V1 and J1 with 2.8/70-210 (example) and pro use with F - mount

[rayman 2]

The picture above is from a coolpix P500...(at 810mm equivalent)

Why am I showing it here is because we got some realy small cameras with realy small lenses with a fraction of the focal length and still get away with pictures
you´d need setups for that you normally couldnt carry....
Big sensors very big lenses and smaller sensors with much shorter focal length.
The V1 and the J1 will be very interesting since if the rumors are right we
can use them with the F mount lenses they will be great for i.e. birding
or taking pictures of aeroplanes....
The 2.7x crop sensor could allow in reality the same quality as a D200.....
thumb rule....
What you gain though is a few stops (about 3 stops) of speed compared to
fullframe same as if you´d use a 2.7 converter in front of your camera
while you´d be using a 500mm 5.6 on a fullframe you´d get away with
a 180mm at around F 2.5......
3 stops is the difference between using 400 iso on the V1 or J1 and using
3200 iso on a fullframe....
So you must in this case compare the quality of the J1 at 400 iso to the quality
of ie your D700 at 3200 iso......or not quite lets say iso 2000....

A popular 70-200 F 2.8 would on a J1 be equivalent to a 180 -540mm with 3 (a little less) stops better then what it would have on FF...... your J1 can be set at iso 400 while your D700 needs 3200 iso....all in a realy compact and portable unit for an additional amount of money a little over what you pay for a converter plus the F mount adapter
... and it doesnt take away much space in your bag...
Peter

 

[rayman 2]

I therefore think that a mount as big as the one on the other Nikons opens
the door for a lot of highspeed lenses and the solid attachment of older
lenses with an adapter...

I think the size of the lens mount is Ok if you look at the use in the above case....
its not just a point and shoot....
If you want or need that you´d go for a P7100 .... and even smaller lens
protrusion.....
Peter

 

[brunobarolo]

... equals a pixel density of about 73mp on FX.

A hypothetical 24mp D400 would equal a pixel density of about 56mp on FX, which is not so much less. Now add a 1.4x TC and you're already beyond the point where a 10mp 2.7x crop will get you (and probably at the very limit of even very good lenses).

I understand your reasoning very well because I thought along these lines also, but I now think a 24mp D400 (with definitely superiour AF) would be the better solution if you want to get as many pixels per duck as possible

 

[rayman 2]

... equals a pixel density of about 73mp on FX.

A hypothetical 24mp D400 would equal a pixel density of about 56mp on FX, which is not so much less. Now add a 1.4x TC and you're already beyond the point where a 10mp 2.7x crop will get you (and probably at the very limit of even very good lenses).

I understand your reasoning very well because I thought along these lines also, but I now think a 24mp D400 (with definitely superiour AF) would be the better solution if you want to get as many pixels per duck as possible

I added a picture made with a consumer lens on the P500 as an example on purpose..
That new camera with a bigger sensor and a high quality prime like a 2/200mm
should get you much better results..

impossible i dont think so..... I can sort of understand the reasoning that Nikon
has behind the 2.7 X crop....
Peter

 

[bobn2]

... equals a pixel density of about 73mp on FX.

A hypothetical 24mp D400 would equal a pixel density of about 56mp on FX, which is not so much less. Now add a 1.4x TC and you're already beyond the point where a 10mp 2.7x crop will get you (and probably at the very limit of even very good lenses).

Not at all, a very good lens will resolve way past that. You never see any tests of what they are capable of, because you always see their resolution in combination with coarse pixel sensors. Every pixel density increase in DSLR's has yielded close to the theoretical resolution increase in the centre of the frame with even just 'good' lenses, there is lots more to come.

I understand your reasoning very well because I thought along these lines also, but I now think a 24mp D400 (with definitely superiour AF) would be the better solution if you want to get as many pixels per duck as possible

It might be a better solution to tracking and focussing the duck but it certainly won't yield as high a resolution image of the duck using the same lens. As for noise, who knows, this depends very much on the efficiency of the sensor that Nikon puts in these cameras.
--
Bob

 

[bobn2]

I therefore think that a mount as big as the one on the other Nikons opens
the door for a lot of highspeed lenses and the solid attachment of older
lenses with an adapter...

