Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
Have you look at the Kodak DCS14n comparison with Canon 1Ds in thisLook, unless I am mistaken (and I may very well be), there is no
lens that resolves a higher resolution than the 1Ds sensor is
capable of. Many photographers complain about the 1Ds "seeing"
limitations in their lenses that film masked. You have to
understand, there are limitations in glass!! Companies have worked
for many years to enhance lens design, but there are still simple
physical limitations.
Many people have pointed out that the Nikon DX lenses more or less
show Nikon's commitment to developing lenses for the current sensor
size they have rather then going Full Frame, as Canon seems to have
committed to... Other then making DX lenses incompatible with
Nikon's film bodies (because of vignetting on the FF size), I am
ignorant of the technical issues, namely:
WHY is FullFrame inherently better then the Nikon way of making
lenses to accomodate sensor size?? Either way, you have a full
range of lenses that give you the desired perspective (WA or
Telephoto) for the pictures you want to take.
I'd really appreciate it if people with more technical know-how
could clear this up for less informed shutterbugs like myself.
Thanks!!
Greg
So again, why do we need the FF? Simply because of the "old" 35mm
format? This seems like a crazy reason, since as time goes on, it
would mean keeping HUGE glass where it wasn't needed.
Right now I think we're wasting a large portion of the glass that's
getting unused in the image. Not to mention the weight penalties
in the glass.
Again - maybe I'm missing something from the "optical" side of things.
--What you are asking is a very good question. The answers may not
be as cut and dry as many here have posted. I think that most
Canonites feel that larger areas equal lesser noise. i.e. 8
million imaginary pixels on an imagingary area of 100 square mm
will be less than 8 million imaginary pixels on an imaginary wafer
of 200 square mm.
The other thing is that a full size digital frame will be 100%
seamless and compatible with film body lenses of the same mount.
I think that what Nikon has asked in its new offering (as has
Olympus) is: how many pixels are enough? It is fairly obvious by
Nikon's new offering that they do not necessarilly feel that more
is better. They also seem to be going ahead and doing what Canon
did when they adopted the EF mount - morphing it for something
newer and better for their current product line.
I personally favor a full frame approach and hope that Canon's
answer will be to produce an affordable FF camera. I believe that
this is very possible. The question is when, and will Canon cave
before it becomes to fruition?
-JM
--Many people have pointed out that the Nikon DX lenses more or less
show Nikon's commitment to developing lenses for the current sensor
size they have rather then going Full Frame, as Canon seems to have
committed to... Other then making DX lenses incompatible with
Nikon's film bodies (because of vignetting on the FF size), I am
ignorant of the technical issues, namely:
WHY is FullFrame inherently better then the Nikon way of making
lenses to accomodate sensor size?? Either way, you have a full
range of lenses that give you the desired perspective (WA or
Telephoto) for the pictures you want to take.
I'd really appreciate it if people with more technical know-how
could clear this up for less informed shutterbugs like myself.
Thanks!!
Greg
http://www.MasterworkPhotography.com
http://www.pbase.com/stevegrillo , Equipment on profile page
I haven't seen any firm evidence of that, and in fact I think it's
incorrect. Remember that the 1Ds is 11MP and the 10D is 6.3MP. If
both images have the same amount of "noisy pixels", the 1Ds image
will look much cleaner. So it depends very much how you define
"noise", but if you define it as how the image looks to the human
eye, the 1Ds produces cleaner images.
My reference was Phil's reviewsI haven't seen any firm evidence of that, and in fact I think it's
incorrect.
"t's important to remember that the 1Ds produces larger images than the D60, this means that at the same print size noise will be less visible for a 1Ds image than a D60."My reference was Phil's reviews
http://www.dpreview.com/reviews/canoneos1ds/page14.asp
the D60, this means that at the same print size noise will be lessMy reference was Phil's review> "t's important to remember that the 1Ds produces larger images than
visible for a 1Ds image than a D60."
I think the argument is larger pixels of the SAME TECHNOLOGY always produce less noise. We saw in the jump from the D30 to the D60 that if you boost technology, you will not see this effect. The only negative is perhaps there is a reason there is noise inferior technology in the 1Ds (eg. micro lens issues with FF) but at this point that's speculation.This is a good point and indeed true. I was mainly addressing the
commonly raised argment that larger pixels will always produce less
noise.
The advantage of the 1Ds is it's sensor size, and that ability to capture more light than a 1.3x, 1.5x, or 1.6x sensor. It's this ability that gives it the best noise performance (at least up to 800). The number of pixels really isn't the thing that saves it, as if it were big 1D sized pixels it'd still have the best noise.I shoudl have mentioned that in my first reply as I did in
another reply to another FF vs DX thread a few days ago. This
advantage of noise in the 1Ds being less visible is an advantage of
pixel count and thus does not give an advantatge to one format over
the other for the same amount of per pixel S/N ratio. We are still
left with the senstivity issue which is a function of the pixel
level S/N ratio. The rest of my argument also still applies. We do
not have any solid info which format has more potential yet. We
could end up better off with different formats for different
applications.
