Making the jump.. seeking some advise

[Mammy645]

I'm with Marko on this one, the moire issue of a 22MP back is extremely exaggerated going by the posts in this thread. I've been shooting almost exclusively fashion and beauty with my P25 for the last 6 years and it's very rarely an issue, and when you do get it it's easily fixable in post. I even shot a textile catalog once and out of about 500 shots there were only a handful with moire and out of those only 1 needed extra attention to fix.

 

[yaya]

Alex Amengual shoots mostly with Aptus-II 6 (28MP) on 645DF

BTW all current Aptus-II backs in M mount allow for 1/1,600 max leaf shutter speed (and sync) on the 645DF

IMO you won't go wrong with amy of the backs discussed. An Aptus-II will integrate better with Capture One compared to an older Aptus and its LCD will also be active while shooting tethered

Looking forward to seeing you MF photos!

Yair

 

[Don Libby]

And now you're getting feedback from actual users!

 

[FredBGG]

Fred, I've been following some of your threads as you give advice on medium format gear and it's got me mildly curious as to what gear you own and use the most. It seems you have vast experience in digital medium format as you offer a plethora of advice.

Don

I currently own Phase One P25+, Phase One AF (still works with film..), Phase One DF, various lenses, but my favorite is the Hassy 110 f2.

I also use the P25+ with the Fuji GX680 I, III and IIIS. This camera gives me
SLR viewing and tilt/shift on lenses from 50mm to 500mm. As well as 6x8 film.
The lenses are sublime and the control afforded by tilt, shift, swing etc is unmatched in any other MF system.

I will rent other backs here and there.

I shoot about 75% digital and 25% film. I generally shoot film for most important shoots. For L'Oreal for example I would often shoot 8x10 film.

I have been shooting for 30+ years. I have consulted for various photography companies and have been sponsored by Polaroid as well as paid to be a testimonial for their pro ad campaign. I also work in motion pictures in various rolls including visual effects supervisor.

 

[FredBGG]

Oh please, the moire issues of the 22mp backs are not that bad. Yes you do get moire, but its not as if it dominates every frame. And when it does occur, the localized adjustments in C1 do an excellent job of getting rid of it. It is a very rare image you cannot fix. And you can always shoot f18 or so, where it disappears from diffraction. The used kit should be wonderful. Spending (a lot) more for the 33mp back would be a serious waste of money in my opinion. The resolution increase is marginal. The only advantages I can find (having owned an aptus22 and tested the II-7) are better high iso performance and indeed more resistance to moire. And don't rule out the smaller sensors. Bigger is better for shallower DOF, but if you stop down it doesn't matter. You get used to the sensor size in either case. In short, I suggest getting the cheapest kit you can find. It'll be great!

Well said....

No need to spend more on the 33MP. It's nice, but the difference is not that big at all. You really need to double the MP count to see noticeable increases in resolution. For a young fashion photographer the difference would be best spent going to shoot in Paris and Milan for a couple of seasons.

 

[FredBGG]

Moire can actually occur on any sensor, including those with a strong AA filter - the AA filter just dramatically reduces it's occurrence (along with softening every image it captures).

I strongly disagree with "just a little less" in your statement. A 33mp sensor will moire far less frequently than a 22mp sensor. If moire is a concern (e.g. shooting people at f/5.6-f/11) then a 33mp sensor is a huge advantage over a 22mp sensor.

Huge difference is as others have stated and exaggeration. The difference in pixel density between these two sensors is quite small. While the 33 will be slightly better you are still talking about a modest difference in linear resolution.

6726 pixels for the 33 MP and 5356 pixels for the 22 MP. That is not a nigh and day difference.

Bayer array sensors , with or without AA filters are prone to moire. Moiree occures because all colors are not sampled in the same spot.

However there are new sensor colors arrays that dramatically reduce moiree.

Multi shot sensors like the Hasselblad and Sinar ones virtually eliminate moiree, but only with still lifes.

Another interesting thing about fabric moiree is that some fabrics will buzz more on higher MP count sensors than on lower MP count sensors. It is a function of what scale the fabrics texture is compared to the sensors array.

various techniques are used to deal with moiree. For example in motion pictures over sampleing is used.

Anyway when it comes to stills it is very easy to control in post.

 

[Shashin]

It is funny how 48x36mm is not a crop sensor and 44x33mm is, especially when 6x4.5 is about 56x42mm and the largest sensor is 53.7x40.3mm. I shoot with a 48x36mm sensor and a 44x33 sensor and there is no real difference.

