[Haakan]

Just to give some examples of Mark's point, here is a link to an exhibition by benjikan where he showed images of about 120x180 inches from the K20D.

https://www.pentaxforums.com/forums/pentax-news-rumors/38006-new-exhibit.html

Best regards,
Haakan

[ricardobeat]

While of course it is true you can't get 300ppi at 16x20, that doesn't make it useless. The general rule of thumb is that the resolution (in terms of pixels per inch) one needs for large prints is less than what one needs for small prints, simply because large prints tend to be viewed from farther away. Sure, we all want all the resolution we can get, but just because it isn't 300dpi doesn't mean it's going to look terrible. Most commercially printed posters are probably far less.

Exactly. Billboard prints routinely go as low as 20dpi, and even lower for facade covers.

[tlwyse]

While of course it is true you can't get 300ppi at 16x20, that doesn't make it useless. The general rule of thumb is that the resolution (in terms of pixels per inch) one needs for large prints is less than what one needs for small prints, simply because large prints tend to be viewed from farther away. Sure, we all want all the resolution we can get, but just because it isn't 300dpi doesn't mean it's going to look terrible. Most commercially printed posters are probably far less.

Nobody said "useless". I'm merely stating that for *photo quality* prints the image needs to be something like 1/2 to 1/3 the output resolution of the printer (by the way, there's some folks that say it needs to be AT the printer's resolution of 600ppi (HP) or 720ppi (Epson) so I think stating a minimum of 1/2 to 1/3 printer output resolution is not being unreasonable).

The way I define "photo quality" is sort of a "no excuses" level of sharpness and detail when viewed with the naked eye (not a loupe)....any MORE resolution and there's little if anything gained...any LESS and you can start to see the degradation on close inspection. I know in my own tests, I didn't feel anything was gained at image resolutions higher than about 240ppi for my Epsons but going lower than that was definitely visible to the naked eye. For the AVERAGE viewer (not a photographer). possibly 1/4 to 1/3 of printer resolution would be acceptable. For a 6mp camera, that is still no more than about an 11x14, give or take....so if you're going to take 6mp to 16x20, don't expect something that's going to knock your socks off on close inspection....unless you're happy viewing the print from several feet away.

What some have described is what I would call "poster quality" or even "billboard quality"...that's a completely different matter. For an Epson, I would say "poster quality" is in the 120-180ppi range...sharpness degradation definitely visible at close range (hold-it-in-your hand range) but might be fine at a distance appropriate for the print size. And I've seen "decent" posters (24x30 or so) done as low as 90ppi...but make no mistake, it is not SHARP when you walk up to it, just merely acceptable for the average person viewing it.

I could go off on the "megapixel" race as well...personally, I had a hard time justifying going from my K10D to the K20D because I *knew* that the (about) 50% increase in megapixels is almost meaningless when it comes to impacting final print size....you need about 2x the number of megapixels to jump to the next common print size...say, from 8x10 to 11x14...and of course *4x* the number of megapixels to truly double your print size from, say, 8x10 to 16x20. As for me, there's not going to be another DSLR in my future until I can afford something in the 24-30 megapixel range...anything less in terms of sheer resolution is not going to mean a helluvalot in terms of what I can *print*.

Regards,
Terry

[palmor]

I think everyone has given you some great advice. Like someone else mentioned try some graduated ND filters for the sunset photos. You'll be amazed how much they help. I used them for the 1st time last month on a vacation to Yellowstone and they'll never leave my bag!

Keep plugging away and learning, if photography was easy it would be boring

John

[Kirivon]

Don't shoot at such narrow apertures. I think you're making the same mistake as I did when I got my first DSLR, which coincidentally was also a K100D.

When I initially started shooting with a digital I carried over the film sunny day f/16 exposure rules, and generally favored higher f-stops for the extra depth of field. After an outing to red rock, I noticed my photos almost looked as if I had applied a watercolor filter to them.

