Has anyone here got the Sigma 30mm f/1.4 for his DSLR?
March 11th, 2010 by mikeIt looks like a beauty, but it seems like Sigma is really proud ($430) that no one else makes one yet...
Any technical reason it should cost double what a 50mm f/1.4 does?
This is good to know. Thanks for taking the time. Until you mentioned that view cameras don't have moving parts aside from the diaphragm, I had forgotten that they focus with bellows. (embarassed) I've only seen one once, when I was dating a girl who was majoring in photography. She was doing a view camera unit.
It is interesting that the same thing that makes them clunky is also what makes them superior.
What use do they have nowadays, when things aren't often blown up beyond 8x10? Just curious.
It would be nice if other companies apart from Sigma made the equivalent of a fast 50mm on 35mm for dSLRs and made it affordable. There must be a demand for it. In fact, I want one and you want one, that should be good enough! Get on with it, lens designers! :D
I didn't even know they made a 50mm f/1.4 autofocus. I've never seen it on the web. Only the 1.4. From f/1.7 to f/1.4 is about half a stop.
How do you like that lens? Does it give good results at all apertures, or can you tell that it wasn't made for this body? It seems like it would make a great portrait lens.
That is a great point, darin. ...and since I'm considering a K1000 as a backup....
I've always heard & read that lenses are at their best, sharpness-wise when stopped down a couple stops from wide open, and that as you stop down more, they suffer a bit in sharpness. Any experience for/against that?
A lot of what you said about image circles in the view cameras makes sense to me. But one thing didn't. You said that they have huge image circles, and that is why they are the sharpest & contrastiest on the planet. I don't think the lenses themselves are any better than the more common 35 mm lenses, it's just that since the negatives are so much bigger, the results are the sharpest on the planet. They are probably more expensive because of the amount of glass involved...
No technical reason. Probably due to the desire to amortize design and development costs coupled with a lack of competition.
You see, that is the tricky thing about Pentax. Their cameras are very good. Their optics are very good. But you won't get as many chances to get good optics at more reasonable prices like you would with Nikon or Canon. I fully believe (and tests generally back it up) that Pentax optics are as good as Nikon or Canon. The better high-end stuff is just as expensive as Nikon or Canon. But entry level stuff is better, and at about the same price as Nikon or Canon.
I thought about all this ahead of time, and decided that I would do at least as well with basic Pentax optics and the anti-shake mechanism instead of more third party optics available and no anti-shake.
I think I will put this lens on hold and think about it a while. Sales of the Pentax K100D and K10D are really starting to take off. When that becomes obvious to other lens makers, there will suddenly be more stuff available for it.
So the next steps for me will be the Pentax 50-200 and a proper P-TTL flash. Fancy stuff like a fast prime and macro will have to wait a while. :D
I wish Pentax would quit jerking around with pancake lenses (as pretty as they are) and make a fast, normal prime at a reasonable price for their digital SLRs.
No, I was testing two different lenses. What does that have to do with anything?You misunderstood. It has a lot to do with this. I know they were two different lenses. I just meant that it would mean more if they had the same effective magnification. (Ex. test a 50mm for 35mm format against a 75mm for APS sized digital.
Both images are made at f8Good to know, thanks.
Because it would prevent my getting smart ass replies when I'm trying to make a point with someone who doesn't want to understand the point?Well, I do understand your point; that 35mm lenses give better edge-to-edge sharpness. My questions were aimed at getting the whole story, not just this narrow viewpoint. Forgive me for being Devil's Advocate. I was teasing a bit, and it apparently went right over your head. I wouldn't suspect a photo-savvy person such as yourself would actually have a fingerprint on a lens while shooting pictures to show lens performance.
Did you read that link I posted earlier? They seem to have a wider breadth of data than you took. If you know the kind of magazine Popular Photography is, they are not afraid to point out the shortcomings of their advertisers. In fact, they are basking in glory when they point out a simple way to save money without losing performance. So I am just questioning things a little to get the full picture. It is possible they have missed something. But they do have quite a comprehensive testing environment there. You didn't make any mention of how each lens performed when wide open, vs. stopped down either. They did.
This large aperture standard lens was exclusively designed for digital SLR cameras featuring APS-C size image sensor
Canon makes a 35mm f/1.4L lens that sells for around $1300.
