[BigMackCam]

My problem is the images not being in focus, and the one in focus are not sharp. The three lenses are F1.4 or f1.8 and stoped down to f4 or more and still not sharp. I am also shooting at fast shutter speeds so it should freeze any movement. I just hate missing so many pictures due to these problems.

OK, this is very helpful.

Firstly, I strongly suggest calibrating the AF fine focus on each camera, with each lens. Your photos look great, but to get maximum sharpness in the right places, you really need to calibrate the fine AF. It can make a *huge* difference shooting at f/4.5 with a close subject. As I mentioned previously, there are articles on how to do this. Let me know if you can't find them and I'll have a look myself.

Next... where are you focusing? On the closest eye? This is typically - not always, but typically - the right place to focus for a portrait, so if you're not focusing there already, that's where you should aim for (you can always focus there and move to recompose if necessary).

To the photos...

1st photo - looks really good. shutter speed looks good at 1/320, maybe slight back-focus (depending on what you focused on, and lens AF calibration)

2nd photo - looks pretty good, but hands blurred. Faster shutter speed needed, perhaps, to capture any motion; and stop down further if hands should be more in focus too

3rd photo - looks pretty good, but some minor blur in face. Again, shutter speed is kind of low here if you want to avoid blur due to movement of the subject. Also, focus looks a little in front of the face (lens AF calibration or focus point issue?)

4th photo - lovely. Nice and sharp - my guess is focus is on the nose? I note that shutter speed is a little faster here - 1/200s - maybe there's something in this shutter speed thing?

I won't go any further right now... I stick with my original suggestion of ensuring your AF is calibrated for each camera and lens combo first. Then, I would suggest shooting at no less than, say, 1/320 (or even faster) - for a while. See what results you get. 1/100s isn't a fast shutter if you're capturing movement (even very slight movement from a typically fidgety child)

---

Last edited by BigMackCam; 02-28-2016 at 05:22 PM.

[kjphilippona]

Thanks very much, I did purchase the Spyderlenscal and thought I did calibrate my cameras to all my lenses. The instructions provided and on line are not very good and must have not done it correctly. I also love the last photo

[normhead]

There's been a couple of comments about high ISO showing that mades wonder. SO i went through my files, a have 1600 high ISO images of 16600 files. SO about 10% of those are 1000 ISO to 3200 ISO. I have files that are useful to 3200 ISO on both K-3 and K5 but as a general rule the K-5 is good to 1600 ISO, the K-3 is good to 1000 ISO. Other peoples standards might vary, but I would expect both values to change.

The K-3 gives me noticeably better resolution at 100-400 ISO.


But it's still functional in low light up to 1600 ISO.







I have a few clean images taken at 3200 ISO.

Normally about 1000 ISO is my high end. 1000 ISO is dependably clean.

I have clean K-5 images taken at 6400 ISO, some at 3200 ISO

With the K-5 about 1600 ISO is the safe high end, and is dependably clean.

So for me the whole thing is, are you willing to trade a stop of high ISO performance, for 20% more resolution. There is no free lunch. If, like me, you manage most of you images between 100 and 800 ISO, the K-3 is a superior camera in every way. If you're spending a lot of time 1600-6400 ISO, the K-5 will make a difference and have better resolution. Over all, the K-3 is a much more enjoyable camera to use. But that's how it breaks down for me. But, if you don't like your pictures with one, you probably won't like them with the other. I get excellent results K-5 , K-01, or K-3. I do use the K-3 much more than the K-5, and Apple Aperture does a great job on K-3 noise, that maybe explains why I don't find it as bad as some do. SO depending on the software you use, your results may vary,

You can get better resolution with a K-3 than you can with a K-5 100-1000 ISO. From 1600-6400 you get noticeably better images with a K-5, most of the time. In my work, I almost always go low ISO hi res.

[TropicalMonkey]

I am wondering if you focus on toddler's eyes using the center focus point and then recompose before actually taking pictures. Whenever I do that, it results in loss of sharpness wide open (since focus plane move farther back than intended).

[audiobomber]

All this talk about the K-5 series having lower noise than a K-3 is misguided. There are two reasons for this misconception.

1. The K-5 series has non-optional noise reduction at ISO 1600 and above
2. The K-3 series has more detail at every ISO. Noise reduction scrubs details to reduce noise

The answer is clear; the K-3 needs heavier noise reduction to equal the K-5 family noise performance, but the fact is, it does. Check the Pentax Forums In-depth review or DXOMark for confirmation.

