Bridge, MFT, achromats, dSLR, primes - a journey of exploration

Thanks Nick. That DPreview thread is nothing short of an epic! Will take some time to read through it though, looks fascinating.

I'll have another go at the sharpening on the flower shot - I've never really got the hang of how much sharpening to add (I use Photoshop Elements). After a while all that seems to happen is you gets lots of noise - which I am not fond of.

More to come in due course :)
 
Thanks Nick. That DPreview thread is nothing short of an epic! Will take some time to read through it though, looks fascinating.

I've certainly found it very interesting and useful. Being in a forum for small sensor Panasonic cameras there is obviously a bit of an emphasis that way in terms of the examples we are working on, but in fact most of what we are talking about there and experimenting with has much wider applicability. Arguably it is the wrong forum to be discussing those things, but there is a good spirit of enthusiasm and cooperation in that forum and I think that is what is making it work.

I'll have another go at the sharpening on the flower shot - I've never really got the hang of how much sharpening to add (I use Photoshop Elements). After a while all that seems to happen is you gets lots of noise - which I am not fond of.

You might find that flower images respond better to contrast/microcontrast adjustment rather than edge sharpening. Here is a version of the image you posted (I had to use a screen grab because I couldn't download from Flickr) where I have applied defogging/clarity. I did this in Elements using Unsharp Mask. For edge sharpening you use a large Amount, a small Radius and a small Threshold. For defogging you use a small Amount, a large Radius and zero Threshold. The numbers you may see quoted on the net are Amount 20%, Radius 60 pixels, Threshold 0. When I was using this (when I was using CS2) I would most often use Amount between 7 and 13 and Radius 30, Theshold 0. In this case I have taken a rather less subtle approach so as to make it clearer what sort of effect it produces. Obviously it is like everything else in post processing a matter of "season to taste".

Hopefully you will see this flip between two versions.


NOT MY IMAGE - Mike flower - PSE Defog
by gardenersassistant, on Flickr

More to come in due course:)

Good. :)
 
Excellent post Nick! Thanks very much indeed. Very interesting to see the two next to each other.

Sharpening is yet another area of processing that I understand very poorly so this is really useful for me. I'm hoping to get some macro shots today and will put the method to use later :)
 
Excellent post Nick! Thanks very much indeed. Very interesting to see the two next to each other.

Sharpening is yet another area of processing that I understand very poorly so this is really useful for me. I'm hoping to get some macro shots today and will put the method to use later :)

I generally use edge sharpening, quite a lot of it, for invertebrates. If you are shooting raw, and since you are using Elements, presumably you are loading the raw files into Elements by way of Adobe Camera Raw. If you are you might want to go to the sharpening/noise reduction tab and set the sharpening to something like Amount 100%, Radius 1 pixel, Mask 50% to 60% or so. If you hold down the Alt key while using the Mask slider you will be able to see what will be sharpened - in white - and what the mask will protect from sharpening - in black, and this masking will reduce the amount of noise caused by sharpening in the plain areas in the background.

If you are shooting JPEG and going straight into Elements then you might want to try Adjust Sharpness, Amount 100%, Radius 1 pixel. There is no Mask facility so the background will be sharpened too and is at risk of accumulating some noise, which is a good reason IMO for shooting raw if you are using Elements.

That is all for invertebrates. For flowers you might want to try turning the Amount down to somewhere between 25% and 60%, depending on taste and depending on the image.

I'm not suggesting these as hard and fast "rules", just something you might like to play with. Apart from anything else it depends on your camera and the sort of files (JPEG or raw) that it produces and how they respond to processing of various sorts.
 

Spider
by Mike Smith, on Flickr


Fly
by Mike Smith, on Flickr

Hmm. Got these today with the RX10iii and Raynox150 - slim pickings up here (Edinburgh) due to cool conditions but I think they are enough to tell me what I need to know... not sharp enough :(

To be expected really but still a bit of a disappointment. So My 760D + 100mm rig will live on until Fuji release a proper macro lens (80mm in the pipeline supposedly) which I can mount on the XT-1. Not a disaster but it would have been nice to consolidate!

I tried the sharpening method you suggested Nick - by heck it works well! That masking thing is witchcraft! Can't believe I didn't know about it already...

Thanks again for the help :)
 
Last edited:

Spider
by Mike Smith, on Flickr


Fly
by Mike Smith, on Flickr

Hmm. Got these today with the RX10iii and Raynox150 - slim pickings up here (Edinburgh) due to cool conditions but I think they are enough to tell me what I need to know... not sharp enough :(

To be expected really but still a bit of a disappointment. So My 760D + 100mm rig will live on until Fuji release a proper macro lens (80mm in the pipeline supposedly) which I can mount on the XT-1. Not a disaster but it would have been nice to consolidate!

I tried the sharpening method you suggested Nick - by heck it works well! That masking thing is witchcraft! Can't believe I didn't know about it already...

Thanks again for the help :)

Don't know that that is what I would have expected. In fact, I'm surprised. I use the 150 almost all the time on the FZ200. The RX10iii should do as well in terms of sharpness at f/16, and better than the FZ200 at larger apertures. How were you focusing Mike? Were these using flash? What was the shutter speed? I wouldn't be inclined to give up on it just yet.
 
I applaud your tenacity Nick :)

I was focusing manually, 1/200, f16, iso100 - with the Mt24 on manual mode at 1/8th I think.

Shutter speed maybe too low? I could try a bit less aperture I guess but doubt diffraction is to blame...

No, not diffraction. I shoot that equivalent aperture almost all the time with invertebrates. It sharpens up ok for the size of output we're looking at here.

Shutter speed to low? Doubtful. If flash was the main source of light, then definitely not, because the flash duration would be the effective shutter speed. If it was sunny and there was a mixture of flash and natural light - 1/200 - well, I suppose it is possible. It really doesn't seem likely though. 1/200 is quite fast and even with the 150 at 600mm equivalent on a 1" sensor we're not talking huge magnification.

Hmm..... What is the sharpness like for ordinary (non close-up) shots? Have you got any with fine detail?

Could it be a post processing issue? Here are reworks. Usual caveats, personal taste etc etc, and I had to work from the screen scrapes.


NOT MY IMAGE - Mike spider
by gardenersassistant, on Flickr


NOT MY IMAGE - Mike fly
by gardenersassistant, on Flickr
 
Hi Nick. Thanks again. Did you sharpen these as per the method described as above? It's definitely an improvement but still not in the same league as what I can get with the 100mm prime (and my 100mm shots are processed with my random sharpening approach - so the unprocessed shots must be miles better!)

The lens itself is sharp - absolutely no doubt about it. I'll post something suitable up to demonstrate :)
 
Hi Nick. Thanks again. Did you sharpen these as per the method described as above?

Partly. I used the approach that I use for my own invertebrate images (although of course I'm starting from raw in that case rather than from JPEG in this case).

I first used DXO Optics Pro to do two things of particular relevance here - (1) microcontrast enhancement and (2) lens-specific "lens softness correction". I then sharpened in Lightroom, for which I did use settings like those I suggested to use in ACR. (Lightroom and ACR are exactly the same in this respect. Lightroom has all the sliders that the version of ACR in Elements has, and they all do the same things. If you set the sliders to the same numbers in ACR and in Lightroom you get the same result.)

It's definitely an improvement but still not in the same league as what I can get with the 100mm prime (and my 100mm shots are processed with my random sharpening approach - so the unprocessed shots must be miles better!)

The lens itself is sharp - absolutely no doubt about it.

Hmm... That is puzzling. I wonder what the problem is.

I'll post something suitable up to demonstrate :)

You've got me interested now!
 