I don't the principle behind it, but 50mm lenses seem to be cheaper and small, and of course more in demand. Lenses such as 85mm or 35mm are bigger at the same apertures. Just using canon as an example, their 35mm f/1.4 is $1300, a 50mm f/1.2 is about the same price although the 50mm is faster
I think its that wide angles need fancier optical formulae to correct for barrel distortion, and telephotos need fancier optical formulae to correct for pincusion distortion.
But this 30 mm is equivalent to about 50mm in 35mm format, so it shouldn't have those problems.
I have heard really good things about this Sigma lens and have considered it myself. I'm not 100% sure, but I think that it's made for APS-C cameras (all Nikon and Canon Rebels & 30D etc.)...and will not work for a full frame camera. To me, that is it's worst trait...even if you don't use full frame.
A Canon 35mm F1.4 or a 28mm F1.8 or any of those prime lenses...are made to be used on full frame (or 35mm film) which means that when used on an APS-C digital...that you are only using the middle of the lens...the sweet spot.
I started writing book here on the subject but stopped myself before I went too far. Look at these two test images.
http://www.foodieforums.com/otherimages/zoomcloseup.jpg
http://www.foodieforums.com/otherimages/testcloseup.jpg
Both are shot with a D50 aimed at the same piece of newspaper. The amount of blowup is different because the lenses are different in focal length. Both lenses are expensive, top-of-the-line Nikkor lenses. Both images are taken from the extreme lower right hand corner of the frame. The second one is a lot better, right? The reason is that it is a lens designed for a 35mm camera. While we are at the extreme corner of the camera frame, we are not at the extreme edge of the image circle.
Putting a lens for a 35mm camera on a digital not only decreases the angle of view, it also provides better corner to corner sharpness. If you need more explanation, let me know and I'll go back to writing the book.
Yes and no... the autofocus 50mm f/1.7 is very much in demand so I've seen them selling used on ebay for the same as a new Canon or Nikon f/1.8. But I went for the 50mm f/1.4, and brand new from a shop it was a whole £75 ($145) cheaper than the equivalent Canon or Nikon. In the US, judging from B&H the price difference is nearer $80/100 (Nikon and Canon respectively) which is still not an insignificant amount. I did have to wait just under two weeks, but I'm not a professional who needs specific lenses at a moment's notice, so I'm happy to wait a bit for that kind of saving.
It would be nice if other companies apart from Sigma made the equivalent of a fast 50mm on 35mm for dSLRs and made it affordable. There must be a demand for it. In fact, I want one and you want one, that should be good enough! Get on with it, lens designers! :D
Just look at most lens tests...they often mention that sharpness, CA & distortion etc. is worse at the edges and corners. If your camera/sensor crops off the bad parts...then you are left with the candy centre...well maybe that's a tootsie pop...but you get the point. :D
This is true for almost every camera lens. As the the aperture closes, the light is narrowed so that it uses a progressively smaller surface area of the lens elements. This has an effect similar to cropping the edges of the image circle and corner to corner sharpness increases. At extremely closed apertures you start getting some interaction between the leaves of the diaphragm causing the light to reflect back and forth from the leaves themselves. The phenomenon is called diffraction. It isn't very severe really and most lenses will produce better corner to corner sharpness completely stopped down than they will wide open. So you get sharper as you narrow the light and use less lens surface until you start suffering from diffraction then performance backs up just a little.
Since the elements inside a photographic lens normally move around relative to one another to change focal length or focal point, the designer optimizes the lens to perform best (to have its ideal correction) at a given zoom setting, focal point and aperture. A couple of stops is common for fixed lenses and 3 is often common for zooms. It is easier to design the lens this way than to design them to peak at 1 stop down or 5 stops down.
Some lenses are designed for wide open use. Some of the Nikkors I have owned that fit this description are the 300 f2.8 ED, the 135mm f2 ED and the 85mm f1.4. The designers decided the users of these lenses want all the light gathering they can get so they desgned them to be best just one stop down from wide open and they are very well corrected wide open. These lenses actually soften up a little as you stop them down (but not much.) I guesss there are always exceptions to every rule.
It is interesting that the same thing that makes them clunky is also what makes them superior.
What use do they have nowadays, when things aren't often blown up beyond 8x10? Just curious.
The same use they have always had - perspective control.
That is what it seems like, but I haven't read yet if that is how it works out in practice. I'm sure the photo magazines have gone over it and I just haven't been reading them lately.