I think the size of the lens mount is Ok if you look at the use in the above case....
its not just a point and shoot....
If you want or need that you´d go for a P7100 .... and even smaller lens
protrusion.....
Peter

...as I promised in the other thread.
Here is a picture of the mount, from Nikon Rumors



Measuring this image, the sensor diagonal (16mm) measures 155 pixels. The mount measures 409 pixels across the throat and 515 across the flange. This gives it a dimension of 42mm across the throat and 53mm across the flange. This establishes the mimimum diameter of the mount and the minimum 'height' of the camera, and it is simply much too big. It will restrict the portability and slimness of cameras built on this platform. Nor is there any gain coming from it. The usual argument in favour of an oversize lens mount is that it allows fast lenses, but in this case it doesn't because the rear element diameter is restricted by the 'mirrorless box' which has been included quite unnecessarily. That measures 183 pixels or 19mm. One of the contributory reasons for this wasted space is the huge and unnecessary contact array, fully 12 contacts, arranged in a 30mm diameter arc. A designed from scratch digital mount really only needs two contacts, and could be designed with just one. All the lens camera protocols can be delivered over a bidirectional signal line (one wire bus) and the lens also needs a power and ground connection. The flange itself can form the ground, so we just need DC power and signal, which can potentially share the same pin, since one is DC and the other AC. If it were me, I would integrate this one pin into the bayonet register pin, then you'd need no additional contacts at all. Additional unnecessary contacts just add potential reliability problems and waste space. In this case they waste a lot of space since they are so huge in relation to the sensor size.
So, that is why I said it is, in my opinion, a really poor mount design.

Again, were it me, I would have designed for a diameter just large enough to take a CS mount adaptor (CS mount is 25.4mm diameter, so let's say 28mm throat, 34mm flange). I would have used the suggested lens/camera protocol integrated with the bayonet locking pin, and I would have left the 28mm throat diameter clear all the way to the sensor. That would have allowed camera/lens configurations 34mm slim, nicely pocketable and it would have allowed 26mm rear elements close to the sensor. All in all, it would produce a much more comat system than Nikon will, with the potential of faster, more compact lenses.
Maybe Canon will do it. Pentax certainly didn't, the Q mount is even worse.
--
Bob

 

[rayman 2]

I therefore think that a mount as big as the one on the other Nikons opens
the door for a lot of highspeed lenses and the solid attachment of older
lenses with an adapter...

I think the size of the lens mount is Ok if you look at the use in the above case....
its not just a point and shoot....
If you want or need that you´d go for a P7100 .... and even smaller lens
protrusion.....
Peter

...as I promised in the other thread.
Here is a picture of the mount, from Nikon Rumors



Measuring this image, the sensor diagonal (16mm) measures 155 pixels. The mount measures 409 pixels across the throat and 515 across the flange. This gives it a dimension of 42mm across the throat and 53mm across the flange. This establishes the mimimum diameter of the mount and the minimum 'height' of the camera, and it is simply much too big. It will restrict the portability and slimness of cameras built on this platform. Nor is there any gain coming from it. The usual argument in favour of an oversize lens mount is that it allows fast lenses, but in this case it doesn't because the rear element diameter is restricted by the 'mirrorless box' which has been included quite unnecessarily. That measures 183 pixels or 19mm. One of the contributory reasons for this wasted space is the huge and unnecessary contact array, fully 12 contacts, arranged in a 30mm diameter arc. A designed from scratch digital mount really only needs two contacts, and could be designed with just one. All the lens camera protocols can be delivered over a bidirectional signal line (one wire bus) and the lens also needs a power and ground connection. The flange itself can form the ground, so we just need DC power and signal, which can potentially share the same pin, since one is DC and the other AC. If it were me, I would integrate this one pin into the bayonet register pin, then you'd need no additional contacts at all. Additional unnecessary contacts just add potential reliability problems and waste space. In this case they waste a lot of space since they are so huge in relation to the sensor size.
So, that is why I said it is, in my opinion, a really poor mount design.

