Originally posted by aurele i guess doing a AF lense for some "rookie" brand is too risky. Moreover, it's hard for any AF to be good at f1.4, the DOF is very tiny.
In one hand, it's cheaper and quite good, but without AF, on the other hand, it could had some AF but unsellable if it would have been a bad AF or something like that.
I guess they choose to sell something, instead of selling nothing !
I think the problem with AF is one of R&D costs and the fact that some (all?) camera makers have proprietary AF algorithms, in many cases protected by patents. For instance, here's the situation with
Canon EF:
"Third-party lenses compatible with EOS electronics are manufactured by Sigma, Tamron, Tokina, and Carl Zeiss. The manufacturers of these lenses have reverse engineered the electronics of the EF lens mount. The use of these lenses is not supported by Canon. Sometimes compatibility problems arise, as no third party has access to Canon's specifications for camera-body communication. These compatibility issues mostly occur when using a newer body with an older third-party lens. Over time, most of these issues have been resolved by the major third-party brands. Nevertheless, it is not accurate to call these lenses EF mount, as that term is reserved by Canon for its own lenses exclusively"
I've come across lots of talk in other forums on the internet about people having to send their Canon Sigma lenses back in to Sigma, more so in years past, to have them re-chipped for compatibility fixes with the latest Canon bodies. (I think it's those third-party lens reverse engineered AF protocol compatibility issues which has to some extent led a lot of review sites to try to include a review of a given lens's AF capability.)
I think it's also fascinating that for Sigma's SA mount bodies, when they chose their AF protocol, they chose to use the
reverse-engineered Canon EF mount signaling and protocol, rather than come up with their own... while perhaps just a cost-cutting measure, I also imagine having to create a body that speaks the protocol perhaps helped Sigma to improve the implementation in their lenses as well by understanding it as a system better.
I get the impression Nikon's AF protocol is similar, and has to be reverse engineered. On top of that, Nikon doesn't put screw-drive AF motors in their low-end bodies which means anyone buying a low-end Nikon body only gets AF if they buy a lens with an in-lens AF motor. (Note that all Canon EF mount lenses have to have in-lens motors, and there is no such thing as an in-body motor for Canon since their mount does away with any mechanical motor connection completely.)
Even when the intellectual property is no longer covered by patents because they've expired, I don't think Nikon/Canon/etc. give out documentation to their AF systems to third parties, so if you are a third party lens maker and want to make AF lenses, it is not a trivial thing to do for many reasons... and if you do try, you need to have a repair network in place so that you can potentially take lenses back in the future for re-chipping if/when new bodies come out that don't work well with existing lenses.
Think about what Samyang has to deal with just in regards to making manual lenses:
1. Different register (flange) distances
2. Different bayonet mount diameters
3. Different bayonet lug designs
Their Pentax lenses have the advantage of auto-aperture, which fortunately for Samyang is pretty simple to implement in KA mount. Now if Samyang wanted to develop an AF lens and sell it for as many mounts as they currently sell their manual focus lenses, they'd have to:
1. Reverse engineer the Canon AF system, develop a compatible in-lens motor
2. Reverse engineer the Nikon AF system, develop a compatible in-lens motor if you want to target the low-end Nikon bodies as well as the mid/high-end bodies
3. Reverse engineer the Minolta/Sony AF system
4. Reverse engineer the Pentax AF system
Those four steps require a lot of R&D, especially if you're starting from scratch and have no help from any of the camera body makers, which I assume is where Samyang would be starting from. They'd probably need someone to come in with a lot of money and be willing to invest in the long-term with the company to make something like that happen, and I imagine it would take a lot of resources and personnel to pull off, and would make their lenses more expensive, likely having to go head-to-head with Sigma and Tamron... at that point, why would someone buy a Samyang lens (or whatever re-brand) over a Sigma or Tamron? There are already people who shy away from long-established third-party brands like Sigma and Tamron over first-party like Pentax, and if Samyang can't seriously undercut Sigma and Tamron on price or consistently coming out with unique or better products, I don't see them making market headway.
Right now Samyang is in a pretty good position - they're making solid products in a price segment that doesn't have a lot of competitors. I'd call it the modern value manual focus lens category. Voigtlander (Cosina) is above them, but no longer makes their lenses in Pentax mount, Zeiss is yet further above them and similar situation, and with the rising popularity of Pentax cameras the used market has seen supply dry up and prices rise to the point where new lenses from Samyang are not just competitive but in many cases better values.
While I can fantasize about inexpensive autofocus Samyang lenses (and maybe they'll surprise me and pull a rabbit out of a hat someday in that regard), I have a hard time envisioning it becoming a practical reality from a business perspective.