[PPPPPP42]

Skipped the 4 pages of technobicker so this may have been said somewhere along the way:

I find it hard to tell between a digital APSC and a digital FF (or even a damn camera phone) when viewed resized on a computer screen.

I love looking at original resolution FF images and seeing all the wonderful details on super wide images. With APSC you seem to either get wonderful details OR super wide images.

I personally like looking at medium format best. Take a picture of the whole damn town and still read peoples license plates. For some reason you can take a picture of your lunch with a medium format and the spectacular level of detail just makes it look great.

That is what makes any format "better" to me than the format a size below it.

[Nicolas06]

I love looking at original resolution FF images and seeing all the wonderful details on super wide images. With APSC you seem to either get wonderful details OR super wide images.

Try it with a good modern prime like Samyang 16mm and you'll see wide ain't that bad !

Samyang 16mm on Canon 600D: www.magezinepublishing.com/equipment/images/equipment/16mm-f20-ED-AS-UMC-CS-5217/highres/Samyang16mmIMGS36343_1373266635.jpg
Samyang 12mm on Fuji XT1: http://pliki.optyczne.pl/samy12/samy12_f5.6_6.JPG

[hoopsontoast]

A try to get a rendering more like your film in DxO:
- Using Color profile Kodak Ektachrome 100 VS Generique
- Tone curve +0.05 on reds.
- exposition +.005
- highlight +10
- shadows -15
- Contrast -49
- Vibrance: -9
- Saturation: +45.

The blues are not really there, but overall it look much more like the film one. Might have been easier to work from raw file.

Thanks, As I say it was more of a comparison of "there was not a 20mm equivalent on APS-C" than trying to compare digital to film.
Good to see I could get close if I played around with the files a lot more. I try not to do 'too' much PP work on the digital files.

---------- Post added 08-20-14 at 08:08 AM ----------

Try it with a good modern prime like Samyang 16mm and you'll see wide ain't that bad !

Samyang 16mm on Canon 600D: www.magezinepublishing.com/equipment/images/equipment/16mm-f20-ED-AS-UMC-CS-5217/highres/Samyang16mmIMGS36343_1373266635.jpg
Samyang 12mm on Fuji XT1: http://pliki.optyczne.pl/samy12/samy12_f5.6_6.JPG

And that was not my quote, not sure what happened there.....

[Nicolas06]

Thanks, As I say it was more of a comparison of "there was not a 20mm equivalent on APS-C" than trying to compare digital to film.
Good to see I could get close if I played around with the files a lot more. I try not to do 'too' much PP work on the digital files.

[off topic]

Yeah sure, I know it wasn't your main point here.

I never really mastered film. I just did a few shoots when I was teenager as my father given me an old Pentax and manual lenses. For me at the time, it was just too much ceremony to get the job done: paying for film, manual focussing, paying again to be able to finilly see some picture out of it in small format (like 10cmx15cm or 13cmx18cm). Anything bigger would have been too expensive as teenager, taking lot of photo or even train too.

I lost interrested and was offered years later a compact. This was a far better experience in usability and ergonomics from what I remembered.

It only recently that I grabbed interrest in a reflex, and for me digital + AF made the deal. Instant review of image, easy focussing, hability to reframe/crop, change the colors.

I wanted to do this exercice as I didn't want to think: digital colors just suck ! To get nice colors shoot film.

I brought DxO Film pack some time ago. At begining it was looking like total garbage and stupid idea. Why one would want to pay to get rendering like old film (something like 100 different rendering, half black & white) ?

I discovered what I didn't know, an additionnal ceremony of film era: you had to choose the film to get the type of rendering you liked... and even more the rendering suited to photo you intended to take. Would you want soft colors for portraiture, vibrant color for a landscape ? Neutral colors ?

And so with this tool I can now fast choose between different color rendering. Portraits look better without spending much time. I can choose between many black & white styles, if I want a contrasted rendering or not, lot of color or not... All in a comprehensive package. This is really a time saver.

In pratice, you just choose a rendering that look good ! Maybe then review exposure/constrast with saturation set to 0 to review the exposure of different part of the image, and then restaure back saturation slowly and stop where it give the best. Just doing that can do more to the pop and attractiveness of your photo than buying the most expensive lenses !

Getting exactly your scanned film might no be that usefull in practice... It doesn't have to be the best possible rendering, but that was the exercice I tried.

