Comparing the M8 to The M7 (Mamiya 7II that is!)

[gdi]

A very timid subject captured with the M8 in low light at f/4. On the Mamiya 65mm lens, very little DOF would have been conveyed at f/4, and in fact this is not the "sweet spot" of the lens. On the M8 with a ZM 35/2 Biogon however, f/4 gives plenty of depth and the lens is uber crisp at this aperture. So with the M8 I think "flexibility", with the M7 I think "extreme quality" in a compact package albeit suited to a narrower range of conditions required for optimal usage. For 4x5, what a PITA, so this is reserved for subjects requiring the utmost detail with the intent of eventual enlargement to 20 x 24 (minimum) and beyond. Again, this applies to the landscape genre.

That is a nice shot! 🙂
But it does look a little weird on the edges, like sever jpg compression or something.

 

[POINT OF VIEW]

Nice, but how does that blown out area on the edge of the game table look in the final print?

I think the photo of the photo that you see caused the hard contrast and excessive light. I just laid the original on the floor with a tape measure. It was shot with a 5 mg. Casio camera. Bottom line the, enlargement picture looks good. I should add, the original was shot hand held in low light, the lens was a Noctilux and the M8 was set on 1250 ISO, fine JPEG. Bill

 

[jplomley]

Thanks gdi. I realized afterwards that I exceeded the file size for a JPEG, so there was some funky down-rezzing going on that furnished the artifacts. Arghhhh! You can check out the image and many others I have posted on the LUF in the People folder. Much higher file size allowance on that forum. Same user name over there as here. Looking forward to your conclusions.

Cheers,
Jeff

 

[gdi]

Thanks gdi. I realized afterwards that I exceeded the file size for a JPEG, so there was some funky down-rezzing going on that furnished the artifacts. Arghhhh! You can check out the image and many others I have posted on the LUF in the People folder. Much higher file size allowance on that forum. Same user name over there as here. Looking forward to your conclusions.

Cheers,
Jeff

I just did jeff - that is a great shot and it looks much better on LUF!

 

[gdi]

Ok - my final conclusions on my little experiment...

I did get a chance to print tonight - at the full target size equivalent to 24x30" at 360 on an Epson 7600. The results are a little disappointing, but not really surprising I suppose.

Undeniably, the scanned 6x7 print is better than the M8 prints. I did a print of the M8 simply uprezzed, then one with some added noise, and finally with a little sharpening.

I think the best M8 is probably the one with just a bit of noise added. The sharpening really results in some slight artifacts that are not effectively masked by the added noise. But the noise does mask the uprezzing artifacts resulting in a noticeable improvement without really being obtrusive.

But, the 7II shot is better in all ways, really. The added detail resolution helps a lot, but a bigger advantage is the tonality and the smooth transitions. The cans in the test shot have some very subtle reflections and metallic sheen. This is very clear and provides a natural and dimensional aspect to the 7II print. This was mentioned earlier by MFogiel and he is right, it is the most impressive difference in my opinion, though the added sharpness and detail is welcome. Also, the noise in this print is practically imperceptible.

The M8 prints loose out on the tonality (maybe the 8bit files contribute to this?) and the prints just don't have the dimensional look to them. The reflections are just not there, or look more like streaks. The fine detail looks fuzzy. There is shading on one can that looks posterized, even with the noise added. While the detail, as pointed out earlier, may not be critical in portrait situations the breakdown of tonality may indeed prove to be a problem.

BUT that said, this is a side by side comparison (by a pretty demanding tester!) and I think there would be plenty of people very happy with the M8 prints prepared in and printed in this way - they do look good and represent a very impressive performance! I can see why people are pleased with their large M8 prints. I would try to scan the prints, but I don't think my little Epson flatbed will show much difference.

As for me, I think my 6x7 kit will remain for a while, but it will get much, much less use than the M8. Thanks to everyone who indulged me in publicly sharing this test!

 

[rxmd]

This is a actual picture of a picture. [...] It had the same sharp appearance as the one shown.

It's easy for a 500x400 pixel downscaled compressed JPEG to appear good. On the contrary, it's difficult for them to appear bad! 🙂

Guess what kind of camera this was taken with (digital/film, film/sensor size).

