Velvia and DSLR test
MXP
the same
frame in the viewfinder. Both cameras mounted on a tripod. Not same lens and
two
different brands of cameras. The film crop is about 8x9mm of film. Scanned
at 4000 dpi.
DSLR image is from RAW format converted to best JPEG.
Remember this is just for fun.......
DSLR:
http://www.pbase.com/mxp/image/52209028
Velvia:
http://www.pbase.com/mxp/image/52208509
This is also a test of PBASE. Have never tried this before.
Max
Kinon O'cann
it this way:
1. Shoot the same scene using lenses of roughly equivilant quality at the
same angle of view. This is critical.
2. Present the full-frame results from both cameras.
3. Show us a crop of the same area of the image, showing a 100% crop of the
larger of the two images.
"MXP" <max_...@post11.tele.dk> wrote in message
news:4377b226$0$182$edfa...@dread11.news.tele.dk...
MXP
"MXP" <max_...@post11.tele.dk> skrev i en meddelelse
news:4377b226$0$182$edfa...@dread11.news.tele.dk...
MXP
You need something like a 24MP DSLR to have about same size.
Both cameras was mounted just beside each other. I have added the
24x36 full frame very compressed. Don't know if I still have the
DSLR full frame image.
DSLR was a 28-70/2.8 pro zoom. Film camera a 50/1.8 lens. The
zoom setting was set so it approx gave the same frame as the 50mm
lens. Don't think the 28-70 is a limiting factor. Should with ease handle
a 8MP sensor?
Max
"Kinon O'cann" <Yes.it's.me.Bowser> skrev i en meddelelse
news:GJGdnchJ-uJ...@comcast.com...
MXP
Skotland......
Velvia scanned from AN glass frame so a little bit of film is hidden in the
frame.
Max
"MXP" <max_...@post11.tele.dk> skrev i en meddelelse
news:4377b580$0$174$edfa...@dread11.news.tele.dk...
Colin D
Why is the dslr crop only half the size of the film crop? makes it hard
to even see the writing at that size. What about a crop 100%, and with
the dslr RAW converted to TIF? What format was the scan of velvia in -
tif or jpeg?
Colin D.
MarkË›
Those crops are SO tiny that it's tough to judge anything.
Try it with a much larger crop, or perhaps even an inflated crop...
MXP
The 4000 dpi scan gives about 24MP .....so therefore larger in 100% viewing.
Text can not be written in the original DSLR raw file (I don't have it) but
I asked my
friend if he could read it. The original film scan was in .tif
We have tested more than this image and if we talk pure resolution 8MP can't
beat
24x36 mm Velvia. That is my own conclusion. I know some claims that the 3MP?
Nikon D1 could show more details that Velvia 50 .....but that sounds
nonsense to me.
Max
"Colin D" <Col...@killspam.127.0.0.1> skrev i en meddelelse
news:4377C47A...@killspam.127.0.0.1...
MXP
"MarkË›" <mjmorgan(lowest even number here)@cox..net> skrev i en meddelelse
news:ewPdf.73$K26.72@fed1read02...
no more information in the DSLR crop than showed.....it is 100% viewing of
this crop.
I have seen many stupid tests where people want to show that a 5 or 6 MP
sensor
outperforms film (Velvia). I have never understood this.....and still does
not understand it.
Max
Canon F1 via PhotoKB.com
>I have seen many stupid tests where people want to show that a 5 or 6 MP
>sensor
>outperforms film (Velvia). I have never understood this.....and still does
>not understand it.
>
>Max
You will never win this arguement with digital shooters. Don't even waste
your typing fingers.
I am with you however. To me, film will out do digital in most cases. I
have a shot of the Teton Mtns. in Western Wyoming that I shot back in August
of this year. I shot the photo from the top of a ski slope on the Idaho side.
It is an uncommon photo. I shot the photo with ordinary Fuji 100 ASA print
film (not slide) out of my Canon F1 through an FL 28mm on f16. I had the
negative scanned and I have a 20 x 30 full frame print matted/framed and
hanging in my living room. There is very little grain in the clouds or blue
sky even this big and with print film. You got to be within inches to see
any grain at all in the photo and then it is mostly in the clouds. You move
back even a foot from the photo and you can't see it anymore.