[BigMackCam]

All this talk about the K-5 series having lower noise than a K-3 is misguided. There are two reasons for this misconception.

1. The K-5 series has non-optional noise reduction at ISO 1600 and above
2. The K-3 series has more detail at every ISO. Noise reduction scrubs details to reduce noise

The answer is clear; the K-3 needs heavier noise reduction to equal the K-5 family noise performance, but the fact is, it does. Check the Pentax Forums In-depth review or DXOMark for confirmation.

The problem here is that many cameras apply noise reduction to RAW images, and that is just as much a part of the image capture system as the sensor. How are we to judge the noise performance of a camera other than what we see in the RAW files? I don't disagree that the K3 captures more detail than the K5 at every ISO. That can be - but isn't always - important. But the fact remains, when I take the same photo, at the same high ISO, with my K5 and K3 and view them side by side at 100% magnification in LR6, the K3 files look a little noisier - not objectionably so, and LR's noise reduction is more than capable of dealing with noise from both, but the difference is there. If I resize the K3 files to more closely match those from the K5's 16Mp sensor, that has some beneficial effect on the perceived noise due to the down-sampling, but at 1:1 viewing the noise apparent in my high ISO K3 images is stronger than in my K5 images.

---------- Post added 02-29-2016 at 08:55 AM ----------

Thanks very much, I did purchase the Spyderlenscal and thought I did calibrate my cameras to all my lenses. The instructions provided and on line are not very good and must have not done it correctly. I also love the last photo

No problem I use the Spyderlenscal also. If you're unsure about the procedure and need help, PM me.

[audiobomber]

If I resize the K3 files to more closely match those from the K5's 16Mp sensor, that has some beneficial effect on the perceived noise due to the down-sampling, but at 1:1 viewing the noise apparent in my high ISO K3 images is stronger than in my K5 images.

Good point. That's the third reason people think the K-5 bodies have lower noise. They view the files at 100% magnification, which is not a fair comparison. The only way to compare noise is at the same pixel size, as shown in the Print screens at DXOMark or DPR:
Studio shot comparison: Digital Photography Review

[BigMackCam]

Good point. That's the third reason people think the K-5 bodies have lower noise. They view the files at 100% magnification, which is not a fair comparison. The only way to compare noise is at the same pixel size, as shown in the Print screens at DXOMark or DPR:
Studio shot comparison: Digital Photography Review

Dan - you're going to have to help me out here... apologies if I'm being dumb (which is more than likely )...

If I take exactly the same shot with both my K5 and K3 at the same ISO setting, bring them into Photoshop via ACR or Lightroom, and size them to equal physical dimensions, I'm going to see more noise with my K3. The size of the pixel doesn't really matter at this stage (or does it?) - what matters is that, at the same viewing size, the K3 image looks noisier than the K5 image...

[audiobomber]

at the same viewing size, the K3 image looks noisier than the K5 image...

And the K-3 image has more detail. Add NR to the K-3 image to reduce both the resolution and noise. (The K-5 raw image has already had noise reduction added by Pentax).

[BigMackCam]

And the K-3 image has more detail. Add NR to the K-3 image to reduce both the resolution and noise. (The K-5 raw image has already had noise reduction added by Pentax).

OK - got it now!

So... is there no noise reduction at all on K-3 RAW images? Or is it merely less aggressive than on the K-5?

[kjphilippona]

The problem here is that many cameras apply noise reduction to RAW images, and that is just as much a part of the image capture system as the sensor. How are we to judge the noise performance of a camera other than what we see in the RAW files? I don't disagree that the K3 captures more detail than the K5 at every ISO. That can be - but isn't always - important. But the fact remains, when I take the same photo, at the same high ISO, with my K5 and K3 and view them side by side at 100% magnification in LR6, the K3 files look a little noisier - not objectionably so, and LR's noise reduction is more than capable of dealing with noise from both, but the difference is there. If I resize the K3 files to more closely match those from the K5's 16Mp sensor, that has some beneficial effect on the perceived noise due to the down-sampling, but at 1:1 viewing the noise apparent in my high ISO K3 images is stronger than in my K5 images.

---------- Post added 02-29-2016 at 08:55 AM ----------



No problem I use the Spyderlenscal also. If you're unsure about the procedure and need help, PM me.

Thanks, i will do that tonight!

