hy6 lenses

[paulmoorestudio]

warning: a pixel question
Is there a theoretical pixel size limit to the film era designed and produced rolleiflex lenses?
Do they or will they hold up to the new small pixels? And is this why sinar has not gone above 33mp?

I have used mine only up to the hasselblad 528 back which gave me fantastic results, however I prefer not to
rely on multishot as my only vehicle to get me there.

[Carl Glover]

Bear in mind that Leaf has a 56 megapixel back that works with the same lenses via the hy6.

I find the lenses remarkeable - especially the 90mm Schneider Macro.

[guano]

Leaf's Afi-II 10 uses a 56MP back and the same lenses as the Sinar and Rollei Hy6 systems. Sinar's latest move with the eSpirit back was a different direction to increased megapixels (and a welcome one at that). Perhaps it's worth asking for samples from forum members using Leaf's 56MP back to test for yourself! Any volunteers?

*edit* Carl got there first... I second the vote for the 90mm Schneider Macro and also suggest the 80mm f2.

[EricWHiss]

The 90mm macro is indeed a fantastc lens - sharp at distance as well as close up.  But the 80mm PQS also fantastic and actually makes a great macro withextension too. The  Schneider 40,60, and 150 are all great too!  I have not tried the 300 yet but have heard great things.   The zeiss 110/2 and 120/4 macro are also nice lenses and the macro is very shArp at close distances 1meter but has a different look than the schneider optics.  I think one would need a multi shot back to make any real comparisons.

[paulmoorestudio]

The 90mm macro is indeed a fantastc lens - sharp at distance as well as close up.  But the 80mm PQS also fantastic and actually makes a great macro withextension too. The  Schneider 40,60, and 150 are all great too!  I have not tried the 300 yet but have heard great things.   The zeiss 110/2 and 120/4 macro are also nice lenses and the macro is very shArp at close distances 1meter but has a different look than the schneider optics.  I think one would need a multi shot back to make any real comparisons.

does sinar or anyone else have any multi shot samples from the 75h? or one shot from the afi10 ..I saw some on the leaf site but no lens or iso info was posted on the images.
To my eye the lenses are great, and do go the distance but I was looking for a scientific viewpoint.. or maybe they defy what is expected.. I was told way back that they would not hold up to the 528 multi..but  they did.. just looking for the ceiling, maybe we haven't got there yet.

[thsinar]

My feeling, and I say purposely "feeling" because I don't have any hard prove of it, is that it will not become easier, for current existing lenses and the increasing resolution of sensors.

The Schneider, as well as the Zeiss lenses are very good lenses, some are little jewels, they have been named here by many. What I can see, when using those lenses on a Hy6 with a 33 MPx sensor, is an obvious improvement as compared to a 22 MPx sensor with the same lenses, in terms of resolution/details.

From the few image samples I have seen taken with those lenses and a CCD with higher resolution, e.g. with 39 MPx, the difference is not longer that obvious, if at all, if compared to a 33 MPx sensor's image.

I can now only guess what happens with sensors of a resolution around 60 MPx: I have seen a few samples, from the same subject, taken with 60 MPx and with 39 MPx sensors/backs, although not with Schneider or Zeiss lenses. I have honestly seen no improvement in terms of resolution and details. May be the Schneider/Zeiss lenses will render details better, but that is not proven yet.

That leads me to think and even believe, that current MF lenses have reached their limit, when associated with digital sensors of the newer generation.

Again, do not take this as a scientifically proven fact, they are my opinions based on my eyes and my logic. But eventually, what counts for me is what my eyes see and can resolve.

Now this to answer Paul's initial question about the capability of the current Schneider/Zeiss MF lenses to cope with the current highest resolution backs (smaller pixels): yes, they certainly can cope with smaller pixels and higher resolution, when we speak about resolution. Is it worth as to what our eyes see and what is necessary? I let anyone doing his own tests and being its own judge.

More importantly, and not to forget is the fact that diffraction will appear at a larger aperture with smaller pixels than with sensors having larger pixels: the limit seems to be reached here as well (f 16, with 39 MPx +?), when speaking about "usable aperture" vs. "needed DOF". That would be my criterion for the decision.

