Do you want to create inspiring imagery of the Milky Way like the image above? Who could blame you…. few things are as compelling as our galaxy. Popularity of Milky Way landscape photography increased greatly the past 10 years, particularly since 2012 as the low-light, low-noise performance of camera sensors has advanced (often referred to as high ISO performance). Furthermore, few genres will challenge your gear and image-processing skills more than Milky Way landscapes.
Yet, compared to landscapes taken in full or partial daylight, there are relatively few photographers that have delved into this genre. This makes it more challenging to learn the tips, tricks and gear needed to competently capture the Milky Way since there’s less information available to would be star photographers, and as you might guess there are few lenses on the market designed with the needs of star photographers in mind. What’s more, lens manufacturers never test and present results under conditions that replicate those when shooting stars (e.g. small, bright, high-contrast points of light over a dark background).
It’s an understatement to say the lighting and environmental conditions when shooting the Milky Way can push sensor and optical technology to their performance limits. Knowing this going into the genre will minimize mistakenly investing in equipment that may not perform up to your expectations.
Some of the most frequent questions I’m asked by would-be star photographers relate to gear, including lenses. Usually an inquiry will begin regarding what settings to use, and since many photographers don’t possess a lens as fast as needed to effectively capture the Milky Way, the discussion ultimately boils down to which lenses they should invest in. Unfortunately, as is the case with many gear questions I receive, the lens to use depends on a number of factors. In this tutorial I’ll provide the list of the lenses I feel are best to capture Milky Way landscapes as of early-2017 based on my experience, some objective testing done by some labs, and testimony from other contemporary star photographers.
But, before I do that, I think it important to give you some context, as well as additional background information that you can use to assess potential star photography lenses since in the coming years there will no doubt be many new offerings that might work well for capturing the stars. Armed with this information, you’ll be better positioned to sort the proverbial wheat from the chaff.
Caveats & Background
Use Low Distortion, Low Aberration Optics
Of course there is no such thing as a perfect performing lens, especially at short focal lengths. Ultimately, what lens you invest in to shoot the night sky will be determined by what level of imperfection you’re willing to tolerate relative to your reproduction goals.
My lens recommendations are the same for capturing northern lights and star trails (Figures 2 & 3), even though the approaches and exposure settings you might use for those can be different from shooting the Milky Way. However, to sufficiently capture northern lights or star trails doesn’t necessitate using the best performing lenses listed below since there’s often much more light in the scene (e.g. northern lights and star trails with moonlight) or you’ll be using longer exposures (star trails with or without moonlight), thereby allowing for smaller apertures that eliminates many aberrations produced when shooting a lens wide-open.
I’ll also only recommend the best performing lenses when shooting stars for purposes of fine art reproduction. When I say perform, I mean they meet my goal to capture as distortion free, tack-sharp star detail as possible with little to no trailing (blurring – relates to the fastness and focal length of the lens), low chromatic aberration (CA, also known as color-fringing), and relatively low astigmatism and comatic aberrations when using the widest apertures.
Of all the potential distortions or aberrations I want to minimize in my star photos, one of the most important is comatic aberration or coma. Coma is not a type of color-fringing, although it may be accompanied by it. When coma is present, detail is usually, but not always, rendered with sharp contrast in the center of the frame, transitioning noticeably to stretched-out, and softer contrast detail toward the edge (Figures 4 & 5) and usually appears as a comet or variation thereof, where it got its name from.
Other aberrations that can be just as distracting if not more so than coma include tangential (meridiontal) and sagittal astigmatism. Different lenses can generate different types and amounts of coma and astigmatism (Figure 5-8). In addition, spherical aberration, when it occurs, usually is found throughout the frame, and that sometimes happens with comatic aberrations on some lenses as well (Figure 9).
Other than spherical aberration, the effects of these other chromatic and monochromatic aberrations are most destructive at the corners and all of these lens anomalies are worse when shooting at wide apertures, especially wider than f/2.8. Coma and astigmatisms are most displeasing in astrophotography, where stars in the image corner (and elsewhere along the edges to a certain extent) appear stretched, squashed, oblong, or diamond-shaped like a bird in flight or flying saucer (Figures 5-8). The best performing lenses will still exhibit some coma and astigmatism when used wide-open, but they keep these aberrations to a minimum, and eliminate them completely from the majority of the frame. If using a good quality, fast lens (e.g. full-frame equivalent maximum aperture of f/1.2 to f/2.0), coma should largely disappear after stopping the lens down by 1 or two stops (Figures 10 & 11), although often they need to be stopped down further to completely control astigmatism.