I have a sea sunset shot with a hanging fisherman's net hanging to dry I
shot in 1990 in Hong Kong I shot with ordinary Fijuchrome 100 ASA film
through an old Canon AE 1 and a 28-105 Knock off lens. With a high res scan,
I have gone to 30 x 40 and there is no noticable grain. I plan on a 40 x 60
of this shot customed framed and matted for the next photo show I go to. It
turns heads and sells at any size, 16 x 24 being the most popular.
Show all the photo's you like but you can't convince digital shooters they
might be on a bad path.
F1
--
Message posted via http://www.photokb.com
MarkË›
Bad path?
Perhaps that characterization is what you find argued, and rightly so.
I don't have a problem identifying where film performs in comparison to
digital.
There are some aspects of each that "out-do" the other. The benefits of
film exist independantly of the benefits of digital. -To characterize
digital as a "bad path" is pretty silly, though.
There is no need for such pronouncements one way or the other, and certainly
not in terms of "good" or "bad."
Each form of photography has areas where it shines, and areas where it
reveals its limitations.
I would suggest that the kind of blanket statements you are making in your
post only fan the flames of a very unnecessary fire.
Canon F1 via PhotoKB.com
>>> I have seen many stupid tests where people want to show that a 5 or
>>> 6 MP sensor
>> Show all the photo's you like but you can't convince digital shooters
>> they might be on a bad path.
>
>Bad path?
>post only fan the flames of a very unnecessary fire.
OK, bad wording. I am sorry. I do beleive that digital photography as we
know it know will likely change into new technologies and all will try to
convert to the newest and latest. I just hope that film is still around.
Sad if my old Canon F1 out lives film.
de...@boisclair.freeserve.co.uk
can view in a largish size)looks warmer - and far more pleasing - in
the digital version
Denis Boisclair
Cheshire, UK
MXP
That kind of work can be done with the Velvia image also. But the Velvia
100F is a bit on the cool side.
Max
<de...@boisclair.freeserve.co.uk> skrev i en meddelelse
news:1131980796....@g43g2000cwa.googlegroups.com...
Kinon O'cann
the digital image to match in size. The DSLR image will appear chunkier with
less resolution, but that's because it is...
Either way, make the crops the same size; it's easier. A little bigger
wouldn't hurt, either.
"MXP" <max_...@post11.tele.dk> wrote in message
news:4377b873$0$179$edfa...@dread11.news.tele.dk...
Scott W
> OK, but blow up the film crop to whatever size you want, and then blow up
> the digital image to match in size. The DSLR image will appear chunkier with
> less resolution, but that's because it is...
>
> Either way, make the crops the same size; it's easier. A little bigger
> wouldn't hurt, either.
I would like to see larger as well, maybe a crop size of around 400 x
400 pixels ?
BTW thanks for posting the photos.
Scott
MXP
I have added a larger crop and upsampled the DSLR image to about same size
as
the film crop.
http://www.pbase.com/mxp/gallery/test
Max
"Scott W" <bip...@hotmail.com> skrev i en meddelelse
news:1132010590.7...@g14g2000cwa.googlegroups.com...
Scott W
> OK.
>
> I have added a larger crop and upsampled the DSLR image to about same size
> as
> the film crop.
>
> http://www.pbase.com/mxp/gallery/test
>
> Max
>
in the past.
Scott
Paul Furman
> OK.
>
> I have added a larger crop and upsampled the DSLR image to about same size
> as
> the film crop.
>
> http://www.pbase.com/mxp/gallery/test
>
> Max
Wow those are very different. The velvia is much more contrasty and
lacks reflections on the windows as if a polarizer was used. The
background trees are much darker and less saturated. The histograms in
PS are really different too: the velvia histogram is 'taller' though I'm
not sure what that means <g>. No doubt there is more detail.
clif
I shoot 8x10 film with ultimate resolution needed. While it takes 3
times as long (sometimes longer) to shoot, there is at least a hundred
fold difference in resolution. I would hate to try to store that shot,
even in the 64bit unix world. I use a modified zone system for
exposure and find that film color depth is much better too. Still,
getting the shot to digital is quite a challenge. The greatest
limitation I seem to run in is in scanning. I frequently find that
scanners cannot provide the color depth of the camera. Perhaps it is
my lack of digital processing knowledge. What scanner was used for
this demo?
MXP
When used correct the 9000ED can pull many details out of the film. It of
course
also requires that the input material is good (tripod, correct focus,
correct exposure,
good camera (flat film and precise mirror box), good lens, good
development).
