Ad
Ad
Ad
Pages: « 1 [2] 3 4 »   Bottom of Page
Print
Author Topic: Film versus Digital: the current comparison  (Read 14297 times)
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7670


WWW
« Reply #20 on: January 10, 2013, 12:36:19 AM »
ReplyReply

Hi,

What I said was that I measured MTF on a Velvia scan Tim sent me and compared it with MTF measured on an IQ180 image that I also got from Tim. MTF data on the Velvia indicated that it would outresolve the IQ180, for high contrast detail. As far as I can recall Tim has samples using microscope demonstrating that this is the case.

Click here for full size: http://www.landscapegb.com/wp-content/uploads/2011/12/miroscope-700.jpg



But image quality is a lot more than resolution. My experience is that scanned film is noisy, but I have no 10000 PPI drum scanner like Tim has.

I would also add that it seem that many photographers prefer the "film look". Personally I prefer digital. The real issue I think is that it takes a lot of involvement to extract good results from film when scanning is involved.

Best regards
Erik

Why do you doubt it?
His test seems pretty clear to me. http://static.timparkin.co.uk/static/tmp/cameratest-2/large.html
That is a good test showing both methods at what they are good at.

The 8x10 film clearly whumps (technical term Wink ) the IQ180 on everything but the shadows. The digital has great ability to show what is in the darker areas.

To my eye the provia 4x5 with NR and sharpening is better than the IQ180. It is also a pain to use relative. If you miss the shot you may have to wait years for another trip. If you go digital and your last shot on the card is botched somehow you can delete the redo. If you are out of film that is it.

The Mamiya 7 is outclassed by the IQ180. Maybe a pure B/W would be similar on resolution. Everything else is lagging.

This test shows someone can get in the game of high res prints with a cheap (now) film camera. They probably need the discipline of a master craftsman to do it all right consistently.

Digital gives you flexibility for a lot of money.

Now in the 35mm game DSLRs are starting to wipe out film. You cant be in the ballpark against the D800. You simply have no ability to do what DSLRs are doing now with quality at ISO 1600-6400.

Thanks Tim for all the work. Its a totally convincing test to me.
« Last Edit: January 10, 2013, 12:57:04 AM by ErikKaffehr » Logged

Fine_Art
Sr. Member
****
Offline Offline

Posts: 1099


« Reply #21 on: January 10, 2013, 03:41:01 PM »
ReplyReply

Yes, scanned film is noisy. A good NR software does wonders on film grain.

The combination of high res scan, good NR, deconvolution sharpening really works well. People do NR and sharpening naturally in their raw conversion. Why feel doing the same on film is cheating somehow? On many shots not doing those two operations on digital would yield rubish. Not doing them on film yields relative rubish unless you need the grain for artistic effect.

Both methods are completely workable for very high quality images.

Logged
Fine_Art
Sr. Member
****
Offline Offline

Posts: 1099


« Reply #22 on: January 10, 2013, 04:52:12 PM »
ReplyReply

From Tim's crops I added NR to the Provia then sharpened, then did a mild soften to remove the sharpening artifacts. That is how I would send it to the printer. The detail is good.

Compare that to the IQ180 shot. We all know the green channel shows off digital at its best due to 2x as many green pixels. It would be a much bigger blowout if the image was red flowers or blue. The digital is nice on shadow areas. It fails with shades of green relative to the film. It all looks a similar yellowish green


http://farm9.staticflickr.com/8336/8369011622_4ede28bb7a_o.jpg


http://farm9.staticflickr.com/8073/8369010544_e52e752210_o.jpg

The links are large versions from Tim's original size.
« Last Edit: January 10, 2013, 05:02:56 PM by Fine_Art » Logged
Fine_Art
Sr. Member
****
Offline Offline

Posts: 1099


« Reply #23 on: January 11, 2013, 06:20:00 PM »
ReplyReply

Added Tim's Velvia 8x10 with my NR, Sharpening.


http://farm9.staticflickr.com/8072/8370929497_a83895788f_o.jpg

Original size linked.

Again, these are crops provided by Tim in his article linked here http://static.timparkin.co.uk/static/tmp/cameratest-2/large.html

Logged
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7670


WWW
« Reply #24 on: January 11, 2013, 09:57:50 PM »
ReplyReply

Hi,

You know, sometimes Swedish winter has a feel of Tri-X underexposed three stops. Other than that there is nothing natural about film. Which film by the way? Ektachrome? Provia? Velvia. By the way, have you ever seen grainy blue California sky?

Does the M6 has better DR than a Canon 5DII? No, it has no DR at all! The film you put into it has DR. Velvia has about 5 stops, a bit less than Canon 5DII at 12800 ISO. Negative color may have 16 stops.

Next question is, what do you do with the stuff? Are you looking at projected slides? Great! If you need prints or screen the image needs to be printed by a lab or yourself. The paper you print on has nothing natural in it. Now, Kodak and Fuji have spent a lot of effort on developing papers, films and chemistry that match and give nice colors.