Unfortunately, very few lenses control these aberrations well, and many of the top-rated, and more expensive, general photography wide-angle lenses in the world generate an unacceptable amount coma and astigmatism when used wide-open to shoot stars in my opinion, much to the chagrin of the uninformed novice photographer who bought a top-rated wide-angle lens for shooting the Milky Way.
For example, scroll down to the star photo example in the review of the Sigma 24mm f/1.4 DG HSM Art lens posted at The-Digital-Picture.com and you’ll see comparisons of both the Sigma 24mm and the Canon 24mm f/1.4L competitor. Both of these lenses are fairly expensive, and highly rated for general photography, but show excessive coma and astigmatism when used wide-open and stopped down by 1 exposure value (EV). Also, Figures 12 & 13 shows another example taken with a very popular, and expensive Canon EF zoom.
Now, yes you can reduce if not eliminate most coma, astigmatism, and essentially all spherical aberration by stopping a lens down (thereby negating the purpose for purchasing a fast lens in the first place), but you’ll need to be willing to spend a fair amount of time dealing with noise during post-production if you have to stop a lens down too much since to get equivalent exposures and maintain short exposure times to keep the stars from blurring, you may need to increase the ISO setting 4 to 8 times. Unless I take 8-20 exposures and stack the images (which is indeed a very effective technique to get clean imagery), I find the noise generated and overall results with ISOs above 6400 to be unacceptable, and generally don’t like to use settings above 3200.
The star photos shown in this tutorial, including the coma and astigmatism examples, were all taken with ISO settings ranging from 1,000 to 3200. When shooting the Milky Way you want to strive to improve exposure using wider apertures first, and only increase ISO as a last resort (see more on apertures below). As implied above, usually exposure times are already set at the maximum for the focal length involved to prevent stars from blurring, so longer exposures aren’t an option under most circumstances. Basically, star photo settings involve a trade-off among light transmission, star trailing/blurriness, lens aberrations and digital noise.
As camera sensors improve, star photography grows in popularity, and the expectations of star photographers become more demanding, we can surely expect better performing (albeit more expensive) low-light glass in the future, although in my opinion there is likely an upper limit to how far portable, rectilinear, spherical lens technology can be taken. But, if you’re serious about using the best glass available, you need to start saving now as such quality glass will take a large bite out of your pocketbook as noted below.
Aside from using metal as opposed to plastic parts and what can be a very expensive R&D process, what makes some lenses so expensive is the number and size of precision elements with expensive coatings, and constructed of special glass that can be more expensive per ounce than many precious gems and metals.
Another important caveat is my recommendations are also based on shooting with standard wide-angle (i.e. 24-35mm) or ultra wide-angle (i.e. 11-20mm) focal lengths that enable inclusion of at least some land in the scene….. after all they’re not landscapes if I don’t include any land in the scene. In my opinion, including some foreground, middle ground or background land makes star photos more compelling pieces of art (Figures 1-3 & 14-15).
It’s true, good Milky Way landscapes can be made with longer focal lengths, but doing so requires taking a great many overlapping frames and stitching them together during post-processing. For example, to get an equivalent ultra-wide 14mm angle field of view to capture the full arc of the Milky Way from horizon to horizon when using a 50mm lens might require taking 50 or more shots. I generally don’t want to spend that much time capturing a Milky Way landscape, and stitching together that many high-resolution images can be cumbersome. Some photographers do just that (e.g. see some of the great Milky Way panoramas by Grant Collier), but they still rely on fast lenses with minimal aberrations.
Also, if you’re more interested in deep-sky astrophotography you’ll most definitely need to use telephoto lenses, resulting in different exposure settings, and to prevent star trailing will need to shoot on a platform that tracks star movement (e.g. equatorial mount). Since I don’t have much experience with deep sky astrophotography I won’t be providing any lens suggestions for that type of work, but for those shooting Canon bodies, I have gotten superb performance when shooting a few deep sky objects with the faster (f/4.0 or faster) Canon L-series telephoto lenses.
With respect to magnification, my recommendations for Milky Way landscapes are restricted to lenses with focal lengths at or below 35mm. Specifically, I recommend using quality glass with focal lengths ranging from 14mm to 35mm for full-frame sensors, and 10mm to 24mm if shooting a smaller ‘crop’ sensor.
