[acoufap]

Has anyone even posted an image taken with a K-1 and TC?

ok - two shots - both taken with DA TC1.4x and DFA100WR @ f11:

near focused


far focused


Same result at f4.

So the pattern for this lens seems to be

• near focusing clearly not FF image circle
• far focusing nearly ok

Same pattern with my Vivtar S1 70-210 - I think at all apertures and FL.

The following image I took with M1.7/50 + DA TC1.4x at nearly nearest focus:



It's ok as other images I photographed - no vignetting, everything's fine ...

and so does the M2.8/28

---

Last edited by acoufap; 05-20-2016 at 10:52 AM. Reason: added hint for M.28/28; added FL for 70-210

[normhead]

Interesting, if you get a chance.

A couple of images, an APS_c image, and an image taken with the same lens with the 1.4 in FF mode would be nice.

[acoufap]

DFA100WR - FF mode


f4 - nearest distance


f7.1 - nearest distance (sorry a bit unsharp - handheld 1/13 at ISO 3200 ... it's early evening now)


f4 - a bit more than 1m distance


f7.1 - ca. 1 m distance

In APSC mode I didn't recognize any problems. Maybe tomorrow I can photograph some of this kind.

---------- Post added 2016-05-20 at 20:21 ----------

And here's one with Vivitar S1 f4@210mm + HD TC1.4:

Same Lens f7.1@210 + HD DA1.4x, APSC mode:

[acoufap]

Interesting, if you get a chance.

A couple of images, an APS_c image, and an image taken with the same lens with the 1.4 in FF mode would be nice.

Hi Norm,

here are two images taken today with K1 APSC mode, DFA100WR and DA TC1.4x - no vignetting problems.

Think that's what you wished to see - don't you?

regards acoufap

[ScooterMaxi Jim]

Norm, I'm wondering how much experience you have with shooting FF?

Your comments exhibit a bias essentially equating image quality to resolution. As you well know, based on the lens tests you have offered here, many factors go into the subjective evaluation of overall image quality.

In answer to the resolution point, the K-5 series (among other bodies prior to the K-3) has a pixel density that correlates closely to the K-1, and many subjectively feel the overall image quality is cleaner and resolution loss is not a significant factor overall (K-5 to K-3 comparison). This analysis can be debated, depending on image subject, need for extended dynamic range, cropping, and required differences in processing. In fact, I come down slightly on the side of the K-3 despite its clearly noisier baseline. Certainly, as we have already seen, the K-1 is clearly capable of clean images, wide dynamic range, and the attributes that come with the larger sensor size requiring less magnification (the real key to matters).

The question, then, which you have virtually eliminated in your analysis, is the impact of added optical complexity and unavoidable image degradation. With the K-1 you get to see the degraded image more clearly. We are already seeing that now that the images - even from a very high quality source lens - are getting posted. Over time, I expect the majority of critical users to agree that the K-1 demands higher quality optics with an adequate image circle to achieve an appreciably higher quality image compared to the K-3. In short, mostly pay to play. That was my experience on the Canon side of things.

[normhead]

Norm, I'm wondering how much experience you have with shooting FF?

Your comments exhibit a bias essentially equating image quality to resolution. As you well know, based on the lens tests you have offered here, many factors go into the subjective evaluation of overall image quality.

In answer to the resolution point, the K-5 series (among other bodies prior to the K-3) has a pixel density that correlates closely to the K-1, and many subjectively feel the overall image quality is cleaner and resolution loss is not a significant factor overall (K-5 to K-3 comparison). This analysis can be debated, depending on image subject, need for extended dynamic range, cropping, and required differences in processing. In fact, I come down slightly on the side of the K-3 despite its clearly noisier baseline. Certainly, as we have already seen, the K-1 is clearly capable of clean images, wide dynamic range, and the attributes that come with the larger sensor size requiring less magnification (the real key to matters).

The question, then, which you have virtually eliminated in your analysis, is the impact of added optical complexity and unavoidable image degradation. With the K-1 you get to see the degraded image more clearly. We are already seeing that now that the images - even from a very high quality source lens - are getting posted. Over time, I expect the majority of critical users to agree that the K-1 demands higher quality optics with an adequate image circle to achieve an appreciably higher quality image compared to the K-3. In short, mostly pay to play. That was my experience on the Canon side of things.

The image degradation is pretty clear from the Nikon side of things. A D7000 16 MP can achieve about 2200 lw/ph, a D810 about 4000. Both using the Sigma 70 macro. SO you double your sensor size, you don't quite double you resolution, but it's pretty close. Again bach gleaned from actual tests at Imaging Resources, not conjecture about what might be. I guess there could be lenses so bad they really mess up on FF, but there is absolutely no data supporting that idea.

[ScooterMaxi Jim]

The image degradation is pretty clear from the Nikon side of things. A D7000 16 MP can achieve about 2200 lw/ph, a D810 about 4000. Both using the Sigma 70 macro. SO you double your sensor size, you don't quite double you resolution, but it's pretty close. Again bach gleaned from actual tests at Imaging Resources, not conjecture about what might be. I guess there could be lenses so bad they really mess up on FF, but there is absolutely no data supporting that idea.

You mean no data that you have quoted - as opposed to all that exists, and further based on your actual experience - quite possibly none, as indicated so far. Again, you're quoting nothing more than test chart resolution from a high-resolution macro lens. So, what teleconverter did imaging-resource.com use to make the comparison we are discussing? I wasn't aware they did any direct testing of the effect of a TC on the D810 (or any other FF), as compared to lens only on a D7000 (or any other crop sensor body).

