S5 Pro 12MP resolution ....

Not sure what you're trying to prove (and the chromatic aberration in
the first sample is at a level I'd find unacceptable in a print,
BTW). The S5 Pro has 6mp resolution, period.
Can you explain this?

Nikon D40: Absolute resolution
1600
1550

Fuji S5 Pro:
1900
1700
I think people have to figure out (if they haven't already) if
they're a commodity shooter (e.g. negative film from a compact or
low-end SLR printed at the one-hour processing place) or a serious
shooter (medium format black-and-white or perhaps pro Velvia printed
by hand).
So people who shoot sports, action, wildlife, PJs... are not serious shooters? I'm asking because I have never seen one with a medium format camera.

--
Regards,

Robert
 
Can you explain this?
Yes. Turn a D40's sensor on its side and see how it resolves. As I've written many, many times, the 45 degree twist on the Bayer pattern has tangible improvements on the vertical and horizontal axis at the detriment of the diagonal. The traditional Bayer is the opposite. Fujifilm's twist is useful for most subjects because of, well, gravity, amongst other things. Basically, on the horizontal and vertical axis you get the equivalent of a traditional 8mp Bayer. The cost is stairstepping on diagonals. Yes, you can alias those out, but that comes at the cost of acuity.
I think people have to figure out (if they haven't already) if
they're a commodity shooter (e.g. negative film from a compact or
low-end SLR printed at the one-hour processing place) or a serious
shooter (medium format black-and-white or perhaps pro Velvia printed
by hand).
So people who shoot sports, action, wildlife, PJs... are not serious
shooters?
There was an "or" in my sentence. The problem with generalizations is that they generalize. Forum posts don't have room for long, considered essays.
I'm asking because I have never seen one with a medium
format camera.
I have. In fact, I tell a story about one of mentors, who sent me out to shoot a track meet using a MF DLR. So I've done it myself. But I've seen Hasselblad and other MF cameras shooting all those things. In fact, I've had several argue with me that their MF choice is better than my best DSLR choice (even after the D3x appeared).

--
Thom Hogan
author, Complete Guides to Nikon bodies (18 and counting)
http://www.bythom.com
 
I think people have to figure out (if they haven't already) if
they're a commodity shooter (e.g. negative film from a compact or
low-end SLR printed at the one-hour processing place) or a serious
shooter (medium format black-and-white or perhaps pro Velvia printed
by hand).
I think your last paragraph is wrong and believe that many others
would agree. Trying to say that art is not produced by "low-end"
equipment sounds like its double rotating and counter intuitive.
I said no such thing.
Errr.....yes you did...or rather, "wrote".

I know you did not intend to say that.......but it slipped by somehow (freudian lingerie, anyone :)
I simply spoke to the acquisition preferences
of different types of shooters.
Now you are more to the point, IMO; this is what you intended to write in the first place; but you did not. Hey we all make mistakes from time to time, you're no exception Thom :¬)
Perhaps you're thinking that only a
"serious shooter" can create art or even interesting photos. I
wouldn't agree.
No Charles wasn't thinking that at all.......you wrote something of the sort....maybe you were? :)
(Just pulling your leg)

Good to know that even Thom has a slip-up from time to time, it makes him more mortal :)
 
I said no such thing. I simply spoke to the acquisition preferences
of different types of shooters. Perhaps you're thinking that only a
"serious shooter" can create art or even interesting photos. I
wouldn't agree.
That is exactly how it sounded though. I think and said no such thing. Just the opposite is what I believe.

Looking at some of your posts you are starting sound like an equipment snob.

--
Charles
My family images are at http://www.stakeman.smugmug.com
Be sure of your subject.
Never, force the shot.
 
There was an "or" in my sentence. The problem with generalizations is that they generalize. Forum posts don't have room for long, considered essays.
Thom: no need to work too hard to defend your statements. I think most of us knew what you intended/meant to say. Being concise in a forum often means being obscure, but I don't think that was the case here.