True, it isn't that much faster than a f/1.7, but I mainly shoot handheld and often in low light, so even a tiny bit faster is good. Plus I'll take all the light I can get when looking through dSLR viewfinders!
Did you do any apples-to-apples comparisons, i.e. with the same equivalent focal length?
No, I was testing two different lenses. What does that have to do with anything?
What about with the lenses stopped down as well as wide open? That second one is certainly very sharp. I can see the pulp in the paper.
Both images are made at f8
Why didn't you clean all the finger prints off of the first lens before shooting? :lmao:
Because it would prevent my getting smart ass replies when I'm trying to make a point with someone who doesn't want to understand the point?
Nor was my post referring to the article you referenced. I didn't read it. Your statement sounded to me like a lens for a 35mm camera that is soft wide open will also be soft wide open when mounted to a DSLR. If that is what your statement said then it is wrong. If that is what the article said, then it is also wrong. I doubt the article said that. I'll go back and read it.
When we talk about lens softness, we aren't talking about the center of the frame. We're talking about the corners - the edge of the image circle. A DLSR crops out the edge of the image circle of lenses designed for 24mm so the corners will be sharper. The image will have a narrow angle of view but it will be sharper in the corners.
My images were just to demonstrate that. Both are corner images, one with a digital format lens and one with a 35mm format lens. Plain and simple. The lens with all the finger prints is quite sharp in the middle. It softens in the extreme corners like virtually every lens ever made. So does the second one. It is just that the soft parts are cropped out.
Another good example of this is the shift lenses. These are wide angle lenses with larger image circles than other wide angles of similar focal length. When they are not shifted, they are sharp as a tack in the corners because the frame crops the edges of the image circle. You don't see the softness until you shift the lens over to the edge.
There you go. No book. Hope that helps.
I've always heard & read that lenses are at their best, sharpness-wise when stopped down a couple stops from wide open, and that as you stop down more, they suffer a bit in sharpness. Any experience for/against that?
A lot of what you said about image circles in the view cameras makes sense to me. But one thing didn't. You said that they have huge image circles, and that is why they are the sharpest & contrastiest on the planet. I don't think the lenses themselves are any better than the more common 35 mm lenses, it's just that since the negatives are so much bigger, the results are the sharpest on the planet. They are probably more expensive because of the amount of glass involved...
The reason they are sharper is that they have no moving parts aside from the diaphragm. No focusing rings, no zooming, no movement. It is easy for a lens designer to nail all the corrections when nothing moves. It's that simple. Projecting the larger image circle actually works against all this, making it harder to correct. Because they are better corrected and simpler, they do outperform 35mm camera lenses in lines per mm resolutions tests.
The reason they are contrastier is that the simpler designs have fewer elements and, hence, less flare. My point was that light scatter due to a large image circle doesn't appear to me to reduce contrast in view camera lenses. If it does, then that would surprise me. It is true that, as you make the camera movements so that the frame comes near the image circle edges, corner resolution does decline as you would expect. That's why LF studio shooters are known for liking to use f64.
Some MF lenses are made this way. A couple of examples are the lenses for the Mamiy RB/RZ 67 and the Rollei SL66. These cameras handle the focusing for the lenses so the lenses have no moving parts. Simple, stunning results.
If 35mm camera lenses didn't have moving parts, they would be even sharper because of the lesser amount of light refraction I referred to earlier. Some of the simpler ones are extremely sharp. A couple of examples would be the famous Nikkor 105mm f2.5 and the famous Leica 90mm f2.8 M. If I remember right these are either 4 or 5 element designs and both of them produce stunning images. Both were designed before the computer age, by the way. Simplicity really helps.
The newer technology that has gone into zoom lens design - things like aspherical elements and APO - have really improved zoom lenses in recent years. They haven't improved them to the point that they can compete with fixed focal length lenses, however. I'm sure they are working on it but simplicity still wins in performance.
Since I own the 12-24 Nikkor zoom I'll analyze this quote from the article:
"Nikon’s new 12–24mm f/4G ED DX-series Nikkor is a digital-only lens. At 24mm, it just edged out (pun intended) the lower-priced 24–85mm f/3.5–4.5G IF-ED Nikkor lens in terms of resolution and light falloff at the widest aperture settings (see chart). Results evened out at higher apertures, and there was little contrast difference between the two lenses at any aperture."
