3

To be able to make videos, I purchased inexpensive ($25) Bower FN72 and FN58 filters. While I haven't expected anything great from inexpensive filters, the result was below my expectations, so I purchased a more expensive ($135) 77mm Hoya Variable Neutral Density ND3-ND400 filter.

I'm disappointed by the more expensive one as well (see the comparison below).

What I need is:

  • A sharp image,
  • Smaller color shift,
  • Reduced cross effect,
  • Less ghosts.

Is it worth spending $300 on a filter, or the results will be similar?

Here is what I get with Bower and Hoya filters:

1. Sharpness and color shifting

This is a cropped part of a photo shot with the sharpest lens I have (AF-S NIKKOR 50mm f/1.8G). Hoya performs well, but it could be better; Bower is terrible. Both Hoya and Bower shift colors toward yellow (white balance is set to a constant value across shots).

Open the images in a new tab to see a slightly larger version; all images contain EXIF.

With no filter, the image is sharp

Fig 1. With no filter, the image is sharp

Hoya filter slightly affects the sharpness of the image

Fig 2. Hoya filter slightly affects the sharpness of the image

Bower filter adds severe blur

Fig 3. Bower filter adds severe blur

2. Cross effect

This is a comparison of cross effect at 18mm. Both Hoya and Bower perform really bad at maximum and maximum-adjacent density.

No larger versions available; all images contain EXIF.

Original image

Fig 4. Original image

Cross effect when using Hoya filter

Fig 5. Cross effect when using Hoya filter

Cross effect when using Bower filter

Fig 6. Cross effect when using Bower filter

3. Ghosts

Without filters, the Nikkor 50mm f/1.8G produces a ghost

Fig 7. Without filters, the Nikkor 50mm f/1.8G produces a ghost

Hoya filter creates an additional ghost

Fig 8. Hoya filter creates an additional ghost

Bower filter creates not one, but two ghosts

Fig 9. Bower filter creates not one, but two ghosts

For Bower, notice two ghosts in the bottom left corner, as well as a ghost near the light source itself.

Comparison

                 Hoya                         Bower
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Sharpness           Slight blur                  Severe blur
Color shift         Important shift to yellow    Important shift to yellow
Cross effect        Severe                       Severe
Ghosts              Clearly visible ghost        Severe ghosting
Vignetting          Unknown¹                     Not noticeable
Uneven darkening    Noticeable in some cases     Noticeable in some cases

While the five times more expensive Hoya gives sharper images and less ghosting, the images are still blurry and ghosting still present. Moreover, the more expensive filter has no benefit in terms of colors shift and cross effect.


¹ Since I only have 77mm filter and my lenses are Ø58mm and Ø72mm, it is impossible to determine the level of vignetting.

  • 2
    You need to watch the "colour shift" thing; most of the "yellow shift" is probably just minus-blue from the polarization (you're shooting dielectric materials under a blue sky). And the cross effect is the result of using two polarizers at nearly 90% to each other; there's no way around it. – user28116 Aug 31 '14 at 18:52
  • Are you using a good tripod? If your tripod is flimsy, you could have camera vibration issues for the longer exposures, which would negatively affect the shots with the ND filter. – Fake Name Sep 1 '14 at 6:49
  • Why don't you get a fixed ND filter? a fixed ND filter will have less effect on the image because there's less glass and no polarization - and for the amount of money you are spending on a variable ND filter you should be able to get a set of reasonable quality fixed ND filters at different strengths – Nir Sep 1 '14 at 7:50
  • voting to close as a dupe but upvoting for a really nicely thought out and presented question. – Paul Cezanne Sep 1 '14 at 13:33
  • @FakeName: the tripod I use is not the best one, but this was not the issue here: the top part of the tripod was collapsed, shutter was activated by a timer and I've done similar tests of the same scene at night with a shutter speed of 5 s. with no noticeable motion blur. – Arseni Mourzenko Sep 1 '14 at 18:18
0

More expensive filters will typically use higher quality glass and coatings. Result is less distortion of the resulting image.
They may (and likely will) also have more true neutral tone, rather than introducing a slight colour cast.

As to whether the difference is worth the price you pay nobody can answer but you.
But I've never seen effects like you show using even cheap Cokin ND gradient filters, certainly nothing as severe. And those are cheaply made, inexpensive, filters.
Maybe if you go pixel peeking at very high magnification you see an effect, but for most practical use they've served me well.

  • I'm guessing your Cokin filters were fixed density rather than variable density? Fixed density filters are fundamentally a lot simpler than variable density filters and therefore should be better - certainly, you should never get anything like a cross-effect with a fixed density filter. – Philip Kendall Sep 1 '14 at 7:31
  • @PhilipKendall no, they were graduated density filters. Fixed gradient though. – jwenting Sep 1 '14 at 7:32
  • 1
    Sure; fixed gradient filters are still simple compared with any sort of variable density filter (which involves two polarizing layers). – Philip Kendall Sep 1 '14 at 7:35

Not the answer you're looking for? Browse other questions tagged or ask your own question.