Do date codes matter?

[Good24]

I've read a little bit about Canon lens date codes and some discussions that they do matter (i.e. that later in time codes give you a better lens because certain initial flaws have been cured) especially for the 50mm L 1.2 which is what I'm considering buying.

Anybody have any evidence that they matter? Or any resources or even just thoughts? And if they do matter, and if I'm buying the lens online (say, from B&H) how would I verify that I'm buying the later / "better" date code version? Thanks for your help.

 

[Loswr]

I think you've been reading the Internet too much.

Canon generally does not make 'improvements' to lenses over time, except as new versions. There were rumors of improvements to the 400/4 DO, Canon confirmed no changes had been made. Canon does fix problems, such as the. 'rays of light' flare issue that affected early runs of the 24-105L (they issued a product advisory for lenses older than UT10, fixed those, and newer lenses were produced with the fix). It's likely they make changes over the years to decrease production costs where possible.

In the case of the 50/1.2L, the focus shift is a result of the optical design. 'Fixing' it would mean changing the optical design of the lens, that's not going to be a 'silent upgrade'. Sure, people say 'I got a newer lens and it's fine'. The issue is there, whether or not you see it depends on how you shoot. Shoot wide open, no shift. Shoot subjects a few feet away instead of at the MFD, you won't notice it.

There's really no way to get a 'newer' lens with an online purchase, but with bigger retailers you tend to because they move their stock. Local stores, if they have high-end lenses at all, often have them sit on the shelf for a while.

 

[Maui5150]

Only thing date code matters for is how old the lens is. most of the time it is inconsequential, but can give you some idea, especially for lenses that have been around for a while, just how long your lens may have been out there.

For most lenses, new, it does not matter much, though if you buy a lens that has been sitting on a shelf for 2 years, you use it a few times and sell it, a buyer may wonder why the lens you say is 6 months old and shot twice has a 3 year old date code. , bets to keep your receipts handy.

For other lenses, it can be important. Take the long in the tooth 100 - 400. This lens has been around for 16 years and while an L lens, the push - pull mechanism seems to be a dust magnet. I would give a 10 year old version of this lens a much thorough looking over for dust and internal issues than say a 1 year old version.

 

[Schruminator]

Canon apparently does roll out some changes to existing models over time. Personally, I'd think that some of these changes certainly have to be for the better (reliability, performance) and cheaper components being available can't be the only reason. That being said, the difference between an older lens and a newer lens with updated circuitry is probably not all that measurable.

In the link below LensRentals notes a change they have seen. Just from a construction standpoint, 5 less soldered wires is 5 less solder joints that can break or come loose. Would they have ever been a problem in the original? Eh, probably not I suppose, but now there are 5 less to worry about

http://www.lensrentals.com/blog/2013/07/silent-changes

 

[Marsu42]

For other lenses, it can be important. Take the long in the tooth 100 - 400. This lens has been around for 16 years and while an L lens, the push - pull mechanism seems to be a dust magnet.

Good point, most like plastic/sealing parts also age with, well, age.

And if they do matter, and if I'm buying the lens online (say, from B&H) how would I verify that I'm buying the later / "better" date code version?

You can try add this kludge to the order ("only if datecode after xyz") I doubt a large retailer will take the hassle unpack the lens & decode the date - but you don't know unless you try. With smaller online retailers or esp. a local shop, try probably want the sale of an expensive L lens bad enough to have a look.

 

[jdramirez]

There is a clicking that happens when you zoom in the 24-70 f2.8L mkii. That was fixed with layer models.

The 24-105 had a rubber band that breaks after five or so years... so getting a newer lens is better than an older.

Other than that.... I don't care.

 

[Loswr]

Canon apparently does roll out some changes to existing models over time. Personally, I'd think that some of these changes certainly have to be for the better (reliability, performance) and cheaper components being available can't be the only reason. That being said, the difference between an older lens and a newer lens with updated circuitry is probably not all that measurable.

In the link below LensRentals notes a change they have seen. Just from a construction standpoint, 5 less soldered wires is 5 less solder joints that can break or come loose. Would they have ever been a problem in the original? Eh, probably not I suppose, but now there are 5 less to worry about

http://www.lensrentals.com/blog/2013/07/silent-changes

One less circuit board and five less wires may increase reliability, but I believe Canon's goal was to reduce parts cost and more importantly the labor cost of those solder points.

 

[Mt Spokane Photography]

Having been in the manufacturing business for over 30 years, I can say that changes to products do happen. Materials are changed due to price or availability, machinery wears out and is replaced by faster and more accurate machines, subtle changes and process improvements happen frequently. However, changes that affect the form, fit, or function are not incorporated unless there is no choice. When that happens, you end up with one of those dreaded situations where replacement parts vary by serial number. If a new process fails to work properly, then its a mess.

Even changing component suppliers can make things different, electronic parts are phased out every few years, and substitutes must be found. I used to do this on a daily basis for a large manufacturing company. Sometimes, a critical part can cost a huge amount of money to tool from a different supplier. I remember having to source critical parts for a cruise missile that was long out of production. The cost per part for a hundred pieces was astounding, I forced the supplier to fully test the part after I discovered that their process for making it had never really been stress tested, it failed, and a redesign had to be done. That led to a recall of 10's of thousands of parts that used the same process commercially. They started failing about 6 months after my testing / redesign was complete, so the manufacturer had already started producing replacement parts. We had notified the commercial users, but they weren't concerned, not until their customers started complaining.

So, I would expect tighter tolerances due to process improvements, and possibly cheaper materials and plastics that are cost savers to the factory. The end user will likely not ever see a difference unless he were to test a large group of lenses and see less variation, for example. Since there is no historical testing of large numbers of lenses that is reliable, we won't have hard evidence.

 

[Good24]

Thanks for all the input. I think I will not allow the issue (if there even is one) to affect any potential purchase.

 

[DRR]

My opinion - individual date codes do not matter. For example, UA vs UB vs UC, is not really going to make any sort of difference at all. However, many lenses are manufactured for a long time... some are even decades old.

A UB (2013) vs a UM (1998) could be a significant difference though. Either way, take your time to evaluate the individual copy of the lens.