Just in case you’re wondering, I capture stars using a Canon EOS 5D Mark III & Mark IV, and over 90% of my Milky Way photos are taken with a prime 24mm or 28mm lens. If I want a wider field of view, and don’t have much subject movement (e.g. clouds or moving vegetation), I will still use these same lenses, but take multiple overlapping, vertically oriented shots and stitch & crop them into the final composition using Adobe Lightroom, Adobe Photoshop and/or PTGui Pro (Figure 15). Ultimately, creating a stitched image with a faster 24mm or 28mm lens gives me a better result (higher resolution, less distortion, less coma, and often better exposure), than if shooting one frame with a slower 14mm f/2.8 lens.
Maximum Aperture Size – Photographing Stars is All About Light
I can’t emphasize enough that anyone serious about shooting the Milky Way should invest in lenses with a maximum ‘effective’ (35mm full-frame equivalent) aperture of f/2.8 or faster. Fast lenses are even more important when shooting with a crop sensor since even 1 or 2 stops of extra light with wide apertures will help minimize the relatively large amounts of digital noise generated when shooting with small sensors at high ISO settings. Very few lenses actually allow in as much light as their maximum aperture implies. Therefore, it’s worth doing your research on transmission or T-values for a lens you’re interested in. For example, most lenses rated at f/2.8 actually transmit light equivalent to f/3.2 or smaller. Also, if using a full-frame lens on a crop sensor, the widest achievable effective full-frame equivalent aperture will actually be smaller (i.e. higher number, transmitting less light) than the lens is rated for when accounting for the crop-factor (e.g. an f/2.8 lens will actually be no faster than f/4.5 on a Canon APS-C sensor even though it’s set to f/2.8!). Of interesting note, this is also true of many lenses designed for crop sensors since many manufacturers don’t honestly advertise the lens’ specifications with the crop-factor conversion applied.
Thus, if shooting a small sensor, keep an eye-out for lenses with a published maximum aperture of 0.95 or faster. As of 2018, there are very few such lenses on the market, and none that perform all that well for shooting stars. Therefore, I won’t be reviewing many lenses designed for crop sensors. But, for some micro four-thirds bodies there are some brands, namely Mitakon and Voightlander, that produce f/0.95 lenses that will provide a full-frame equivalent f-stop of f/2.0 or so. Stopping these lenses down further to reduce aberrations should then meet the requirment of an effective aperture of f/2.8 or wider so that a sufficient total overal exposure will reach the sensor.
The reasons why a 35mm full-frame equivalent of f/2.8 is my threshold requirement for apertures is that the light transmission of most lenses is lower than the maximum aperture implies as noted above, and almost all need to be further stopped down a little to minimize the aberrations mentioned above. Furthermore, since exposure times need to be short enough to prevent stars from blurring, and ISO can only be increased to a certain threshold before degrading image quality too much by revealing noise in the dark tones, effective apertures smaller than f/2.8, and especially smaller than f/3.2, simply transmit too little light to the sensor to capture the Milky Way well. Therefore, unless or until sensors are made even better than today’s top rated full-frame cameras (i.e. with greater sensitivity, dynamic range, and signal to noise ratio), thereby allowing the use of smaller apertures for photographing the night sky, the best way to improve light transmission is to use optics that employ the widest possible apertures.
Obviously, what’s considered acceptable from an image quality standpoint varies with the photographer and their goals, but in general I’ve always been frustrated shooting Milky Way landscapes whenever I’ve needed to use apertures slower than f/2.8 to mitigate aberrations since the image invariably is too noisy for my tastes due either to under-exposure and/or needing to raise the ISO significantly. I’ve found this to be true even when shooting stars on new hi-resolution Sony sensors that I’ve rented (which contrary to what some claim, are not truly ISO invariant).
For example, and despite the fact that many astrophotographers shoot at extreme ISOs above 3200, as of 2018 no camera manufacturer has yet produces a digital sensor rated to perform acceptably well above ISO 4000 based on DXOMark’s low-light, high ISO testing standards. Although sensor technology has come a long way over the past decade, it usually takes about 10 years for sensors to advance by one effective stop of light with respect dynamic range (DR) and signal-to-noise ratio (SNR). And, in recent years the rate of advancement in sensor performance has slowed, suggesting we may be reaching the limits of what the physics of light capture will allow current technology to acheive. So, if physically possible, it’s likely going to be quite a while before we’ll see acceptable DR and SNR ratings approaching 6400 or above. This leaves us to rely on optics and other creative capture and processing techniques to get the most out of our star photo pixels.
