Ad
Ad
Ad
Pages: [1] 2 3 »   Bottom of Page
Print
Author Topic: Digiatl vs. film another comparison  (Read 11595 times)
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7252


WWW
« on: July 30, 2009, 01:12:31 PM »
ReplyReply

Hi!

I visited a place where I have been taking pictures with my Pentax 67 a couple of years ago and reshot the same subjects with my new Sony Alpha 900. These pictures are a couple years apart but still should give some info on the capabilities of the two media. To make it short, here are my samples (67 on the left Alpha 900 on the right).

[attachment=15800:Default1.jpg]

These images are very small crops. The image from the Alpha 900 was upscaled using bicubic smoother to have the same dimensions as the Pentax 67. The Alpha 900 image may need some sharpening because of the upscaling. A sharpened version is below:

[attachment=15801:Sharpened.jpg]

My take:

In my view the two pictures are about on par. Both can be adjusted significantly. IMHO the 67 image still has an edge in sharpness and maybe color. Both can be adjusted quite a lot, however.

Sample files can be downloaded here: http://83.177.178.241/ekr/LLStuff/P67_VS_Alpha900/index.html

Technical details for 67:

Pentax 67
Fuji Provia 100
Scanned at 3200 PPI on Minolta Dimage Scan Multi Pro

Provia image has "Landscape sharpening" while Alppha 900 image has default sharpening.

Best regards
Erik
« Last Edit: July 30, 2009, 02:10:18 PM by ErikKaffehr » Logged

Guillermo Luijk
Sr. Member
****
Offline Offline

Posts: 1274



WWW
« Reply #1 on: July 30, 2009, 01:41:23 PM »
ReplyReply


I think the digital image clearly keeps more detail and texture. Grain on the film version can be interpreted as extra sharpness, but if you look closely you'll see it's just an illusion and the digital camera wins.

Regards.
Logged

Gemmtech
Sr. Member
****
Offline Offline

Posts: 526


« Reply #2 on: July 30, 2009, 01:47:42 PM »
ReplyReply

Quote from: GLuijk
I think the digital image clearly keeps more detail and texture. Grain on the film version can be interpreted as extra sharpness, but if you look closely you'll see it's just an illusion and the digital camera wins.

Regards.

I had to read the original post twice because for me the image on the right is clearly superior, is this the A900?  No contest.  I guess it's true, "It's all in the eye of the beholder"

Logged
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7252


WWW
« Reply #3 on: July 30, 2009, 01:49:04 PM »
ReplyReply

Thanks Guillermo, nice to have you watching over me!

Regarding the issue at hand I may or may not agree with you. My comparison is somewhat sloppy, but I wanted to share my findings ASAP.

The issue is partly that sharpening really matters and the images are different both in tonality, color and size. Your observation is much appreciated, and I'm probably going to revisit the issue in the coming couple of days. As you may have noticed I added a link where the images can be downloaded.

Best regards
Erik

Quote from: GLuijk
I think the digital image clearly keeps more detail and texture. Grain on the film version can be interpreted as extra sharpness, but if you look closely you'll see it's just an illusion and the digital camera wins.

Regards.
Logged

MarkL
Sr. Member
****
Offline Offline

Posts: 332


« Reply #4 on: July 31, 2009, 06:44:09 AM »
ReplyReply

It would be interesting to see if the 6x7 improves much by using an imacon and also a drum scanner.

I've moved over to stitching with my D700 for most landscapes from shooting 6x7 provia. I haven't done any direct comparisons with my own gear but I have to stitich a number of D700 frames togther to feel I'm getting close to my imacon scanned 6x7 film.

Logged
Anders_HK
Sr. Member
****
Offline Offline

Posts: 1001



WWW
« Reply #5 on: July 31, 2009, 09:40:55 AM »
ReplyReply

@ Erik,

Read the post I linked above of my comparison ZD vs. Velvia 50 in Mamiya 7. I am rather certain the ZD at low ISO is more sharp than the Sony. Thus, proper scanned on a drum your 6x7 slide is destined to beat the Sony in resolution, colors etc hands down. Care to make a bet?Huh It serious would be interested see that slide drum scanned, if original is sharp.

I must honest state that not even my 35mm slides from Minolta Dimage SE were lack of detail as that scan of 6x7.

Above per my honest experience. Thus I shoot more film in addition to digital; now also 4x5.

