Newbie questions (FF vs. crop, extension tubes) & snowfl

[dudski]

Helllooo,

First off, this place is awesome. Gotta say, you guys are a very talented and helpful bunch! The only thing this site is missing is a more advanced search engine. I'm sure the questions I'll be asking have been answered before, just can't find the threads.

So my main (macro-related) interest lies in snowflake photography and I'll be using stacked lenses (105mm f2.8 macro & reverse-mounted 40mm f1.8 ) for this application. I'm quite eager for the winter, as it was only toward the end of last winter that I started to get a hang of the technique, lighting, etc. for snowflake photography. In addition, I've geared up a little, though about my setup, and discovered both focus stacking techniques and this website!! So you better believe I'm looking forward to some nice snowflakes this winter.

First two questions will be basic - and I'm assuming that both have been discussed at length and answered before, but I just can't seem to find them!

#1: I've got two bodies at my disposal: a crop-sensor Nikon D200, and the FF D700. My question simply boils down to whether it's worth using the D200 for the crop factor? My assumption is NO and the reasoning behind this is that the number of resolved pixels with the equipment that I am using (it's sharp, but obviously not as sharp as a dedicated lens or enlarger on bellows, etc.) is not high enough to make use of the increased pixel density of the crop sensor, bringing it down to light-gathering ability and the better quality sensor of the D700. Am I correct in this assumption? Also, I would imagine that diffusion would kick in sooner due to the smaller effective aperture of the crop-sensor setup (or is this completely off since a crop sensor doesn't actually magnify?)

#2: I've got some Kenko extension tubes at my disposal. So far my (somewhat basic) tests have reveiled 1) a decrease in image quality (assumedly since the macro lens is no longer in it's optical prime) 2) more vignetting (not surprising with the stacked lens setup).
Now do extension tubes not simply magnify the image, without changing the effective resolution?? So by similar logic to #1, if I'm not resolving more pixels AND reducing the effective aperture, then why should I even bother with these.. which boils down to: what are extension tubes for in the first place??

So confused.

Ah, and here is a link to last years efforts, all taken with the reverse-mounted 40mm on a crappy 70-300 bought at a garage sale for 5$: http://www.pbase.com/damian_turski/snow

As you'll probably remark, the IQ isn't good at all.. so looking to improve on that this year.

Thanks in advance for the help!

[rjlittlefield]

dudski, welcome aboard and thanks for the kind words.

Answering your questions...

#1. Go with the D700. As I read the specs, the D700 has more total pixels than the D200 (12.1 megapixels vs 10.2), and the larger sensor will capture more light and give you better gradation. The D200 could be better in special cases if you have an issue with framing, for example being able to fill the whole frame of the D200 but not the D700. Otherwise the D700 has all the advantages.

I would imagine that diffusion would kick in sooner due to the smaller effective aperture of the crop-sensor setup

I think you meant diffraction, not diffusion. If so, then one good answer is that sensor size makes no difference in diffraction when you set up to get the same framing and same DOF with both sensors. In that case the effective aperture will end up being proportional to the sensor size -- bigger sensor, bigger f-number. At same light level and ISO setting, this will require a longer exposure time, and that's what allows the larger sensor to capture more light and give you that better gradation.

#2. Extension tubes give you more magnification from the same lenses. Contrary to advertising claims that trumpet "nothing to degrade your image!", extension tubes often do degrade your image. This happens because the added extension pushes the lens outside its design space into a focusing relationship where it has significant aberrations. How much the image degrades depends on the lens and the amount of extension, and there's no good way to tell without trying it.

A couple of other points...

If you're new to combos (stacked lenses), then be aware that you can get significantly different results depending on how you stop them down. See FAQ: Stopping down a lens combo for some discussion and illustration of that.

About searching in the forum, yeah, that's a problem. Even those of us who know exactly what to look for sometimes have trouble finding the right posts because so many of the same words appear in lots of postings. You might have more success using Google's advanced search engine (http://www.google.com/advanced_search) with the "site or domain" restricted to photomacrography.net .

--Rik

[dudski]

Rik,

Thanks for the quick response!

First off, yes, I meant diffraction.. don't know why I wrote diffusion.

