comparison

[robirdman]

They don't seem very sharp. I don't usually shoot at F1, as compared to F32, so I don't know how much area should be in focus though. I was using live view this time.

Previously I was shooting at 1/3" and this time 1/100" which would seem to make it sharper.

Could it be the change in flashes. the R1s were on the lens and not touching the subject are. One SB800 was on the camera and the other was on a small stand next to the subject. Could its firing caused vibration?

I went back and tried moving the one SB800 off camera. I created a 3/4 bowl with added translucent material, but I kept getting way too much exposure, even after I put white dust masks around each flash and moving back. Then I tried different apertures, and images seem much sharper at f32 than f8. the smaller the number, the less in focus even with this relatively flat surface.

then I just hung th R1 ring on the barrel, with the 3 flashes, still triggered by the off camera 800 and the pictures look sharper. Also when I use a slower shutter, seem to be getting a halo. I wonder , as I noticed in some post if the slaving is not synchronous enough.

[rjlittlefield]

Previously I was shooting at 1/3" and this time 1/100" which would seem to make it sharper.

Hmm... I think I should review how flash photography is supposed to work.

The idea behind flash photography is to illuminate the subject only by flash. In that case the effective exposure time is the duration of the flash. That duration is around 1/1000 second at full power, and proportionally shorter at lower powers. The mechanical shutter speed is irrelevant as long as (1) it's long enough that the whole sensor is exposed when the flash goes off, and (2) it's short enough that ambient light makes no significant contribution to the overall exposure.

Every camera has a certain shortest shutter time that meets criterion (1). Using that time also gives the best chance of meeting criterion (2). Check the manual for your camera. It's probably around 1/200 or 1/250 of a second. That's the shutter speed you should be using for now. (Later we might want to get into second-curtain flash with a longer shutter time, but at this point that's an unnecessary complication.)

In most cases, the short effective exposure time with flash photography will "freeze out" whatever vibration is present. This may not happen if the vibration is unusually fast or intense, but that's unusual.

I was using live view this time.

Again, there are devils in the details. On the Nikons that I know, shooting in live view is the very worst thing you can do from the standpoint of vibration. During live view the mirror is up and the shutter is open. When it's time to take an exposure: the shutter closes, the mirror drops, the mirror rises again, and the shutter opens, all of this occurring just before the actual exposure. This is inviting trouble.

It is far better to turn off live view while shooting, and use mirror lockup to give yourself several seconds delay between mirror up and start of exposure. That way the only vibration-inducing event that affects the exposure is the opening of the shutter. Again, check your camera manual for details. Probably mirror lockup is operated by two presses of the shutter release. If that's the case, then you'll need to program your StackShot controller accordingly, for two shutter pulses per step, with an appropriate delay between them.

Could its firing caused vibration?

Firing of stobes can cause vibration, but given what I'm hearing about the rest of your setup, I'm guessing strobe vibration is the least of your concerns.

What I think happened is that you had your camera set to get a lot of its exposure from ambient light, and then all the mirror and shutter shocks that resulted from shooting in live view caused the camera to be moving during your entire 1/100 second exposure. In this situation 1/100 second is actually less sharp than 1/3 second because a larger fraction of the exposure happens right after the shocks.

so I don't know how much area should be in focus though.

It doesn't matter how much area is in focus -- that's what stacking takes care of. What matters is how sharp the images are, at whatever places are in focus.

Learn to shoot sharp first, then use stacking to put the sharp bits together. You should never shoot and process a stack without confirming that you can get sharp individual shots.

--Rik

[robirdman]

Actually, when I started the stack shoot, the flash wasn't going off, and it appeared not to do so with live view on, so I restarted with it off. I also turned off the light in the room just before the stack started as I thought the fluorescent light behind was creating a bit of a bright spot.

The previous time with the 1/3 I had the light on and that extra light was adding enough to use the F32 which otherwise was too dark at higher speeds.

Normally in the field with birds, I use FP so it shoots at any speed. With insects, it seems the R1 slave doesn't sync right about 1/250.

I usually use rear curtain and it was set thus in each stack series

I noticed when I went back down that I had pushed the tripod so it was actually touching the desk with the subject, so maybe that is why.

I seem to have trouble finding much sharp at the large apertures.

I'm definitely having more of a problem lighting without the R1s.

I just checked the other series that I did with the same setup but the other closeup lens which gave a 20mm field, with a smaller beetle on a triangle. I had 94 images at f8 and the result is just as bad, so I'll delete the whole day's work.