Again, were it me, I would have designed for a diameter just large enough to take a CS mount adaptor (CS mount is 25.4mm diameter, so let's say 28mm throat, 34mm flange). I would have used the suggested lens/camera protocol integrated with the bayonet locking pin, and I would have left the 28mm throat diameter clear all the way to the sensor. That would have allowed camera/lens configurations 34mm slim, nicely pocketable and it would have allowed 26mm rear elements close to the sensor. All in all, it would produce a much more comat system than Nikon will, with the potential of faster, more compact lenses.
Maybe Canon will do it. Pentax certainly didn't, the Q mount is even worse.
--
Bob

I heard that the Cx mont has a rotary component.. the sensor is actually rotary..
could that possibly be a reasoning for the dimensions.. ?
Peter

 

[bobn2]

I therefore think that a mount as big as the one on the other Nikons opens
the door for a lot of highspeed lenses and the solid attachment of older
lenses with an adapter...

I think the size of the lens mount is Ok if you look at the use in the above case....
its not just a point and shoot....
If you want or need that you´d go for a P7100 .... and even smaller lens
protrusion.....
Peter

...as I promised in the other thread.
Here is a picture of the mount, from Nikon Rumors



Measuring this image, the sensor diagonal (16mm) measures 155 pixels. The mount measures 409 pixels across the throat and 515 across the flange. This gives it a dimension of 42mm across the throat and 53mm across the flange. This establishes the mimimum diameter of the mount and the minimum 'height' of the camera, and it is simply much too big. It will restrict the portability and slimness of cameras built on this platform. Nor is there any gain coming from it. The usual argument in favour of an oversize lens mount is that it allows fast lenses, but in this case it doesn't because the rear element diameter is restricted by the 'mirrorless box' which has been included quite unnecessarily. That measures 183 pixels or 19mm. One of the contributory reasons for this wasted space is the huge and unnecessary contact array, fully 12 contacts, arranged in a 30mm diameter arc. A designed from scratch digital mount really only needs two contacts, and could be designed with just one. All the lens camera protocols can be delivered over a bidirectional signal line (one wire bus) and the lens also needs a power and ground connection. The flange itself can form the ground, so we just need DC power and signal, which can potentially share the same pin, since one is DC and the other AC. If it were me, I would integrate this one pin into the bayonet register pin, then you'd need no additional contacts at all. Additional unnecessary contacts just add potential reliability problems and waste space. In this case they waste a lot of space since they are so huge in relation to the sensor size.
So, that is why I said it is, in my opinion, a really poor mount design.

Again, were it me, I would have designed for a diameter just large enough to take a CS mount adaptor (CS mount is 25.4mm diameter, so let's say 28mm throat, 34mm flange). I would have used the suggested lens/camera protocol integrated with the bayonet locking pin, and I would have left the 28mm throat diameter clear all the way to the sensor. That would have allowed camera/lens configurations 34mm slim, nicely pocketable and it would have allowed 26mm rear elements close to the sensor. All in all, it would produce a much more comat system than Nikon will, with the potential of faster, more compact lenses.
Maybe Canon will do it. Pentax certainly didn't, the Q mount is even worse.
--
Bob

I heard that the Cx mont has a rotary component.. the sensor is actually rotary..
could that possibly be a reasoning for the dimensions.. ?
Peter

No, look at the mount design, you can rotate the sensor within that 19mm 'mirrorless box' without requiring a 52mm diameter mount. You could do it with my 34mm diameter mount.
--
Bob

 

[brunobarolo]

... equals a pixel density of about 73mp on FX.

A hypothetical 24mp D400 would equal a pixel density of about 56mp on FX, which is not so much less. Now add a 1.4x TC and you're already beyond the point where a 10mp 2.7x crop will get you (and probably at the very limit of even very good lenses).

Not at all, a very good lens will resolve way past that. You never see any tests of what they are capable of, because you always see their resolution in combination with coarse pixel sensors. Every pixel density increase in DSLR's has yielded close to the theoretical resolution increase in the centre of the frame with even just 'good' lenses, there is lots more to come.