The scan could probably be a bit better if I used a Gepe mount with only AN
glass
on the top so the emulsion side is free from glass. If you look in the high
contrast
transitions there is a bit of red/blue CA (about 1 pixel row). It ammost
disappears if I
do that and also running the scanner in the fine CCD mode where only 1 CCD
row
are used (scanner has a CCD with 3 rows). Then scan time takes 3 times
longer.
There are many options when scanning. In this scan I used x8 multi sample. I
can
chose from 1x, 4x, 8x and 16x multi sample. This is to outcompensate noise
from
the CCD. Then you get dark areas very clean and free from noise.
Max
MXP
"clif" <cl...@posey.org> skrev i en meddelelse
news:1132026363.7...@o13g2000cwo.googlegroups.com...
The scanner I use now is a Nikon 9000ED. I think you get good price/quality
when buying this scanner. Next step is Imacon which is a big step.....very
expensive.
They are good for 4x5" ......but don't know if they make 8x10" scanners.
The largest format I use is 6x6. Then I don't count vintage 6x9 cameraes I
have.
The scanner can do 6x9 so I am looking for a good quality 6x9 camera. E.g. a
Linhof with 6x9 Rollex could be fun to play with.
When I scan 6x6 i 16 bit color mode (4000 dpi) the file size is 400 - 500 MB
pr. scan.
Then Photoshop adjustments takes some time also.......need a lot of RAM.
Max
MXP
"Paul Furman" <paul-@-edgehill.net> skrev i en meddelelse
news:FYCdnf1LdoJ...@speakeasy.net...
I used a polarizer set to max. polarizing effect. I am not sure about my
friends DSLR. I know he has a filter...but if it was used and how it was
set I don't know.
Paul Furman
> "Paul Furman" <paul-@-edgehill.net> skrev i en meddelelse
> news:FYCdnf1LdoJ...@speakeasy.net...
>
>>MXP wrote:
>>
>>>OK.
>>>
>>>I have added a larger crop and upsampled the DSLR image to about same
>>>size as
>>>the film crop.
>>>
>>>http://www.pbase.com/mxp/gallery/test
>>>
>>>Max
>>
>>Wow those are very different. The velvia is much more contrasty and lacks
>>reflections on the windows as if a polarizer was used. The background
>>trees are much darker and less saturated. The histograms in PS are really
>>different too: the velvia histogram is 'taller' though I'm not sure what
>>that means <g>.
Here's the histograms for those:
http://www.edgehill.net/1/?SC=go.php&DIR=Misc/photography/histogram
> No doubt there is more detail.
>
>
> I used a polarizer set to max. polarizing effect. I am not sure about my
> friends DSLR. I know he has a filter...but if it was used and how it was
> set I don't know.
I don't think he used a polarizer. I'm not sure how much difference that
makes. Another thing I notice is your velvia shot shows highlights in
the hand rails in the foreground which are just flat gray on the DSLR
image. Overall much more contrasty.
MXP
OK.
To make more justice to the DSLR image I have made level adjustment and
USM in Photoshop and uploaded the image. The JPEG converted RAW
was probably direct from the camera and you always do some adjustments
in Photoshop or another program.
MXP
"MXP" <max_...@post11.tele.dk> skrev i en meddelelse
news:437a23d1$0$38722$edfa...@dread12.news.tele.dk...
Velvia
still has more details. I could probably make the Velvia better by
rescanning
it and be a little more careful with levels and USM in Photoshop. I have
blown
out the highlights what you also see in the histograms.
For this image I did not save the original "raw" scan which would have been
nice. When we get 18-20 MP DSLR's I probably will get one but still use
film. Waiting for the new Fuji S4 Pro.......
Paul Furman
I also played with levels, curves & sharpening without any huge
improvement. I'm still impressed with the Velvia. It's possible the
polarizer had a big impact though if this was a little competition
between you & your buddy, you clearly won <g>.
HvdV
> It is 100% crops. The 24x36 scanned at 4000 DPI will appear larger then.
> You need something like a 24MP DSLR to have about same size.
>
> Both cameras was mounted just beside each other. I have added the
> 24x36 full frame very compressed. Don't know if I still have the
> DSLR full frame image.
