Canon 1DSMK3 test image compared to Phase Backs

[Frank Doorhof]

I saw an incredible boost in my pictures when I switched from the 20D to the 5D, the pictures were more 3D and much more "real".

When I switched to MF it was an even bigger jump forward.

I think that lenses can still perform very well, people are talking about 1:1 resolution but they forget that you need much more pixels than 1:1 to really capture a scene, for example for audio we need at least 44.1 Khz to make a sound sound natural, although most people don't hear above 15K.
This is called the nyquist factor.

When we look at lenses I THINK that the same rule applies.
So if you outperform your lens on a 1:1 pixel ratio there will still be more detail than people think.

I do STRONGLY believe however that especially on cameras like the 5D/1DsII/1DsIII the lensquality is much more important than on a 4MP camera.
But I do not believe you are outperforming the glass with a 22MP sensor, but I could be wrong.

But the better glass, the better the picture that stands without a doubt.
Some people however claim that lenses can only perform up untilll XX megapixels, but that story has been going on since as long as I know.
The 1DsII would outresolve any lens, and now that the 1DsIII is here you can see a big boost in detail even with lensen like a Tamron 28-75 which shows alot more on the 1DsIII than on my 5D
Let alone the real good and exotic glass.

People however have to learn to not only look at the paper with specs, real life is so much different. That's the reason I don't respond much more in this thread, everything has been said.

[uaiomex]

Frank:
I've read most posts already and I completely agree with you. I'm not equalizing 35 with MF, IQ likewise. "Dangerously aproaching" means mainly, it will erode more and more from dmf sales.

This is bad and good.

For the good part, Nikon and mainly Canon are providing pretty stable and reliable cameras that can do most things mf cameras can at a price range where most photogs around the globe can afford without a mortgage.

For the bad part, if this happens, lower number of mf sales will keep prices high or even will increasing them.

As for me, in order to save some money, for years I've bought from US vendors, first by fax, now online. This means I need to buy pretty reliable tried&true equipment, on this department, digital 35 is light years ahead of mf digital gear. So, currently I am considring investing in 1DsIV and 3 top notch quality primes. Not cheap either, but it will allow me to offer my clients a better product yet. I could do it within reasonable spending without the lacerating, painful, bleeding spending of going medium format. On top of that, maybe in a few months we'll have a 5D successor "dangerously aproaching" the 1Ds4 image quality.

Color lens casts, centerfold issues, tethered shooting problems, close systems&open systems, back compatibility, clumsy software, etc, etc, etc, have put me off in the last months, after seriously considering through all this year to jump to medium format digital, I think I'll pass for the time being.

Best everything for 2008 to all
Eduardo

[Frank Doorhof]

That's looking at : Paper.
When you look at real good photographs you will still see a much more profound difference than what I have seen so far in tests.

As mentioned before I have travelled recently with the 5D and Leaf Aptus 22 and shot almost everything with the Leaf for the simple reason that even on my laptop screen the MF files looked so much better in depth and feel of being there.

The fun thing is that we made a small book in which we mixed the shots of the 5D and with the Leaf on normal sized prints, my mother (no photographer) picked out the leaf shots as being special, when I asked her why ?
she said, with those shots I have the feeling of being there were the other pictures look flat, not bad but more flat.

Maybe I got the depth thing from my mother
But it's fun that she commented on it.

[Ray]

As mentioned before I have travelled recently with the 5D and Leaf Aptus 22 and shot almost everything with the Leaf for the simple reason that even on my laptop screen the MF files looked so much better in depth and feel of being there.
[a href=\"index.php?act=findpost&pid=163406\"][{POST_SNAPBACK}][/a]

Frank,
Perhaps a similar comment could have been made by someone travelling with a Canon D60 and a 1Ds which has almost double the pixel count and a sensor 2.6x the size.

Unless these mysterious qualities you refer to are due largely to the lack of an AA filter, one might expect that owners of a 5D could claim the same additional depth, reality and 3-dimensionality when comparing 5D images with those from a 40D or Olympus E-3.

There has to be a reason why you are getting this MF effect. If it's entirely due to sensor size, then the same effect should be apparent if one goes down a couple of format sizes. Remember also, the difference in sensor size between a 40D or E-3 and FF 35mm is greater than the size difference between FF 35mm and MFDB sensors.

[Frank Doorhof]

Well I never said otherwise

When I switched from the 20D (1.6) to the 5D (FF) I saw a huge jump in image quality but also in depth. When chaning to MF it's the same PLUS all the extra things like 16 bits, DR, no AA etc.

So like I said before for me a large part is in the sensor size that sets the systems apart.
You can NEVER get the same FOV with a FF 35mm camera you get with the MF system, and visa versa offcourse that's why it's comparing apples and oranges.

