Spring Flowers using the Sigma 24-105mm F4 DG (OS) HSM and the the Sigma 70mm F2.8 EX DG Macro lens

Started Jun 8, 2014 | Discussions
Nancy and Pete Spader Senior Member • Posts: 2,050
Spring Flowers using the Sigma 24-105mm F4 DG (OS) HSM and the the Sigma 70mm F2.8 EX DG Macro lens
1

We have recently spent most of our time testing the quality of our new Sigma 24-105mm F4 DG (OS) HSM lens on our SD1M camera. This involved exploring different focal lengths and also comparisons with our old primes, especially our Sigma 70mm f2.8 EX DG Macro. In doing this we have ignored posting any of the spring flowers as a primary subject. Since the number of people enjoy Nancy’s presentation of the spring flowers as they developed, here are a few of the spring flowers that have finally shown up now that the cold weather has finally begun to ease up. We supply links below each picture to our PBase gallery, since there you will find Information and the original sized images.

We hope you enjoy them!

All comments are welcome.

f8 at 70mm using the Sigma 70mm F2.8 EX DG Macro lens

http://www.pbase.com/pspader/image/156006186

f8 at 70mm using the Sigma 70mm F2.8 EX DG Macro lens

http://www.pbase.com/pspader/image/156006183

f8 at 50mm using the Sigma 50mm f2.8 EX DG lens

http://www.pbase.com/pspader/image/156006179

f5 at 68mm using the Sigma 24-105mm F4 DG (OS) HSM lens

http://www.pbase.com/pspader/image/156006177

Pete and Nancy

Sigma SD1 Merrill
If you believe there are incorrect tags, please send us this post using our feedback form.
larryj Forum Pro • Posts: 10,265
Re: Spring Flowers

Nice selection of flowers Nancy and Pete:  I really like the second shot, it captures  a nice balance of subtle colors and rich detail.  Thanks for sharing

-- hide signature --

Cheers,
larryj
If you can see the light, you can photograph it
Quote from Myron Woods

 larryj's gear list:larryj's gear list
Sigma SD1 Merrill Sigma 24-70mm F2.8 EX DG HSM Sigma 150-500mm F5-6.3 DG OS HSM Sigma 105mm F2.8 EX DG OS HSM Sigma 18-35mm F1.8 DC HSM Art +1 more
OP Nancy and Pete Spader Senior Member • Posts: 2,050
larryj

Thanks Larry. This is my favorite as well. This Peony has only been blooming a few years and this year it is special.

Nancy

xpatUSA
xpatUSA Forum Pro • Posts: 13,151
Re: The Red Flower

Nancy and Pete Spader wrote:

<snip>

All comments are welcome.

f8 at 70mm using the Sigma 70mm F2.8 EX DG Macro lens

http://www.pbase.com/pspader/image/156006186

In the linked image, there seem to be some purples around the red petal edges. So I took the liberty of opening the image in ColorThink. a) Looks like pbase strips out the ICC profile but mainly b) the reds and greens are quite heavily color-clipped to the 3D sRGB gamut boundary - perhaps due to the direct sunlight illumination?

I can also get that effect in SPP by editing to perfection in ProPhoto working space and then 'saving image as' sRGB . . .

-- hide signature --

Cheers,
Ted

 xpatUSA's gear list:xpatUSA's gear list
Sigma DP2 Sigma DP1 Merrill Panasonic Lumix DMC-G1 Sigma SD14 Sigma SD15 +16 more
OvinceZ
OvinceZ Senior Member • Posts: 2,725
Camera settings for flower photos.

I know you and others on this forum have done tests and concluded that the Foveon sensor underexposes so you use +.3 EV. From my experience shooting flowers I always use at least -.3. If I am shooting red flowers I use -1 EV or more minus EV. That way I don't get burned out colours.

I also highly recommend using ISO 100 if the camera is on a tripod. I prefer f/11 aperture to get more depth of field.

Just the way I shoot flowers.

Vince

 OvinceZ's gear list:OvinceZ's gear list
Sigma DP3 Merrill Canon EOS 5D Mark II Canon EOS 60D Sony Alpha a7R II Sony Alpha a7R III +15 more
OP Nancy and Pete Spader Senior Member • Posts: 2,050
Re: The Red Flower

Thanks Ted. Yes, it was at a time of day where the sunlight is very bright and it does cause problems with the reds and green especially, though we find the SD1M does a better job than the earlier cameras. I had been using this image to demonstrate the sharpness of the image so I did not color-correct it extensively. I do use SPP 5.5.4 and can use it to do some more correcting, but I can use Picture Windows Pro also to color correct these problems as well. An alternative is to use a better color-space than sRGB with its limitations. But sun that bright is always going to cause problems.

