm f1.4 wide open using Metabones Speed Booster = 35mm f1.4) |
![]() |
Look at the straight line on the left: Not bad here too using "Sigma 50mm f1.4 Art Profile" in Capture One (using Canon Mount Sigma 50mm f1.4 wide open using Metabones Speed Booster = 35mm f1.4) |
Normally I'd say unfortunately I don't have any Canon mount lenses that have strong distortion to test. But of course, I'm lucky to have a wonderful stable of lenses.
The only lenses I have with serious distortion are native Sony e-mount lenses. And they were designed knowing that such flaws can now be corrected in camera on the fly and in post. Where as Canon lenses can still be put on film cameras where that's not possible.
Without regard to distortion, I found that the optical performance was pretty good in the center, OK at the edges. And AF was OK. Not
I hope to test this a little further later. But feel like I learned a little bit at least here.
(full resolution samples are here: https://urbantexture.smugmug.com/Hidden/LensbustersSamples/Metabones-Speed-Booster-and-Capture-One-profiles/n-kZcfGk/)
Comments
Post a Comment