The 35 and 65 could conceivably have the same AF issues as the 45. The first reviews for the 45 didn't convey the scope of the AF issues. That only happened when consumers got their hands on the lenses. If you look at the Dustin Abbot or Christopher Frost reviews, they both show AF performance in well lit scenes. You might not notice it without using the lens in a wide range of situations.
Both the 35 and 65 are soft at minimum focus distance wide open, just less so than the 45. If the new lenses have no AF issues, that will beg the question of why the issues aren't fixed on the 45.
Dec 13, 2020 at 04:25 PM
imagesfromobjects Offline • • • • Upload & Sell: Off
Again - there's no fixing AF-C issues on Sony because it is PDAF-only. Sigma can't change that.
Foresska wrote:
The 35 and 65 could conceivably have the same AF issues as the 45. The first reviews for the 45 didn't convey the scope of the AF issues. That only happened when consumers got their hands on the lenses. If you look at the Dustin Abbot or Christopher Frost reviews, they both show AF performance in well lit scenes. You might not notice it without using the lens in a wide range of situations.
Both the 35 and 65 are soft at minimum focus distance wide open, just less so than the 45. If the new lenses have no AF issues, that will beg the question of why the issues aren't fixed on the 45....Show more →
I'll chime in on the AF accuracy of 45mm debacle, as I grabbed this lens during BF sale (that was the only offer that caught my eye and I also wanted to see for myself how AF-C is in practice, because the opinions ranged from unsuable to completely okay).
So, in my limited testing on A7III of AF-C vs AF-S in range of 0.5m (that's where the wide open glow gets less noticeable imho) to a couple of meters maximum I came to the following conclusions:
* AF-C is simply unreliable, I'd say in 9/10 cases AF-S gives noticeably better focused image. I don't have any other lens that behaves like that (some actually seem sometimes worse in AF-S).
* AF-C is still usable (45mm @2.8 is not that demanding even around 0.5m), but the chance you will have tack sharp image focused exactly where you wanted (e.g eye via eyeAF) is small (but we're talking mostly about pixel peeping at 100%). I took a lot of pictures in AF-C, that are worse than AF-S but still completely usable to me. The thing is you don't know when taking the shot how close (or far) AF-C precision will be compared to AF-S
* AF-C + slow/mid continuous drive is now the only way I use AF-C, so instead of taking 1 shot I take around 3-5, which in most cases gives me at least one focused "close enough" to AF-S.
* AF-S is too slow to use with my kids , so I hope Sony improves AF-S operation in next iteration of cameras (as I've read about other cameras) . What I mean, in most cases it's easier to take multiple shots in AF-C hoping at least one will be good enough, than to wait for the long enough "still" moment when AF-S manages to acquire focus.
So, to sum it up, I'd say AF-C is usable but luck is involved (i.e. you need some means to minimize its impact) ...
Maybe an example of AF-C (click the link to view the original size), shot at about 0.8m wide open, 3 shots exif claims one second apart from each other, taken from 100% crop, eyeAF I hope: AFC_1sec_series_100 by Jan U, on Flickr
So, to me they are all acceptable, with the first being the worst, but none of them is really tack sharp on the eye (actual focus is rather on the eye lashes and still not perfect)... For this kind of shot, this is an acceptable result for me though .
goldb wrote:
I'm curious what you would say regarding the picture above (of the young boy). Perfect example of a half body shot from probably 2-3 feet away? Personally its a good example of a lovely shot that would be, for me at least, one where I'd go "ugh, if only this were sharper." Hard to say without seeing the raw obviously, but in a browser at least, its a good example of a shot I wouldn't be happy with because of the softness.
I guess you and I just have different tastes. I find the sharpness more than acceptable and the rendering exceptional. I think it is a fantastic shot and shots like this one are exactly why I like the lens. Of course you may not feel the same way, and it sounds like you don't. We don't have to like the same things, but this sort of shot with this amount of sharpness and that rendering is exactly why some of us really like the lens.
imagesfromobjects wrote:
Again - there's no fixing AF-C issues on Sony because it is PDAF-only. Sigma can't change that.