This opinion does of course not take in account that there are many more factors than simply resolution, when it comes to newer sensors technology, and they are IMO more important to consider. To name a few, and not necessarily in order of importance:

- sensor size
- dynamic range
- sensitivity
- base noise level
- Read-out technology
- etc ...

This all being said, my dream, and it is really and only a dream: having a 9 x 12 cm sensor, with around 90 MPx, which leads to a pixel size of about 11 micros. That would be the ideal sensor for me, allowing again to make use of the large format lenses and stop them down as we used to do on a view camera (f22 - f45 max), with possibly a higher and better response for long exposures, ground-glass viewing/setting, etc ...

Best regards,
Thierry

[EricWHiss]

I'd much rather see larger sensors than a finer sensel pitch.  The problem with smaller pixels isn't so much if the lenses are good enough because they might be, but only at 1 or 2 stops from wide open due to diffraction limits.  I'd much rather see larger sensors with larger sensels (or have pixel binning like the new phase p65+).   How about a 6x6 sensor for the Rollei/Hy6/AFi?  Now that would be something!   I think the AFi-10 is a step in the right direction.  

[yaya]

I'd much rather see larger sensors than a finer sensel pitch.  The problem with smaller pixels isn't so much if the lenses are good enough because they might be, but only at 1 or 2 stops from wide open due to diffraction limits.  I'd much rather see larger sensors with larger sensels (or have pixel binning like the new phase p65+).   How about a 6x6 sensor for the Rollei/Hy6/AFi?  Now that would be something!   I think the AFi-10 is a step in the right direction.

What I can see so far is that the Schneider cope very well with the 6 microns on the AFi-II 10.

I was shooting jewellery the other day with a customer, using the 90mm Macro with/ without ext. tubes and also with the macro bellows attachment. This in my view is the ultimate practical test for the ability to control sharpness Vs DOF.
Bare in mind that this is not a view camera so controlling the focus plain is nearly impossible however we had no diffraction issues at f16 and even f18 and the SensorFlex feature allows you to pull back quite allot and still shoot 36MP square for a <100MB file. At f22 there is a noticeable loss of sharpness but given the 160MB file when downscaling it to an A3/ A4 page it looks absolutely fine.
The 90mm, in terms of sharpness/ resolution is IMO head and shoulders above any other lens from any brand with the exception of the new Rodenstock 70mm which I've tried on the Sinar ArTec but these are two different lenses for different applications.

The 80mm Schneider AFD and the 180mm Schneider AFD I find also very, very good and definitely capable of coping with the 6 micron pixels.

On another occasion we were shooting White Leather Italian furniture in a White cove with the AFi-II 10 and the 90mm alongside an Aptus 75 on a Cambo with a Digitar 90mm. We also tried the AFi-II 10 with the Digitar and although the differences were small, we could definitely see the step up in resolution not only between the 33MP and the 56MP but also between the APO-Symmar and the Digitar.

On our website, the "cowboy" shot was taken with the 80mm at f16, 1/250 and 100iso
Frank Doorhof's shot was taken with the 180mm at f11, 50 iso and 1/125
I took the cityscape image with 40mm Super Angulon handheld at 50iso, 1/60 and f11

I hope this helps

Yair

PS I should have an Aptus-II 10 (56MP) in my hands soon so I can test it on the Mamiya AFDIII. I'm expecting the 120mm to be very good and it'll be interesting to see how the 80mm and some other new D lenses cope.

[BJNY]

Thanks for the GREAT info, Yair.

Am I reading it wrong?
How can the 90mm Makro be head and shoulders above any lens
with the exception of the Rodenstock 70mm
and still be inferior to the 90mm Digitar?

P.S. I always thought the 90mm Makro & Digitar were the SAME optics?

[BJL]

What I can see, when using those lenses on a Hy6 with a 33 MPx sensor, is an obvious improvement as compared to a 22 MPx sensor with the same lenses, in terms of resolution/details.

