Saturday 4 March 2017

Review -Viltrox Focal Length Reducer









The fact that I'm writing this review is the result of someones slip up, a happy accident and unexpected windfall for a mate.

Here's the backstory, a close friend ordered a Viltrox basic adapter to use Nikon Ai lenses on his Sony A6000. We got him a great price on eBay, about $32.00 as I recall, and the adapter promptly arrived almost exactly on the due date.  It sat in his bedroom cupboard along with a little Nikon 35-70mm zoom he bought at the same time for several weeks, then a few days ago he finally brought the camera/lens/adapter combo to the coffee shop so I could show him how to shoot using manual focus with his A6000.


Most folk would think that shooting in manual focus would be a simple affair that needed no training, well yes it would be for most people, but "The Edgar" is one of those people that needs a constant digital assistant to get through even the most unchallenging tech day.  The joke is that somewhere along the way he paid me an $18.50 training fee with a "whole of life free tech upgrade", and it's now my responsibility to sort anything that goes wrong, looks like it might be going wrong or just looks wrong!




It's nicely made and finished; the tripod foot is removable.

At first glance, I was a little bemused as the adapter came in a nice plastic box that looked, well pretty much the same as the ones that Metabones adapters ship in, in particular, my bemusement was due to the fact my own Viltrox adapters didn't come in any plastic boxes at all.

So I opened the neat little box and straight away I could see this wasn't just a regular adapter but rather it was a focal length reducer, heck right up till that minute I didn't even know Viltrox made such things.

My initial impressions were pretty positive, the adapter/reducer was well made and like my regular Viltrox adapters had a solid removable tripod foot. "The Edgar" was now a little perplexed as to why I was looking perplexed and said, "don't tell me they've sent me the wrong thing, that'd be right, I always end up with the wrong thing........"(ranting and internal raving ensued).

Calm down Ed, it's not what you ordered, but it's better than what you ordered....you've had a win, not a loss!

Well, it could have been a loss if it performed badly, but I wasn't game to mention that, it doesn't do to stress "the Edgar" when he's trying to relax over a cuppa. Trust me pretty quickly his stress becomes my stress and then pretty soon anyone nearby is stressed and then before you know it we all have to go out the back of the coffee shop to the bar and replace the coffees with beer or rum.

So about this cut-price focal length reducer, is it any good I hear you say.  Well since I don't own any focal length reducers it's a little hard for me to make a fully conclusive comparison. I've often been tempted to bite the bullet and buy a pair of Metabones Speed Boosters for my Sony E and M4/3 gear, but frankly, the price has frightened me, I mean six hundred....(choke, choke), I mean, man are those things expensive for what they are!  They may be good, great even I'm sure, but heck I could just buy another fast lens second hand for that sort of money.

I did a little price peeking on eBay, and it seems the Viltrox focal length reducer typically sells for around the $130.00 US mark for the EF to Sony E mount version with the simpler Nikon to Sony E job pulling in around $80.00 to $120.00 US.  

The Nikon is manual focus only and provides no contact with the lens, so no EXIF data is provided.  The Canon EF should, in theory, operate with all the regular options, i.e. EXIF, autofocus, stabilisation but as I understand it, there are many lenses that simply don't play the full ballgame with the Viltrox, so you'd best check that compatibility carefully if you're tempted.  Anyhow, I had the Nikon version to test and just as well as I don't own any Canon EF glass these days, though I do have a couple of Canon FD lenses and Viltrox have a version for these as well, though it's not as easy to find, oh and by the way you can also get a Minolta MD version.

Before proceeding I'd like to say sorry on two counts. One, none of these pics are great works of art, I just needed to get some quick images that tested the ability of the adapter and to demonstrate the possibilities it might offer for owners, so they are just pics of what is available close to home.

The second thing is that I did not take the same pics without the adapter for comparison. I did consider doing so but decided they'd be quite meaningless; you see exactly how would you make the comparison, the reducer changes the angle of view and effective aperture so direct comparisons are not possible. 

I can think of one way to make some reasonable comparison, but it would involve shooting with the adapter a 24 megapixel sensor Sony a6000 and then cropping out a section to match the framing on a 16 megapixels Sony NEX 5N then comparing the two images. Despite the complexity even that approach would not be a truly accurate comparison, and in any case I don't own an a6000.


On the other hand, I have a very deep level of experience with the Sony NEX 5N and have shot many thousands of frames using the same lenses used in this test I have a clear understanding of what the resulting lenses are capable of when shot without the focal length reducer.






Not a great pic, just a quickie in the front yard to give an idea of the bokeh you might expect, in this case, a Nikkor 50mm f1.8 wide open, so it works like 35.5mm f1.3.  It's a little busy I feel but actually quite OK, and certainly, there's plenty of separation.




Why use a focal length reducer?