Regards
Anders
Logged
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7252


WWW
« Reply #6 on: July 31, 2009, 11:23:49 AM »
ReplyReply

Hej Anders,

Just a few observations:

1) There is of course the question of weather the 67 slide is critically sharp.
2) The 67 slide is taken with a Pentax 67 with a 90/2.8 lens and it is very well possible that it is less sharp than your Mamiya 7. The tests I have seen on the Mamya lenses were generally excellent while the Pentax is at best middle of the road.
3) The Pentax 67 has issues with vibration from both mirror and shutter, but I controlled this as well I could.
4) The image you see is a very small crop at actual pixels. The original image is about 7000x8500 pixels the crop is about 1000x1000 pixels. Grain is clearly visible, so the scanner has resolved that.
5) The image is shot on Provia not Velvia (Velvia has better grain in my experience)
6) Regarding colors IŽd suggest that you are slightly in error. Colors are created both in capture (and determined by the filters in the bayer array) and in postprocessing, where you can essentially achieve any color.
7) The engineers at Fuji developed a color rendition for Velvia which is much liked by many photographers, but far from all. Joseph Holmes does not like Velvia at all, for instance. I'm pretty sure that digital capture can be processed into Velvia like color.

I could unfortunately not find the link to your post, I sure would like to check.


These images were not intended as test images, but I happened to have shot the same subject with the two different systems. A side by side comparison shooting the same subject at the same time may have been less error prone.

A side note, I have noticed that some photographers don't like digital, I have respect for that. It seems that a majority of photographers prefer digital and I also have respect for that.

Best regards
Erik

Quote from: Anders_HK
@ Erik,

Read the post I linked above of my comparison ZD vs. Velvia 50 in Mamiya 7. I am rather certain the ZD at low ISO is more sharp than the Sony. Thus, proper scanned on a drum your 6x7 slide is destined to beat the Sony in resolution, colors etc hands down. Care to make a bet?Huh It serious would be interested see that slide drum scanned, if original is sharp.

I must honest state that not even my 35mm slides from Minolta Dimage SE were lack of detail as that scan of 6x7.

Above per my honest experience. Thus I shoot more film in addition to digital; now also 4x5.

Regards
Anders
« Last Edit: July 31, 2009, 11:57:10 AM by ErikKaffehr » Logged

Pete Ferling
Guest
« Reply #7 on: July 31, 2009, 12:19:07 PM »
ReplyReply

No doubt a decent drum scan could yield a different result.  However, Erik's test is here most likely a real case scenario for the majority of us who shoot film and do not have access to expensive drum scanners.
I figured this lesson out when shooting a favorite winter scene of mine, and my film Mamiya froze up on me.  So I dragged out my 40d and performed a 3-pan and stitch.  The 40" wide print was very revealing; just as sharp and detailed of the same scene done earlier with MF film.

However, there are times where I know that a 3pan stitch, including +/-2 stop bracket grouping with digital to preserve hightlights and open shadows, can be nailed with a single shot of MF 160VC.  Of course I'm using a crop sensor and it requires work.  Therefore, sharpness and pixel counts aside.  Do you have any examples in regards to lattitude or range?
Logged
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7252


WWW
« Reply #8 on: July 31, 2009, 01:43:00 PM »
ReplyReply

Hi,

I am not shooting print film so I cannot speak about that. Regarding slide film I'd say that it has a five stop dynamic range, like two stop + and 3 stop - . The Velvia has a very high density, about 4, meaning that black is 10000 times darker then clear film, that is about 13 stops, but it corresponds to a much smaller dynamic range on the subject side. I guess that both the Imacon and a real drum scanner can extract much more of that dynamic range than a normal CCD scanner. Drum scanners use photo multiplier tubes.

On the other hand I'm pretty sure that noise is going up in the dark areas whatever technology is used.

You may check this article: http://www.clarkvision.com/imagedetail/fil...l.summary1.html

It's interesting to note that, according to the above article, Velvia is significantly better than Provia that I used in my comparison. I have the same impression. That may contribute to Anders's preference for analogue Velvia, other factors may be better technique, especially in scanning, but also taste. I have seen some of Anders work and find it pretty impressive!

Best regards
Erik




Quote from: Pete Ferling
No doubt a decent drum scan could yield a different result.  However, Erik's test is here most likely a real case scenario for the majority of us who shoot film and do not have access to expensive drum scanners.
I figured this lesson out when shooting a favorite winter scene of mine, and my film Mamiya froze up on me.  So I dragged out my 40d and performed a 3-pan and stitch.  The 40" wide print was very revealing; just as sharp and detailed of the same scene done earlier with MF film.