With the smaller sized APS-C sensor, diffraction takes its toll on image quality much sooner than when shooting with 35mm film. Most lenses reach their peak sharpness around f/4 - f/5.6, and by shooting at f/11+ your lens is probably producing a softer image than it would wide open.

Pentax SMC DA 35mm f/2.8 Limited macro - Review / Test Report

If you look at the MTF chart you can see what I'm talking about. Don't worry too much about MTF ratings, though. They don't really tell you anything about how the lens renders colors and micro-contrast, which in my opinion, is more important than outright sharpness. It doesn't hurt, however, to glance over some of the reviews to see where the strengths and weaknesses of your lenses lie.

Personally, when I need depth of field I opt for f/6.7 or f/8 at the most.

Diffraction Limited Photography: Pixel Size, Aperture and Airy Disks

Here's a site that explains diffraction in greater detail, and also has a nifty diffraction limit calculator at the bottom.

Also, keep in mind the reciprocal rule for shutter speeds. Basically, take the reciprocal of your focal length, and that's the minimum shutter speed you need to get a blur-free image. The effective focal length of a 35mm lens on a APS-C sensor is 53mm, so you'd choose 1/60th of a second since it's the closest shutter speed. When shooting with Shake Reduction it's easy to forget this rule, but it's paramount for landscapes where the detail you are trying to capture is very far away, and therefore very very small.

Regarding the 18-250, I think you may be disappointed by the results. The Sigma 10-20, Tamron 28-75, and the 35 f/2 are all fairly excellent lenses. Super zooms tend to be a compromise for the sake of convenience, and will be unlikely to match the quality of your current setup. I'd only suggest a super zoom if you find yourself missing photo opportunities due to frequent lens switching. If you're looking to fill in the telephoto end in your setup, the DA 50-300 is a good lens for the money.

Finally, as to get a K7 or not, I think it depends on what you want it to do. If you want it solely to take better landscapes, then I don't think it'd be a justifiable purchase. Upgrading your tripod will have a greater impact on your landscape photos than upgrading your camera body.

If you're traveling with someone who isn't interested in waiting around for you to set up photos, therefore making a tripod impractical, then you may benefit from the higher usable ISO of the K7. However, the ISO performance of the K7 isn't particularly improved over the K20D, which can be had new for half the price of a K7. The primary advantages of the K7 are better white balance and focusing under artificial light, improved focus speed, and video capability. If you don't need or want any of those things, then I'd wait until the price comes down before upgrading.

[Marc Sabatella]

Nobody said "useless". I'm merely stating that for *photo quality* prints the image needs to be something like 1/2 to 1/3 the output resolution of the printer (by the way, there's some folks that say it needs to be AT the printer's resolution of 600ppi (HP) or 720ppi (Epson) so I think stating a minimum of 1/2 to 1/3 printer output resolution is not being unreasonable).

I agree. It's clear you know what you're talking about. I'm not trying to "correct" you here, since you are absolutely right in what you say. I'm just trying to clarify, for the benefit of those who are unfamiliar with these concepts, that what you call "photo quality" is an extremely high standard when talking about large prints, since large prints are normally viewed from farther back the sort of distance you view at when tying to judge if something is "photo quality" or not. "Poster quality" is likely to be sufficient when printing poster-sized, "billboard quality" when printing billboard-sized. Even magazines - which of course are not large at all - are typically only the equivalent of 150ppi. Not "photo quality", but for the intended purpose, good enough.

So you shouldn't *necessarily* spend a lot of money on a camera upgrade unless you know for sure what sort of resolution you need to get the results you expect. If you need it to look "photo quality" from up close, then yes, you want 240-300ppi at your intended print size. But if you just want a magazine-sized print to look as good as a typical magazine, a poster-sized print to look as good as a typical poster, or a billboard-sized print to look as good as a typical billboard, then you don't need that much resolution. How much resolution *do* you need, and how much are you willing to pay to get it? Well, those are questions everyone has to answer for themselves based on their expected print sizes, viewing distances, and personal standards for quality.