But, novice star photographers have to start somewhere, and you’ll still learn much if experimenting with sub-optimal optics. In general, if you don’t need to meet the highest output quality standards, any lens meeting the focal length recommendations listed above that also has a maximum aperture of at least f/2.8 (regardless of aberrations) will enable you to capture neat photos of the Milky Way under the right circumstances, no matter what sensor you use. It really just boils down to how much aberration, geometric distortion, and noise you’re willing to tolerate.
Lens Type (Prime vs. Zoom)
As with other genre, using a prime (fixed focal length) lenses to shoot the Milky Way will generally provide better results than zoom lenses. However, there are some excellent wide-angle zooms that take great star photos, and regardless of quality, their flexibility is an advantage to consider. Ultimately, as with most things, your individual needs and goals will determine the type of lens you invest in.
Electronic vs. Manual Apertures
For the greatest flexibility and convenience when shooting subject matter during the day, it’s great to use a lens with electronic aperture control. However, some of the best performing glass for shooting the stars do not have this functionality (no electronic chip; which reduces the cost of the lens), requiring the use of the lens’ aperture ring.
Fortunately, the convenience of electronic-apertures isn’t important when shooting stars. In general, I usually set my camera to manual or bulb exposure mode, and often use only one aperture (e.g. f/2) all night anyway, so the manual aperture control on some of my lenses isn’t an issue for me. The biggest downside by far when using fully manual lenses that don’t have an AE chip is that the aperture setting and lens brand & model won’t be recorded in the image file’s metadata.
Auto vs. Manual Focus
As with electronic aperture control, AF capability is nice, and almost indispensable for certain daytime work (action, sports and wildlife subjects). Thus, if you can get a good star photography lens with AF (e.g. Zeiss Batis 25mm f/2.0 for Sony E-mount) you’ll have a superb all-around lens. But again, this feature isn’t relevant for shooting stars, and many of the best star photography lenses do not have AF capability. Importantly, even the best sensors in the world can’t auto-focus well at night, and I find it more precise, if not easier to focus the lens manually anyway; so don’t let manual focus dissuade you from considering a lens for shooting stars.
For the most part image stabilization (IS) isn’t needed when using focal lengths shorter than 50mm regardless of the genre involved, and that’s why we don’t find IS as a feature on many wide-angle primes and zooms, including most of the top star photo lenses. But, it’s even more irrelevant for photographing the night sky since exposure times are too long to shoot without a tripod. Nevertheless, there are some recent ultra wide-angle zooms on the market that include IS and if you use them you should disable the feature when shooting the Milky Way since it can introduce vibration when shooting from a tripod.
What About Consulting Formal Lens Reviews and Testing Scores?
Lens reviews abound on the Internet. Some are good, some not so. All too often you’ll see reviews based on word of mouth or regurgitation of information provided by the manufacturer that isn’t helpful at all in assessing whether it will perform well when shooting stars.
One site that provides objective test data on lens performance is DxOMark (Figure 16). The site’s extensive database enables comparisons among different lenses, as well as lens test scores when shooting different camera sensors. DxO Labs provides an overall lens rating, as well as separate scores on sharpness, transmission, distortion, vignetting, and chromatic aberration for the lens tested. In general I don’t find the overall rating very useful, and of the five individual scores, sharpness is probably the least useful as a star photographer. But, all the other test scores provide a fairly good assessment of a lens’ ability to transmit light efficiently without distortions or aberrations.
If using DxOMark to assess a potential star photography lens, you should look for those that have the best transmission and vignetting scores (although remember that light falloff is something that can’t be completely avoided with wide-angle lenses). But most importantly, you want to look for lenses that demonstrate high marks for lack of distortion and chromatic aberrations.
However, once you’ve narrowed down a potential list of lenses based on DxO Lab’s data, you still have some digging to do. Unfortunately, DxO Lab’s testing mainly targets general, all-purpose photography, and doesn’t test specifically for coma or other aberrations when shooting small points of light. For example, as of early-2016, the top 2 scoring 24mm lenses for Canon dSLR cameras are the Sigma f/1.4 Art, and Canon’s own EF f/1.4L II USM. While these lenses score well for lack of chromatic aberration on DXOMark, and can indeed be used for shooting stars, they nevertheless exhibit a horrible amount of coma and astigmatism in the corners and along the edges when shooting wide-open. The next lens in the list is the Samyang f/1.4, and in my experience generates more acceptable coma and astigmatism than the Canon or Sigma alternatives, especially when stopped down to f/2.0. Thus, once I’ve narrowed my list of potential candidates, I either have to borrow and test the lenses myself, or begin searching for more detailed online reviews where I can better assess the level and type of aberration generated by the lenses I’m comparing.