A phenomenon that is easily demonstrated when comparing DSLR sensors to point 'n' shoot sensors, or between point 'n' shoot sensors of different sizes..... I was mainly addressing the
commonly raised argment that larger pixels will always produce less
noise.
But larger format leads to an advantage in pixel count! If you have identical technologies, i.e. pixels of the same size and level of noise, and you apply that technology to a small and large sensor, then the large sensor will have more pixels, and thus the per-pixel noise will be less noticeable.This
advantage of noise in the 1Ds being less visible is an advantage of
pixel count and thus does not give an advantatge to one format over
the other for the same amount of per pixel S/N ratio.
Answer this question. If in a few years the successor to the 10D (30D?) still has a small sensor, with say 10-12MP, ISO 6400 and has demonstrably superior image and noise quality than the current 1Ds, how many people here would care that it's not a FF sensor. Just imagine those images would be of superb quality and how much more could you want, even if the 1Ds' successor is also much better.The real winner in this battle will be the person who can make a
smaller sensor with low noise and plenty of pixels. With
advancement in technology comes minaturization. Already it's no big
deal to pack 6 MP in a 1.6X sensor. And the new Nikon has ISO
sensativities up to 6400 or something like that so already they've
figured out how to beat the noise.
Won't be long before they have 12 MP on these smaller sensors and
running up to ISO speeds unheard of in the film world. The smaller
sensor will undeniably be cheaper to produce and thus the FF won't
be able to compete. FF will become the medium format of the
future, the other 99.99% of us will use smaller sensors.
Very true. And that is exactly what Canon gives us with their current line-up of three very different D-SLRs (new arrivals later this year notwithstanding).We
could end up better off with different formats for different
applications.
I have to agree with this. People are always saying stupid things about images shot with the 1Ds. The high resolution hides noise very well, especially if you're going to print the photo at 10x8 on a Frontier printer - noise? what noise?! People also complain about lenses (16-35L) being soft on the 1Ds like using a D60/10D would be better, I think not!All my tests indicate that the 1Ds is better than most people give
it credit for. I believe it's the best camera available up to and
including ISO 800. It's surely better than the 1D after you either
print the image or downsample it to the same 4MP resolution. I did
a test yesterday (not nearly as controlled... different scene) that
said I can't see any improvements of a D1h over a 1ds at ISO 800
either.
Here you are talking about pixel level noise not percieved noise and this puts my 10D and 1Ds example back to action. These cameras probably have similar technologies and at the pixel level the 10D is cleaner at every ISO despite the smaller pixel and more sensitive (again related to the better noise performance). The reason is unkown, probably it is due to some compromises you have to make when doing a FF design. My point was (which btw is just a different version of what you said) we do not know which format technology development will give the upper hand, if any. That is, will it be easier to make a more sensitive FF sensor as you increase its pixels or put more pixels on a DX format sensor and get comparable (to FF) noise, senstivity and dynamic range from it? Both routes are going to progress. We do not which will progress faster or which will hit a barrier first. Right now I see no clear winners. I am waiting for the next Nikon high res DSLR to get a feel of what can be accomplished.I think the argument is larger pixels of the SAME TECHNOLOGY always
produce less noise. We saw in the jump from the D30 to the D60
that if you boost technology, you will not see this effect. The
only negative is perhaps there is a reason there is noise inferior
technology in the 1Ds (eg. micro lens issues with FF) but at this
point that's speculation.
Again you are talking about pixel level noise and the smaller sensor in the 10D is the winner here. It is the pixel count that gives the 1Ds the larger picture output and hence the less percieved noise.. It is just like 35mm film and large format. Both have the same noise per unit area (grain) but the larger film does not show as much noise because it does not have to be enlarged as much when enlarging. In the case of sensors the larger pixel count also gets compressed more when making a small picture hence suppressing the noise better than the small pixel count.The advantage of the 1Ds is it's sensor size, and that ability to
capture more light than a 1.3x, 1.5x, or 1.6x sensor. It's this
ability that gives it the best noise performance (at least up to
800). The number of pixels really isn't the thing that saves it,
as if it were big 1D sized pixels it'd still have the best noise.
I was definitely not talking about the ultra tiny pixels in the p&S cameras..A phenomenon that is easily demonstrated when comparing DSLR
sensors to point 'n' shoot sensors, or between point 'n' shoot
sensors of different sizes.
I doubt that a 6 months difference can make all that difference in high ISO sensitivity ... very unlikely. The info I have been reading in this forumn indicate that the 1Ds is also produced in Canon's fab.We can call into question whether the same phenomenon is
demonstrable with the 10D vs the 1Ds, but I guess it's at least
fair to point out that the 10D is about 6 months newer than the
1Ds, and I think it was the first to be produced on Canon's own fab.