[/off topic]

[Lowell Goudge]

Skipped the 4 pages of technobicker so this may have been said somewhere along the way:

I find it hard to tell between a digital APSC and a digital FF (or even a damn camera phone) when viewed resized on a computer screen.

I love looking at original resolution FF images and seeing all the wonderful details on super wide images. With APSC you seem to either get wonderful details OR super wide images.

I personally like looking at medium format best. Take a picture of the whole damn town and still read peoples license plates. For some reason you can take a picture of your lunch with a medium format and the spectacular level of detail just makes it look great.

That is what makes any format "better" to me than the format a size below it.

The problem is that what you want is the effect that MF film shots have, where the base resolution of the recording medium remained the same but the size of the medium increased. Sadly this is not the case. FF cameras have less resolution than the APS C counterparts. Take a K3. At 24 MP, you would need a 54 MP sensor, but we are not being offered that in FF.

---------- Post added 08-20-2014 at 07:21 AM ----------

In Bold above, I have exactly this (taken minutes apart)

APS-C - (K-01) DA 14mm f2.8


FF - (MZ-5N + Fuji 100F) FA 20-35mm f4 (@20mm)


Now you cant compare the two exactly, the K-01 Image has been processed from RAW with Lightroom (highlights, shadow recovery), the MZ-5N straight from the Commercial Scan and little added contrast. But the point is there are equivelent, even the FF lens is much lighter but just as sharp, if possibly sharper at the expense of the all metal build.
The main difference in that was the colour from the film was so much more life like, even playing around with the digital file extensively in white balance/settings I could not match it without bumping up saturation.

The main reason I would love a Pentax FF is just that I can carry one set of lenses to use with 135 Film and Digital.
Also the potential for other advantages mentioned before like lower noise, better high ISO performance, the lens being used as designed etc.

I think one of the issues between the two shots is that the exposure and WB are different between the two. When doing sunsets, I find that the best exposure is to spot meter off the brightest part of the blue sky, shooting with Daylight setting for WB. This pushes the overall exposure down and gives you the deep reds and yellows in the cloud detail.

The other thing I see, and this also contributes to the difference between film and digital, is your scanned shot seems to show , overall less exposure lattitude. This is the result of shadow and highlight protection, durn off the shadow protection and it will darken the portions of the shot below 100 greyscale by about a stop, getting closer to the film shot.

[Chris Mak]

The main difference in that was the colour from the film was so much more life like, even playing around with the digital file extensively in white balance/settings I could not match it without bumping up saturation.

The main reason I would love a Pentax FF is just that I can carry one set of lenses to use with 135 Film and Digital.
Also the potential for other advantages mentioned before like lower noise, better high ISO performance, the lens being used as designed etc.

I shoot the Pentax K3 (and before that the K5/K5IIs), and since a month or so also the Sony A7r, both with the same (Zeiss ZK) lenses. One of the first things that struck me, is how much color information there is in the Sony Raw files. On a non-technical level this translates (for me that is) into much more involving and life-like images, that also hold up superbly in Raw editing and other PP. I have gone through lots of hoops to get color out of Apsc digital that involved me like Film photography, DxO film pack, tonal and saturation masking in photoshop, colorspaces and what not. But it was only with the Sony A7r that the color that I would like is simply there. My experience with the (seemingly) bloated files (resolution wise) of the A7r, is that apart from benefits of cropping and printing big, the real advantage lies in how the enormous amount of tonal and color detail translates into images that are very pleasant to look at, and also allow for extensive editing without losing their overall appearance.
I still like Apsc as a wildlife and practical system, and the K3 is one fine camera, but even with the same lenses, the Sony A7r gets to where I truly enjoy the images.

Chris

[rawr]

how much color information there is in the Sony Raw files

DxoMark bears this out - A7R 25.6 bits vs 23.7 bits of colour depth for the K-3. That seems a solid margin.

But colour depth seems to depend as much on sensor quality and type of technology as sensor format size per se. At the moment when you sort all the colour depth results for all the DxOMark measured cameras, the top results are all certainly FF and MF. But the newest generation of APS-C sensors, led by the new Sony Alpha 77 II and a few others like the Nikon D3300 and D7100, perform the same for colour depth as FF Nikon D4s, Df, and even the MF 645D. So while most FF's may perform better than APS-C on this metric, it's doesn't seem a straightforward FF vs crop issue.

