[Khashoggi]

How did you do that? Tried that in lightroom and elements 10&couldn't get it any better. Thanks!

But Labrador race are there in 3 colours, brown(like those two), black&blonde. With the light, couldn't really get it in my back on that moment, so had to search for some darker places.

[UnknownVT]

How did you do that? Tried that in lightroom and elements 10&couldn't get it any better.

I used Curves (or Tone Map) in my olde editor PhotoImpact 8 that I got for Free with a UK computer magazine years ago -
in this case it allowed me to pull up the shadow areas (the dark brown Labrador) without blowing out the highlights -



In that editor I can point to an area and it gives the RGB readouts and I can find that on the diagonal and lift that curve up.
This photo seems to work well from anywhere near the bottom (where I illustrated) to the mid-tone point.

[rawr]

Great continuation of your low-light shooting masterclass UnKnownVT. Some excellent, vibrant shots.

[UnknownVT]

Great continuation of your low-light shooting masterclass UnKnownVT.

Thank you for your kindness rawr

Hardly...
mostly I'm figuring out things as I go along

Curves or tone mapping can help even in high-key photos

---

Last edited by UnknownVT; 04-30-2013 at 02:51 PM.

[UnknownVT]

I use a compact camera a lot in smaller clubs - but for low available light I much prefer using my dSLR the K-5 (and previously for the past 3 years the K-x).
For backup I usually carry my compact - often I will take shots with the compact as well just to compare.

Over this weekend I took pics of a jazz band in a fairly low light venue - but the difference was mixed lighting - not just different colors but incandescent and LED.

K-5 shot (EXIF attached)


Compare to a similar shot taken on a compact (Canon G15) - EXIF attached -


Note the difference between the details of the pianist in the center of the pictures -
how the magenta (LED) light plays havoc with the K-5 shot.

details 100% crops (EXIF attached)
K-5


Canon G15


Normally for available light (no flash) shots my choice is the K-5 - because I prefer its rendition and HighISO performance.

This one case because of the mixed lighting the results of the more humble compact was a lot more preferable, to me.

[UnknownVT]

Note the difference between the details of the pianist in the center of the pictures -
how the magenta (LED) light plays havoc with the K-5 shot.
...
Normally for available light (no flash) shots my choice is the K-5 - because I prefer its rendition and HighISO performance.
This one case because of the mixed lighting the results of the more humble compact was a lot more preferable, to me.

I've pondered this "magenta problem" for quite some time (please see: Modern LED Stage Lighting & photography problems ( 1 2 3 ... Last Page) )
I think I may have stumbled on an explanation of sorts.

The hint was a while back in Post #185 in the thread above - in using "Muted Color".

This would suggest that Pentax dSLRs (at least K-5 and K-x) tend to over saturate some of the colors to cause some of this LED magenta problem (ie: red and blue)
most of the time this is not a problem under more natural lighting - as this would tend to avoid any sickly greenish tint to flesh tones etc.
in fact, I would go as far to say the K-5 and (previously the K-x) have given me some of the very best renditions at the low light jazz club using just AWB (and subtle setting) -
for lots of samples please see earlier in this thread and for 3 years' worth in Kx in Use ( 1 2 3 ... Last Page)

How do I illustrate this?

Actually I found a possible example at dpReview's comparison studio test images -
in that shot is a test chart of graduated colored panels without any separators
I selected the bottom right corner where the magenta patches were -
I used ISO3200 as that is better but close to the normal ISO I shoot at
but for the Canon G15 compact I used ISO800 as that has a much faster lens at f/1.8, which allows me to shoot at 2 stops lower sensitivity.
(Just for completeness I chose the closest Canon and Nikon dSLRs to compare at the same ISO3200)



Looking at the 100% crop patches one can see how the lowest right most two patches almost merge on the K-5 sample - and the colors are more saturated -
whereas the Canon G15 patches are quite distinct - despite being a tiny sensor in comparison -
the Canon and Nikon dSLR aren't that much better - but at least one can make out the difference between those two patches more easily.

So I guess I could try a custom color and reduce the red and blue saturation to about the same degree as the Muted color setting -
and use it at the next gig I have that "features" magenta lighting and see if the K-5 does any better.

[rawr]

This issue brings to mind a recent conversation I had online with a guy who used to be a pro photographer (weddings etc) using a Canon 5D2.

The conversation I had with him turned to some of the colour temperature problems I had been having shooting a recent show of his that used LED lighting, flooding everything with magenta or green, and which then required lots of laborious colour corrections in post. He said the following in response to one of my flickr images:

"Up until recently I was working as a photographer, and the issues of LED lighting is an interesting one. I had two Canon bodies. A 21MP Canon 5dMk11 which was a full frame sensor, and a 15 MP canon 50d with the smaller sensor. The 50d could not handle LED lighting at all, but the 5d full frame was fantastic, so it was evident that a full size sensor made all the difference under these difficult conditions."

