New gig, four thirds micro

[Ben Rubinstein]

With no mirror and a small shutter for that small sensor it's certainly going to be quiet!

Of course I assume it will have a hot shoe so no problem using accessory viewfinders. Hope they include a hyperfocal focus setting!

What I really hope is that they design it as a rangefinder/p&s and not as a DSLR, hyperfocal setting and a decent auto iso is far more important to me than 5fps!

Jono, will I really be annoying you if I say that my father will be going on a lecture tour of the states after September so methinks I'll have one in hand first? :ROTFL::ROTFL::ROTFL:

 

[Lili]

Droooooooooooool!!!!!
This could be so sweet; raising the bar on IQ, Versatility and Size equation!

As to killing Bridgecam, perhaps, however the new cams will still be as vulnerable to dust when one changes lenses; which is just one reason for the Bridges...
Bridges lately also have not been that compact, they are seen as an all-in-one-solution not so much as compact cameras.
I suspect the EVIL and the Bridge will appeal to different needs.

 

[Lars]

I'd love to see a nice rangefinder using this lens standard - just give me an optical viewfinder as well. Sensor quality on 4/3 format is of course critical for demanding use, but I understand it is decent today and will get better over time. This is an interesting development.

 

[ecliffordsmith]

Hi All,

This is really exciting news, the possibility to have a very compact system with significantly better image quality and fast primes is most compelling. A new market and a new alternative.

I wonder when we will see the fruits of this?

 

[Ocean]

It will be interesting to see the lens line-up from Olympus and Leica for this new format. I don’t know how they are going to design the wide-angle lenses since the new format has the same sensor dimensions as the original 4/3 format but the flange distance is cut by 50%. The original 4/3 is about having light path arriving at a somewhat perpendicular angle to the sensor surface. How would that be achieved with a shorter flange back distance? Unless the sensor technology has advanced to the point where new sensors can accept light path coming at a steep angle.

Kind regards,

 

[jonoslack]

It will be interesting to see the lens line-up from Olympus and Leica for this new format. I don’t know how they are going to design the wide-angle lenses since the new format has the same sensor dimensions as the original 4/3 format but the flange distance is cut by 50%. The original 4/3 is about having light path arriving at a somewhat perpendicular angle to the sensor surface. How would that be achieved with a shorter flange back distance? Unless the sensor technology has advanced to the point where new sensors can accept light path coming at a steep angle.

Kind regards,

Hi there

Well parallel is parallel! So if all the stuff about telecentricity is to be believed then it doesn't matter about the flange distance

 

[Lars]

Flange distance will matter if a lens design is used where the rear element is closer to the image plane thus increasing the largest angle of the light path hitting the image surface. Leica's solution for its M8 (which was necessary to keep compatibility with older non-retrofocus lens designs) was to add a microprism layer in front of the sensor. Micro-4/3 lenses will be designed to avoid such problems - retrofocus designs will be used for wideangle lenses rather than more symmetrical lenses like classic wideangle lenses.

For a good example of an extreme retrofocus design, se Schneider Digitar 2.8/28 - even though the focal length is 28 mm, the flange focal distance is 70 mm. http://www.schneideroptics.com/ecommerce/CatalogItemDetail.aspx?CID=1325&IID=1920

 

[jonoslack]

Flange distance will matter if a lens design is used where the rear element is closer to the image plane thus increasing the largest angle of the light path hitting the image surface. Leica's solution for its M8 (which was necessary to keep compatibility with older non-retrofocus lens designs) was to add a microprism layer in front of the sensor. Micro-4/3 lenses will be designed to avoid such problems - retrofocus designs will be used for wideangle lenses rather than more symmetrical lenses like classic wideangle lenses.

For a good example of an extreme retrofocus design, se Schneider Digitar 2.8/28 - even though the focal length is 28 mm, the flange focal distance is 70 mm. http://www.schneideroptics.com/ecommerce/CatalogItemDetail.aspx?CID=1325&IID=1920

Hi Lars
I obviously didn't get the point across - the whole philosophy of 4/3 is to have true telecentricity (it is actually why Olympus have made nothing faster than f2 for 4/3 - those who HAVE are almost certainly breaking the telecentriclty rules).

