Saturday, July 11, 2009

The Vivitar 500/8 Mirror Lens

I recently bought a Vivitar 500/8 mirror lens (also sold as Samyang, Phoenix, Opteka, Quantaray, and perhaps other brands). Although it's not a great lens, it's not nearly as bad as some say, once you figure out how to use it. So I thought I'd share some of my experiences. Here's a sample image:

First, though, some background.

If you use an SLR (digital or otherwise) and have ever tried shooting birds or other wildlife, you've probably thought about getting a longer lens. And chances are you've seen ads for "mirror" lenses (aka catadioptric, aka reflex) that seem too good to be true - under $100 for a brand new 500mm lens that's barely bigger or heavier than the "kit" zoom that came with your camera. If you've read up on the topic at all, though, you know that mirror lenses come with some major disadvantages: a small and fixed aperture (usually f/8), poor contrast and sharpness, and very odd-looking donut-shaped "bokeh" (the appearance of out of focus areas).

Now, every once in a while you come upon someone who manages to get good results from a mirror lens, and they'll often steer you toward the more expensive ones, saying the cheap ones are to be avoided. But if you're like me, you can't help but be curious. After borrowing one of the "good" ones (the Sigma 600/8) for a few weeks, I found that mirror lenses were not all bad, although they do take some work to get the hang of - both in use and in post-processing. Here's a sample image that I like from the Sigma:

I'm not going to try to convince you this image stands up to viewing at 100% on my 10MP Pentax K200D as well as an image from top quality glass telephoto lens. And if I'm able to get close enough to shoot with at 200mm or 300mm and still fill the frame the way I want, I can get better results with my 200m prime, or my 50-200 or 70-300 zoom. But if I end up needing a teleconverter and/or having to crop significantly in post-processing to get what I want with one of these shorter lenses, the Sigma mirror lens wins over any of the other options.

Unfortunately, this particular model was discontinued long ago and is kind of hard to find and relatively expensive when you do find it. Worse from my perspective, it is bulky for a mirror lens, and around three times the weight of one of the smaller models. It might seem petty to complain about the size and weight of a mirror lens, because at around 2 lbs even the heaviest are much smaller and lighter than a regular telephoto lens of that focal length would be. For me, though, that Sigma was in a bit of a no-man's land. If I'm going to carry a 2lb lens that doesn't fit in my camera bag, I might as well carry around something even bigger that will do the job with fewer compromises. But will I normally bother to carry any lens that big? I know I'm much more likely to actually use a lens that comes in at under a pound and fits in my bag.

There are other discontinued models from Tamron, Tokina, and others that are considered among the better mirror lenses and are probably worth seeking out, but they may also be harder to find, more expensive, and heavier than one might prefer. Frankly, my experience with the Sigma just made me that much more curious about the cheap models one sees advertised everywhere.

There are basically two 500mm mirror lenses currently on the market, although they are sold under a wide variety of different brand names. There is a 500/8 (normally black) made by Samyang and sold under that brand name as well as Phoenix, Opteka, Vivitar, Quantaray, and probably others. Then there is a 500/6.3 (normally white) that is usually sold under the name Pro-Optic but sometimes Kenko and perhaps others. Both are T-mount lenses, meaning that with a proper adapter, they can work with any camera mount. Usually, these lenses are sold with an adapter for a specific camera, but be aware that you can actually buy one for any camera brand if you don't mind picking up an adapter for your own camera separately (only $15 or so at your local camera store), and you might get a better deal on a used one that way.

I thought for a while about which of these lenses to try. Although the f/6.3 sounds better than f/8, one thing I learned from the Sigma is that DOF is so shallow at extreme focal lengths that a fixed f/6.3 as opposed to f/8 could create more problems than it solves. Plus the f/6.3 version is twice the weight of the f/8, and somewhat more expensive to boot. So I became fixated on the Samyang-made 500/8.

Unfortunately, as you also may have discovered in the search process that presumably brought you to this article, there just aren't many detailed reviews or even sample images from this lens, under any of its various names. What I managed to find didn't seem to resolve anything for me - some folks who hated the lens, others who thought it was surprisingly good, and only a small handful of sample images to support either opinion. Most of the references to the lenses came from people who never used either but dismissed them out of hand. Which left me still wondering how one of the Samyang-made 500/8 lenses would work for me.

I knew from the Sigma that one shouldn't expect stellar results right out of the box. Many people have little experience with manual focus, and the small fixed aperture of most mirror lenses means the viewfinder is quite dim, making focus harder than usual. Many do not realize that long telephoto lenses will create very shallow depth of field (DOF), and that avoiding camera shake at those focal lengths can be very difficult too. Many people don't realize they really need to use a hood with these lenses, or that they are supposed to always use the small clear filter that can be inserted in the rear of most mirror lenses. Many are put off by the donut bokeh that is inherent in all mirror lenses. Many see the low contrast and sharpness on the images straight from the camera and don't realize just how much this can be improved via post-processing.