It got me wondering: how could that be? Why should sensor size make a difference?

Your G15 images got me thinking the same thing.

The whole FF sensor size issue seems to have some validity when I look at the work of local gig photogs like (Johnny Au), who shoots with FF Nikons (D4, D700) and gets excellent natural colour temperatures, especially of skin tones, seemingly in every shot, in every gig - eg these gigs:

Photo Gallery: Sydney Comedy Festival - Paul McDermott Performs Paul Sings - Enmore Theatre (03.05.13) | the AU review

Photo Gallery: Stone Music Festival Day 1 - ANZ Stadium (20.04.13) | the AU review

Photo Gallery: Byron Bay Bluesfest 2013 Day 5 Part 2 - Tyagarah Tea Tree Farm (01.04.13) | the AU review

Maybe it is a LED light wavelength vs pixel pitch issue, but then something like the Powershot G15 should have even more problems than APS-C cameras like the K-5 or the 50D. It is all very mysterious.

[UnknownVT]

It got me wondering: how could that be? Why should sensor size make a difference?

Your G15 images got me thinking the same thing.
...
Maybe it is a LED light wavelength vs pixel pitch issue, but then something like the Powershot G15 should have even more problems than APS-C cameras like the K-5 or the 50D. It is all very mysterious.

Once again good to hear from you rawr.

I really don't think it's the sensor size -
I feel this may be a well intentioned but erroneous association of cause and effect -
I don't think FF makes any difference -
other than perhaps less noise reduction - hence less processing and loss of information,
also FF are more "professional" - and will tend to give less saturated images than most consumer and pro-sumer cameras.

The fact that an intrinsically noisier tiny sensor, in the Canon G15 compact, can also yield acceptable results -
shows that FF may not necessarily be the answer.

I'll use the 4 dSLRs you mentioned -

Canon 5DmkII (FF) and 50D (APS-C)


Nikon D4 and D700 (both FF)


Notice how much more saturated the K-5 patches are -
I already mentioned how the bottom right most patches almost merge together for the K-5 -
but also look at the next row up see how much deeper and more saturated those patches are compared to any of the other cameras.
The Canon 50D image is more saturated than the 5DmkII and that may be the cause of the problem.

My take is that Pentax dSLRs tend to saturate images - and that probably aggravates the LED/magenta problem -
but most of the time this is not a problem as it does give us nice rich colors - and most of my shots are very pleasing to me -

The whole FF sensor size issue seems to have some validity when I look at the work of local gig photogs like (Johnny Au), who shoots with FF Nikons (D4, D700) and gets excellent natural colour temperatures, especially of skin tones, seemingly in every shot, in every gig.

Those are good photos -
However I think it may have more to do with the photographer and his processing -
(otherwise anyone who uses the same equipment would be as good? )
Most photographers do NOT post their failures.

also go look at our own PentaxForums gallery - there are plenty of super results with super flesh tones and not a single LED or magenta problem......

It is only because I came across the LED and magenta problem a that I started to post failures as examples, and have been talking about this now for over 3 years.

Most of my shots do not have that problem - in fact look at your own examples - you probably never had much of a problem with Pentax APS-C dSLRs
and probably got good flesh tones (our eyes tend to be very sensitive to flesh tones - so anything slightly amiss and we would notice it)

So, do you think it was my cameras that were the primary cause of the problem?

Well, yes, and no....

Yes, because I now know Pentax dSLRs (at least the K-5 and K-x) tend to give saturated images - even when set on Natural Color,
and that tends to aggravate the problem - and I have been shooting in clubs that have a prevalence of LED and magenta lighting.

No, because - you have/had the same Pentax dSLRs -
and you had been happy with them for almost 2.5 year longer than me......

[rawr]

Thanks for your thoughts, and providing those DPR image swatches. I agree - FF sensors probably aren't automatically better at handling colour temperature than any other sensor size. There's lots more to it, including photog skills and PP.

On the subject of photog skills, I found it interesting after studying the EXIF of some of Johnny AU's shots that he uses spot AE metering almost exclusively at gigs, which surprised me. I have avoided using spot AE because it's very small 1-5% area of coverage could be a problem when shooting dynamic subjects, unless they remain centred in the frame.

But in the K-5 manual, I see an option to use multi-segment metering whilst linking the AE to the AF point (p118 of the manual), producing a kind of 'dynamic balanced spot AE'. Maybe Nikon has the same. I will give this feature a try with 5 point AF to see if it works any better under complex lighting than the centre-weighted AE metering and centre-spot AF I usually use at gigs. Since I usually try and focus on faces, maybe this will improve the skin-tones I get.