The point is that the light out of the rear element is parallel - and, as I said, parallel IS parallel - however close to the rear element you put the sensor it's STILL going to be parallel. It also explains why current Zuiko designs for 4/3 are such good corner and edge performers (and they really are good - MUCH better than anything that Nikon has to offer (at least for the zooms))

As far as I can see Olympus is really the only company who can just do this (i.e. change the distance from the rear element to the sensor). I would definitely envisage difficulties with lenses made by others in the 4/3 consortium (notably pana-leica and sigma) who have clearly flouted the telecentricity rules. As for Nikon or Canon - I don't think it would be possible for them to follow along (at least for existing lenses) We'll see!

 

[Riley]

Jono thats an excellent synopsis

 

[Lars]

Jono,
Got it I stand corrected.
Lars

 

[jonoslack]

Jono,
Got it I stand corrected.
Lars

Thank you . . . . but, like most things it isn't quite that simple. There are quite a lot of examples of Olympus using the words 'near telecentric'.

But I'm assuming that the principle is what makes this possible - and also why it will be tricky for others to follow (at least with existing lenses).

Personally I hate LCD and EVF viewfinders, so I'm hoping that Leica find a way to do it better (I can at least dream about a rangefinder type viewfinder with accurate lines for different lenses).

I think the point about this is that it's a small chink of light coming in from what could be a very big world . . . . we haven't even mentioned video!

 

[Robert Campbell]

While it pains me Jono to agree with you - and it pains me a lot - I think your analysis is correct.

 

[jonoslack]

While it pains me Jono to agree with you - and it pains me a lot - I think your analysis is correct.

You old bugger - I often agree with you, and I've always thought you were a lovely bloke too. And then this . . .
:ROTFL::ROTFL::ROTFL:

 

[Robert Campbell]

You old bugger

Not so much of the old, young man :ROTFL:

 

[jonoslack]

Not so much of the old, young man :ROTFL:

oops, sorry

you bugger

is that better?

 

[Robert Campbell]

The 4/3 format is roughly half 35mm size. However, half-frame cameras such as the Olympus Pen series were only about 2/3 the size of full frame 35 mm cameras - excepting of course the Olympus XA series and the Minox 35mm series. This new 4/3 micro seems to lend itself to rangefinder or direct vision cameras with interchangable lenses, yet I don't think that they will be half the size of present DSLRS or even half the size of a Leica M8.

The telecentric lens design is predicated by the general design of sensors, though clever arrangement of micro lenses can negate this to some extent - makers such as Nikon have stuck with a throat opening designed in the 1950s, when the problems with non-parallel lens design and ultra-wide angles where unthought of, and so need to look for an alternative if they are to keep compatability with older cameras and lenses - telecentric lenses need a fairly large lens throat - so Nikon are to some extent prisoners of their own unwillingness to change - and you might say that their lenses etc are triumphs of design over common sense. I'm not an Olympus owner, but I appreciate the concept behind starting from a blank sheet of paper when it comes to integrating lens and sensor design - no other DSLR maker has done this, and it is arguable that they are in part lumbered with yesterdays technology.

Olympus seem nowadays to have largely overcome the noise problem associated with their relatively small sensors - and lenses for this new design ought to be considerably smaller than the retrofoucs designs for DSLRs - just compare the enormous sizes of lenses for the average DSLR to those of comparable focal length for a rangefinder - I don't see any advantage in largeness in lenses just for its own sake [unless you want to make a priapatic statement ]

However, the viewfinder for any presumptive rangefinder camera looks like being a weak area at present; electronic viewfinders have a rather coarse pattern, and 'chimping' has its own disadvantages. Any optical viewfinder must allow for variable focal lengths, and yet be big enough to satisy today's users - the mini viewfinders of the Olympus XA or Minox 35mm are no longer acceptable.

 

[Robert Campbell]

oops is that better?

