[haroon88]

Hello dear Pentaxians,

so previously I reported the unhappiness on the use of Sigma lenses on my K50. I use Sigma 17-50 f2.8 and Sigma 30mm f1.4 (not the Art one). Both have some problems for which I am thinking to do some changes. Let me summarise that the 17-50 at wide angles have problems of sharpness at edges of the photo, center is sharp as I want. Also the edges show some sort of distortions but this thread is particularly dedicated to the Sigma 30mm f1.4.

I went out for some low light street photography of Melbourne's street using my K50 and the prime lens. I am really really disappointed by doing the photography at f1.8. OK I know that f2.8 is the sweet spot for more sharp photos but what's the point of having f1.4 then when f1.8 can't do the work. I had used (€30) Pentax M 50mm f1.7 and it was extremely better in terms of sharpness than (€250) the Sigma 30mm f1.4. Now through expert opinions, I can perform some definite actions to resolve the problem.

So having tried multiple times to calibrate the lens, I am still not convinced of having paid €250 euros for this beautiful yet useless lens on K50. What should I do. Let me show some results I got recently. Some are at f1.8 and some at f2.8 handheld. I am pretty much familiar about using tripods for razor sharp photos and I am not looking for them here so kindly no suggestions of using tripods. The point I want to discuss is the sharpness possible by handheld photography in which just a very few are acceptable but rest are pretty bad to look at even.

Should I sell my prime lens which I bought from Japan in immaculate condition or should I change the body?

---

Last edited by haroon88; 12-10-2019 at 05:09 AM.

[JohnX]

Have you calibrated the lens using the AF Fine Adjustment?

Have you tried shooting the same scenes using Live View?

Have you tried the same scenes with the camera on a tripod? (You say you don't want to discuss tripods, but are any shots better with a tripod? Please post them here)

Have you tried the same scenes doing any of the above with lower ISO?

You need to ensure you eliminate possible bad technique, non-optimum settings, etc. especially if shooting hand-held.

[Kozlok]

I think you have multiple issues. I think it's a combination of technique, expectations and limitations of any lens.

Technique, I think you have some motion blur in these shots, and some of them are shot at very high ISO. Either of these will make a shot look soft. Shooting at lower ISO and smaller (higher number) aperture will give you sharper images, and a tripod is generally needed to steady the camera at slow shutter speeds. I know you said you don't want to hear that, but the reason people suggest tripods is that they work. I'd guess you are shooting too fast and moving your body while shooting. You can try practicing slow-shutter-speed camera stabilization; there's an article here on PF written by Heie that addresses that technique. .

Limitations: at very larger apertures, the depth of field is small. Back focusing even slightly causes problems. Sigma lenses don't focus as reliably as genuine Pentax lenses. Try focusing on something 3/4ths of the way into the frame. See if you can get something sharp in the frame.

Expectations: The Sigma 30/1.4 is one of the sharpest lens you can get, but only in the very center of the frame. It's still not amazingly sharp until past f2. The corners of this lens will be soft at any aperture. If you can get something sharp in the center of the frame, that's the best this lens will ever do. Also, you won't ever get a sharp shot at ISO3200, they will always be a bit soft. For pixel-peeping sharpness, you need to be below 800 or so. A K-70 will improve that a bit, but it's not magic.

Edit: Thinking about this, I gather you don't own a tripod. Try this experiment, set you camera on a wall, use 2s delay, set ISO at 400 or so, focus just a little less than infinity and shoot a shot like any of these. See if you can find a band of sharpness in that shot. If so, then it's some combination of what I wrote. If that shot has no sharpness anywhere in the frame, then you might have a defective lens. Also, post full resolution samples on Flickr or something similar so we can better see what's going on.

[haroon88]

Have you calibrated the lens using the AF Fine Adjustment?

Have you tried shooting the same scenes using Live View?

Have you tried the same scenes with the camera on a tripod? (You say you don't want to discuss tripods, but are any shots better with a tripod? Please post them here)

Have you tried the same scenes doing any of the above with lower ISO?

You need to ensure you eliminate possible bad technique, non-optimum settings, etc. especially if shooting hand-held.