A 24mp DX sensor is MTF 50 diffraction limited at around f5.6. So you need a lens that is better at f4 (in the centre) than at f5.6, if you want to achieve higher resolution. That will surely work with the 200mm f2, it may even work with the 300mm and 400mm f2.8, but I'm quite sure it will not work with my 300mm f4 + 1.4x TC at f8 (where this combination resolves best)

I understand your reasoning very well because I thought along these lines also, but I now think a 24mp D400 (with definitely superiour AF) would be the better solution if you want to get as many pixels per duck as possible

It might be a better solution to tracking and focussing the duck but it certainly won't yield as high a resolution image of the duck using the same lens. As for noise, who knows, this depends very much on the efficiency of the sensor that Nikon puts in these cameras.
--
Bob

Using a 24mp DX sensor with my 300mm f4 + 1.4x TC I'd have more pixels on the duck than a 10mp 2.7x crop (without TC). So I'm not sure where the higher resolution on this Nikon mirrorless sensor might come from.

 

[bobn2]

... equals a pixel density of about 73mp on FX.

A hypothetical 24mp D400 would equal a pixel density of about 56mp on FX, which is not so much less. Now add a 1.4x TC and you're already beyond the point where a 10mp 2.7x crop will get you (and probably at the very limit of even very good lenses).

Not at all, a very good lens will resolve way past that. You never see any tests of what they are capable of, because you always see their resolution in combination with coarse pixel sensors. Every pixel density increase in DSLR's has yielded close to the theoretical resolution increase in the centre of the frame with even just 'good' lenses, there is lots more to come.

A 24mp DX sensor is MTF 50 diffraction limited at around f5.6.

No it isn't. A sensor isn't at all diffraction limited, only a lens is diffraction limited. Whan you build a system from a lens and sensor/AA filter, the MTF's of the two multiply, the higher resolution sensor produces a higher resolution at every f-number. If you don't believe me, here is some evidence from photozone.de.
On 10MP DX



On 16MP DX



Where is this diffraction limit you are talking about?

So you need a lens that is better at f4 (in the centre) than at f5.6, if you want to achieve higher resolution.

No you don't, as the (correct) theory and the Photozone tests show, you achieve higher resolution at every f-number.

That will surely work with the 200mm f2, it may even work with the 300mm and 400mm f2.8, but I'm quite sure it will not work with my 300mm f4 + 1.4x TC at f8 (where this combination resolves best)

No point discussing what it will work with because your theory is wrong.

--
Bob

 

[brunobarolo]

... equals a pixel density of about 73mp on FX.

A hypothetical 24mp D400 would equal a pixel density of about 56mp on FX, which is not so much less. Now add a 1.4x TC and you're already beyond the point where a 10mp 2.7x crop will get you (and probably at the very limit of even very good lenses).

Not at all, a very good lens will resolve way past that. You never see any tests of what they are capable of, because you always see their resolution in combination with coarse pixel sensors. Every pixel density increase in DSLR's has yielded close to the theoretical resolution increase in the centre of the frame with even just 'good' lenses, there is lots more to come.

A 24mp DX sensor is MTF 50 diffraction limited at around f5.6.

No it isn't. A sensor isn't at all diffraction limited, only a lens is diffraction limited.

Yes, I know, so to put it correctly for you, at f5.6 a lens cannot deliver more than 50% of the original contrast at the resolution that is necessary to resolve 24mp on a DX sensor. Alright?

But please never talk about a setting sun! Because that's plain wrong, it's just the earth that's turning! The sun doesn't set!

Whan you build a system from a lens and sensor/AA filter, the MTF's of the two multiply, the higher resolution sensor produces a higher resolution at every f-number. If you don't believe me, here is some evidence from photozone.de.
On 10MP DX



On 16MP DX



Where is this diffraction limit you are talking about?

So you need a lens that is better at f4 (in the centre) than at f5.6, if you want to achieve higher resolution.

No you don't, as the (correct) theory and the Photozone tests show, you achieve higher resolution at every f-number.

That will surely work with the 200mm f2, it may even work with the 300mm and 400mm f2.8, but I'm quite sure it will not work with my 300mm f4 + 1.4x TC at f8 (where this combination resolves best)

No point discussing what it will work with because your theory is wrong.