Therefore, it doesn't surprise me that some think even the better mirror lenses are "junk". But more importantly, if someone reports a bad experience with one of the cheaper models, I take that with a grain of salt.

So as I said in the beginning of this article, I did finally pick up a Vivitar-branded version of the lens. I bought it used, as even the $99 brand-new price tag seemed like too much of a gamble. My assumption was that there are a lot of these lenses on the used market that are perfectly fine but are being being sold by people who just don't "get" it.

I'd say the lens has pretty much met my expectations. Not as good out of the box as the Sigma, but capable with a little work of producing results "almost" as good. As with the Sigma, the results are not nearly as sharp as they would be if I were able to shoot from closer with one of my other telephoto lenses. But depending on the resolution of your camera and the quality of your other telephoto lenses, you may still find that mirror wins if you'd need a teleconverter or heavy cropping to get the composition you want at 200mm or 300mm. Here's an (appropriately post-processed!) example I like from one of my first outings with the Vivitar:

But unless you're careful, it's just as easy to get terrible results that would make you assume the lens was not capable of anything like that. So let's look at some of the things that can make the difference.

As I alluded to earlier, technique is a big part of it if you are not used to manual focus or extreme telephoto lenses in general. Depth of field at 500mm and f/8 is very shallow, so you have to really nail the focus or the image will be out of focus. Fortunately, the focus ring on the Vivitar is very smooth, with good resistance and a long throw. If your experience with manual focus is limited to the "kit" zoom lens that came with your camera, you will probably find the actual mechanics of it easier with the Vivitar, even if the viewfinder is dimmer because of the fixed f/8 aperture.

Holding a camera steady with a 500mm lens can be more difficult than you might imagine. The old "1 / focal length" rule is probably on the optimistic side in this case, especially when used with a "cropped" sensor that makes the field of view narrower than it would have been on the 35mm cameras for which that rule was invented. A good tripod helps a lot. But one of the advantages of a relatively small and light lens like this is portability. And many people wanting that in a lens would be reluctant to use a tripod or even a monopod, even though the fixed f/8 aperture means you may often be looking at shutter speeds below 1/500".

For those wanting to handhold the camera with this lens, systems like Pentax, Sony, or Olympus that employ camera-based stabilization provide a significant advantage over Canon or Nikon. This is true even though camera-based stabilization might not be as effective as lens-based stabilization in general, and even though camera-based stabilization systems tend to be less effective at longer focal lengths. But any stabilization is better than none. I find it awkward, imprecise, and often misleading to express the effectiveness of stabilization as being "X" number of stops better, but I'd say that without stabilization, shutter speeds slower than 1/500" are very likely to be noticeably blurry for me; with stabilization, I have a fair chance at a decently sharp picture at 1/90".

If you don't have camera-based stabilization but still wish to handhold the camera, see if you can find a rock, table, or other surface to rest the camera on, as this can control shake at least as well as a monopod. Otherwise, you will need to increase ISO to get shutter speeds above 1/250" (and preferably to 1/750"). You will want to experiment for yourself to find what shutter speeds you need to get get the results you want. Unfortunately, while the aperture is technically f/8, you will find the lens absorbs so much light that shutter speeds are more consistent with what you'd expect of f/11. I find that on a typical sunny day, ISO 400 can give me shutter speeds of 1/500" or better in direct sun but not in the in the shade, and shooting on an overcast day is similar to shooting in the shade. So you might often be looking at ISO 800 or even 1600, depending on how steady your hands are.

Some mirror lenses come with hoods, but this one did not. When shooting anywhere near the direction of the sun, lens flare is pretty unmistakable - simply using your hand to shield the lens from the sun yields an improvement that is obvious just looking through the viewfinder. I bought a cheap 72mm collapsible rubber hood that might not eliminate as much flare as a hood designed specifically for the lens, but it works better than nothing. Still, it is better to avoid shooting in the general direction of the sun.

The famous donut bokeh is going to be more troublesome in some situations than others, and that requires some experimentation to figure out. Small bright highlights create the most obvious donuts. But the prominence of the donuts also has to do with how far you are from your subject relative to the out of focus areas. In a lot of shots - such as most of the ones I've posted here - I am barely aware of it. It gets more prominent the closer the background is to your subject. A very distant background will be so blurred you will probably not notice. And my impression is, the Vivitar seems less prone to donuts than the Sigma in any situation.

Even if you are careful about all of this while shooting, you will probably be disappointed with how the images look straight out of the camera. Contrast, sharpness, and saturation are all relatively low, and your camera probably doesn't have strong enough settings to fully address this. You will almost certainly be needing post-processing to improve this - and you will be needing to increase these parameters far more than you are probably accustomed to. Also, when using high ISO and then increasing contrast and sharpening in PP, noise becomes even more visible than usual for that ISO. You may need to experiment with both the noise reduction and sharpening settings in your PP software to combat this. I find that doing NR as usual for the ISO but setting a higher than usual threshold for sharpening and being relatively aggressive with amount and to a lesser extent with radius works well for me.