Earlier today I tested this lens at 2 meters to compare focusing at f/2.8 and f/5.0 (more details in earlier post). I couldn't tell any difference in tracking in a dimly lit scene. I would expect it should've done better at f/5.0. However, I didn't take any pictures, so accuracy may have differed.
I could be interpreting it wrong, but it looks to me like the Sony FE 85mm f/1.8 also has some, perhaps less, spherical aberration. However, that lens is considered to autofocus well.
chez wrote:
Take a look at Fred's test of the AF using AF-C. and upto 5 meters distance. I would not call that sharp. Did you use AF-C in your selfie or AF-S?
See my previous post. AF-C is inconsistent. I always use AF-S, where it is the same as any other lens. If you want to use AF-C, I think you will find the lens maddening. I don't think anyone should try to hide or diminish that fact, like the light loss associated with the 100 STF, it's a factor that someone needs to consider in their purchase decision. I did, quite seriously, but I'm and undemanding AF user and don't use AF-C often at all, so I'm very happy that took a chance that the so-called sharpness issues were not much of an issue. But the AF-C is, IMO.
Steve Spencer wrote:
I guess you and I just have different tastes. I find the sharpness more than acceptable and the rendering exceptional. I think it is a fantastic shot and shots like this one are exactly why I like the lens. Of course you may not feel the same way, and it sounds like you don't. We don't have to like the same things, but this sort of shot with this amount of sharpness and that rendering is exactly why some of us really like the lens.
Agreed. For me, it comes down to what draws my attention / distracts. Like was said a few posts ago - we all see differently. For me, a single, very sharp and contrasty point in the image draws my eye in more so than a busy background bokeh distracts from it. I do quite like the rendering the 45mm gives (including in the previous image), but my brain has a tendency to look at the in-focus areas and gets put-off if those areas aren't tack sharp. I also tend to like intimate shots where the subject is close.
It is a balance - and sharpness at all costs is definitely not ideal. For me, a lens like the samyang 85mm gives very sharp images while still having very nice rendering (to my eye). It strikes what to me is a good balance; its not the sharpest lens ever but still is quite sharp, even close to MFD, with nice rendering. From what I've seen of the sigma 45mm, the close-up performance is a little bit softer than I personally like. Obviously, different strokes for different folks. It looks like the 35 /65mm may be close to the 85 samyang in terms of performance - which is exciting for my tastes.
Edit: sigma's new 105mm macro also looks great! Tack sharp but great rendering from what I've seen. Lenses like that make me wonder why it is we need substantial SA in order to get the rendering right. Perhaps 105mm is easier optically than 45mm in that regard.
Foresska wrote:
Earlier today I tested this lens at 2 meters to compare focusing at f/2.8 and f/5.0 (more details in earlier post). I couldn't tell any difference in tracking in a dimly lit scene. I would expect it should've done better at f/5.0. However, I didn't take any pictures, so accuracy may have differed.
I could be interpreting it wrong, but it looks to me like the Sony FE 85mm f/1.8 also has some, perhaps less, spherical aberration. However, that lens is considered to autofocus well.
If SA is the issue, these new lenses should have issues closer to MFD. They also have significant SA closer to MFD, just less so than 45mm....Show more →
On L mount, all AF lens focus WO. Maybe Sony doing the same. I no longer follow how sony implement this. (before it is kind of all over the place depend on light condition, particular lens and setting) So f5 will not change SA. I think from people's experience, SA might be only one of the issues, the small AF motor simply doesn't act fast enough, it will not play well with super fast AF system Sony capable of. Just need align your expectation.
Sony's 35mm use a top line dual motor on a small lens like that. It should be a choice if you value high speed AF higher than other property.
Dec 13, 2020 at 05:42 PM
imagesfromobjects Offline • • • • Upload & Sell: Off
I don't know if one can compare results from these two lenses to assess SA or AF-C performance. The Sony 85mm only focuses down to 0.8m. It's also an 85mm. Being such, it's going to be used in a completely different way.
