j4nu wrote:
With my SY's (recent AF ones) purchases, I have not noticed any more deviation than in Sigmas or Sonys for that matter...
Edit: I did get my only so far DOA with SY85F1.4 though...
keepcoding wrote:
Good point. I've had slightly more issues with QC with Samyang lenses than with Sigma lenses. However, I always try to buy from a place where I can return the lens since in my experience about every third lens (regardless of manufacturer) comes with QC issues. Even though this is annoying and I hate to return a brand new lens (which will potentially be scrapped), I kind of got used to it.
Well, luckily for me I've never gotten a decentered lens in like... 6 purchases?
Not many retailers in Sweden, so returning can be a bit of a pain if you don't live within a city that has one. Bought my recent 40 G through American relatives when they came to visit, but unfortunately they couldn't figure out how to manually focus lens/do a decentering test. Fortunately for me... my copy is 100% perfectly centered .
Heck, my 20 G is still centered even after I dropped it.
My next lens will probably come destroyed for karma
JVan_02 wrote:
Well, luckily for me I've never gotten a decentered lens in like... 6 purchases?
Not many retailers in Sweden, so returning can be a bit of a pain if you don't live within a city that has one. Bought my recent 40 G through American relatives when they came to visit, but unfortunately they couldn't figure out how to manually focus lens/do a decentering test. Fortunately for me... my copy is 100% perfectly centered .
Heck, my 20 G is still centered even after I dropped it.
My next lens will probably come destroyed for karma
Yeah, I think it's mostly luck of the draw for regular customers like us. Now, I also think more than the brand, it's the particular lens model that affects the chances of getting a good one. There were times when I've sent multiple GMs and ARTs back but kept the cheapest SY on the first try...
I hope Sigma maintains a pragmatic approach in this lineup rather than faithfully segmenting the Sigma I-Series as speculated earlier in this thread:
- Small size, small aperture and less clinical
- Medium size, medium aperture and sharp
This has been a good year for consumers with Sony and Sigma sometimes breaking out of the typical lens design molds.
- Primes with apertures f/2.8 and smaller can be aimed at professionals
- Long and short lenses shouldn't necessarily have the same aperture sizes
- Zoom lenses can have shorter ranges
- Lenses don't have to be sharp at their widest apertures or shortest focus distances
I personally hope lens manufacturers start disregarding lens reviewers and aim their focus towards specific user groups and use cases.
It would be amazing if Sigma created some lenses sharing the characteristics of the 45mm f/2.8, but with wider apertures for the more wide angle lenses.
Charlie N wrote:
Updated with the latest nokishita posting
The 90mm lens sounds good--but there's this: "The lens also uses your camera's optical correction functionality, that way it improves rendering performance while reducing the size and weight of the lens." I suppose that that caveat suggests in part that distortion correction may often be useful and that the expected (uncorrected) high resolution results in initial reports will require some interpretation.
zugzwang2 wrote:
The 90mm lens sounds good--but there's this: "The lens also uses your camera's optical correction functionality, that way it improves rendering performance while reducing the size and weight of the lens." I suppose that that caveat suggests in part that distortion correction may often be useful and that the expected (uncorrected) high resolution results in initial reports will require some interpretation.
The camera’s optical correction function can also mean correction of lateral CA. Notice how only correction of axial CA was mentioned in the text, though five SLD elements and aspherics mean that even lateral CA may be well corrected optically. This leaves distortion, but a 90mm is not likely to have more than 2% native optical distortion, which shouldn’t be a deal breaker.
The Sigma 24/3.5 has very little optical distortion, particularly compared to the Sony 24/2.8 G at a staggering ~8%. The 45/2.8 is also nice in this regard, so in summary I think Sigma will strike a good balance
The optical design of the 90/2.8 sounds outright ambitious with five SLD elements, and a statement describing something that sounds like an apochromat stopping short only of saying that word.
trogdon wrote:
The 90 is pretty exciting with those specs, but as an LA-EA5 owner I’d probably just go back to the 85 SAM which I’ve used previously and liked.
The 24 is very odd coming out so soon after the 3.5 version.
If we think of the I-series as two lines in one, a line of f/2 lenses and a line of slower lenses the 24/2 joins the 35/2 and 65/2, while we get 24/3.5, 45/2.8 and 90/2.8 as the slow speed series.
The 24/2 consists of 13 elements in 10 groups, which is more than the other 24/25mm E-mount lenses, except the Sony 24/1.4 GM. E.g. the Zeiss Batis 25/2 is 10/8 elements/groups. Hopefully the new 24/2 turns out great.
The actual 'need' for corrections will be of great interest to purists. Sigma have these AF issues going back, and have shown they are willing to stretch other aberrations, some OK, some not so much. A little breathing is OK by me - you get a 90mm for landscapes and a little longer for portraits.