Pete and Nancy

OP Nancy and Pete Spader Senior Member • Posts: 2,050
OvinceZ

Thanks for the good advice. We find +0.3 to be a good general setting, but you are correct, no setting is good for everything. We  will try the -1 range for these types of flower shots. Yes, we have used iso100 to allow us to get better DOF and really should have dne so with this closeup shot.  I was doing things faster than I should have and whenever you are doing things undeer time pressure  you forget things.

Thanks again, Vince

Pete and Nancy

xpatUSA
xpatUSA Forum Pro • Posts: 13,151
A Caveat . .

Nancy and Pete Spader wrote:

Thanks for the good advice. We find +0.3 to be a good general setting, but you are correct, no setting is good for everything. We will try the -1 range for these types of flower shots. Yes, we have used iso100 to allow us to get better DOF and really should have done so with this closeup shot. I was doing things faster than I should have and whenever you are doing things undeer time pressure you forget things.

Thanks again, Vince

Good advice indeed, provided that the problem being avoided is "burnt out" colors caused by sensor saturation in one or more of the layers.

The Caveat

Since color rendition is so important in flower shots, another problem is color saturation in the scene itself - leading to color clipping when the captured image is transformed eventually into sRGB for posting. Color clipping, in extremis , causes false colors - or it caused the far more subtle effect of a flat appearance in highly saturated color areas. Here's an exaggerated illustration of that flat appearance taken from a X3F which was axtually quite under-exposed (about -1EV):

Saturation adjusted in PhotoPro Working space then 'saved as' sRGB. +0.7 at left, -0.3 at right

Of course +0.7 sat. was a bit much but this X3F, when saved as zero sat. sRGB, showed color clipping in ColorThink even though it was originally underexposed -1EV (according to RawDigger).

-- hide signature --

Cheers,
Ted

 xpatUSA's gear list:xpatUSA's gear list
Sigma DP2 Sigma DP1 Merrill Panasonic Lumix DMC-G1 Sigma SD14 Sigma SD15 +16 more
xpatUSA
xpatUSA Forum Pro • Posts: 13,151
Re: A Caveat . .

xpatUSA wrote:

The Caveat

Since color rendition is so important in flower shots, another problem is color saturation in the scene itself - leading to color clipping when the captured image is transformed eventually into sRGB for posting. Color clipping, in extremis , causes false colors - or it caused the far more subtle effect of a flat appearance in highly saturated color areas. Here's an exaggerated illustration of that flat appearance taken from a X3F which was axtually quite under-exposed (about -1EV):

Saturation adjusted in PhotoPro Working space then 'saved as' sRGB. +0.7 at left, -0.3 at right

One thing I just noticed in the above comparison is that brighter levels in the over-saturated image still have detail whereas it is lower levels that have lost detail. Which supports my point that lower exposure does not necessarily prevent over-saturated colors.

-- hide signature --

Cheers,
Ted

 xpatUSA's gear list:xpatUSA's gear list
Sigma DP2 Sigma DP1 Merrill Panasonic Lumix DMC-G1 Sigma SD14 Sigma SD15 +16 more
OvinceZ
OvinceZ Senior Member • Posts: 2,725
Sample photo using my recommended settings.

Waratah:   f/11  1.3 sec  ISO 100  -1 EV  Sigma DP3M

 OvinceZ's gear list:OvinceZ's gear list
Sigma DP3 Merrill Canon EOS 5D Mark II Canon EOS 60D Sony Alpha a7R II Sony Alpha a7R III +15 more
xpatUSA
xpatUSA Forum Pro • Posts: 13,151
Re: Sample photo using my recommended settings.

OvinceZ wrote:

Waratah: f/11 1.3 sec ISO 100 -1 EV Sigma DP3M

Very nice shot. Outstanding detail in the bright areas.

Just for interest, here it is compared with the sRGB color space:

Plotted on the familiar 'horseshoe' 2D diagram.

The red color clipping looks worse that it really is because the clipping is occurring in lower tones:

The somewhat less familiar 3D Lab diagram

Here we see that the red color-clipping occurs below about 35% lightness with very little effect on that particular image but, still, it is there.

-- hide signature --

Cheers,
Ted

 xpatUSA's gear list:xpatUSA's gear list
Sigma DP2 Sigma DP1 Merrill Panasonic Lumix DMC-G1 Sigma SD14 Sigma SD15 +16 more
OP Nancy and Pete Spader Senior Member • Posts: 2,050
The limits to gamut

Thanks all for your comments. Some of you are aware these debates have on before and are always useful.