From the few image samples I have seen taken with those lenses and a CCD with higher resolution, e.g. with 39 MPx, the difference is not longer that obvious, if at all, if compared to a 33 MPx sensor's image.

I have seen a few samples, from the same subject, taken with 60 MPx and with 39 MPx sensors/backs ... I have honestly seen no improvement in terms of resolution and details.

It does not surprise me that the step from 33MP to 39MP shows little difference: that is only a 9% increase in sensor resolution as gauged by "pixels per frame height". From 22MP to 33MP (9 micron to 7.2 micron cell size) is a bigger 25% increase, and Dalsa's latest step from 7.2 microns to 6 microns is a 20% increase.

A comparison of 60MP vs 39MP, or 33MP to 56MP for the Dalsa sensors options for Hy6 lenses, has more potential to show resolution gains, but of course requiring larger prints sizes than needed to see the 22MP vs 33MP difference. Mere 16"x20" prints might not be enough of a test.


How about this idea:
1. print the higher MP image at about 240 to 300 PPI and the lower MP image at equal size, so lower PPI. (Using higher PPI and thus smaller prints might hide differences.) With the new 60MP sensor, that means print widths of about 30" to 36", or about A1.
2. Judge how much difference you see.

Of course you might also want to:
3. decide how much use, if any, you have for prints that large and detailed.

[BJL]

I'd much rather see larger sensors than a finer sensel pitch.  The problem with smaller pixels isn't so much if the lenses are good enough because they might be, but only at 1 or 2 stops from wide open due to diffraction limits.

That diffraction limitation on the aperture sizes and DOF at which one can get "full sensor resolution" is almost entirely related to increasing image resolution (pixel count). A larger sensor with an equal number of larger photosites will be equally limited on the DOF possible while getting full sensor resolution: just at larger f-stops combined with larger focal lengths.

Also, "one or two stops from wide open" is a bit of an exaggeration, unless the lenses are f/5.6 or slower. With the new 6 micron cell size MF sensors, diffraction does not start to significantly cut into the sensor's resolution potential until about f/11 or beyond. (Not the word "significantly", as opposed say "detectably".) [Edit: Yair above suggests the threshold is between f/18 and f/22, so my f/11 is quite cautious, even if some simplistic theories predict problems as early as f/8.]


I do agree though that MF pixel counts might be approaching fundamental optical limits on resolution in high DOF images, where f/11 or smaller apertures were often used with MF film, and where even smaller apertures might be needed with the greater enlargements needed to reveal all the detail recorded by the new 6 micron cell size MF sensors.

[yaya]

Thanks for the GREAT info, Yair.

Am I reading it wrong?
How can the 90mm Makro be head and shoulders above any lens
with the exception of the Rodenstock 70mm
and still be inferior to the 90mm Digitar?

P.S. I always thought the 90mm Makro & Digitar were the SAME optics?

Sorry Billy I should have made it clearer...in the furniture shoot the 90 APO-Symmar on the AFi was better than the Digitar on the Cambo

Yair

[thsinar]

BJL,

That is absolutely my understanding and the tests I would do, when comparing the resolution and details.

But for me, that is only one (little) factor in the whole quality chain, among many others and when it comes to these newer sensors. One speaks only about "sensors with x MPx", emphasizing only the resolution increase, when there is IMO at least 1 more important technologic step with these new CCD's, the increase of the capture size.

Year after year, since the beginning of digital, the emphasis has been put on resolution only, being it by manufacturers, dealers and photographers, as if this was the most important factor in the whole quality chain, forgetting that there is much more behind it.

And what would be as much a criterion of decision and choice for me, much more than resolution, would be the fact that diffraction increases with pixels becoming smaller. Yair seems to confirm what I was guessing: f 18 (f 16 1/3) seems to be the limit above which noticeable diffraction "kills" details. That's not so much of a problem when shooting on a view camera and when able to adjust the sharpness plane and DOF, but with a camera with fixed lens and image planes, that can be of concern.

Best regards,
Thierry

A comparison of 60MP vs 39MP, or 33MP to 56MP for the Dalsa sensors options for Hy6 lenses, has more potential to show resolution gains, but of course requiring larger prints sizes than needed to see the 22MP vs 33MP difference. Mere 16"x20" prints might not be enough of a test.