As I said, I believe you when you say that print quality on your M8 prints is good. All I want to say is that this kind of quality is impossible to illustrate with downscaled Internet pictures, which can be used to illustrate almost anything. (That's why all the usual discussions on microcontrast and tonality etc. on the basis of Flickr images are somewhat pointless).

Philipp

 

[POINT OF VIEW]

Very true

Very true

Very true, but when you think about it, comparing the medium film camera to a M8 digital camera makes the hole thing pointless. I guess you could say I am comparing absurdity with absurdity. Bill

It's easy for a 500x400 pixel downscaled compressed JPEG to appear good. On the contrary, it's difficult for them to appear bad! 🙂

Guess what kind of camera this was taken with (digital/film, film/sensor size).

As I said, I believe you when you say that print quality on your M8 prints is good. All I want to say is that this kind of quality is impossible to illustrate with downscaled Internet pictures, which can be used to illustrate almost anything. (That's why all the usual discussions on microcontrast and tonality etc. on the basis of Flickr images are somewhat pointless).

Philipp

 

[gdi]

Very true, but when you think about it, comparing the medium film camera to a M8 digital camera makes the hole thing pointless. I guess you could say I am comparing absurdity with absurdity. Bill

It not absurd or pointless at all, in fact it is the only common sense to compare. If you look you will find numerous posts stating that the m8 produces prints equal to MF and even 4x5.

If you were like me and owned both of these quite expensive systems and felt there was a reasonable chance that one system would satisfy all requirements it would be irrational not to compare.

You may think it absurd to hope the M8 could be up to the task, but the final results show that it performs a good deal beyond what most would expect. No 10MP camera I have used could do as well and it definitely does as well as the 13mp 5D I compared head to head with the 7ii.

I am glad I finally did the test and again I thank you and the others who contributed with comments and photos!

BTW- here is a photo i took at a fair game booth with the M8, and a fast lens. (But its no award winner!)

 

[2XLX2]

nice shot, gdi

what aperture, shutter-speed and iso? what lens?

 

[2XLX2]

another newbie question: how does the M8's dynamic range compare with color negative and transparency [35mm or larger] ?

 

[gdi]

That shot is F1.2 @1000th, Hexanon 50/1.2.

I haven't done direct comparisons with slides or color print film, but I haven't found it to be an issue with most scenes - IF I properly expose (no overexposure). I think the M8 should be close to print film and probably a little better than E6?

 

[Richard Marks]

It not absurd or pointless at all, in fact it is the only common sense to compare. If you look you will find numerous posts stating that the m8 produces prints equal to MF and even 4x5.

If you were like me and owned both of these quite expensive systems and felt there was a reasonable chance that one system would satisfy all requirements it would be irrational not to compare.

You may think it absurd to hope the M8 could be up to the task, but the final results show that it performs a good deal beyond what most would expect. No 10MP camera I have used could do as well and it definitely does as well as the 13mp 5D I compared head to head with the 7ii.

I am glad I finally did the test and again I thank you and the others who contributed with comments and photos!

BTW- here is a photo i took at a fair game booth with the M8, and a fast lens. (But its no award winner!)

It is a very relevent comparison. The M8 certainly has resolution beyond 35mm film so the obvious question is can it replace one's medium format gear. Nearly but not quite is the answer if we are talking pure resolution terms. However when you bring hand held performance, low light levels, ease of use, playback and post processing editing the argument is more difficult. The gear I used to use for portraits was a Hassy 503 with 120 Makro Planar. I have been more productive with mt M8 and 75 lux Similar focal length. I never really hit it off with Mamiya 7 for portraits as the 150 lens did not focus close enough to use the full frame of the 6x7 format.

Great thread

Richard

 

[TJV]

GDI

I'm wondering a couple of things about your scanning AND M8 workflow that may / may not make a difference.