Like this, yes

 

[jonoslack]

The 4/3 format is roughly half 35mm size. However, half-frame cameras such as the Olympus Pen series were only about 2/3 the size of full frame 35 mm cameras - excepting of course the Olympus XA series and the Minox 35mm series. This new 4/3 micro seems to lend itself to rangefinder or direct vision cameras with interchangable lenses, yet I don't think that they will be half the size of present DSLRS or even half the size of a Leica M8.

Well, I guess it depends whether one is speaking of volume or length, but I do agree in principle - still, I'd estimate that they should be able to make a body around the size of the Canon G9 (which must be about 2/3 the size of the Olympus E420).

The telecentric lens design is predicated by the general design of sensors,

I think I quite disagree with this - telecentric design is about getting the light out of the rear of the lens parallel . . nothing to do with the sensor except that sensor design is much easier if you assume telecentricity.

though clever arrangement of micro lenses can negate this to some extent - makers such as Nikon have stuck with a throat opening designed in the 1950s, when the problems with non-parallel lens design and ultra-wide angles where unthought of, and so need to look for an alternative if they are to keep compatability with older cameras and lenses - telecentric lenses need a fairly large lens throat - so Nikon are to some extent prisoners of their own unwillingness to change - and you might say that their lenses etc are triumphs of design over common sense.

Quite agree - I think this is where Olympus have it right, and why it may be difficult for others to follow

I'm not an Olympus owner, but I appreciate the concept behind starting from a blank sheet of paper when it comes to integrating lens and sensor design - no other DSLR maker has done this, and it is arguable that they are in part lumbered with yesterdays technology.

I think so, and if this is a success . . .

Olympus seem nowadays to have largely overcome the noise problem associated with their relatively small sensors - and lenses for this new design ought to be considerably smaller than the retrofoucs designs for DSLRs - just compare the enormous sizes of lenses for the average DSLR to those of comparable focal length for a rangefinder - I don't see any advantage in largeness in lenses just for its own sake [unless you want to make a priapatic statement ]

Only OLD buggers would do that:ROTFL:

However, the viewfinder for any presumptive rangefinder camera looks like being a weak area at present; electronic viewfinders have a rather coarse pattern, and 'chimping' has its own disadvantages. Any optical viewfinder must allow for variable focal lengths, and yet be big enough to satisy today's users - the mini viewfinders of the Olympus XA or Minox 35mm are no longer acceptable.

Again, I quite agree, although it is possible to make good screens (witness the Nikon D3/D700). I'd be surprised if Olympus or Panasonic bothered much beyond offering a hot shoe and a good quality LCD. Seems this might be a point for Leica to make a real hit.

 

[Robert Campbell]

I think I quite disagree with this - telecentric design is about getting the light out of the rear of the lens parallel . . nothing to do with the sensor except that sensor design is much easier if you assume telecentricity.

I don't think we are disagreeing: non-telecentric design needs micro-lenses or equivalent to avoid purple fringing from photon overspill; but this problem can be overcome. Telecentric design - though I gather this is relative rather than abolute - simplifies the overall design of sensors and micro-lenses and surely simpler is better [and cheaper?]. Alternatively, complex engineering solutions [micro lenses] can more or less overcome inherent deficiencies [non telecentric lenses, small throats].

Anyone else brave enough to start from scratch?

 

[jonoslack]

I don't think we are disagreeing: non-telecentric design needs micro-lenses or equivalent to avoid purple fringing from photon overspill; but this problem can be overcome.

It can . . . but I'm not sure that the IR problem can be overcome so easily.

Telecentric design - though I gather this is relative rather than abolute - simplifies the overall design of sensors and micro-lenses and surely simpler is better [and cheaper?]. Alternatively, complex engineering solutions [micro lenses] can more or less overcome inherent deficiencies [non telecentric lenses, small throats].

Anyone else brave enough to start from scratch?

Well, we certainly agree that simpler is better - it would appear that Olympus was brave 5 or so years ago when they announced 4/3. Let's hope they are going to reap the benefit of it by producing something that it's hard for others to follow. . . . and let's hope that Leica realise what a big part they can play in this, both in terms of lenses and high quality bodies.