How about this idea:
1. print the higher MP image at about 240 to 300 PPI and the lower MP image at equal size, so lower PPI. (Using higher PPI and thus smaller prints might hide differences.) With the new 60MP sensor, that means print widths of about 30" to 36", or about A1.
2. Judge how much difference you see.

Of course you might also want to:
3. decide how much use, if any, you have for prints that large and detailed.

[Guy Mancuso]

What I can see so far is that the Schneider cope very well with the 6 microns on the AFi-II 10.

I was shooting jewellery the other day with a customer, using the 90mm Macro with/ without ext. tubes and also with the macro bellows attachment. This in my view is the ultimate practical test for the ability to control sharpness Vs DOF.
Bare in mind that this is not a view camera so controlling the focus plain is nearly impossible however we had no diffraction issues at f16 and even f18 and the SensorFlex feature allows you to pull back quite allot and still shoot 36MP square for a <100MB file. At f22 there is a noticeable loss of sharpness but given the 160MB file when downscaling it to an A3/ A4 page it looks absolutely fine.
The 90mm, in terms of sharpness/ resolution is IMO head and shoulders above any other lens from any brand with the exception of the new Rodenstock 70mm which I've tried on the Sinar ArTec but these are two different lenses for different applications.

The 80mm Schneider AFD and the 180mm Schneider AFD I find also very, very good and definitely capable of coping with the 6 micron pixels.

On another occasion we were shooting White Leather Italian furniture in a White cove with the AFi-II 10 and the 90mm alongside an Aptus 75 on a Cambo with a Digitar 90mm. We also tried the AFi-II 10 with the Digitar and although the differences were small, we could definitely see the step up in resolution not only between the 33MP and the 56MP but also between the APO-Symmar and the Digitar.

On our website, the "cowboy" shot was taken with the 80mm at f16, 1/250 and 100iso
Frank Doorhof's shot was taken with the 180mm at f11, 50 iso and 1/125
I took the cityscape image with 40mm Super Angulon handheld at 50iso, 1/60 and f11

I hope this helps

Yair

PS I should have an Aptus-II 10 (56MP) in my hands soon so I can test it on the Mamiya AFDIII. I'm expecting the 120mm to be very good and it'll be interesting to see how the 80mm and some other new D lenses cope.

Grab a 150mm 2.8 D to go with that.  What Jack and I noticed with the 80d and 150 d when going from the P25 , P45 and P65 that the lenses did well with all of them but the sensor resolution was more the determining factor. I think that would be the basis for any of the MF lenses to start with. Now between brands and such i don't know but honestly not sure any of the lenses out there of let's say with newer technology for lack of a better description that they would fail the acid test with the bigger backs. I honestly think any of the MF lenses in general will pull it off. Now there maybe like anything else some real stellar ones in the crowd and I really don't think it would be much different from what you see today with the current backs. Even the Mamiya older 55mm lens did very nicely with the P65 Plus . Now different sensor microns i just don't know from a technical standpoint. Heck i hate testing to begin with and that part gets pretty scientific between sensor microns. Bottom line the great lenses that you mention are still great lenses but even the lessor known ones can pull a lot out of that P65 which i was quite surprised it did. I thought some would fall apart at the seems. Now these are eye tests and not scientific studies mind you but I think the top end lenses will still be the top end lenses and with the bigger backs you will see more resolution because of the back. I don't think we hit the sensor limit yet and by the eye it seems to appear that way. But maybe scientific testing that could be proven wrong. For me i am a shooter and go by what i see and really the only thing I trust and as photographers I think that is what really counts. The pixel peepers may argue that point and I can see that side of it also. Just some thoughts. Now maybe the new high res. backs have finally buried some old relics in glass design and that is something I have not tried yet or actually seen.

BTW when we did some of that testing it was around F11 to avoid any diffraction and my guess would be after f16 or so.