I have already done more than my share of arguing the virtues of the Sigma 45mm. I even debated with Dustin Abbott - who I admire and respect a great deal. One of the only people on YouTube I can stomach anymore, in fact. He has a good eye and tests thoroughly, but I still feel like he missed the point - and clear advantage - of the Sigma vs the Samyang.
It doesn't really matter - if someone likes the lens and can live with its quirks and play to its strengths, they'll be happy. If someone wants an AF speed demon or a lens that will destroy an MTF chart, they would be best off looking elsewhere.
Foresska wrote:
Earlier today I tested this lens at 2 meters to compare focusing at f/2.8 and f/5.0 (more details in earlier post). I couldn't tell any difference in tracking in a dimly lit scene. I would expect it should've done better at f/5.0. However, I didn't take any pictures, so accuracy may have differed.
I could be interpreting it wrong, but it looks to me like the Sony FE 85mm f/1.8 also has some, perhaps less, spherical aberration. However, that lens is considered to autofocus well.
If SA is the issue, these new lenses should have issues closer to MFD. They also have significant SA closer to MFD, just less so than 45mm....Show more →
I thought the same about Dustin Abbott's review. It was fair, but I was a little surprised that he wasn't blown away by the lens.
I don't mean to be a stickler or nitpicker about this subject. It's more a point of curiosity for me. It will be interesting to see what kind of AF performance these lenses have, including the 24mm, when people get their hands on them.
Personally, I was a disappointed by the announcement of these lenses. They will probably sell better than what I hoped for, though.
chez wrote:
I thought you weren't fond of the OOF images the 45 was producing when using AF-C at close distances. I don't think this is image character we are talking about...but just straight back focus. Wasn't that your conclusion?
This is actually interesting. When I first tested the 45/2.8C on my A7R3, I was getting phase detect AF inconsistencies at any distance -- even at long distance, so I never thought this was caused by SA.
I believe my review was the first one that reported this: https://www.fredmiranda.com/forum/topic/1608234/0?keyword=fm,review#chapter9
I do think under-corrected SA will cause accuracy issues at MFD because it's very close for a 45mm lens at 0.24m and phase detect AF can't be very precise there.
Sold and re-purchased it a year later and now with the A7R4 and the new firmware, I no longer have the issues with AF-C. It's very rare the lens misses focus at mid-distance and even close distance (>1m). Sigma didn't officially acknowledge any AF fixes with the firmware but something changed -- perhaps it was the tweaked AF algorithms going from A7R3 to A7R4. -- or maybe my first copy was faulty.
Honestly I don't know but currently, I don't even think there is inaccuracy in phase detect AF with this lens, except when focusing close to MFD.
imagesfromobjects wrote:
It struggles with AF-C on a Sony, which - correct me if I'm wrong - only uses PDAF in AF-C. SA and PDAF don't play well. So it's not entirely the fault of the lens.
In AF-S and average light, it's great. Dim light and it starts becoming a problem. Often it will lock on a vertical edge (CDAF just LOVES vertical edges) and not necessarily where you meant to focus. For human faces, often this is the hair, so it presents as "back focused" and the eyes/nose will be slightly soft. Wives tend to love this lens, test charts, not so much.
In very low light or with fast-moving subjects, I prefer to just use it as a manual focus lens.
How is the manual focus, btw. Focus by wire, linear or accelerated. One thing I've noted about my 70 Macro, is that I don't care for the focus by wire when trying to make small adjustments. Seems to want to "not move", then "jump" with too big a move over my small adjustment. Wondering if the manual focusing (close focusing) on the 45/2.8 is improved, or is it still a trick to use FBW manually?
RustyBug wrote:
How is the manual focus, btw. Focus by wire, linear or accelerated. One thing I've noted about my 70 Macro, is that I don't care for the focus by wire when trying to make small adjustments. Seems to want to "not move", then "jump" with too big a move over my small adjustment. Wondering if the manual focusing (close focusing) on the 45/2.8 is improved, or is it still a trick to use FBW manually?
zhangyue wrote:
On L mount, all AF lens focus WO. Maybe Sony doing the same. I no longer follow how sony implement this. (before it is kind of all over the place depend on light condition, particular lens and setting) So f5 will not change SA. I think from people's experience, SA might be only one of the issues, the small AF motor simply doesn't act fast enough, it will not play well with super fast AF system Sony capable of. Just need align your expectation.
