35MM SUMILUX

[ho_co]

Advertisement (gone after registration)

Uh-oh. I'm getting worried. The fact that Lars hasn't responded yet can only mean he's reloading.

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[adan]

Jeez - it is obvious that Leica "recomputed" this lens - even if they made no changes whatsoever to either the shape of the elements or the glass used.

 

The only way one can figure out how much to move the floating element (actually 5 elements in 3 groups), and in what relationship to the rest of the lens and to the changes in focused distance, is to run the optical formula through the software again to see what happens to the light rays at the various locations of the FE.

 

Is this a different formula to the "old" ASPH? Obviously - at some focused distances the elements are no longer spaced as they were in the old lens.

 

Does the new lens use the same glass cut to the same shapes as the old lens? Until and unless someone can list and compare exact curvatures, thicknesses and catalog glass types for both lenses, everyone is just pulling guesses out from where the sun don't shine.

 

If I multiplied 34,296 x 41,947 seventeen years ago, and multiply them again today, I may get the same answer (1,438,614,312) both times - but I still recalculated it.

[Rick]

Jeez - it is obvious that Leica "recomputed" this lens - even if they made no changes whatsoever to either the shape of the elements or the glass used.

 

The only way one can figure out how much to move the floating element (actually 5 elements in 3 groups), and in what relationship to the rest of the lens and to the changes in focused distance, is to run the optical formula through the software again to see what happens to the light rays at the various locations of the FE.

 

Is this a different formula to the "old" ASPH? Obviously - at some focused distances the elements are no longer spaced as they were in the old lens.

 

Does the new lens use the same glass cut to the same shapes as the old lens? Until and unless someone can list and compare exact curvatures, thicknesses and catalog glass types for both lenses, everyone is just pulling guesses out from where the sun don't shine.

 

If I multiplied 34,296 x 41,947 seventeen years ago, and multiply them again today, I may get the same answer (1,438,614,312) both times - but I still recalculated it.

 

This is dead on. I'm going to measure the curves on the front and back of the new FLE and I'll post them. Maybe, someone with the old ver. can do the same.

[lars_bergquist]

Dear Howard,

 

I do respect Mr Puts, but sometimes he is in too much of a hurry. The former, network version of his Compendium is full of errors of a kind that one might term "editorial" – Mr Puts knows better, but sometimes he does not check up on what he writes. Add to that the fact that words do not seem to have a generally accepted meaning in these matters. For instance, what is the difference between a new lens, and a version of a new lens? And how different must a version be to be a version?

 

For instance, does a change to the mount only qualify as a "version"? (See 35mm Summicron "v.2" and "v.3") And why are the two long-focus 90mm Summicron lenses, SOOZI (1957–59) and SEEOF (1959–1979) treated as one version, though not only the mount was completely re-designed, but also the optical performance changed markedly due to an optical re-design? As I wrote before, it's a mess.

 

We do probably agree on the factual parts of it. Making the rear half of the lens focus differentially does of necessity entail a quite thorough re-computation of the lens. After all, the performance prognosis has to be worked out anew for the whole range of focusing. Also, after a production run of sixteen years, one would expect that a health check would be called for.

 

So it all boils down to the use of language- When Mr Puts says that the lenses are "optically identical", what does he mean? That the full set of specs for the optical elements of the two lenses are identical? That would be an impossibility. Or that Mr Puts deems the performance to be identical? (Which, by the way, is NOT the case.) I do fear that Mr Puts has allowed his keyboard to go on automatic.

 

Lars B

[Xmas]

 

For instance, does a change to the mount only qualify as a "version"? (See 35mm Summicron "v.2" and "v.3")

...

Lars B

 

For a Jurno his english is poor, (and he does not employ a proof reader,) just about as bad as mine.

In one of his books he indicates the diameter of the (internal) optics of the V.3 35mm cron was changed, increased from the V.2 cron, but he does not imply that it was either the same lens or a modification merely the same 4 gp 6 el double Gauss design, as I read it, YMMV.

 

Noel

[jaapv]

Jurno??

[giordano]

Advertisement (gone after registration)

Jurno??