But while you're at it, you should also be asking if your lenses and tripod are of sufficient quality to actually produce meaningful results at these kinds of resolutions. I suspect the average person using the kit lens wide open handheld on a 6MP camera would gain more by using a prime and a tripod than by going to 10 or even 14 MP when it comes down to how good the print actually looks.

I could go off on the "megapixel" race as well...personally, I had a hard time justifying going from my K10D to the K20D because I *knew* that the (about) 50% increase in megapixels is almost meaningless when it comes to impacting final print size....you need about 2x the number of megapixels to jump to the next common print size...say, from 8x10 to 11x14...and of course *4x* the number of megapixels to truly double your print size from, say, 8x10 to 16x20.

I'm glad you pointed this out - that's another fact often not fully appreciated by newcomers to digital imaging. It takes four times as many pixels to double the dimensions of a print at a given ppi - meaning that as far as I know there is *no* APS-C camera (?) that will produce as much resolution at 16x20 as a 6MP camera does at 8x10. And yet, people have been printing posters for years, which illustrates the fact that for printing posters, "poster quality" may indeed be good enough/

[Das Boot]

Again, Enfuse can take a 3 image bracket from a 6mp camera and increase the resolution, sharpness, and dynamic range without looking phony. 16x20 will knock your socks off because the upscale is not interpolated.

[gfmucci]

Does Enfuse work only with Lightroom and Photoshop - not Elements? Is there a freestanding version?

[tlwyse]

Even magazines - which of course are not large at all - are typically only the equivalent of 150ppi. Not "photo quality", but for the intended purpose, good enough./

Hate to be Mr. NitPick but you might be confusing "ppi" or "dpi" with "lpi". Magazines are typically printed at 150 lpi or "lines per inch" screening but thats not the same as ppi or dpi. In fact, printing plates are typically imaged at 2400ppi, meaning that images are rendered at 2400ppi but then halftone screened at 150lpi (typically).

Rule of thumb for image resolution to be screened and printed for publication is that the image should be roughly 2x line screen resolution...so, typically, 300ppi images are suggested for 150lpi printing. As with anything, there's some latitude as you can go as low as about 1.5-1.7x line screen and the image will still print acceptably...so for magazine publication, 225-300ppi is generally OK for 150lpi publication printing. That's for conventional halftone (AM) screening. Interestingly (or not), FM or stochastic screening which uses a randomized dot pattern similar to an inkjet does not require as high an image resolution as conventional screening.

Regards,
Terry

[Marc Sabatella]

Hate to be Mr. NitPick but you might be confusing "ppi" or "dpi" with "lpi"

I tried not to; that's why I said "the equivalent of". I'm aware the printing process is more complicated than that. But while I'm no expert on different print technologies, it *is* my understanding that for all *practical* purposes, the bottom is line is that what you are actually getting resembles 150ppi. I know when I've had CD covers printed from my own artwork, the print quality is clearly way below what I'd expect of 300ppi, and that was submitting artwork capable of being printed 300ppi at that size, and this is exactly as the printer had explained to me up front would be the case. But a CD cover is not a magazine, to be sure; and if there is something different about the latter that causes the "actual" resolution to be higher, then I stand corrected.

But FWIW, looking at pictures in, say, the current issue of Shutterbug up close, they are also very clearly nowhere near 300ppi in practice. Putting a loupe up to one of those images versus the output from a 4x6" inkjet print made from a 1200x1800 image makes this very obvious. Unless those images were actually *submitted* at less than 300ppi, then I'd say what I said holds in practice - the printing process resulted in the "equivalent" of about a 150ppi print.

[Shashinki]

those pics look more like weird/botched PP resizing for web than any camera's fault to me.

i could be wrong though, i usually am.

i make up for it by being a jerk.