--
SteveG
'When love and skill work together, expect a masterpiece.'
— Found in a Chinese Fortune Cookie
http://www.stephenmichaelgarey.com
 
nothing really ... except that an interpolated 12MP Fuji S5 Pro image
is quite a bit better than an interpolated 6MP image from any other
6MP camera,
It should be almost exactly as good as 12mp upsized from an 8mp
camera on the horizontal and vertical axis. That stairstepping you
saw? The 6mp camera wouldn't have it (or would have less of it on the
uprez). The problem I have is that it isn't as good as a real 12mp
camera ; ).
So Thom ... it's Saturday morning, and I'm sitting here finally getting around to reading more/much of the S5 Pro User's Guide I bought from you when I get to this paragraph: "Overall, the integrity of the Fujifilm 12mp interpolation is quite good, and coupled with the slight detail bump gained by the on-angle photosite design, the 12mp JPEGs coming out of the S5 Pro are darned close to being what you might expect out of a standard Bayer camera with 12mp."

My point exactly.

But from what you've posted here, it sounds like you might have had second thoughts about that statement since writing it. If so, I don't see why. I really do think the S5 is what that statement says it is.

--
SteveG
'When love and skill work together, expect a masterpiece.'
— Found in a Chinese Fortune Cookie
http://www.stephenmichaelgarey.com
 
nothing really ... except that an interpolated 12MP Fuji S5 Pro image
is quite a bit better than an interpolated 6MP image from any other
6MP camera,
It should be almost exactly as good as 12mp upsized from an 8mp
camera on the horizontal and vertical axis. That stairstepping you
saw? The 6mp camera wouldn't have it (or would have less of it on the
uprez). The problem I have is that it isn't as good as a real 12mp
camera ; ).
So Thom ... it's Saturday morning, and I'm sitting here finally
getting around to reading more/much of the S5 Pro User's Guide I
bought from you when I get to this paragraph: "Overall, the integrity
of the Fujifilm 12mp interpolation is quite good, and coupled with
the slight detail bump gained by the on-angle photosite design, the
12mp JPEGs coming out of the S5 Pro are darned close to being what
you might expect out of a standard Bayer camera with 12mp."
Steve if you view them critically I would say they hold up at about 8mp but......
But from what you've posted here, it sounds like you might have had
second thoughts about that statement since writing it. If so, I don't
see why. I really do think the S5 is what that statement says it is.
I've just recieved back a completed a storybook wedding album today, shot totally with S5's, for a client and it looks superb.....the files print beautifully having a richness about them I love....
best
--
Geoff_R

'Always look on the bright side of life...'
http://www.fightwireimages.com
http://www.sportsshooter.com/members.html?id=7656
 
Look time no see, indeed. I didn't know you were shooting with an S5
(among other cameras). Do you still use your Sigma?
I've got a little DP1 for pleasure that also gets used for candids of groups at weddings and love it......but Dslr's are a whole different ball game these days.
If you've read through some of my posts, you can see I'm new to the
Fuji way of photography. I don't know what took me so long. I've
pretty much experimented with everything else out there .... well,
not EVERYTHING. Haven't used a Pentax or Sony or Leica M8. I do enjoy
the S5. I enjoy the S2 as well. Terrific little camera, though short
on features and not quite the resolver the S5 is.
Bit like me really....came to the Fuji party late...glad I did though and of course managed to get them at the right price:-)
Anyway ... and as always ... good to hear from you.
And you.
The D700 soundsgreat. After a fling with a D2x (loved, loved, loved that > machine!), I've been looking into some of the newer Nikons, ie D300 and > D700 ... plus I'm considering going back to a D2x, the "s" version. We shall
see. I'll look forward to your posts as well.
If you liked your D2X you'll love the D700....its been a bit of a revalation for me....its almost good enough at high iso not to use flash again:-) I dont see much difference in IQ between the D300 and D700 with the exception of DR and high iso performance....so if you dont need high iso the D300 isnt a bad buy and of course Nikon is a bit lacking in good FF glass.....nothing like the Canon F4 range for example.

