RFF to encode leica and Zeiss lenses

Status
Not open for further replies.
The camera will carry a table of lenses it knows about - lens name, focal length, their maximum aperture, image correction characteristics - so that when a particular lens is mounted, the correct processing can be applied.

Mount a lens with a code which the camera doesn't know about and it will either ignore it or put up some sort of message.

If Leica come out with new lenses after the M8 is released, I expect the lenses will come with an SD card to teach the camera about the new lens through a partial firmware upgrade.

Mount a lens from another manufacturer and the camera will recognise it as whatever Leica lens is represented by the code, whether or not it's correct. The optical characteristics of a Zeiss 50mm lens are different from a Leica one so it makes little sense to code non-Leica lenses if you are looking for image improvements which will be tailored to the lens.

By the way, has anyone bothered to look at a Zeiss or CV lens to see if there is space on the lens mount to code it, or are there fixing screws in the way?
 
Mark Norton said:
The camera will carry a table of lenses it knows about - lens name, focal length, their maximum aperture, image correction characteristics - so that when a particular lens is mounted, the correct processing can be applied.


Mount a lens from another manufacturer and the camera will recognise it as whatever Leica lens is represented by the code, whether or not it's correct. The optical characteristics of a Zeiss 50mm lens are different from a Leica one so it makes little sense to code non-Leica lenses if you are looking for image improvements which will be tailored to the lens.
Click to expand...
Mark, I think you are right. it might be easier to look at the 75mm lenses, since there are only three with very different optical designs, the 'Lux, 'Cron and Heliar. The exit pupil and rear element diameters are probably different and the sensor fall off could be different enough to be seen. Any in-camera corrections might not be generic enought to handle all three. EXIF would show 75mm, but on the Heliar it wouldn't show f/2.5. The TTL flash might work fine, though. The EXIF info isn't important to me, as I use manual focus lenses on my E-1 all the time and haven't missed the data. The TTL flash also isn't a big issue for me, as I tend to prefer the "flashmatic" auto with flash and definitely dislike pre-flash.
A note on the "patent " issue, a check should also be made with Trade Mark and Copyright offices, just to be sure.
Bob
 
Can somebody explain how this encoding works? With six bits there are only 64 unique values. So is it a system like:
1 = XX focal length lens with YY f-stop
2 = XXX focal length lens-with YYY f-stop, etc.

That is each value encodes a specific focal length and f-stop combination. If so, then any lenses which aren't in the list (or aren't equivalents) can't be encoded.

I would have thought that a system where some bits were used for focal length and some other bits were used for f-stop would have been better. Then any combination could have been set up.
 
Robert, they could give 64 types of lenses a unique value with the rest of the information stored in the cameras rom.
 
Socke said:
Robert, they could give 64 types of lenses a unique value with the rest of the information stored in the cameras rom.
Click to expand...

I think that is what they are doing, each specific lens formulation has its own value. As we build a picture of all the codes in use, we may see a pattern, but you run into the problem of multiple 35mm f2 or multiple 50mm f2 which suggests they will not be allocating 3 bits to focal length (there are more than 8 anyway) and 3 to aperture (there are at least 5: 1, 1.4, 2, 2.8, 4).

By identfying the lens as a "type 31", the camera will have been programmed with everything it needs to know about this lens.

But, we do not expect the camera to report working aperture, not unless they've come up with a clever way of figuring it out.
 
I may be missing something, but from the photos I've seen this looks like a series of black or white ovals painted on the lens mount.

I don't see why someone couldn't produce a sheet of self stick labels with all the possible combinations. A simple jig of cardboard that fits around the mount with an opening to show where to apply the appropriate label would then be used for positioning. If these were printed on mylar they would stand up for quite awhile.

If they eventually got worn, just peel off at put on a fresh one. When the DX codes came out for film there were similar products offered to tag refillable cartridges.
 
robertdfeinman said:
I may be missing something, but from the photos I've seen this looks like a series of black or white ovals painted on the lens mount.

I don't see why someone couldn't produce a sheet of self stick labels with all the possible combinations. A simple jig of cardboard that fits around the mount with an opening to show where to apply the appropriate label would then be used for positioning. If these were printed on mylar they would stand up for quite awhile.

If they eventually got worn, just peel off at put on a fresh one. When the DX codes came out for film there were similar products offered to tag refillable cartridges.
Click to expand...

Actually, there are 6 pits milled into the surface of the lens mount and filled with paint to make them resistant to wear. I'm quite sure a thin self-adhesive film could be used but it will eventually wear. New lenses have the code as standard and the cost of upgrading old lenses is the cost of a new bayonet ring, fitting it, testing the lens, administration, shipping and tax.

$125/€95? It's a bargain.
 
Jorge,

Why would someone who spends $5000 for a M8 try to save a few bucks on lens coding? It sure beats me. Also I cannot see how you could do the machining, supply a new bayonet, test, etc for much less than $125.

Also, I would not be so sure about possible patent issues. Just because they have not yet said anything to you means nothing. Obviously the patent would be pending at this point so its enforceability is moot.

Perhaps your energy might be better devoted to encouraging your sponsors to pass on film/shipping or whatever saving to RFF paying members.
 
There speaks a voice of reason, thanks Jon. Doing this separately from Leica is a pointless exercise which will cause Jorge no end of hassle.