[normhead]

You mean no data that you have quoted - as opposed to all that exists, and further based on your actual experience - quite possibly none, as indicated so far. Again, you're quoting nothing more than test chart resolution from a high-resolution macro lens. So, what teleconverter did imaging-resource.com use to make the comparison we are discussing? I wasn't aware they did any direct testing of the effect of a TC on the D810 (or any other FF), as compared to lens only on a D7000 (or any other crop sensor body).

We do know that from Pentax the 1.4 TC costs about 3% resolution so w could count on the Nikon using the Sigma 70 with the D810 to be about 3880. Say they are overly optimistic and it's 5%. The D810 will come in at 3800. Do you even have any math at all to support your conjecture?

In other words, you've got nothing.
You could have just said that. You continue to want to live in a world of conjecture.
Shown some viable facts I'm able to do an about turn quite quickly, I just haven't been shown any viable facts, here. At least I have some data supporting my opinion.

Look at the numbers on photozone for say the Sigma 85. You'll see the TC would have to cost the lens 30% of its resolution to match the FF even on the edges. More like 45% in the middle. So yes the resolution is weaker on the edges in both, , but FF resolution never degrades enough to be in range of an APS-c sensor, even if on the extreme edges.

I can't believe we are even arguing this.

[ScooterMaxi Jim]

We're arguing about it because the actual testing of this sort (FF with 1.4x vs. APS-c directly) hasn't been posted. Even the Acoufap photos are not showing direct comparison of the same scene. All we know for sure is that the DA TC does quite poorly in the corners at distances approaching infinity - the situation for which it most likely will be most heavily used. That somewhat answers it at face value. In my times that I used the 20D and 5D, it never really struck me as useful to do that sort of testing with the TC, and my overall impression favored what I am stating here - especially regarding edge sharpness and CA issues. However, the Pentax TC is better, so that introduces uncertainty. At the same time, the Pentax is an unmatched TC, and will vary considerably based on the lens used. Sorry to put it conditionally, but this is a topic those of us who have a deep understanding need to explain in detail. Not sure you're quite there, Norm.

The statistical comparisons you have cited have no applicability to this discussion because they don't include the comparative elements; i.e. no TC in the test so you're just throwing out some red herrings.

[normhead]

Really, no applicability? Who made you the judge of what is applicable and what isn't? For the purpose I was using them for, to demonstrate how unlikely it is that your assertion is correct, they are quite applicable.

And if you want to talk about red herrings... how about someone who asks about Pentax equipment, and someone answers with their experience with Canon equipment. Ok, now is this the same person declaring some one else's data irrelevant? If true, it's hilarious.

I know you're trying to be reasonable Jim, but the only way that the FF image is better than the APS-c image, is if the HD just barely covers the APS-c sensor, with no wiggle room. If the FF image has an extra 200 pixels each way to work with using the TC and avoiding vignetting, it will be better. It can't be noticeably less.

And you can declare my point of view as irrelevant as you want. That just makes me think perhaps you just don't understand the relevance. Did I not explain it clearly enough to you? Maybe you can find another way of expressing yourself, where that is not quite so obvious.

It's funny how we look at Acufap's images, and come to totally different conclusions. But that's what biases do. They don't help you interpret data, they actually cause you to mis-interpret data.

[ScooterMaxi Jim]

Yep, that's what biases do. We agree on that.

[normhead]

Yep, that's what biases do. We agree on that.

Thats why I run tests and polls. For me, what I think is only part of the equation. Some of us are pretty much caught up in the universal relevance of our personal opinions.

That helps me eliminate biases. SO, i don't really trust anyone who doesn't do the same. If you don't work to understand your own biases, you'll always be subject to them.

[ScooterMaxi Jim]

Let's be very clear what the OP asked in this particular thread. No one with a good understanding of optical science denies that a good lens and a good TC can, and does, improve resolution and overall IQ on the same sensor (as is the basic supposition in another thread since started). This is true. As a user of several TCs of varying quality, the benefits and trade offs become clear over time.

Using a TC to overcome the limitations of lenses that perform poorly on FF edges is the subject of this thread, and is a less-settled discussion. Pentax engineers have been clear that the DA TC, in particular, is not overall well suited for FF. Obviously, there will be exceptions to the rule. Possibly, marginal added detail can be derived in the center of the image area when using good optics because the lens and quality TC perform best in their sweet spot and provide beneficial magnification. However, edges and corners will tend to be compromised, with CA and other imperfections magnified. Sure, you can crop out the bad parts of the image, but what then have you accomplished by going to FF in the first place?

Testers who don't have an agenda to put forward will have answers and add some clarity soon enough. Pretty clearly, the answer will be somewhere in between the idea that "TCs will salvage most DA lenses for superb FF use," and "TCs are just plain bad." No doubt, this is worth some trial and error, but keep in mind that TCs are not designed to work with wide angle optics, so don't expect expanded image circles for UWA lenses to yield anything approaching quality images. Light rays just don't happily bend that way (unless you happen to be riding a well-tuned super collider).

[normhead]

I have several DA lenses and am considering ordering a K-1. Wondering if the use of the 1.4x Pentax rear converter makes these lenses more compatible with the full-frame sensor by increasing the image size, therefore covering more of the sensor area.

"In the case of the DA *55-300, yes it does, so it is a distinct possibility. But don't count on it. There are as many factors working against it making the lens more useful on FF than there are for it. Without having the ability to anayse every lens design for the lenses that it might be attached to, it's probably going to be a trial and error process. "

How's that for s short answer.?

[Fredooo]

I have my K 1 with the 100 mmDFA macro WR and the diffraction appears beyond f/9,if i use an converter what could happen ,anyone has tried?