There were quite a lot of discounts this year. You are probably correct about the vagueness. However, there could be something else, for example they made some secretive change in strategy that might be easier to notice if they estimated the entire market. Yes, I'm wearing a tinfoil hat, how about you?
Canon indicated a 1% growth in y/y unit sales for 2023. Full year CIPA data for 2023 aren't available and in any case report production and shipments, not sales. But for ILC shipments Jan-Nov y/y there was a 2% increase. That means Canon's 2023 market share is likely ±1% of their 2022 market share, i.e. not a significant change. Since they're already hovering at just below 50% market share, it means they probably didn't break that threshold but also means their dominance of the market is unchanged.
You mean, the fact that, in their financial documents Sony never reveals unit sales of their interchangeable-lens cameras? Canon sold 2.88 million units in 2023 and forecasts 2.80 million units in 2024.
Correct... well Canon hasn't made a legal fuss with the OEMs (yet) for manual focus lenses.
I believe that vignetting is an issue with all very wide angle lenses. Canon adds stretching via algorithms as well but optically or mathematically (or both) is okay for me. Coma is the main thing for astro and more noise in the corners can be mostly corrected in post.
They can't make a legal fuss about the use of the RF mount for manual focus lenses - you can't claim intellectual property for accessories that fit on to your patented gizmo. All Canon can do is to prevent the illegal use of their communication protocols, and they can't stop genuine reverse engineering of those protocols. I don't see why there are not TC's from competitors - all they have to do is have direct contacts going the the TC, and all it would fail to do is to report the modified f-number and focal length to the EXIFs - I could live with that.
If the modified focal length isn't reported to the camera, then the cooperative IS/IBIS would be broken so you would have turn off stabilization. That is a pretty small market and most would not understand so the griping level would be very high. A pure manual lens is sold as just that and creates no false expectations.
It would report the original focal length and the IS/IBIS would cooperate based on that. How much difference would that make in practice?
The IBIS would be under-correcting by a linear factor of 1.4 or 2 depending on the TC. The magnitude of the problem would depend on how much of the correction Canon automatically assigned to the IBIS. If you could turn off IBIS independently of lens IS, then no problem, but that is not an option. I suspect there isn't enough money in TCs to make the headache worthwhile to 3rd parties. Even in EF, the only meaningful TC vendor was Pentax/Tokina/Kenko, who are not currently rolling in development cash. Tamron and Sigma made TCs specifically for a few of their own lenses and AFAIK, most of the Tamrons were relabeled Kenkos. Metabones does make an RF speed booster for EF lenses and Red uses the RF mount on some of their cameras, but I suspect both of those may be bilateral connections, as in Metabones may be making the Canon branded speed booster for the C70 and good odds that Canon is making some sensors for Red. Further, the Metabones speed booster only needs to use EF protocol, since it is for use with EF lenses, so in theory, no special deal would be needed. Metabones also makes RF speed boosters for other Cine lens mounts like PL, but none seem to have electrical connections.
Yes, that lens is the only native lens that has close to a reasonable value (price, aperture, coma performance) choice. Weight is slightly less than my Sigma EF20/1.4 but not by too much.
For long lenses, IBIS makes only a very minor contribution to overall stabilization.
The Sigma DG DN lens looks very nice from a coma perspective, but the EF version is not nearly as good as the RF 15-35. Just not that many fast, wide lenses with good coma performance. Have you looked into any of the Cine lenses to see if there are some options there? Doesn't seem like AF is necessary for what you are doing although the R5 is very good at focusing on celestial objects even with a long lens. In this shot I took during the Jupiter/Saturn conjunction in 2020, the R5 was reliably autofocusing on Callisto. Shot was with the EF 800 5.6 L and a 2x TC, so the AF only had f/11 to work with. Actual shot was at f/16 and 1/30 at ISO 400 with quite a bit of post processing to bring Saturn and Jupiter's moons to a similar brightness level with the big dog. I don't have an equatorial mount, so had to keep the exposure short, but Jupiter was really bright, so short was needed anyway to avoid burnout. Still some motion blur and CA (thanks to the TC).
True, but 3rd party TCs tend to get used on some shorter lenses due to the fact that they don't have the forward projection and will physically fit, but I think the bottom line is that Kenko is really the only player and that is probably not a hill they want to die on. If Tammy and Sigma were making RF telephotos, then I could seem them adding TCs, but they would still likely be specific to a few lenses as in the past.
Nice shot!
Yes that lens is good but unfortunately, the Sigma 20mm F1.4 DG DN Art is for E and L mount only. If only they released it in manual focus in EF (or R mount) :-(
I haven't looked at cine versions to be honest. Their pricing has put me off and I guess that coma isn't a priority vs other performance aspects. I will have a look.
The only one for EF is
Sigma 20mm f/1.4 DG HSM ART for EF Canon
https://dustinabbott.net/2016/04/sigma-20mm-f1-4-dg-hsm-art-lens-review/
reviews the coma performance.
That said, for a multi-panel panorama it still can produce good results especially with StarXterminator
R5, 9 shots @ 20s, ISO5000, f2 (I think - not in exif).
Menindee Lakes, NSW. Bortle 1 sky
. Pannys are much nicer to drive than Sonys and the S1R dynamic range is very close to the R5 other than at ISO 400 where the dual gain kicks in on the R5 and it has a 1-1/2 stop advantage, but that re-normalizes at ISO 800 when the Panny dual gain turns on. There is a mint S1R on eBay right now for about the price of an R7. You are probably right about the cine lenses as the big issues in cine are focus shift and of, course, parfocal behavior with zooms. I just thought there might be a sleeper out there somewhere as there really are a LOT of cine lenes these days. We just don't see them much in the still segment because they are so specialized. Interesting note that the RF 15-35 is almost perfectly parfocal at least according to TDP and its coma behavior is very good, but two stops is a lot for what you are doing.The Sigma 14mm f/1.4 DG DN Art also looks nice for E/L but the Sony FE 14mm f/1.8 GM size and weight would be nice for blue hour foregrounds.Very pretty panorama. Yes, I knew the 20mm DG DN was only available in L and E, but it has about the best Coma performance I have seen. If the lens defines the job, you could spring for an S1R