I had some time today to play with this image using a color space (ProPhoto RGB) with a wider gamut than sRGB. It did allow me to come closer to capturing the color I see in this particular rose, though when I translated it to sRGB to post I lost most of what I had gained. Also, despite the fact that the wider gamut allowed me to come closer, I still could not reproduce what I saw with my own eyes/brain.

I suspect the extremely wide gamut which human beings have access to “naturally” will always allow us to see colors no camera or processing software will be able to capture and reproduce, and this rose has an example of such a color(s). Nancy has been trying to capture this particular "red" on this type of rose from the days of film and has never been able to do so.

Incidentally, you will notice that in very bright sunlight even our naturally extra-wide gamut is not wide enough to capture all of the color that is “there.” I believe this is what is happening when you stop seeing individual colors and see “white,” a white that disappears as the intensity of the light diminishes. Even our naturally wide gamut is not infinite.

So until I find a way to process these RAW files to get more than I can now I will accept the limits that are there. Incidentally, in this image the edges of the rose that are “light” colored are what I actually see. I can, of course, manually replace “white”or "lighter" with a color that I judge is more accurate, but doing so gives an artificial sense to the image. And it is truly artifical if what we are trying to captre and present is what we really see.

xpatUSA
xpatUSA Forum Pro • Posts: 13,151
Re: The limits to gamut

Nancy and Pete Spader wrote:

Thanks all for your comments. Some of you are aware these debates have on before and are always useful.

<snip>

I suspect the extremely wide gamut which human beings have access to “naturally” will always allow us to see colors no camera or processing software will be able to capture and reproduce, and this rose has an example of such a color(s). Nancy has been trying to capture this particular "red" on this type of rose from the days of film and has never been able to do so.

Incidentally, you will notice that in very bright sunlight even our naturally extra-wide gamut is not wide enough to capture all of the color that is “there.” I believe this is what is happening when you stop seeing individual colors and see “white,” a white that disappears as the intensity of the light diminishes. Even our naturally wide gamut is not infinite.

So until I find a way to process these RAW files to get more than I can now I will accept the limits that are there. Incidentally, in this image the edges of the rose that are “light” colored are what I actually see. I can, of course, manually replace “white”or "lighter" with a color that I judge is more accurate, but doing so gives an artificial sense to the image. And it is truly artifical if what we are trying to captre and present is what we really see.

Hi Pete,

I have an X3F from some years ago of a sun-lit sunflower. Yellow is one color that is 'close' to the edge' of the human gamut - along with ProPhoto and sRGB whose red-green edges are almost tangent to the familiar horse-shoe shape. So it comes as no surprise that, in ColorThink, my flower's yellow when saved out of SPP as a ProPhoto .tif is hard up against - and clipped by - the ProPhoto gamut boundary.

The flower colors themselves are well within the sensor's color space though. So there may yet be a processing path for these difficult to process shots, IMHO.

Good luck with your quest!

-- hide signature --

Cheers,
Ted

 xpatUSA's gear list:xpatUSA's gear list
Sigma DP2 Sigma DP1 Merrill Panasonic Lumix DMC-G1 Sigma SD14 Sigma SD15 +16 more
OP Nancy and Pete Spader Senior Member • Posts: 2,050
Re: The limits to gamut

Thanks Ted. I have not given up on getting a processing path that will do better than what we have now and will capture more than we can now capture (though as long as band-width is a problem for web-sites we will always have problems sharing what we get beyond sRGB). It is just that we have to recognize there are and always will be limits to what we can get. Since we do not know exactly where they are it is still worth pushing the envelope as far as we can!

Pete

xpatUSA
xpatUSA Forum Pro • Posts: 13,151
Re: The limits to gamut

Nancy and Pete Spader wrote:

Thanks Ted. I have not given up on getting a processing path that will do better than what we have now and will capture more than we can now capture (though as long as band-width is a problem for web-sites we will always have problems sharing what we get beyond sRGB). It is just that we have to recognize there are and always will be limits to what we can get. Since we do not know exactly where they are it is still worth pushing the envelope as far as we can!

Indeed, one such limit is our expectation of rendered image color vs. scene color, I find. Even though our camera sensors can capture 'colors' far outside of the xyY horseshoe it is of course impossible to render on, for example, a sRGB monitor - and the so-called 'wide gamut' monitors are not that much better, IMHO, not that I've ever seen one.

So we can never reproduce the exact look of cactus flowers but, as you say, we do our best.