[Oldbayrunner]

OK my 2 cents... A key element everyone seems to be over looking here is what was the Distance the OP's photos were taken at? All of the lens adjustment in the world isn't going to correct photos not taken within the correct depth of field for the aperture and distance used. Not knowing that lets go through a presumption of apprx 10ft. The first photo shows a very narrow frontal depth of field so at F4.5 with a 55mm lens the depth of field @ 10ft, near acceptance is 9.18ft, the feet are sharp so they appear within acceptable sharp distance. The far limit is 11 feet so of course the back ground is sufficiently blurred. That leaves a depth of field of 1.79ft, which is still pretty narrow. The subject is sitting down adding farther distance from the feet, I presume the focal point was on the face so sitting down and somewhat moving as normal children do I suspect she wasn't within that narrow depth of field rendering her slightly out of focus. Now some fine folk on here would look at this photo and see the sharpness at the feet and suggest this lens is front focusing not taking into consideration the depth of field. Don't forget on these wide aperture lenses the depth of field at close range remains very narrow from F1.2 to F5.6 so controlling your depth of field is critical.

So for the OP run these photos on a depth of field calculator using each lens with the EXIF apertures at what you remember your distance taking into consideration your shooting a moving target that can move in and out of a narrow depth of field and I think you may find some insight to why these may not be as sharp as you'd like. Use larger apertures instead of trying to get that every so thought of creamy Bokeh and I believe you will encounter sharper images. if you want to use wider apertures you need to control the depth of field to achieve the look and sharpness you want.

PS... Any of these I could run in photos in my Photoshop and with a slight bit of smart sharpen or unsharp mask render these photos exceptionally sharp. I personally catagorize the

[kjphilippona]

I am wondering if you focus on toddler's eyes using the center focus point and then recompose before actually taking pictures. Whenever I do that, it results in loss of sharpness wide open (since focus plane move farther back than intended).

I use back button focus and face recognition when shooting portrait pictures.

[BigMackCam]

OK my 2 cents... A key element everyone seems to be over looking here is what was the Distance these were taken at? All of the lens adjustment in the world isn't going to correct photos not taken within the correct depth of field for the aperture and distance used. Not knowing that lets go through a presumption of apprx 10ft. The first photo shows a very narrow frontal depth of field so at F4.5 with a 55mm lens the depth of field @ 10ft, near acceptance is 9.18ft, the feet are sharp so they appear within acceptable sharp distance. The far limit is 11 feet so of course the back ground is sufficiently blurred. That leaves a depth of field of 1.79ft, which is still pretty narrow. The subject is sitting down adding farther distance from the feet, I presume the focal point was on the face so sitting down and somewhat moving as normal children do I suspect she wasn't within that narrow depth of field rendering her slightly out of focus. Now some fine folk on here would look at this photo and see the sharpness at the feet and suggest this lens is front focusing not taking into consideration the depth of field. Don't forget on these wide aperture lenses the depth of field at close range remains very narrow from F1.2 to F5.6 so controlling your depth of field is critical.

So for the OP run these photos on a depth of field calculator using each lens with the EXIF apertures at what you remember your distance taking into consideration your shooting a moving target that can move in and out of a narrow depth of field and I think you may find some insight to why these may not be as sharp as you'd like. Use larger apertures instead of trying to get that every so thought of creamy Bokeh and I believe you will encounter sharper images. if you want to use wider apertures you need to control the depth of field to achieve the look and sharpness you want.

PS... Any of these I could run in photos in my Photoshop and with a slight bit of smart sharpen or unsharp mask render these photos exceptionally sharp. I personally catagorize the

Agreed - but before taking depth of field into account, it's important that the point being focused on is accurately in focus (hence lens calibration); especially if there's a back-focusing situation, as that will render the acceptably-sharp area in front of the focus point (presumably the eyes) even shallower. And if shutter speed is too slow, all the depth of field calculations in the world won't stop motion blur. That said, depth of field is of course important in getting enough of the subject in focus, both in front of and behind the immediate focus point.

[Oldbayrunner]

Agreed - but before taking depth of field into account, it's important that the point being focused on is accurately in focus (hence lens calibration); especially if there's a back-focusing situation, as that will render the acceptably-sharp area in front of the focus point (presumably the eyes) even shallower. And if shutter speed is too slow, all the depth of field calculations in the world won't stop motion blur. That said, depth of field is of course important in getting enough of the subject in focus, both in front of and behind the immediate focus point.

True but looking at the exif in each case the Shutter speed was at a sufficient setting to stop motion so I took that into account when writing my post and mention only the aperture and distance for further consideration.