In my experience, and I haven't experimented with the Nikon 9000ED scanner I'm scanning my M7ii slides yet, is that with any scan there needs to be a small amount of selective channel noise reduction (and then subsequent REsharpening) to get rid of colour noise introduced in the scanning stage. I guess it's similar to de-noising a digi M8 file. With the M8 I usually left noise reduction to be done solely in Photoshop, employing Bruce Fraser's techiques outlined in his many brilliant Photoshop books. The same exact process I use for my film scans, although I usually find I need to apply slightly more because grain aisling complicates the appearance of regimented scanner noise. The M8, believe it or not, I found required very little noise reduction, and if done properly in conjunction with intelligent two stage sharpening (both source AND print optimized,) produced some absolutely stunning 12x18" prints at 1250ISO. 2500ISO was, for me, as waste of time unless I was printing small. Film scans in my experience have a unique tonal quality that I couldn't replicate (my own lack of skill level here,) with digital files in print.

Anyway, maybe you can explain more about your two types of workflow in regards to how you process your film scans after the scanning stage to get the most out of them compared to the M8? Maybe the processes can even up the playing field? Maybe these techniques make the gap bigger if done properly?

 

[gdi]

GDI

I'm wondering a couple of things about your scanning AND M8 workflow that may / may not make a difference.

In my experience, and I haven't experimented with the Nikon 9000ED scanner I'm scanning my M7ii slides yet, is that with any scan there needs to be a small amount of selective channel noise reduction (and then subsequent REsharpening) to get rid of colour noise introduced in the scanning stage. I guess it's similar to de-noising a digi M8 file. With the M8 I usually left noise reduction to be done solely in Photoshop, employing Bruce Fraser's techiques outlined in his many brilliant Photoshop books. The same exact process I use for my film scans, although I usually find I need to apply slightly more because grain aisling complicates the appearance of regimented scanner noise. The M8, believe it or not, I found required very little noise reduction, and if done properly in conjunction with intelligent two stage sharpening (both source AND print optimized,) produced some absolutely stunning 12x18" prints at 1250ISO. 2500ISO was, for me, as waste of time unless I was printing small. Film scans in my experience have a unique tonal quality that I couldn't replicate (my own lack of skill level here,) with digital files in print.

Anyway, maybe you can explain more about your two types of workflow in regards to how you process your film scans after the scanning stage to get the most out of them compared to the M8? Maybe the processes can even up the playing field? Maybe these techniques make the gap bigger if done properly?

TJV,

I normally don't have to worry about noise on my 6x7 scans at the max size I can print - especially if I don't apply sharpening to the entire file. Using a selective/edge sharpening, the grain is not usually noticeable in even large prints. Sometime it takes some tweaking to get that sharpening right

Now 35mm is a different story and, while I don't normally print over 8x12 with them, I may try your noise reduction/resharpen technique. I actually have a B&W 35mm shot that I will be printing tomorrow for a local business owner, and I may try some NeatImage on it.

As for M8 files, I haven't used any noise reduction on them - for high ISO shots I do no sharpening and no noise reduction unless it is a 2500 or maybe 1250 shot and then only color NR. My hope is that if I don't use noise reduction in my workflow, I won't have to use sharpening - and vice versa. This has been successful for me for printing M8 files that didn't require significant uprezzing.

I have been pretty conservative to this point with my max print sizes, so I may need to apply NR eventually. Of course the comparison I did here was at 160 so noise reduction wouldn't be an issue.

That said, I am always open to ways to improve my workflow -especially with the more onerous scanning process! So any suggestions are welcome and I happy to give them a try.

Thanks for the comments and tips.

 

[Gary Sandhu]

Nice comparison! I just bought the Mamiya 7 with 80mm lens and am waiting for some roll film to arrive.

 

[aizan]

i have several questions. first, how did you add grain to the m8 shot? truegrain or something else? what do high iso shots look like with added grain? and at what print size would you say they even out? the largest printer i will likely ever get is an epson 3800 or hp b9180, so that's 17'' and 13'' tall, respectively.

 

[gdi]

i have several questions. first, how did you add grain to the m8 shot? truegrain or something else? what do high iso shots look like with added grain? and at what print size would you say they even out? the largest printer i will likely ever get is an epson 3800 or hp b9180, so that's 17'' and 13'' tall, respectively.

I did not use anything sophisticated for adding noise, just PS. I have never added noise to M8 High ISO shots, they seem to have enough - but if you are enlarging it may make sense.

I doubt many people could tell a significant difference in an M8 print vs a 6x7 at sizes up to A3. If you compared side by side, there my be a noticeable difference in smoothness of tonality in the MF.