Well, most folk in inter-web camera land and seem rather too pre-occupied with super shallow depth of field these days, which I suspect is a symptom of only looking at pictures as small web sized images on Facebook and Instagram.  The use of a focal length reducer can help with that little dalliance when you wish to use your lenses on smaller formats than their initial design intent. For most buyers I guess getting this shallow DOF look is probably their main aim but a reducer can do so much more than that.






Again the Nikkor 50mm f1.8 wide open and while not exactly creamy there's a strong DOF effect and reasonable 3D pop.  The purple fringing on the Hyundai sign on the back of the car is not from the adapter, this lens always displays this type of fringing on bright highlights unless it's closed down a stop and half or so.



More Light, Of course, shallow DOF is related to wider apertures and as we all know the focal length reducer gives the equivalent of an extra f-stop larger so, for example, your f2 lens now works like an f1.4 lens.  The upshot of this being you can extend your shooting envelope a little into the darker regions, which is nice.

But I think there are a few other benefits in using a focal reducer to consider, though possibly not all realised in reality due to possible build quality and design limitations.......

More Resolution:  Metabones certainly claim that their reducers, (the nicely named "speed boosters") can increase the resolution seen at the sensor, we'll see about that but in theory yes that should be an option and the difference would likely show up in particular on the denser 24mp APS-C sensors.

Field of View: The field of view of the legacy lens ends up pretty close to what it was on its original film body or full frame digital body.  That's a particular benefit for those who want to use legacy wide angle lenses on crop sensor bodies or make better use of the old 50mm standard lenses.

Easy access to maximum lens performance: Say what! Bear with me as I hadn't thought of this till I ran this little evaluation, but think about it, f8 passes light like f5.6 from an exposure point of view and most lenses are in their absolute prime even if they are not a prime at f8. You get to shoot at a shutter speed twice as fast, and you may get sharper results for two reasons, one the lens is more likely working in its design sweet spot or the use of a faster shutter speed may reduce camera movement a bit better or you can set your ISO one step lower.

Might Provide an option that doesn't exist otherwise:  In the case of smaller formats there are few options for very fast wide angles, there are no STF type lenses in M4/3 and Sony E mount (yet) and so on.  

Probably one of the best arguments for speed boosters on the M4/3 kit is the Sigma 18-35mm f1.8, it becomes a kind of hyper zoom in combination with the Speed Booster, in fact, it's beyond even that on the Blackmagic Cinema cam formats with the appropriate 0.64 version of the Speedbooster. Unfortunately, at this point, Viltrox doesn't do an M4/3 focal length reducer, but I suspect they soon will.

Added Options

You already have a lens or two, why not get a little more life out of them for not too much outlay. 

For example my Nikon 55mm f2.8 can be used on full frame, on my APS-C Sony gear as an equivalent to an 82mm f2.8 tele, and on M4/3 as a 110mm f2.8. A focal length reducers would gives me the option to also have the equivalent of a 58mm f2 on the Sony NEX and a 78mm f2 on the M4/3. 

Special Note: I have actually tried the Sigma 18-35mm f1.8 on the Viltrox with the Sony NEX, yes I know it's not meant to work as the Big Sig is actually an APS-C lens, but it actually does offer something a bit unique, there's a bit a vignetting as you'd expect but the central core is very sharp and even the edges are fine, and you can effectively get a 24mm f1.2, an 18mm f1.2 and even a usable 15mm f1.2 out of this combo.  Now that's a pretty unique alternative to say the least.  Sorry, no pics on this in this post but I think it might be worth a separate blog entry.







This pics is quite telling, it's one of the standard test scenes I use for testing all lenses, it includes lots of beautiful detail with distant and moderately close objects, high contrast and plenty of straight lines.  In this case the lens is the el cheapo 35-70mm Nikkor zoom (f3.3-4.5 Ai version) shot wide open at 35mm, which is equal to 25mm f2.4.  No it's not fully sharp to the corners, but interestingly with a little bit of extra sharpening the image ended up with plenty of cross frame clarity to make say a 14-inch wide print. The main negative issue is micro flare on the highlights, which you might just see on the white lights, but this is present when the lens is shot wide open at 35mm so it's definitely not an issue with the adapter.  






Now here we have the same scene (sorry about the slight framing inaccuracy) stopped down to f5, which is effectively f4, the whole frame is very sharp and only items right at infinity are ever so slightly soft due to DOF issues, but you'd need a very big print to see the softness and again its nothing to do with the adapter.




OK, so what about the deficits?

Well like I say I don't own any focal length reducers at this point, but I know enough about optics (and certainly have been able to confirm with testing this bargain basement Viltrox) to realise that there are a few niggling issues that may crop up for you.

You will most likely get an increase in lateral chromatic aberration.