Interesting about the sharpening I use a two stage method
In Lightroom on the raw I set sharpening to 37%. detail to 15 think its called radius at 1.2 and masking to 40 this is presharpening
I would rather slightly under sharpen than over sharpen and try to get the shot when I take it as sharp as I can
Then in Photoshop just before converting to jpeg the last thing I do is smart sharpen again just slightly at 65% and a radius of 1 pixel
I don't really understand the process of sharpening but found by trial and error that this seems to work well :)
 
Interesting about the sharpening I use a two stage method
In Lightroom on the raw I set sharpening to 37%. detail to 15 think its called radius at 1.2 and masking to 40 this is presharpening
I would rather slightly under sharpen than over sharpen and try to get the shot when I take it as sharp as I can
Then in Photoshop just before converting to jpeg the last thing I do is smart sharpen again just slightly at 65% and a radius of 1 pixel
I don't really understand the process of sharpening but found by trial and error that this seems to work well :)

Interesting. I use a two stage process as well (for edge sharpening - that is after the microcontrast enhancement and lens-specific deconvolution deblurring in DXO). In Lightroom I use presharpening Amount 66% at 1 pixel masking 50-60 as a default but for invertebrates often up that to Amount 100%. For the second stage I use Lightroom Output sharpening, Sharpen for screen, Amount standard.
 
Interesting. I use a two stage process as well (for edge sharpening - that is after the microcontrast enhancement and lens-specific deconvolution deblurring in DXO). In Lightroom I use presharpening Amount 66% at 1 pixel masking 50-60 as a default but for invertebrates often up that to Amount 100%. For the second stage I use Lightroom Output sharpening, Sharpen for screen, Amount standard.
Yes sounds very similar to what I do :)
Apart from sometimes when I do a bit of simple focus stacking of a couple of shots I normally just convert the raw sharpen send to photoshop and sharpen again then convert to a jpeg :)
 
Couple of RX10iii (non-macro) samples which I hope will demonstrate the sharpness of the lens :)

Both shot at nearly full reach with the lens wide open so I should be able to do better stopped down a bit (and maybe also with less focal length)

Be interested in your thoughts Nick...

Hmmm..... Don't know. What are we looking at here Mike - Out of the camera JPEGs, processed raw files, processed JPEGs? Any cropping? In any case what we see is a combination of the lens and processing; I can't separate out these two factors.

As far as the bird goes, I'm afraid I'm not a good judge of bird images because I am so unfamiliar with them. I don't know how much detail I should expect to see, or how well-defined the plumage should be.

For the butterfly, which I am more familiar with, the DOF is very narrow, which doesn't leave much available for assessment. From what I can see, looking at the largest size over at Flickr, I wonder if the head is a little "sparkly". If so, would this be the lens, the illumination, or the processing?

Sorry if this sounds unhelpful, but I'm struggling a bit here I'm afraid. I suppose assessing the lens sharpness would be easier using some sort of "like for like" comparison (although quite possibly that would be not at all easy to come by).

I keep coming back to the thought that, because small aperture diffraction is a great equaliser, at f/16 with the RX10iii (which is where we started with this, with the spider and the fly) you should be getting pretty much as good as with minimum aperture on any other system and lens. Are you comparing those images with what you would expect at minimum aperture with the 760D and 100mm? If not, it isn't a fair comparison - I would expect the 760D and 100mm to be sharper at less than minimum aperture than the RX10iii (or anything else) at minimum aperture.
 
You're never unhelpful Nick!

They are processed RAW files with sharpening added and a modest crop.

I am comparing them with my 'normal' macro which is around f11-F14 on apsc sized sensor so actually not very fair as you suggest (especially once effective aperture is added to the mix?). The f4 on the butterfly was a mistake... Should have been shooting at around f5.6 - f8 but got too excited! Oops.

I'm no birding expert but the sparrow is sharp - lots of feather detail.

So maybe il will try some shots around f8 and see how they look as a next step.... At the moment my 100mm setup is miles ahead though looking back at old shots.

Thanks again, Mike.
 
You're never unhelpful Nick!

It is a pleasure Mike. I love thinking through examples and talking about these issues. But by now I think you know that. :)

They are processed RAW files with sharpening added and a modest crop.

I don't know if you have a "standard" amount of sharpening that you apply to all your images, but if you do bear in mind that some camera's images can take (or need) more sharpening than others. That said, I don't think I'd want to sharpen the butterfly image any more, or possibly a bit less because of the "sparkly" look when you look closely (although if you can't see it at normal viewing size it doesn't matter). So it's probably not a sharpening issue.

Actually the butterfly image is one that might respond better to more microcontrast enhancement rather than edge sharpening.

I am comparing them with my 'normal' macro which is around f11-F14 on apsc sized sensor so actually not very fair as you suggest (especially once effective aperture is added to the mix?).

If you are using a close-up lens on the RX10iii and a macro lens on the APS-C then effective aperture would reduce the difference in diffraction losses as magnification increases, since the macro lens suffers from a reduction in effective aperture as magnification increases but the close-up lens does not.

For example, If you were operating at 1:1 on the APS-C then a nominal f/11 - f/14 would become effective f/22 - f/28, which is equivalent to nominal (and effective) f/16 on 1".

The f4 on the butterfly was a mistake... Should have been shooting at around f5.6 - f8 but got too excited! Oops.

I'm no birding expert but the sparrow is sharp - lots of feather detail.

So maybe il will try some shots around f8 and see how they look as a next step....

f/11 on Canon APS-C is equivalent to f/6.5 on 1", and f/16 is equivalent to f/9.4 (the Cambridge in colour calculator that I use - the second calculator on this page - only does whole stops), so f/8 on the RX10iii seems about right for comparison with f/11 - /f/14 on APS-C.

At the moment my 100mm setup is miles ahead though looking back at old shots.

I'd expect that increasingly as you move away from minimum aperture. (Remember that the only reason the small 1/2.3" sensor works so well for me is that I use it for subjects where I use minimum aperture almost all the time so as to get lots of DOF, and if I use equivalent apertures on other systems it reduces their resolution to about the same as I get with the FZ200. Move away from minimum aperture and they become increasingly unequal - larger sensor better - especially with a macro lens which can resolve lots of detail, which as far as I understand it is the case with pretty much all of them.)
 
Right - another attempt tonight... still not a lot around here in the way of insects to shoot but a few decent sized flies as below. Looking a lot more promising I would say! Basically the same technique so I'm not sure why the last ones were so poor.

I sharpened the RAW file in ACR as you suggested then used the 'dehaze' function in photoshop elements to bump up the micro-contrast - worked ok I think?


Sony RX10iii Macro
by Mike Smith, on Flickr

Peek-a-Boo by Mike Smith, on Flickr
 
Last edited:
Right - another attempt tonight... still not a lot around here in the way of insects to shoot but a few decent sized flies as below. Looking a lot more promising I would say! Basically the same technique so I'm not sure why the last ones were so poor.

I sharpened the RAW file in ACR as you suggested then used the 'dehaze' function in photoshop elements to bump up the micro-contrast - worked ok I think?

I think so too. :)

In fact, rather tasty detail in those eyes.
 
Hi Nick - will post this in here as my 'Odds and Ends' thread seems to be gathering nothing but tumbleweed!

Got my 72mm Marumi Achromat +5 through now and am using it with the RX10iii - results are promising so far.

Top shot below was with RX10iii + Marumi + MT24 flash - f14 + 1/2000 and focal length around 200mm equivalent (no vignetting now so no need to shoot at 600mm!). Am using autofocus now as per your comments the other day - works fine.

For contrast the lower one is with a Canon DSLR + 100mm - the quality looks very similar to me... any thoughts ?


Bee
by Mike Smith, on Flickr


Bee - Bombus lucorum
by Mike Smith, on Flickr
 
Hi Nick - will post this in here as my 'Odds and Ends' thread seems to be gathering nothing but tumbleweed!

Got my 72mm Marumi Achromat +5 through now and am using it with the RX10iii - results are promising so far.

Top shot below was with RX10iii + Marumi + MT24 flash - f14 + 1/2000 and focal length around 200mm equivalent (no vignetting now so no need to shoot at 600mm!). Am using autofocus now as per your comments the other day - works fine.

For contrast the lower one is with a Canon DSLR + 100mm - the quality looks very similar to me... any thoughts ?

Caveats: Different subject may have different characteristics. The capture illumination and the processing may have been different. I can't tell what the effective aperture is for the 100mm shot (I know the nominal aperture was f/11, but I don't know what the magnification was), so I can't take diffraction effects into account.

However, that said ..... the bee's fur in the RX10iii image looks to have better microcontrast to me and the hair/fur ends that are contrasted against the background look sharper to me in the RX10iii image. The antennae look better defined too.

Translation. The RX10iii image looks better to me.
 