Here's the problem. It is easier to design a shorter lens with a smaller image circle than a longer one with a larger image circle. The reason is that the refracted light doesn't have to bend as far so it is easier to correct. When they compare a small image circle lens to a larger image circle lens of lesser quality, the smaller image circle lens is outperforming it slightly wide open. I think the difference would be more than slight if the edges of the image circle on the longer lens weren't cropped. I don't question their finding and it doesn't really relate to the point I was making. Take a 100mm digital image circle lens and compare it to a comparable 150mm 35mm image circle lens to product a similar angle of view and the longer lens will beat the shorter lens because the edges of the image circle are cropped out. Every time for lenses of similar design and manufactured quality. Same with wide angles if it were possible to get them together for a test. I have no problem with that part of the article.
The part of the article I really disagree with is their point that the wider image circle of 35mm camera lenses causes light to scatter around inside the camera body reducing contrast. I have a really hard time with this particularly since I spent years making images with view cameras. View camera lenses have gigantic image circles to provide for the camera movements. View camera lenses are the sharpest, contrastiest lenses on the planet. Their performance is simply breathtaking.
Consider, also, the 35mm shift lens. I used to have a 28mm PC Nikkor. It had a large image circle also to provide for the vertical or horizontal shifts. When this lens was centered (not shifted) it was sharper than any other wide angle lens I ever used on a Nikon in the corners of the frame. That's because the edges of the image circle were cropped out. Nikon has been famous over the years for making fairly lousy 28mm lenses. That's why Nikon shooting pros all used the 24mm which was very good. The 28mm PC Nikkor is the exception to that reputation.
This idea makes sense on the surface. Light can scatter inside a lens and reduce contrast (it is called flare.) Why couldn't it scatter inside a camera body? I suppose it could but I'm skeptical that it could cause a reduction in contrast like it does in a lens. The reason I'm skeptical is that I've never encountred it personally and my experience with view cameras and 35mm shift lenses certainly hasn't indicated a contrast problem of any sort. I'd need to see some tests to accept the premise.
In other words, I don't question the observations the testers made. I do question the explanation for the reasons behind them and have offered my own, which is a guess just like theirs. They blame it on light scatter. I blame it on lens design.
At this point I was going to explain why lenses get sharper as they are stopped down. I'll let it go. I'll just say that it is because it uses less of the lens surface. Basically it is similar to cropping out the edges of the image circle.
http://www.bhphotovideo.com/images/largeimages/394223.jpg
However, it is an FA lens and therefore optimised for the last generation of film cameras although that means it would work on your new Pentax dSLR as well as a backup K1000. On the bright side it is one of the FA Limiteds (high quality). It too is dang expensive though and almost entirely unavailable.
Canon makes a 35mm f/1.4L lens that sells for around $1300.
I don't the principle behind it, but 50mm lenses seem to be cheaper and small, and of course more in demand. Lenses such as 85mm or 35mm are bigger at the same apertures. Just using canon as an example, their 35mm f/1.4 is $1300, a 50mm f/1.2 is about the same price although the 50mm is faster
The better '35mm' lenses do look pretty good wide open...especially the high end lenses. My 'digital only' kit lens is much, much better when stopped down.
From what I've read...some lenses are just better than others when wide open...digital optimized or not. The big difference is that the newer lenses have more anti-reflective coatings, especially on the rear element...because the reflectivity of the sensor can cause a loss of contrast.
Just look at most lens tests...they often mention that sharpness, CA & distortion etc. is worse at the edges and corners. If your camera/sensor crops off the bad parts...then you are left with the candy centre...well maybe that's a tootsie pop...but you get the point. :D
FYI:
http://www.popphoto.com/cameralenses/849/digital-lenses-the-f-stops-here.html
The 35mm optimized lenses don't work very well when shot wide open. If you have to stop them down, you might as well just buy the slow digital-only lens.
But this 30 mm is equivalent to about 50mm in 35mm format, so it shouldn't have those problems.
You mean on digital? The sigma lenses also work on full-frame, so they need to address barrel distortion problems if you shoot with it on a film camera
Did you do any apples-to-apples comparisons, i.e. with the same equivalent focal length?
What about with the lenses stopped down as well as wide open? That second one is certainly very sharp. I can see the pulp in the paper.
Why didn't you clean all the finger prints off of the first lens before shooting? :lmao:
#If you have any other info about this subject , Please add it free.# |
Posted in enart.gcgx.com | edit
