IR foreheads?

[Guest guy_mancuso]

Advertisement (gone after registration)

Little wonder brush and all gone and it's good to leave a little there as a highlight

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[jaapv]

Howard, this is certainly doable, but doesn't capture pictures that look alike. In burst mode (bracketing function), the pictures will all occur in 1.5 seconds. The sitters' expressions are more likely to be similar.

 

Yup, it's clugey and asks Leica to give us a digicam function that is probably not on their radar, but it has the specific function of bracketing so we can control blown highlights while, at the same time, getting close to similar expressions. In the interest of controlling lighting variables, Leica might be interested in this type of bracketing.

 

We don't need to do this for grass, trees, cows, or Morgan sports cars.

 

Regards,

 

I guess I'm just an old fogey set in his manual ways... I am no fan of these newfangled machines with little goblins sitting under the hood driving the photograph. Which doesn't mean I don't see where you are coming from, Bill. But I fear Leica philosophy is on my side

[Guest sirvine]

Has any tried using the recovery tool from Lightroom to address this?

 

I find that using the exposure slider to recover highlights is hopelessly inferior to whatever tone mapping magic Adobe put into the recovery slider.

[wparsonsgisnet]

Howard and Jaap, I am in agreement with your statement of Leica's minimalist philosophy. I suggest the bracketing function only in the sense that (1) it could help with the blown highlights problem, and (2) it is not so outlandish a suggestion for a 21st-century camera.

 

Leica always finds very imaginative ways to design and build their cameras. I have used M's, exclusively, since 1970.

 

What I would really like to do here is to register my amazement at this forum, at Leica's answering Guy's list of issues, and at their approach to us.

 

There have been mentions above that this is unusual. It's so unusual that it's off the chart. Thanks to Leica for providing this forum and for being so open and accepting with us. Also ... thanks for the M8.

 

Regards to all,

[rwfreund]

They are all specular highlights.

They can be one or MORE stops brighter then "near white"

If the exposure is such that the highlight is recorded as the sensor's maximum value, then ther is no amount of digital processing that can recover it.

So it comes down to what dynamic range the medium can record and then how we meter and expose to fit that range within the capability of the media. I did a lot of spot metering and zone system developing of 4x5s in the past to manage this same problem in silver chemistry film.

I think that the digital equivalent of this is dynamic gamma, which seems to be implemented in some digital cameras (the D200 for instance, but not the 5D). The other solution is to meter carefully so that the highlights are not blown out and then do the gamma game in PS.

I can live with or without the automatic assist, although the automatic contrast control makes it easier for more casual photos.

-bob

[Jamie Roberts]

Yes, they're speculars, for sure.

 

The M8 is fine with highlights relative to other digicams, and blows to white much more gradually and seemlessly than cameras with lower-bit-depth AD converters.

 

I've compared the M8 with the DMR and it apparently has a stop or two on it. This is likely because the noise floor of the DMR is so much higher than the M8 (and the noise floor means less DR).

 

This is a "quality of light" issue. Yes, amateurs don't have soft boxes, or those kinds of choices, but you need to learn to see light that looks like a softbox. Sometimes gesture or moment will outweigh the need for good light, and that's the compromise we all make all the time.

 

But working with available light means looking for low contrast, very soft light for most portraits; or knowing how to use hard light and exposing (like slide film) for the highlights. Don't need bracketing there... just remember that the M8 meter is pretty darned accurate (I've found a lot of digicam meters expose "to the left" to preserve highlights, as others have said. As Guy pointed out, the DMR and M8 let you mess up )

[ho_co]

Advertisement (gone after registration)

This is the wrong place to raise this question, so it's about time I raise it--I'm ready for the defamation of my character and intellect that will follow.

 

Isn't this "blown highlights" issue just a photographer's issue? Does anyone not photographically trained care?

 

A photo dealer I called on with Leica kept showing me an 8x10 b/w print (nice snapshot) of his son made with another brand of equipment, and telling me that no camera could do better than that. Finally in exasperation I pointed out a large and very distracting area where sunlight had blocked up the negative and all he had was paper white.

 

I shouldn't have shown it to him. He was crestfallen. All he had noticed up to that point was that his little boy had a nice expression and that the print was sharp.

 

But we "as photographers" prove our abilities by the efforts named here. (BTW, some very good pointers, guys!) But does the client care? Or just the other photographers you know?

 

So, please: Shoot me down.

 

--HC

[sdai]

I agree, Howard ... sometimes, the subject/main character in the photograph can be very distracting. LOL

[ho_co]

I suggest the bracketing function only in the sense that (1) it could help with the blown highlights problem, and (2) it is not so outlandish a suggestion for a 21st-century camera.

Bill--

You defend your point excellently. Point 2 is superb for its irony.