Sony's 35mm use a top line dual motor on a small lens like that. It should be a choice if you value high speed AF higher than other property. ...Show more →
+1 regarding expectation alignment.
If you're comparing a price point with dual (or triple) motors vs. a price point with single motor, thar goes back to my point earlier about putting more money into both, gears and glass. Here again, "Contemporary" build to a price point. The Sony 35 goes about 30-35% more. Hard to say how much of that is glass and how much is gears (generic term).
IQ
Build, Performance, Speed
Cheap
Pick any two of the above three. But, you're not likely to find all three in the same place.
Heck, some lenses have good IQ, without any AF (good build, MF), and still aren't cheap.
Fred Miranda wrote:
You should try it before dismissing it. What you think is "poor" should read "intentional". It'd say the lens is not for everyone but some of us love the images it produces.
Exactly. I only recently acquired a copy of the Sigma 45 because, foolishly, I was put off by the ill-informed criticism of the lens that punctuates this thread. After a couple of weeks of use, it’s now one of my all-time favourite lenses. But as Fred, Michael, Steve, and other members have pointed out, the Sigma 45 is not for everyone. It has, however, brought me back to shooting with a Sony after nearly two years of using Leicas exclusively.
goldb wrote:
Reviews / images I've seen say the poor correction extends out to about 2 feet away - not just at MFD. 2 feet away on a 45mm is not all that close... If you like the soft look - then go for it. But its not just a normal instance of "slightly soft at MFD."
genji wrote:
Exactly. I only recently acquired a copy of the Sigma 45 because, foolishly, I was put off by the ill-informed criticism of the lens that punctuates this thread. After a couple of weeks of use, it’s now one of my all-time favourite lenses. But as Fred, Michael, Steve, and other members have pointed out, the Sigma 45 is not for everyone. It has, however, brought me back to shooting with a Sony after nearly two years of using Leicas exclusively.
So, with that in mind ... would you consider it a "poor man's" choice for a "Leica-like" rendering? By that, I mean do you think that there is any appearance of Leica design influence to Sigma as part of the L mount Alliance?
That's not to suggest that the Sigma is a Leica ... just that they might be starting to think a bit more in that direction for lens design?
Yes, I think we also need understand inconsistent may mean different things to people. If subject not moving, you hit AF 10 times get different results means inconsistent. Both sigma FP and SL2 works fine in this case. After that, If subject moving slow at a few meter, AFC can’t nail, that is move to different scenarios. SL2 seems still work well under this case. FP will fail half of the time in my case.
If subject move fast, focus distance is close, it will fail. In my case I think it maybe the camera or the both. For Sony user, it could be lens as bottle neck.
zhangyue wrote:
Yes, I think we also need understand inconsistent may mean different things to people. If subject not moving, you hit AF 10 times get different results means inconsistent. Both sigma FP and SL2 works fine in this case. After that, If subject moving slow at a few meter, AFC can’t nail, that is move to different scenarios. SL2 seems still work well under this case. FP will fail half of the time in my case.
If subject move fast, focus distance is close, it will fail. In my case I think it maybe the camera or the both. For Sony user, it could be lens as bottle neck. ...Show more →
Yes, everyone has different requirements for focus and what might be ok for one just does work for another. For me, this lens would be used to photograph subjects that move and I love using eye-AF for these conditions. If the focus is unreliable then it really does not matter how good the rendering is if the focus is missed.
Most images I see with this lens are of stationary objects where it shines...but for moving subjects it just might fall short.
I'm back to looking at the Batis 40 as I love what I get from the 25 and 85.
, so I hope Sony improves AF-S operation in next iteration of cameras (as I've read about other cameras) . What I mean, in most cases it's easier to take multiple shots in AF-C hoping at least one will be good enough, than to wait for the long enough "still" moment when AF-S manages to acquire focus.
.