Caveats: Different subject may have different characteristics. The capture illumination and the processing may have been different. I can't tell what the effective aperture is for the 100mm shot (I know the nominal aperture was f/11, but I don't know what the magnification was), so I can't take diffraction effects into account.

However, that said ..... the bee's fur in the RX10iii image looks to have better microcontrast to me and the hair/fur ends that are contrasted against the background look sharper to me in the RX10iii image. The antennae look better defined too.

Translation. The RX10iii image looks better to me.

Thanks Nick. If the Sony + Marumi is up there even in the ballpark with the Canon + 100mm then I am well chuffed as my pursuit of photographic nirvana can go on! (ie one camera and lens to do it all) . The post sharpening will be better as I am following your recommendations now (it was just random before to be honest *hangs head*) but that's by the by really as the basic shots are very good at least to my eye.

So...more test shots needed! And then I need to look at lighting... the MT24 although effective is big, heavy and clunky to use... what about something like a Nissin I40 (with appropriate diffusers) - think I could get it up to the same standard as the MT24 is delivering now? Or maybe just a pipe dream?

Thanks again for all the help :)
 
Thanks Nick. If the Sony + Marumi is up there even in the ballpark with the Canon + 100mm then I am well chuffed as my pursuit of photographic nirvana can go on! (ie one camera and lens to do it all) . The post sharpening will be better as I am following your recommendations now

Did the 100mm bee shot use the old or new sharpening approach? If it used the old approach then that pair wouldn't be a fair comparison.

(it was just random before to be honest *hangs head*) but that's by the by really as the basic shots are very good at least to my eye.

So...more test shots needed!

Indeed so. Preferably matched pairs where possible, or similar-ish pairs otherwise, if possible. I know it's difficult to arrange that with insects though.

And then I need to look at lighting... the MT24 although effective is big, heavy and clunky to use... what about something like a Nissin I40 (with appropriate diffusers) - think I could get it up to the same standard as the MT24 is delivering now? Or maybe just a pipe dream?

Tricky. With the MT24 having two flash heads, and the flash heads being near to the subject, you may have difficulty matching it.
 
Thanks again Nick - yes, the 100mm shot was with the old method so not really fair but I don't mind, the fact that they are even close is good enough :)

Take your point about the flash - need to have a think about that one. May just accept that I have gone as far as I can with my downsizing drive and continue with the RX10iii + Marumi + MT24.
 
I have modified my camera/lens/software mix a bit. When I was last writing here, for close-up/macro work I was shooting exclusively raw, hand-held, using:
  • For medium-sized invertebrates, a Panasonic FZ200 bridge camera with Raynox 150 or less often Raynox 250 close-up lens on an adapter tube with Venus Optics KX800 twin flash with individually diffused flash heads and a separate concave diffuser
  • Only occasionally, for smaller invertebrates, a Panasonic G5 micro four thirds camera with the KX800 and 45-175 lens and Raynox 150+250 stacked, Raynox MSN-202 or Raynox MSN-505
  • For natural light, for flowers and (occasionally) larger invertebrates, a Canon 70D dSLR with 55-250 STM lens with and without a Canon 500D close-up lens.
For processing I was using (probably, I don't remember the exact timing of the changes) batch processing in DXO Optics Pro 10 using the same settings for all images, exported as TIFF files which were imported into Lightroom 6 for image-specific adjustments and creation of 1300 pixel high JPEG images.

For initial triage (rapidly rejecting obviously unusable images) I was using Faststone Image Viewer or, where the raw files needed an initial boost in brightness, Lightroom.

I was using Faststone Image Viewer for moving files around and renaming files.

And now?

I am still shooting exclusively raw. I am mainly shooting hand-held but I have been experimenting with using a steadying stick some of the time, as described in this post.

After 101,000 shutter activations the older of my two FZ200s has died. It is repairable. I was going to do the repair myself (having been pointed to a maintenance manual which explains in detail how to replace the non-functioning part) but I could not source the part in the UK. I decided not to pay for a repair as the camera has been heavily used and would quite probably develop another fault before too long. I therefore considered a replacement.

For some time I have been considering trying my close-up lenses with a bridge camera with a 1 inch sensor (much larger than the 1/2.3" sensor in the FZ200). None of the existing 1 inch bridge cameras is suitable (either because they don't have an articulated screen or, in the case of the Pansonic FZ1000, they don't have a small enough minimum aperture). At the time I was deciding what to do next a replacement of the FZ1000 (which might have a suitably small minimum aperture) had not been announced (it has now). The Nikon 24-500 was the other potentially suitable candidate, but this has been delayed for many months and is now rumoured to be not available until next year.

Readers who have been paying close attention to this thread (if there are any!) may wonder why I would consider either of these cameras, as both have extending lenses which change the distance between close-up lens and subject when zoomed (unlike the FZ200 when fitted with an adapter tube or the G5 with its non-extending 45-175). I have previously said how important this is to me, especially at higher magnifications, the main issue being that finding very small subjects at high magnifications is difficult, and is much easier if you can start with a low magnification to find the subject and then zoom in on it without having to move the camera. I have changed my opinion on this.

My preference for retaining a fixed distance to the subject as magnification changes stemmed from when I previously worked with a tripod almost all the time, in which case, even when using a focus rail, moving the camera back and forth to get to the correct working distance is a nuisance. However, after working more and more hand-held I slowly came to realise that this was less important. I started by using the Raynox 150 directly attached to the FZ200 camera lens. I found that moving the camera when changing the magnification was no problem when working hand-held. And, with the FZ200, attaching the close-up lens directly to the camera lens has the great advantage of radically reducing the amount of vignetting and thus allowing a much greater range of magnifications without needing to change to a different close-up lens. This is true for all of my close-up lenses, not just the 150.

I progressively discovered that I could also do this when working hand-held with my more powerful close-up lenses - in order of power Raynox 250, 150+250 stacked, 202 and 505. I think my technique must have improved because when some time ago I tried using the MSN-202 on an expanding lens I found it pretty much impossible to use. I think I probably wrote about that experience some way back in this thread. Successful use at higher magnifications can be much improved by a coordinated combination of zooming (to change magnification) and movement (to keep the working distance somewhat constant). I haven't got that working smoothly at higher magnifications, but (when I do occasionally do higher magnification stuff) I'm working on it, practising.

Framing the shots and gaining good focus, centred exactly where required, does become progressively more difficult as magnification increases. This is where the idea came from for the steadying stick (based on the pole and stick used respectively by two of my macro heroes, Brian Valentine and Mark Berkery). Like working hand-held it makes it quick and easy to adjust the distance to the subject, and also to move the camera around to track a moving subject. But like working hands-on with a tripod it damps down hand-shake and makes it easier to frame subjects, track moving subjects and hold positions for multiple shots of a stationary subject.

So, no longer being so concerned about keeping the close-up lenses in a fixed position while changing magnification, my options for other cameras and lenses have opened up.

The FZ200 has proved hugely important to me. It works so well for my purposes I would be seriously bothered if I didn't have something as good, or better, at some point in the future. New FZ200s are still available at the moment, but I didn't know in what numbers and how soon stock might run out. I was nervous about waiting for the (hypothesised) FZ1000 replacement or the Nikon 24-500. This left the FZ200 and the newer FZ330, at about twice the price of the FZ200. The FZ330 has the same lens and quite probably (Panasonic has remained silent on this score) the same sensor as the FZ200. So why pay twice as much for the FZ330?

The FZ200 uses contrast detect focusing. This is more accurate but slower than the phase detect focusing used in most dSLRs. And I use autofocus a lot so this matters to me. The FZ330 still uses contrast detect focusing, but it is enhanced with a "Depth from Defocus" (DFD) technology developed by Panasonic which can work out which way and how far to move the focusing by looking at the out of focus areas of an image. It is said to be much faster than normal contrast detect focusing, and is sometimes claimed to be almost as fast as phase detect focusing.

Faster focusing could be very helpful for my close-up/macro work, especially now I am working more hand-held. This is because the faster the focusing and capture is, the less time there is for the subject to move out of focus before the shot is captured. What I didn't know is whether the DFD focusing on the FZ330 was significantly faster than the FZ200, and no one could tell me with certainty whether DFD focusing would work at all with close-up lenses. I decided to buy an FZ330 to find out. What happened next is described in some detail in this thread in the Panasonic Compact Cameras forum at dpreview.

