Tilta Mirage Matte Box and Variable ND filter review

The most frustrating issue I have with any DSLR type cameras like the Sony A series or the Blackmagic Pocket 4K is that they do not have internal neutral density filters. Especially when shooting documentary style work external ND’s are cumbersome and time consuming to use. I was very exited to hear about the Tilta Mirage matte box with variable neutral density filter as it seemed to be just what I was looking for; quick to put on and take off with very easy density adjustment.

The build quality is amazing and feels solid with an adjustment ring that is smooth as butter, but unfortunately the VND filter optics are lacking.

I’ve used many different kinds of ND’s, single 4 x 4″ ND filters and a few variable ND’s. For the past two years I’ve been shooting two long form documentaries and I’ve been using a Tilta Mini matte box with a Heliopan Variograu slim variable ND on my Blackmagic Pocket 4K camera. The Heliopan VND is great quality, but especially in the winter when you have your gloves on it’s quite difficult to attach and detach from the lens; it can sometimes get a little stuck because of temperature changes. In comparison the Tilta Mirage with the variable ND filter is extremely fast to change and very easy to adjust. But as I found in my tests there is a drawback though with the image quality. Normally I don’t make reviews of gear that I don’t actually use but this time I thought it was important to put this review out so that you can make an educated decision if this product is for you or not. Let’s have a look at the comparisons and the tests I made.

Variable neutral density filter tests and comparison

Tilta VND’s scale goes from 1 to 9 stops. This is quite a large scale. Most variable ND’s come in much smaller scale such as 1-6 stops because at larger stops the image quality is usually negatively effected. The x-pattern is an inherent problem with all variable ND’s and can not be completely eliminated. The Polar Pro Basecamp matte box VND system uses two separate VND filters, one from 2-5 stops and one from 6-9 stops ND’s, to help minimise the issue. Usually the x-pattern is quite visible at higher ND stops on VND filters. I would personally never use any VND at the 7 – 9 stop range as this creates a very nasty visible x-pattern on the image. Unfortunately based on the tests I’ve done it seems that the Tilta VND doesn’t really perform that well even at lower stops.

Above is an image without any ND’s for reference shot at f14.9 iso 400. You can see the blue sky is very uniform and this is apparent on the scopes as well. All the images were shot with my Blackmagic Pocket Cinema Camera 4K with the Metabones EF – BMPCC4K T Speed Booster XL 0.64x. The lens used was the Canon 24-105mm f4 L at the widest at 24mm. With the Speed Booster XL this gives you a full frame equivalent of a 31mm and the aperture speed is increased by 1 1/3 stops. That is why the fastest aperture is f2.6.

Above are two images to exhibit the worst case scenario and to demonstrate what the x-pattern problem is with variable neutral density filters. I’ve increased the iso to 3200 to be able to get both the Tilta VND and the Heliopan to their maximum ND which is 8 stops on the Heliopan and 9 stops on the Tilta. Both are unusable to my eye, but this is to be expected and happens with all variable neutral density filters. So to be clear I would not expect any VND to work at these stops. Below we start looking at real world scenarios at levels you would actually want to use these VND’s at.

Above you can see both the Heliopan VND and Tilta VND at about 1 1/2 stop ND with the aperture at f8 and iso at 400.  Both images look good to my eye although the Tilta VND exhibits a little dip in the center. It’s not very obvious though and you have to look at the the scopes to see the x-pattern. This is a very usable image with both VND’s so no worries here.

Above you can see both the Heliopan VND and Tilta VND at 2 1/2 stop ND with the aperture at f5.6 and iso at 400.  The Heliopan shows somewhat of a dip in the middle on the scopes but it still looks very good to my eye and is absolutely usable. The Tilta VND is starting to exhibit more of the x-pattern and the already the usability begins to be questionable. We are only at a 2 1/2 stop ND so it is not looking good for Tilta.

Above you can see both the Heliopan VND and Tilta VND at 3 1/2 stop ND with the aperture at f4 and iso at 400. The Heliopan is starting to show a little dip in the center in the scopes but it still looks ok to my eye. The Tilta VND is now exhibiting quite a bit of the x-pattern. I would already be hesitant to use the Tilta VND at this level.