But there really is not such thing as a "crop sensor." Now, size matters, but silly definitions do not. Just like film before it, digital is creating its own formats and it is better to learn to work with the format you have than trying to understand it by referencing it to a format you don't.

BTW, my 44x33 sensor performs much better than my 48x36 sensor even with a smaller pixel pitch. I love my p25+ back and use it, but the additional +4mm and +3mm are not really doing much.

Get the best sensor in your budget, whatever that may be.

BTW, the difference in resolving power between a 33MP and 22MP sensor is about 22%. Also look at long exposure limits and noise.

 

[dougpeterson]

Huge difference is as others have stated and exaggeration. The difference in pixel density between these two sensors is quite small. While the 33 will be slightly better you are still talking about a modest difference in linear resolution.

6726 pixels for the 33 MP and 5356 pixels for the 22 MP. That is not a nigh and day difference.

Frequency of Moire occurrence is not linear to sensor resolution. That assumes the lens is perfectly sharp, the DOF covers all subjects all the time, and there is no diffraction in any shot. Moreover, since I'm the guy that ends up getting the call when a customer has a technical question or issue, I have experience with many many photographers with various resolution backs, so I have a broad sample size of real world experience to draw on.

In the real world moire happens far less frequently as you increase resolution. Any sensor can moire, and it's even very possible (as you say) that a specific subject will moire on a high res sensor and not a low-res sensor. But when measured by "how often it's a big issue" a 33mp back is a significant improvement over a 22mp back.

Maybe I shouldn't have said "huge" as I have no way to directly quantify it. But I would argue to the death it's not "small" and is well worth mentioning to someone if they shoot fabrics and are considering an entry-level back.

Here I am in the odd position of arguing that an issue with a product we sell is moderately more annoying than others are saying it is. Who woulda thunk. That's the company ethos though at Capture Integration, tell it like it is :toocool:.

Though I do agree with your overall point that moire is more often talked about than experienced and the vast majority of the time it occurs the software fix is very simple.

Doug Peterson (e-mail Me)
__________________

Head of Technical Services, Capture Integration
Phase One Partner of the Year
Mamiya Leaf, Leica, Arca Swiss, Cambo, Profoto, LaCie, Canon, TTI, Broncolor & More

National: 877.217.9870 | Cell: 740.707.2183
Newsletter | RSS Feed
Buy Capture One 6 at 10% off

 

[dougpeterson]

It is funny how 48x36mm is not a crop sensor and 44x33mm is, especially when 6x4.5 is about 56x42mm and the largest sensor is 53.7x40.3mm. I shoot with a 48x36mm sensor and a 44x33 sensor and there is no real difference.

I try to remember to refer to the 48x36 as a 1.1 crop sensor. If I ever fail to remember to do so please let me know why!

Regarding "full frame"
41.5 x 56.0 - Nominal film gate size of a Mamiya/Phase 645
40.4 x 53.9 - P65+

That leaves half a mm on the left and right and a smidge more than a mm on the top and bottom.

More over the viewfinder of the Mamiya 645 was very close to 100% coverage but not exactly 100%, so the end result is that if you put a P65+ on a Mamiya 645 the viewfinder will be exactly what you get (without a mask).

In other words What You See is What You Get. If that doesn't qualify as "full frame" I don't want to be involved in a conversation about what would.

Now all that said the practical differences I find in using a 1.0, 1.1, or 1.3 crop sensor:
- the larger sensors see more of the weak areas of the less-sharp wide angle lenses. So if you go large sensor and you want to shoot wide angle you need to invest in the best wide angle glass available to take advantage of it. At mid-wide and longer this is a moot point.
- when shooting wide open a 1.0 or 1.1 sensor has a significantly shallower depth of field when filling the frame with the subject, as compared to a 1.3 sensor. If you rarely shoot wide open this won't be especially important. The look of the 150mm f/2.8 lens on a 1.0 or 1.1 sensor, shot wide open, is very hard to match.
- it's nice to look through a larger viewfinder. You notice this the most when, like me, you work with a dozen cameras a month. 1.3 to 1.0 especially is a pretty large change in the size of the viewfinder.
- if your back is against the wall (literally - as in you're shooting e.g. an interior of a yacht) and you're using the widest lens you have in your bag, if you're using a 1.3 crop sensor you will REALLY wish you had a 1.1 or 1.0

I also feel the "medium format" ness of the image is strong on the 1.0 and 1.1 crops vs. 1.3. But I've never tried to directly compare this, and to be honest the terminology and effects involved are very vague "feeling" is the best you say. So I would not make a big deal of this aspect.