Furthermore, there aren't any Canon's in the top 25 camera rankings on colour depth on DxOMark, including ANY of their full-frames. Meanwhile, the K-5IIs and a bunch of other crop bodies have greater colour depth than the Canon 1Dx, Canon's full-frame flagship, according to DxO. So the sensor leadership of Sony may be playing more of a role in some metrics than mere FF vs crop sensor size.

[Nicolas06]

I shoot the Pentax K3 (and before that the K5/K5IIs), and since a month or so also the Sony A7r, both with the same (Zeiss ZK) lenses. One of the first things that struck me, is how much color information there is in the Sony Raw files. On a non-technical level this translates (for me that is) into much more involving and life-like images, that also hold up superbly in Raw editing and other PP. I have gone through lots of hoops to get color out of Apsc digital that involved me like Film photography, DxO film pack, tonal and saturation masking in photoshop, colorspaces and what not. But it was only with the Sony A7r that the color that I would like is simply there. My experience with the (seemingly) bloated files (resolution wise) of the A7r, is that apart from benefits of cropping and printing big, the real advantage lies in how the enormous amount of tonal and color detail translates into images that are very pleasant to look at, and also allow for extensive editing without losing their overall appearance.
I still like Apsc as a wildlife and practical system, and the K3 is one fine camera, but even with the same lenses, the Sony A7r gets to where I truly enjoy the images.

Chris

Could you please then when you have time get your K3 and your A7r, shoot the same subject at roughly same time with same field of view and so on (like 50 f/5.6 and 75mm f/8), and maybe share the A7r jpeg taken from raw without PP and K3 jpeg taken without PP so we can see this difference? Admitting it is visible on a 1024x768 on screen... My understanding color difference would show...

Also what are the lenses you have for your K3 and the A7r? All A7r lenses are very highend Zeiss lenses. One should compare it with an FAltd or at least a DA*. Was it the case?

[LFLee]

OK, many of you said format doesn't matter - then go buy a Q and trash your K.

Personally, I observed and conclude the following general tread:
If you sucks at photography, format doesn't matter: gave you a FF camera your still take cat photos.
If you good at photography, format matters: give you a FF camera your photos dramatically become even better!

of the following also true:
If you sucks at photography, format matter: upgrading from phone camera to DSLR immediately improve your photos.
If you good at photography, format doesn't matter: still take stunning photos using phone camera.

Buy what you can afford and enjoy shooting!

[mecrox]

I shoot the Pentax K3 (and before that the K5/K5IIs), and since a month or so also the Sony A7r, both with the same (Zeiss ZK) lenses. One of the first things that struck me, is how much color information there is in the Sony Raw files. On a non-technical level this translates (for me that is) into much more involving and life-like images, that also hold up superbly in Raw editing and other PP. I have gone through lots of hoops to get color out of Apsc digital that involved me like Film photography, DxO film pack, tonal and saturation masking in photoshop, colorspaces and what not. But it was only with the Sony A7r that the color that I would like is simply there. My experience with the (seemingly) bloated files (resolution wise) of the A7r, is that apart from benefits of cropping and printing big, the real advantage lies in how the enormous amount of tonal and color detail translates into images that are very pleasant to look at, and also allow for extensive editing without losing their overall appearance.
I still like Apsc as a wildlife and practical system, and the K3 is one fine camera, but even with the same lenses, the Sony A7r gets to where I truly enjoy the images.

Chris

Yes, this is indeed a good point. Do FF files require less PP, on average, than ones from APS-C for a given level of quality. I know that is entirely subjective, but folks who use both systems must know how much time they typically spend in PP for each. With my K5, I am certainly tiring of the amount of PP involved in getting what I feel to be an acceptable image. Lightroom, Photoshop, Topaz Labs, back to Photoshop again, etc. Maybe my standards are getting higher. Who knows. Perhaps it's the generation of sensor used or perhaps it's the good glass argument again. A very good lens will produce colour, contrast and sharpness which won't need so much attention in PP. But there again, maybe a good lens does more of its goodness on FF than on APS-C. I'll admit that whenever I see an image on Flickr which just goes Kaboom! it usually turns out to have been taken on FF with a top-quality lens. There have been some killer ones recently on a Sony A7 using an smc-A 50mm f1.4, courtesy of a forum member. Probably the real reason is that he is a top-quality photographer. Oh well, round in circles as always. Pretty well the only way is to try it and see, I guess.

[BrianR]

OK, many of you said format doesn't matter - then go buy a Q and trash your K.