The best online reviews are those done by practicing photographers who have thoroughly test the lens under the most demanding circumstances. Even then, since relatively few photographers specialize in star photography, they will rarely test for or assess the degree and type of coma by taking star photos or using simulated points of light. If you have the opportunity to chat with an experienced star photographer who’s used the lens you’re interested in, you’ll likely get the best sense for how it performs.
Fortunately, The-Digital-Picture.com and LensTip.com have completed some of the more useful online reviews I’ve found in recent years. Full reviews on these sites are very thorough and usually address aberrations if reviewing a fast wide-angle or ultra wide-angle lens, often including at least one image demonstrating whether coma or astigmatism is a problem when shooting wide-open, and sometimes presents images taken at -1 and -2 EVs for comparison. In addition, they sometimes present comparisons with 1 or 2 competitors. The very best assessments will also test for spherical aberration throughout the frame, which can be a real problem for many lenses with apertures wider than f/2.8.
Please note that recommendations below aren’t necessarily presented in any order of quality or performance. Furthermore, the best lens for you may or may not be equivalent to the lens that I consider to be the best, as there are many factors, including overall versatility and other shooting circumstances to consider. Unless otherwise noted, prices in USD noted in this review are those listed at Amazon.com as of January 2016.
Since none of the lenses I recommend are truly cheap (e.g. <$100), and since you might want to compare multiple lenses before making a final decision, I strongly advise you borrow copies of lenses you’re interested in so you can actually test them with your camera body before spending a lot of cash. There are a number of online rental warehouses, two good ones in North America include BorrowLenses.com and LensRentals.com. I’ve found that LensRentals.com usually has a slightly larger selection.
So, without any further ado, immediately below are the makes & models of SLR lenses that I consider to be the best night sky performers. With appropriate adapters they can also be used with some mirrorless camera bodies. I’ve used some of these lenses, and the others listed are ones many of my contemporary star photographers recommend. Also, my recommendations are mainly for full-frame sensors, although the same lenses can just as easily be used on crop-sensor bodies, and when doing so some of the aberrations discussed above are reduced if not eliminated from the frame due to the crop-factor involved (but remember as pointed out above that light transmission across the crop-sensor will be lower, and hence noise levels higher due to slight underexposure or by the need to compensate with the ISO setting).
Zoom Lens Recommendations
In my opinion there are few acceptable star photography zoom lenses on the market today. The first two listed below are the only ones I feel quite comfortable recommending, and in general perform similarly. The last one listed just makes the cut to warrant consideration.
Tamron SP 15-30mm f/2.8 Di VC USD (Figure 17)
Released in 2014, this moderately expensive lens ($1,199) is a very good performer overall, especially for the price. Based on both DXOMark and Lenstip.com testing it is probably the best ultra wide-angle zoom lens for star photography on the market, and competes well on the low end of the focal range with many prime lenses of similar magnification. Although some might argue the Nikon 14-24mm f/2.8 presented below is better. The lens is extremely well built, but isn’t quite up to the more rugged and weather resistant build-quality of the Nikkor or Canon zooms below.
This lens performs particularly well at the lower end of the focal range, on crop and full-frame sensors alike, the variable focal length of the zoom is convenient, and it is a fully automatic lens with IS capability for broader usage outside of astrophotography. Although it should be noted that IS and associated weight isn’t needed for such short focal lengths. This model is widely available for Canon, Nikon and Sony camera bodies.
Unfortunately, the large front lens element prevents screw-on filters, so if you wish to use polarizing, ND, or grad-ND filters for other work, you’ll need to invest in special large filters and holders, which are much more expensive than smaller screw-on types.
As with almost all wide-angle lenses, there are some issues with coma and astigmatism when shooting this lens wide-open, although it is actually quite low and performs as good or better in this regard than most competitors, including many prime lenses. I purchased a copy to replace my Canon EF 16-35mm f/2.8L USM II for use during the day. But, I have absolutely no qualms shooting stars with this lens wide-open at night, where it beats the Canon equivalent hands-down, including the recently released EF 16-35mm f/2.8L USM III.
The biggest drawbacks of this lens are the low transmission, poor vignette scores, and relatively high distortion at the short end of the focal range. However, while the relatively low transmission is consistent across the focal range, it does transmit more light than the Canon and Nikon zooms. Excessive vignetting and poor transmission can be important if you underexpose a star photo, and if you need better transmission at slightly longer focal lengths you would be far better off investing in a fast fixed prime lens. As expected, this lens also isn’t as sharp as the best performing prime lenses of equivalent focal lengths, but it’s among the best in it’s class.