[BJL]

Year after year, since the beginning of digital, the emphasis has been put on resolution only, being it by manufacturers, dealers and photographers, as if this was the most important factor in the whole quality chain

Although I agree that there is in some quarters an excessive emphasis on image resolution as indicated by pixel counts, it is a great exaggeration to say that this is the only thing being emphasized. Putting aside the even more pervasive over-emphasis on noise levels at high ISO speeds as a measure of overall virtue, note that Dalsa has made several important innovations in its most recent sensors:
- micro-lenses that work with almost all lenses, thanks to very good sensitivity to highly off-perpendicular incident light, allowing wider use of micro-lenses in MF, and thus about a doubling of sensitivity
- reduced dark noise (mainly through reduced amplifier noise, it seems) leading to increased dynamic range for its new photosites despite their smaller size,
and the most obvious one,
- increased sensor size.

... the fact that diffraction increases with pixels becoming smaller.

How many times does this need to be said:
diffraction is due entirely to the lens, not the pixels, and so when images from sensors of equal size but different resolution ("pixel size") are displayed at the same size, the diffraction effects wil be equal, and so will at most reduce the extent of the resolution advantage of the sensor with more, smaller pixels. That is, smaller pixels in the same sensor format do not in any reasonable sense make anything worse when it comes to overall image resolution/sharpness/detail.

[James R Russell]

Although I agree that there is in some quarters an excessive emphasis on image resolution as indicated by pixel counts, it is a great exaggeration to say that this is the only thing being emphasized.

If your shooting still life, landscapes or shooting for pleasure rather than commerce or editorial, then just about any camera/back will work, obvously some better than others.

If your working professionally, then the "important" list should be much longer than defraction, pixel count, or image resolution.

1.  Is the camera and back reliable.  I mean does it have to go into service every year or for that matter ever?  Does it overshoot, jam, have to be reconnected, burn through batteries.

2.  Does it tether reliably.  Do you need special firewire powered repeaters to work with a variety of computers or when tethering will the back allow itself to be self powered so you can run a 30 ft. cable to the computer.  Trust me on this, nobody wants to get stuck at 10 or 15 ft. as through the day you just won't move around  to the positions you need to be in to explore something unique.

3.  Is the software stable.  Can you plug and unplug withouit issue, restarts or crashes.  Nothing is a buzz kill worse than watching a tech muddle through the ups and downs of trying to fix anything on set.

4.  Is the software intuitive.  Honestly can you plug the camera into the computer, set a collection folder and shoot sans digital tech?  

5.  Does the lcd work when tethered.  Medium format lcd's are challanged at best, but during pressured production running over to the computer to see if the last frame was ok will wear you out.

6.  Is the file format out of camera generic.  Will it work in almost any 3rd party software without any type of conversion regardless of compression.  Nothing adds to workflow like taking raws, converting them, then trying out other software.  Nothing doubles storage space like a system that requires you to convert the files. Do you then toss the originals, or keep them just in case.  The ability to just drag a file into lightroom, RD or photoshop without drama is almost mandatory in my world  because regadless of look, style, post processing, any file that goes to a retoucher usually requires a raw file so they can add detail when needed.  Can a retoucher process your original file.

7.  Are the file formats all the same, tethered/non tethered.  Once again step back and think worse case sceanrio.  Your shooting tethered, but you see a beautiful image of a model in makeup.  Can you yank the firewire cord, throw in a cf card and just just a handfull of images without having to go back into the makers software to convert them for use. Worse case sceanrio 2.  The generator dies, the computer fails,  can you just shoot non tethered drag and copy images without worrying about the format.

8.  Is the camera maker transparent.  Will they actually take a phone call, answer a problem, get back on an e-mail in a day, not a week.  Believe you when you say you have half black frames, or green previews, or do they say check the firewire cable, or talk to you like it's the first time you've held a camera in your hand. Do they have expert tech on hand in most major countries or do you have to file support cases and wait for replies.

9.  This holds true for dealers.  Price savings is one thing, interfering with a project is another.  Does the dealer answer their phone that minute, give you cell phone numbers, late night numbers, do they stand behind everything they promised at purchase.