Doug Peterson (e-mail Me)
__________________

Head of Technical Services, Capture Integration
Phase One Partner of the Year
Mamiya Leaf, Leica, Arca Swiss, Cambo, Profoto, LaCie, Canon, TTI, Broncolor & More

National: 877.217.9870 | Cell: 740.707.2183
Newsletter | RSS Feed
Buy Capture One 6 at 10% off

 

[Shashin]

Doug, I have no argument that size matters. I have no real problem saying the 80MP sensors are 6x4.5. But I am starting to feel that the term "cropped sensor" is just a disparagement--I doubt most photographers worked in 6x4.5 where they would really understand the meaning of the crop factor, in fact, I have seen more reference to 35mm equivalents for angle of view than 6x4.5 equivalents. 4x5 is no more a "crop" format of 8x10, as 44x33 is somehow less than MFD. And I am getting not much more "medium formatness" with a 48x36 sensor? And then image qualities cut both ways where having more DoF or less DoF, for example, is rather personal, especially since most of the work I see around here has really deep DoF.

As far as your rounding goes, I really am not impressed. Why do you round down for the 44x33 format to 1.3, when the crop factor is 1.27, but you round up for 48x36 to 1.1 when the crop factor is 1.16 (based on the 70mm diagonal of the 6x4.5 image area). It should be 1.2. So the question is how big a deal is the difference between 1.16 and 1.27? If we use your numbers, 1.1 and 1.3, the difference is 18%, if mine, the difference is 9%.

But buying a MFD back/camera is complex. Size is important and the bigger the better. Pixel resolution, pixel pitch, ISO limits, long exposure limits, and microlenses can be equally important and in some cases more than a 9% difference in crop factors.

 

[Guy Mancuso]

That is the way the industry rounds the numbers up for 1.3 and 1.1 that is not a Doug call but a industry one that comes from the OEM's. As far as Moire the bottom line and I tested this the bigger the Microns (9) the more susceptible to Moire given the same subject than compared to a 7.6, 6. and 5.2. It seems from what I have seen the lower the Microns the less the effect even though they all may Moire on the same subject the effect is less with the lower Micron sensors. I tested this awhile back with a 22, 31,39 and 40mpx sensor and the P40 at the time was far less evident in the file which is a 6 micron. It also has a lot to do with the lens performance the better the lens the more you can Moire as it hits the Nyquist limit. Now on crop sensors it is actually quite meaningless in some cases the P40+ performs exactly like the P65+ except the difference being 20mpx has been cut out. In this case size only matters as more MPX are being thrown at the file. It would have nothing to do with Moire as they are the same pixel pitch. Now take a 31mpx sensor compared to the 39mpx Kodak sensor than you will have a difference as micro lenses have been added to the 31mpx. So the size thing has some meaning but only if there is a change to the sensor itself which in this case both H and P did add micro lenses so the performance is different than the 39mpx sensor and in this case it does affect the ISO noise levels as well. The P40+ and P65+ outside the size than it is identical in performance.

 

[dougpeterson]

Doug, I have no argument that size matters. I have no real problem saying the 80MP sensors are 6x4.5. But I am starting to feel that the term "cropped sensor" is just a disparagement--I doubt most photographers worked in 6x4.5 where they would really understand the meaning of the crop factor, in fact, I have seen more reference to 35mm equivalents for angle of view than 6x4.5 equivalents. 4x5 is no more a "crop" format of 8x10, as 44x33 is somehow less than MFD. And I am getting not much more "medium formatness" with a 48x36 sensor? And then image qualities cut both ways where having more DoF or less DoF, for example, is rather personal, especially since most of the work I see around here has really deep DoF.

As far as your rounding goes, I really am not impressed. Why do you round down for the 44x33 format to 1.3, when the crop factor is 1.27, but you round up for 48x36 to 1.1 when the crop factor is 1.16 (based on the 70mm diagonal of the 6x4.5 image area). It should be 1.2. So the question is how big a deal is the difference between 1.16 and 1.27? If we use your numbers, 1.1 and 1.3, the difference is 18%, if mine, the difference is 9%.

But buying a MFD back/camera is complex. Size is important and the bigger the better. Pixel resolution, pixel pitch, ISO limits, long exposure limits, and microlenses can be equally important and in some cases more than a 9% difference in crop factors.

No argument from me about any of this. We post the specific sensor sizes on our website for those that want to be very precise (Leaf / Phase), and, more to the point, in our Printed Product Guide we just show a scale-model of the various sensor sizes (much more informative than the numbers). We really need to get that diagram on our website; I'll talk to our Web Master (o crap, that's me).