Anyway its good to touch base with you and I'll keep an eye out for you.
best
--
Geoff_R

'Always look on the bright side of life...'
http://www.fightwireimages.com
http://www.sportsshooter.com/members.html?id=7656
 
So Thom ... it's Saturday morning, and I'm sitting here finally
getting around to reading more/much of the S5 Pro User's Guide I
bought from you when I get to this paragraph: "Overall, the integrity
of the Fujifilm 12mp interpolation is quite good, and coupled with
the slight detail bump gained by the on-angle photosite design, the
12mp JPEGs coming out of the S5 Pro are darned close to being what
you might expect out of a standard Bayer camera with 12mp."
Now here is a quote from Thom's S5 product review:

"Just as with the S3 Pro, my conclusion is that the Fujifilm 6mp SuperCCDs really resolve most detail at about the level you'd expect an 8mp camera to."

So you can see the quote from the e-book and the quote from the S5 review are at odds with one another. I would believe there is an error in the e-book which Thom will correct shortly to get all of us on the same page again. No one has ever promoted the idea the S3 or S5 has 12 mpix of resolution and that includes Prof. Hogan.

I guess if we get the chance to look at a 12mpix EXR sensor in a Fujifilm we will get a chance to see what a 12 mpix Fuji sensor looks like. That will depend upon you choice or resolution, dynamic range or low noise settings you choose for your photos. I say let that debate begin!

--
Alan, in Montana
Photos are cached here,

http://radphotos.net/index.php?option=com_copperminevis&Itemid=33&place=gallery&cat=10069
 
So Thom ... it's Saturday morning, and I'm sitting here finally
getting around to reading more/much of the S5 Pro User's Guide I
bought from you when I get to this paragraph: "Overall, the integrity
of the Fujifilm 12mp interpolation is quite good, and coupled with
the slight detail bump gained by the on-angle photosite design, the
12mp JPEGs coming out of the S5 Pro are darned close to being what
you might expect out of a standard Bayer camera with 12mp."
Now here is a quote from Thom's S5 product review:

"Just as with the S3 Pro, my conclusion is that the Fujifilm 6mp
SuperCCDs really resolve most detail at about the level you'd expect
an 8mp camera to."

So you can see the quote from the e-book and the quote from the S5
review are at odds with one another. I would believe there is an
error in the e-book which Thom will correct shortly to get all of us
on the same page again.
You see, this is what happens when human beings get involved in a process. Errors! Cameras and computers, of course, never ever Err. :-)

--
SteveG
'When love and skill work together, expect a masterpiece.'
— Found in a Chinese Fortune Cookie
http://www.stephenmichaelgarey.com
 
Yes. Turn a D40's sensor on its side and see how it resolves. As I've
written many, many times, the 45 degree twist on the Bayer pattern
has tangible improvements on the vertical and horizontal axis at the
detriment of the diagonal. The traditional Bayer is the opposite.
Fujifilm's twist is useful for most subjects because of, well,
gravity, amongst other things. Basically, on the horizontal and
vertical axis you get the equivalent of a traditional 8mp Bayer. The
cost is stairstepping on diagonals. Yes, you can alias those out, but
that comes at the cost of acuity.
So it DOES have more resolution than 6MP-cameras and your first statement was wrong? BTW, my D50 shows more stair stepping on diagonals than my S5.
I think people have to figure out (if they haven't already) if
they're a commodity shooter (e.g. negative film from a compact or
low-end SLR printed at the one-hour processing place) or a serious
shooter (medium format black-and-white or perhaps pro Velvia printed
by hand).
So people who shoot sports, action, wildlife, PJs... are not serious
shooters?
There was an "or" in my sentence. The problem with generalizations is
that they generalize. Forum posts don't have room for long,
considered essays.
I'm not a native English speaker, so could you please tell me what you wanted to say? What does the "or" change? For me it sounded like "if you are a serious shooter, you need a medium format camera. Sorry if I misunderstood you.
I'm asking because I have never seen one with a medium
format camera.
I have. In fact, I tell a story about one of mentors, who sent me out
to shoot a track meet using a MF DLR. So I've done it myself. But
I've seen Hasselblad and other MF cameras shooting all those things.
In fact, I've had several argue with me that their MF choice is
better than my best DSLR choice (even after the D3x appeared).
Does anyone know a professional sports shooter (NBA, NHL, NFL, Soccer Worldcup, Olympic Games) who uses a Hasselblad for his work? I'd love to see his pictures.

