New Lens (Irix 15mm Firefly) Apparently Defective

[Thlayle]

I just bought an IRIX 15 mm Firefly to replace my Rokinon 14 mm that I bought approximately five years ago. My use of this lens is intended primarily for night/astro photography.

My first test shots looked great in terms of sharpness and overall appearance. However, many shots turned out to have a black triangular area in the upper right and lower left corners. I am not able to attach an example image since the CIC site keeps erroring out in my Safari browser every time I try to do so. The most extreme cases of this problem show a well defined, black triangular area in the upper right and lower left corners of the image.

I expected some vignetting with this lens but I don't think this is vignetting but some other problem. Also, I will add that these two observations: when shooting, a darkened area is noticeable in the upper right portion of the view finder (an area that roughly matches the problem area in the resulting image) AND for some reason the maximum aperture shows on my Canon Camera as f 2.5 (instead of the actual 2.4 max).

I am returning this lens. My question comes down to this: should I even try exchanging for a different one or just give up and look for a different lens? I have not seen this issue reported for this lens, just the vignetting.

I ask for this input since I am not experienced with very many different lenses and I just don't know what to expect from here now that I have what I assume is a defective lens.

Beyond all that, I am open to suggestions for a different UWA for my night photography, although I am trying to stay in the low budge range (which I understand will likely leave me dealing with more issues!).

Thanks,

Randy

[Thlayle]

I will be the first to reply to my own post: what a laugh --after multiple tries at attaching an image, failing, giving up, closing my browser and starting over AND NOT attaching an image, the example file now shows up as an attachment. Go figure.

[Manfred M]

Those marks look like vignetting from an improperly mounted lens hood. As we see this on two opposite corners, that is suggestive of the hood being askew.

If you have a removable hood, then try taking a shot without it and see what the image looks like. If the lens hood is a permanently mounted design, then the lens is defective and needs to go back.

[Thlayle]

Those marks look like vignetting from an improperly mounted lens hood. As we see this on two opposite corners, that is suggestive of the hood being askew.

If you have a removable hood, then try taking a shot without it and see what the image looks like. If the lens hood is a permanently mounted design, then the lens is defective and needs to go back.

Wow. That will be some thing if that is all it is. Didn’t actually think to check for that. Will let you know, unless I cannot overcome my embarrassment...

[pschlute]

It appears that your lens also has a rear filter slot. You should check that area too.

[DanK]

A test will tell, but I'm skeptical. If the hood were askew, I'd expect more vignetting in one corner and less in the other. I can't see a way to make opposite corners protrude too far without making the entire hood protrude. But as they say, it's an empirical question.

[xpatUSA]

A test will tell, but I'm skeptical. If the hood were askew, I'd expect more vignetting in one corner and less in the other. I can't see a way to make opposite corners protrude too far without making the entire hood protrude. But as they say, it's an empirical question.

I thought hood parts subtended equal distances with respect to the center-line. In which case, opposite corners would have equal vignette patterns. 3D ray diagram needed ...

[Manfred M]

Wow. That will be some thing if that is all it is. Didn’t actually think to check for that. Will let you know, unless I cannot overcome my embarrassment...

This is a shot at a 14mm focal length on my Panasonic GX-7. The lens hood on the 14 - 140mm lens is a bit prone to getting knocked loose. Have a look at the top right and bottom left corner. That vignette should look somewhat familiar...

The vignette size is dependent on how far I have knocked the lens hood askew.

[DanK]

Well, empirical evidence is empirical evidence, but I'm still confused by the geometry. When you say askew, in what way? The hood is presumably rigid. If the hood is tilted relative to the axis of the lens, meaning that one point is pulled away from the camera more than others, shouldn't the opposite point be the least pulled out, with all points in between, including the other two corners, being pulled away by amounts between the two extremes? And if it is moved perpendicular to the axis of the lens, the opposite corner should also be least vingnetted. I'm just having trouble visualizing this.

In any case, we'll see whether this fixes Randy's problem.

[xpatUSA]

Well, empirical evidence is empirical evidence, but I'm still confused by the geometry. When you say askew, in what way? The hood is presumably rigid. If the hood is tilted relative to the axis of the lens, meaning that one point is pulled away from the camera more than others, shouldn't the opposite point be the least pulled out, with all points in between, including the other two corners, being pulled away by amounts between the two extremes? And if it is moved perpendicular to the axis of the lens, the opposite corner should also be least vingnetted. I'm just having trouble visualizing this.

By "askew"- I believe that Manfred means rotated about the lens axis. Hence my previous post.

[dje]

I was able to replicate this effect on my 24-105 (at 24mm) by rotating the lens hood. Most lens hoods have smaller "petals" on the short side of the frame. (Except for my 45mm Sigma which has a cylindrical lens hood!)

Dave

[DanK]

Ah. Thanks. That makes perfect sense. I didn’t think that the discussion was about rotation.


Sent from my iPad using Tapatalk

[Manfred M]

I was able to replicate this effect on my 24-105 (at 24mm) by rotating the lens hood. Most lens hoods have smaller "petals" on the short side of the frame. (Except for my 45mm Sigma which has a cylindrical lens hood!)

Dave

That makes perfect sense as the the petals are cut back to eliminate vignetting on the long side of the sensor. In the example that I posted, the lens hood was in the correct position, but not properly mounted / locked so the vignette is partially due to what you have found (rotation) and partially due to off-axis positioning. This would explain why the upper right corner is more pronounced than the lower left.

If I go searching through my images some more, I'm sure I can find different variations of these impacts.

[Thlayle]

Ok, I think I'm convinced this was an issue with the lens hood, user error if you will.

Still, there's some consolation for me in this. Just knowing that Manfred ran into this is some help. Bonus to me if he also had a bit of a time figuring it out at first.

Also, I already sent my lens back and re-ordered with a better deal: same pricing & delivery but an bundle of things to go-with (filter, etc.).

Real irony in this: I hardly ever can be bothered to put the lens hood on. But with the UWA, I am thinking the hood may be even more useful/important.

[Thlayle]

Any thoughts though on why my Canon 5d reads the aperture of 2.4 as 2.5 instead? For that matter, In the viewfinder and on the back screen quick controls it shows no f-stop lower than 2.8 when I put this lens on.

[dje]

Any thoughts though on why my Canon 5d reads the aperture of 2.4 as 2.5 instead? For that matter, In the viewfinder and on the back screen quick controls it shows no f-stop lower than 2.8 when I put this lens on.

All I can think of is that IRIX haven't quite got their firmware fully compatible with the Canon lens-camera communication protocol.

Dave

[pschlute]

Real irony in this: I hardly ever can be bothered to put the lens hood on. But with the UWA, I am thinking the hood may be even more useful/important.

I never use a lens without its hood. Stray light and reflected light can lessen your image by producing flare and low contrast image. Digital sensors are more prone to this than film was. Hoods have the added bonus of offering protection to the front element too in case you bump it against something, or in my case it also stops a dogs wet nose print (sometimes).

[DanK]

I never use a lens without its hood. Stray light and reflected light can lessen your image by producing flare and low contrast image. Digital sensors are more prone to this than film was. Hoods have the added bonus of offering protection to the front element too in case you bump it against something, or in my case it also stops a dogs wet nose print (sometimes).

I agree. I usually leave one on except when it’s impractical, e.g., when doing macros with tiny working distances. However, back to the OP: the wider the angle of view of the lens, the less useful a hood is. That’s why hoods for wide angle lenses are petal-shaped: if they were simple cylinders and cut to the length that avoids vignetting in the corners, where the angle of view is greatest, they would offer almost no protection at all.


Sent from my iPad using Tapatalk

[Thlayle]

And my old Rokinon 14mm has a permanently attached hood, so I've never given this much thought. Certainly, I was not aware that it was possible to attach incorrectly by having it rotated incorrectly.

[xpatUSA]

I agree. I usually leave one on except when it’s impractical, e.g., when doing macros with tiny working distances. However, back to the OP: the wider the angle of view of the lens, the less useful a hood is. That’s why hoods for wide angle lenses are petal-shaped: if they were simple cylinders and cut to the length that avoids vignetting in the corners, where the angle of view is greatest, they would offer almost no protection at all.

A further thought is that a 135-format lens used on a smaller format camera could have a longer hood than the one it came with.

For example, I just ordered an M42-mount 400mm tele for use on a micro-four-thirds camera - comes with a round sliding hood - I bet I could add to that hood's length without incurring corner vignetting.