An odd question

I own a Nikkor 20-35 f2.8 D lens. My camera tech and I scratch our heads as to why my D750 (and his D800) identify it as a "Tokina AT-X 235 AF PRO (AF 20-35mm f/2.8" in the data. The lens works beautifully with the camera and the results are excellent, but why the odd ID?

Anyone have a thought on this?

It's not the camera that's reporting your lens as a Tokina; it's the software you use to view/edit photos that's doing so.

I'm going to guess that Tokina's engineers, when they reverse-engineered the Nikkor to come up with the content for this lens's CPU program, went as far as to copy the lens ID from the Nikkor. With the two lenses sharing the same ID (a number or nondescriptive string), perhaps the makers of photo viewers/editors got confused and started mapping that ID to the Tokina lens description instead of the Nikkor one.

It's not the camera that's reporting your lens as a Tokina; it's the software you use to view/edit photos that's doing so.

 

I'm going to guess that Tokina's engineers, when they reverse-engineered the Nikkor to come up with the content for this lens's CPU program, went as far as to copy the lens ID from the Nikkor. With the two lenses sharing the same ID (a number or nondescriptive string), perhaps the makers of photo viewers/editors got confused and started mapping that ID to the Tokina lens description instead of the Nikkor one.

This makes perfect sense, of course. I've not handled the Tokina, but I've seen them and they certainly are similar. Reverse engineering another company's software is a bit of an iffy notion....

I'm going to guess that Tokina's engineers, when they reverse-engineered the Nikkor to come up with the content for this lens's CPU program, went as far as to copy the lens ID from the Nikkor.

I've never seen this in Tokina lenses I own, but there is no doubt that many of the something-or-other to EOS adapters do sometimes use circuitry that has a 'residual' identity.

I don't think the Nikon cameras would record any brand at all. They assume all lenses you use on them are Nikon.

Used on a digital camera the make and model of the lens appears in the data associated with each image as long as the lens has an electronic connection with the camera. Even lens as old the the D series (pre-digital) show this information. For example the data for my 300mm shows up as "Nikon D750 AF Nikkor 300mm f/4 IF-ED".

Fiddlefye - what is the software giving the wrong exif information?

Both Photos (Mac) and Photoshop/Lightroom.

I note that my D7100 does not give the lens manufacturer in the in-camera EXIF data, but that the program "Exiftool" does. That's a free program that works in Dos. If you set it up right, you can drag any image to it, and it reads more stuff than any other program I've found.

Works in Linux and on the Mac too (at least from the command line). It did identify my Nikkor 50mm f/1.8G AF-S as such, which kind of surprised me - unless it's cross-referencing a database, I wasn't expecting that much information. Older lenses may share less, and rely on the software working out from the provided details (aperture and focal length as a start; it wasn't until AF-D that focus distance got involved, so the difference between two lenses of the same basic spec might get missed). Assuming the cameras have an internal database and a third-party lens would benefit from in-body lens corrections applied to the on-brand one, it's in the interests of Tokina to report an identical id (and Nikon are not known for going out of their way to support third-party glass); if the Tokina then becomes more popular than the Nikkor, software may assume that's what it's got. Or that may be unrelated.

exif data is for tossers

Care to elaborate on that?

As to the OP-I've had Lightroom throw out some really bizarre identities on older lenses for which it didn't already have a profile. I forget what it identifies my 105mm f/2.8D Micro as-I think some sort of Tokina macro.

Something to do with pasta or salad, perhaps? I'm quite keen on it, partly to remember what I was doing/using and partly because I use DxO by default, and that applies lens-specific corrections based on the image EXIF data. I'll admit to having to strip the EXIF data occasionally when wanting to share online. It's been very useful for diagnosing problems on this forum, though.

I'm quite grateful to all the people (well, most of them) on Wikipedia and IMDB who collect data just because they can, for what it's worth.

