Posts Tagged ‘sharpness’


Battle of the VariNDs: What’s the Best Variable ND?

Friday, June 14th, 2013


(Caution: this blog post contains serious camera geekery. If that’s your thing, read on!)

Pros have long known that Neutral Density filters are really important for shooting video on DSLR cameras (because they let us open up the lens to nice wide apertures and get that nice shallow depth of field). The most convenient way to use ND filters on DSLRs is the “VariND,” or variable ND filter, whose filtration can be adjusted with a simple twist of the filter ring. The problem with variable NDs though is that up until now they’ve usually been quite soft and tend to degrade image quality. Recently though a number of companies have released new models of VariND filters designed to address the softness. I am a stickler for tack-sharp imagery, so when I heard good things about a couple of the new VariNDs, I wanted to test them, to see which one was the new king of sharpness. After a thorough testing process I had a winner, but it was NOT a result I expected.

The two new variable ND filters that have come out recently that I had heard good things about were the Genustech Eclipse and the Tiffen Variable ND. Take a look at the video below for my intro to my testing.

Here’s what the test shots looked like:

©2013 Chris Conti Photography. All Rights Reserved.

If you’re interested in doing tests like this yourself, you can download the full-resolution test chart that I used here (it’s also handy for testing focus, adjusting autofocus microadjust, etc!).

I originally planned to stop both lenses down to f/11 for the tests to make sure that I eliminated as much softness coming from the lens as possible, so that I could really isolate softness being introduced by the filters. After conducting all the tests at f/11 though and inspecting the RAW files on my editing machine, I have to say I was completely blown away with the results: I couldn’t see almost ANY softness introduced by EITHER filter even at maximum filtration strength!

Even though the lens aperture shouldn’t matter in terms of the filter sharpness, I decided at that point that since I would likely be using these filters at wide apertures, I might as well test them at wide apertures. So then I opened the lenses up to f/2.8 and ran through all the test shots again. Again, the results were astonishing. Take a look at the 1:1 crops of the center of the resolution charts below:

©2013 Chris Conti Photography. All Rights Reserved.

Remember, these are 1:1 (i.e., full size) crops of a 21 megapixel frame! Both the Genus and the Tiffen were set at maximum ND filtration (or close to it), and compared to the shot with no filter, there is almost no perceptible softness. The jagged lines of the resolution chart are due to the resolution of the laser printer I used to print the chart (which was printed at 300dpi!). While neither of the filters introduced any significant softness, they both added about 20 points of green tint (thankfully though, that is very easily correctable either in the camera via custom white balance or in post). Also, while the image above is of the 200mm test, the results on the 24mm test were the same: nearly zero perceptible softness.

So Were the Two Filters Equal? NO.

Even though both filters were amazingly sharp (gone are the days of soft VariNDs! Hooray!), they were NOT equals. The Tiffen Variable ND was the clear winner. Here’s why:

As I said in the video, as a result of their opposing-polarizer design, ANY VariND filter is going to get the cross-type pattern of doom eventually. Here is what the pattern looks like, if you’re not already familiar with it:

The cross of doom. ©2013 Chris Conti Photography. All Rights Reserved.

The difference between the Genus and the Tiffen was that the Genus started displaying the cross pattern above substantially earlier in its filtration range (in other words, had less usable range) than the Tiffen. Compared to the Genus, the Tiffen had an extra stop or two of usable range before it too started showing the cross pattern. It is a subjective judgement how much cross-pattern darkening is acceptable before the image becomes unusable, but using the same lighting and camera settings, it was clear that the Tiffen had substantially more range (I’d estimate about two stops) before it started exhibiting the same level of the pattern above as the Genus.

The Tiffen Variable ND Wins

So the test concluded that both the Genus and the Tiffen were tack-sharp filters. Congratulations to both manufacturers for achieving that. The Tiffen filter though is the hands-down winner. In the end, there were plenty of reasons why: In addition to the added usable ND range, the Tiffen was both ergonomically superior (the build quality is better; the rotating ring is very nicely dampened and almost feels fluid whereas the Genus rotates somewhat loosely) AND less expensive ($150 vs. $165 for the Genus). The Tiffen is also easily available at Amazon, whereas the Genus is harder to get: the company distributes its products in the United States only through a small handful of distributors (and my local distributor, Rule Boston Camera, told me they couldn’t even get the Eclipse), and I ended up ordering it directly from the company in Australia.

So there you have it. It seems, for the time being at least, the Tiffen Variable ND is the sharpest, best variable ND filter you can buy… and it’s not even very expensive. Well done, Tiffen.

(P.S.- If you want to have a look at the RAW files to see the results for yourself, I’ve put them all online. You can download them here (be aware though that it’s about a 600MB download))

