Black & White -- Which Cameras?

[dg1]

<p>Couldn't read all the preceding, so maybe someone already mentioned this. If you are talking about in-camera B&W, the Ricoh GRD1 produces some of the nicest looking B&W I've seen. Also the Panasonic L1 has very nice in-camera B&W, with 3 different settings. These are also available on the Panasonic LX3 I believe.</p>

 

[dg1]

<p>Couldn't read all the preceding, so maybe someone already mentioned this. If you are talking about in-camera B&W, the Ricoh GRD1 produces some of the nicest looking B&W I've seen. Also the Panasonic L1 has very nice in-camera B&W, with 3 different settings. These are also available on the Panasonic LX3 I believe.</p>

 

[rustys pics]

<p>We could all post a gazillion bad jpegs until time ends and learn nothing about a good B&W print. If all you want is a decent B&W jpeg for your Flickr account, then anything will do. The suggestion that you spend some time looking at B&W prints ON paper is a good one. When I was in grad school, we all thought we were ace printers....until Nathan Lyons sent us to Eastman House to look at actual vintage prints. We all returned a humbler group and ready to learn.<br>

So really you got a range of answers. But it may be that your question was vague and those are the appropriate responses.</p>

[dan_kreithen]

<p>Hi John -<br>

That's a very nice photo. I've attached one...D700 ISO 2800, 35mm ZF Distagon, 1/40 sec, f2.8. I'll be the first to admit that you can't really tell anything about the quality of the photo at 600 pixels wide (which is I think the limit here at photo.net), but to my eye, printed, it looks very close to my Rolleiflex (with lower ISO film). Personally, I was blown away by the low-light capabilities of this camera.</p><div></div>

[rogerjporter]

<p>Godfrey, that Turbine image is beautiful. In response to the original question, i had good luck with an Olympus E-500 I had for a short while. It can be picked up rather inexpensively, has filter color control, sharpness, and contrast control right in the camera. I was always happy with the jpeg results straight from the camera. Now the flash system left something to be desired. </p>

[godfrey]

<p>Thanks Roger!</p>

<p>G</p>

[mauro_franic]

<p>Nothing like some good dear Kodak TMX:<br>

http://shutterclick.smugmug.com/gallery/4811519_F9MBv#286378736_Sqezo-X3-LB</p>

<p> </p>

[mauro_franic]

<p>You can notice something difficult to explain in B&W from digital, something that looks a bit too smooth and makes you a bit dizzy. It makes no sense this is perceived even in the small crops posted because of their tiny size they should have enough luminosity and color detail density. But that dizzy feel still hits me, and every person I discussed this that also shoots B&W film. </p>

<p>I do not have an explanation for it, just an observation to the OP: shoot a couple rolls of B&W film and you will see what I mean, and most likely never make another B&W picture from a digital camera again.</p>

[Troll]

<p>Does every thread eventually evolve into a "digital vs film" thread?<br>

Why don't you people post on the APUG Forum, where nobody notices?</p>

[Sanford]

<p>Weston's photograph of a pepper is not a great photograph worth many thousands of dollars because of it's technical excellence, it's a great photograph because it's a photo of a pepper and no one thought of photographing a pepper before.</p>

[marcodavi]

<p>For B&W out of the camera I enjoy using a Ricoh GX200. </p>

[rustys pics]

<p><em>"Does every thread eventually evolve into a "digital vs film" thread?<br /> Why don't you people post on the APUG Forum, where nobody notices?"</em><br>

<br /> I don't think it has to. I don't think anyone is telling you to shoot film, but perhaps to consider it. Why do you ask questions when you don't want to listen? In the end it's up to you which camera you choose, and in the end the tool is only part of the equation. Would you build a house without looking at existing structures? Too often the digital group gets defensive any time film is even mentioned. Please just listen and consider the responses here, and perhaps next time you'll listen to the voices you don't agree with.<em><br /> </em></p>

[godfrey]

<p>Bill, <br /> <br /> Adam Maas posted this photo to one of my other mailing lists today. I thought you might be interested: <br /> <br /> On Flickr.com:<br /> <a href="http://www.flickr.com/photos/mawz/3215481554/" target="new1">http://www.flickr.com/photos/mawz/3215481554/</a><br /> <br /> <img src="http://farm4.static.flickr.com/3093/3215481554_9fbeb8330e_o.jpg" alt="" /><br /> <br /><em>Panasonic G1, G Vario 14-45, ISO 400, 1/6, f4.5@14mm, 16:9 Aspect Ratio,<br /> Dynamic B&W mode, in-camera JPEG with sharpening only in post</em><br /> <br /> Godfrey</p>

