[tennjed]

Boy, I sure wasted time and anxiety for the last couple of days.

I recently decided on buying the Pentax 50-200 WR lens and noticed that the images produced by the K3 appeared noticeably softer than the images produced by the k5IIs- using the same lens. So I have spent the last couple of days fooling around with the AF fine tune on the K3 with no other result than going from bad to worse. I was on the verge of sending the K3 back to Pentax.

Until this evening, I was reviewing images on the camera LCD and on my MacBook (Non Retina) laptop. A short time ago I loaded the K3 images on my Ipad (Retina Display); lo and behold, the same images appear about as sharp as you could expect from a bargain priced lens.....On par with the images from the K5IIs using the same lens. Maybe a bit better.

After my initial relief, I began to wonder why the images from the K5IIs were OK on the lower resolution monitors, but the K3 images were not. Is the resolution capacity of the K3 so much greater than the K5IIs that this sort of thing is to be expected? Can a higher resolution monitor mask a soft focus problem with a camera? I am kind of confused by this. Does the fact that the K3 images, when viewed on the non Retina MacBook, are noticeably softer than the K5IIs images indicate that I might well have an AF problem with the K3?

[Rimfiredude]

I can show you why this happens but put it in words is another story. Sorry

[stevebrot]

Can a higher resolution monitor mask a soft focus problem with a camera?

A pixel is a pixel is a pixel. Think of it in those terms.

Now consider the number of pixels per running inch on your retina display vs. the number of pixels per running inch on a less capable monitor. Perception of sharpness is related to magnification and viewing distance. The finer the dot pitch the smaller an image of defined pixel dimensions will appear for a given viewing distance (fewer degrees of arc) and the sharper that image will appear. If normalized to the same degrees of arc, both the retina and a less capable display will appear equally sharp.


Steve

[tennjed]

Thanks Steve. So the fact that the K3 images appear noticeably softer on the MacBook (non retina) does indicate that the K3 has an AF problem?

[osv]

Thanks Steve. So the fact that the K3 images appear noticeably softer on the MacBook (non retina) does indicate that the K3 has an AF problem?

you need to be looking at the images at the full 100% size to determine focus and sharpness... did you do that?

[nomadkng]

Thanks Steve. So the fact that the K3 images appear noticeably softer on the MacBook (non retina) does indicate that the K3 has an AF problem?

If normalized to the same degrees of arc, both the retina and a less capable display will appear equally sharp.

^^^^That is the most important statement.

There's a lot of math involved and I may not have it all down, but it starts with the 240 pixels per inch of both the K5 and K3 native sensor image.
You then have to equate the viewed image to the pixel density of the monitor, it might be something like 90 for your non retinal display and 130 for the retinal display.
There's a formula out there for monitor size and pixel density etc.

So basically until you can look at your monitor replicating 240ppi based on your MONITORS ppi you aren't comparing images equally.
At this risk of being really wrong, I feel it's akin to JPEG compression because at 90ppi you are seeing only 1 out of every 3 (rounded up) pixels of the native 240ppi image.
What this means is that your monitor at say 90ppi is using hardware to extrapolate an image that's at 240ppi. Some monitors do this better than others.
I think this means 240ppi images at 185% on a 4k monitor (130ppi) = 240ppi images at 267% on a standard monitor (90ppi)

What does that mean for your 50-200? I'm not sure because you are talking about MacBooks versus iPads and different resolutions. Pixel density is a factor of resolution AND diagonal so the different platforms are apples to oranges.

I also don't know how many images you took, at what shutter speeds and what apertures and iso, but if I had to make a guess, I'd say the 50-200 is not playing well with the K3. The K3 is very unforgiving of lenses and technique (you can find 3 dozen "K3 sucks..oh wait it was me" threads on PF). I had 3 lenses that played very well with a K5iis, a K30 AND a K-01, but they became unusable when put on a K3, no matter how much I tried to tweak them. I think the K3 may be exposing flaws in the 50-200 or your technique, or both.

[stevebrot]

So the fact that the K3 images appear noticeably softer on the MacBook (non retina) does indicate that the K3 has an AF problem?

No. It tells you nothing other than the two displays are different.


Steve

---

Last edited by stevebrot; 04-14-2015 at 01:04 PM.

[stevebrot]

240 pixels per inch of both the K5 and K3 native sensor image

Not quite:

Pentax K-3 horizontal resolution = 6539 pixels/inch
Pentax K-5 horizontal resolution = 5356 pixels/inch

For a quick comparison, the OP could simply up-sample a real good image from the 50-200 from the K-5 to the K-3 resolution of 6061x4000 using their favorite image manipulation software. I think they will find that the up-sampled image will contain the same or less detail as a native K-3 capture of the same subject. View them both on the same monitor at full resolution.


Steve

[stevebrot]

BTW...I would suggest that you benchmark the performance of your DA 50-200 on the K-3 before fiddling with the AF fine tuning. Do this with the camera on tripod using a flat subject with lots of detail (e.g. cereal box). Focus manually using magnified focus peaking in live view with the lens at 100mm focal length at a distance of 2 meters (78") from the subject. Use the 2 second delay to disable SR and minimize vibration from mirror slap.

That should give you a good idea as to the best performance you can expect from the lens. Having done so, please consider the following bullet points regarding shooting with the K-3. More pixels means:

• Greater attention must be paid to minimizing camera motion. This was noticed early on by users on this site when the K-3 came out. This is not a failing of the SR feature. Nikon users with VR lenses reported similar problems on the D7100.
• While the K-3 will not make any lens in your bag "worse", it may take a lens to its performance limits with the result that the image may appear softer at full resolution than a full resolution crop from the K-5 with the same lens. The actual detail captured will be the same or greater, but it doesn't appear that way. If you must compare, do so with an up-sampled K-5 image.

Steve

[tennjed]

Thanks for all of the help with this. I guess the question I should have asked is the one that Nomadking and Stevebrot have answered: Does the K3 sensor require greater resolution from lenses used? The fact is, I may have been giving the 50-200 more credit for resolving capacity than it deserves. I understand- and believe- it is a great lens, but it is obviously has limits when compared to higher priced primes and zooms.

Today I did something that I all to infrequently bother doing: I took the tripod out of the corner and shot the K3/50-200 combination on the tripod. Results are markedly better. Just for kicks, I attached the 43 1.9- my only other AF lens- and shot for awhile with it. The focus performance with that lens seems to be good.

Thanks again. I guess now all I need to do is figure out a way to use my Ipad as a monitor for the MacBook Pro.