If you want to use just one lens as opposed to 2 or more primes, one at the best price possible for a zoom in this class, and one that performs exceptionally well at night, this may be the lens for you.
Nikon AF-S Nikkor 14-24mm f/2.8G ED (Figure 18)
Similar comments as with the Tamron just described, but my contemporaries and the review on LensTips.com indicate it suffers from slightly less coma than the Tamron on a full-frame sensor when shooting at the short end of the focal range. However, the difference really isn’t noticeable, and overall any difference is relative to the type of coma and astigmatism you prefer. Personally, I feel the Nikon generates about the same amount of aberration. Until the Tamron was released, this lens (first released in 2007) was top-rated in it’s class and is used by a great many astrophotographers around the globe. Like the Tamron, it is AF capable, and has electronic apertures for broader use during the day. It can also be used on Canon and Sony cameras with an appropriate adapter (although AF might not be possible).
This lens is fairly expensive ($1,900), and exhibits only fair light transmission and high vignetting at best when shooting wide-open at the shortest focal lengths, although some of these shortcomings can be mitigated during post-processing if you use the best rated full-frame sensors on the market. But because of this, it generally must be used wide-open when capturing the Milky Way, so completely eliminating coma and astigmatism further by stopping the lens down isn’t feasible. This lens isn’t as sharp as the best performing prime lenses of equivalent focal lengths, but for a zoom lens is exceptional, and the overall build quality is a little better than the Tamron alternative.
As with the Tamron zoom, specialized, and expensive filters are required if you need filter capability for other work. But, if you want to carry just one lens for star photography, especially if shooting a top-rated low-light camera sensor, and can afford the hefty price tag, this is definitely the lens for you due to its versatility, low coma performance & build quality.
Canon EF 16-35mm f/2.8L USM III (Figures 19 & 20)
The version II of this lens, while a significant improvement from the original offering, has always been a relatively poor astrophotography lens, and in general few Canon lenses perform well at night. However, I’m finally comfortable recommending the 3rd generation of this lens as a legitimate contender, especially for those who like shooting Canon glass.
To be true, Canon wide-angle primes and zooms have a long way to go before they can compete at night with the Tamron and Nikon zooms on my list, but given the relatively limited pool of ultra-wide angle zooms for star photography, this recent model (released in August, 2016) performs just well enough to make the cut. As with its predecessors, vignetting is significant, and transmission poor, but consistency in sharpness across the field of view at all apertures has improved dramatically. In this regard, it excels over the Nikon and Tamron alternatives. However, overall it’s center sharpness is no better than the competition. Most importantly though, comatic and astigmatic aberrations, while still significant in the corners, has improved greatly and can be easily combated by stitching multiple images. Nevertheless, coma in the corners, especially on a full frame sensor, is far worse than the Tamron or Nikon.
Build quality is excellent of course, it’s AF capable, threaded filters can be used, it weighs much less than the competition, and the improved consistency in sharpness will certainly make this lens versatile for all-around landscape photography. But, as the poorest performing zoom lens on my list, it also is the most expensive at $2,200.
Prime Lens Recommendations
Samyang/Rokinon/Bower 14mm f/2.8 ED AS IF UMC (Figure 21)
Almost all lenses marketed under these brands, including this model, are affordable compared to the name-brand lenses, as well as the higher end art models recently produced by Sigma. This lens, available since 2009, is an extremely good performer considering its price ($320), and has been the go-to ultra-wide angle rectilinear lens for a high percentage of star photographers. This lens is well-known for exhibiting almost no coma when shooting wide-open, which is very important selling point considering the light-falloff it suffers from. It does exhibit noticeable astigmatism in the corners when used wide-open, but it’s of a type that most star photographers can definitely live with. It was the lens I used most when first starting to shoot the stars, and one that I continued to grab to obtain ultra wide-angle views in just one frame until I eventually decided to replace it with the equally good if not slightly better performing, and much sharper premium f/2.4 version.
This is a manual lens (only a disadvantage if you plan on using it routinely for general landscape photography during the day). Since I purchased my copy, a version with auto-aperture control has been released for both Nikon and Canon mounts. The downside of the fully manual versions of these lenses is that EXIF data is not transmitted to the camera body. This requires keeping track of the lens and exposure settings used for each photo taken and later adding it to the image header using EXIF-editing software. While it may not be a deal-breaker for most, it can be frustrating to contend with if you take a lot of photos and require EXIF data for all your shots.