Yes I did many times AF Fine Adjustments to verify if my previous adjustments were wrong.
Using Live View required having tripod, I did had with my in my room but didn't bothered to take it with me since I wanted to do street photography with open liberty.
YES YES YES I know that tripod helps but let me ask why is this formula to avoid camera shake exists in which a shutter speed should be faster than "1/ (effective [full-frame] focal length)". Most of my photos I posted here are faster even that of 1/125 isn't that great either with ISO 1600.
Lower ISO would then make the shutter speed slow and hence obviously creating the camera shake issue. Better remind here that I do have SR functionality on all the time.

I think you have multiple issues. I think it's a combination of technique, expectations and limitations of any lens.

Technique, I think you have some motion blur in these shots, and some of them are shot at very high ISO. Either of these will make a shot look soft. Shooting at lower ISO and smaller (higher number) aperture will give you sharper images, and a tripod is generally needed to steady the camera at slow shutter speeds. I know you said you don't want to hear that, but the reason people suggest tripods is that they work. I'd guess you are shooting too fast and moving your body while shooting. You can try practicing slow-shutter-speed camera stabilization; there's an article here on PF written by Heie that addresses that technique. .

Limitations: at very larger apertures, the depth of field is small. Back focusing even slightly causes problems. Sigma lenses don't focus as reliably as genuine Pentax lenses. Try focusing on something 3/4ths of the way into the frame. See if you can get something sharp in the frame.

Expectations: The Sigma 30/1.4 is one of the sharpest lens you can get, but only in the very center of the frame. It's still not amazingly sharp until past f2. The corners of this lens will be soft at any aperture. If you can get something sharp in the center of the frame, that's the best this lens will ever do. Also, you won't ever get a sharp shot at ISO3200, they will always be a bit soft. For pixel-peeping sharpness, you need to be below 800 or so. A K-70 will improve that a bit, but it's not magic.

Edit: Thinking about this, I gather you don't own a tripod. Try this experiment, set you camera on a wall, use 2s delay, set ISO at 400 or so, focus just a little less than infinity and shoot a shot like any of these. See if you can find a band of sharpness in that shot. If so, then it's some combination of what I wrote. If that shot has no sharpness anywhere in the frame, then you might have a defective lens. Also, post full resolution samples on Flickr or something similar so we can better see what's going on.

Well regarding technique, let me clarify that I am a 2 years old photographer and have been using 3 lenses with 2 tripods and 1 camera body. I have learned over the years how to adjust the body, breathing technique and stable myself by getting a support of tree, electric light pole etc. Tripods do work, YES I KNOW, but the street photography shouldn't always be done with a tripod to get good photos (I am not talking about crisp photos). Just an example, look at the clocks in the photos except for the first one, I placed the center point of the camera to focus it and none of the times it ever managed to focus it good enough for me to smile. I never shoot too fast and never moved my body. I would however have a look on the slow-shutter-speed camera stabilisation.

Talk about limitations, I used large apertures to counter camera shake and wanted to test if it worked at 1.8, but when I tested at 2.8, it still never became better, check the last photo. Hence that sweet spot of 2.8 is fairly tested here and flawed, OK I will try tonight or tomorrow night with tripod and will come up with results to see if I always need tripod or not? I will try the 3/4th of the way into the frame.

Talking of expectations, when you say at the center of the frame, is it day light hand held photography? ISO3200 was used to increase shutter speed at f2.8 (again last photo) but softness shouldn't have misfired focus at least. I have learnt over the years that always a prime lens should give ideal results either handheld or tripod supported but I am evaluating here where the problem is existing

P.S. I have two tripods which can be stabilised by weight as well. In these days, I'll do a comprehensive testing using tripod and will vary the ISO/aperture/shutter speed to confirm our claims.

---

Last edited by haroon88; 08-19-2019 at 01:03 PM.

[JohnX]

The questions I raised are to help eliminate the various factors that might be causing the problem.You need to work through all the possibilities, then whatever is left is the answer!

[Serkevan]

I get the feeling that the lens is misfocusing. Several shots look pretty decent on the road in the foreground for 3200 ISO... Clearly the front row of cars is orders of magnitude better than the background building. Maybe the K-50 is having trouble with the AF in low light? Or were those shots manually focused? I'm asking because yesterday night I went shooting fireworks and had to tell the AF of my K-7 to pack up and go home, it refused to work properly.

On another note I also find it interesting that you get such clear sunstars from the light posts at 1.8... is that normal behaviour for the 30mm 1.4?

[TerryL]

To answer your question, I would say an upgrade to a KP would be nice.

[haroon88]

The questions I raised are to help eliminate the various factors that might be causing the problem.You need to work through all the possibilities, then whatever is left is the answer!