10.  Is the system full featured.  Are all the lenses and accessories available.  I hear all the time the Contax is a dead system or nobody wants to use an H series on anything but a blad back because it is now a locked down system.  Forget that thought because there is so much product from those two systems floating around it's harder not to find something than to actually find it.    With any new system you going to have to wait on lenses, finders, grips.  Do you need to send it in to update firmware.  Who can guess when or if they need their system intact.

11.  Is the software released on time, or when the maker promises.  Now in medium format the answer is NO because all of the medium format makers have missed their own deadlines, though does the older software still work reliably and if so then you can continue working while the makers figure out how to get their software out in a stable fashion.

12.  Can you use this camera/back for 5 years.  These things, even at today's reduced prices are expensive and we're getting past the point of always buying new stuff for an extra 20% in detail, or 14% faster.  

13.  Will the dealer/maker give you real world references from people they don't have deals with.  Can you call Joe or Jill and find out what their real experiences are like.  Can you get the complete unvarnished truth.

14.  Which camera/back or dslr do you use for the job of your life.  The carreer changing job that makes or breaks you.  Which camera do you pull out of the bag and do you trust the camera without backup.  I personally wouldn't shoot anything without multiple backups, but when you look around the room which one is the most reliable, which one will not keep you from working.

15.  Can you rent one if yours goes down, or even worse case scenario 3 can you be in any major world city and buy one off the shelf.  

[BJL]

James,

   that is a great list, but dare I suggest that its spirit can be summarized in one important idea: the mistake of measuring photographic equipment solely or predominately by the attributes of the sensor, disregarding the rest of the body, the available lenses and accessories, and so on.

As I have said a number of times before, film cameras using identical "sensors" (35mm film) vary hugely in price: from about ten dollars to some thousands. This tells me that people who declare that a DSLR or digital back is overpriced simply due to sharing a sensor with another less expensive alternative are missing a lot of the picture.

[thsinar]

Yes, that was actually my message. Although you are right that other "issues" are also emphasized (noise level, high ISO, etc ...), this holds true when in a discussion forum like here. As soon as you go out and speak with photographers the picture is completely different, and the first question ALWAYS asked is "how many mega-pixels does this back/sensor have", which in return suggests to me the importance of this for most photographers. Very rarely am I asked about other issues and factors all being an important part of the quality chain of an image.

Although I agree that there is in some quarters an excessive emphasis on image resolution as indicated by pixel counts, it is a great exaggeration to say that this is the only thing being emphasized. Putting aside the even more pervasive over-emphasis on noise levels at high ISO speeds as a measure of overall virtue, note that Dalsa has made several important innovations in its most recent sensors:
- micro-lenses that work with almost all lenses, thanks to very good sensitivity to highly off-perpendicular incident light, allowing wider use of micro-lenses in MF, and thus about a doubling of sensitivity
- reduced dark noise (mainly through reduced amplifier noise, it seems) leading to increased dynamic range for its new photosites despite their smaller size,
and the most obvious one,
- increased sensor size.

That is well understood and obviously true, that diffraction comes solely from the lens (AND how much it is stopped down), and not from the pixel itself: I did not mean or want to suggest such, but may be expressed myself in a way to make one think so. What I meant to say was that the EFFECT of diffraction increases with pixels becoming smaller. So the pixel size does effectively have an indirect effect. And yes, it will affect and reduce the extent of the resolution: that is all what it is about and what I was meaning to say. The real question being and remaining: is this increase in resolution worth it? My opinion is that I would rather be pleased by the other improvements like sensor size increase, etc ...

BTW: there is a nice tutorial about "Diffraction & Photography" here:

http://www.cambridgeincolour.com/tutorials...photography.htm

Best regards,
Thierry

How many times does this need to be said:
diffraction is due entirely to the lens, not the pixels, and so when images from sensors of equal size but different resolution ("pixel size") are displayed at the same size, the diffraction effects wil be equal, and so will at most reduce the extent of the resolution advantage of the sensor with more, smaller pixels. That is, smaller pixels in the same sensor format do not in any reasonable sense make anything worse when it comes to overall image resolution/sharpness/detail.