 

Journo.

 

[edit] Sorry, that should have been

 

He means "journo", mate.

[lct]

"Journo" for journalist?

[Xmas]

"Journo" for journalist?

Yes

[lct]

Recalls the french "journeaux" (newspapers)

[ho_co]

Lars, I didn't mean to offend. (That's the only time I've seen you sign your name.) And I know Mr Puts can be sloppy.

 

You're right, he points out that the new lens has different performance from that of the old; that would hold also in the scenario I sent you privately.

 

I'm interested in working forward from the facts of the two lenses, the LFI remark and Mr Puts' comment. That event sequence would do just this.

 

I point out that the ray-traces would have to be run twice; but that's not the same as re-computing the lens. When the mount is changed to incorporate a floating section, the lens is a different lens, but not automatically a re-computed one, as I explain.

 

Some, I think (may be my misperception) want to prove that Mr Puts or I am wrong; that's fine, but I'm trying to build on what we've been given and build a plausible theory to explain it. Different purpose.

 

We're all in the same place with this: A lens with floating elements can't be optically identical to a lens without, can it? That's my starting point as well as yours.

 

It's that strange LFI remark that bothered me when it came out that got me interested; and I think the supposition I sent you and Jaap works and makes all the inconsistencies balance out.

 

That's for others to judge, of course.

[ho_co]

... it is obvious that Leica "recomputed" this lens - even if they made no changes whatsoever to either the shape of the elements or the glass used.

 

The only way one can figure ... is to run the optical formula through the software again to see what happens to the light rays at the various locations of the FE....

Getting a little sloppy here, Andy. To me, running a ray-trace on a given formula is not the same as "recomputing" it.

 

... Is this a different formula to the "old" ASPH? Obviously - at some focused distances the elements are no longer spaced as they were in the old lens....

By that definition, one could argue that the new lens has a different formula at each focus distance.

 

... Does the new lens use the same glass cut to the same shapes as the old lens? Until and unless someone can list and compare exact curvatures, thicknesses and catalog glass types for both lenses, everyone is just pulling guesses out from where the sun don't shine....

I would also allow the statement of authority. If the lens designer says all the glasses and curvatures are the same, I would accept that.

 

As Jaap and Lars said, we're getting into trouble because we're using the same words differently.

[Nikkor AIS]

One thing that kind of concerns me about trusting "any" sourse on how good a lens is the number of sample's they test. And frankly, I could care a less what there name is or where there getting there info ( Leica or not) unless they test multibal sample's.

 

It just dosnt seem very scientific of a test, unless you get more sample's than just one.

Now you can say "my copy" of so and so lens is... but without mutibal sample's, I just don't think to be trusted.

 

As far as the discussion in changes in optical design, it's very interesting.

 

 

 

35 1.4 Summilux on Leica M7

 

Gregory

[lars_bergquist]

Howard, I was not offended.

 

And yes, if the optical performance had been identical, what was the reason for the remake? Especially the floating element? No point in it.

 

And yes, I agree that just repeating the ray traces would not have amounted to a recomputing. But that too would have been pointless. The point of the exercise had to be that you wanted to see what changes to make (or perhaps none, to be sure).

 

When you make a change as big as making the entire rear half of the lens focus differentially, you simply have to check up on the result, and you have to check the entire lens. And remember, the old lens was a compromise in the matter of focus shift. Now you drastically reduce the shift – that means that you change the compromise, and I would be very much surprised if that did not entail adjustments of the front half too.

 

Here we have a passenger aircraft in for repairs to the starboard engine. Then the plane is sent off, full of passengers, and the engine has not been tested before the flight. I would not fly with that airline. Neither would you.

 

The old man who remembers

[ho_co]

... it is obvious that Leica "recomputed" this lens - even if they made no changes whatsoever to either the shape of the elements or the glass used....

 

That puts us on the same side, Andy: Mr Puts said "Leica states that the SX35 FLE is optically identical to the previous version." You say that's possible and could still result in a new design.

 

That was my point from the beginning, and I think we are all now pretty much in agreement that

1. my initial phrasing was bad; and
   
2. it's possible that the new 35/1.4 uses precisely the same elements as the previous one.
   