Thanks for your help.

I get the feeling that the lens is misfocusing. Several shots look pretty decent on the road in the foreground for 3200 ISO... Clearly the front row of cars is orders of magnitude better than the background building. Maybe the K-50 is having trouble with the AF in low light? Or were those shots manually focused? I'm asking because yesterday night I went shooting fireworks and had to tell the AF of my K-7 to pack up and go home, it refused to work properly.

On another note I also find it interesting that you get such clear sunstars from the light posts at 1.8... is that normal behaviour for the 30mm 1.4?

That is what I assume as well regarding misfocusing. Someone in this forum previously suggested a link where it is reported about misfocusing on K50 (it was the in depth review article on Sigma 30mm f1.4 on PF).
Yes, I also noticed the front row being better focused but I remember that always I had made the camera focus the building. In low light, Sigma 17-50 at f2.8 does work better even at high ISO. I would try to compare them these days in low light condition. These shots were focused by camera. I will try to get the sunstars from light posts tomorrow. I will try to replicate the situation which I had in Melbourne.

[Serkevan]

Yes, I also noticed the front row being better focused but I remember that always I had made the camera focus the building.

In my experience with the K-7, the camera is perfectly capable of ignoring a very obvious building completely . In low light it can be frustrating...

[ChristianRock]

In low light, Sigma 17-50 at f2.8 does work better even at high ISO.

I don't think this has been brought up before, but there is something else that might be in play here in addition to the things that have been brought up before (which include that the Sigma 30 non-art is always soft off center, wide apertures will give a low in-focus area, slow shutter speeds will need a better technique to avoid motion blur even with SR, and higher ISOs will give you softer images, and Sigma 30mm lenses all usually have some autofocus issues, except maybe the very old screw drive ones).

This comment that I just quoted reminded me of my own battle with the Pentax shutter shake issues. The K-50 has some of it, but not as bad as my K-S1. The K-S1 is pretty bad in regards to that. The K-50, which I also own, has some as well.

The issue has been well documented in the forums. The K-1 also has it. The camera I started out with, the K20D, did not have any. My K10D doesn't either, but pretty much every camera since, with the possible exception of the K-3, has had it.

In the K-50 and K-S1, I get it from shutter speeds of about 1/60s to about 1/200s. The worse range is about 1/100s to 1/125s, it's very visible there. So in your examples posted, that would include pictures #2, #3 and #5. Pictures #1 and #4 should be free of shutter shake and might have other issues (#1 has a pretty slow shutter speed compared to the others but actually sort of looks the sharpest to me!)

There are two things that help control shutter shake: using a flash on to camera (the added weight/mass reduces the shake), or use a heavier lens - especially one with more weight at the front. The Sigma 17-50 2.8 is a pretty heavy lens compared to the 30 1.4, and is quite front heavy, so naturally it will reduce the shutter shake quite a bit.

I first noticed this back when I still has a K-r (which has a bit of shutter shake but also not quite as bad as the K-S1, which is probably the worst in the Pentax range in that regard). I was in my basement taking shots with my Tokina 19-35mm (a fairly large full frame wide angle zoom made of lots of glass and metal) and my Pentax DA 35mm f/2.4 "Plastic Fantastic" with both at 35mm and f4.5 (which is the longest focal length of the Tokina, and wide open!) and I could not figure out why the Tokina was consistently pin sharp but the super-lightweight DA 35 2.4 wasn't! And I used that DA 35mm on the K20D a lot prior to this and it was always pin sharp from f2.4. This bothered me for days if not weeks until I found out about shutter shake. And now I know that the featherweight DA 35 2.4 is the worse lens when it comes to shutter shake, it adds no mass nor weight to help it at all.

I would suggest that you do another shoot at night like the one you just did, set the camera to something like 1/25s or 1/30s in Tv mode, and then set your ISO range to top at ISO 1600 or even 800. Then let the camera select aperture. You might end up with some f/2 shots, and some f/4 or f/5.6. Just remember to use good technique - have your elbows brought in to touch your body for stability when shooting. With this you would bypass the shutter shake and the higher ISO issues and see if there are any issues left that related to the lens, AF or the K-50.