[mauro_franic]

<p>I believe a partial explanation of the problem with B&W from digital is that once color detail is upsampled in the demosaicing algorithm, luminosity per channel is warped and even downsizing the picture cannot fix that.</p>

<p>A foveon type sensor at its native resolution (not 3x) could probably be less impacted by it.</p>

<p>The above is just an attempt at expaining it, not a statement of fact. The fact is though, in my own perception, that B&W from digital just doesn't look apealing compared to the same picture taken with film.</p>

[godfrey]

<p>mauro, </p>

<p>Sorry, but I have no idea what you mean by that. Presuming that you can produce accurate color with a bayer mosaic sensor (and all indications based upon calibration exercises prove that you can), I can see no basis for saying that the "luminosity per channel is warped" when rendering to grayscale. It just makes no sense at all. </p>

<p>There's no significant benefit to the Fovean technology for monochrome rendering work. Bayer mosaic sensors provide more spatial resolution and more than adequate chroma information to create the monochrome tonal scale.</p>

<p>Godfrey</p>

[robert lee]

<blockquote>

<p>... B&W from digital just doesn't look apealing compared to the same picture taken with film.</p>

</blockquote>

<p>I use (B&W) film and digital capture equally. What I've come to conclude is that much of the "character" generally attributed to film are really artifacts of its deficiencies, specifically the much higher noise floor and much earlier sharpness roll off. </p>

<p>Practically, a low enlargement print or contact print from film has the same crisp, grainless quality as that of a low ISO digital capture. Personally, I find these qualities generally appealing.</p>

[Troll]

<p><em>RUSS ROSENER RESONDS:</em><br>

<em>Does every thread eventually evolve into a "digital vs film" thread?<br />Why don't you people post on the APUG Forum, where nobody notices?"<br /><br /></em>I don't think it has to. I don't think anyone is telling you to shoot film, but perhaps to consider it. Why do you ask questions when you don't want to listen? In the end it's up to you which camera you choose, and in the end the tool is only part of the equation. Would you build a house without looking at existing structures? Too often the digital group gets defensive any time film is even mentioned. Please just listen and consider the responses here, and perhaps next time you'll listen to the voices you don't agree with.<br>

Russ, I am not interested in "film vs digital." I have already been there and done that and moved on. I have definitely made the switch from film to digital (except for 4x5).<em><br /></em>I have been shooting film for over 60 years, and doing my own darkroom work, everything from B&W platinum contacts to Dye Transfers. Almost any DSLR can produce color prints superior to anything since they took Kodachrome II and Dye Transfer off the market, and I believe that digital B&W has also reached that stage. The time has come for me to simplify. I just need practical information, not pontification.</p>

 

[mauro_franic]

<p>Godfrey, the Bayer Mosaic is most efficient I agree. The problem (once again just my theory) is with the demosaicing algorithm that interpolates colors. Say you only have red and yellow but it get's interpolated to orange where the colors meet. Then even downsizing carries this problem to a smaller size.</p>

<p>Robert, "a contact print from film has the same crisp, grainless quality as that of a low ISO digital capture", have you ever made a contact print? how many times you need to enlarge a print from B&W film before you see grain? Not sure what you mean by your comment, can you provide specifics?</p>

<p>Bill, "Almost any DSLR can produce color prints superior to anything since they took Kodachrome II and Dye Transfer off the market, and I believe that digital B&W has also reached that stage." Sadly you are misinformed and seem to have no desire to learn. Not sure why you would ask for people's advice. Can you post a side by side of any picture from your 60+ years of experience that justifies a comment like this. I promise you not.</p>

<p> </p>

[mauro_franic]

<p>Bill,<br>

this is how the Canon 40D looks against a 35mm TMX Scan. Upsampled to the same size. Feel free to convert to B&W. <br>

http://www.shutterclick.smugmug.com/gallery/6560295_hraSq#417849621_RBnJg-O-LB</p>

<p>here is the 40D agains a 67 TMX scan:<br>

http://www.shutterclick.smugmug.com/gallery/6302153_PLzKe#401256905_rVxpZ</p>

<p>and this is how the 40D compares to a 67 scan of TMX, Velvia 50, Velvia 100 and Provia:<br>

http://www.shutterclick.smugmug.com/gallery/6302153_PLzKe#407096809_FeZJ4-O-LB</p>