Next, I should have mentioned that on a full-frame camera and my current setup which, from what I understand gives approximately a 2.5x magnification (105 / 40.. or slightly less due to the shortened focal length of the macro lens at closest focusing distance), this is generally not enough to fill the frame with your typical snowflake. Hence the question of whether to use the D200 for this application.

So, here is where some of the confusion kicks in (especially since you've described this exact scenario where the D700 cannot fill the frame): why even touch on total megapixels if the driving factor is the resolved megapixels? And seeing as I'm probably not resolving even 10MP (assumption), then why would I even think of using the D200?

My understanding of a good occasion to use a crop sensor camera (such as the D200), for example, would be if I am certain that I could obtain a higher effective resolution photo (by using better optics) AND I am not filling the frame.


My thinking is similar for the extension tubes - which are, with my current understanding, effectively "cropping" the image without increasing the number of resolved pixels.. so if 2 lines separated by a small distance appear to be 1 (not resolved by optics), then it doesn't matter how close you get (via simple magnification), they'll still appear as 1..

With regards to stopping down a lens combo, I learned the hard way last winter!

[Chris S.]

Welcome, Damian! You've gotten some pretty nice results so far. Are you using a lighting aproach similar to Kenneth Libbrecht's?

As I read the specs, the D700 has more total pixels than the D200 (12.1 megapixels vs 10.2), and the larger sensor will capture more light and give you better gradation.

A couple of years ago, I read the specs and came up with the same analysis as Rik. But in later testing with both cameras in hand, I was surprised to find the opposite to be true: The D200 produced very noticeably better gradations than the D700. I was so surprised that I kept checking the metadata, thinking I had surely mixed up the images. My first test was of a closet full of clothing (representing many colors and textures) illuminated by a large north-light window (providing very smooth gradients from highlight to shadow). I used a collared zoom lens mounted on a tripod, so that the only changes between shots were the camera body and zoom setting (to adjust for equal framing. Since I was testing at base ISO (where any camera should deliver its best performance), and the base ISO of the D200 is 100 and the D700 is 200, I needed to adjust shutter speed to keep exposures consistent. On some other shots, I also adjusted the aperture setting to keep DOF visually consistent. Not that it matters, but I of course used a cable release and mirror-lock-up. Shot raw+jpg.

I expected the D700 shots to look better than the D200. Instead, they looked dramatically worse. Quite an eye-brow-raising thing to see from one's new $3000 camera! At first I was looking at the jpegs, and figured this to be some artifact from the jpeg engine. But it held true for the raw files, too--I could consistently make better images out of the raw files from the D200. The difference was in the subtle gradations of rendering.

That was over three years ago, and constant use of both cameras over that time has convinced me that the original test told the truth. Don't get me wrong--I'm a huge fan of the D700. Any time I can't control the light, I'd much rather have it than the D200. It can handle high contrast situations far better than its older sibling (that is to say, it has a much higher dynamic range). The D700 has produced images that would never have been possible with the D200: Milky Way rising in the desert; nighttime shot of a little girl touching a live bat (held by a biologist) under mixed illumination of headlamps and flash; wedding under trees with bright spots of dappled sunlight mixed with open shade and balanced with fill flash. If you want my D700, you will have to pry it out of my cold, dead fingers.

But under conditions where I can keep the light within the more limited dynamic range of the D200, this camera produces better rendering. I suspect the difference may involve the fact that the D200 has a CCD sensor, the D700, a CMOS sensor. This suspicion has kept me from purchasing a D300 (CMOS) for the Bratcam.

Since you have both bodies, you can easily test this yourself. That said, the lighting paradigm you are using seems to enhance contrast, rather than produce fine gradations of light, so it may not matter in your application.

Regarding your 105 macro looking worse with extension tubes, have you tried reverse-mounting it? This is likely needed for best quality above1x. That said, most snowflake photography is done in the neighborhood of 4-10x, which is way beyond the sweet spot for a reversed 105mm.