We'll see soon enough - my rule is: no necessary 'corrections' and no firmware updates, that is a deal out for me. If it needs more than CA and vignette to taste in post, I'll wait on one from someone else, or make do. No computer-dependent lenses, not even once! My three MF lenses in this FL region all have 1% or less (pincushion), so do the good other ones.
None of the DG DN Contemporary have that low distortion, granted this FL could be a wild card. But given how small the lens is......
philip_pj wrote:
The actual 'need' for corrections will be of great interest to purists. Sigma have these AF issues going back, and have shown they are willing to stretch other aberrations, some OK, some not so much. A little breathing is OK by me - you get a 90mm for landscapes and a little longer for portraits.
We'll see soon enough - my rule is: no necessary 'corrections' and no firmware updates, that is a deal out for me. If it needs more than CA and vignette to taste in post, I'll wait on one from someone else, or make do. No computer-dependent lenses, not even once! My three MF lenses in this FL region all have 1% or less (pincushion), so do the good other ones....Show more →
zz wrote:
I've been reading comments on the 90/2.8 how it might be a macro option. That would make sense given the focal length, although Sigma already has the 105/2.8 for that of course. That's in the Art series however, so there might be a C option on the cards.
At 0.2x max magnification, this is not the lens you are looking for IMO.
tsdevine wrote:
None of the DG DN Contemporary have that low distortion, granted this FL could be a wild card. But given how small the lens is......
I can see where Philip's coming from, specifically on the 90. My only interest in this lens is for landscape and pano, and I appreciate a lens that, at 5.6 or so, has low to no distortion, very well-corrected CA and coma, low to no vignetting, strong retention of contrast in backlighting, even resolution across the field, and a flat field from infinity down to 10M. It's not difficult to find a short-tele prime that meets these qualifications. There's a ton of Sonnar inspired short teles from 85-135 that can get me what I want. They're all bigger and lack AF, but they can be had for half or less than the price of the Sigma 90. I'd happily pay for a size and weight reduction if it doesn't compromise on what they do very well. Otherwise, I'll stick with what I've got.
I'm not, however, a hard-liner about profiled corrections in post. I know there are going to be some compromises with the 24 but, if the performance is like the 35 and it's as-well profiled, then I won't have any problem buying in.
I'm thinking of something like the 100MP, optical design a first priority. But we have to be realists, I suppose. We'll see. I saw the expected date was 'in the next few months' but with all the info around maybe sooner? Juha will know soon enough.
I’ve probably said this before, but we need to remember that optical distortion correction is not free. Other things being equal, a lens that corrects distortion well trades that off against resolution. Or size.
So should we software correct for distortion? Well that depends. Is the post correction resolution and contrast better or worse than it would have been had it been designed, for the same budget and size, with optical correction? If so, yes, software correction is the correct choice, and optical correction would be a nostalgic compromise. But if not, then the optical correction is the best choice.
Of course we will never know those counterfactuals about any particular lens. But the best we can do is see how the lens performs after correction. Is the post correction contrast, resolution and geometry attractive for the form factor and cost? If so, it may be something you want.
Not arguing or disagreeing with Philip, nor am I saying that distortion and correction come for free, there is definitely is cost.
I was just was saying that none of the Contemporary so far have shown < 1% distortion, maybe this FL will break the trend of the other Contemporary lenses.
freaklikeme wrote:
I can see where Philip's coming from, specifically on the 90. My only interest in this lens is for landscape and pano, and I appreciate a lens that, at 5.6 or so, has low to no distortion, very well-corrected CA and coma, low to no vignetting, strong retention of contrast in backlighting, even resolution across the field, and a flat field from infinity down to 10M. It's not difficult to find a short-tele prime that meets these qualifications. There's a ton of Sonnar inspired short teles from 85-135 that can get me what I want. They're all bigger and lack AF, but they can be had for half or less than the price of the Sigma 90. I'd happily pay for a size and weight reduction if it doesn't compromise on what they do very well. Otherwise, I'll stick with what I've got.
I'm not, however, a hard-liner about profiled corrections in post. I know there are going to be some compromises with the 24 but, if the performance is like the 35 and it's as-well profiled, then I won't have any problem buying in. ...Show more →
Maybe we'll a compromise? Many lenses make this possible, you can either use the image more or less as it emerges, or make slight changes for your own tastes, or use the profile. And neither one looks too different from the other.
Here, they have a lot of good optical design material to work with in a high element count and special glass, so we might see that middle ground come to pass. It's sure to be a good lens either way.
As 'lenses have just been added to overseas retailers' we won't have long to wait, it seems.
In design tradeoffs I am perfectly happy to reduce all the primary aberrations at the expense of distortion. Distortion doesn't affect the point spread function - all the other aberrations blur the image directly, while transforming an image to un-do distortion results in a tiny loss of information but does not broaden points of light or sharp edges. To me only the final image matters. If the combination of optics (glass) and software produces a better image than glass alone can do because of the limitations of aberration balancing, then I'm all for it.
. 
zz wrote:
