RCicala

DPReview Contributor
Joined on Oct 7, 2011

Comments

Total: 64, showing: 1 – 20
« First‹ Previous1234Next ›Last »
In reply to:

Jon Stern: It's interesting that they're testing lenses on modified cameras, rather than using a Trioptics machine.

Well, they are doing center and subassembly tests on an interferometer. But the off-axis camera testing dies hard.

Link | Posted on Mar 20, 2017 at 22:41 UTC
In reply to:

Entropy512: This is mainly for @RCicala (maybe I should post it as a blog comment instead?)

"the PCB is small and only has one large flex connecting it to all the camera’s electronics. A typical 70-200 f/2.8 would have at least half-a-dozen connections" - I see two large flexes there? The 10-pin E-mount and then another large flex going deeper into the lens.

It would be interesting to know what is on that particular board though - in future teardowns, would you mind taking detailed photos of all of the PCBs (if possible - detailed enough such that IC markings can be read) for the curious EEs in your audience? :) (Your picture of the "big" PCB is much better - the markings are a bit hard to read, but that's likely because not all IC markings are easy to read even with the naked eye...)

I probably have some better images, and certainly some that aren't cropped. You can email me roger at lensrentals and I'll send you some.

Link | Posted on Mar 9, 2017 at 03:13 UTC
In reply to:

OlyPent: Eventually, will this thing end-up on the used lens market?

Depending on the lens from 30% to 75%. It's not a great indicator of quality, though. For example, we'll quickly do a $500 repair on an Otus, but not a Canon 85 f/1.8. There's also the parts thing. We can buy Canon parts by the gross at reasonable prices, so a Canon lens has little value as a parts donor. We can't buy Nikon parts without going through shady repair shops in iffy parts of the world, so a Nikon lens may be worth $$$ in parts which makes that $500 repair less attractive. We can buy some Sony parts, but they're pricier and many are in big assemblies, so a Sony lens with, say, a broken mount might donate enough parts to fix 4 or 5 other lenses. So lots of Canon lenses live to be sold, but fewer Sony and Nikons, but it's more about economics than simply reliability.

Link | Posted on Mar 6, 2017 at 22:02 UTC
In reply to:

Ultimax: Maybe put a drop sensor device in an easy to access panel. I think this would keep the renters and second hand market honest.

Actually the surprise was stuff didn't get dropped all that often. You know, looking on the internet I thought every package got dropped 12 times, but it turned out it wasn't worth sensors.

Link | Posted on Mar 6, 2017 at 13:29 UTC
In reply to:

Ultimax: Maybe put a drop sensor device in an easy to access panel. I think this would keep the renters and second hand market honest.

We've actually done that with shipping cases. :-)

Link | Posted on Mar 5, 2017 at 19:48 UTC
In reply to:

weathersealed: Praise, speculation, criticism of built products carries much less weight when then techs are not certified by Sony to repair this lens or sat down with the design team and gone over everything why and what for.

You are correct, we are not Sony Certified. We could be if we wished, but it's very expensive to buy all the equipment it requires. We sit down with their engineers regularly and have excellent give-and-take both by email and in person. BUT there are things they don't tell us and other things I can't share because of nondisclosure agreements.

So far I've been able to tread the fine line of continuing the blog despite the nondisclosures. For example, on several occasions I've done contract work for a manufacturer, then gone back and repeated the same tests on our own gear so we could write it up. That's getting more difficult and may come to an end for some brands this year, partly because being 'factory authorized' for most brands, requires complete nondisclosure.

That being said, this was, as you suggest, the first time we'd ever seen the inside of this lens. And our reason for doing it is primarily so that we can keep lenses in operation to our standards.

Roger

Link | Posted on Mar 5, 2017 at 19:48 UTC
In reply to:

OlyPent: Eventually, will this thing end-up on the used lens market?

Damn, I knew I shouldn't have shown the serial number in all those pictures :-) Actually, this one is going to be a 'learning lens' for us. We'll spend time off and on for a month moving optical adjustments and seeing what effects they have, documenting, etc. writing our own manual so we can work on these. It may go back into stock some day, it may become a parts donor. If it does go back to stock, it will be one of the most heavily tested and closely analyzed lenses in the fleet.

If it does sell in two years (only about half our Sony lenses make it to sell -- see parts donor above) we'll mention it because linking to the articles will probably make it somebody's collector's item. You'd be surprised how that works.

Link | Posted on Mar 5, 2017 at 19:32 UTC
In reply to:

Angrymagpie: Just in case I'm misunderstanding this, the mystery is basically: how come such an expensive and market-as-innovative-and-professional lens did so poorly in the test?