Also, I'm not a big fan of stacked lens combos. Been there, done that, never again. I wonder how your 40mm would perform alone, reversed on those extension tubes or a bellows? It might be much better than the combo. Many of us here have used an old EL-Nikkor N 50mm f/2.8 enlarging lens, reversed on a bellows, for magnifications that would be good for snowflake photography. Mine knocks the socks off any lens combo I've tried. These EL-Nikkors are very inexpensive on the used market. Don't have a bellows? A cardboard tube lined with black felt will work just fine.

Cheers,

--Chris

[dudski]

Chris,

Those are some interesting results! Eventually I'll do some tests of my own with the snowflakes this winter to see which body handles the situation better. At the same time, I'm very interested in the theory behind it all - and as you could probably tell it's making my head spin a little bit.

Between the two cameras, I rarely ever pick up the D200 unless I'm in a situation where I fear for the safety of my camera.. otherwise the D700 simply whups the D200 for all other areas of photography that I do. In fact, sometimes I feel that I'm "cheating" with the D700 on auto-iso set to go up to iso6400... it's kind of like shooting on automatic!

I haven't tried reverse mounting the 105 as I currently don't own a reverse-ring - though I'm seriously considering picking up the whole kit (el-nikkor 50mm, bellows) and taking it from there. I'm a little bit concerned about the price of the bellows though and whether I would need to go with something of quality like the Nikon PB-6. I need to do some research on this stuff, but I'm already looking to pick up a el-nikkor for $45 locally.

The whole lens combo thing for me started a few years ago when I read a tutorial on how to do it.. a few weeks later I had found, at a garage sale for $5 each, a 70-300 and the 40mm f1.8 - but was so discouraged by the results (I had really no clue what I was doing) that I ditched the effort for long time. Then last winter, with not much to do, I picked things up and started slowly crawling along. In fact, up until about a month ago when I discovered this site, I had no clue what bellows, enlarger lenses, etc. was! (snowflakes are in general my only macro-related interest.. though that might change if I manage to hook up a good setup) And yes, I am using a lighting approach similar to Kenneth Libbrecht - it's his website that taught me that it's even possible to take photos of snowflakes. Though last year was a little bit of a free-for all with a very very very umm crappy setup (not even a roof over my head and taking pictures outdoors... I couldn't even see through the viewfinder much of the time because it was filled with snow!)

We're starting to get some snow up in my neck of the woods, but it will still be a good couple of weeks before it gets cold enough for any interesting snowflakes to appear.

Cheers,

Damian

[Harold Gough]

The only thing this site is missing is a more advanced search engine. I'm sure the questions I'll be asking have been answered before, just can't find the threads.

Have you tried Boolian searching. If you want to find terms together insert AND between them: term1 AND term2.

For three terms: term1 AND term2 AND term3.

Harold

[ChrisRaper]

The only thing this site is missing is a more advanced search engine. I'm sure the questions I'll be asking have been answered before, just can't find the threads.

What Harold said ... but also remember that you can use Google to search single sites:

term1 term2 term3 -term4 (term5,term6) site:www.photomacrography.net

[Harold Gough]

Here is some previous discussion:

http://www.photomacrography.net/forum/v ... t=searches

Harold

[Chris S.]

I'm seriously considering picking up the whole kit (el-nikkor 50mm, bellows) and taking it from there. I'm a little bit concerned about the price of the bellows though and whether I would need to go with something of quality like the Nikon PB-6. I need to do some research on this stuff, but I'm already looking to pick up a el-nikkor for $45 locally.

Given your (current) concentration specifically in snowflake photography, this sounds like a very good plan. In this scenario, I don't think you need something like the Nikon PB-6, though you certainly might like one. I have one, as well as a simpler, cheaper off brand. Both work, but the Nikon is more convenient, and provides a mechanism to avoid having front of the unit hit the subject at less than full bellows extension. But you've said that for now, your sole macro interest is snowflakes, which limits your current needs considerably. I'd imagine that with a 50mm enlarging lens, you'll mostly be working at nearly full bellows extension, where the placement of the front standard is rarely a problem. And most snowflake work seems to be done around 5x, where mechanical concerns are not the issue they are at higher magnifications. If the cost is no great worry, I recommend getting a PB-6 (or possibly a PB-4, which I haven't used but understand to be highly regarded). It's a great piece of equipment, and I doubt the value will decline. But a budget option will likely serve your needs pretty well, too.