[UnknownVT]

On the subject of photog skills, I found it interesting after studying the EXIF of some of Johnny AU's shots that he uses spot AE metering almost exclusively at gigs, which surprised me. I have avoided using spot AE because it's very small 1-5% area of coverage could be a problem when shooting dynamic subjects, unless they remain centred in the frame.

But in the K-5 manual, I see an option to use multi-segment metering whilst linking the AE to the AF point (p118 of the manual), producing a kind of 'dynamic balanced spot AE'. Maybe Nikon has the same. I will give this feature a try with 5 point AF to see if it works any better under complex lighting than the centre-weighted AE metering and centre-spot AF I usually use at gigs. Since I usually try and focus on faces, maybe this will improve the skin-tones I get.

Or use Locking Exposure when the Focus is Locked on page 131 of the manual -

Locking Exposure when the Focus is Locked
Set [5. AE-L with AF Locked] in the [Custom Setting 1] menu (p.91)
to lock the exposure value while the focus is locked.

2 | On | Exposure is locked when the focus is locked.

There is no restriction on which metering mode is used -
One can set spot metering with spot focus
then when (spot) focus is locked, (spot) metering is also locked.

I just tried it and it works fine.

The reason why I can't use it is because in a lot of my shots in the dark jazz club -
the light is so low I have focus on whatever the camera can manage
and that is often other than the face.

[rawr]

Or use Locking Exposure when the Focus is Locked on page 131 of the manual

Thanks for that extra tip. And I thought I knew the camera options pretty well by now... Clearly not. I'll give that a try too and see how it works in the field.

[UnknownVT]

Thanks for that extra tip. And I thought I knew the camera options pretty well by now... Clearly not. I'll give that a try too and see how it works in the field.

No probs - that's why we're on this forum - always a great source to pick up information that we may have missed - but others have assimilated.
Locking metering when locking focus - may seem to be the way your Johnny Au uses spot metering?

[IchabodCrane]

Interesting study. I did a little playing around myself with the DPR studio tool and found that the RAW images produced more expected results on that particular magenta patch comparison with the Pentax K-5 and K-5II (those are the two I looked at). Similar to the results with a 60D and D7100. It's the JPEG image comparison where the two magenta patches almost merge on the Pentax images. This sure tells me it's the Pentax JPEG engine that is blowing out and muddying the reds.

[UnknownVT]

Interesting study. I did a little playing around myself with the DPR studio tool and found that the RAW images produced more expected results on that particular magenta patch comparison with the Pentax K-5 and K-5II (those are the two I looked at). Similar to the results with a 60D and D7100. It's the JPEG image comparison where the two magenta patches almost merge on the Pentax images. This sure tells me it's the Pentax JPEG engine that is blowing out and muddying the reds.

Great observation, thank you.

Here are the patches using RAW at ISO1600 (the 3200 are a bit too noisy) with the Canon G15 compact at 2 stops lower ISO400, due to its 2 stop advantage in a faster lens.


Indeed the K-5 RAW is much less saturated than the JPG (which probably is using Bright Color - known for saturated images).
However the K-5 is still more saturated than the G15 or 60D - surprisingly the Nikon D7000 image is almost as saturated.

Basically LED magenta causes quite severe problems anyway - on almost any camera -
even on Pentax dSLRs set on Natural Color - the only non-ideal way of mitigating this is to use Muted Color - which we know is less than satisfactory for regular general shots - that's how severe the problem is.

I used the patches merely to illustrate that Pentax dSLRs do tend to saturate the images - here we see even default ACR RAW conversion is still more saturated than other cameras -
which only goes to aggravate the situation with LED lighting - especially magenta (made up of discrete red and blue LEDs).

Thank you.

[UnknownVT]

My take is that Pentax dSLRs tend to saturate images - and that probably aggravates the LED/magenta problem -
but most of the time this is not a problem as it does give us nice rich colors - and most of my shots are very pleasing to me

I don't want all this to be taken out of all proportion.

Just want to emphasize that LED stage lighting and especially magenta (made up of red and blue LEDs) is relatively speaking a fairly specialized problem.

For most people this is not a problem, and I would say for the majority of my shots it is not a problem.

Although LED stage lighting is more and more prevalent many of the more savvy venues are using incandescent for the front lighting of the acts and now using the LEDs for the back color washes -
to me this is almost ideal - since incandescent lighting is much easier for our dSLRs to handle - and the LED back color give nice strong saturated colors that give the photos some extra snap.

From the same night as the problematic photos in post #65 above -

magenta LED, yes, but it's on the background the front lighting on the people was incandescent.


in this shot one can see quite clearly the LED lighting was high in the back....
(....there was a joke in there....)


lots of LED magenta..... but not on the person.
Notice the use of LED bars over the backdrop curtain.

same night, same venue...



the dynamics/contrast range was too high -
had to mask the front face to bring up the rest of the photo......

Different night, different venue -
direct comparison -

K-5


Canon G15 (compact)