 

When I worked for Leica I presented the R3 to my biggest account. The dealer was unimpressed ("Looks like a Minolta," "Doesn't feel like a Leicaflex," "Too cluttered").

 

My response was, "But that's not the point! Don't you see? It has an automatic mode! Leica has entered the 20th Century!"

 

That one point took us past all his objections, and he became a big supporter of the camera.

 

Despite my justifications, Bill, you're right. Bracketing wouldn't add that much in complication and would add a lot in convenience.

 

--HC

[ho_co]

I agree, Howard ... sometimes, the subject/main character in the photograph can be very distracting. LOL

Simon--The moderators ought to award you 50 points for that one!!!

[carstenw]

Howard, you worked for Leica??? Tell all!

[rwfreund]

Long term, better than bracketing would be an extended sensitivity range sensor and 16 bit output. If linearly encoded, each extra bit is a stop worth of bracket.

-bob

[rosuna]

I've compared the M8 with the DMR and it apparently has a stop or two on it. This is likely because the noise floor of the DMR is so much higher than the M8 (and the noise floor means less DR).

 

That is very interesting. Thanks Jamie. Dou you think the M8's images would benefit from DNGs with a wider tonal space (linear 16-bits or non-linear 10 or 12 bits)?

[ennjott]

IR-radiating foreheads are caused by a sickness called Morbus Rosso. It can be treated with mashed garlic applied to the shiny spots.

 

Of course you cannot recover highlights from the 8 bit DNG. That's exactly what I have been saying all along. It was already clear from the first DNG that appeared on the net (from Photokina).

Although even with a "true" 12 or 14 bits linear raw, you only get about one more stop of DR. Most of the time, this means the spec(tac)ular lights shrink a bit, but don't entirely disappear. To fix this, you must either use negative film or proper lighting like Guy..

[Jamie Roberts]

@ Rubén, I'm really not technically sure there aren't more than 8 bits worth of data in the current compressed M8 DNG files.

 

I've asked Leica directly for some technical advice, and they will get back to me on this. In the meantime, yes, I'd like to have the option of an uncompressed 14bpp file (padded to 16bpp) file out of the M8, like the DMR.

 

I suspect, however, that unless you're manipulating skies or other natural gradients (which I still do quite a lot!), you won't see much difference between the compressed or uncompressed files.

 

Typically, too, IIRC, you get more useful "stops of exposure" in the shadows relative to the noise floor when you add levels in terms of bit depth. Yes, you have more levels everywhere, including the highlights, but it's when you get to extreme shadows and you're only left with 4 (!!) levels over noise that you encounter banding, posterization, etc...

 

Of course, all of this is bounded by noise. If you have a high bit depth sensor, but high noise as well, you won't see the extra information anyway.

 

@ NJ, as for holding highlights, the M8 does as well--or better--than anything else I've seen.

 

But on any system--regardless of relative DR--when you blow them they're blown--you can't recover clipped highlights at the RAW level.

 

You can, however, recover RAW data that is there but not represented in the JPEG representation, or RAW highlight data that is clipped due to the tone or gamma compensation of the RAW converter itself.

 

This is why, for example, ACR does a better job with Canon highlights (slightly) than C1, and both do a much, much better job than DPP. This is why you need multiple RAW programs sometimes.

 

Overall, I still find C1, in "linear" mode (which isn't really linear at all since it's still gamma-corrected) maintains the best highlights, unless you want to use something like DCRAW (which will output uncorrected linear information from the RAW files).

 

How do you get around this clipping limitation of the sensor? Well, with something like an M8 (and the DMR, for that matter) you can actually underexpose then push the images, again, at a low enough noise level (ISO) that you're not obliterating information. Since the M8 has much less noise than the DMR, it is simply more flexible in this regard.

 

But more importantly, you need to see the right kind of natural light for digital. It's a bit different than film, because you can mess up so much more easily with highlights and upper mid-tones.

 

Again, for portraits, I would look for very soft, very diffuse light. You'll be surprised what happens there. It's also all around you, even on very sunny days, but you need to look for natural reflectors, like concrete, glass windows, or aluminum.

 

Or if you're out in the full sun, you have to use that harsh light to your advantage. Either by adding light (with, say, a flash or reflector) or by subtracting it from the subject.

 

You absolutely don't need softboxes or studio light or any other to get "proper lighting" to get this light, but you need to look for it, and you need to see it, too. It is everywhere but, perhaps, a desert at noon Or some of the outdoor receptions I've shot (also much like a desert at noon).

 

You also need to keep in mind everything about ratios, position of key light and fill and incidental light when you're outside--just like you are in a studio.

[stunsworth]

Jamie, I find nothing to disagree with in what you've written. It's all about light and understanding how it works in the final image. Which is why when you first start taking photographs you look for bright sunny days, and as you get older you start to look forward to a few clouds :-)