>
> DSLR was a 28-70/2.8 pro zoom. Film camera a 50/1.8 lens. The
> zoom setting was set so it approx gave the same frame as the 50mm
> lens. Don't think the 28-70 is a limiting factor. Should with ease handle
> a 8MP sensor?
scans with a Minolta 5400-II of velvia 50 do show grain. Did you use use an
anti grain filter, and if so which?
Thanks for posting this, Hans
MXP
"HvdV" <noh...@svi.nl> skrev i en meddelelse
news:57490$437a3685$3e3aaa83$20...@news.versatel.net...
My scanner has GEM but I did not use it to blur out the grains as my
experience is that it also affects the details. GEM is pure software I think
and the software can't tell the difference between very fine details and
grains.
I only used digital ICE set to normal to remove the dust. I have checked
that
it does not affect the details.
There is a bit of visible grains in the sky but I think the AN glass from
the
scanning side (emulsion) has worked like a soft filter. The slide was
mounted
in a Gepe AN glass frame. If you look at the Gigabit scans you will see
grains
especially in the sky. For these scans I mounted the film in a Gepe frame
only
with AN glass on the glossy side (opposite the emulsion side).
I will try to rescan the Velvia using the same type of Gepe mount.
On the paper the Velvia 100F should be a bit more fine grained than the
Velvia 50.
Max
MXP
I think this is the best I can get out of the Velvia image with my 9000ED.
Re-scanned crop has been uploaded.
"Paul Furman" <paul-@-edgehill.net> skrev i en meddelelse
news:y92dnb8s-cwEtufe...@speakeasy.net...
HvdV
>
> My scanner has GEM but I did not use it to blur out the grains as my
> experience is that it also affects the details. GEM is pure software I think
> and the software can't tell the difference between very fine details and
> grains.
> I only used digital ICE set to normal to remove the dust. I have checked
> that
> it does not affect the details.
I thought you were using something like an adaptive gaussian or median filter.
> There is a bit of visible grains in the sky but I think the AN glass from
> the
> scanning side (emulsion) has worked like a soft filter. The slide was
> mounted
> in a Gepe AN glass frame. If you look at the Gigabit scans you will see
> grains
> especially in the sky. For these scans I mounted the film in a Gepe frame
> only
> with AN glass on the glossy side (opposite the emulsion side).
Were these the gigabit scans from the lighthouse?
> I will try to rescan the Velvia using the same type of Gepe mount.
ok!
> On the paper the Velvia 100F should be a bit more fine grained than the
> Velvia 50.
Yes, but isn't the 100F 'newer and better'?
-- Hans
MXP
> MXP wrote:
>>
>> My scanner has GEM but I did not use it to blur out the grains as my
>> experience is that it also affects the details. GEM is pure software I
>> think
>> and the software can't tell the difference between very fine details and
>> grains.
> Indeed, some grains are larger than some details...
>> I only used digital ICE set to normal to remove the dust. I have checked
>> that
>> it does not affect the details.
> I thought I could see some more noise or grain near contrasty boundaries,
> so I thought you were using something like an adaptive gaussian or median
> filter.
>> There is a bit of visible grains in the sky but I think the AN glass from
>> the
>> scanning side (emulsion) has worked like a soft filter. The slide was
>> mounted
>
>> in a Gepe AN glass frame. If you look at the Gigabit scans you will see
>> grains
>> especially in the sky. For these scans I mounted the film in a Gepe frame
>> only
>> with AN glass on the glossy side (opposite the emulsion side).
> Were these the gigabit scans from the lighthouse?
Yes, the BW scans I have uploaded is from Gigabitfilm.
>> I will try to rescan the Velvia using the same type of Gepe mount.
> ok!
I have now uploaded a re-scan of the crop. Think you can see a bit more
grains here.
>> On the paper the Velvia 100F should be a bit more fine grained than the
>> Velvia 50.
> Yes, but isn't the 100F 'newer and better'?
>
> -- Hans
Velvia 100F is newer. But many still likes the old Velvia 50. The yet newer
Velvia 100 (no F) should be the best of them both. Fine grained like 100F
and
colors like Velvia 50. I look forward to try the new Velvia 100. It is then
assumed
that the old Velvia 50 will go out of production.
Max
MASL
MXP Wrote:
>
> Remember this is just for fun.......
>
Well thanks, I had fun. Glad to see that 35mm holds it's own vs the
1DII. I can see why you were out there with your cameras too, pleasant
scene. Thanks for the effort,
-Mark
--
MASL