It turns out that for my purposes the FZ330 has several advantages over the FZ200 in terms of usability. For example, when using autofocus, which I do most of the time, the focus box can be smaller and positioned more easily and faster than with the FZ200. The manual focus implementation has been improved, so that for example a magnified area of the screen can be seen as a picture in picture so as to aid with the focusing but retaining the whole-frame context for checking the edges. It is easy to move the area of the whole frame covered by the picture in picture so you are not restricted to a magnified view of the centre of the frame. You can also easily change the magnification used for the picture in picture. The FZ330 has focus peaking, and for natural light working it has zebras, both of which I find helpful sometimes. The FZ330 is more configurable than the FZ200 and I have managed to arrange easy (button, wheel, lever) access to just about everything I typically need. The FZ330 is also water and dust sealed, which makes it practical to place it briefly on damp ground while I need both hands to find/adjust other things.

Oh, and the focusing probably is faster too, although that is difficult to pin down given the inherent variability involved in close-up/macro shooting. Whether or not that is the case my early experience is of some quite good success rates and pleasing image quality. There are a few examples in this post in the dpreview thread, which also contains links to five albums at Flickr containing around 250 examples using the FZ330 and close-up lenses. The bee images in this post in this forum were FZ330 images, and I will be posting more in the forum of course.

My 70D died as well, a little before the FZ200. I had, unusually, taken out an extended warranty on the 70D, and it died within the three year warranty period. I therefore had it mended under warranty. (It needed a new mirror assembly.)

My preferred hardware mix is now looking like this:
  • For medium-sized and small invertebrates, a Panasonic FZ330 bridge camera with Raynox 150, 250, 150+250, 202 or 505 close-up lens attached directly to the camera lens, with Venus Optics KX800 twin flash with individually diffused flash heads and a separate concave diffuser, used hand-held or with a steadying stick
  • When using the FZ330, for subjects of different size for what I have the FZ330 rigged for, either the FZ200 or 70D, depending on the nature of the subjects I will be looking for/hoping to encounter.
  • For natural light, for flowers and (occasionally) larger invertebrates, 70D with 55-250 STM lens with and without a Canon 500D close-up lens, used hand-held.
  • For very early morning still air slow exposure natural light shots (I haven't had any of these sessions this year), either the FZ330 or 70D with Raynox 150 or 250, or possibly Canon 500D, using a hands-off tripod approach with cable remote shutter release.
In terms of software, I have had to upgrade to DXO Optics Pro 11 in order to process native FZ330 raw files. I am now exporting from DXO as DNG rather than TIFF because this enables the receiving application to deal with the white balance and (I haven't been able to pin this down) DNG may have another advantage in terms of dynamic range retention. After a lot of experimentation, for close-up/macro I am now importing the DNG files from DXO into Silkypix Developer Pro 7 rather than Lightroom 6 for image-specific adjustments. I find I can reveal more fine detail using Silkypix. I also like the way it deals with colour retention in highlights and prefer the overall look I get with it compared to Lightroom. This isn't always the case though and I do sometimes use Lightroom, although this is more often for non-close-up/macro images.

Rather than exporting from Lightroom as JPEG at my target 1300 pixels high, I am now exporting from Silkypix in TIFF format in case I want to make further adjustments in another editor. One example of this is using Dfine from the now free (Google-owned) NIK collection for easy and rapid reduction of noise/fine-scale artefacts that my agressive use of Silkypix sharpening sometimes creates in plain backgrounds. Cloning in Silkypix is only effective for very small areas and so I use my old copy of Photoshop (CS2) for any more extensive cloning (although I don't do much cloning these days). I occasionally use graduated filters in Lightroom, which handles this better, with more flexibility than Silkypix. I might very occasionally use other editors on the TIFF files.

When I have finished adjusting the TIFF files (which is most often immediately after they come out of Silkypix) I produce 1300 pixel high JPEGs using Faststone Image Viewer, which I also use for renaming files and moving files around.

For initial triage I am still using Faststone Image Viewer whenever possible. That is because Faststone deals with the embedded JPEG files and this makes it really fast to get through the large numbers of images that I still routinely capture. Instead of using Faststone for triage, I can use Lightroom or Silkypix to do bulk (for example) exposure increases to make it easier to see what is going on in my often under-exposed images, but that slows down the triage considerably and complicates the subsequent file handling. I generally find it preferable to use Faststone and, where I can't see enough to know if an image might be usable or not, to leave the image in the set and see how it responds to the initial DXO processing, which levels up the brightness, and then decide in Silkypix (or Lightroom) whether the image is usable or not.

Why not simply start by putting all the images through DXO and then do the triage? Time. DXO Optics Pro Version 11 is a bit faster than Version 10, but even so it takes 30 seconds or so for each FZ200/330 raw file, and longer for 70D raw files. So 600 FZ200/330 raw files (not an exceptional number from a session) would take 5 hours to process in DXO. I could leave them processing overnight, but DXO stresses the PC, using all four processors and the graphics card at full pelt, making the fans whine quite loud. I'd rather not leave that running overnight.
 
An experiment - More DoF vs more processing, FZ50 and FZ330

I recently bought a 2006 vintage FZ50 bridge camera. It's main attraction for me is that it has a small minimum aperture compared to my other cameras. It goes down to f/11 which, given its 1/1.8" sensor, is equivalent to f/53 on a full frame camera. My other cameras have minimum apertures which are around f/45 equivalent.

The FZ50 therefore has a smallest aperture of about 1/2 stop smaller than my other cameras, and should therefore give a greater depth of field at minimum aperture.

On the other hand, images from my other cameras can be enhanced in terms of sharpness and detail using DXO Optics Pro's Microcontrast enhancement and camera/lens-specific Lens softness correction. DXO cannot apply these enhancements to FZ50 images - the FZ50 is not one the cameras DXO knows about.

Given that my go to camera for insects, spiders etc is my FZ330 (with close-up lenses), the question in my mind was "What gives an image that is better looking - the FZ50 with its smaller aperture of f/11 and larger DoF, or the FZ330 with a larger aperture of f/8 but enhanced with DXO Optics Pro?"

The following images were captured raw using a Raynox 250 close-up lens on the FZ50 and the FZ330, with my KX800 twin flash and diffusers also used in both cases. I captured several raw images with each camera and picked the best one from each camera. I used a small central focus area (the only focus area available on the FZ50) focused on the eye, and then recomposed to get the whole fly into the frame. The images don't exactly line up but I think they are close enough for this comparison, the results of which are consistent with two other series of images I captured of the same subject, one of those sets using the Raynox 150 rather than the Raynox 250.

In terms of post processing, I loaded the FZ50 (f/11) image into Silkypix and used the default settings (apart from adjusting the white balance to try to get the colours more similar to the FZ330 colours). I did the same for the FZ330 (f/8) image. I also processed the FZ330 image in a different way, first applying my normal pre-processing in DXO Optics Pro, exporting as DNG and loading the DNG into Silkypix using the defaults. I exported from Silkypix as TIFF and then resized to 1300 pixels high using Faststone Image Viewer.

The first comparison flilps between the FZ50 f/11 image and the FZ330 f/8 image, both processed with Silkypix defaults.


1011 Comparison 1
by gardenersassistant, on Flickr

Looking at the nearer feet, I think the centre of focus is slightly nearer the camera for the FZ330. Because of that it is difficult to be sure , but I do have the impression of greater DoF for the FZ50.

The next comparison flips between the FZ50 image, processed with Silkypix defaults, and the FZ330 image processed first in DXO and then with Silkypix defaults.


1011 Comparison 2
by gardenersassistant, on Flickr

Looking at the thicker hairs on the back of the fly, it still appears to me that the FZ50 has greater DoF (which is what you would expect - these are the same two images we are comparing after all). However, the areas of the FZ330 image that are within the DoF are now so much sharper and more detailed that the DoF appears to be larger than it was. For example, look at the detail in the front (top as we look at it) of the wing. In the rest of the in-focus areas, away from the edges of the DoF, there is much more detail and also better clarity in the FZ330 image. For detail for example, look at the eye. (These differences might stand out more clearly by going to Flickr and looking at the "Original" size images. They are the same size as shown here but they appear to show more detail, for example the cell structure of the eye.)