You will probably see more vignetting than you are typically getting with your lenses when mounted on the regular adapter, probably quite a lot more in fact.

Since there is now an extra 4 elements of glass involved in the light pathway, it's entirely possible that overall contrast may be lowered a little, which by the way is not necessarily a bad thing in some instances, but the extra flare might cause grief under very low but highly contrasty lighting situations.







Here we have the 55mm f2.8 Micro Nikkor wide open (effectively 39mm f2), you might notice how the plane of sharpness has a little curvature to it, (the focus seems to come a little forward in the corners,) this is not the way the Micro Nikkor usually works so I assume it's a product of the adapter. What is rather nice is the bokeh, it looks pretty smooth in the full res images and to my eye a fair bit nicer than that produced by the 50mm f1.8 Nikkor.




Stop it down a couple of stops and the curvature is a non-issue and there is no lack of clarity.


But, What You Really Need to Know

Not all lenses are ideal candidates for working with focal length reducers! You say, sure that's obvious.  Well yes, it is but bear with me a little.  

You may have lenses you've used on your crop sensor camera that worked fine because the crop sensor eliminated the nasty bits of the lenses performance out towards the edges and corners.   Now you're going to be using almost all of the donor lenses coverage and probably uncover just how diabolical that lens actually was on the original film body or a full frame digital sensor, the adapter will certainly not cure the ills of a rotten old school stinker.

So the long and short of it, many lenses are just not going to be very good when mounted on the reducer because, well, they weren't very good lenses to start with and some lenses may also interact oddly with the reducer due to factors relating to their optical design.  Got it, that doesn't mean the reducer is not good but rather the original lens is going to be the ultimate limiting factor....in other words don't expect optical miracles that somehow deny optical physics.

On the other hand...lenses of relatively normal design that performed to a very high standard on their original bodies, be they film or digital, will most likely work pretty well on a reducer.  For example, if the lens had very little vignetting, very low chromatic aberration and was sharp right out to the corners then it's going to be an excellent candidate.




This is the full frame shot with the 55mm f2.8 Micro Nikkor at f6.3, which is effectively 39mm f5, even in this small size it looks very sharp, but check out the 100% crop below from the centre of the frame, this is resolving right down to the pixel level, if you had a 24 megapixel sensor it would be sharper again by a handy margin I expect.






Enough Enough, is the Viltrox Any Good

From my tests it seems to have the ability to be very good indeed, assuming you mount the right lens on it.

Allow me to pull that performance apart for you.



Build

Like I said, it's nicely made and comes in a neat plastic box.  The alignment seems fine and the mounting faces of excellent quality.  The action of the aperture control ring is really lovely and normally much nicer than using the aperture ring on the lens itself, though it won't tell you the actual f-stop you are set on.

The only blight, and a slight one at that is the tripod foot never mounts really tightly to the adapter body, this seems to be an issue across the board as my regular Viltrox adapters are the same and I've modded them so the foot is a nice tight secure fit.



Chromatic Aberration

Regardless of what lens you place on the adapter, it will increase or perhaps even introduce some degree of chromatic aberration.  Some of the aberration will have been in the original lens and you may never have noticed it because you were only cropping out the central frame, but I'm sure the adapter does add some of its own residual aberration regardless.

The CA is not a deal breaker, generally, it's easy enough to sort it in the Raw conversion software and up with very little or no CA in the final file.  Just to be clear the Viltrox does not introduce horrific levels of CA and having looked at a few tests with other brands of reducers online I'd say it is no worse than the best of them.



Vignetting

Yes, it absolutely does increase vignetting at wide apertures and even when the primary lens is stopped down quite a bit you can still see that the vignetting is stronger than it would be for the lens used in its native state.  The wider the angle of the donor lens the more obvious the vignetting.

Is it a drama? From my perspective, not so much as I often add vignetting in post and the levels introduced by the adapter are not so great as to cause major noise issues when fixed in post.



Field Curvature

I have no doubt now after having closely examining many images and especially images taken with the Nikkor 55mm f2.8 Macro that the reducer introduces a small degree of field curvature which means the focus on the edges and corners of the frame is closer to the camera than the central portion.  The shots with the Micro Nikkor prove the case for me as this lens normally has a very flat field rendering...like most macro lenses.

Is it an issue, maybe, but only at very wide apertures as it can affect the apparent cross field clarity of the image and the edges/corners will be a bit softer than they would otherwise be.

The good news, once you have stopped down a couple of notches the increased depth of field covers any disparity up.  Only those shooting wide open and expecting edge to edge, corner to corner clarity will have much to worry about.  In truth, if you're shooting wide open and going for a super shallow depth of field rendering it's unlikely the lack of corner or edge resolution will factor into your image, in fact, it could be a bonus.

