Using a DSLR for landscape photography

[stan_belyaev]

Godfrey,

You don't need to be so emotional.... It is just photograpgy. It is not a matter of life and death.

 

Pls, check the histogram and see for yourself that the highlights (sky) are gone. You may want to apologize after you do that.

 

There are plenty of books on how to make a good photo. I realy like Thornton's "Elements" (I was lucky to find one in England) and "Egde of darkness". Todd's "Elements of B&W photography" is also good. Disregard everything that related to film and concentrate on what one should be able to see on a photograph. You will find a lot of useful information on highlights, shadow and everything in between.

After you done you may want to look at your pictures one more time...

Stan

[godfrey]

Yes, Stan ... I know the pixel values in several of the white sky regions are 255,255,255. Does

it occur to you that if the sky is white and without detail that those are the proper pixel

values?

 

Does your film render white, blank sky with interesting details? Quite fascinating if it does.

 

Godfrey

[godfrey]

BTW, can I presume from your comments that you have never seen a white sky? or fog? Or

would you prefer the photograph to contain artificially constructed gray sky areas of pillowy

clouds, or grainy texture? They weren't there but I can produce both for you, if you prefer.

 

If you don't like the photograph, just say so. Don't tell me what the scene looked like, you

cannnot know that. The photograph renders the scene I was looking to capture with exquisite

fidelity.

 

Godfrey

[stan_belyaev]

Another example....

Picture was taken with Contax TVSIII+Kodak UC400. (a tiny camera with a fantastic lens)

Try to reproduce it with any digital camera and any software.

It could be a good lesson.<div></div>

[stan_belyaev]

I really don't want to carry on this discussion any further. It is pointless. I can not/do not want to explain basic things about the exposure and histograms here...

Sorry, I didn't expect that you would bo so frustrated/upset.

S

[godfrey]

Thanks for your example. Clean sky, sun values at 255,255,255, no definition on the

surfer/bather.

<br><br>

Sorry for a low quality JPEG:

<center>

<a href="http://homepage.mac.com/ramarren/photo/PAW4/large/14-half.jpg"

target=new><b>Click to see "Santa Cruz Sunset" in a new window</b></a><br>

Canon EOS 10D + 50mm f/1.4 lens<br>

©2004 by Godfrey DiGiorgi<br>

</center>

[godfrey]

<i>Stan:

<br>

I really don't want to carry on this discussion any further. It is pointless.... </i>

<br><br>

Yes, your comments have been pointless.

<br><br>

Godfrey

[mike_foley2]

<P><i>Take for example your first response above, you present an image from your 10D (nice shot BTW) then state, "But much very good work can be done with 35mm film, and better work can be done with DSLRs in my opinion" so if your point is that a 6MP is better then 35mm film, where is it?</i></P>

<P></P>

<P>color. noise. sharpness. and low contrast detail. all of which is demonstrated in your own tests.</P>

<P></P>

<P>the finer films win in terms of high contrast detail, the aspect of your tests you like to focus on.</P>

<P></P>

<P><i>I have presented all sides in the best that they can be and provided more then enough information to have them indepently verified by anyone anywhere in te world.</i></P>

<P></P>

<P>anyone who looks up some of your past threads on this site knows this to be...um...less than completely true and honest. i recall you setting a 20D to a high iso, 800 or 1600, and presenting it alongside 100 film, until i called you on it, after which you silently removed the image.</P>

<P></P>

<P>http://www.photo.net/bboard/q-and-a-fetch-msg?msg_id=00Hk4X</P>

<P></P>

<P>that's to say nothing of some of the raging fights you've had, after which you've been known to swap digital images for improved versions.</P>

<P></P>

<P>further, you never scale your scan and digital crops to the same size. you know full well that the bigger square biases the observer.</P>

<P></P>

<P>sorry, but you are very biased when it comes to this debate. why i do not know.</P>

<P></P>

<P><i>Canon itself only compares their fullframe DSLRs to 35mm film stating, "Canon?s full-frame sensors have reached another image quality milestone as well. Their gradations and dynamic range are now the equal of the best positive films, and their resolution and lack of grain are superior. No smaller sensor has achieved this level of performance." You'll notice that they lumped yours as well as my DSLR as not being equal to 35mm film and not even the latitude of slide film.</i></P>

<P></P>

<P>oh good grief. they say their full frame sensors are superior in resolution and lack of grain. they say nothing about other sensors being inferior in these respects.</P>

<P></P>

<P>you tend to read want you want les. i'm surprised you haven't starting talking about reichmann yet.</P>

<P></P>

<P><i>This also addresses the claim above that a 5D - or in your case that a 1Ds MKII, that only compares to 35mm somehow matches or even exceeds MF. You may want to reevaluate that position about MF as you can see examples of it in my Resolution album as today's DSLRs are not in that realm as you correctly ponted out in your first response.</i></P>