I don’t have to. People who collect data just because they can are known as tossers. It’s a fact.

I don't necessarily see this as a fruitful line of discussion, especially since you've already shut it down, but I'll bite...

First of all, data makes the world go round. The fact that photography even exists is the result of folks compiling a lot of data over the past 200 years and working that in to technology to allow us to take images with simple to use devices.

Many of us care a whole lot about having recorded information about the photos we've taken. Digital makes this easy-it beats carrying around a notebook for every shot, and is still a lot less kludgy than high tech film cameras that do record the information but require you to manually retrieve it and match it up with your photos.

You seem to have a really bad habit on this forum of summarily dismissing anything that doesn't agree with your world view, whether if it's discussion of historical photographers or even techniques for photographing specific subjects(and sharing the results of those)! If you have an issue with people discussing EXIF data, there's no requirement for you to participate in the thread.

Moderator note: This has gone well off topic and I would agree,

I don't necessarily see this as a fruitful line of discussion,

Let's get back to the OP or drop it.

Moderator note: This has gone well off topic and I would agree,

 

Let's get back to the OP or drop it.

I posted simply because it was an oddity I found interesting and a bit puzzling. I'm guessing that in the early days of putting chips loaded with data in lenses all manner of things happened that no one really expected would come to light. Lo! and Behold! the new day dawned and the digital light shone into a few darker corners.

FWIW - the same issue has been reported here: wrong lens exif info with Apples Aperture 3: Nikon SLR Lens Talk Forum: Digital Photography Review

It most likely is the fault of the reporting software - as suggested above, try using exiftool to read the EXIF information directly instead of some software that may rely on a comparison with an intrinsic database. Tokina may have indeed copied the Nikon lens ID number (if not the entire chip) - and some software might have the Tokina in their database but not the Nikon 20-35 (though both appear to be of the same vintage).

Works in Linux and on the Mac too (at least from the command line). It did identify my Nikkor 50mm f/1.8G AF-S as such, which kind of surprised me - unless it's cross-referencing a database, I wasn't expecting that much information. Older lenses may share less, and rely on the software working out from the provided details (aperture and focal length as a start; it wasn't until AF-D that focus distance got involved, so the difference between two lenses of the same basic spec might get missed). Assuming the cameras have an internal database and a third-party lens would benefit from in-body lens corrections applied to the on-brand one, it's in the interests of Tokina to report an identical id (and Nikon are not known for going out of their way to support third-party glass); if the Tokina then becomes more popular than the Nikkor, software may assume that's what it's got. Or that may be unrelated.

Exiftool, like other EXIF readers such as the ones in Lightroom and Photos, does of course use a lookup table to provide a more useful lens descriptor than the bare lens ID, which (at least in the case of one of my lenses) is composed of eight pairs of 2-digit hexadecimal numbers. I happened to notice just now that exiftool and Lightroom provide different lens descriptors for my Tokina 100mm macro.

Incidentally, a sixteen-digit hexadecimal number is enough for 18,446,744,073,709,551,616, or about 18.4 quintillion, unique IDs. Tokina could have chosen a different ID; the reason they chose the one for the Nikon 20-35mm might have been in order to take advantage of the distortion correction profile that Nikon cameras have for the Nikkor—if, that is, the two lenses have similar geometric distortion.

FWIW - the same issue has been reported here: wrong lens exif info with Apples Aperture 3: Nikon SLR Lens Talk Forum: Digital Photography Review

It most likely is the fault of the reporting software - as suggested above, try using exiftool to read the EXIF information directly instead of some software that may rely on a comparison with an intrinsic database. Tokina may have indeed copied the Nikon lens ID number (if not the entire chip) - and some software might have the Tokina in their database but not the Nikon 20-35 (though both appear to be of the same vintage).

Exiftool also uses a lens ID lookup table. But it's a different one from those used by Apple and Adobe.