--
Regards,

Robert
http://www.sondek.smugmug.com
 
Yes. Turn a D40's sensor on its side and see how it resolves. As I've
written many, many times, the 45 degree twist on the Bayer pattern
has tangible improvements on the vertical and horizontal axis at the
detriment of the diagonal. The traditional Bayer is the opposite.
Fujifilm's twist is useful for most subjects because of, well,
gravity, amongst other things. Basically, on the horizontal and
vertical axis you get the equivalent of a traditional 8mp Bayer. The
cost is stairstepping on diagonals. Yes, you can alias those out, but
that comes at the cost of acuity.
So it DOES have more resolution than 6MP-cameras and your first
statement was wrong? BTW, my D50 shows more stair stepping on
diagonals than my S5.
you quoted him above....

He said 'equivalent'.. not native.. the 6mp of s sites are not for resolution. it is a 6mp camera that interpolates in camera to 12Mp.. the trade off is more stairstepping on diagonals.

My experience with the S2/3/5 is that it is/has the appearance of more res than my D50 (side by side comparison) but not the D200 or D300. the apparent res figures of the Sx Pro series, keep creeping up.. 8Mp then some said near 10Mp.. well it's only a smidgen from 12Mp.. why leave it there?

it's strengths are elsewhere... even against current releases. be content until/if Fuji does enter the DSLR market again.
I'm not a native English speaker, so could you please tell me what
you wanted to say? What does the "or" change? For me it sounded like
"if you are a serious shooter, you need a medium format camera. Sorry
if I misunderstood you.
I'm asking because I have never seen one with a medium
format camera.
I have. In fact, I tell a story about one of mentors, who sent me out
to shoot a track meet using a MF DLR. So I've done it myself. But
I've seen Hasselblad and other MF cameras shooting all those things.
In fact, I've had several argue with me that their MF choice is
better than my best DSLR choice (even after the D3x appeared).
Does anyone know a professional sports shooter (NBA, NHL, NFL, Soccer
Worldcup, Olympic Games) who uses a Hasselblad for his work? I'd love
to see his pictures.

--
Regards,

Robert
http://www.sondek.smugmug.com
--
There is a very fine line between 'hobby' and 'mental illness.' :'!':
 
He said 'equivalent'.. not native.. the 6mp of s sites are not for
resolution. it is a 6mp camera that interpolates in camera to 12Mp..
the trade off is more stairstepping on diagonals.
Sorry, Alan, but he wrote:
"The S5 Pro has 6mp resolution, period."

That's wrong, period. It has more resolution than a 6mp camera. It's visible and measurable.

--
Regards,

Robert
http://www.sondek.smugmug.com
 
That's wrong, period. It has more resolution than a 6mp camera. It's
visible and measurable.
There's only one problem with both your statement and mine: you can't isolate to the pixel count. So we're both wrong.

Two big things come into play, the angle of the Bayer distribution and the antialiasing filter on top of the sensor. You can have two 6mp cameras that, on a test chart, show different values. In essence, that's what's happening with the S5 versus D50. Nikon has tended to use overly aggressive antialiasing IMHO. Funny thing is, take the AA off, and suddenly that D50 looks a lot like the S5 Pro on the test charts when you twist them 45 degrees.

I don't know where it's still available, but Fujifilm themselves many years ago wrote a white paper talking about the resolution difference (I think it was presented at SPIE). Basically: something like a 1.3x increase on the horizontal and vertical axis, 1.3x decrease on the 45 degree diagonal. The reason why I see it resolves like an 8mp camera is that both the world and our eyes orient more towards the horizontal and vertical axis. Fujifilm's observation about Bayer was quite clever. But the problem is that Fujifilm has been stuck at 6mp for three generations now while their competitors have continued to bump up. Thus, whatever advantage they had when cameras were all 6mp is now completely gone.

