PhotoZone starts testing lenses on 5DSr

[ahsanford]

We often take for granted that the 1Ds III, 5D2 and 5D3 were all within 1 MP of each other. That allowed Canon users to tap into lens review sites that used a 'reference' camera for many years, which allowed us to (within that 21-22 MP FF testing ground) compare lenses relatively consistently against each other.

Oh, the times they are a changin'.

Disregarding the useful but likely one-off test on the 5DS rigs by LensRentals, PhotoZone is (to my knowledge) the first site that has published resolution numbers on a 5DS/5DSr rig. To them, LensTip, and others, this migration suuuuucks pretty hard* as it represents a need to retest everything. They'll be at this all year, I'd guess.

To the point, though: Photozone's reference camera is now the 5DSr, and they have two lenses to report on with the 50 MP:

• Sigma 24mm f/1.4 Art (the resolution trends vs. aperture are the same as other reviews, but the scale of the resolution is obviously increased)
• Canon 11-24 f/4L USM (impossibly, the center performance is best wide open, but it's an odd duck in the corners at those wide FLs)

To PhotoZone's credit, they refuse to give a 1-5 star rating to the optical quality of those two lenses until they've had more lenses to compare against.

- A

*DXO, despite being batsh-- crazy, oddly gets full marks here as they do update their lens testing as new bodies are released. But that's another conversation.

 

[Proscribo]

*DXO, despite being batsh-- crazy, oddly gets full marks here as they do update their lens testing as new bodies are released. But that's another conversation.

I thought they calculate most of the values and doesn't actually test them? At least that's what I've read & it sounds possible as re-testing everything well.. is quite a job to do. I could be wrong tho.

 

[ahsanford]

*DXO, despite being batsh-- crazy, oddly gets full marks here as they do update their lens testing as new bodies are released. But that's another conversation.

I thought they calculate most of the values and doesn't actually test them? At least that's what I've read & it sounds possible as re-testing everything well.. is quite a job to do. I could be wrong tho.

I've not heard that they estimate values -- that's interesting.

But Roger Cicala's 5DS vs. 5DSR vs. 5D3 resolution figures would imply that's not just a matter of sensor resolution at play. More pixels reward all lenses -- but how much they are rewarded and where in the frame they are rewarded is somewhat inconsistent. I imagine that might complicate running calculations.

- A