[Spock]

I have a FA*28-70 F2.8 in near mint condition.

It's a great lens although I've always known it was a bit soft wide open - especially at 70mm.
Recently I was taking some indoor pics in low light using my K3, and at 1600 ISO the wide open shots were very disappointing - almost unusable.

Here are some test shot examples to illustrate the problem:

1/80 sec. f/2.8 @ 70mm 1600 ISO - note the softness and 'glow' in this image



1/25 sec. f/5.6 @ 70mm 1600 ISO - two stops smaller and a bit improvement in image quality.



But this got me asking what's the point of a fast lens if I have to use it stopped down? I started thinking about replacing my FA* with something newer.

Before making any rash decisions, I tried taking some test shots in bright daylight at 200 ISO and I'm pleased to say the difference between F2.8 and smaller apertures is much less in bright light with low ISO, so that the wide open shots are quite acceptable.

Now I'm confused. Why would this be the case?

---

Last edited by Spock; 09-01-2020 at 02:11 AM.

[Serkevan]

That sounds like it could be a combination of the AF being a tad less accurate in low light (which is more noticeable wide open, of course) with 1/80 being one of the shutter speeds that can more easily be affected by mirror shock IIRC. The ISO setting should not be causing undue glow or aperture-dependant softness...

Can you try tripod shooting with 2-second delay, as well as manual focusing in magnified LV?

(I'm assuming you shot handheld, I can't see any EXIF from Postimage, at least on the work computer)

[Spock]

Yes, handheld. I will certainly use a tripod and do some more serious tests.

[photoptimist]

First, the optical behavior of a lens does not change with light level or ISO. A lens has no idea how much light is passing through it. Only the aperture changes the amount of softness or glow caused by the lens.

Second, lighting can make subjects look soft even with a sharp lens. The lighting in that picture looks pretty diffuse and the white-ish walls would turn the entire place in to a giant softbox.

Third, is there any chance that the lens was fogged by humidity? That picture looks like it is some sort of gymnasium or sports facility. If the lens had been in an air-conditioned office or car and was brought into a hot/humid room, the front element would fog as might the internal elements if you zoom the lens.

[pschlute]

Looks underexposed too, which will add to noise

[angelo9978]

I've had that lens and my copy was sharp wide open. If you don't mind my two cents, try shooting on a tripod, use a minimum shutter speed of at least 1/125 f2.8, and manual focus using live view to get critical focus on whatever you are focusing on, all at different focal lengths. if the image is still soft using the above then chances are then you have a bad copy. if the results are sharp, maybe your autofocus is off and you need to try the fine adjustment.

Heres a photo I took with that lens on a k1 @ f2.8 @28mm, skin is pretty sharp

---------- Post added 09-01-2020 at 05:25 PM ----------

actually I just took the time to look at your photo and there is plenty of sharpness there. The teeth look sharp and I can see the small wire on right side of picture twisting, so that's plenty sharp. I agree with Serkevan. I would also try using iso 100-400, shoot at a lamp if you can't wait until daylight or don't have flash. There is something making your pic look hazy...not sure if its fog like photoptimist said or a setting you have on, or real haze on the lens? try shining flashlight through lens to check all the elements?

That sounds like it could be a combination of the AF being a tad less accurate in low light (which is more noticeable wide open, of course) with 1/80 being one of the shutter speeds that can more easily be affected by mirror shock IIRC. The ISO setting should not be causing undue glow or aperture-dependant softness...

Can you try tripod shooting with 2-second delay, as well as manual focusing in magnified LV?

(I'm assuming you shot handheld, I can't see any EXIF from Postimage, at least on the work computer)

[Spock]

Thanks everyone for the replies. The photo was taken in a fitness centre (foyer) under fluoro lights but there was no humidity and my lens has no haze. If it did why would the f5.6 pic be so sharp compared to the f2.8? I'm thinking more that it could be a focus problem. I doubt it would be a handheld shake issue given the picture taken at 1/25 is sharper than the one taken at 1/80.

[Spock]

I took some pics just now - again at 70mm at both f2.8 and f5.6 at iso100. This time, notwithstanding a slightly softer image wide open, both images are sharp.
So I'm none the wiser as to why my promotional pics for my gym (as illustrated by the the skeleton photos above) varied so much in IQ from stopped down to wide open.

70mm @ F2.8 1/250



70mm @ F5.6 1/60

[yucafrita]

I think the DOF is pretty narrow. even with 2.8 (which is not 1.4). So maybe you just ever so slightly moved out of focus with the camera?

[mikesbike]

I have a FA*28-70 F2.8 in near mint condition.

It's a great lens although I've always known it was a bit soft wide open - especially at 70mm.
Recently I was taking some indoor pics in low light using my K3, and at 1600 ISO the wide open shots were very disappointing - almost unusable.

Here are some test shot examples to illustrate the problem:

1/80 sec. f/2.8 @ 70mm 1600 ISO - note the softness and 'glow' in this image



1/25 sec. f/5.6 @ 70mm 1600 ISO - two stops smaller and a bit improvement in image quality.



But this got me asking what's the point of a fast lens if I have to use it stopped down? I started thinking about replacing my FA* with something newer.