[interested_observer]

I am in somewhat the same boat as you, liking to take landscapes. Several months ago after using my K100 for about 4 years, I upgraded to a K20. It is an amazing camera and I have been enjoying it very much. The primary reasons for the upgrade were ISO 100, additional resolution and the additional features. Primary the additional features have helped increase my skill level. However, having said that, my wife and I recently finally, after 10 years picked up a large screen 52 inch 1024p TV. It has a USB port for images, so I loaded up on a thumb drive a number of K100 and k20 images. I was fully expecting to have the K20 images blow away the K100 images purely on resolution. Now, I do not know if it was the TV's rendering of the images or just the images themselves, but the K100 pictures were very unexpectedly stunning in their detail and resolution. The K20 as expected were also wonderful. Based on the difference in resolution, I was expecting a much larger difference, especially in image quality, especially at that display size. The point is, that my appreciation for the K100 has only increased with using the k20 - its a great sensor, low noise, and does a great job. I am glad I kept it as a backup. So, do not feel that the K100 is inferior and not up to the task.

Your picture is wonderful, and I need to drive north a bit and start using the equipment I have. Several suggestions come to mind. The first suggestion as was made before, is to use f8, and or find the best aperture for the lens, and use it. The next one is to use the depth of field charts for the sharpest aperture. This will give you a distance to manually set the focus to so that you keep everything in focus out to infinity. I would then adjust the exposure time based on the selected aperture, and if you need a tripod - then invest in one - it does not have to be massive, but a reasonable travel one, I find works well for me. As long as I am on the topic of tripods, use the 2 second mirror lockup so as to let the vibration from the mirror dampen out, and for a couple of dollars, get a wired external shutter release. I am suggesting a wired one, because on the K100 function menu (up arrow), you can either select the 2 second mirror lockup or the IR shutter release, but not both. So with the wired external release, you can essentially do both.

Here is a depth of field table...

Depth of Field Table
Online Depth of Field Calculator

I am using your Sigma 10-20 as an example. Find the aperture that provides the best result (IQ) and use that. For this lens it appears to be 20mm at f8. For the FA35/2, I think that you can search the web to find what other users have found to be the sweet spot for that lens, through experience.

Sigma 10-20mm f4-5.6 EX DC HSM Lens Review: 3. Test results: Digital Photography Review

For the Pentax 18-55 kit lens, it appears that 24mm at f5.6 is the best for that lens.

http://www.dpreview.com/lensreviews/pentax_18-55_3p5-5p6_p15/page3.asp

Now I am going to go way out on a limb here (and probably saw it off behind me), but there is the same analysis for the Nikon 35mm f1.8 prime available, so I am going to assume that the Pentax FA35 would perform in a similar fashion. It looks like it is at its optimum at around f5.6, so I would guess that the FA35's prime optimum point may be similar - at around f5.6.

http://www.dpreview.com/lensreviews/nikon_35_1p8g_n15/page3.asp

... and it actually appears that f4 is the spot for the FA35/f2 -

http://www.photozone.de/pentax/123-pentax-smc-fa-35mm-f2-al-photozone-review...report?start=1

http://ca.geocities.com/spirope/infinitytest.htm

One of the main reasons why I choose Pentax was the in body image stabilization, since I was interested in wide angle shots and none of the other makes even supports lens based IS for wide angle lenses. However, there are limits to it, and that is where the tripod comes in. I find it especially useful when taking HDR or bracketing. With the K100, I would get to the third image and have to wait for the prior two images to be written out so that the third one could then be taken, thus trying to keep the camera still - handheld was an issue - hence the tripod.

Here is a post of mine - somewhat similar to yours, from earlier this year. Some of the same problems, and similar suggestions, with improved results.

https://www.pentaxforums.com/forums/general-photography-techniques-styles/653...otography.html

Going back to your original question new body or lens, I would opt for lenses, since they last the longest, and it appears Pentax equipment is rising in price. Electronics always become cheaper - especially when the model goes into closeout. I slowly acquired the lenses I really wanted, while I had the K100, and then upgraded when the K20's price came down. One of the reason for the K20 was I did not want to wait an additional 18 - 24 months for the K7 to become more affordable (i.e., at the end of its model run).

Now, that the summer is starting to end, and the heat will moderate a bit, I really need to go up north and take advantage of the scenery. I lived up in Seattle for a number of years, and there are quite a few places I would want to go with a camera in hand....

hope that helps....