[Ray]

Well I never said otherwise

When I switched from the 20D (1.6) to the 5D (FF) I saw a huge jump in image quality but also in depth. When chaning to MF it's the same PLUS all the extra things like 16 bits, DR, no AA etc.

So like I said before for me a large part is in the sensor size that sets the systems apart.
You can NEVER get the same FOV with a FF 35mm camera you get with the MF system, and visa versa offcourse that's why it's comparing apples and oranges.
[a href=\"index.php?act=findpost&pid=163551\"][{POST_SNAPBACK}][/a]

There's something here that doesn't seem quite right, Frank. Having myself used a 6mp D60, an 8mp D20 and and a 12.7mp 5D, I have not been struck by any extra sense of reality or 3-dimensionality from the 5D.

The differences I notice are basically higher resolution and more detailed images from the 5D; a necessity to use f16 with the 5D instead of f10 or f11 that I would use with the D60 or 20D for the greatest DoF without noticeable or significant sacrifice of resolution and detail; a shallower DoF with the 5D at the same f stop which gives the 5D a slight advantage when shallowness of DoF is sought, and a wider FoV with the same lenses.

I simply don't understand your comment that you can never get the same FoV with a FF 35mm as you would get with a 36x48mm sensor. Are you referring to differences in aspect ratio? If so, you simply have to crop one format to the same aspect ratio as the other and use the appropriate lens.

For example, the widest FoV you would normally get with 35mm would result from use of a 12mm lens. If you crop the 35mm format to the same aspect ratio as a Phase DB, you'd need to use an 18mm with the Phase back.

I'm no expert on the range of MF lenses avbailable, but the widest Digitar lens with the smaller image circle ideal for the 48x36mm format seems to be 24mm.
If you want the same horizontal FoV that 35mm can give you with a 12mm lens, then you need a 16mm Digitar lens with reduced image circle.

It looks to me as though MFDBs are at a distinct disadvantage when it comes to maximum FoV.

[Dinarius]

I think digital MF is a work in progress, while Canon and Nikon are playing at the top of their game.

Edmund
[a href=\"index.php?act=findpost&pid=163258\"][{POST_SNAPBACK}][/a]

An interesting point.

However, if in (say) four years time, pixels were 2 microns wide instead of the current 6(ish), wouldn't that rewrite the rule book? (And I'm speaking as an MFDB advocate.)

As regards sensor size, Hassie and the like have quite a way to go before they use up the space available to them. i.e. 6x6cm.

D.

ps....what MFDB system are you using?

[Frank Doorhof]

When you crop you can, but why would you crop
That's the whole reason to go to a bigger sensor.
Heck when I crop enough I can emulate my cellphone with my 5D.....

This is not going anywhere.

Just mount a 35mm on a medium format back and shoot, and you will NEVER ask these questions again
You will not even have to compare it to the DSLR with a 17-20mm

[Ray]

When you crop you can, but why would you crop
That's the whole reason to go to a bigger sensor.
Heck when I crop enough I can emulate my cellphone with my 5D.....

This is not going anywhere.

Well, it is going somewhere. We've just learned that you don't crop. I frequently crop my images. Landscapes often require a wider aspect ratio than portraits. I crop for the purpose of creating a more pleasing composition regardless of the size of the sensor.

Just mount a 35mm on a medium format back and shoot, and you will NEVER ask these questions again
You will not even have to compare it to the DSLR with a 17-20mm

Not sure I understand your point here, Frank. Are you saying that a 35mm frame on a medium format back is good, or not good? Some of the first DBs, up to around 16mp, were 35mm format with MF lenses, weren't they. It must have been very difficult to get a wide FoV.

[nicolaasdb]

It amazes me that the DB vs 35mm digital is constantly debated........most "photographers" couldn't get a nice image with a snap shot camera....so what does it matter that your image has more depth??? So you can see even better that the images suck??

Seriously.......stop debating and get your ass back behind your camera's (whatever kind it is) and train your EYE!!!!

Just had to vent this before 2008 starts. I see a lot of very badly/poorly composed/lite and retouched images on these forums! Why not start a tread with some quality images??

[Frank Doorhof]

Well to answer both questions.

Try this with a 17mm on a DSLR ?



No retouching to straighten lines.

[Ray]

Well to answer both questions.

Try this with a 17mm on a DSLR ?

 .

[Frank Doorhof]

Well, I can't make anything else from it.
This is shot with a 35mm on the Mamiya.

[Ray]

Seriously.......stop debating and get your ass back behind your camera's (whatever kind it is) and train your EYE!!!!
[a href=\"index.php?act=findpost&pid=163769\"][{POST_SNAPBACK}][/a]

I'm trying to train my eye to discern these subtle qualities of extra reality and 3-dimensionality attributed to the larger sensor. There's nothing like a properly conducted comparison to bring out such differences, but they seem to be in short supply.