UPDATE: Do NOT buy the Genus filter from the Genus website under any circumstances. Here’s why: After performing this test and finding the Genus Eclipse Fader more expensive than the Tiffen and with a smaller ND range, I went back to the Genus website where I had originally bought it to return the filter (the company’s website clearly spells out their return policy, which I had checked before purchasing the filter, which says that Genus accepts returns of items within 30 days of delivery for full refund). I couldn’t find any way on the website of initiating the return, so I emailed the customer service email address listed. After hearing nothing back for several days, I contacted the company’s Twitter account asking about how to do a return. The person in charge of the Twitter account got back to me relatively quickly saying they’d check in on it for me. After another couple of days I got a response to my original email from an individual named Mark asking why I wanted to return the filter (this itself was slightly unprofessional as any company of any size has enough experience that they simply accept returns without demanding to know the customer’s reasoning first). I cheerily responded to Mark that I had conducted the test described here and found another competing filter to be better, but that had read Genus’s returns policy on the website before I bought it in the first place. After another day Mark then angrily emailed me that if I wanted to return the filter not only would I have to ship it to an address in Hong Kong (which alone essentially prevents returns, since the cost to ship anything there is about half the cost of the filter itself), but “all costs relating to the sale” would be deducted from my refund, including Paypal fees (Genus doesn’t accept credit card payments), shipping (which I’d already paid upon purchase), an “inspection” fee and a restocking fee (note that he didn’t tell me how much any of these fees would be). I then emailed back that none of those fees were mentioned in the company’s published returns policy. At that point Mark then emailed back and said “Until we receive the unit back and have the unit checked we cannot determine whether the unit is damaged or not.” My clear interpretation from that statement was that despite the fact that the filter was in immaculate condition, when it arrived Genus was going to claim it was damaged.

From this exchange it was very clear that despite their published returns policy Genus was not accustomed to refunding customers’ purchases and were not inclined to give me my money back. So I can only recommend that readers stay far, far away from Genus’s store, either for VariND filters or for anything else.

Tags: , , , , , , , , , , , , , , , , , , , , , , ,
Posted in Uncategorized | 14 Comments »

Zaza Gallery Canvas Photo Print Review

Wednesday, July 7th, 2010

So about a month ago (jeez, it’s been a month already…!) I wrote that a company called Zaza Gallery that makes canvas photo prints had offered to give me a free canvas if I would review the product here on the blog. The technology that drives many of today’s photo products is evolving so rapidly that I’m always interested in hearing about and trying out new vendors, so I was happy to take them up on it.

Their initial offer was for a 16″ x 20″ photo canvas. I shoot on Canon cameras, whose sensors are built with a 3:2 aspect-ratio frame (meaning the width of the image is 1.5x the height) and like many pros, whenever possible I use the entire frame when composing my shots (this is a good practice, as it maximizes the sensor area that you’re using for your final composition, thereby maximizing the image quality). As a result, the 16×20 canvas was a different aspect ratio (5:4) than my intended composition. I brought this to the attention of the company, and they generously offered to instead provide me with a 16×24 canvas, which matched my images’ 3:2 aspect ratio.

Zaza directs that for best quality, the image file that customers provide for printing have a resolution of 300 dpi in order to preserve detail in the final print. This is good, because in order to achieve the great detail of true professional-quality prints, high resolution is essential. For a 16×24 print though, this works out to 4800×7200 pixels, or approximately 35 megapixels, which is a higher resolution than even the best pro cameras widely used today (there are a small number of exotic systems that can achieve this resolution natively). What this means is that to make a Zaza canvas print properly, photo-editing software such as Adobe Photoshop or Lightroom is needed to increase the resolution of image files (known as “up-resing”). Since up-resing can result in nasty pixelation, artifacting and other image degradation if not done carefully, images that will be used to print high-quality prints such as canvases must be originally captured in the highest resolution possible.

In any case, for this test canvas, I chose to use one of my more recent photos (featured previously on the blog as the first Photo of the Day, here) which I called “Daybreak”. Here is the photo:


I chose to use this photo because it would really put the Zaza printing process through its paces: With its heavily saturated colors it would test Zaza’s ability to color-match, the absolute blacks in the silhouettes and sky would test the ability to achieve true black, and the smooth fade-to-black in the sky would test Zaza’s ability to print smooth gradients.

I prepared the image file according to Zaza’s specifications, with the appropriate resolution, format (Zaza takes standard JPG files) and embedded sRGB color profile (color profiles are essential for accurate color reproduction) and sent the image off. In about a week and a half (which is a normal turnaround time for canvas prints like this) I received the finished canvas. It was packaged well-protected, in a cardboard box in a plastic bag covered in bubble-wrap. Here is the finished canvas:


Zaza offers a number of different wrapping style options – a traditional “gallery wrap,” in which the image extends beyond the edges of the frame and continues on the sides, white and black wraps, in which image extends only to the edge of the frame and the sides are white or black, and finally “mirror wraps” and “blur wraps,” which are the best of both worlds: the image extends only to the edges of the frame (meaning the image is not cut off), but the edges are colored either by a reflection of the edge of the image or a blur of the edge of the image (which is nice so that the sides, if visible when hung, have some color and look like a real gallery wrap). I elected for the blur-wrap style. I haven’t seen this option with other canvas print vendors, and it is really nice. Your image doesn’t get clipped, but you still get nice coloration on the sides of the frame.

The quality of the final product is very good. The frame is sturdy and the canvas is stretched quite taut and stapled very securely. As far as the print quality:

-The color reproduction is very good. The colors matched the file I provided, and the saturation and vividness are excellent. Neither over- nor under-saturated.

-The blacks are truly black, the white truly white. Overall, contrast is excellent.

-Detail sharpness is average. On extremely close inspection I can make out a bit of fuzziness in the details, but this is to be expected from a file that was up-res’ed. And in any case no one viewing the print on a wall will get close enough to see the level of detail that I was inspecting. No complaints here.

-Like nearly all canvas prints I’ve seen, the print reflects a moderate amount of glare light, so care must be taken in regard to where the canvas is hung to avoid glare light. But again, this is common for canvas prints.

So there you have it, that’s my review! This canvas will hang proudly on my studio wall. Good job Zaza!

Tags: , , , , , , , , , , , , , , , , , , , , , , , , , , ,
Posted in News | No Comments »