Probably of more importance is the means you employ of controlling the distance between camera rig and snowflake. But you may have that worked out already.

Cheers,

--Chris

[ChrisLilley]

#1. Go with the D700. As I read the specs, the D700 has more total pixels than the D200 (12.1 megapixels vs 10.2), and the larger sensor will capture more light and give you better gradation.

Yes, you read the specs correctly (12.1 MPx full frame vs. 10.2 MPx 1.5 crop). But there is more to it than that. The D200 is a soewhat old camera - very well built, but with a CCD sensor. It was replaced by the D300, then the D300s, and arguably superceeded by the D7000. So it is a few generations older in sensor technology than the D700 which (while it is expected to be replaced fairly soon) is still current.

DxOMark has publishd comparisons of the sensorsi n these two cameras - have a look especially at the dynamic range and the colour rsolution tests:

http://www.dxomark.com/index.php/Camera ... 2%29/Nikon

(The forum software has decidsed to break th e URI so copy and paste the whole thing rather than just clicking on it).

[Oskar O]

I find a bellows and enlarger lens handy for photographing snowflakes. But the big disclaimer is that I use different techniques than Librecht, I'm not interested in duplicating his results. Still, such a combination might be handy, as you could setup the camera without having to calculate what tube length you need. Snowflakes vary in size a lot depending on weather, but depends on where you live which conditions you encounter and you may want to photograph only specific types.

With a crop sensor camera, something like 3x-5x should be pretty convenient, depending on flake size and how frame-filling you want it to be.

[Chris S.]

But there is more to it than that. The D200 is a somewhat old camera . . . with a CCD sensor.

In parallel reasoning, the Nikon N Plan CF Apo 4x 160/0.20 is a somewhat old lens. . .and a finite optic. What of it? This lens is among the best optics, new or old, yet tested at 4-5x. Let's not drink the Kool-Aid--newer is not an automatic pass to "better."

CCD vs. CMOS: Which is better, oil paints or watercolors? Kodachrome or Ektachrome? Mozart or Beethoven? Chicken or fish? An assertion that either CCD or CMOS technology inherently produces better images is difficult to defend. Differences among particular embodiments strongly involve the optimization choices of the designers who integrated whichever chip into the imaging system.

It was replaced by the D300, then the D300s, and arguably superceeded by the D7000. So it is a few generations older in sensor technology than the D700. . .

Another attempt at a parallel assertion: Tony Bennett is a few generations older than Britney Spears. Whose music is better? Older is not necessarily worse. I have a choice of available camera bodies, and after testing and extensive use, find the D200 to produce better results than the D700 in controlled-contrast conditions (though find it very much the opposite for conditions with challenging contrast). Any photomacrography under artificial light strikes me as being in the "controlled-contrast" category.

The OP also owns both of these bodies. If he wants best results, he should test for himself. Part of being a great painter is testing the paint.

The Nikon D700 is almost a see-in-the-dark camera. It is a high-contrast champion. Conversely, the Nikon D200 is horrible in dark, high-contrast situations. Which is better? Depends on what you are doing with it. Within its dynamic range sweet spot, I find the D200 to render color gradients with more fidelity and subtlety than the D700. Is this so surprising? Remember in film days, when we made choices among film stocks that emphasized either expousre latitude or quality of color rendition? Today's digital imaging engineers may have to make similar tradeoffs to those formerly made by film engineers.

Studio photomacrography--including snowflake photography with something like Rheinberg illumination, even when done outside--is a controlled contrast situation. We need not make color rendition sacrifices in order to capture large contrast ranges that do not exist in these scenarios.

DxOMark has publishd comparisons of the sensorsi n these two cameras - have a look especially at the dynamic range and the colour rsolution tests. . .

This strikes me as an example of why limited measurements, such as those done by DxoMark, should at best be a starting point for making one's own observations. Those of us who are also interested in high-end audio know that objective tests may provide suggestions for components we ought to go out and listen to; but when purchasing, it is one's ears, rather than limited test data, that should be followed. I hope and expect that someday, objective measurements will improve in concept and precision enough that they predict how the human eye/brain system processes images. But so far, DxoMark's tests fall short.