[Ray]

Well, I can't make anything else from it.
This is shot with a 35mm on the Mamiya.
[a href=\"index.php?act=findpost&pid=163774\"][{POST_SNAPBACK}][/a]

You could probably straighten the horizon just a touch. Good job you weren't using a 25mm on the Mamiya (is there one?). The shot could have been quite unflattering.

[Ray]

By the way, Frank, this is a good example of the use of f22 with MF. You can use it without resolution falling noticeably. That's not possible with 35mm and with a cropped format 35mm, f22 would produce a very soft result.

On the other hand, you wouldn't need to use f22 with the smaller format to get the same DoF. F14-16 would be sufficient with FF35mm and f10 or 11 with a 20D.

[Frank Doorhof]

And don't get me wrong, the written word is often more harsh than meant.

1. HOW can you judge difraction on a webformat picture ??
You are right that difraction is less with the MF system without a doubt, but on webformat ??

2. F22 is used to keep the sky/sun from blowing out, not for DOF.
I could understand the question when the sun is not in the frame, but here I think it's quiet obvious.

3. There is a 28mm for the Mamiya but I hardly use my 35mm (had to look for this shot ) so 28mm is not my choice, however I know the lens and it's straight in the geometry so also that lens would give me a good picture.
Whilst on the 35mm a 15mm will give me funky distortions and a curved horizon.
Remember that a 35mm is still a 35mm lens but with more space arround it, just like a 50mm still is a 50mm on a CROP camera and on a FF camera.
It's behaves exactly the same, but you get a crop.

4. Even f22 on the DSLR would not give me this picture, you really need ISO50 to get this effect and as far as I know there is NO DSLR with a REAL ISO50.
All are exposing the highlights on ISO100.

Again don't get me wrong, I'm not attacking you.
But you are giving me the idea (and maybe also others) that the equiptment I use is better due to:
Commercial brainwashing, placebo effect.
OR
Marginally better than a DSLR, not economical feasable etc.

While in practice and what I have been posting all along they are two different systems with TOTALLY different results for different situations.

[Ray]

1. HOW can you judge difraction on a webformat picture ??
You are right that difraction is less with the MF system without a doubt, but on webformat ??

Frank,
F22 is included in the metadata in ACR. I merely raise the issue because some of you DB users seem to have forgotten that smaller formats require larger apertures. I'm not of course referring to you   .

There is a 28mm for the Mamiya but I hardly use my 35mm (had to look for this shot ) so 28mm is not my choice, however I know the lens and it's straight in the geometry so also that lens would give me a good picture.
Whilst on the 35mm a 15mm will give me funky distortions and a curved horizon.
Remember that a 35mm is still a 35mm lens but with more space arround it, just like a 50mm still is a 50mm on a CROP camera and on a FF camera.
It's behaves exactly the same, but you get a crop.

That's quite true. I've used my Sigma 15-30 zoom on my D60 and 20D and it is effectively cropped to a 24mm lens. If you could fit the lens to an MF camera with 48x36mm sensor, at 15mm the zoom would be like an 11mm lens (in 35mm terms)but with horrible vignetting.

However, since it is possible to get reasonably good 12mm lenses for 35mm, I think one would have to say that regards a wide FoV, 35mm has the advantage.

Would you be able to take the following shot with an MFDB, for example? I sure would appreciate the extra detail that an MFDB could give me, but alas! I don't think your widest wide angle lenses are quite wide enough.

This is another shot at 15mm with a cheap lens and a measly 12.7mp. I've shown a 100% crop plus the area it occupies in the image. On my laptop, the crop represents a 36"x24" print of the full image.

As you will gather, I like messy scenes. I did some perspective correction to the verticals in PS but not much. If a few parts look as though they are leaning a bit, they actually are   .

[attachment=4457:attachment]  [attachment=4458:attachment]  [attachment=4459:attachment]

[favalim]

...
As you will gather, I like messy scenes. I did some perspective correction to the verticals in PS but not much. If a few parts look as though they are leaning a bit, they actually are   .

[attachment=4457:attachment]  [attachment=4458:attachment]  [attachment=4459:attachment]
[a href=\"index.php?act=findpost&pid=163799\"][{POST_SNAPBACK}][/a]

Ray just one observation I have to say looking at your picture: looks very digital
If you still keep on pixel counting and what a 35mm can do in extremes situations youll'never face the real issue of 35mm: it's not at the same level of MFDB in 3D effect, colour fidelity, very good wide angles etc.

[Frank Doorhof]

The calculation is also off I'm afraid.
A 15mm on 35mm is about equal to a 30mm on a MFDB (roughly).
But than as mentioned before it's not comparable because of a totally different FOV and DOF.

That's one of the reasons why the MF system is great for wideangle work, we don't have to work with all the nasty distortions you get from the smaller sensors.

That said I would love to do some portrets with a LARGE format cam, I think that would turn out even more that feel of reality.