<P></P>

<P>http://www.photo.net/bboard/big-image?bboard_upload_id=32819284</P>

<P></P>

<P>6x7 seems to resolve some of the letter edges better, but the d2x image has better color and contrast. the d2x advantage in color and contrast is much greater than the 6x7 advantage in fine detail at this enlargement. unless i needed to print larger than about 60 inches, i would take the d2x file any day. heck, i would probably choose it for even larger sizes.</P>

<P></P>

<P>if the d2x is clearly in the realm, which it is, then so are the 5d and 1ds.</P>

<P></P>

<P>while we're at it, i would take the d2x shots over any 35mm scan you've presented.</P>

<P></P>

<P><i>Rob, I know for a fact that sharpening does not add missing detail nor for that matter any version of Genuine Fractals or other similar program/plugin.</i></P>

<P></P>

<P>http://www.clarkvision.com/imagedetail/image-restoration1/index.html</P>

<P></P>

<P><i>The characteristic of latitude is a little tougher to quantify but anyone who has ever used a digicam knows that it blows out highlights.</i></P>

<P></P>

<P>http://www.clarkvision.com/imagedetail/dynamicrange2/index.html</P>

<P></P>

<P>he didn't use a portrait or bw film unfortunately. but he still blew the myth that film in general has more latitude right out of the water. dslrs do pretty well on latitude when exposed correctly.</P>

<P></P>

<P>dslrs blow highlights because people don't expose properly. with the low noise and the dramatic cutoff on the high end, you make sure you don't clip your highlights and worry about recovering shadow detail, if necessary, in post. learn that and you will rarely have a latitude problem with digital. certainly less often than you would with slide and some neg films.</P>

<P></P>

<P><i>These two qualities are less for any DSLRs as compored to 35mm film and as I have already pointed out Canon itself has acknowledged it is so for anything less then only its full frame DSLRs - which you again chose to ignore.</i></P>

<P></P>

<P>he's ignoring it because you're misrepresenting it.</P>

<P></P>

<P><i>It is intimidating to call them amateurish I suppose</i></P>

<P></P>

<P>they are amateurish. a professional test would have used measured line charts so the differences could be quantified and all guesswork removed.</P>

<P></P>

<P>do you think test engineers for nasa have people give their impressions of engine thrust based on how loud a test is, or do you think it is measured and quantified?</P>

<P></P>

<P>"gee bob, let's go with rocket X. the flames off rocket Y just didn't look as pretty, so i'm betting rocket X has more thrust."</P>

<P></P>

<P><i>For instance Robert's test above is fine, if it were conducted at the same time with the same lens at the same settings, with the same weather conditions and more importantly at the same location - provided it hasn't changed, while mine can be conducted anywhere, anytime by anybody.</i></P>

<P></P>

<P>robert's test made the film look sick. his results are typical of the real world, where film is grainy, soft, and dull, while digital jumps off the page at the viewer.</P>

<P></P>

<P>"coming from a background of over two decades of being a test engineer."</P>

<P></P>

<P>which tests have you professionally performed not by measuring and quantifying the data, but by simply giving an opinion on what seemed better?</P>

[mike_foley2]

<P><i>I have been a film camera user for a number of years, and, like many others, I am considering moving to the digital realm. One of my favorite types of photography is landscapes, and I want to be able to achieve at least as good image quality in 13x19 inch prints as I now do with Astia 100f. However, I have read on more than one occasion that DSLRs are still not as good as 35mm film for landscapes. Is there any truth to this statement?</i></P>

<P></P>

<P>overall they are better, because at the end of the day their advantages out weigh 35mm film's advantages.</P>

<P></P>

<P>BUT...you will have to be careful about the lenses you choose. wide angle is a bit more challenging for most dslrs because of their smaller sensors and the cropped view it produces. there are now some good, sharp, made-for-aps-digital lenses for wide angle work.</P>

<P></P>

<P>you can meet your goals (13x19 comparable to astia prints) with any of the latest dslrs.</P>

[mike_foley2]

<P><i>Another example.... Picture was taken with Contax TVSIII+Kodak UC400. (a tiny camera with a fantastic lens) Try to reproduce it with any digital camera and any software. It could be a good lesson.</i></P>

<P></P>

<P>stan - do you ever thumb through the photos on photo.net? there are plenty of digital beach sunset photos, most with equal or better latitude and detail.</P>

<P></P>

<P>why would this be hard to reproduce digitally?</P>

[rob_landry]

<i>robert's test made the film look sick. his results are typical of the real world, where film is grainy, soft, and dull, while digital jumps off the page at the viewer.</i><p>

 

Mike, I dunno about Robert's results, but you may want to take a look at the results of the test I posted a few lines up. The reason for most of the poor film results you see is because scanning is not easy; it's much harder to achieve a good scan than it is a good capture from a DSLR.<p>

 

Regardless of the results I achieved, I still shoot much more digital these days than 35mm as the need to enlarge beyond 11x17 in the minature formats is few and far between. So, in large part, even my test is acedemic at best. Still though, even when shooting with small formats, when I need Velvia's palette, I shoot Velvia and leave the DSLR in the bag.