There's an overriding context here that you can't ignore- the smaller your final image size the less you'll notice the difference in formats. Cell phone vs MF for a highly compressed 80x80 pixel avatar meh take you pick... print the same images at 3 or 4 feet wide and it will be much more noticeable...

But otherwise...

Buy what you can afford and enjoy shooting!

...yep.

[stevebrot]

Try it with a good modern prime like Samyang 16mm and you'll see wide ain't that bad !

I don't have experience with the Samyang, but in the last day or so I have judged four photos for the PEG taken with the DA 15/4 Limited. The DA 15 is a decent lens (I have shot with it), but three of the four were sadly soft at other than the center to the detriment of the photo. This may have been a reflection of the technique used or maybe it just proves the point that it is very difficult to produce a good rectilinear lens below 24mm focal length, regardless of format.


Steve

---------- Post added 08-20-14 at 08:54 AM ----------

There's an overriding context here that you can't ignore- the smaller your final image size the less you'll notice the difference in formats.

Absolutely! Snapshot film cameras were designed around the knowledge that nobody was going to print larger than 4x5.


Steve

---------- Post added 08-20-14 at 09:04 AM ----------

FF cameras have less resolution than the APS C counterparts. Take a K3. At 24 MP, you would need a 54 MP sensor

Yep...All the more reason to buy a 645Z

That response is a little tongue-in-cheek since in the real answer is that many current FF cameras meet or exceed the resolution of the K-3. In the digital world 24 Megapickles APS-C resolution is the same as 24 Megapickles FF (36x24) resolution. There are nominally the same number of pixels on each axis. The K-3 image is composed of 6016x4000 pixels while the Sony A7 (FF) image is composed of 6000x4000 pixels.*


Steve

* The pixel dimensions for my scanned 35mm negatives are somewhat less than that

[Lowell Goudge]

I don't have experience with the Samyang, but in the last day or so I have judged four photos for the PEG taken with the DA 15/4 Limited. The DA 15 is a decent lens (I have shot with it), but three of the four were sadly soft at other than the center to the detriment of the photo. This may have been a reflection of the technique used or maybe it just proves the point that it is very difficult to produce a good rectilinear lens below 24mm focal length, regardless of format.


Steve

---------- Post added 08-20-14 at 08:54 AM ----------

or maybe it is more that it is difficult to produce a good retro-focus lens. Note 24 mm is shorter than the regestry distance by half, and therefore the retro focus elements add their one problems

Yep...All the more reason to buy a 645Z

That response is a little tongue-in-cheek since in the real answer is that many current FF cameras meet or exceed the resolution of the K-3. In the digital world 24 Megapickles APS-C resolution is the same as 24 Megapickles FF (36x24) resolution. There are nominally the same number of pixels on each axis. The K-3 image is composed of 6016x4000 pixels while the Sony A7 (FF) image is composed of 6000x4000 pixels.*


Steve

* The pixel dimensions for my scanned 35mm negatives are somewhat less than that

This could be a great debate. What people really want is the same FOV of a FF camera, but the pixel pitch, hence increased resolution of APS C.

Consider the extreme, my original Q and it's pixel density, would require roughly 170 MP in APS C and 300 MP in FF for the same absolute resolution. That is really the basis of the comment about MF vs FF

[Rondec]

OK, many of you said format doesn't matter - then go buy a Q and trash your K.

Personally, I observed and conclude the following general tread:
If you sucks at photography, format doesn't matter: gave you a FF camera your still take cat photos.
If you good at photography, format matters: give you a FF camera your photos dramatically become even better!

of the following also true:
If you sucks at photography, format matter: upgrading from phone camera to DSLR immediately improve your photos.
If you good at photography, format doesn't matter: still take stunning photos using phone camera.

Buy what you can afford and enjoy shooting!

I think clearly the more difference there is in sensor size, the more likely you will see a difference in final output. The difference between a V1 or Q and a full frame camera's output is much more likely to be seen in real world shooting than the difference between a full frame and APS-C or medium format and full frame.

I don't have experience with the Samyang, but in the last day or so I have judged four photos for the PEG taken with the DA 15/4 Limited. The DA 15 is a decent lens (I have shot with it), but three of the four were sadly soft at other than the center to the detriment of the photo. This may have been a reflection of the technique used or maybe it just proves the point that it is very difficult to produce a good rectilinear lens below 24mm focal length, regardless of format.