Despite being a great fan of maximum DoF and the fact that the FZ50 has greater DoF, I still prefer the look of the more detailed FZ330 image (and that is a rather strong preference too).

The final comparison shows the difference between the FZ330 image as it looks processed with the Silkypix defaults and how it looks when processed by DXO and Silkypix.


1011 Comparison 3
by gardenersassistant, on Flickr

For my purposes, the ability to use DXO with the FZ330 overrides the benefits of the FZ50's smaller aperture and greater DoF. (And because I can't distinguish between FZ330 and FZ200 in terms of image quality, the same is true for the FZ200.)

I think this probably completes my testing of the FZ50. I may keep it as an interesting historical item, but I don't think it will see active use for any of my day to day photography.
 
Gardenershelper, this is some thread, fantastic.
I have just started using an rx 10 iii + a raynox 250.
Havent gone smaller than f8 yet, but i will need to, to get a greater dof, also need to get a proper ring flash or some sort of macro flash.
Here are 2 of mine

https://flic.kr/p/NkaTdb

https://flic.kr/p/Nua44d

Any thoughts ?

It looks like that combination of camera and close-up lens works well. Very good clarity and detail on the eye in the first one, especially as you weren't using flash. Presumably that is direct sunlight on the fly?
 
Hi, that's spot on,yes direct sunlight. I am interested in putting some sort of rig together but am unsure as to what way to go, as there is a limited budget after getting the rx 10 iii.
I have had a dslr and 100mm macro lens in the past and used extension tubes and loved experimenting, but thats all it was, experimenting. I can see from reading through this thread that you actually understand the magnification ratios and sensor ratios that to me might as well be double dutch :eggface::eggface::eggface:
So any advice would be greatly appreciated, as I am aware of my short comings in this area.
Derek
 
Hi, that's spot on,yes direct sunlight. I am interested in putting some sort of rig together but am unsure as to what way to go, as there is a limited budget after getting the rx 10 iii.
I have had a dslr and 100mm macro lens in the past and used extension tubes and loved experimenting, but thats all it was, experimenting. I can see from reading through this thread that you actually understand the magnification ratios and sensor ratios that to me might as well be double dutch :eggface::eggface::eggface:
So any advice would be greatly appreciated, as I am aware of my short comings in this area.
Derek

The cheapest approach to using flash (if you haven't already got an external flash) is to use the on-board flash and a home made "snoot". This is what Mark Berkery uses (he is one of my macro heroes). Have a look at this page where he explains how he works. (He uses close-up lenses btw). If you search for "Velcro" on that page you will see the snoot he uses. You will see from the many images on that page that a snoot can provide excellent light. (This is not "I suggest you use what I use" advice by the way - I use an external flash and I don't use a snoot. :))

That is a long page but, especially if you are new to close-up/macro, I think it is probably worth reading all of it. (Probably worth reading for everyone who is into this stuff actually, beginner or not.)
 
[There is more detail about this stuff in this thread at dpreview.]

In Seven stacked botanical close-ups I posted seven images created from natural light 4K videos captured using Panasonic's post-focus technology with an FZ330 bridge camera and a mild Canon 500D close-up lens, and using Helicon Focus to extract and stack images from the videos. Since then I have spent a lot of time experimenting with using the same basic technique (but using an LED light rather than natural light because of low light levels) for insects. Unlike with the flowers, this proved to be very difficult and I only had limited success. This post contains some of the more successful, albeit for the most part flawed, results that I achieved.

Despite the overall lack of success with this particular technology for insects, what I did achieve gave me some insight into what this sort of technique might deliver if I could get it to work more reliably, possibly with a different technology such as focus bracketing (which would need a different camera). Strange as it sounds, I suspect that in some cases different subjects might give better results. I may start using a tripod again for insects etc for this sort of work - most of the experiments were done hand-held, with some quite slow shutter speeds. I'm also still learning about some of the ins and outs of stack processing and ways to work round some of the issues.

So I expect I will continue to experiment with stacking for insects etc. That will probably have to await Spring before suitable subjects become available. I was fortunate to find two insect subjects that I could experiment with, one dead (a fly, which I took indoors) and one that was out in the open on a stone on the patio, moving very little for several days and completely immobile on some cold days. I also had a brief opportunity with a beetle, which is one of the images below.

With the exception of #4, which is a single, diffused flash image, these are stacks created from 4K videos captured with my FZ330 and Raynox 150 or 250 close-up lenses.

There are 1300 pixel high versions of these images over at Flickr.

I was pleased with the first one until I noticed a glitch in the ground running across the image at about the level of the insect's mouthparts. This capture was hand-held (kneeling, probably with my hands resting on my thighs, as probably were the wasp images below) and I suspect that using a tripod might have avoided that issue. As it was I only had a few attempts before the beetle moved away, so I might not have got any shots at all had I been setting up on a tripod.

This image is stacked from 113 4K (3328 x 2496) frames, which at 30 frames per second would have taken around four seconds to capture. It was captured using ISO 100, f/2.8 (equivalent to f/10 on APS-C, f/16 full frame), with a shutter speed of 1/60 sec, using a diffused LED light.

#1

1045 1 2016_11 STACK 261 113f (C)x(A) SP7 LR6 Df
by gardenersassistant, on Flickr

Now two stacks of the wasp, also using the LED light. This one was a stack of 90 frames, ISO 100, f/2.8, 1/50 sec.

#2

1045 2 2016_11 STACK 266 90f (C)(a) SP7 LR6 Df
by gardenersassistant, on Flickr

#3
77 frames, ISO 100, f/2.8, 1/60 sec

1045 3 2016_11 STACK 216 77f (C)(a) SP7 LR6 Df
by gardenersassistant, on Flickr

The problem with these two, as with a lot of others, was noise. One of the stacking methods in Helicon Focus is good for detail, but creates lots of noise. Another is fine in terms of noise but poor with details. What I did with these was to use the noisy method and after doing the stacking in Helicon Focus and processing the stacked versions in Silkypix and Lightroom, I then denoised the backgrounds using Nik Dfine. That was ok for the plain areas, but as an observant member at dpreview pointed out (which I hadn't noticed) the shadows were noisy too. (Perhaps that was with another image, but the general point remains - noise is an issue.)

For comparison purposes, here is a single shot of the wasp captured using my usual technique of single shot using diffused flash and a very small aperture to maximise the DoF at the cost of loss of sharpness/detail.

#4
ISO 100, f/8, 1/1600 sec

1045 4 2016_11 SINGLE IMAGE P1180271_DxO RAW01a100 SP7 LR6
by gardenersassistant, on Flickr

Looking at the antennae, the DoF is obviously much less with the single shot, and I don't think the hairs on the head and thorax show up as well either. On the other hand, this is a shot that could not have been captured with a stack as the small fly in the background was moving rapidly (it just happened to line up for the shot. I had no idea it was there until it wasn't). I also much prefer the gentle fall-off between the in-focus and out of focus areas. I really don't like the rapid, unnatural transitions in the stacked versions. For one or two of them I did try softening the edges a bit in post processing, but there wasn't much I could do about them.

Now a couple of the fly. Not a pretty sight I'm afraid, but quite good as something to practice with with all that fine detail. I don't recall seeing as much texture/detail in a wing as in this one. My hands were probably resting on the desk for this one.

#5
Around 40 frames. ISO 125, f/4, 1/40 sec

1041 36 ISO 125, F4, 1-40 sec 426 (A,Radius3,Smoothing4) SP7 1300h
by gardenersassistant, on Flickr

I think I used a tripod for the next one.

#6
Around 110 frames. ISO 100, f/4, 1/50 sec.

1041 40 ISO 100, F4, 1-50 sec 667 (C) SP7 LR6 Df
by gardenersassistant, on Flickr

Now another example of the sharp focus drop-off (on the wing this time) that I think really spoilt this image (and given the angles I don't think I'd want to crop enough to get rid of it).

#7
Around 65 frames. ISO 125, f/4, 1/30 sec.

1041 37 ISO 125, F4, 1-30 sec 428 (A,Radius3,Smoothing4) SP7 1300h
by gardenersassistant, on Flickr

And finally, one more of the wasp, this one on a damp, foggy evening. Another one where the rapid focus drop-off (this time on the far antenna) puts me off. Perhaps this was the one where I got the (perfectly justified) comment about noisy shadows.