I have to repeat all the testing on Lensrentals-owned copies, and we're kind of backed up with some contract work. A couple of weeks at least, maybe a month, I'm afraid.

Link | Posted on Mar 4, 2017 at 00:35 UTC
In reply to:

Angrymagpie: Just in case I'm misunderstanding this, the mystery is basically: how come such an expensive and market-as-innovative-and-professional lens did so poorly in the test?

You're right and I apologize. I wanted to just say this teardown had nothing to do with MTF testing and that was not the mystery. We were just fixing a broken lens and had no idea what was wrong with it.

Link | Posted on Feb 27, 2017 at 13:00 UTC
On article Things we found at CP+ 2017 which had been cut in half (89 comments in total)

About the M9 review -- so you're saying there's a chance, right?
That was fun, Barney, made my morning.

Link | Posted on Feb 26, 2017 at 16:13 UTC as 38th comment
In reply to:

Angrymagpie: Just in case I'm misunderstanding this, the mystery is basically: how come such an expensive and market-as-innovative-and-professional lens did so poorly in the test?

I can't say yet, some of it is still under contract. But in general it's a variance thing.

Link | Posted on Feb 26, 2017 at 15:44 UTC
In reply to:

Angrymagpie: Just in case I'm misunderstanding this, the mystery is basically: how come such an expensive and market-as-innovative-and-professional lens did so poorly in the test?

Nah, we figured that out. The mystery was basically what broke inside to make it stop focusing.

Link | Posted on Feb 26, 2017 at 12:24 UTC
On article Prime or zoom? LensRentals investigates (232 comments in total)
In reply to:

shademaster: stupid question: why is the subject distance not an important parameter? e.g. nobody ever talks about a given prime being soft for close objects but sharp for far ones. why not? (sorry if this is basic optics)

in reply to shademaster: I'm always concerned that wide-angles tested on DxO or Imatest are being tested too close (it can be 3-4 feet testing distance, but even 6 concerns me). On the other hand Macro lenses tested on the optical bench at infinity are not at their best either. For most lenses in the standard to telephoto range, though, I doubt it matters much.

Link | Posted on Feb 11, 2017 at 21:55 UTC
On article Prime or zoom? LensRentals investigates (232 comments in total)
In reply to:

FLruckas: "damn time"
Is that the time zone for Lens Rentals?
Just kidding.....
Someone can sell it to you for less even if its 1 cent less.
But it's not really what you're thinking of less unless it'd hundreds or thousands of dollars less.
There is variation from lens to lens.
You can get a bad copy.
Do lens manufacturers need to sell "certified" lenses?
How many of us have access to tools like Rodger?
Do we loose sight of anything when we know a good zoom can have some red areas where it's not as sharp?
In previous posts it was pointed out that F stop is a factor as well.
I'm assuming this post is done at the sharpest F stop for each lens.
So it becomes a multidimensional mine field of uh ohs.
F stop. Good copy. Zoom or not. Focus distance.
Since Rodger has access to so many lenses from so many manufacturers I wonder if he's ever published a sheet like
80% of Sigma ART lenses need to be calibrated etc.

I do agree with that, but with two additions.

1) Give a number for each aspect. Perhaps a sharpness number, a focus accuracy number, Lat. color. etc. Don't lump everything together because different photographers have different areas of interest. That's why I said my comments aren't trashing DxO = they give that information and I appreciate that.

2) Don't use insignificant digits in the overall rating. Saying it's 8.34523 is repeating 'insignficant digits'. How can they tell? Simple, test 6 or 8 copies of a lens and see how much they vary. 8.3 might be significant if numerous copies of the lens ranged from 8.26 to 8.34.

But then the Nikon, Canon and Sony might all be 8.3 and there wouldn't be a zillion posts among arguing fanboys about who had the better lens. :-)

Link | Posted on Feb 11, 2017 at 19:10 UTC
On article Prime or zoom? LensRentals investigates (232 comments in total)
In reply to:

FLruckas: "damn time"
Is that the time zone for Lens Rentals?
Just kidding.....
Someone can sell it to you for less even if its 1 cent less.
But it's not really what you're thinking of less unless it'd hundreds or thousands of dollars less.
There is variation from lens to lens.
You can get a bad copy.
Do lens manufacturers need to sell "certified" lenses?
How many of us have access to tools like Rodger?
Do we loose sight of anything when we know a good zoom can have some red areas where it's not as sharp?
In previous posts it was pointed out that F stop is a factor as well.
I'm assuming this post is done at the sharpest F stop for each lens.
So it becomes a multidimensional mine field of uh ohs.
F stop. Good copy. Zoom or not. Focus distance.
Since Rodger has access to so many lenses from so many manufacturers I wonder if he's ever published a sheet like
80% of Sigma ART lenses need to be calibrated etc.

My thoughts are that a lens test (what I do on a specific aspect) or review (a complete evaluation of a lens) is worth looking at and gives some worthwhile information. But pretending it's so accurate that a number difference means this lens is better than that lens is silly.

Looking at half a dozen reviews should give you a general feeling about the lens. Looking at one number is just Fanboy Fuel and fairly meaningless. Saying third-graders are about a meter tall is useful to know. Saying we measured a third grader with laser accuracy and they are 2.975438 meters tall is silly. Saying "a third grader at school A was 2.975438 meters tall and one from school B was 2.932345 meters tall, so you should send your child to school A, the lunch nutrition is better" is just ridiculous.

Link | Posted on Feb 11, 2017 at 18:10 UTC
On article Prime or zoom? LensRentals investigates (232 comments in total)
In reply to:

shademaster: stupid question: why is the subject distance not an important parameter? e.g. nobody ever talks about a given prime being soft for close objects but sharp for far ones. why not? (sorry if this is basic optics)

It can be, although not very often. I think the bigger reason is testing equipment is at a set distance: Optical bench is at infinity, Imatest and DxO are close up, etc. So no one really checks a single lens at multiple focusing distances.

Link | Posted on Feb 11, 2017 at 18:05 UTC
On article Prime or zoom? LensRentals investigates (232 comments in total)
In reply to:

Rick Knepper: I scanned the article for tests with specific lenses and didn't see anything so I didn't read the article. No need to because I know what I know. If the following matches the article then great..

What I know is the best zoom cannot match the best prime (I'll throw the Zeiss Otus 55mm out as an example) but the best zoom can match some primes, especially budget primes or primes from a generation ago (for some brands at least) when the entire frame is taken into consideration.

IMO, the new standard for premium lenses is sharp across the frame wide open. The lens doesn't have to be uniformly sharp across the frame but the edges and corners have to be sharp with minor fall-off from the center.

Well, you guys said in 2,000 less words than I did. But that's about it.

Link | Posted on Feb 11, 2017 at 14:10 UTC
On article Prime or zoom? LensRentals investigates (232 comments in total)
In reply to:

RichyjV: This should be compulsory reading for any forum post including the words
"DXOMark Score", and a few others.

Actually wasn't aimed particularly at them. They do have lots of useful sub-data that I look at and refer too. Not the overall rating, of course, which I truly believe is there to start internet arguments that gets the name repeated 6 zillion times. I don't think DxO believes in that number any more than I do. But they provide lots of other information if you read deeper that is really good stuff.

Link | Posted on Feb 11, 2017 at 14:05 UTC
On article Lensrentals tears down the Nikon 105mm F1.4 (157 comments in total)
In reply to:

WIMorrison: For the uninitiated can you tell me what the GMR unit is and why we mustn't look at it let alone touch or breathe on it?

Some of you have had a lens that developed 'hunting AF', it might or might not lock on but mostly just focuses all over. That's often a failed GMR unit.

Link | Posted on Dec 5, 2016 at 13:16 UTC
In reply to:

Frank_BR: Great work by Roger Cicala, as always.

The difficulties experienced by Roger Cicala to dismantle the Canon 5D Mark IV came basically from not having access to the service manual of the camera. Manufacturers are increasingly reluctant to provide information and parts for maintenance by anyone other than authorized centers. The consumer is treated like a king when he/she spends his/her hard-earned money on a product. Sadly, immediately after the purchase manufacturers start to consider the consumer as an idiot who must be away from any maintenance, however simple it may be. In fact, the maintenance of a digital camera is not so complicated if you have the right information and right parts.

Frank is exactly correct. We generally have to 'reverse engineer' everything. Access to manuals isn't happening so we have to make our own. A certified repair shop would have the manuals, and next time we do it we'll know what to expect.

An orderly disassembly is the key. You can't usually tell from the teardown articles but every piece is laid out in a specific order, every screw kept by it's hole, etc. If not we'd have zero chance of reassembling. If it's laid out properly, another tech who had never seen the camera should be able to sit down and reassemble it without difficulty.

Actually one of the reasons we started publishing teardowns is we would take pictures of 'first disassemblies' so we'd have reassembly documentation.

Link | Posted on Sep 12, 2016 at 13:51 UTC
Total: 64, showing: 1 – 20
« First‹ Previous1234Next ›Last »