[ehrwien]

Just a few ideas: How have you set the AF-S setting in Custom Setting 2 menu, Focus Priority or Release Priority? Have you been using AF-S with these shots? Spot focus point (just the central point)? Do you focus with the shutter button half pressed? How long do you keep it half pressed/ how long does it take from you beginning to press down the shutter button to it being fully pressed? Are you waiting for AF confirmation in the viewfinder halfway through?
Also, you said you were focussing on the clock? I think the clock is rather small / far away / overexposed / not very contrasty. Look for clearly defined, contrasty edges for the AF to catch on to, and remember that the focus point is in reality a bit larger than it is shown in the viewfinder, so it may choose to select something else in the vicinity of where you want it to focus. When it's very dark, the camera may have trouble finding the correct focus and hunt back and forth, sometimes not finding anything to focus on at all. Try a different spot to focus until the camera confidently chooses focus without much hunting.

All that said, I do have the Sigma 30/1.4 EX DC myself, but have only shot it once so far without fine tuning the focus before, and found it to miss at f/1.4 way more often than at f/2.0. I think I recognized this while chimping and continued to shoot at f/2.0 or f/2.8 afterwards, which greatly improved my keeper rate.

[mikesbike]

Two years experience is not that long, but better than just starting out. You will continue to make good strides forward.

As CristianRock brings up, especially the K-x, and the K-S1 have quite often been reported having "shutter shake", and/or more commonly mirror slap shake. I don't know about the K-50, but the K-r I owned was designed for an improvement in this regard, along with other improvements over the K-x, which largely succeeded. The K-S2 was an advancement in many ways. My K-5 and K-5 IIs have no such issue, in fact have the quietest shutter/mirror operation, I think even quieter than my KP.

And, yes, having the camera in contact with your hands and body helps to damp out such vibrations, more reliably so than using a tripod. Even using a larger, heavier lens helps. However, aside from this issue, very often softness from shake is due to hand-held shots being done at a shutter speed that is a bit low. That is also a variable from one person to another. I am seeing evidence of some of that in the above examples and also there appears to be some focus issues with related depth-of-field (DOF) issues from using very wide apertures. As Terry suggests, the KP has a newer and superior AF system, as well as having a new and better 5axis SR system for hand-held shots, and has greatly upgraded performance for using higher ISO settings in low lighting.

Using the center-only spot focus setting, AF-S, then recomposing to get your shot while holding that focus is the proper technique for accuracy, but be sure to "catch onto an edge" when focusing on a particular spot or area.

All that said, it is rather common for fast lenses, even prime lenses, to be less than sharp when wide open, to some degree or other. Your Pentax 50mm f/1.7 is one lens that is capable of good sharpness even wide open. But it is not f/1.4 and on APS-C it functions in the telephoto range so an image from it at the same scene would capture only a small section of that scene, not several blocks of it hoping to get edge-to-edge sharpness at a wide aperture. This might be possible if everything in the frame were at some distance. Why pay for a very fast f/1.4 if it is not sharp wide open? Sometimes it is the only way to get a higher shutter speed and at least a fair result, especially with a live subject. And the viewing through the VF will be brighter.

The shots you present are a poor test for edge-to-edge sharpness form a lens. Too much DOF issues to deal with, along with points of focus in the scene. To do an easy, much more reliable test, it must be in good lighting and lower ISO, while getting a good, high shutter speed. This with a solid, well-defined subject against a not-busy background providing good contrast against your subject. The subject should be some mid-distance away yet large enough in the frame for good visibility. Shoot wide open, center-only AF, placing your subject in the center of the frame. Use the half-press of the shutter release button to make sure you get good focus. Do the half-press more than once to make sure, then immediately carefully switch your camera to manual focus which will preserve your focus setting. Being careful not to touch the lens focus ring, then take more shots of your subject, moving the camera to place the subject very near the right edge of the frame, then in the right upper and lower corners. Do the same placing the subject on the left side. This test will also reveal a decentered lens, which will exhibit sharper results on one side vs. the other side.

Even with a very good lens, the centered subject will likely appear sharper than when it is at the edge, and with yet more difference when in the corners. But with a very good lens, the disparity will not be huge. Stopping down a little should sharpen up edges and corners accordingly.

[TER-OR]

What Mikebike suggests is the key, I think.
You note that on the last two you're trying to focus on the clock face, but it's the street right in front of you which is in focus. I'm sorry if I missed it, but are you running center focus point only, or some other mode where the camera can pick another focus point? That's particularly tricky at night when the contrasts are confusing.