[KJon]

Despite the technical nature of this thread, I didn't notice that anybody mentioned turning SR off when using a tripod.

Best to all,
Kevin

[rparmar]

I was fully expecting to have the K20 images blow away the K100 images purely on resolution. Now, I do not know if it was the TV's rendering of the images or just the images themselves, but the K100 pictures were very unexpectedly stunning in their detail and resolution. The K20 as expected were also wonderful. Based on the difference in resolution, I was expecting a much larger difference, especially in image quality, especially at that display size.

This expectation is based on the premise that we all start out with: 14MP is more than twice as big as 6MP and so should result in lots more detail. This is based on two false premises. First, if we are doubling area (say for an enlargement) we need four times the pixels, so really 24MP is the next distinct step up from 6MP. Second, more pixels does not necessarily equal more detail, for that we need a larger sensor. So we can expect more detail by going to full-frame, 645, 6x7 and so on up to large format. And that's why landscape photographers work in those sizes.

Despite the technical nature of this thread, I didn't notice that anybody mentioned turning SR off when using a tripod.

On a tripod I'm always using a timer, which turns off SR anyway.

[interested_observer]

This expectation is based on the premise that we all start out with: 14MP is more than twice as big as 6MP and so should result in lots more detail. This is based on two false premises. First, if we are doubling area (say for an enlargement) we need four times the pixels, so really 24MP is the next distinct step up from 6MP. Second, more pixels does not necessarily equal more detail, for that we need a larger sensor. So we can expect more detail by going to full-frame, 645, 6x7 and so on up to large format. And that's why landscape photographers work in those sizes.

I'm glad you pointed this out - that's another fact often not fully appreciated by newcomers to digital imaging. It takes four times as many pixels to double the dimensions of a print at a given ppi - meaning that as far as I know there is *no* APS-C camera (?) that will produce as much resolution at 16x20 as a 6MP camera does at 8x10. And yet, people have been printing posters for years, which illustrates the fact that for printing posters, "poster quality" may indeed be good enough/

Wonderful observations! - I have dealt with adaptive beam forming sensors long enough that I just let the computer figure it out, rather than start with the basics and think it through.

Anyways, back to the question. Maybe the biggest issue is my photography skill (e.g. underexposure,) plus lack of practice in PP. Still, the white details on the rock surfaces looks so grainy to me. Not out of focus, but grainy. I did learn Hyperfocal focusing a while back. Would this be clearer with a K20?

At f11 it looks like you were trying to reach for depth of field and in doing so gave up light and thus were forced to rely on stabilization hand holding the camera. With the best aperature at f4 and using the appropriate DOF / hyperfocal distances, you could have had just about everything going in your favor - more light gathering ability, better sharpness, and the depth of field - along with a faster shutter speed. You also had two really great lenses the 35 and the 10-20. So, I think that we are both somewhat in the same boat, having to do a better job of taking the original image (with what is at hand - thinking the problem through), essentially improving skills.

I also believe that its not necessary skill at PP, as you really need good original material in order to make it better through software manipulation. So, in my opinion - it comes back to the skill in taking the original image, with good optics, and capturing it off the sensor.

I like the original picture very much. Somewhere, I have seen a (and I want to say somewhat famous) oil painting that is very similar. Also, I can say - that it is much easier (at least for me) to analyze what I should have done, sitting after the fact looking at a large monitor, than when I was actually taking the picture. That is why, to some degree - I am trying to just have somewhat of a standard process to go through - especially when I have a shot that I think maybe is (or can be) something special. I have tried to go through and determine the best aperature and focal length for each lens. Make a note of them, and make a trimmed down DOF chart to stick in my backpack.

In my mind at least, all of this just reinforces the concept, that the K100 with good glass has both tremendous capability and potential. Few can afford to go to a substantially larger physical sensor in terms of cost, weight and size (lenses), so with the high quality glass that is currently available for the APS-C, this has the best combination available of all the attributes, and to be really honest (at least for my needs, wants, checkbook and skill set), this is it.

hope that helps...