For this purpose, I find ColorThink invaluable but it does cost a buck or two. In their 3D view, it is easy to see over-saturated colors crammed up against a standard color space boundary and, by experimentation, back off the ProPhoto working space saturation, re-save as sRGB until the crammed color(s) in ColorThink spread out a bit. I find that as little as about -0.3 on the saturation slider does the job for many 'saved as sRGB' shots. The main point here is having the feedback from ColorThink rather than gazing at an image and wondering.

A bit like my diabetes: I can use the tester to see the blood sugar level or I can guess from how I feel

 xpatUSA's gear list:xpatUSA's gear list
Sigma DP2 Sigma DP1 Merrill Panasonic Lumix DMC-G1 Sigma SD14 Sigma SD15 +16 more
xpatUSA
xpatUSA Forum Pro • Posts: 13,151
Re: The limits to gamut

xpatUSA wrote:

Nancy and Pete Spader wrote:

<snip>Since we do not know exactly where they are it is still worth pushing the envelope as far as we can!

<snip> For this purpose, I find ColorThink invaluable but it does cost a buck or two. In their 3D view, it is easy to see over-saturated colors crammed up against a standard color space boundary and, by experimentation, back off the ProPhoto working space saturation, re-save as sRGB until the crammed color(s) in ColorThink spread out a bit. I find that as little as about -0.3 on the saturation slider does the job for many 'saved as sRGB' shots. The main point here is having the feedback from ColorThink rather than gazing at an image and wondering.

Today I found a sunflower X3F with highly saturated yellows. I opened it in ProPhoto working space and saved as four sRGB jpegs with saturations of 0, -0.3, -0.5 and -1.0. The jpeg output files looked like this:

View original size to see which is which . .

Then I opened them all simultaneously in ColorThink's 3D Lab view and also opened the sRGB color space for comparison:

See the plot colors for which is which, if it's not obvious

We see that, at 0 sat., the yellows are crammed hard to the right. At -0.3, a slight improvement especially lower down. At -0.5, all is well! At -1, the image is obviously de-saturated too much. Of the four images I would used the -0.5 sat. and upped the sat. a bit in post. Or, I would gone to SPP and tried -0.4 saturation.

Now lets look for any hue shifts by using the 2D view:

As the 0 sat. image yellows head up the yellow(ish) axis we see that they take a sharp left and head off to true yellow on the boundary line. That is to say, there is a significant color change which we may not like. The -0.3 sat shifts similarly at the boundary, albeit not so bad. Again, the -0.5 image would be my choice for final processing.

I can put the X3F up on my site if anyone wants to play - it's a LO res SD10 snapshot - not Merrill-huge 

-- hide signature --

Cheers,
Ted

 xpatUSA's gear list:xpatUSA's gear list
Sigma DP2 Sigma DP1 Merrill Panasonic Lumix DMC-G1 Sigma SD14 Sigma SD15 +16 more
OP Nancy and Pete Spader Senior Member • Posts: 2,050
ColorThink

Thanks for suggesting this tool. It does look very impressive indeed. But, as you point out it does cost.Sometimes you have to pay to get something really useful assuming you can afford it. We cannot at this point, but I will keep it in mind.

Pete

xpatUSA
xpatUSA Forum Pro • Posts: 13,151
Re: ColorThink

Nancy and Pete Spader wrote:

Thanks for suggesting this tool. It does look very impressive indeed. But, as you point out it does cost.Sometimes you have to pay to get something really useful assuming you can afford it. We cannot at this point, but I will keep it in mind.

In truth it is quite a luxury which I bought for a different but related line of research. And the gain from the time and money is not huge. In fact the difference between a color-clipped image and one produced with the help of ColorThink can be quite subtle as illustrated here:

Both images post-processed a bit. Shown 200% in FastStone Viewer comparison view.

The hue shifts at brighter levels can be seen in one of the images and the 'good' one is a little less garish. Some may not even see the difference

-- hide signature --

Cheers,
Ted

 xpatUSA's gear list:xpatUSA's gear list
Sigma DP2 Sigma DP1 Merrill Panasonic Lumix DMC-G1 Sigma SD14 Sigma SD15 +16 more
OP Nancy and Pete Spader Senior Member • Posts: 2,050
Re: ColorThink

Thanks for this report since we casnnot get it aat this time. But as a research tool I can see how it would be of help.

For the time being we will focus on the practical attempts to get as accurate abimage as possible. Nancy has a great color sense and if what we end up with is not at least extremely close to spot on we just do not post it, except as an example of "close but no cigar."

Thanks again!

Pete and Nancy

Keyboard shortcuts:
FForum MMy threads