[NikonUser]

As I recommended f/8 I thought I should try and see what my 200mm Nikon micro does at f/8.
D600 200mm at closest focus (1:1 mag), single SB900, styrofoam cup diffuser; 0.2mm steps for beetle increased to 2 mm for pin shaft to get pin head in focus.
57 frames, ZSMax
original is horizontal, rotated here

I would never use this setup for such a beetle, likely use a reversed enlarger lens or a 105mm Nikon Micro

Quick and dirty setup; full frame and then a crop for actual pixels.


NU140108

[robirdman]

So do you use 2 series, to allow for the pin and merge them? Thanks for the followup.

I have a 60mm macro. I forgot that I don't need to use the 200mm, since I'm not using the 4x microscope objective I got for it.

[robirdman]

I tried the 2nd beetle again today, with f11, 1" ISO 64. I took some text focus shots, and then I found a 1/2" layer of styrofoam that I wrapped around a translucent plastic container that Chinese soup came in. This seemed to get rid of most of the reflections. I used steps of .067mm.

I thought the beetle initially looked cleaner but sitting overnight maybe picked up even more debris, which is another problem of how to clean such.

Again, CS6 , which had done a pretty good job on the first beetle the first time with F32 produced a totally useless image.

These are pretty awful too because of the debris, butthe main thing I am wondering about is the comparison of the focus practice shot, without such diffusion over the stack. though less is in focus, what is seems much sharper. I am not improving over my initial results, but doing worse, which is frustrating.

[ChrisR]

which is frustrating.

Not sure I'd be only frustrated.
I can't think of anything which can be consistently wrong, other than a faulty lens. What magnification are the last ones?
I'd stick to flash while proving things, just to eliminate vibration. At say f/11 -f/16 marked (may report as f/22-32 on that lens, I'm not sure if it's one which does that) you should be able to beat the shaprpness you're showing, without paper-thin dof, easily. Even if you don't stack, there should be SOMETHING sharp!
A flat subject may make testing-time easier too. Graph paper with scribbled ink perhaps.

[ChrisR]

Had to try...
I don't have a 200mm micro, this is a 180mm zoom micro at closest, 44mm field width cropped down to about 28mm. Marked f/22 so someting about f/50 effective. D700 (24 x 36mm). Single exposure handheld, flash on hotshoe bounced of a piece of paper

This often-spotted Qtip bird has some parts which appear to be a heck of a lot sharper than what you're showing us?

You mentioned a D4, also 20mm field width, which means you're cropping or using something else??

[robirdman]

On the 200 macro was a 6T closeup lens on this beetle. On the previous Bombardier was the 5T, both on the awful shots and the previous better ones.

The aperture was from the readout, not the lens barrel.

Perhaps the subject position is not stable enough.

[robirdman]

Here are 4 different processing of the first stack I had done before with the 5T. That was with the F32 and 1/3", same stack as previously posted with different processing softwares.




They were with the 3 R1's though, not the SB800 with diffuser tunnel.

[Chris S.]

Robirdman,

Another Chris, here.

I think you should stop stacking--temporarily--and concentrate on getting a single shot that is very sharp within its limited zone of focus. This is a much quicker way to find and eliminate problems you may be having, rather than doing stacks of what may well be flawed source images.

If you are using flash--and only flash--you should be able to get a sharp (in a single plane) image even if your camera or subject are not perfectly steady. But for this kind of shooting, it's important to make sure that no other light (from room lighting or window lighting) is contributing to the exposure. If it is, you have a recipe for trouble. To test this, turn off your flash and take a picture just like one from some of the stacks you showed--same setup, same room conditions, 1-second exposure, f/11. The result you want will be a completely black image with absolutely nothing recorded on it. If you see anything but pure black, you have room lighting mixing with your flash lighting. If so, make your shooting environment much darker and try again until you get pure black.

Now try a single picture with flash, focusing on one particular detail that you can examine on the computer screen. Is this detail razor sharp?

If not, I'd suggest removing those close-up adapters and trying another single shot. Sharp yet?

If not, we have to suspect a problem with your lens. I also have a Nikkor 200mm micro f/4 IF-ED, and it's a very sharp "macro" lens. Since you also have a 60mm micro--which I know only by its good reputation--you have a nice basis for comparison. Either lens, if it is working properly and used appropriately, should be sharp enough to cut you at 1x. Forget about f/32--no decent lens is its best at such a small (diffraction producing) aperture. On the 200mm lens, I'd guess your best results at 1x will be around f/8 or f/11, as reported by the readout on your camera. (If your lens is sharpest at f/32, it's broken.)

When your workflow is reliably producing very sharp single images, it is time to start shooting stacks. But until this point is reached, stacking only adds confusion.

(And when it does become time for stacking, I highly recommend you download the free trial of Zerene Stacker. The stacking capability of Photoshop CS6 is, by comparison, quite primitive.)

Please let us know the results of these tests!

Cheers,

--Chris S.