Before making any rash decisions, I tried taking some test shots in bright daylight at 200 ISO and I'm pleased to say the difference between F2.8 and smaller apertures is much less in bright light with low ISO, so that the wide open shots are quite acceptable.

Now I'm confused. Why would this be the case?

I have experienced some similar though not exact situations. I think the answer in this case is probably a simple one, and not at all a lens defect. First of all, you are shooting with a first-class fast zoom lens, but any zoom lens and most fast prime lenses will not be near their best performance wide open- this is not a defect. Secondly, combine (important) that with the fact you are shooting with the K-3, a very fine DSLR but not the best out there for higher ISO and likely to lose edge sharpness and fine detail at higher ISO settings and perhaps contrast as well, compared to the best models in this regard. You have your own proof of how this is probably the cause here, in that your results wide open with the same lens and camera at ISO 200 provided far better results.

If you can access a KP, try the same shot at ISO 1600 with that camera and your FA* 28-70mm lens. I would bet you'd see a difference. You might well have that or a K-new in your future!

One experience I had was a bit different. Yet it did show how a combination of factors together ganged up for a noticeable degradation of quality. I shot a river scene at night and decided to give a try with my fast 24mm prime, the Sigma 24mmm f/1.8 EX DG on the KP at ISO 3200. The shot included a bridge with supporting structures. In the past, this FF lens on APS-C has supplied "fair" results wide open, since the outer edges of the glass are not in use, and quite decent just stopping down a little to f/2.2- but this was with lower ISO and with less-able cameras than the KP.

With the KP, noise was easily low enough, but so was sharpness! Especially in the detail of the solid supporting structures, which has some lighting on them. Good enough to be usable, but only for viewing on a smaller print or screen. I also had with me the DA 40mm Limited and tried both lenses at f/2.8- wide open for the LTD, which blew away the Sigma for sharpness, even though the Sigma was just slightly improved over the f/2.2 outcome. Not too surprising. But there was less difference between them upon lowering the ISO to 1600!! However, the LTD was still better at ISO 3200 than the Sigma at ISO 1600. The lesson was twofold- Even with the KP, which is great for higher ISO, there is still some degree of disadvantage going up in ISO beyond a certain point. There was far less difference between ISO 1600 and ISO 800. Yet, there was less difference between ISO 1600 and ISO 3200 when using this sharp LTD lens! One would think the sharper lens would reveal more of the negative effects of the higher ISO, but this was not the case. So it was the 2 factors working together- less capable lens + higher ISO disadvantage- even though slight, teaming up to make the difference greater than slight.

[Spock]

Thanks again for your comments. Yes I think a large part of the problem is the less than stellar high iso performance of the K3. I may have to reconsider using iso higher than 800. (But then of course low shutter speeds will cause their own problems). Also I agree that focus becomes quite critical at f2.8.

I'm disappointed that I can't get good images in low (artificial) light conditions, but relieved there doesn't seem to be an issue with my FA*28-70 at least.

[leekil]

Could the SR not be completely settled by the time the first photo was taken? The EXIF says that the SR was on, but I thought there was a second EXIF value that indicated if it was completely settled or not -- or maybe that was only on older Pentaxes?

---------- Post added 09-01-20 at 10:38 PM ----------

Looking at it more closely, I think that it is a SR issue. It is not an ISO difference.

I think it is an SR issue, as I often take candid photos, and sometimes I take two because it seems like I get cases where there is some blurriness due to camera movement. If I take two, the second is usually sharp if the first one isn't. This seems like the same type of thing, but I could only be completely sure if the second one had been taken in quick succession, which wasn't the case here.

[Spock]

Could the SR not be completely settled by the time the first photo was taken? The EXIF says that the SR was on, but I thought there was a second EXIF value that indicated if it was completely settled or not -- or maybe that was only on older Pentaxes?

---------- Post added 09-01-20 at 10:38 PM ----------

Looking at it more closely, I think that it is a SR issue. It is not an ISO difference.

I think it is an SR issue, as I often take candid photos, and sometimes I take two because it seems like I get cases where there is some blurriness due to camera movement. But if I take two, the second is usually sharp, if the first one isn't. This seems like the same type of thing, but I could only be completely sure if the second one had been taken in quick succession, which wasn't the case here.

There was a gap between the photos while I changed the aperture so I doubt it is an SR issue. Besides, I had the issue with many photos; those two skeleton photos I took to confirm my suspicion that there was a problem wide open images.

[yucafrita]

Here's an old thread about the SR: SR Icon EXIF - PentaxForums.com

It might be that the SR did not already settled (completely).

[leekil]

There was a gap between the photos while I changed the aperture so I doubt it is an SR issue. Besides, I had the issue with many photos; those two skeleton photos I took to confirm my suspicion that there was a problem wide open images.

I'm not suggesting that the second photo was taken soon after the first -- I already knew it wasn't. I am just suggesting that the SR hadn't completely taken effect when you took the first photo. I was describing my practice of taking two photos because often the SR hasn't settled when I take the candid shot.

It could be the 2.8 aperture, but I don't think it is the ISO performance of the K-3.