Steve[COLOR="Silver"]

I like the DA 15, but it needs to be shot at f8 and manually focused to infinity (or close to it) to see good results. Anything else and field curvature and border softness are problematic.

[Chris Mak]

Yes, this is indeed a good point. Do FF files require less PP, on average, than ones from APS-C for a given level of quality. I know that is entirely subjective, but folks who use both systems must know how much time they typically spend in PP for each. With my K5, I am certainly tiring of the amount of PP involved in getting what I feel to be an acceptable image. Lightroom, Photoshop, Topaz Labs, back to Photoshop again, etc. Maybe my standards are getting higher. Who knows. Perhaps it's the generation of sensor used or perhaps it's the good glass argument again. A very good lens will produce colour, contrast and sharpness which won't need so much attention in PP. But there again, maybe a good lens does more of its goodness on FF than on APS-C. I'll admit that whenever I see an image on Flickr which just goes Kaboom! it usually turns out to have been taken on FF with a top-quality lens. There have been some killer ones recently on a Sony A7 using an smc-A 50mm f1.4, courtesy of a forum member. Probably the real reason is that he is a top-quality photographer. Oh well, round in circles as always. Pretty well the only way is to try it and see, I guess.

To my experience so far with the Sony A7r, comparing to the (Pentax) Apsc family of the last three years, yes I do spend less time in the raw converter and much less in Photoshop trying to get it right. To begin with (and yes: this is with a quartet of quality Zeiss ZK lenses that I bought three years ago and used on Apsc all this time), the A7r files capture a large dynamic range. I know that per DxO the Sony A7r has the same 14.1 score on dynamic range that the Pentax K5 had, and of course the K5 could bring shadows back seemingly endlessly. Still there is a difference, the images seem to have a wider scope at default, and more often only need brightening up with the exposure tool (I use CO1), whereas the Apsc images need more fiddling around with shadow raising or highlight reconstruction and such, which tend to need all sorts corrections e.g. color wise. Although I could capture quite a large DR with the K5 at iso 80, there was always some sort of sense of compression at default, and colors not really evenly saturated, just needing a lot of work to get a pleasant and natural looking image.
I'm not saying that you can as well shoot the A7r in jpeg, but with a good raw converter it is relatively easy and definitely less time consuming to get balanced, pleasing images with all the tonal details and color nuances. Of course that adds to the satisfaction of shooting with the camera.

Chris

EDIT:
Here is an example with 100% crop and screenshot from the raw converter.
From the screenshot you can see that the whole histogram at default setting, other than a 0,8 ev exposure lifting, easily falls within the 0-255 scale, I can even move the black point as well as white point inwards a bit. This may seem no big deal, with the shot seemingly not containing a large DR, but I have shot many similar lighting situations with all three Pentax crop cameras, and in these circumstances the skies were always blown, or the black point could not be brought back into the 0-255 scale anymore, at least not with the normal exposure tool, needing all sorts of substantial shadow lifting. The sun is lighting up the clouds, and background skies like that are harder than the Sony A7r makes them look, especially with the Zeiss makro planar 50/2, with which this image was taken, a lens that easily blows skies. The great thing is that you can significantly underexpose, and totally bring the image back with a natural look with one slide of the exposure tool. The Sony A7r has this tendency to underexpose around 0,7ev by default, hence the +0,8 ev on this image. You can see from the 100% crop how much real detail is retained. The way in which this makes the image pleasing to me, is that shadows and light tones are all there in a natural way, a bit like my eyes would see them in the real situation.







---------- Post added 08-20-14 at 10:12 PM ----------

Could you please then when you have time get your K3 and your A7r, shoot the same subject at roughly same time with same field of view and so on (like 50 f/5.6 and 75mm f/8), and maybe share the A7r jpeg taken from raw without PP and K3 jpeg taken without PP so we can see this difference? Admitting it is visible on a 1024x768 on screen... My understanding color difference would show...

Also what are the lenses you have for your K3 and the A7r? All A7r lenses are very highend Zeiss lenses. One should compare it with an FAltd or at least a DA*. Was it the case?

Nicolas, see my response to mecrox.
When I have the time, I will do a comparison. I use the Zeiss 28/2, 35/2, 50/2 and 85/1.4. These are the MF Zeiss lenses, on a voigtländer adapter.
These may seem exotic (price wise), but that's not really so: e.g. the 35/2 cost me 799,- euros new at a respectable camera store, the HD DA35/2.8 ltd goes for 729,- euros

Chris