This brings up the obvious question: is 6mp enough? And the answer is yes and no. Yes in the sense of casual shooting and use and at modest print sizes, no in the sense of carefully controlled local contrast and larger print sizes. The other thing that worried me about Fujifilm's DSLRs is that I actually think my S2 Pro has more resolvable detail than my S5 Pro. I was reminded of this when I was going through images for a project I'm working on. Someday I'll have to get the two out and test them with a flat field lens on the charts and see what I get.

--
Thom Hogan
author, Complete Guides to Nikon bodies (19 and counting)
http://www.bythom.com
 
So Thom ... it's Saturday morning, and I'm sitting here finally
getting around to reading more/much of the S5 Pro User's Guide I
bought from you when I get to this paragraph: "Overall, the integrity
of the Fujifilm 12mp interpolation is quite good, and coupled with
the slight detail bump gained by the on-angle photosite design, the
12mp JPEGs coming out of the S5 Pro are darned close to being what
you might expect out of a standard Bayer camera with 12mp."
But from what you've posted here, it sounds like you might have had
second thoughts about that statement since writing it. If so, I don't
see why.
Yes, I'd write it differently today. The statement would be "darned close to what you get out of several of the Nikon DSLRs in JPEGs at up to about 10mp."

At the time I wrote that statement, many of the Nikon DSLRs had suboptimal JPEGs. I can clearly resolve more out of NEF files on a D100, for instance, and that continued to be true up through about the D200 generation. Things are a little different today. Plus, I'm talking about JPEGs in that statement, not best possible image.

--
Thom Hogan
author, Complete Guides to Nikon bodies (19 and counting)
http://www.bythom.com
 
So you can see the quote from the e-book and the quote from the S5
review are at odds with one another.
You see, this is what happens when human beings get involved in a
process. Errors! Cameras and computers, of course, never ever Err.
It's not just humans, it's time. I can only write about my view at the point in time I'm writing it as the time machine I've been working on in my basement isn't done yet. And I continue to grow and discover more as do (most) of the camera makers. I'm better at testing today than I was a few years ago.

However, I'll say that one of the things that has delayed my deploying a new Web site is that I've been working backwards through all the legacy material trying to bring all the statements as much up to the present as I can. The "resolves about like an 8mp" statement isn't getting changed. Some others that are comparative in the narrow time frame of the original article are.

--
Thom Hogan
author, Complete Guides to Nikon bodies (19 and counting)
http://www.bythom.com
 
My understanding of resolution is how the camera resolves light in a scene, ie the tonal depth and range from the lightest to the darkest. In my experience S5 outresolves all Nikons that I owned/tried, D70, D2x, D300, D3 and the same is true with the Canons that I owned, 20D, 1dsmk2. None can match Fuji's tonal depth and range specially with high end glass. DR is also part of a camera's resolution, no dslr that I know of today can touch Fuji yet. I haven't tried the D3x yet but the Sony A900 with the same sensor can't. Isn't photography about light? Doesn't the machine that captures the most light actually have the highest resolution? IMO 6mp, 12mp or 24mp only tell part of the story, maybe they can be used for print sizing but can't define real resolution.

david
http://www.pbase.com/ddk
 
That's wrong, period. It has more resolution than a 6mp camera. It's
visible and measurable.
There's only one problem with both your statement and mine: you can't
isolate to the pixel count. So we're both wrong.
Of course I could be wrong, but as far as I can remember, I only said the S5 would have more resolution than 6mp cameras. And if you look at the test charts, I'm right. It doesn't matter why it is better (different angles, antialiasing...).
Two big things come into play, the angle of the Bayer distribution
and the antialiasing filter on top of the sensor. You can have two
6mp cameras that, on a test chart, show different values. In essence,
that's what's happening with the S5 versus D50. Nikon has tended to
use overly aggressive antialiasing IMHO. Funny thing is, take the AA
off, and suddenly that D50 looks a lot like the S5 Pro on the test
charts when you twist them 45 degrees.
First, the D50 (and D40, D70) has an extremely weak antialiasing filter, not an aggressive one (like the S5). Second, as a photographer it doesn't matter how you theoretically could get more resolution out of your camera if practically you can't.

--
Regards,

Robert
http://www.sondek.smugmug.com
 

Keyboard shortcuts

Back
Top