However, there are times where I know that a 3pan stitch, including +/-2 stop bracket grouping with digital to preserve hightlights and open shadows, can be nailed with a single shot of MF 160VC.  Of course I'm using a crop sensor and it requires work.  Therefore, sharpness and pixel counts aside.  Do you have any examples in regards to lattitude or range?
Logged

Rob C
Sr. Member
****
Offline Offline

Posts: 12215


« Reply #9 on: July 31, 2009, 02:05:11 PM »
ReplyReply

[quote name='ErikKaffehr' date='Jul 31 2009, 04:23 PM' post='301221']
Hej Anders,

Just a few observations:

1) There is of course the question of weather the 67 slide is critically sharp.
2) The 67 slide is taken with a Pentax 67 with a 90/2.8 lens and it is very well possible that it is less sharp than your Mamiya 7. The tests I have seen on the Mamya lenses were generally excellent while the Pentax is at best middle of the road.
3) The Pentax 67 has issues with vibration from both mirror and shutter, but I controlled this as well I could.




Erik

I also had a Pentax 67 ll and after a few months decided to let it go. I went back to Nikon. The reasons? Mostly your third point (shutter), and I proved it to myself because with that camera, I only operated it on a huge Gitzo with mu in operation. But I also had issues with the 55mm wide and the fact that 135 film really was a better option for me came with the realisation that IŽd changed to 6x7 for the wrong reasons: imagined advantage in the stock world, just as digital was starting to impact! Timing has often let me down. :-(

Rob C
« Last Edit: July 31, 2009, 02:06:24 PM by Rob C » Logged

Anders_HK
Sr. Member
****
Offline Offline

Posts: 1001



WWW
« Reply #10 on: August 01, 2009, 08:53:59 AM »
ReplyReply

Quote from: ErikKaffehr
Hej Anders,

Just a few observations:

1) There is of course the question of weather the 67 slide is critically sharp.
2) The 67 slide is taken with a Pentax 67 with a 90/2.8 lens and it is very well possible that it is less sharp than your Mamiya 7. The tests I have seen on the Mamya lenses were generally excellent while the Pentax is at best middle of the road.
3) The Pentax 67 has issues with vibration from both mirror and shutter, but I controlled this as well I could.
4) The image you see is a very small crop at actual pixels. The original image is about 7000x8500 pixels the crop is about 1000x1000 pixels. Grain is clearly visible, so the scanner has resolved that.
5) The image is shot on Provia not Velvia (Velvia has better grain in my experience)
6) Regarding colors IŽd suggest that you are slightly in error. Colors are created both in capture (and determined by the filters in the bayer array) and in postprocessing, where you can essentially achieve any color.
7) The engineers at Fuji developed a color rendition for Velvia which is much liked by many photographers, but far from all. Joseph Holmes does not like Velvia at all, for instance. I'm pretty sure that digital capture can be processed into Velvia like color.

I could unfortunately not find the link to your post, I sure would like to check.


These images were not intended as test images, but I happened to have shot the same subject with the two different systems. A side by side comparison shooting the same subject at the same time may have been less error prone.

A side note, I have noticed that some photographers don't like digital, I have respect for that. It seems that a majority of photographers prefer digital and I also have respect for that.

Best regards
Erik

Hej Erik,

Here http://luminous-landscape.com/forum/index....showtopic=20970 or search "MFDB challenge".

"I'm pretty sure that digital capture can be processed into Velvia like color." - Noone yet has to my knowledge succeeded, since it is very difficult to replicate the rendering of Velvia 50 and for various shots in different conditions. And processing with many adjustments are time consuming. On contrary with Aptus 65 and Leica M8 colors are already at a very pleasing departure point already at default settings in e.g. CS3, and such departure point  (though not Velvia like) is something which is unlike any other digital cameras I have used and saves time.  It is true that not everyone likes Velvia, but for landscape photography it has a truly amazing and beloved rendering of colors in a much pleasing way and enhances just the right colors etc, and... low and behold it has black levels. To be frank, contrary to what many think in digital terms, it is pleasing to have a certain level of black levels in a photo, also s digital one!

"I have noticed that some photographers don't like digital, I have respect for that." - So do I and for vice versa, but I think the key thing to realize is that film and digital have a different rendering and it depends also what we shoot. I like both, and in different ways. Yet Velvia has a means of rendering that is truly magic for landscapes! - also to point of pleasingly absorbing and enhancing colors that you barely saw.