 

 

Lars has responded to my PM in the forum publicly. Jaap's response to my PM was generally positive, but with reservations, as he expressed publicly here as well.

 

Thanks.

[ho_co]

Lars, thanks for the note.

 

I still think the suggested scenario I sent you by PM was a viable explanation, though speculative.

 

Clearly, Leica had to know how the new lens would respond, and wouldn't have released it without doublechecking. The fact that the new version outperforms its predecessor is proof of that and IMO goes without saying.

 

My sole point was that Mr Puts' remark could be accurate. I didn't express that well and led us off into useless byways.

[Xmas]

That puts us on the same side, Andy: Mr Puts said "Leica states that the SX35 FLE is optically identical to the previous version." You say that's possible and could still result in a new design.

 

That was my point from the beginning, and I think we are all now pretty much in agreement that

1. my initial phrasing was bad; and
   
2. it's possible that the new 35/1.4 uses precisely the same elements as the previous one.
   

 

 

Lars has responded to my PM in the forum publicly. Jaap's response to my PM was generally positive, but with reservations, as he expressed publicly here as well.

 

Thanks.

 

I'd suggest the 'It's possible' quoted above should read 'very unlikely'

 

For three reasons

 

i) if they say it is a new lens with floating elements that does not mean that they switched to the new lens to include a floating element or to cure the focus shift symptoms. It would be more normal that a glass type is no longer available, or that you have new production equipemt, e.g. cause the previous had a problem beyond economic repair, or could not achieve better tolerances, to avoid focus shift, if it make you feel happier.

 

ii) The lenses are designed by optimisating software where the deigner tells a computer the constrants including the glass catalogue, size of front element etc, and says compute. The computer does a hill climb to creae a design matching the costraints but may come back with less performance then desired, The process is repeated until the designer is happier, the production manager (or his team are involved).

 

But it is rather unlikely that the Type I lux design could tolerate movement of a group of lenses and still produce optimum results, that would not have been one of the origonal contraints of the type I design, so the production people would have vetoed, flaoting a group arbitarily. The problem is not just the optical performance or linear position of the group, the floating group will need greater tolerances in axail and skew, as the lens is not 'solid' metal tube any more...

 

iii) The MTF curves for the type II seem different from the type I, rather better, so unless the type I had its rear group suboptimal for infinity and the shifted group has improved performance at the MTF distance (infinity) the lens havs been recomputed, i.e. redesigned.

 

I'd suggest a more heavily figured asph or higher refractive index glasses, the latter the most likely, the glass catalogue has changed since 92-93.

 

It is a better lens, I'd be happy with a type I, but if all you guys (and gurls) dump your type I like they were horsie dung, I still wont be able to justify, to big.

 

Noel

[Rick]

I have completed my measurements of the base curves of the front and back lens on the new SX35 FLE. I have chosen to represent the radius of curvature translated to its dioptric power. The front lens is probably the only measurement that is of importance here, because we know the back elements had some degree of redesign. The front elements may or may not have been redesigned. If, someone could at least measure the front element of the older version we may see a change, although I'd imagine that the front surface may stay the same even if other changes were made in the rest of the front elements. But, if the old lens has a different front surface, then the lens was redesigned for sure. I can't believe I spent time doing this.

 

Front: -10.50

Back -20.00

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[Rick]

And remember, the old lens was a compromise in the matter of focus shift. Now you drastically reduce the shift – that means that you change the compromise, and I would be very much surprised if that did not entail adjustments of the front half too.

 

The old man who remembers

 

I agree, and also the old lens was a compromise between the plane of best contrast and the plane of best micro-detail. They aren't at the same place as I understand it. The new design would have to incorporate decisions about these design goals.

 

Or, on the other hand, they could have just slapped the old rear elements into the lens and only changed the mechanics of the focusing portion. What do I know?

[adan]

Rick - sorry, I didn't mean you personally needed to measure lenses.

 

I presumed that those arguing so forcefully for a new computation and design actually had some evidence (like specs for the glass types and curvatures and such.) Apparently not.