So it does - there's a lot of commentary data encoded, so I wondered whether the manufacturers had started actually including the branding names in plaintext. You could get about 12 characters in there (if I can do maths; say 37 values for alphanumerics plus a space), which wouldn't get you full branding (with "NIKKOR" taking up half), but would distinguish a lot of Nikon's alphabet soup given that the min/max focal lengths and apertures are recorded separately.

A little rummaging uncovered this, which is moderately fascinating if you're the kind of geek I am, and probably incredibly dull if you're not. This indeed show that lens id "2F 40 30 44 2C 34 29 02" means both the "AF Zoom-Nikkor 20-35mm f/2.8D IF" and the "Tokina AT-X 235 AF Pro (AF 20-35mm f/2.8)".

While it's possible, I suspect it's unlikely that the Nikkor and Tokina were so similar that in-camera corrections should be identically applied to both - especially for a lens launched in 1994 (when the F4 was current). I admit it might be used to affect matrix metering, if focal length is insufficient for that. As far as I can tell, neither the data backs nor Photo Secretary actually record a lens ID (although I also didn't know that Photo Secretary could show you the output of the matrix meter - apparently I own an earlier Nikon digital camera than I thought I did); the F6 records a little more, but still not the ID, AFAICT. I don't know if it's actually possible to get the lens ID off any film body (or the Kodak digital backs based on them), so I assume Tokina weren't too fussed at the time, but probably also decided that Nikon were unlikely to add either in-camera or software support for a Tokina lens, and therefore that being identified as the Nikkor was likely to be more useful than nothing.

(although I also didn't know that Photo Secretary could show you the output of the matrix meter - apparently I own an earlier Nikon digital camera than I thought I did);

I REALLY wish I could find a copy of Photo Secretary.

I have an MC-34(grabbed it when it popped up for sale from Japan on Ebay several months ago) and of course have Macintoshes to which I can connect it and run the software on out the rear end. There's a PowerMac 9600 sitting right next to my Mac Pro-it's mostly a stand for one of my scanners, but does work and I power it up occasionally-mostly to play Civ II or some other old game but it's actually quite a capable computer for late 1990s programs.

There's a PowerMac 9600 sitting right next to my Mac Pro-it's mostly a stand for one of my scanners, but does work and I power it up occasionally-mostly to play Civ II or some other old game but it's actually quite a capable computer for late 1990s programs.

Wow.

Wow.

Old Macs are a bit of a hobby of mine-I have computers ranging from a "Macintosh" made in May of 1984 up to the MacBook Pro I use every day(that I'm typing from). I have a bunch of 68K, PowerPC, and Intel desktops and laptops, including a bunch of real "hotrod" PowerPC computers.

If I actually managed to find a copy of PhotoSecretary, I might have to migrate some of my 35mm scanning to a PowerPC system. I have ways to put a serial port in a G4 tower so that the MC-34 will connect(Stealth Serial Port), and can run either my Coolscan III or Coolscan V easily from OS 9 with Nikon Scan(serial ports are a pain to get working in OS X).

It would all be a fun and interesting experiment-if I could just find the software.

BTW, I do all that stuff now with Meta35, but I REALLY want to play with looking at the matrix output from the F5 now that I know it can be done-Meta35 won't do that. The only thing is-I don't THINK there was ever a version that would work with the F6. I've never looked in detail at the data I can pull from the F5 vs. F6, but I think the only "official" Nikon way to get data from the latter is to connect the MV-1 and have it spit out a .csv. I've never played with an MV-1 since I can't bring myself to spend $200 on a glorified CF card reader...

Old Macs are a bit of a hobby of mine-I have computers ranging from a "Macintosh" made in May of 1984 up to the MacBook Pro I use every day(that I'm typing from). I have a bunch of 68K, PowerPC, and Intel desktops and laptops, including a bunch of real "hotrod" PowerPC computers.

Oh, I wish I could say that my amassing of a "history of Macintosh" was a "hobby"