 

[godfrey]

<p><em>> the Bayer Mosaic is most efficient I agree. The problem (once<br /> > again just my theory) is with the demosaicing algorithm that<br /> > interpolates colors. Say you only have red and yellow but it<br /> > get's interpolated to orange where the colors meet. Then even<br /> > downsizing carries this problem to a smaller size.</em><br /> <br /> I'm glad you admit that this is your personal theory as it is obviously incorrect given that, as I said before, proper color calibration to very exacting standards can be achieved with current bayer matrix sensors. If you can calibrate to a color standard, calibrating to B&W tonal requirements is easy. <br /> <br /> No amount of example files you produce to prove your point changes this fact one iota. There are so many variables in the process of exposing, processing and scanning film, and for that matter in setting exposure properly with a Canon 40D, that supplying a bunch of sample files really proves nothing but the skill of the person doing the image processing to make the one look like the other.<br /> <br /> Which, for any skilled image processing person, is a piece of cake.<br /> <br /> Godfrey</p>

[robert lee]

<blockquote>

<p>... how many times you need to enlarge a print from B&W film before you see grain? ... can you provide specifics?</p>

</blockquote>

<p>My short stint contact printing 135 for proofing decades ago doesn't count for much, and in interest of full disclosure, I haven't been near an enlarger for at least that long either. However, I have failed often enough at it to appreciate what an excellent traditional wet darkroom print looks like, contact and otherwise. </p>

<p>The ability of film to record with high spatial fidelity starts to fall off earlier (relative to a digital sensor) as a function of increasing source image detail. Practically, this means an 8x10 from 135 Tri-X does not look as sharp as that from a 5D, but a 6x7 TMY sourced scan would be indistinguishable from the digital. Basically, if the goal is a print of high sharpness and low noise, it's easier to push the envelope with a digital source. It's instructive to look at the MTF curves for some common B&W films; Google around for that of a typical DSLR small format imager.</p>

<p>Digital post B&W conversion makes a lot of sense given the additional flexibility of filter selection and faster workflow. In camera B&W conversion is fine too for that matter; the full color file is always available from the RAW anyways.</p>

<p>As for appearance of grain, it depends on the film and development doesn't it. For Tri-X in Xtol 1:1, grain starts to appear on the print at more than 2X or 3X enlargement. TMY is better; Acros is much better.</p>

[Brad_]

<p>Getting *outstanding* B&W today is SO easy. Don't understand why there is so much hand-wringing over this today.</p>

<p>Not rocket science. Sheesh...</p>

[bill_tuthill]

<p>Well Brad, somebody was belly-aching in another thread about how 16-bits per color are required for B&W, and had samples to prove it [<a href="../digital-darkroom-forum/00Rmgv">add by Benjamin Schaefer</a>]. Seems ludicrous, but perhaps Mauro's theory helps explain it. Many imaging algorithms are optimized for color and when dealing with grayscale allocate only ~ 256 tones. For this reason, we can't really argue about this problem by posting B&W images on the web.</p>

[godfrey]

<p>It more than *seems* ludicrous. It is ludicrous. </p>

<p>G</p>

[Ed_Ingold]

<p>I don't see any quality issues rendering B&W in 8 bits rather than 16 bits. However where you start and how you get there can make a significant difference.</p>

<p>I've never had good results scanning B&W at 8 bits/channel. The tonality tends to be very contrasty compared to a 16 bit scan. When you shoot B&W in a DSLR, the original image is in 16 bit color, then converted to B&W and resampled at 8 bits/channel for a JPEG output. While that may work well, a 16 bit/channel master gives you a lot more room to make adjustments, including the use of the Channel Mixer.</p>

<p>If the color image is simply desaturated, as in older DSLRs or inept use of Photoshop, the results tend to be flat and skin tones rather dark. B&W film is nearly twice as sensitive to blue light as to red or green, close to 50/25/25, whereas a DSLR is tuned to be equally sensitive to all bands of light. This response and the effects of common filters are easily emulated in Channel Mixer. Results I've seen from a D300 and D700 are very believable - consistent with a conversion taking the sensitivity of B&W into account. Once you have the dynamic range of the subject tucked comfortably into a 16 bit/channel file, downconverting to 8 bits changes the appearance very little, color or B&W.</p>