Another problem with the D700 vs. the D200 or D300 series Nikon bodies is that the large sensor makes it difficult to use many of the optics so far demonstrated to be optimal for photomacrography. The needed image circle for the Nikon D700 is simply too large. I love, love, love, my D700 for general photography, but for photomacrography, consider it a poor choice.

The Nikon D7000 is an interesting piece of disruptive technology. I have never had one in hand, but would enjoy borrowing one to test. The many images I have seen from it look over-processed--like something I could do with raw files from the Nikon D200, but with overdone elements that I would avoid. But no way can I confirm or deny this without having a D7000 body in hand to experiment with. And the cost of the body is too high to justify such tests.

Cheers,

--Chris

[ChrisLilley]

But there is more to it than that. The D200 is a somewhat old camera . . . with a CCD sensor.

In parallel reasoning, the Nikon N Plan CF Apo 4x 160/0.20 is a somewhat old lens. . .and a finite optic. What of it? This lens is among the best optics, new or old, yet tested at 4-5x. Let's not drink the Kool-Aid--newer is not an automatic pass to "better."

Chris, I hope you know me better then that by now.

My argument ws not 'oh, its new, it must automatically be better' or 'technology A from whatever manufacturer is always better than technology B from whatever manufacturer'.

My argument was that, following the development of this line of cameras over time to aid my purchasing decisions, saw improvements over those generations. I also own Nikon DSLR of those same generations - D40 (6MPx CCD), D90 (12 MPx CMOS) and D7000 (16MPx CMOS) and in each case compared the performance of that sensor to those of its predecessors and of higher ad lower models in the same line (D40 vs D40x, I picked the D40 as beter; D90 vs D200 and D300, picked D90 as better; D7000 vs D300s, picked D7000 as better).

An assertion that either CCD or CMOS technology inherently produces better images is difficult to defend.

Luckily then, that wasn't an assertion that I was making; so I have no interest in defending it.

I have a choice of available camera bodies, and after testing and extensive use, find the D200 to produce better results than the D700 in controlled-contrast conditions (though find it very much the opposite for conditions with challenging contrast). Any photomacrography under artificial light strikes me as being in the "controlled-contrast" category.

That is a useful finding, and would be even more useful with side by side comparisons of the same subject with the same lighting.

The OP also owns both of these bodies. If he wants best results, he should test for himself. Part of being a great painter is testing the paint.

Again, you seem to be trying to stuff strawman arguments into my mouth; this time, 'you should not do comparative tests of products you already own'. I would rather you didn't do that.

Yes of course the OP shoud do tests on their own equipment. But they also asked for advice, and I gave some based on my onw experience and closely following the development of Nikon DSLR sensors over the last 6 or so years.

Within its dynamic range sweet spot, I find the D200 to render color gradients with more fidelity and subtlety than the D700. Is this so surprising?

Yes, it is surprisig to me, since you ask; given the dynamic range test results at base ISO of the various cameras being discussed.

This strikes me as an example of why limited measurements, such as those done by DxoMark, should at best be a starting point for making one's own observations.

Yes. And those tests have been a starting point for mine; and having purchased equipment based on those tests, I have so far found them to be an accurate indicator when it cam to testing and shooting the new camera alongside the previous ones I have owned.

Do those tests tell you everything? No, clearly not. On the areas tested though, I do find them to be a useful predictor.

[dudski]

If the cost is no great worry, I recommend getting a PB-6 (or possibly a PB-4, which I haven't used but understand to be highly regarded).

There is someone close to home selling a PB-4 for 75$, and from what I gather this is an excellent deal. Does anyone have any experience mounting either a D200 or a D700 to such a bellows?

[ChrisLilley]

If the cost is no great worry, I recommend getting a PB-6 (or possibly a PB-4, which I haven't used but understand to be highly regarded).

There is someone close to home selling a PB-4 for 75$, and from what I gather this is an excellent deal. Does anyone have any experience mounting either a D200 or a D700 to such a bellows?

75 USD is a good price for a PB-4. A D200 and D700 should mount without problems unless you have a battery grip installed,in which case a small extension spacer may be needed.

Note that you mount the camera in portrait orientation, then swing it wound to landscape if needed.