#8
Around 60 frames. ISO 100, f/2.8, 1/50 sec.

1041 35 ISO 100, F2.8, 1-50 sec 601 (C) SP7 1300h Df
by gardenersassistant, on Flickr
 
In this thread I posted my first attempts at focus bracketing with my new G80 and Olympus 60mm macro and some background on focus bracketing. I have since then done some indoors experiments to try to understand better what causes the halos I see in my stacked images, and to work out how best to deal with the halos that I can't (or didn't) avoid.

Sometimes the halos can be reduced by using a different stacking method or changing the parameters used by the stacking process, but there are trade-offs. The methods and parameters that reduce halo formation also tend to reduce the amount of detail visible, sometimes drastically (as in turning a detailed area almost completely smooth), which is a rather serious shortcoming for close-up/macro. Also, the method that is usually best in terms of the detail it reveals has a horrible effect on plain areas, making them very noisy (not just a little noisy if you look hard and pixel peep, but sometimes grossly noisy at normal viewing size - you don't need to look for it, you can't avoid it).

These two issues put a premium on something that both Zerene Stacker and Helicon Focus let you do - as well as allowing you to paint from individual images on to the stacked image, they let you paint from one stacked image on to another stacked image. This means that you can make two or more stacked versions using different methods and/or parameters and then switch between these stacked versions to pick halo-free outlines, smooth backgrounds and subject detail from one or the other as appropriate. Like painting from individual images on to a stacked image it can be time-consuming, but as I get more familiar with the tools (I'm mainly using Helicon Filter) I'm getting quicker at this sort of retouching.

So it turns out that stacking, at least for the sort of subject matter I want to stack at the moment (I don't know about invertebrates yet) is a much slower process than the photography I am used to. Indeed, you could argue that it isn't really photography at all in the sense of capturing a scene. Obviously there is a large element of that, but there is also a large element of constructing the images rather than capturing them. I know that this is true with post processing generally, but it feels like a step up in complexity and time commitment, and for me it seems to tip the balance from mainly capturing to mainly constructing.

That, such as it is, is the good news. The bad news is that there are some halos that are impossible to get rid of using these techniques. And I do mean impossible, not difficult, or very difficult, or extremely time-devouring, but simply impossible as far as I can see. The problem arises when you have two areas, one behind the other, and you want both of them in focus. If there is enough distance between them (and it doesn't have to be a huge distance), what happens is that when the front one is in focus and has a sharp edge the rear one is out of focus. That is ok. But when the rear one is in focus the front one isn't, and it occupies a larger area than when it is in focus and this extra area hides the in-focus part of the rear area that is around the edge of the front area. This means that there will be no images in the sequence in which that part of the rear area is in focus.

Here is an example. In the following crop the top left hand end of the yellow petal with a pink edge that is above the water droplets is in focus. The green leaves behind it are out of focus.


1058 6 2017_01_08 Halo problem 1 - Left top edge of front area in focus
by gardenersassistant, on Flickr

Next we have the leaf behind the petal in focus. The petal is out of focus, and larger.


1058 7 2017_01_08 Halo problem 2 - Rear area in focus
by gardenersassistant, on Flickr

Here is what one of the stacking methods made of this area. Lots of halos.


1058 8 2017_01_08 Halo problem 3 - A stacked version
by gardenersassistant, on Flickr

Here is what another stacking method made of it. Different halos.


1058 9 2017_01_08 Halo problem 4 - Another stacked version
by gardenersassistant, on Flickr

Some sort of halos are inevitable in areas like these were there is not, and cannot be, in-focus versions of the area.

There are I think two options for dealing with this. One is to shorten the bracketing sequence so as to remove the images where the rear areas are in focus. This means removing everything behind them as well. Doing this can sometimes leave you with a usable image, but for the most part, for the type of images I want to make, I think I would be throwing away a lot of stacks. I would rather not do this.

The other alternative is to use cloning. Sometimes that will not be possible as there is no suitable source area to draw from. In other cases, like this one, it is possible. But it requires a certain amount of skill and patience and makes the process even more time-consuming. When I first realised that this was likely to be a common problem for my botanical images my immediate thought was to give up stacking on the grounds that it would be more trouble than it is worth. However, I really liked the look of some of my early botanical stacks and so I decided to take my time and have a serious go at learning how to use the available tools to see if it would become workable. After all, I nearly gave up close-up/ macro when I started out with a Raynox 250 and couldn't get anywhere with it.

After doing some indoor experimentation with test scenes to try to understand what was going on I went out into the garden and captured some stacking sequences to play with. Even though from what I have read Zerene Stacker seems to be regarded as the better product I found that I was more comfortable using Helicon Focus' tools. It has three stacking methods compared to two in Zerene stacker and what is probably more important to me it has a cloning tool, which Zerene doesn't. The Helicon Focus cloning tool also has a rather clever "Colour tolerance" setting which can in some cases make it much easier to apply changes on only one side of a boundary between two differently coloured areas.

By the end of the exercise (which involved multiple attempts with several images) I felt I was getting more of a feel for how to go about this sort of retouching and getting a bit faster (more accurately perhaps, a bit less slow) at doing the retouching, and a bit better at noticing what needed doing.

I have posted five images from this botanical subject learning session in this thread in the forum.

More in the next post about the implications of all this for my choice of equipment, which looks like changing yet again.
 
Last edited:
Equipment. Not settled after all. :)

I have been thinking of:
  • my Canon 70D, with and without a Canon 500D close-up lens, as my preferred camera for natural light shots of botanical subjects and large insects,
  • my Panasonic FZ bridge cameras (FZ200 and since last summer FZ330) with Raynox 150 and 250 as my camera for flash images of medium sized invertebrates, and
  • my Panasonic micro four thirds G5 with 45-175 lens and Raynox 150+250 stacked, MSN-202 and MSN-505 for flash images of small invertebrates.
The flash for both FZ and G5 cameras is still the Venus Optics KX800. For a while now I have been working hand-held almost all the time.

I have dabbled with multi-image techniques (a few very small stacks and occasional exposure blending), but the very great majority of my stuff until now has been single images. The FZ330 and now the G80 have made me wonder whether it is now practical for stacking to play a larger role. I already know that stacking has great potential, and I have seen great stacks from other people here and elsewhere, but it has never really clicked for me. I think that what the FZ330 showed me with Post Focus, and the G80 has shown me even more so with focus stacking, is that, for me, the benefits of stacking in terms of image quality and (for botanical subjects) aesthetics may justify the slower pace and greater per image commitment needed.

I think that in terms of basic image quality the G80 is probably a match for the 70D. And it can do focus bracketing (and post focus). The 70D can do neither. This mandates the G80 or FZ330 for botanical stack captures. However, the FZ330 only offers the small image size of 4K post focus, which is limited to JPEG only. The G80 can do post focus but also focus bracketing, which uses full size images and can use raw as well as JPEG, and also provides more control over the sequencing of the captures (number of captures, focus intervals between captures, positioning and range of distances focused on). This makes focus bracketing preferred (for me) over post focus, and makes the G80 the preferred camera for capturing image sequences for stacking. If I am using the G80 for stack acquisition anyway, and it can produce much the same quality for single shots as the 70D, I may as well use the G80 for botanical single shots as well as for stack sequences.

I have a macro lens for the 70D, a Sigma 105 macro, but I have never got on with it, and only used it for testing and experiments. It may seem strange therefore that I have purchased another macro lens, this time an Olympus 60mm macro, for use with the G80. My thinking here was that it didn't really make any difference to image quality whether I used the Sigma 105 Macro or not because I use very small apertures for single-image close-up/macro shots, and this brings all lenses to a common, low level in terms of sharpness and detail capture. My comparisons documented earlier in this thread went through that in some detail. However, for stacks one uses larger apertures, around the sweet spot for sharpness. In this case there should be a significant advantage in using a macro lens because at these larger apertures it should out-resolve my zoom lenses and achromats by a significant margin. And this is indeed what I believe my early experience with these botanical stacks has shown.

I still don't much like using a prime lens. I miss the flexibility of a zoom. But the 60mm macro is tiny and light, and very sharp, and colours and clarity look pleasing to me too. (The lack of an antialiasing filter in the G80 helps with detail capture too.) As with the extra time needed per image with stacking being justified by the outputs that become possible, I think the loss of flexibility and convenience of a zoom + achromat may also prove to be a worthwhile trade-off for improved image quality.