--
Bob

Diffraction isn't a brick wall. I was talking about MTF 50 explicitely. You cannot get more than 50% of the original contrast on a 24mp DX sensor at f5.6. But if you get only 25%, that doesn't mean the image is unsharp.

In any case, and back to my point and the OP, the 10mp resolution on the rumoured Nikon mirrorless sensor is not enough to resolve higher than a 10mp crop from a 24mp DX sensor with a 300mm f4 + 1.4TC.

That may be different with better lenses, and it may change with higher resolving 2.7x crop sensors, but a 10mp 2.7x crop sensor will not do it.

 

[bobn2]

... equals a pixel density of about 73mp on FX.

A hypothetical 24mp D400 would equal a pixel density of about 56mp on FX, which is not so much less. Now add a 1.4x TC and you're already beyond the point where a 10mp 2.7x crop will get you (and probably at the very limit of even very good lenses).

Not at all, a very good lens will resolve way past that. You never see any tests of what they are capable of, because you always see their resolution in combination with coarse pixel sensors. Every pixel density increase in DSLR's has yielded close to the theoretical resolution increase in the centre of the frame with even just 'good' lenses, there is lots more to come.

A 24mp DX sensor is MTF 50 diffraction limited at around f5.6.

No it isn't. A sensor isn't at all diffraction limited, only a lens is diffraction limited.

Yes, I know, so to put it correctly for you, at f5.6 a lens cannot deliver more than 50% of the original contrast at the resolution that is necessary to resolve 24mp on a DX sensor. Alright?

And so?

But please never talk about a setting sun! Because that's plain wrong, it's just the earth that's turning! The sun doesn't set!

Not a sensible metaphor. The pint is, what is being 'limited', and what is the consequence of that 'limit'. The fact is, as I said, that a 24MP sensor will extract more resolution from any lens at any aperture than will a lower pixel count sensor. "$MP on APS-C is way below the limit at which further increases in sensor density fail to deliver more resolution.

Diffraction isn't a brick wall. I was talking about MTF 50 explicitely. You cannot get more than 50% of the original contrast on a 24mp DX sensor at f5.6. But if you get only 25%, that doesn't mean the image is unsharp.

Exactly.

In any case, and back to my point and the OP, the 10mp resolution on the rumoured Nikon mirrorless sensor is not enough to resolve higher than a 10mp crop from a 24mp DX sensor with a 300mm f4 + 1.4TC.

Well, for a start, this is not the right comparison to make, the question is what will it resolve relative to a 2.7x (FF) crop from the 24MP sensor. That crop will be about 7MP, and the 10MP of the CX will resolve more simply because it puts more pixels behind the subject.

The 10MP sensor on 2.7x has the same pixel pitch as a 32MP sensor on APS-C. It will deliver more resolution, likely significantly more, even into the region that you are claiming as 'diffraction limited'. Look at f/8 on the photozone chart, into so-called 'diffraction limiting' for the 16MP sensor, yet that still delivers shedloads more resolution than the 10MP sensor. It even delivers more resolution at f/11 than the 10MP does at f/8.

That may be different with better lenses, and it may change with higher resolving 2.7x crop sensors, but a 10mp 2.7x crop sensor will not do it.

Certainly will.
--
Bob

 

[nkarasev]

I still don't get why did you post so dull and extremely over-sharpened picture (#2) If this is what 2.7 delivers (I hope it does better), I will stay miles from it.

Nik

 

[xrdbear]

I still don't get why did you post so dull and extremely over-sharpened picture (#2) If this is what 2.7 delivers (I hope it does better), I will stay miles from it.

I don't think it is over-sharpened. I suspect the dark line around the edge of the building and statues may be the result of selecting and darkening the sky. The rest of the image doesn't seem to have much in the way of sharpening halos.
--
Brian
Fine Art Print sales of the Isle of Skye at:
http://www.eyeofskye.co.uk/

 

[rayman 2]

I still don't get why did you post so dull and extremely over-sharpened picture (#2) If this is what 2.7 delivers (I hope it does better), I will stay miles from it.