This lens generates a fair amount of geometric distortion compared to some of its competitors, but that isn’t a huge drawback since it is corrected using lens profiles during post-processing. The lens also suffers from a high amount of vignetting, relatively low light transmission when shooting wide-open (both of which are typical for ultra wide-angle prime lenses), and also generates a fair amount of chromatic aberration.
Depending on the shooting circumstances, the vignetting and poor transmission makes shooting wide-open barely feasible, and generally forces you to push the limits of exposure time (which could cause star trailing), and/or push the ISO (which will make for much more noisy images, especially in shadow regions). However, this won’t be as much of a problem if using a top-end full-frame sensor (provided your shooting in RAW which you should always do with star photos anyway), or willing to blend multiple exposures during post-production.
Generally this is a great performing lens, and relatively sharp for the price (although most competitors are sharper), but to get the most out of it you should use it with a full-frame sensor, try to use exposing-to-the-right (ETTR) technique to the extent possible, or use other techniques such as stitching multiple images to overcome the light fall-off and vignetting problems with this lens. But, to a certain extent the same caveat goes for the Nikon and Tamron zooms listed above.
Finally, as with the Tamron and Nikon zooms noted above, specialized and expensive filters are required if you want to use polarizing, ND, or grad-ND filters for daytime shooting.
Samyang/Rokinon/Bower 24mm f/1.4 ED AS UMC & 35mm f/1.4 AS UMC (Figure 22 & 23)
Released in 2011, these two lenses are very fast and bright. They perform better than the majority of lenses on the market today from the standpoint of coma, astigmatism and CA when shooting wide-open, but such aberrations are still present.
What aberrations are generated wide-open is reduced and fairly well controlled when shooting at f/2.0, which is still 1 full stop brighter than shooting at f/2.8. Most fast wide-angle lenses on the market have a maximum aperture of f/2.8, and many with a maximum aperture of f/2.0 need to be stopped down to at least f/2.8 to mitigate coma and astigmatism. The low aberrations of these two lenses at f/2.0 is a huge advantage when shooting the Milky Way.
Compared to almost all good alternatives, these are affordable lenses ($480-550). Both are also among the best wide-angle lenses on the market today in terms of relatively low vignetting and good light transmission when shooting at wide apertures.
As with most lenses marketed with these brand names, these two models are fully manual with associated drawbacks, although that may not be an issue if you treat them as dedicated astrophotography lenses. Since I purchased my copies, versions with auto-aperture control have been released for both Nikon and Canon mounts. These lenses can’t resolve nearly as much detail as other competitors, and are noticeably softer on the edges of the field of view. But, few competitors of the same focal length, and none at such a low price, can outperform the low coma and control of astigmatism of these lenses.
Unlike the previously mentioned lenses, these models will take standard, and much more affordable 77mm screw-on filters for daylight work.
One thing to be aware of regarding all lenses in the basic line by this manufacturer, is that there is a lot of performance variability among individual copies of the various models, probably due to the lower build quality compared to other brands. I’ve seen imagery from the 24mm f/1.4 lens on at least one good online review showing that some aberrations aren’t controlled well until stopping the lens down to f/2.8, but my copy has always controlled aberrations acceptably well at f/2.0. I’ve also come across other reviewers who had to exchange their original copies since they exhibited various alignment and focusing issues Also, the accuracy of the infinity mark on the focus ring is variable among copies. Therefore, if you invest in one of these lenses, I advise you test them thoroughly on stars soon so that you can exchange them for another copy if needed. I find the biggest issue when shooting these lenses wide-open is the fair amount of spherical aberration throughout the field of view on high magnitude stars. Some don’t mind that effect, but I prefer to avoid it so typically shoot my copy at f/2.0.
Considering quality and especially price, these two lenses, and possibly the 14mm f/2.8, are the ones I recommend above all else as the best all-around, yet very affordable Milky Way landscape lenses. You could get all 3 lenses for less than it costs to buy the Nikon 14-24mm f/2.8 (which costs even more if you don’t shoot a Nikon body since you’d have to also invest in an adapter) or the Canon 16-35mm f/2.8. What’s more, if you shoot the 24mm & 35mm f/1.4 models and stitch multiple images, you can achieve an ultra-wide field of view, but with superior image quality and improved resolution without investing in the 14mm f/2.8. In fact this is what I do in practice, and I find that I rarely need to shoot stars with any lens other than my 24mm and 28mm lenses.