What are you going to use to verify the camera can read the code before it is sent back? What's the lead time going to be keeping in mind the mounts will have to be sent for machining in batches and then matched with the lenses on return? How are you going to replace a batch of rings which are incorrectly machined or damaged or lost in the mail? Who are the RFF members going to sue (given you Americans are a litigious lot) when the lens they get back is not the lens they sent or claim that some other damage has been introduced? What's the impact on resale value going to be when you try to trade one of these lenses at a dealer?

And so on. It's a non-starter IMHO. Get Real!
 
Just a quick status report.

The manufacturer of the machine has just informed me that they have successfully machined a test mount. I should receive the sample tomorrow. If it meets my expectations, I will probably be getting the machine shortly and begin modifications soon.
 
Jorge, just a few questions...

How are you going to verify the coding is correct?

You have a handful of codes from pictures I have posted, where are you finding out the other codes from?

What are you going to do about lenses which have mounting holes where the code is supposed to go, such as the 35mm Summicron pre-ASPH?

As I understand it, you want to charge $60 + shipping, compared to Leica's $125 including shipping. The difference (let's say $50) is little more than 1% of the cost of the M8 it will go on (say $4500 + sales tax). Are Americans really that (as we would say in the UK) tight?

Are you taking out insurance for when things go wrong, for whatever reason, as they inevitably will from time to time? "My lens got scratched while you were modifying it"

I suppose the good news is that since the work only involves modification to the bayonet ring, which is discarded by Leica anyway, unhappy customers can still send their lenses to Leica to be modified properly. It will at least brighten the Leica technician's day.
 
Mark, I agree with your aguments against Jorge's proposed scheme but don't you think it's time to give it a rest? I'm sure Jorge understood your objections the first time. Personally, I'm not sure why all the excitement? I (like, no doubt, a few here) have ponyed up a deposit on the M8 but I don't expect to have one in my hands until October at the very earliest. If it turns out that the lens codes are going to make a significant difference to image quality (i.e. do something that can't be done during the RAW conversion) then I guess I'll think about sending my lenses to Leica for the coding. I don't really see the need for rushing into this now - it's not as if, without the codes, the M8 is going to self-destruct at Christmas.
 
Ian, you're probably right, but I can guarantee that Jorge did not know (and nor did I until Bob Ross pointed it out to me) that one lens has a hole where the code should go and he will not therefore be able to grind the required depressions in that bayonet ring.

So in a sense, I'm just pointing out the pitfalls.
 
Mark Norton said:
Ian, you're probably right, but I can guarantee that Jorge did not know (and nor did I until Bob Ross pointed it out to me) that one lens has a hole where the code should go and he will not therefore be able to grind the required depressions in that bayonet ring.

So in a sense, I'm just pointing out the pitfalls.
Click to expand...
Hi Mark,
I'll probably take Ian's approach, because EXIF info isn't important to me and I don't tend to use flash for the TTL zoom feature, but where I think Jorge's idea has the most merit is for lenses not on the lists and those using the SrwMt-M mount adapters, or Zeiss glass. Since we all have individualized kits, we'll have to figure out if it has any usefulness. I think all my Canadian Crons have that mount screw in the dot area (35, 50 & 90).
Bob
 
I get the lenses coded mainly for the expected improvement in image quality. Life is too short to sit in front of a PC processing all the images so I prefer to get them as right as I can out of the camera and process the RAW files only if I need to.

I had forgotten that the coding will be available for non-Leica lenses though whether the firmware will yield the same improvements is an open question.

In the meantime, I've just bought another (uncoded) lens. It arrives on Friday and will go off by DHL to Solms on Tuesday.
 
Mark,

To answer some of your questions.

Codes will be determined by a source which is having the entire range of lenses coded. You were also another source for the codes when you posted the images of of your coded lenses. There is no secret to the code. it is available from any camera store that has the new lenses. simply look at a lens and copy the code for that specific focal length/lens version.

The lens with a screw hole where the code goes is still in question. Once I see how Leica does it, O will be able to duplicate it exactly unless it involves a different mount. At that point I probably will not address that lens.

My cost for doingthe mods will be based on the lens owner emoving the lens mount and sending it to me. This will eliminate the need for added insurance and remove the risk of damage in shipping. However, I will accept the entire lens for those who do not wish to remove the mount.

Money is money. some people are really pulling all strings to buy the M8. Anyway you look at it, having two lenses encoded for the price of one is a good deal. Even the richest of the rich are "tight" with their money.

I willplace a photo of a machined mount soon to show progress in getting this done right.

I am more than glad to answer your questions when I have an answer. I surely dont expect people to have their lenses altered without asking first.

Thanks to people like you, everyone is well informed.
 
Mark Norton said:
I get the lenses coded mainly for the expected improvement in image quality. ..
Click to expand...

Not me. If I get my lenses encoded I will do so only to get the EXIF metadata information. I personally am a 100% RAW user and don't want the camera firmeware messing with my data. It would be nice, however, if the encoding would provide me with the focal length and F# of the lens. That's all I want.

It's also more important that I can turn OFF any features that try to improve the vignetting, chromatic abberation, etc. First of all, I dont see how the supposed feature could actually work in practice. For instance, reducing vignetting always increases noise in the corrected area. So I have to use my own judgement in deciding how much, if any, vignetting to correct. How can the firmware "know" what my judgement would be?

Rex
 
I see your point but I trust Leica to employ a lightness of touch which is mainly concerned with obtaining even coverage across the frame, such as you would get with film. The natural vignetting of the lens should be left untouched.
 
Status
Not open for further replies.
Back
Top Bottom