If you are into serious photography with film you either have a dark room or scan. Scanning is a can of worm of it's own. Most scanners don't have enough dynamic range to handle Velvia. My experience may even be that they even don't have enough DR to correctly handle Ektar 100. Some people have drum scanners that can eak out every detail from a slide or negative. I'm pretty sure that scanning has a steep learning curve.

My take is really that digital is mostly preferable. Large format film in combination with an excellent scanner and scanner operator can give excellent results.

Best regards
Erik


This is a nubie question but I need to ask it to help resolve a discussion with a friend who has recently switched from digital - a Leica M9- to film - a Leica M6 I think.
He says that the "truly great photographers" only use film and I found this astonishing given the huge possibilities of digital technology.

His arguments for film are

- film has a more natural look
- film has greater dynamic range
- film images are more beautiful. They have a certain something that is difficult to pinpoint.

So I wonder what your opinion is. For instance is it really true that a Canon 5d Mark 2 has less dynamic range than a Leica M6?
Aren't there techniques - such as adding grain or noise - that make digital images look as natural as film images?
Why do so many people say that film has that certain something? Why isn't this purely a matter of taste?


Logged

Fine_Art
Sr. Member
****
Offline Offline

Posts: 1099


« Reply #25 on: January 12, 2013, 02:20:23 AM »
ReplyReply

On the whole I agree digital is preferable. That is, it is much easier to use. I have not taken the cinder block out to shoot in a couple years.

135 film I really enjoy when I do it. It is refreshing to be focused on the process of making the shot. Everything culminates in the click then it is over, you forget about it until the delight of picking up your roll. I find I am in the zone with film far more than digital. Digital often feels mechanistic. Take shot. Check histogram. Check focus.

My scanner has no problem with Velvia. Every scanner can probably handle slide film. Ektar is negative if I remember. The DR of negative is much greater than slide. I find negative is too compressed in the tones.

Logged
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7670


WWW
« Reply #26 on: January 12, 2013, 02:59:57 AM »
ReplyReply

Hi,

What scanner do you have? Velvia is known to be difficult to scan.

You are mixing up DR and DMAX, by the way.

Best regards
Erik


On the whole I agree digital is preferable. That is, it is much easier to use. I have not taken the cinder block out to shoot in a couple years.

135 film I really enjoy when I do it. It is refreshing to be focused on the process of making the shot. Everything culminates in the click then it is over, you forget about it until the delight of picking up your roll. I find I am in the zone with film far more than digital. Digital often feels mechanistic. Take shot. Check histogram. Check focus.

My scanner has no problem with Velvia. Every scanner can probably handle slide film. Ektar is negative if I remember. The DR of negative is much greater than slide. I find negative is too compressed in the tones.


Logged

Fine_Art
Sr. Member
****
Offline Offline

Posts: 1099


« Reply #27 on: January 12, 2013, 03:08:54 AM »
ReplyReply

Epson Perfection 4490
« Last Edit: January 12, 2013, 03:15:24 PM by Fine_Art » Logged
Fine_Art
Sr. Member
****
Offline Offline

Posts: 1099


« Reply #28 on: January 12, 2013, 03:13:57 AM »
ReplyReply

How? DR is a function of the film or sensor. Dmax is a function of a scanner.

My scanner has a dmax around 4. 3.8? A negative film that can be 16 stops has to be compressed into the 3.8. Black still shows as black, white is still white in the output. Tones look very similar. With slide film tones look dynamic.
Logged
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7670


WWW
« Reply #29 on: January 12, 2013, 04:20:01 AM »
ReplyReply



How? DR is a function of the film or sensor. Dmax is a function of a scanner.

My scanner has a dmax around 4. 3.8? A negative film that can be 16 stops has to be compressed into the 3.8. Black still shows as black, white is still white in the output. Tones look very similar. With slide film tones look dynamic.
Film has a DMAX (maximum density) it is given in spec sheets. If you check the enclosed see that maximum density (DMAX) is around 3.8 for the blue channel and 3.4 for the red channel. So DMAX for that film is 3.8.

DR for film is the range of exposure that the film can separate, so you would check Exposure axis to calculate dynamic range. DR is not easily defined, because of the inverted S-shape, but I would say about 1.5 that is about 5 stop.

The enclosed figure comes from Velvia-100 spec sheet.

I also enclose a screen dump of a scanned Velvia 50 slide (with very high contrast) the one on the left was drum scanned while the one on the right is scanned on my Minolta Dimage Scan Multi Pro. If your scanner handles a DMAX of 4.0 it should be able to handle darkest detail on Velvia. Most scanners don't.

Best regards
Erik
Logged

Fine_Art
Sr. Member
****
Offline Offline

Posts: 1099


« Reply #30 on: January 12, 2013, 03:21:13 PM »
ReplyReply

Thanks, I have never seen DMAX in reference to film before. I have seen it on scanner reviews.