"It seems that a majority of photographers prefer digital and I also have respect for that." - Yes, on LL and a number of other forums because it is primarily centered about the latest gear, and frank many of us are sold on that, are we not? Following website have many very nice film images, check threads under "On Photography"; http://www.largeformatphotography.info/forum/

As I said, I use both film and digital but for capturing in different ways. Right now I am struck by how amazing simple it is with 4x5, and how accurate one can control the image by taking time and slow for each image, and spot metering accurate. Of course, the last can also be done with digital...  

Regards
Anders
Logged
Plekto
Sr. Member
****
Offline Offline

Posts: 551


« Reply #11 on: August 01, 2009, 01:35:10 PM »
ReplyReply

I still prefer grain over blurring.  The Sony is obviously suffering from a very heavy AA filter.  The film also has better colors, IMO.  But they are indeed very close.
Logged
Rob C
Sr. Member
****
Offline Offline

Posts: 12215


« Reply #12 on: August 01, 2009, 04:11:15 PM »
ReplyReply

Quote from: Plekto
I still prefer grain over blurring.  The Sony is obviously suffering from a very heavy AA filter.  The film also has better colors, IMO.  But they are indeed very close.



Have to agree about grain looking nicer than gloop. Colour depends also on lighting - canŽt make a real choice on that score as the light isnŽt the same. Snag is, digital is just so convenient...

Rob C
Logged

ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7252


WWW
« Reply #13 on: August 01, 2009, 05:44:40 PM »
ReplyReply

Hi!

Sony is upscaled to same resolution, sharpening matters a lot. I'll probably redo the comparison with another crop and Photokit sharpener applied to both.

Best regards
Erik

Quote from: Plekto
I still prefer grain over blurring.  The Sony is obviously suffering from a very heavy AA filter.  The film also has better colors, IMO.  But they are indeed very close.
Logged

ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7252


WWW
« Reply #14 on: August 01, 2009, 05:46:55 PM »
ReplyReply

Thanks, I recall the pictures!

I even made a raw conversion just to find out that my interpretation was pretty boring!

By the way, Velvia was my choice before going into digital, I liked it a lot.

Best regards
Erik

Quote from: Anders_HK
Hej Erik,

Here http://luminous-landscape.com/forum/index....showtopic=20970 or search "MFDB challenge".

"I'm pretty sure that digital capture can be processed into Velvia like color." - Noone yet has to my knowledge succeeded, since it is very difficult to replicate the rendering of Velvia 50 and for various shots in different conditions. And processing with many adjustments are time consuming. On contrary with Aptus 65 and Leica M8 colors are already at a very pleasing departure point already at default settings in e.g. CS3, and such departure point  (though not Velvia like) is something which is unlike any other digital cameras I have used and saves time.  It is true that not everyone likes Velvia, but for landscape photography it has a truly amazing and beloved rendering of colors in a much pleasing way and enhances just the right colors etc, and... low and behold it has black levels. To be frank, contrary to what many think in digital terms, it is pleasing to have a certain level of black levels in a photo, also s digital one!

"I have noticed that some photographers don't like digital, I have respect for that." - So do I and for vice versa, but I think the key thing to realize is that film and digital have a different rendering and it depends also what we shoot. I like both, and in different ways. Yet Velvia has a means of rendering that is truly magic for landscapes! - also to point of pleasingly absorbing and enhancing colors that you barely saw.

"It seems that a majority of photographers prefer digital and I also have respect for that." - Yes, on LL and a number of other forums because it is primarily centered about the latest gear, and frank many of us are sold on that, are we not? Following website have many very nice film images, check threads under "On Photography"; http://www.largeformatphotography.info/forum/

As I said, I use both film and digital but for capturing in different ways. Right now I am struck by how amazing simple it is with 4x5, and how accurate one can control the image by taking time and slow for each image, and spot metering accurate. Of course, the last can also be done with digital...  

Regards
Anders
« Last Edit: August 01, 2009, 05:51:02 PM by ErikKaffehr » Logged

Plekto
Sr. Member
****
Offline Offline

Posts: 551


« Reply #15 on: August 02, 2009, 01:08:19 PM »
ReplyReply

Quote from: ErikKaffehr
Hi!

Sony is upscaled to same resolution, sharpening matters a lot. I'll probably redo the comparison with another crop and Photokit sharpener applied to both.

Best regards
Erik

I'd almost rather see a 1:1 crop of each, because I've found that sharpening and tweaking in most programs does far more harm than good.  That said, I wish there was a way to turn it all off on the Sony, but as mentioned in previous threads, there is some software AA going on as well as a fairly heavy AA filter as well.  The photos out of the A900 look smooth and clean.  Very much so.  But it's because it's effectively down-sampling/glossing over resolution until you get about the same results as a 16MP or so DB.