[robirdman]

My first 3 comparison images were with Zerene, Helicon and Photoshop. At first I was not able to get a result with CombineZ but later succeeded and posted the same stack with 4 different methods. All 7 seemed to have rather similar results to my untrained eye, and those were with F32, I think.

When I started the bigger stacks with the smaller aperture of 8 or 11, I started getting the bad results.

I tried again today, not the best subject, this Pelidnota, because it is so deep, and I didn't try to get all the legs in focus. But it was cleaner, so I tried it. I thought I improved on the lighting setup somewhat after I finally found things that would position the flash where I wanted it.

The smaller images are with the 60mm macro and are less than 1:1. The first image, I used autofocus and then I backed off slightly to begin the stack that doesn't include that one. this time I did tiffs to avoid the time of converting RAW.

the larger images, I did the same thing with the 200mm macro at about 1:1. First image, autofocus to get starting position, then the stack on manual, and not including that shot.

I viewed with a hoodman, liveview to set the end point near the labrum. I called Nikon to ask about how to view on a computer, and the tech said you can't do that even with camera control, you could only view the shots taken.

When I zoomed the liveview I was surprised to see that it was looking at vibrating atoms, not motionless. Is this normal?

Actually I realized as I was about to upload, that I needed to make jpgs, and cropped, so I added 200mm to the names of those. I see names don't show. 1st 2 are 60mm. then 2 200mm.
I didn't notice that it sorted alphanumerically. The 2 with the more mottled background are the 60mm.
Each was 87 steps at .16 mm.

60mm
200mm
60mm
200mm

[Chris S.]

My first 3 comparison images were with Zerene, Helicon and Photoshop. At first I was not able to get a result with CombineZ but later succeeded and posted the same stack with 4 different methods. All 7 seemed to have rather similar results to my untrained eye, and those were with F32, I think.

Those comparisons probably didn't reveal much about the capabilities of the stacking software, due to issues with the input photographs.

When I started the bigger stacks with the smaller aperture of 8 or 11, I started getting the bad results.

To get top quality results, you need to be shooting at something closer to f/8 or f/11--definitely not f/32. The fact that you're having trouble as these larger apertures points to problems with your lens or your technique.

One the "lens problem" side, f/32 can make a broken lens look better, because you're using the tiny opening in the iris rather like a pinhole. If you've ever done pinhole photography, you know that you don't actually need a lens to take a picture, if you are willing to accept the diffraction fuzz that is inevitable with a tiny aperture. At f/32, this is pretty much what you are doing.

On the "technique problem" side, since f/8 represents a much larger opening in the lens than f/32, much more light gets through. Since you're shooting with flash, this means that at f/8, you are turning down your flash output (or if shooting with some form of TTL, the camera is dialing down the flash output for you). At the same time, if you are getting any leakage of room or window light onto your subject, you are increasing the degree to which this adds to your exposure, since, while flash output may be automatically decreased, room light is not decreased, and with your lens being opened much wider, far more room light can get in.

This could certainly be a problem for you, because, at the slow shutter speeds you are using, continuous light can be a huge source of image softness under any but the most rigorously-controlled circumstances.

Have you done the "flash turned off, is the image black?" test I suggested?

Come to think of it, why are you using shutter speeds as long as one second? Your flash sync should be much higher, 1/250 of a second iirc? What happens if you use a shutter speed closer to your max sync speed, which should minimize the effect of room light? (At this shutter speed, if you see black bands at the top or bottom of your image, your camera and flash are having trouble syncing--in which case, drop to 1/200 or 1/160 second.) Do the fuzziness problems at f/8 go away?

I used autofocus. . . . This time I did tiffs to avoid the time of converting RAW.

Even recognizing that you turned autofocus off after initial focus acquisition, I strongly suggest you avoid autofocus for all further macro shooting. It can create errors, and has very little value at anything near the 1:1 scale. Near 1:1, I would set the lens for the needed field of view, then move the lens/camera to obtain focus.

Tiffs make sense in your situation, as your workflow suggests that raw is giving you little or no value. (Raw can be very useful, if one is tweaking the conversion in good raw conversion software--but without such tweaking, raw has little inherent benefit. If you are willing to accept default raw conversion from a stacking program, there is no real tweaking going on during conversion.)

I called Nikon to ask about how to view on a computer, and the tech said you can't do that even with camera control, you could only view the shots taken.

It appears that somewhere, a misunderstanding has occurred. Nikon Camera Control Pro II will most assuredly show the live view image in real time on your computer screen. You can even zoom in on a detail of your choice to do critical focusing. Then after you take a shot, the software can also display the image just taken. So will ControlMyNikon--which does an even better job for much less money. There are other programs with similar claims, and some of them may be good--but I'm personally familiar with these two.