I remember comparing velvia scanner output to visual inspection of the film with a high magnification telescope eyepiece. The detail seemed the same in darker areas.

Maybe you need Hamrick's vuescan. it does a multi-pass with different scanner light intensities.
Logged
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7670


WWW
« Reply #31 on: January 12, 2013, 04:10:51 PM »
ReplyReply

Hi!

I used Vuescan. It does multipass, but there is a tendency to double contours.

By the way I happened to work with Ed Hamrick on Vuescan, the drivers for the DSMP are developed for my scanner ;-)

Best regards
Erik

Thanks, I have never seen DMAX in reference to film before. I have seen it on scanner reviews.

I remember comparing velvia scanner output to visual inspection of the film with a high magnification telescope eyepiece. The detail seemed the same in darker areas.

Maybe you need Hamrick's vuescan. it does a multi-pass with different scanner light intensities.
Logged

Fine_Art
Sr. Member
****
Offline Offline

Posts: 1099


« Reply #32 on: January 12, 2013, 05:09:48 PM »
ReplyReply

Hi!

I used Vuescan. It does multipass, but there is a tendency to double contours.

By the way I happened to work with Ed Hamrick on Vuescan, the drivers for the DSMP are developed for my scanner ;-)

Best regards
Erik


Then I don't see why your dark areas are blacked out. What is the dmax on your scanner?
Logged
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7670


WWW
« Reply #33 on: January 12, 2013, 06:20:48 PM »
ReplyReply

I don't know. According to the manual it is 4.8 but I have seen some test where they put it much lower. I guess 3.5.

"Bcooter" recommended Iconlab. They say their Imacon has DR of at least 3.2.

"The Imacon scanner can handle standard film formats by the roll or individual frames; film does not need to be cut. This scanner has a dynamic range of at least 3.2, which ensures that negatives are scanned with almost as much detail as a drum scan. It is able to do an adequate job scanning chromes, but the extended range of a drum scanner will capture additional highlight and shadow detail. Color references should be provided when scanning negatives to achieve the best results."

The Imacon is a step up from the scanner I have.

Iconlab also offers drum scans: "Our drum scanning department is able to capture full tonal range and all the detail in any original. We can scan any size original up to 11x14 transparency and 16x20 reflective. The entire frame of any format, including rebate or borders, can be scanned. This is the only way to scan 6x8 or panoramic film. Virtually unlimited resolution results in file sizes up to 2GB from any original."

 
Best regards
Erik

Then I don't see why your dark areas are blacked out. What is the dmax on your scanner?
« Last Edit: January 12, 2013, 10:51:37 PM by ErikKaffehr » Logged

BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3768


« Reply #34 on: January 12, 2013, 06:33:14 PM »
ReplyReply

I used Vuescan. It does multipass, but there is a tendency to double contours.

Hi Erik,

You could try treating 2 separate scans as a kind of HDR image bracketing, and assemble an aligned HDR version from 2 scans. The same applies to your camera repro version.  Film tends to require some time acclimatizing in the scanner when they are not fluid mounted, otherwise the dimensions could change during (or between) the scan(s).

Cheers,
Bart
Logged
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7670


WWW
« Reply #35 on: January 12, 2013, 10:59:13 PM »
ReplyReply

Hi Bart,

I have tried that but it didn't work out well. I'm quite happy with the drum scan I have.

Best regards
Erik


Hi Erik,

You could try treating 2 separate scans as a kind of HDR image bracketing, and assemble an aligned HDR version from 2 scans. The same applies to your camera repro version.  Film tends to require some time acclimatizing in the scanner when they are not fluid mounted, otherwise the dimensions could change during (or between) the scan(s).

Cheers,
Bart
Logged

louoates
Sr. Member
****
Offline Offline

Posts: 780



WWW
« Reply #36 on: January 13, 2013, 08:41:35 PM »
ReplyReply

None of you know what you're talking about. Digital and film both suck. I'd go on to discuss the issue further but I've got to go coat my glass plates.
Logged
Ben Rubinstein
Sr. Member
****
Offline Offline

Posts: 1733


« Reply #37 on: January 14, 2013, 11:37:29 AM »
ReplyReply

None of you know what you're talking about. Digital and film both suck. I'd go on to discuss the issue further but I've got to go coat my glass plates.

Cheesy
Logged

Fine_Art
Sr. Member
****
Offline Offline

Posts: 1099


« Reply #38 on: January 14, 2013, 04:08:52 PM »
ReplyReply

Can you email a sample?
Logged
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7670


WWW
« Reply #39 on: January 14, 2013, 10:35:31 PM »
ReplyReply

Hi,

Whom are asking for a sample of what? Sorry for asking but I cannot see which post you have been responding to.

Best regards
Erik


Can you email a sample?
Logged

Pages: « 1 [2] 3 4 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad