Schneider 38 Angulon or 35 Digitar for P45+

[tashley]

Hey Guys,

I'm just placing an order for a Silvestri Bicam for use with my P45+ back - aim is to do some stitching but mainly use it straight or with some rise. I've been recommended the 38 because of its wider image circle and been told that its not being a specifically designed digital lens is no loss.... however the Schneider website has pages of 'why you should use digitar' diagrams that seem irrefutably clear on the issues of pixel pitch and LPM and so on.

Whichever I choose would be used in a helicoid mount,  without bellows.

Head scratch. Grrr. Anyone got a clue here?!

t

[thsinar]

I would also consider the Rodenstock Digital HR 4/35 equivalent.

As you have seen it from the charts, the difference in resolution power is noticeable and although the 38mm has a bigger image circle, the resolution decreases from the center to the edge, making it very unlikely that you can use it fully.

The 35mm HR Rodenstock has a smaller circle than the 35mm Digitar, but on paper only: the values given by Rodenstock are very "conservative" and on the very secure side. One can shift a few mm more without any noticeable degradation at the edge, and this at full open aperture, which is useful.

Best regards,
Thierry

Hey Guys,

I'm just placing an order for a Silvestri Bicam for use with my P45+ back - aim is to do some stitching but mainly use it straight or with some rise. I've been recommended the 38 because of its wider image circle and been told that its not being a specifically designed digital lens is no loss.... however the Schneider website has pages of 'why you should use digitar' diagrams that seem irrefutably clear on the issues of pixel pitch and LPM and so on.

Whichever I choose would be used in a helicoid mount,  without bellows.

Head scratch. Grrr. Anyone got a clue here?!

t

[jonstewart]

Hey Guys,
however the Schneider website has pages of 'why you should use digitar' diagrams that seem irrefutably clear on the issues of pixel pitch and LPM and so on.


t

This only goes to show why you should get both lenses and test them for yourself. I was all fixed on getting the 35, for the same reasons as Tim above, until I actually tested both.

The 38 is perfectly fine out to the edges, despite all the nice little charts and mtf data. I tested both of these lenses mentioned and preferred the 38SA for a number of reasons, one of them being that you could actually do simultaneous shifting without hitting the edge of the image circle very quick. Other reasons included better shadow saturation, and more gentle IQ falloff over f11.

@Thierry: Are Rodenstock's 'conservative figures' covering shifting in one direction only at a time, or simultaneous shifting? With an image circle smaller than the Schneider, it's likely to make the Rodenstock 35 quite unsuitable for the Bicam (which, with the stitching back has about 35mm movement in both X and Y directions). What you think?

[tashley]

The 38 is perfectly fine out to the edges, despite all the nice little charts and mtf data. I tested both of these lenses mentioned and preferred the 38SA for a number of reasons, one of them being that you could actually do simultaneous shifting without hitting the edge of the image circle very quick. Other reasons included better shadow saturation, and more gentle IQ falloff over f11.

Thanks (yet again!) for your advice!

There are trade-offs to be made between shiftability (image circle, both size and quality at the margins) and what the edges are like in a 'straight shot' and issues of overall IQ, which is an intangible. Since in the real world, unless one's dealer has unpackaged versions of both, one simply can't try everything, I'll rely on the experience and opinions of others here in assessing those trade-offs and in this instance my criteria are firstly for the best IQ unshifted. I believe that with tripod mounting and careful focus, a P45+ file can be upresed to 400% for BIG enlargements but that this will be telling of any edge lens weakness. The reason for adding a tech cam to my armoury is that even the best retrofocus lenses struggle here so I do need to ensure that my alternative performs as needed.

Having stated that intent, the next thing I'd like to be able to ensure is the creation of 'flat stitched' images (i.e. no parallax movement) of an excellent quality sufficient to give an eventual file of 2:3 rather than 3:4 ratio, using the full frame height. That represents about 6.25% extra width to each of the left and right. Anything beyond that is a much appreciated bonus! But as I say, what I really need is excellent corners and edges, especially at the sides...

[thsinar]

The Rodenstock HR 4/35 mm has an image angle of 90° for an image circle of 70 mm (stopped down 2 f-stops = at f 8 and at infinity), allowing 6 mm shift up (or down), resp. 5 mm for shift for left (or right), or if you prefer, totally 12 mm vertically, resp. 10 mm horizontally. This for a 36x48 mm sensor. Those are the figures written in the technical data.

When using this lens, one can easily go 12 mm up (or down), resp. 10 mm left (or right), or if you prefer, totally 24 mm, resp. 20 mm, without noticing any loss of quality on the edge of the image (though with a slight distortion).

That is why I was saying that the figures in the technical data are conservative.

The Schneider Digitar 5.6/35 mm is given with an image angle of 88° full open and 102° stopped down at f 11 (2 f-stop), with an image circle of 70mm at f 11, and a total shift of 20 mm vertically (or 10 mm up, resp. 10 mm down) and 16 mm horizontally (or 8 mm left, resp. 8 mm right), for f 11 and at infinity.

Therefore I conclude from it that both lenses have similar shift ways, +/- and at the same given f-stop of f 8 or f 11. But the Rodenstock has IMO the advantage that it suffers much less when used full open at f 4.

Best regards,
Thierry

@Thierry: Are Rodenstock's 'conservative figures' covering shifting in one direction only at a time, or simultaneous shifting? With an image circle smaller than the Schneider, it's likely to make the Rodenstock 35 quite unsuitable for the Bicam (which, with the stitching back has about 35mm movement in both X and Y directions). What you think?

[jonstewart]

The Rodenstock HR 4/35 mm has an image angle of 90° for an image circle of 70 mm (stopped down 2 f-stops = at f 8 and at infinity), allowing 6 mm shift up (or down), resp. 5 mm for shift for left (or right), or if you prefer, totally 12 mm vertically, resp. 10 mm horizontally. This for a 36x48 mm sensor. Those are the figures written in the technical data.

When using this lens, one can easily go 12 mm up (or down), resp. 10 mm left (or right), or if you prefer, totally 24 mm, resp. 20 mm, without noticing any loss of quality on the edge of the image (though with a slight distortion).

That is why I was saying that the figures in the technical data are conservative.

The Schneider Digitar 5.6/35 mm is given with an image angle of 88° full open and 102° stopped down at f 11 (2 f-stop), with an image circle of 70mm at f 11, and a total shift of 20 mm vertically (or 10 mm up, resp. 10 mm down) and 16 mm horizontally (or 8 mm left, resp. 8 mm right), for f 11 and at infinity.

Therefore I conclude from it that both lenses have similar shift ways, +/- and at the same given f-stop of f 8 or f 11. But the Rodenstock has IMO the advantage that it suffers much less when used full open at f 4.

Best regards,
Thierry

Thanks, Thierry. Good job I'm generally not using it at f4, but at f11-f16. With the 38SA, I can go simultaneously 16mm vertical and horizontal in either direction without loss of quality (there is some falloff, corrected with a CF). So, both lenses would be usable on the bicam.

Any chance you can answer my question though? I am assuming you have this lens and use it. Forgive me if it is a wrong assumption.
How much simultaneous shift can you do with the Rodie lens? The Schneider figures on their website and publications are for shift in ONLY ONE direction, and I might assume Rodenstocks are quoted in the same way.

TIA
J

[filmcapture]

T
The Schneider Digitar 5.6/35 mm is given with an image angle of 88° full open and 102° stopped down at f 11 (2 f-stop), with an image circle of 70mm at f 11, and a total shift of 20 mm vertically (or 10 mm up, resp. 10 mm down) and 16 mm horizontally (or 8 mm left, resp. 8 mm right), for f 11 and at infinity.

Apo-Digitar 5.6/35 mm XL Technical Info:

Aperture: f 5.6 - f 22
Image circle at full aperture: 90 mm
Image circle at f 11: 90 mm
Max. angle: 102 °
Distance settings: 0.35 m - infinity

[adammork]

The Schneider Digitar 5.6/35 mm is given with an image angle of 88° full open and 102° stopped down at f 11 (2 f-stop), with an image circle of 70mm at f 11, and a total shift of 20 mm vertically (or 10 mm up, resp. 10 mm down) and 16 mm horizontally (or 8 mm left, resp. 8 mm right), for f 11 and at infinity.

Therefore I conclude from it that both lenses have similar shift ways, +/- and at the same given f-stop of f 8 or f 11. But the Rodenstock has IMO the advantage that it suffers much less when used full open at f 4.

Best regards,
Thierry
[/quote]

Dear Thierry

maybe on paper but not in real life......

with the Schneider Apo Digitar 35mm you can shift about 18mm in all directions 18mm up, or 18mm down, or 18mm left or..... on an Aptus 75.

If you remove the centerfilter it gives you a couple of extra mm.

the sharpness will drop towards the edge of the image circle. No problem if it's in the sky, maybe a problem if there is fine texture that are importen to you.

the Schneider Apo Digitar 35mm is also practically free of distortion.

All the lenses have strong and weak sides, if possible test them and make your own opinion based on your use.

Very best,
Adam

[BJNY]

How would Hasselblad's 40mm IF and Rollei's 40mm Super Angulon compare?

[schaubild]

....

maybe on paper but not in real life......

with the Schneider Apo Digitar 35mm you can shift about 18mm in all directions 18mm up, or 18mm down, or 18mm left or..... on an Aptus 75.
.....

18mm in all directions? How should that work with a non square sensor?
Geometry tells something different.

[thsinar]

That cannot be, the same (usable) image circle at 2 different f-stops, even if written somewhere, same with the image angle.

Here they have what seems to be the correct (logical) figures:

https://www.vistek.ca/store/ProPhotoLargeFo...lens/Specs.aspx

Image angle = 102° is given for f 11. At f 5.6 it is 88°

Best regards,
Thierry

Apo-Digitar 5.6/35 mm XL Technical Info:

Image circle at full aperture: 90 mm
Image circle at f 11: 90 mm
Max. angle: 102 °

[thsinar]

Yes, I did use the Schneider 35mm in a way that corroborates what I am writing: both lenses have similar shift ways in both directions. And of course, the Rodenstock can as well be used on the Bicam (though I did not use it on it). Adam, in his post above and responding to mine, seems to say that he can use a shift of up to 18 mm with the Schneider 35 mm. I believe him, I don't have the experience he has with this lens. But he says also, that the sharpness drops towards the edge: then it depends effectively if this drop is in a "sky" part of the image, in which case it doesn't really matter, or in a part where you need maximum details and resolution.

Yes, I have the Rodenstock HR 4.0/35 mm (Sinaron Digital 4/35 mm HR) and have use it. It has some visible distortion when shift to the maximum mentioned, but it can be corrected. And yes, that's what I meant, the Schneider figures are in one direction (up or down, resp. left or right), and that's why I have added both directions. And Rodenstock ALSO give the figures in one direction, BUT my findings, and this of many other users of this lens, is that you can easily go 4 mm more in both directions.

But one advantage, IMO, is that you can use the Rodenstock digital HR lenses full open without loss, or very minimal. The Schneider Digitar need, again IMO, to be stopped down by 2 f-stops.

That's all what I wanted to say, and it does by no way mean that I find the Digitar 35mm to be a bad lens or even inferior. It is of course an outstanding lens.

Best regards,
Thierry

Thanks, Thierry. Good job I'm generally not using it at f4, but at f11-f16. With the 38SA, I can go simultaneously 16mm vertical and horizontal in either direction without loss of quality (there is some falloff, corrected with a CF). So, both lenses would be usable on the bicam.

Any chance you can answer my question though? I am assuming you have this lens and use it. Forgive me if it is a wrong assumption.
How much simultaneous shift can you do with the Rodie lens? The Schneider figures on their website and publications are for shift in ONLY ONE direction, and I might assume Rodenstocks are quoted in the same way.

TIA
J

[thsinar]

Dear Adam,

yes, you are certainly right, I believe it (though it must be a bit less in one direction, due to the rectangular sensor). It all depends in which part of the image the sharpness/resolution drop is located.

And as you say it, a lens should be tested, under one's own needs and conditions: you will find out that they are both excellent lenses, with both strong AND weak sides.

Best regards,
Thierry

Dear Thierry

maybe on paper but not in real life......

with the Schneider Apo Digitar 35mm you can shift about 18mm in all directions 18mm up, or 18mm down, or 18mm left or..... on an Aptus 75.

If you remove the centerfilter it gives you a couple of extra mm.

the sharpness will drop towards the edge of the image circle. No problem if it's in the sky, maybe a problem if there is fine texture that are importen to you.

the Schneider Apo Digitar 35mm is also practically free of distortion.

All the lenses have strong and weak sides, if possible test them and make your own opinion based on your use.

Very best,
Adam

[jonstewart]

That's all what I wanted to say, and it does by no way mean that I find the Digitar 35mm to be a bad lens or even inferior. It is of course an outstanding lens.

Best regards,
Thierry

Thanks, Thierry.

[filmcapture]

That cannot be, the same (usable) image circle at 2 different f-stops, even if written somewhere, same with the image angle.

Here they have what seems to be the correct (logical) figures:

https://www.vistek.ca/store/ProPhotoLargeFo...lens/Specs.aspx

Image angle = 102° is given for f 11. At f 5.6 it is 88°

Best regards,
Thierry

OK, based on Schneider's MTF spec (http://www.schneideroptics.com/ecommerce/CatalogItemDetail.aspx?CID=1325&IID=1886), Schneider 35XL digitar image circle = 70mm @ f/5.7 and 90mm @ f/11.

However, as an owner of this lens, I believe Schneider's numbers are very conservative.

[adammork]

Dear Adam,

yes, you are certainly right, I believe it (though it must be a bit less in one direction, due to the rectangular sensor). It all depends in which part of the image the sharpness/resolution drop is located.

And as you say it, a lens should be tested, under one's own needs and conditions: you will find out that they are both excellent lenses, with both strong AND weak sides.

Best regards,
Thierry

18mm in all directions? How should that work with a non square sensor?
Geometry tells something different.

You are both absolutely right - I'm sorry for not being precise, the +18mm I was referring too is in one direction only, along the long side of the sensor - when I'm back in my studio I will test it and post the precise numbers.

Best Adam

[archivue]

i you entend to stitch... i will go with the Sironar digital 35... not the HR.

Considering the Hassy 40... nice colors, but far more distorsion !

The new and expensise rodenstock 40 may be a challenger... but the regular digital 35 is a good value for money !