One of the issues with the 60mm in terms of flexibility is that, with it being a prime, I can't frame shots easily by just zooming in and out. This doesn't matter so much when working hand-held, but it does make quite a difference when using a tripod. And one of the things my most recent bracketing exercise has shown me is that a tripod is definitely a good idea for bracketing (we are talking 100 shots a time here), as well as being pretty much necessary in the low light levels on the overcast days I'm working in at the moment.

Another issue with the 60mm macro is that it doesn't have the reach of my zoom lenses. I used a 55-250mm with the 70D, which is around 80 to 400mm full frame equivalent. I encounter botanical subjects that I can only just reach with the 70D. No matter how sharp the 60mm is it will not be able to handle such subjects. My micro four thirds 45-175mm goes out to 350mm equivalent and my 45-200mm goes out to 400mm equivalent, although I haven't used that lens since I got the 45-175. I am much more likely to carry the 45-175 around with me (which is delightfully small and light) for longer reach botanical shots than carry the extra weight of the 70D and 55-250 just in case I need the extra reach.

It rather looks like the 70D may be phased out for close-up/macro. I do have one use for it, which is occasional forays into birds in flight at the local pond and marina. It is far superior to any of my other cameras for that.

I rather doubt I will be using the 60mm macro for invertebrates. As you get towards 1:1 the working distance gets quite short, around 85mm at 1:1 (18mm scene width on micro four thirds). This compares to around 200mm with my Raynox 150, which gets me down to about 14mm scene width on the FZ330, and around 120mm working distance with the Raynox 250 on the FZ330, which gets me down to about 9mm scene width. As long as I am using very small apertures I see no advantage in using the 60mm macro with the G5. I will do some comparisons though, between the FZ330 and G5 +45-175, both with Raynox 150 and 250, both at minimum aperture, and the G5 with the 60mm macro at f/11 and f/16 and cropping for DoF.

There is another issue with the 60mm macro though, which is getting beyond 1:1. This is obviously possible, to some extent, with extension tubes and/or close-up lenses. However, I work quite fluidly either side of 1:1 and I can see this being awkward with the 60mm macro. My guess is that, unlike with botanical work, improved image quality with invertebrates won't justify the inconvenience. However, that is something for another day when I have had a chance to try it rather than just thinking about it. :)

Another modification to my equipment setup is the use of an LCD light. I can't use flash for focus bracketing (or post focus). I don't know how useful it will be out of doors (it depends on the working distance), but I have used this LCD light setup for some indoor test shots, and for a few post focus insect images shown in this thread.


1051 19a LCD light
by gardenersassistant, on Flickr
 
Last edited:
the G80 [is] the preferred camera for capturing image sequences for stacking. If I am using the G80 for stack acquisition anyway, and it can produce much the same quality for single shots as the 70D, I may as well use the G80 for botanical single shots as well as for stack sequences.

Further investigation has shown this is not the case, at least not for a few months. The reason is the DXO Optics Pro does not yet support the G80. I did some like for like comparisons between the G5, using DXO in the processing mix, and the G80, without DXO. I preferred the G5 results - better detail/microcontrast/clarity. So until DXO supports the G80, the G5 is preferred to the G80 for single shots. (For focus bracketing I can only use the G80 as it is the only camera I have which does focus bracketing).

The question then arose of whether the G5 is better than the 70D for botanical shots (DXO supports both cameras). I did some like for like botanical shots with the G5 + 45-175 and the 70D + 55-250 STM (I don't know why I haven't done a test like this before, at least I don't think I have.). These were hand-held with some quite slow shutter speeds using ISO 800 and 1600, which is a realistic setup for overcast days. I used electronic first curtain with the 70D and electronic shutter with the G5, in both cases using the LCD. I liked the G5 shots better - more of the shots had better overall sharpness/detail (although the 70D had more "crunchy" petal textures in some area, but I'm not sure I liked that anyway), and I preferred the way the colours came out with the G5 (not so much the hues, which can be tweaked anyway, to some extent, but the colouration of lighter areas that seemed to retain colour better with the G5 rather than going towards light/white with the 70D). So, I am going to switch to the G5 for botanical work until the G80 is supported by DXO.

I also prefer the lighter weight of the G5 and G80 compared to the 70D, although that wasn't a key factor in the decision. There were also, rather to my surprise, hints that the noise characteristics of the G5 were marginally better than the 70D, although the sample size was too small for me to be convinced about that. The 70D + 55-250 goes slightly wider and longer than the G5 + 45-175 but this too was not a key factor in the decision.

It looks like the 70D is not going to get used much for now. Occasional bird in flight sessions at the marina and boating lake may be about all I do with it. Perhaps I'll put my 10-20mm on it, which I have hardly ever used, and use that for sunsets and cloudscapes.

for stacks one uses larger apertures, around the sweet spot for sharpness. In this case there should be a significant advantage in using a macro lens because at these larger apertures it should out-resolve my zoom lenses and achromats by a significant margin. And this is indeed what I believe my early experience with these botanical stacks has shown.

I still don't much like using a prime lens. I miss the flexibility of a zoom. But the 60mm macro is tiny and light, and very sharp, and colours and clarity look pleasing to me too. (The lack of an antialiasing filter in the G80 helps with detail capture too.) As with the extra time needed per image with stacking being justified by the outputs that become possible, I think the loss of flexibility and convenience of a zoom + achromat may also prove to be a worthwhile trade-off for improved image quality.

Having tried again with the 60mm macro I have changed my view of this. For single shots I really do prefer, strongly prefer, the reach and flexibility with a zoom, and I think the image quality is good enough. I think the 60mm macro will be used only for stacks, and that is only some stacks, where reach and - for invertebrates - (rather small) working distance permit.

Another constraint on using the 60mm macro is lens-changing. It is inconvenient to handle out in the field. I can live with that, but as with the 70D I have been having problems with the G5 and G80 with dust on the sensor. Even though I have all manner of materials and equipment for sensor cleaning I have had difficulty getting the sensors completely clear, and have had to use tediously many attempts. In fact, I have one mark (not a dust spot by the look of it) showing on the G5 that I have given up trying to move (perhaps it is an immovable flaw rather than something on the sensor). Fortunately it is near a corner and shouldn't be problematic. But this issue does discourage me from changing lenses once I have got the sensors clear of spots.
 
Three months on and things seem to be clarifying (for now at least!) on the equipment and post processing front.

There has been a firmware upgrade for the Panasonic 45-175 lens that I like so much. This means that the 45-175 now works with "dual IS" on the G80, with the camera's in-body image stabilisation (IBIS) working cooperatively with the lens' optical image stabilisation (OIS). I have been doing hand-held natural light botanical work with the G80 and 45-175 for the past 6 weeks or more and I very much like the usability of this combination.

DXO Optics Pro now supports the G80 and 45-175 and I have done a three way comparison (results in this album at Flickr) for 40 botanical images I shot a couple of days ago before the (much needed) rain started. I processed each capture with:
  • RAW into Silkypix Developer Pro 7 for batch processing, out as TIFF to Lightroom 6 for image-specific adjustments and output to 1300 pixel high JPEG
  • RAW into DXO Optics Pro 11 for ISO-specific batch processing, out as DNG in to Lightroom for image-specific adjustments and output to 1300 pixel high JPEG
  • RAW into DXO Optics Pro 11 for ISO-specific batch processing, out as DNG in to Silkypix for batch processing, out as TIFF to Lightroom for image-specific adjustments and output to 1300 pixel high JPEG (One very heavy crop, image 3, also had Dfine 2 noise reduction on the background for this DXO+Silkypix+Lightroom version.)
In every case I preferred the third of these options: DXO followed by Silkypix followed by Lightroom.

Given that I like using the G80 and 45-175 for botanical work and like the results I can from it, the G80 + 45-175 has definitely become my go to setup for botanical work. I usually use it with or without a Canon 500D close-up lens, but on a few occasions have used a Raynox 150 for small subjects.