[Benz3ne]

As above, and as you already know, from a novice's perspective it seems as though the focus is 'off' in the sense that it's picking up on foreground rather than more distant objects. To reiterate, it's worth checking whether there's spot focus activated or not, and to try again. Alternatively, if you're on a tripod and can see decently through the viewfinder, you could try focusing manually and see how that affects the images.
The suggestion of motion blur I think isn't much of an issue, as it's typically shot around 1/125 of a second, so only faster moving cars should really give any motion blur.
Your aperture is quite 'open' - between f/1.8-2.8 from what I can see. This would undoubtedly give you a shallower DoF, especially with the closer focus, which throws the background out of focus as you're witnessing. It might be worth looking up hyperfocal/hyperlocal distances to see where is approximately best to focus or playing with a DoF simulator. If in doubt, try aiming 2/3 of the way up the landscape and see where that takes you.
In these instances, I think a smaller aperture (e.g. f/5.6 or something?) + longer exposure combination would give you a better result, as a bit of motion blur wouldn't detract too much from a night-time scene, and can look quite nice with respect to car lights.

To answer the question in the title - I'd sell neither until I've tried out manual focusing, playing around more with aperture and exposure settings, ensuring that there's a 2s timer on button press before the photo is taken (to reduce movement during shutter press) to see what gives the best results with what I've got. Only then would I really be able to pinpoint what the issue is, whether it's down to the effective EV at which focus works accurately (which would be vastly improved with a K-70 or KP vs the K-50) or whether there's a definite issue with the lens, which would also show issues during daytime photography.

[haroon88]

Thanks to all my pentaxian friends for their very kind info/critiques/suggestions above. I would reply to particular questions below. Then a surprise problem and its easy solution is mentioned in the following post which for me is strange and funny.

Just a few ideas: How have you set the AF-S setting in Custom Setting 2 menu, Focus Priority or Release Priority? Have you been using AF-S with these shots? Spot focus point (just the central point)? Do you focus with the shutter button half pressed? How long do you keep it half pressed/ how long does it take from you beginning to press down the shutter button to it being fully pressed? Are you waiting for AF confirmation in the viewfinder halfway through?

Well to answer your kind questions, I'll reply in sequence to the questions: It is Focus priority always. Also AF-S always for focusing the shots. Spot focus as well in which then I can tell the camera to focus the object. I always focus with the shutter button half pressed and I keep it half pressed for at least one second and steadily continue to fully press it to take the shot. Yes I wait for the AF confirmation.

As above, and as you already know, from a novice's perspective it seems as though the focus is 'off' in the sense that it's picking up on foreground rather than more distant objects. To reiterate, it's worth checking whether there's spot focus activated or not, and to try again. Alternatively, if you're on a tripod and can see decently through the viewfinder, you could try focusing manually and see how that affects the images.
The suggestion of motion blur I think isn't much of an issue, as it's typically shot around 1/125 of a second, so only faster moving cars should really give any motion blur.
Your aperture is quite 'open' - between f/1.8-2.8 from what I can see. This would undoubtedly give you a shallower DoF, especially with the closer focus, which throws the background out of focus as you're witnessing. It might be worth looking up hyperfocal/hyperlocal distances to see where is approximately best to focus or playing with a DoF simulator. If in doubt, try aiming 2/3 of the way up the landscape and see where that takes you.
In these instances, I think a smaller aperture (e.g. f/5.6 or something?) + longer exposure combination would give you a better result, as a bit of motion blur wouldn't detract too much from a night-time scene, and can look quite nice with respect to car lights.

To answer the question in the title - I'd sell neither until I've tried out manual focusing, playing around more with aperture and exposure settings, ensuring that there's a 2s timer on button press before the photo is taken (to reduce movement during shutter press) to see what gives the best results with what I've got. Only then would I really be able to pinpoint what the issue is, whether it's down to the effective EV at which focus works accurately (which would be vastly improved with a K-70 or KP vs the K-50) or whether there's a definite issue with the lens, which would also show issues during daytime photography.

More or less according to your suggestion, I have done some testing and in the following post I am going to share the results.

Just to get myself as well clarified that if my handheld photography technique is flawed, I found some photos of my trip to Melbourne which were handheld, when I say handheld, I mean that no tripod is used. So in different ways (like placing the camera and pressing the shutter button without timer etc) I have shot some photos which I am sharing here. They did remove the doubt that I was having that probably my handheld technique has to be corrected or not!