Picking a back is complex and I totally agree sensor size is only one of important factors!

(as a trivial aside, I'm using the film gate of the Mamiya 645 as my reference for "full frame" which leads to slightly different numbers than yours and only serves to show that there are many valid ways of thinking about it)

 

[dougpeterson]

Doug, I have no argument that size matters. I have no real problem saying the 80MP sensors are 6x4.5. But I am starting to feel that the term "cropped sensor" is just a disparagement--I doubt most photographers worked in 6x4.5 where they would really understand the meaning of the crop factor, in fact, I have seen more reference to 35mm equivalents for angle of view than 6x4.5 equivalents. 4x5 is no more a "crop" format of 8x10, as 44x33 is somehow less than MFD. And I am getting not much more "medium formatness" with a 48x36 sensor? And then image qualities cut both ways where having more DoF or less DoF, for example, is rather personal, especially since most of the work I see around here has really deep DoF.

Very valid points. DOF (when you want more) is the big trade off with larger sensors. It definitely cuts both ways.

But the "cropped" in this case basically just refers to the fact that most of the bodies that these backs are used on were designed for 645 film and have viewfinders, lenses, and accessories designed around a full 645 frame. Take a Mamiya 645AFD and slap a digital back on with a smaller-than-645 sensor and you are essentially cropped in.

But it's really just semantics. "1.1 crop" is the best way (read: a succinct and relatively clear way that requires little explanation) I've found to describe that size sensor. Using the mm sizes make it very hard (and requires active thought) to visualize the relative sizes for most users (I've found).

Maybe I should just attach the sensor-size-comparison graphic to every post where I reference a sensor size (which is by far the best way to show and describe a certain sensor size IMO), but that seems a bit overboard.

Anyway it sounds like we agree on most points and at the end of the day there is no perfect way to quickly summarize detailed information. We just disagree on what the less imperfect way would be.

 

[FredBGG]

Here I am in the odd position of arguing that an issue with a product we sell is moderately more annoying than others are saying it is. Who woulda thunk. That's the company ethos though at Capture Integration, tell it like it is :toocool:

Then I think it is also very important to tell a young photographer on a budget that moire removal is a few clicks away. Capture One does a great job and it is general not a particularly destructive correction..... and it's free when you buy a back.

Another important point to make is that once you have a MF format camera with a close to full frame sensor you can easily and relatively inexpensively rent a 65 or 80 MP back. The camera will feel just the same. here in LA there are many product photographers with high MP count backs that also work as digital techs/assistants and can come along with a gear package too. Having a digital tech with state of the art gear on the right job can be far more important than owning "a few more megapixels".

Now let me add that despite the animated discussion.... I find it is very productive to have people like you with a good forum presence taking part in these discussions.

While you have plenty of feedback from your clients I have 30 plus years behind cameras shooting fashion and beauty and the A-list. Most of my shoots in Europe take part in large rental studios where there will be three or more other fashion photographers shooting in the studios across the hall.

Also I lived in Italy for 25 years, so I LIKA TO AVVA A GOOD ARM WAVING DISCUSSION... generally starting with "MAMA MIA!"

Now I'm off to the CNC machine shop to work on my on my tilt and shift bellows unit to use my Fuji gx680 lenses on the Phase One DF
in auto stop down mode.

 

[dougpeterson]

All fair points. I live for animated discussion and it's been great to read your experience and opinions on the matter.

 

[Paul66]

One suggestion is to keep the 5D2 for high ISO shots!!!!!!!!!
Second suggestion is to get the camera, or rent one and shoot a lot of practice shots before you do a paid gig, there is some differences in the handling in my opinion.

I would be happy to post shots from my first few shoots with the 645DF/28DM if you'd like

 

[Shashin]

That is the way the industry rounds the numbers up for 1.3 and 1.1 that is not a Doug call but a industry one that comes from the OEM's.

I am sorry, I was not really implying Doug, but Phase One. If Doug took it that I thought it was he that was setting the specs, then I apologize.

But which industry? Not the camera industry, surely. I worked in that industry and my experience is that they know how to round. No engineer I worked with would round 1.1666* down to 1.1. Marketers on the other hand...

 

[Shashin]

Anyway it sounds like we agree on most points and at the end of the day there is no perfect way to quickly summarize detailed information. We just disagree on what the less imperfect way would be.

It is all shades of grey/gray, and I know you understand that. I just react to exaggeration and absolutes. To simply blast a sensor size for being 9% smaller is a bit harsh.

But that D800 is going to kick MFD butt....

:ROTFL::loco::deadhorse: