Solved, me thinks: Strange Bokeh on new Nikon 70-300mm VR

Seems to be the case that if you don't like 'wierd' bokeh, don't get a 70-300 or a 100-400 lens...lol
 
Christ, I hadn't thought my thread would garner such a huge response - lots of helpful folk certainly on this forum. :clap:

Just to clarify VR was definitely OFF as I was on a tripod.

The filter I'm using is a Hoya Pro UV filter. The exact same physical filter I was using on my 18-135 before I put it on this new lens. I had not seen a problem like this on my other lens at any focal length or f/number.

I also own the non-VR version of this lens (albeit about ~15 years old) and certainly never noticed any harsh bokeh to the same affect I'm seeing on this new lens.

Obviously before purchasing this lens I had read quite a few reviews (Ken Rockwell's fairly in-depth one included) and none of them mentioned awful bokeh - KR gave it 'neutral to fair' in bokeh I believe.

Here's what I'm going to do: hopefully tomorrow I can go out and test this again (at the same place as well for consistency) and see if I get the same results. I'll try it with the filter on AND off, with VR on AND off, and also try the non-VR version of this lens on the same body with filter on AND off. I'll also try the lens on both my D90 and D40x bodies. Then I'll post the results here after my findings.

Cheers guys for the help, some very helpful people on this thread. :)
 
Well if it's anything like the 100-400L it wasn't a big problem - you'd just see the effect now and again and always with the same type of background- related I feel sure, to the shooting distance, and the distance from the subject to the bg. If that's the type of situation where you occasionally get the problem I honestly wouldn't worry about it- it's a small price to pay for the convenience of your zoom.

If it happens with a wide variety of subjects and different shooting distances then I'd be wanting Nikon to check it out as it could be something amiss like a displaced element.

My gut feeling is it's small potatos. ;)
 
CT said:
4663592094_a0440abc49_b.jpg
Click to expand...

Aye, you've come on a bit since them days Cedric! :D
 
CT said:
4663592094_a0440abc49_b.jpg
Click to expand...

I am fairly certain that the problem (as has been stated) is down to the blades and their configuration. Many a lens produces beautiful focused subject matter but is spoilt by 'eratic/not easy on the eye' bokeh. Obviously, this is highlighted with or at certain DOF shots.
 
Basket Case said:
Christ, I hadn't thought my thread would garner such a huge response - lots of helpful folk certainly on this forum. :clap:

Just to clarify VR was definitely OFF as I was on a tripod.

The filter I'm using is a Hoya Pro UV filter. The exact same physical filter I was using on my 18-135 before I put it on this new lens. I had not seen a problem like this on my other lens at any focal length or f/number.

I also own the non-VR version of this lens (albeit about ~15 years old) and certainly never noticed any harsh bokeh to the same affect I'm seeing on this new lens.

Obviously before purchasing this lens I had read quite a few reviews (Ken Rockwell's fairly in-depth one included) and none of them mentioned awful bokeh - KR gave it 'neutral to fair' in bokeh I believe.

Here's what I'm going to do: hopefully tomorrow I can go out and test this again (at the same place as well for consistency) and see if I get the same results. I'll try it with the filter on AND off, with VR on AND off, and also try the non-VR version of this lens on the same body with filter on AND off. I'll also try the lens on both my D90 and D40x bodies. Then I'll post the results here after my findings.

Cheers guys for the help, some very helpful people on this thread. :)
Click to expand...

Thanks for putting in the extra time with this - it should help clarify all our speculation and give us something to work with...
 
I don't know why you seem to spend so much time defending filters. I'm sure there's a really good reason all the other elements in the lens are precisely curved etc. It stands to reason that if you put a piece of flat glass in the path of light travelling into the lens it's going to cause some sort of degradation of quality/refraction.
Obviously, there are times when it's better to use one, but it's still going to be a compromise.
 
Arkady said:
If that were solely to blame why do we not see this with all telephoto lenses over 200mm?
Click to expand...

There are very few telephoto lenses over 200mm really...

Off course 300mm f/2.8, 400mm f/2.8, 500mm f/4 don't take UV filters in any case.

I personally think lens designers spend a great deal of effort designing lenses for optimum performance, with elements just so then some dozy clot plonks a generic flat filter onto a lens and all bets are off...
 
ruvor said:
I don't know why you seem to spend so much time defending filters. I'm sure there's a really good reason all the other elements in the lens are precisely curved etc. It stands to reason that if you put a piece of flat glass in the path of light travelling into the lens it's going to cause some sort of degradation of quality/refraction.
Obviously, there are times when it's better to use one, but it's still going to be a compromise.
Click to expand...

I agree (sorry Rob...) and despite what has been said, there HAS been numerous times where filters have been proven to degrade quality. Hoppy even took his own comparison pictures to prove it. I too can see their advantages but I don't see what the big deal is. If lenses were designed to have a filter constantly stuck to them, they'd come with - you guessed it - filters constantly stuck to them. Some of Canons teles even have an easily replaceable front element meaning a protection filter is completely unecessary.
 
trencheel303 said:
I agree (sorry Rob...) and despite what has been said, there HAS been numerous times where filters have been proven to degrade quality. Hoppy even took his own comparison pictures to prove it. I too can see their advantages but I don't see what the big deal is. If lenses were designed to have a filter constantly stuck to them, they'd come with - you guessed it - filters constantly stuck to them. Some of Canons teles even have an easily replaceable front element meaning a protection filter is completely unecessary.
Click to expand...

I'll let Hoppy tell you the full story, but his results weren't as conclusive as he'd hoped...

What and miss out on being able to chage us for filters and/or repair to front elements...

If I buy a second-hand lens and it comes to a choice between one that's had a filter on and one that hasn't... I know which one I'll be putting my money on...

Your lenses - you do what you like... I'll keep mine in good order thanks...
 
Well I wasn't able to get to the same location today, but I managed a quick (but, fairly conclusive IMO) test in the back garden.

Both of these taken at 300 f/5.6: (click for more detailed view)


WITH UV Filter:






WITHOUT UV Filter:





Now I don't know about anyone else here, but I can certainly see a big difference in bokeh between having the UV filter on and having the UV filter off.

In the first picture with the filter ON I can definitely see diagonal lines running bottom-left to top-right in the OOF bokeh areas. This is non-existent in the second picture without the UV filter.

I honestly did not think it would come down to the filter, but the results speak for themselves I suppose. Now, if only I hadn't paid £50 for the damn thing in the first place. :razz:

I guess in the future I'll take the filter off at such long focal lengths as I can't really see it affecting the image too much below 100mm.


Just as an aside, my ~15 year old lens of the same type (Nikon 70-300) with an equally old Jessops UV filter doesn't exhibit any of these artifacts so maybe Jessops filters aren't that bad after all - at least the ones that are 15 years old anyway!
 
Well...it's definitely the filter...lol

Although to be fair I can still see it - though not so pronounced - in the 'without' shot as well...
 
Just curious, you didn't buy that Hoya on ebay? Because they are very popular with counterfeiters and you may have been sold some cheapo knockoff...
I'm using a Hoya UV on the same lens as well and haven't noticed anything in the bokeh. Could be a production fault in the filter as well of course.
 
the filter it is :)

btw where did you buy the filter? I don't think a £50 Hoya UV filter would be that bad ..
 
Rob - yeah I can still see it a tiny bit, but it's nowhere near as distracting.

The filter was purchased from Jessops about a year ago - could just be a badly produced one that 'slipped through' - although IQ is fine with it on my 18-135 so I think I'll chuck it back on that lens. However for a £50 filter you should expect it to be a lot better than that.

Well at least it's not a misaligned element.
 
Basket Case said:
I honestly did not think it would come down to the filter
Click to expand...

I did.

Glad you ran this all the same. I get quite annoyed when folks buy expensive lens then start a "What filter?" thread... thats why God invented the lens hood :lol:
 
puddleduck said:
I did....Glad you ran this all the same. I get quite annoyed when folks buy expensive lens then start a "What filter?" thread... thats why God invented the lens hood...
Click to expand...

Some people have different priorities based on their shooting preferences...it's not for you or I to judge... I can understand you being annoyed if people insisted you use a filter against your wishes, but it's really none of your business what other people choose to do.

If I and others are happy using them (and I might add, with no bokeh problems at all on any of my lenses), then who are you to take such a self-righteous approach?

There's a hint of puritanism in your posts that sits ill with me I'm afraid...
it reminds me of the same flavour of attitude that insists that no post-production is necessary on real photographs, or that digital isn't real photography and only film will do...

Not using filters works for you, fine, I get it, so do we all...

We now have one instance with one lens-type (mid-priced zooms in the 80-300/100-400 range) where filter use is demonstably affecting IQ.

Hardly the end of the world is it...?
 
Arkady said:
Some people have different priorities based on their shooting preferences...it's not for you or I to judge... I can understand you being annoyed if people insisted you use a filter against your wishes, but it's really none of your business what other people choose to do.

If I and others are happy using them (and I might add, with no bokeh problems at all on any of my lenses), then who are you to take such a self-righteous approach?

There's a hint of puritanism in your posts that sits ill with me I'm afraid...
it reminds me of the same flavour of attitude that insists that no post-production is necessary on real photographs, or that digital isn't real photography and only film will do...

Not using filters works for you, fine, I get it, so do we all...

We now have one instance with one lens-type (mid-priced zooms in the 80-300/100-400 range) where filter use is demonstably affecting IQ.

Hardly the end of the world is it...?
Click to expand...



Or you could just have the good grace to admit you were wrong about this? Rather than inventing stuff that no one said (digital isn't real photograph) to defend your viewpoint?

I refer you to post #7 and then post #9 - you seem to be very Pro filter, and seem blind to the fact that these do damage IQ.

If you are going to quote me, call me a puritan is fine, but don't put words into my month to try to back this up.
 
redhed17 said:
After reading through the thread, I'm leaning towards the filter. There is differences between filters (obviously or they would all be the same price;)), and having a cheap one on a big zoom lens can show up the defects at maximum zoom of the filter. I had Jacobs/Cokin Polarizer on a Sigma 70-300mm and the camera was unable to focus at 200-300mm, :eek: and the image in the viewfinder was distorted. More extreme than what Basket Case has had, but it is an example of the Filter affecting the image.
Click to expand...
Well I never, looks like I may have been right. ;)

Arkady said:
We've established here and elsewhere that it's not a filter problem - it occurs whether a filter is fitted or not...try to keep up please - dropping posts like this in only serves to muddy the waters...
Click to expand...
Basket Case hadn't confirmed that it was not the filter at this point. So much for muddying the water. ;)

I was correct then! :eek: :D

Sorry, couldn't resist after being shot down so emphatically. ;)

Like I said, if you want to use a filter on a high zoom lens, you need quality filters because there is high magnification of the filter which will amplify any imperfections. Better to not use the filter sometimes maybe.

It could still be a lens/aperture blade/stabilisation problem in some cases too. Needs more tests by people who have similar problems.
 
Guys, this just came up to me, it may be stupid, but @ Basket Case, did you CLEAN your filter?
Those strange bokeh looks like fingerprint to me :thinking: :bang:
 
Oh yes, I made sure that filter was damn clean before I started testing. Just a dud one I think - although it looks perfectly fine on the 18-135. Just that the longer focal length of the 70-300 seems to 'magnify' the problem of the filter so that it is only now visible.

I think I'll just leave it off this lens now and stick with just the hood - I'd like to think I'm fairly careful when it comes to handling my equipment (ooh, matron!) so I doubt I'll put a scratch in the front element. Now having said that....
 
My my, it must be divine intervention that I somehow managed to miss this thread :lol: So it's the filter then, Rob :D

My take on this is that a) it's the filter that's the primary cause, b) it is a reflection off the surface of the sensor which then bounces directly back off the flat rear surface of the filter (which is why you don't see it with film), and c) it only happens with certain lenses in certain situations/subjects when there is a coincidence of various 'events' within the design and construction of the lens as the light travels up and down and back and forth again and again.

You only see it with long lenses as by definition they have a very narrow angle of view so the light is always very close to the optical axis. In this way, the reflections are directly overlaid as they bounce back and forth.

Lenses with already dodgy bokeh, one of the inherant design compromises, seem to be the culprits. This must be contributing factor. We see it a lot with consumer zooms and lenses like the Canon 100-400L because they are both slightly compromised and also very popular. Canon probably sells more 100-400L zooms than all off its other 300mm-plus lenses put together. I was unable to replicate this effect with my 70-200L 4 IS, plus 1.4x telecon, despite sticking a horrible filter on the front. But that is very different design to any consumer grade lens. The 100-400L also has a unique optical layout. The big Canon primes all have an integral protection 'filter' on the front, which is deliberately curved for exactly this reason. Canon's technical literature explains why.

All of which seems to fit as an explanation to me. I don't buy the interference patterns thing (and anti-aliasing should get rid of that anyway) nor anything inherantly to do with IS/VR. Take the filter off, problem goes away. Sure, you're still left with slightly odd bokeh from time to time, but that's just the basic lens design. I don't even think it's a 'cheap' filter thing although that obviously won't help (didn't the OP have a Hoya Pro-1 on?). It's basically the combination of two flat reflecting surfaces and a light beam that comes directly straight down the pipe.

On a related subject, I did some other tests with the 70-200 + TC combo and a cheap polarising filter. It was horrific. A dramatic reduction in sharpness - nothing to do with this bokeh effect. It was the poor quality polarising layer affecting sharpness (as distinct from the actual glass) and it just got worse as focal length increased. I couldn't really see it at all below 50mm. This is becuase long lenses only look through a small section of the filter and any defects are greatly magnified compared to shorter lenses. Fitting a high quality Hoya HD polariser fixed it completely.
 
Adam/BasketCase, as a double check on the explanation I posted above, does this bokeh phenominon appear in the viewfinder, or only in final images?

If it is a combination of both the filter and the sensor, then you should not be able to see it in the viewfinder. But when shooting in live view (or final images) you would. Equally, you wouldn't get it with a film camera.

If you could confirm this, that would be great. Or anyone else? Thanks! :)
 
puddleduck said:
Or you could just have the good grace to admit you were wrong about this? Rather than inventing stuff that no one said (digital isn't real photograph) to defend your viewpoint?

I refer you to post #7 and then post #9 - you seem to be very Pro filter, and seem blind to the fact that these do damage IQ.

If you are going to quote me, call me a puritan is fine, but don't put words into my month to try to back this up.
Click to expand...

I did: Post #54

The other stuff isn't related to this and it's nothing you have commented on directly, but it's something your post reminded me of... I should have made that more clear...

I am pro-filter...the fact that I can't reproduce this effect on my lenses indicates that while it is a filter problem (that's me admitting I was wrong, see?), it's limited to certain lenses only.
I get no noticable IQ degradation on any of the lenses I own (or have access to) when using a variety of filters, from old and new Hoya HMC UV to Nikon UV (old, very old) and new Hoya Pro-1 UV...

What I don't use are the type of lenses we're seeing this phenomenon mostly occurring with, which is why I took the stance I did.

Which is also why I'll still recommend thier use in certain shooting situations for those lenses that are seemingly (for now) unaffected.





...and since I hate being proved wrong about anything, I'm now going to sulk in a corner for a few days...:suspect:

pfft...tzzz...
 
So to sum up...

Its not a filter problem.
Its not a lens problem.

Its a filter and lens combination problem. That way everyone is happy. Unless your filter causes your lens to do that!
 
menthel said:
So to sum up...

Its not a filter problem.
Its not a lens problem.

Its a filter and lens combination problem. That way everyone is happy. Unless your filter causes your lens to do that!
Click to expand...

An alternative summary might be:

Filters cause problems.
They create flare.
They cause ghosting of bright light sources.
They cause bokeh problems with most popular long zooms.
They can reduce sharpness.
They are expensive and very profitable for manufacturers, distributers and dealers.
They are a con.


And because I'm enjoying Rob's self-flagellation so much, a poll:

Arkady took it like a man.
Arkady took it like a lamb.
Arkady took it lying down.
Arkady took it with good grace and humility.
Arkady choked on his dummy.
Arkady is a big girl's blouse.

:D
 
fark...grumble-grumble...I'm off to polish the car and waste more petrol...
 
HoppyUK said:
An alternative summary might be:

Filters cause problems.
They create flare.
They cause ghosting of bright light sources.
They cause bokeh problems with most popular long zooms.
They can reduce sharpness.
They are expensive and very profitable for manufacturers, distributers and dealers.
They are a con.


And because I'm enjoying Rob's self-flagellation so much, a poll:

Arkady took it like a man.
Arkady took it like a lamb.
Arkady took it lying down.
Arkady took it with good grace and humility.
Arkady choked on his dummy.
Arkady is a big girl's blouse.

:D
Click to expand...

I feel your summary comes from a less biased standpoint than mine though! ;)


Not that any of my lenses have filters on anyway.......
 
You must log in or register to reply here.
Back
Top