My processing workflow for G80 natural light botanical work will (for now!) be
  • Initial selection in Faststone Image Viewer, the selected raw files moved to a subdirectory
  • DXO Optics Pro, Silkypix and Lightroom processing as described above, with further selection during the Lightroom stage to produce a shortlist of 1300 pixel high JPEGs
  • XnView MP to check the image quality of the shortlist JPEGs and reject or reprocess any unsatisfactory images. Reprocessing may occasionally involve the use of Dfine 2 on a 1300 pixel high JPEG to smooth out noisy/crunchy backgrounds (which can arise for example from extremely aggressive shadow raising and/or severe cropping).
Although I have been capturing flowers etc for the past 6 weeks or more I have only just started processing the images. However, looking back on the shooting sessions it is striking that I haven't done any stacks. It has often been breezy anyway which militates against the use of stacks, but I think the real reason is simply that I like the flow of single-shot shooting and don't find stack-shooting so enjoyable. I'm also put off too by how fiddly it can be to process stacks.

The FZ330 is still very much my go to camera for invertebrates, mainly with the Raynox 150 attached. I have arranged it so that I can add a second Raynox 150. This is much easier than dismounting the Raynox 150 and replacing it with a Raynox 250, one reason being that the concave diffuser has to be detached when replacing the 150 with a 250, which is fiddly. I have a smaller concave diffuser for use with the Raynox 250 and changing the diffusers adds to the time and effort needed to up the magnification. I have only used the double 150 a few times, but when I did I didn't change the concave diffuser, which could therefore stay attached to the 150 mounted on the camera lens, meaning that all I had to do to change magnification was to screw on the second 150. This is much easier and quicker than a change to a 250. It gives slightly more magnification than with the Raynox 250. The range of scene widths is from about 37mm down to about 7mm, with a working distance of about 90 mm independent of magnification.

Adding a Raynox 250 rather than a second 150 gives a range of scene widths from about 26mm down to about 5mm, with a working distance about 70mm independent of magnification. I haven't used this combination yet.

Autofocus works throughout the range with both a second Raynox 150 and with a Raynox 250 added to the 150.

I have not tried photographing any tiny invertebrate subjects for some time. If I do go out specifically to look for tiny subjects I may well use the Raynox MSN-202 and/or MSN-505 on the 45-175 on the G80. This is because the fact that the 45-175 doesn't extend when zooming makes subject acquisition and changing magnification much easier at high magnifications. (The KX800 twin flash works fine on the G80, with exactly the same diffusion arrangements as on the FZ330.)
 
It's good to (as it were) see you back again Mike. I lost interest completely for several months. From the start of the year I photographed the flowers in the garden as they came out, not wanting to waste the opportunities, but I couldn't bring myself to process any of them. It was only when the first invertebrates appeared that I started doing a little processing. I'm well back into capture, processing and posting now, but it took several weeks of cautious engagement to get there.

So, perhaps you can gently ease yourself back into it Mike. I hope so. :)
 
Adding a Raynox 250 [to the FZ330] rather than a second 150 gives a range of scene widths from about 26mm down to about 5mm, with a working distance about 70mm independent of magnification. I haven't used this combination yet......I have not tried photographing any tiny invertebrate subjects for some time. If I do go out specifically to look for tiny subjects I may well use the Raynox MSN-202 and/or MSN-505 on the 45-175 on the G80. This is because the fact that the 45-175 doesn't extend when zooming makes subject acquisition and changing magnification much easier at high magnifications.

Well, I did go out a couple of days ago specifically looking for tiny subjects. Rather than using one of the most powerful achromats, Raynox MSN-202 or MSN-505, on the G80, I stacked a Raynox 250 on a 150 on the FZ330. I'm not sure why I decided to do that, but I'm glad I did.

I used autofocus without problem for the whole of the session.

The fact that the FZ330 lens extends proved not to be a significant issue. I think this was mainly because I was working hand-held, which makes it easy and quick to adjust the distance to the subject. (I have previously used a tripod to help stabilise the camera to assist with subject acquisition, focusing and scene framing. Framing obviously has a larger random element when working hand-held, but especially given the advisability of cropping for DoF with smaller subjects, it proved not to be too much of a problem. Subject acquisition was not more difficult than when using a tripod, just differently difficult. I suppose the really surprising thing is that autofocus worked ok hand-held. I imagine this is because the FZ330 focuses a bit faster than my earlier cameras.)

The extra working distance of the 150 + 250 combination compared to the MSN lenses made the rig easier to use than with the shorter MSN working distances.

Another oddity was that I didn't change the concave diffuser. I have a set of concave diffusers, progressively smaller for the more powerful lenses so as to match the decreasing working distance. But for this exercise, and again I'm not sure why, just forgetfulness I suppose, I left the largest of the concave diffusers in place, the one that I use with the Raynox 150. There was an issue a couple of times with the concave diffuser getting in the way, but for the most part, and much to my surprise now I think about it, it worked fine. I think that must have been because I was working close to the ground or on fallen tree trunks most of the time and there was no vegetation hitting the diffuser, unlike the case when working with subjects on foliage on a bush, which is where I did encounter the couple of problems.

Because of the way close-up lenses work I was able to work out the size of the subjects when processing the images. I found a globular springtail and was happy that I had found something small enough to do a good test, something - I assumed - with a diameter in the one millimetre range. I was therefore disappointed when I discovered from my measurements that it was in fact twice that size. However, I did find one smaller subject, of around 1.2mm body diameter - a red mite. Most of the rest of the subjects were in the 2.5mm to 6.5mm range, exceptions being a fly and crane fly both around 9mm head to tail and a wood louse of around 11mm length. There are 1300 pixel high images from the session in this album at Flickr. I have posted a selection of the images in[TO BE DONE] this post in the forum.

The raw files were batch processed in DXO Optics Pro and Silkypix, with image-specific adjustments in Lightroom.

I cropped a lot for the smallest images, going down almost to a 100% crop for all the images of the springtail. This is a big ask for a camera with a 1/2.3" (smart phone size) sensor, especially as I use some quite aggressive processing to bring out detail. This (and in particular the Silkypix element of this) creates a "crystalline" structure on a small scale in smooth areas, which heavy cropping makes unpleasantly visible. I have in the past used Nik Dfine on the final JPEGs to deal with this, but I was not satisfied with the results in this case. I went back to a technique that I have used before which involves hitting the whole image with very strong noise reduction in Lightroom and then painting over the subject and any other in focus areas with a brush reversing the noise reduction. Alternating between an intelligent ("auto-masking") brush and a dumb feathered brush turned out to be quick to use for these images and as far as I could see, effective. In fact, I went through all 57 of the processed images and applied this technique to them, whether or not I had been conscious of the "crystalline noise" or not. I think it is something that I may apply more often in future, and I think Nik Dfine may get dropped from the toolkit.

On the basis of this exercise I think that for now at least I will use the FZ330 for all my flash-based work, including very small subjects. A Raynox 150 will stay attached to the camera and another 150 or a 250 added when I need to increase the magnification. Doing this without changing the concave diffuser makes it a quick and easy operation (which isn't the case when taking off the 150 to replace it with a 202 or 505, which is a fiddly process). I'll see how this approach pans out with more extensive use.

As well as the working distance advantage over the MSN lenses, I didn't notice any of the chromatic aberration that I have come across with the MSN lenses, especially the 505. This may simply be because of the subjects I was photographing not being prone to this problem by virtue of having strong dark/light boundaries, and also because I wasn't using the highest magnifications I had available. I'll need to use the 150+250 more to see how it does with that type of subject.

Given the convenience and speed of adding an extra lens, the benefit of a longer working distance and the possible benefit of reduced chromatic aberration, I think the 150 with and without another 150 or 250 may make the MSN lenses pretty much redundant for me. I would need to use one of them if I wanted to tackle smaller scene sizes than I can get with the 150 + 250, but for this session the maximum focal length I used was 400mm equivalent compared to the maximum 600mm equivalent of the FZ330, so I think the need for greater magnification than I can get with the 150 + 250 would probably be a rather rare occurrence.
 
Last edited:
Thanks Nick. I've a new mini soft box to try out with my Fuji so maybe that'll get me interested :)

Wow, that is big. And big is good in this context, as long as (a) the light gets spread somewhat evenly over the large surface and (b) it is practical to use despite the size. I'll be very interested to hear how it goes. A quick little indoor experiment perhaps to begin with? :)
 
Back
Top