[robert lee]

The thing to keep in mind about the three images I posted up thread is that these are 100% crops. On the typical 17 inch LCD display, these are life size sections of a image that would be 36 inches by 24 inches. Very, very few 135 color film can be enlarged 25x and maintain critical sharpness; 12x18 is about the reasonable max.

 

"...scanning is not easy; it's much harder to achieve a good scan than it is a good capture from a DSLR."

 

This is true. However, neither is scanning rocket science. Using a "good" film can make life lots easier if maximizing resolution is a goal. Take a look at the beach scene scans from Gold 100 at http://www.photo.net/bboard/q-and-a-fetch-msg?msg_id=00HEsT

 

I've seen film datasheet numbers to be a pretty good predictor of scan quality. In general, 35mm reversal film lags negatives in usable resolution. Even Velvia isn't stellar, relatively, in this respect.

[Allen Herbert]

My cam is better than yours, so there. I'll puff and puff until i blow your house down.

 

So there.

[Allen Herbert]

BTW, I am considering one of the new generation of 10MP DSLRs.

 

Consider a new generation of vision.

 

A toy is nice, sort of like eating candy, lacks substance.

[rob_landry]

<i>This is true. However, neither is scanning rocket science.</i><p>

 

Well, rocket science it ain't for sure, but it's still much harder to get decent top-notch results from film scans in 35mm. I don't know if you looked at the results I posted in the link above, but I included my test parameters in the gallery when you click on the link. To obtain the best results from 35mm, you really need to start with the best scanner you can lay your hands on. Before I purchased my last scanner, I did a lot of testing and the Minolta 5400 (MK I) was the best of the bunch and bested all of the 4000 ppi units out there and even some drum scanners. The extra pixels really do make a difference if your film has the detail. Second of all, I temporarily glass mount any frames I am going to scan (for print) to ensure optimum flatness. Third, I always focus the scanner manually and make multiple test crops to ensure I determine the optimum focus point. Fourth, I always run my scans through Noise Ninja for grain reduction and sharpening.<p>

 

By the way, the Provia scan crop you posted above does not seem to be very sharp; you should see more grain aliasing than what appears there. Also, I noticed that your film scan crops are almost the same magnification as the XT crop. Did you upsample the XT crop to match the pixel dimensions of the film scan or change your lens focal length/shooting position to compensate for the FOV difference?

[robert lee]

"...the results I posted in the link above, but I included my test parameters in the gallery when you click on the link."

 

No, the link doesn't work for me.

 

From your description, you're a lot more meticulous on the scans than I am.

 

I have a CS5000. The machine was used primarily to archive decades worth of family photos, so high throughput was important. Generally, I run everything on default. The autofocus, for example, was spot on the few times it was manually verified, so I don't bother anymore.

 

I definitely agree though, that there's more information on some films above 4000dpi. Even if there isn't for the particular emulsion, the higher the scan resolution, the more effectively grain reduction can be applied in post; I'd do a 8000dpi scan everytime if my equipment were capable of it.

 

"By the way, the Provia scan crop you posted above does not seem to be very sharp..."

 

Hmmm... I really didn't give this a second thought. The image is pretty typical of my Provia 100 scans, both mounted, and unmounted through the strip feed. The strip feed is used for the negative scans as well, so the sharpness difference is probably not attributable to the scanning workflow.

 

"Did you upsample the XT crop ... focal length/shooting position to compensate for the FOV difference?"

 

No, definitely sloppy here, eh. I was interested in comparing Reala to Provia, so the XT shot was just by happenstance; I already had a pretty good feel for the particulars of the digital gear by this time. Check the original thread though. The 100% crop features are about the same sizes relative to the full frame in both.

[rob_landry]

<i>No, the link doesn't work for me.</I><p>

 

Sorry, I forgot to make the presentation viewable publicly. You should be able to view it now.<p>

 

 

<a href="http://www.photo.net/photodb/presentation.tcl?presentation_id=332985">LINK</a>

[robert lee]

No, it's not working. Still gets an "Invalid presentation id" error.

[robert lee]

Oh, it's working now, kinda. Quite the pronounced resolution difference I must say.

 

Actually, I'm a bit surprised by the seeming roughly comparable dynamic range between the D50 and Velvia. This is an artifact of this particular conversion to JPG, no? I would expect the RAW file to have more depth.

[rob_landry]

Yeah, you have to be logged in to access the link as it's arranged as a presentation. I had to do it this way so I could include the text.

 

In regards to the seemingly similar contrast between the two, I just checked the EXIF and it was shot with the normal contrast setting and auto tone response curve. Auto white balance was used. It was an NEF (RAW) file processed in Nikon Capture for output to TIF. Since I was only concerned with resolution for this test, I chose the normal settings on the camera.

 

Anyway, despite the lower resolution, I still use and enjoy the DSLR immensely. Of course, Velvia is also something I enjoy which is why I am glad Fuji has announced that they are re-releasing the original RVP 50 in the spring.

[rob_landry]

Okay, I have placed the images in another, more easily accessible folder. They can be seen at:<p>

 

<a href="http://www.photo.net/photodb/folder?folder_id=660906">NEW LINK</a>