When I zoomed the liveview I was surprised to see that it was looking at vibrating atoms, not motionless. Is this normal?

Hard to say, without seeing what you are seeing. Particularly in low light, visual noise appears and disappears that might be described as vibrating atoms. If so, this is normal. But if you see your subject moving around, this is not a good thing, if you have any continuous light reaching your subject during the exposure. A flash-only approach, though, might tolerate moderate apparent subject movement.

The fact that you can use live view suggests that some continuous light is indeed hitting your subject, and may be the cause of your lack of sharpness. (Unless you turn off your focusing light prior to shooting your stack.)

Actually I realized as I was about to upload, that I needed to make jpgs, and cropped, so I added 200mm to the names of those. I see names don't show. 1st 2 are 60mm. then 2 200mm.
I didn't notice that it sorted alphanumerically. The 2 with the more mottled background are the 60mm.

I'm confused about what is being demonstrated in each shot. Here are the names of your images in the order displayed (obtainable at my end by quoting your post during reply):

3832_900_9538_1.jpg
3832_900_9613_1.jpg
3832_20140110163737_ZS_DMap200mm_1.jpg
3832_20140110170923_ZS_DMap_200mm_1.jpg

From appearances, are the latter two shots stacked images with the 200mm lens, the earlier two shots perhaps single images? Can you clarify, and explain at what parameters each images was made?

Though I'll say again that at this stage, stacking probably wastes your time and ours. Only after you can make razor sharp single images, is it time to stack. Likewise, though your efforts at more pleasing light are both admirable and sound, it's probably out of order to worry about this until you can obtain sharp stacks. As someone once told me, "Don't try to boil the ocean." In other words, I was being told to solve one problem at a time.

Each was 87 steps at 1.6 mm.

An increment of 1.6m is pretty large--perhaps appropriate to shooting at f/32, but not for larger apertures such as f/8. So are these more shots at f/32? To repeat, if you are shooting at f/32 and unhappy with the level of sharpness, stop shooting at f/32.

But something is strange, here. If you took 87 shots and moved the camera/lens 1.6mm between each shot, you should have moved the camera a total of 139.2mm--or about 5.5 inches! From what I see, no way did your camera move that far.

--Chris

[robirdman]

Sorry for the confusion. I was confused because I didn't notice my images were resorted when the software renamed the stacks.
Top image was 60mm at f11. I trusted autofocus to get the top in focus. In the earlier series I had used manual to start and since they were determined to be unsharp I thought this might be more dependable. The autofocus shot was not included in the stack, just used as comparison single shot.
2nd image was 200mm at f11. Likewise all the other procedure.

3rd image is stack of 60mm f11. The stack was started with the starting point backing off slightly from the autofocus point and camera was switched to manual. Liveview was used with the rail to set the end point near the labrum. When I zoom in, what I see isn't a shaking but the pixels seem to be a seething mass. Because of the fluorescent illumination? Never noticed before. The room light was turned off before each shot, and without the flash I would get a black picture.

The 4th shot is stack 200mm f11. Ditto everything for the 60mm stack. I used the slower shutter speed, as Rik said my 1/3" exposures in my first test, would ironically be sharper than my 1/100' in the 2nd blurry series.

It did not seem to make sense, what the Nikon help tech told me. There is an extreme variance in knowledge, when calling, I've found. Once with a previous problem I was told it was due to using an SD card with the D4, and that couldn't be used. When I checked to see what I could use, that was the only one that could be, confirmed by a 2nd tech.

[robirdman]

Sorry for the confusion. I was confused because I didn't notice my images were resorted when the software renamed the stacks.
Top image was 60mm at f11. I trusted autofocus to get the top in focus. In the earlier series I had used manual to start and since they were determined to be unsharp I thought this might be more dependable. The autofocus shot was not included in the stack, just used as comparison single shot.
2nd image was 200mm at f11. Likewise all the other procedure.

3rd image is stack of 60mm f11. The stack was started with the starting point backing off slightly from the autofocus point and camera was switched to manual. Liveview was used with the rail to set the end point near the labrum. When I zoom in, what I see isn't a shaking but the pixels seem to be a seething mass. Because of the fluorescent illumination? Never noticed before. The room light was turned off before each shot, and without the flash I would get a black picture.

The 4th shot is stack 200mm f11. Ditto everything for the 60mm stack. I used the slower shutter speed, as Rik said my 1/3" exposures in my first test, would ironically be sharper than my 1/100' in the 2nd blurry series.

It did not seem to make sense, what the Nikon help tech told me. There is an extreme variance in knowledge, when calling, I've found. Once with a previous problem I was told it was due to using an SD card with the D4, and that couldn't be used. When I checked to see what I could use, that was the only one that could be, confirmed by a 2nd tech.