Serious Help Needed

[PhilH]

The Spider here is obviously dead so I took the opportunity to do a little practising.
I used several settings and these were a couple of the results.
What is dissapointing is that the best image (IMO) was taken on auto.

No.1. f/22.0 - Shutter 2.5 - ISO 400 - Focal Length 60MM - No flash



No.2 taken using the cameras (Canon 350D) auto macro setting. f/4 - Shutter 1/60 - ISO 400 - Focal Length 60mm - On board flash. I know the DOF isn't as good but the overall quality is far superior.
What am I doing wrong?

[Ken Ramos]

Most of the settings I use are between f/5.6 and f/11 at 1/200 sec. ISO 100 or 400 and I use the camera in the manual mode, stay away from auto . These settings I have found to be the most useful, though that is not to say they are set in gold. A lot of things can be considered at times to depend on what you set your camera at but until you get used to yours these may work for you in the manual mode, in the mean time.

Your camera also has a series of auto focus sensors visible in the view finder, turn all them off except for the center sensor, your manual will explain and try not to use auto focus, pressing the shutter button halfway will cause the center sensor to beep and light up at the correct focal point as you manually focus your camera. Also you can set your cameras shooting paremeters by going into the cameras menu. Again your book will explain how to do this but try setting the contrast paremeter to the next lowest setting below the default setting.

This should give you better results and I am sure there are others who can advise you of other things to do, much better than I. Hope this helps you out Phil.

[PhilH]

Thanks Ken. I have set the camera's focus sensors before but not on this occasion.
I really want to stay away from auto but I found these results very dissapointing. The manual settings you see above are the best from several tries.
Grrrrr. How hard is this photography larck

BTW. The little chap above is only about 4mm (3/16 inch) in size.

[Ken Ramos]

Oh and depending on the light, you may or maynot want to use a flash at those settings I wrote. I once had a 350D and got good results from it, it was a very nice camera but a bit small and light for me.

[Bruce Williams]

Hi Phil and welcome to the forum.

A spider that small makes quite a challenging subject. I assume from the very slow shutter speed on shot1 that you had the camera mounted on a tripod?

Hmmm...well I actually like shot1 (without flash) better than the second shot using flash. It shows more detail on my monitor and IMO has a generally more pleasing and natural look - to my eye anyway.

Look forward to seeing your next posting.

Bruce

[rjlittlefield]

Grrrrr. How hard is this photography larck

Sorry to say, it can be pretty hard for stuff this small. But hey, if it were easy, it wouldn't be any fun, right?

Now, about the pictures. We need a few more details. How much of the full frame are we looking at here? Are these images full frame width resized to 750 pixels wide, or cropped so that we're seeing the camera's pixels, or something in between? And when you say "overall quality is far superior", can you say more precisely what aspects you're looking at?

What I notice in the second picture is that the color balance is different, colors are more saturated, contrast is higher, and subject features exactly in the focal plane are sharper.

The first three of these -- color balance, saturation, and contrast -- could well be due to the different lighting.

The last one -- better sharpness in the focal plane with a wider aperture -- could be due to diffraction effects. A marked f/22 at 1:1 on your camera is definitely getting into the area where diffraction can cause problems.

--Rik

[PhilH]

Rik, what you're looking at are staright from the cam. I have made no adjustments in any way apart from cropping. The pic below is the full version of pic 2 above. The manual image is the same size.

[Mike B in OKlahoma]

No.1. f/22.0 - Shutter 2.5 - ISO 400 - Focal Length 60MM - No flash

Do I understand that your shutter speed was 2.5 seconds for the first shot?! If so, that's the problem.

My advice would be to keep your shutter speed down! :-)

[rjlittlefield]

Thanks, Phil. This tells me what I needed to know about the cropping.

I'm fairly sure that the softness in your picture #1 is due to diffraction. Very small apertures are guaranteed to do that, and once your aperture gets small enough to see the effect, it gets worse in direct proportion to the f-number. You can compensate quite a bit for that softness by sharpening in post-processing. I do that routinely. For example the aphid predators that I just posted, got sharpened by Photoshop unsharp mask at about 70% with 1 pixel radius (working on the camera resolution images).

I'm not so concerned about the long exposure time as Mike is, but it is something to think about. If your setup is dead stable, then you can expose as long as you like. Before I got my nice bright halogen illuminator, I routinely used several second exposure times with small apertures and basic photoflood lighting. But if there's any vibration, the image can get a little fuzzy, and if you're really unlucky, certain patterns of vibration can look just like diffraction blur! (I know this, of course, because it's happened to me. )

Again, I recommend to run test sequences isolating each aspect as much as possible. Make sure there's no vibration, set up the lighting and don't touch it, and run a sequence stepping aperture from wide open to full closed, adjusting exposure time to keep the image brightness constant.

You ought to get a sequence like what's summarized at http://www.janrik.net/insects/ExtendedD ... deoff.html. Chances are that full open will be fuzzy from aberrations, someplace around f/8 will be as sharp as it gets, and from there down the images will first lose sharpness and then get frankly fuzzy.

Once you do that, you'll have a good start on understanding the effect of aperture.

Then you can start to work on lighting, which is an art form of its own!

Hang in there -- you're doing fine!

--Rik

[DaveW]

Sorry Ken,

What is the point of switching off autofocus then using focus confirmation? Focus confirmation is linked to the autofocus circuitry so if that is inaccurate close-up then so will be focus confirmation using manual focus.

I completely ignore focus confirmation and just rely on focusing the image on the focusing screen. If you are using focus confirmation close-up you might as well leave autofocus on!

DaveW

[rjlittlefield]

What is the point of switching off autofocus then using focus confirmation?

I regularly use focus confirmation with autofocus off. The reason is that given the viewfinder and focusing screen of my Canon 300D, focus confirmation can do a better job than my eye can, of telling exactly when best focus is reached.

It's annoying that focus confirmation only works with AF-capable lenses. I'd really prefer to have it available even with bellows setups, but for some unknown reason, Canon disables it except with AF lenses.

--Rik

[DaveW]

But Rik it's the same sensor you consider too unreliable to work the autofocus! It was never intended to replace autofocus with autofocus lenses as it is no more reliable and part of the same system. The sensor also lights when your autofocus locks on to the subject, so it is obvious they are one and the same circuit. You and Ken then may as well leave your lenses on autofocus all the time if you rely on the confirmation light rather than the screen image.

DaveW

[rjlittlefield]

1. Autofocus often goes crazy hunting. Focus confirmation with manual focusing does not.

2. Where did I say that I "rely on the confirmation light rather than the screen image"? It's another piece of information. I use 'em all.

The confirmation light tells me quite precisely what's happening in that one little area that it's checking. The viewfinder image tells me about everything else, and I use it a lot.

--Rik

[MacroLuv]

... You and Ken then may as well leave your lenses on autofocus all the time if you rely on the confirmation light rather than the screen image.

DaveW

It would be nice to have so much clever lenses, knowing what you have in mind and which autofocus wouldn't easily escape from desired place to some more near or more distant point. In that case this approach could be beneficial.

[rjlittlefield]

Add a couple more reasons why focus confirmation is more useful than autofocus for macro:

3. Autofocus happily changes the magnification, making it rather more difficult to know how long the scale bars should be.

4. Autofocus gets very twitchy (or doesn't work at all) when the lens runs up against the end of its focusing range. Of course that occurs at maximum magnification -- not a happy loss.

--Rik