Now, for a use and forget about it camera, that's great.  Less processing and tweaking - just drop in a raw converter and tweak the levels a bit - done and off to the client or emailed off to the editing department.  For speed and production work, this is probably a great thing.  But for artist types, where we're squeezing every pixel, it's a bit of a letdown.  

That said, I still am firmly of the camp that thinks that Bayer patterns are the major problem now.  It was a kludge to begin with that worked for many years, but now it's holding everything back.  But nobody is making an alternative that competes.  9-10MP equivalent isn't close, though.  I keep waiting but so far... yeah... always "next year"
Logged
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7252


WWW
« Reply #16 on: August 02, 2009, 11:21:24 PM »
ReplyReply

Hi,

I'll fix another crop, partly because there is something odd with the detail I choose. I can make 1:1 crops, without scaling. I don't think that there is software AA on the Sony, as raw processing was done with Lightroom. Regarding the issue Gabor Reported I don't think it is relevant in this context, it's more related to the handling of very dark pixels.

Regarding resampling the Sony image was not downsampled but upsampled. I would say that the comparison was tilted in favor of film, except capture.

Some sharpening is always needed. The ideas whit using Photokit Sharpener is that it has among other things well tought out sharpening for film. I'd certainly use it for film.

Best regards
Erik  

Quote from: Plekto
I'd almost rather see a 1:1 crop of each, because I've found that sharpening and tweaking in most programs does far more harm than good.  That said, I wish there was a way to turn it all off on the Sony, but as mentioned in previous threads, there is some software AA going on as well as a fairly heavy AA filter as well.  The photos out of the A900 look smooth and clean.  Very much so.  But it's because it's effectively down-sampling/glossing over resolution until you get about the same results as a 16MP or so DB.

Now, for a use and forget about it camera, that's great.  Less processing and tweaking - just drop in a raw converter and tweak the levels a bit - done and off to the client or emailed off to the editing department.  For speed and production work, this is probably a great thing.  But for artist types, where we're squeezing every pixel, it's a bit of a letdown.  

That said, I still am firmly of the camp that thinks that Bayer patterns are the major problem now.  It was a kludge to begin with that worked for many years, but now it's holding everything back.  But nobody is making an alternative that competes.  9-10MP equivalent isn't close, though.  I keep waiting but so far... yeah... always "next year"
Logged

barryfitzgerald
Sr. Member
****
Offline Offline

Posts: 566


« Reply #17 on: August 03, 2009, 07:25:04 AM »
ReplyReply

Interesting post.

However as is always the case, so many variations that make life hard, type of film used, processing, scanner, software for scanning.
At the end of the day 1:1 crops will never reveal why some of us like film, it's not something you can "test" just a love of the "look" you get.

Logged
Hank
Sr. Member
****
Offline Offline

Posts: 679


« Reply #18 on: August 03, 2009, 10:49:32 AM »
ReplyReply

There's a huge difference for us in the photo business, even with comparable results.  In the days of yore we spent between $25,000 and $30,000 a year on film and processing.  Last year we bought and processed a little less than 100 sheets of 4x5 film and no 120, 220 or 35mm film.  The $27,500 savings (by actual accounting for 2008) went into our pockets.  We reserve film use to 4x5 only, which we use on jobs that require the movements of the camera rather than the film itself.  Give us an affordable digital camera with movements and our film use will be exactly 0.
« Last Edit: August 03, 2009, 11:34:52 AM by Hank » Logged
Rob C
Sr. Member
****
Offline Offline

Posts: 12215


« Reply #19 on: August 03, 2009, 12:10:04 PM »
ReplyReply

Quote from: Hank
There's a huge difference for us in the photo business, even with comparable results.  In the days of yore we spent between $25,000 and $30,000 a year on film and processing.  Last year we bought and processed a little less than 100 sheets of 4x5 film and no 120, 220 or 35mm film.  The $27,500 savings (by actual accounting for 2008) went into our pockets.  We reserve film use to 4x5 only, which we use on jobs that require the movements of the camera rather than the film itself.  Give us an affordable digital camera with movements and our film use will be exactly 0.




Hank, does that 27,500 dollar saving include the time/expense of doing the digital "processing" instead, or have you not put a price against that in the calculation?

Rob C
Logged

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

Ad
Ad
Ad