I was talking about per pixel noise and I gave the example of the 10D and 1Ds, which I guess have very comparable noise reduction technology, to show that the issue of pixel size, format and noise is not as simple as some people like to make it.But larger format leads to an advantage in pixel count! If you have
identical technologies, i.e. pixels of the same size and level of
noise, and you apply that technology to a small and large sensor,
then the large sensor will have more pixels, and thus the per-pixel
noise will be less noticeable.
Very true. And that is exactly what Canon gives us with theirWe
could end up better off with different formats for different
applications.
current line-up of three very different D-SLRs (new arrivals later
this year notwithstanding).
Greetings
Stefan
With the new D2H, you can now have Matrix/Centre/Spot metering with input of the lens aperture/focal length info. Will Nikon retrofit this to the older DSLRs? Dunno, but would be great if it could be done (technically possible).I realize that you can put a manual focus Nikkor from
the mid 70's on a digital Nikon slr, but believe me your options
are limited (probably need stop down metering, no display of data
in viewfinder, etc.)
Sorry, beg to differ - Nikon's F-mount today does everything that the EF mount can do - AFS(USM), VR(IS), Distance readouts, Aperture readouts - while maintaining backwards compatibility (albeit with caveats in certain cases) with virtually ALL previous Nikkor lenses out there - manual AIS, non-AIS, AF, AF-D, AF-G, AFS, etc.Canon invested in the future with the EOS system and they're
actually the more compatible system now. I respect Nikon's loyalty
to their user base, but it's hurting them now.
If it was truly a problem, then it would have also showed up in 35mm film as the exact same conditions apply for the light striking silver-halide film. In any case, microlenses on the sensor (1Ds) can compensate for the off-angle light.had to do with a bit of the above how the light hits the sensor and
how the sensor needed more straight on view or something ?? could
not remember exactly
but would cause soem problems it was doable but with quality loss ???
so maybe they didnt want to chance people putting on cheap lenses
and then bitching that the camera sucks ??
you hear some of that now with the 1Ds
So again, why do we need the FF? Simply because of the "old" 35mm
format? This seems like a crazy reason, since as time goes on, it
would mean keeping HUGE glass where it wasn't needed.
Right now I think we're wasting a large portion of the glass that's
getting unused in the image. Not to mention the weight penalties
in the glass.
Again - maybe I'm missing something from the "optical" side of things.
--What you are asking is a very good question. The answers may not
be as cut and dry as many here have posted. I think that most
Canonites feel that larger areas equal lesser noise. i.e. 8
million imaginary pixels on an imagingary area of 100 square mm
will be less than 8 million imaginary pixels on an imaginary wafer
of 200 square mm.
The other thing is that a full size digital frame will be 100%
seamless and compatible with film body lenses of the same mount.
I think that what Nikon has asked in its new offering (as has
Olympus) is: how many pixels are enough? It is fairly obvious by
Nikon's new offering that they do not necessarilly feel that more
is better. They also seem to be going ahead and doing what Canon
did when they adopted the EF mount - morphing it for something
newer and better for their current product line.
I personally favor a full frame approach and hope that Canon's
answer will be to produce an affordable FF camera. I believe that
this is very possible. The question is when, and will Canon cave
before it becomes to fruition?
-JM
--Many people have pointed out that the Nikon DX lenses more or less
show Nikon's commitment to developing lenses for the current sensor
size they have rather then going Full Frame, as Canon seems to have
committed to... Other then making DX lenses incompatible with
Nikon's film bodies (because of vignetting on the FF size), I am
ignorant of the technical issues, namely:
WHY is FullFrame inherently better then the Nikon way of making
lenses to accomodate sensor size?? Either way, you have a full
range of lenses that give you the desired perspective (WA or
Telephoto) for the pictures you want to take.
I'd really appreciate it if people with more technical know-how
could clear this up for less informed shutterbugs like myself.
Thanks!!
Greg
http://www.MasterworkPhotography.com
http://www.pbase.com/stevegrillo , Equipment on profile page
What I'm saying is that larger sensors can provide lower noise via one of two paths. One is by having the same number of pixels, but larger pixels, which is generally considered to provide lower per-pixel noise. But if you don't accept as a given that this is true, then it should be simple enough to consider the second approach, which is to provide pixels of the same size on a larger sensor, and thus provide the same per-pixel noise, but more pixels, thus lessening the impact of the same per-pixel noise.What we need to do is get Canon to make us a 1.6X sensor with the
10D technology and with 11.5 MP and compare its senstivity and
noise performace to the 1DS then we can tell.
I think there's sufficient knowledge to state that larger sensors will provide lower noise, keeping all other factors the sameYou might not agree with everything I said but I guess you will
agree with my main point and that is we just are too short on
knowledge to even speculate.