Above you can see both the Heliopan VND and Tilta VND at 4 1/2 stop ND with the aperture at f2.6 and iso at 400. Both clearly show the x-pattern on the image but the Tilta VND is definitely worse here. I find that the Heliopan x-pattern is more centered as well which gives it a little more leeway and is easier to correct in post. I would still give the Heliopan a usable check mark here. On the Tilta VND the x-pattern is not in the center which makes the problem more noticable. It’s more of a diagonal dark patch that cuts through the image unevenly. You can see clearly in the scopes as well that the Tilta VND passes much more light from the left side than the right side.

Tilta states on their website that: “The [VND] module has built-in hard stops between ND 0.3 ~ ND 2.7 to ensure no risk of an X-Pattern Cross Shadow. Based on golden ratio design and optimized color rendition, to maximize the ND function our variable ND has a larger axis between the filter elements. To prevent vignetting, use ND 0.3-2.1 [1 to 7 stops] for wide lenses and use ND 2.1-2.7 for focal lengths beyond 35mm.” In essence Tilta recommends you to use 1 to 7 stops ND for lenses wider than 35mm and only use the 8-9 stops beyond 35mm.

I did these tests at 24mm (full frame equivalent of 31mm) so it should look good up to ND 2.1 (7 stops). I don’t find this is accurate though. In my opinion the Tilta image quality is already negatively effected at much lower ND levels starting at 2 1/2 stops, and from 3 1/2 stops onwards I would just not be comfortable using it. Of course it is very rare you would just film a completely uniform sky so these examples and this test of course accentuates the problem. In normal shooting situation the problem would be less visible. But if I have to choose between the convenience of the Tilta and the quality of the Heliopan VND I will definitely go with quality.

The Heliopan VND that I was hoping to retire but proved to have superior image quality.

Below are all the test images side by side in small size so it’s easier to compare.


The modularity and ease of use

Let’s put aside the quality of the Tilta variable ND filter for a while as the rest of the product receives nothing but praise from me. The modularity of the Tilta Mirage is a huge advantage compared to other products. You can go all in and use it with the matte box or have the lightest and smallest setup with just the VND attached to the lens and nothing else. You can disassemble almost all the parts separately from the matte box. It’s almost like you have the Revoring, Polar Pro Basecamp and a Tilta Mini matte box all in one.

The packaging is great and the accessories that come with the Tilta Mirage are all excellent quality. I love the fact that you have the choice and can mount the matte box to rods or directly to the lens.

You can either attach the matte box directly to the lens with the included adapters 67mm, 72mm, 77mm, 82mm or to rods with the included rod mount. I use the 77mm adapter and connect it directly to the lens because I’m mostly using a zoom lens that extends quite a bit, so anything connected to the lens must move with it. The Tilta matte box and the VND are incredibly lightweight so the lens can hold they without any problems even when fully zoomed.

I paid 457 € for the  Tilta Mirage matte box and the VND so it is not the cheapest product. I did not purchase the motorised version (that kit costs 742 €) as I do not really have the need for it. Compared to the set up I have now (a 100€ Tilta Mini matte box and the Heliopan VND that cost 160€) it is quite a bit more costly but the convenience of being able to quickly take the VND in and out is worth it. The Tilta Mirage matte box alone costs 150€ which is a very good price. I own the Tilta Mini Matte Box and I would recommend this over that one for sure.

It would have been great if there was a way to use 4 x 4″ filters with the Tilta matte box. As of now you can only use 4 x 5,65″ filters. I guess you could pretty easily make a 4 x 5,65″ to 4 x 4″adapter though. Another small inconvenience is that the rotating adjustment wheel of the VND does not have any indication at what stop it is at but I don’t see this as a big issue.

My documentary rig set up with the Tilta Mirage Matte Box and variable ND on the Blackmagic Pocket 4K.


The Tilta Mirage matte box with the VND is pretty much a perfect product concept and I had very high hopes for it. The modularity is amazing and the build quality is top notch all around. But the variable neutral density filter image quality is what matters the most so unfortunately that is a deal breaker for me personally. The Tilta Mirage Matte Box that you can buy separately is a great deal however your existing filters fit into it.

Scroll to top