In all my tests the curvature was not visible by the time the lens was stopped down to f5.6 which in practice is effectively f4 from an exposure perspective.



Bokeh?

Bokeh will largely be a product of the donor lens and the way it's designed but the reducer will have a sort of amplifying effect on whatever is present, so the takeaway is that lenses with great bokeh may end up looking a little better still, but those with rough, messy, agitated bokeh are probably going to create a bokeh result that looks a little annoying.





This is 50mm f1.8 Nikkor wide open and on a close-up subject, seriously could you ever really need a more shallow DOF rendering than this?




But this pic really tells me something, 50mm f1.8 Nikkor wide open, what really strikes me is the superb central sharpness on the metal stays, in the full res image the texture is fully rendered, I really do like the fall-off of sharpness in this one.




Distortion

As far as I can tell the adapter is not introducing any additional distortion into the images other than what's already present in the donor lens, I have not been able to conduct specific tests on this since sadly my Sony NEX 5n suffered a catastrophic shutter failure during the review process.  (Don't feel too bad it's done a massive amount of work and hey, now I can justify updating to a newer second-hand version, the 5T, which means I can still use all my accessories and DIY bits and by the time you read this my 5T will probably have arrived).

The shots were taken with the 55 mm Micro Nikkor make a good test case for distortion checking as the lens is pretty much devoid of any natively, anyhow I couldn't detect any distortion in the shots I took with it mounted on the Viltrox.  

Again I imagine if you end up with distortion it's most likely due to the adapter revealing the inherent distortion present in the full coverage of the lens when not mounted on the adapter.


And tell me about the resolution.

I have some good news here, really good news.  Assuming the lens you place in front of the adapter is very sharp to start with you may actually see that theoretical increase in resolution become a practical reality.

With all of the lenses I tried on the adapter the centre seemed to be a little better resolved, that might be enough for most peoples needs..... but bear in mind you have to stop the lens down a stop or two to see the improvement, you can't change the laws of physics and you certainly can't change the optical characteristics of the original lens.

The great news is that I found in the case of one lens I tested, the Micro Nikkor 55mm f2.8  (a well-deserved legend by the way)  the whole image once the lens was stopped down to the optimal aperture was stunningly sharp.  I mean not just sharp, I mean cut your fingers with a razor sharp and sharp right out to the corners with it.

I must point out that your mileage may and probably will vary a bit as I shoot only RAW and convert the files in Iridient Developer with a very finely tuned process I've developed over many years, if there is detail there I will get it with this fantastic application.  

But.... the fact remains the reducer with the right lens at sensible and appropriate aperture settings can and does deliver images with increased detail and overall clarity.  

Phew, that has surprised me and is justification enough for me to buy my own copy of the little adapter. 






So try this one on for size, Micro Nikkor 55mm f2.8 at f8 (effectively 39mm f5.6), now check out the right lower corner crop below, you can easily see every blade of grass! The centre crop is below and is obviously scalpel sharp.








Just a bit on the aperture

Now, remember that you are getting one stop more light on the sensor, so that means that f11 kind of works like f8...got it.  Ok so that means you can stop the lens down more than you otherwise would, but you don't seem to get the same diffraction effects you normally get unless you go to the very smallest apertures.

The thing I noticed was that if I stopped the lens to f11 I almost universally got excellent results with all lenses and most focal lengths and there was very little diffraction softening at all.

The side benefit (and I appreciate this might be somewhat hard to grasp) was that I could reduce the radius of the sharpening tool in Iridient Developer and slightly increase the edge detail settings and radically improve the micro detail setting then liberate some extraordinary finely detailed 3D like images.

My overall point here is that optical and sensor interactions are quite symbiotic, it's often tiny issues that make the difference but take it from me, with the right lens and raw converter settings, yes, you really can get better detail!






Just for fun, I took a shot the spider web between some rocks, wide open on the Micro Nikkor (effectively 39mm f2), now just check out the 100% centre crop below and take a peek at the way the spider, web and seed trapped in the web are rendered! 





Finally, we have a boarded up door shot at f5.6 (effectively 39mm f4), the textural rendering in the 100% centre crop is superb, I doubt I can get more resolution than this from the 16mp Sony NEX sensor








So the summary

For the money I reckon this is a very good product, the results were better than I expected and at this price, you can afford to have one even if you are not going to be using it regularly.

There are no real downsides, provided you mount the right lens on it, and for me the fact that it extended the performance envelope of my favourite 55mm Micro Nikkor is enough to make me buy one, sadly, I guess I will have to pay the regular price and I've already had to pay for that replacement Sony 5T body.









1 comment:

  1. It's good reading, thanks! I didn't knew Viltrox, but interestingly, they make an EF/EFs adapter to m4/3 ... much cheaper than Metabones, but ... can it be any good?
    Frank.

    ReplyDelete