I don't think it is over-sharpened. I suspect the dark line around the edge of the building and statues may be the result of selecting and darkening the sky. The rest of the image doesn't seem to have much in the way of sharpening halos.
--
Brian
Fine Art Print sales of the Isle of Skye at:
http://www.eyeofskye.co.uk/

Hey its a 300$ P500 it was just an example of what size of reproduction you´d get
I could have used anything..

A camera with about double the sensor size of the P500 must have a better image quality..

And what the quality of the picture is concerned I shot it to make a blurred background for a real shot ... i dont put up my real work for grabs...
I had a real bad experience with some of the people here especially those
that dig out my old threads where I just wanted to help you guys and they
let me look like an id*ot....
Peter

 

[BrunoS]

If the 70-200 f/2.8 is good enough to give sharp pictures wide open with a 2.7x 10MP crop sensor, would a 24MP+ DX camera be even better? You could crop to 2.7x and have all the advantages of a dSLR (e.g optical viewfinder, comfortable grip, etc). IMHO 10MP on 2.7x crop will be too much for FX lenses.

 

[bobn2]

If the 70-200 f/2.8 is good enough to give sharp pictures wide open with a 2.7x 10MP crop sensor, would a 24MP+ DX camera be even better?

No, a 24MP DX sensor cropped to 2.7x would have only 7.4MP

You could crop to 2.7x and have all the advantages of a dSLR (e.g optical viewfinder, comfortable grip, etc). IMHO 10MP on 2.7x crop will be too much for FX lenses.

It will still produce more detail than a 7.4MP equivalent crop. The sensor Nyquist is about 150 cycles per mm. Many FF lenses are producing appreciable contrast at that frequency.
--
Bob

 

[rayman 2]

Looks like the specs of the new camera and the samples are even better then expected !

New autofocus worlds fastest with phase detection (worlds first) with most af points plus the 60 frames per second and the 16000 th of a second electronic shutter speed and the imagequality near the D200 also at 400 iso are something quite interesting for action shootings...

put a 2.8 300mm lens on that beauty and you got an equivalent of an 800mm lens....
at f 2. 8 !
Peter

 

[Neil Morgan]

Agreed on paper there is lots to like.

What you wont get though is great night time performance. Certainly not D700/7000/ or D3 series worthy.
So regardless of FL conversions, high ISO is high iso and it wont be pretty.

Having said that, if its got half the potential of the D7000, then all things considered it will be decent.

ALL of these cameras are great pocket/travel light cameras and to some point are more useful than DX/FX cameras because you will be more inclined to take it anywhere because of their size and weight. My D700 kit bag with 3 lenses is 15kg and its like carrying small kid around plus tripod !
So this sort of thing has its appeal.

It will be interesting to see low light images, i am sure the day time shots will be good.

--
A Birth Certificate shows that we were born.
A Death Certificate shows that we died.
Pictures show that we lived!

http://www.flickr.com/photos/knumbnutz/

 

[xrdbear]

I still don't get why did you post so dull and extremely over-sharpened picture (#2) If this is what 2.7 delivers (I hope it does better), I will stay miles from it.

I don't think it is over-sharpened. I suspect the dark line around the edge of the building and statues may be the result of selecting and darkening the sky. The rest of the image doesn't seem to have much in the way of sharpening halos.
--
Brian
Fine Art Print sales of the Isle of Skye at:
http://www.eyeofskye.co.uk/

Hey its a 300$ P500 it was just an example of what size of reproduction you´d get
I could have used anything..

A camera with about double the sensor size of the P500 must have a better image quality..

And what the quality of the picture is concerned I shot it to make a blurred background for a real shot ... i dont put up my real work for grabs...
I had a real bad experience with some of the people here especially those
that dig out my old threads where I just wanted to help you guys and they
let me look like an id*ot....
Peter

Actually Peter what you said above is all fairly obvious. I thought I was posting something in your favour. Your posts are greatly appreciated by me.

--
Brian
Fine Art Print sales of the Isle of Skye at:
http://www.eyeofskye.co.uk/