When applying this much adjustment in PP, you can get much better results from RAW than from JPEG. I have created a preset that increases contrast and sharpness while adding a bit more vibrance too. By making this a preset, I can easily apply the same settings to all images shot with this lens in one click. In a matter of seconds, the results improve dramatically. Of course, I can also custom process images after applying the preset - there are adjustments that are best made on a per-image basis using levels and curves adjustments, local contrast enhancement, hue adjustment (color is often cooler than I would like), and so forth. But just applying my basic preset instantly makes the results from the Vivitar much more impressive than they would otherwise be.

When you are careful to observe this advice about shooting and post-processing, the Vivitar can definitely beat the results you might get using a 50-200, 70-300, or another similar telephoto lens, and then cropping to yield the same field of view as a 500mm lens. Of course, you'd hope this would be so, or there wouldn't seem to be any reason to use the Vivitar if you already own another telephoto lens. On the other hand, there is something to be said for actually seeing your subject large in viewfinder - particularly if you enjoy just looking at wildlife. And in situations - like with birds in trees, or birds in flight, even - where you might find yourself focusing manually even with an autofocus lens, having the subject appear large in the viewfinder can help. So even if the Vivitar did not actually beat the results from cropping an image from a shorter focal length, it wouldn't be without its appeal.

But if you need to be convinced, here is a comparison. First is a 100% crop of an image from my 50-200 at 200mm. That means it is blown up as big as it can be without getting "pixelation". Following that is a corresponding crop from the Vivitar, resized to match. The original image is the one shown at the top of this article:

If you look closely, you will see the Vivitar image clearly contains more detail. I should point out, though, that this is apparent only in cases where you need to crop beyond the field of view of a 500mm lens as I did above, or in cases where you wish to print the full image larger than 4x6". If I crop the image from the 50-200 to exactly match the full image from the Vivitar and then print both at 4x6", the results are practically indistinguishable. So don't completely discount the value of cropping. It's only on larger prints or heavier crops where the Vivitar wins.

Still, there is no denying the fact that the Vivitar is capturing more detail, or the fact that it provides a much larger viewfinder image, which is nice in itself. And mirror lenses can be surprisingly versatile. Here's a portrait made at that same location (Arches National Park) minutes earlier:

I imagine the most common use for this lens will be for wildlife, though, and birds in particular. I don't think I have quite reached the potential of this lens for that purpose yet. But here is a shot that came out reasonably well considering it was handheld at 1/180":

I don't particularly recommend this lens for concert photography. 500mm is usually far more than you want, and f/8 (more like f/11 in practice) is usually much too small an aperture to get a fast enough shutter speed to stop subject motion. But I have observed that in most venues where you might need very long focal lengths (200mm or more), the stage lighting is often good enough that you can get away with smaller apertures than one needs in smaller clubs. I tested this idea recently with the Vivitar at a concert in a 300-seat recital hall, shooting from the rear balcony. At ISO 1600, all I could manage was a shutter speed of 1/45". But I'm used to shooting at that speed with my shorter lenses, so I know that stopping subject motion is possible with good timing. By resting the camera on the balcony railing, I was able to get the camera steady enough to take this shot:

For large outdoor concerts, I could see this lens being quite useful.

Now, I won't lie - I have taken sharper landscapes, portraits, wildlife pictures, and concert shots with my 200mm and 300mm telephoto lenses. But I cannot stress enough that in order to get those results, I had to be much closer to my subject. If I tried shooting from the distances I was dealing with in these examples, I would need to crop the image very heavily in PP to get the same field of view. And as the comparison I posted above suggests, the Vivitar is capable of giving better results than simply cropping an image from a shorter lens. BTW, I find teleconverters usually give results that are not even as good as cropping, although this depends on the specific lens and TC involved.

On the other hand, if you don't have enough light to get a fast enough shutter speed to stop subject motion or camera shake, a mirror lens is not the best answer. Again, neither is a teleconverter, since they will reduce most telephoto lenses to f/8 or worse as well. Cropping a shot taken at 200mm or 300mm can be a very viable way to get a usable image as long as you don't need to print too large. But mirror lenses like the Vivitar are a fun and effective way to get the shot without cropping.

In conclusion, I hope I have set your expectations for what this inexpensive mirror lens can and cannot do, and how to get the most out of it. A lot of people are extremely disappointed by this lens, and of course, it does not come close to matching a high quality 500mm glass telephoto lens. But I do think most of the naysayers are not taking full advantage of the potential of the lens - shooting with questionable focus and stability, without a hood, and not enhancing for contrast or sharpness or color. It is capable of taking pictures you are not likely to get any other way from a lens of this price, size, or weight.