Samyang XP/Rokinon SP 14mm f2.4 (Figures 24 & 25)
Hope springs eternal, and it will be interesting to see what the future holds over the next few years for Samyang/Rokinon. With the recent advent of their Premium or Special Performance lens line, designed to take advantage of the latest high-resolution sensors, and other than the general lack of AF capability, we could see Samyang/Rokinon supplant other manufacturers not only in affordability but also in overall optical quality & performance.
Enter the impeccable Samyang XP/Rokinon SP 14mm f2.4 premium-grade lens. Test versions of this lens, briefly mentioned above, have been in the hands of a few photographers since late-2016. Within the U.S. it first hit the scene in March or April 2017 under the Rokinon brand for $1,299, marketed as their Premium or Special Performance line, but is currently retailing at Adorama and B&H for only $799.
DXOMark has yet to test this lens, but fortunately Daniel Gangur at Gippsland Images has published an excellent review with images that provides a good sense of how well it performs when shooting the night sky. Christopher Frost has also posted an excellent video review of this lens on his YouTube page at https://www.youtube.com/watch?v=PMfshUJGZow. It is a manually focused lens, but unlike most offerings from Samyang/Rokinon, it deploys electronic apertures, and a significant improvement in overall build quality (all-metal construction) compared to other Samyang/Rokinon models, somewhat similar in appearance to the Zeiss Milvus and Otus line-ups (although significantly cheaper).
Based on initial reviews, I decided to invest in a copy to replace the previous generation 14mm f/2.8 lens. My experience thus far suggests this premium lens sets a new standard for low coma, especially of the night sky, outperforming its little sister, which was considered by many to be the best performing ultra-wide angle astrophotography lens on the market. In fact, I would say this model demonstrates close to zero comatic aberration when shot wide-open, and it appears to handle vignetting and geometric distortion much better than previous models. This is impressive considering the 14mm focal length and that it boasts a wider/faster maximum aperture of f/2.4 compared to its f/2.8 predecessor.
Samyang XP/Rokinon SP 14mm f2.4 (Figure 26)
For those really wishing that Samyang/Rokinon would produce a fully automatic version of their lenses, you now have some options available as of early-2018. Beginning in 2017 AF versions of some Samyang/Rokinon models have been available for Sony E-mount mirrorless cameras, and as of 2018 Canon EF mount shooters now have the Samyang/Rokinon 14mm f2.8 EF lens to choose from.
Unfortunately, as of March 2018 there are no independent reviews, nor rigorous testing that’s been done on the 14mm AF version, especially from any astrophotographers. However, based on the manufacturer’s MTF charts, this lens does appear to outperform the fully manual original version (14mm f/2.8 ED AS IF UMC). However, there are no test images available showing the level of astigmatism or coma. Also, MTF charts show clearly that this version of the 14mm can’t come close to matching the performance of the premium XP/SP 14mm f/2.4 mentioned above.
As of March 2018 this model is retailing for about $799 at B&H and Amazon, which is the same as the Rokinon branded SP permium version that performs much better. However, if AF is a necessity for you, and since the MTF data suggests it performs better than the original fully manual version of the lens (which is still considered one of the best ultra-wide astrophotography lenses), and since the price tag is much cheaper than the Sigma 14mm f/1.8 lens mentioned below, then this is likely the best ultra-wide option for most photographers to consider due to its combined price-point and all-around useability with built-in AF.
Using Samyang/Rokinon/Bower Lenses on Crop Sensors
Other Samyang/Rokinon/Bower lenses are rated very highly for the low CA and coma they exhibit, but most are only usable on crop sensors. The best image quality is provided by full-frame sensors, but if you use a crop sensor, I highly recommend you consider a crop-sensor lens from the Samyang/Rokinon/Bower line-up, especially if it’s a model with f/2.0 or wider maximum aperture. In that case, choosing models with focal lengths of 10mm to 16mm will get you reasonably wide fields of view as if you were shooting a full-frame sensor.
Sigma 14mm f/1.8 DG HSM Art (Figure 27)
This lens, announced in early-2017 and available late-spring, is by far the best performing model for astrophotography in Sigma’s art lens line up. Recent reviews of this lens have been posted on dpreview.com by Francisco Slgado, and lenstip.com.
The relatively low coma and astigmatism in the corners and edges of the full-frame compared to longer focal lengths in the art line-up is impressive considering how difficult it is to avoid aberrations in ultra-fast (e.g. faster than f/2.8), ultra-wide angle (e.g. < 24mm) rectilinear lenses.
However, while this model is definitely a worthy, serious astrophotography lens contender in the ultra-wide angle focal length range, and is more versatile than some competitors in this class due to its AF capability, its performance still can’t match that of the aforementioned Samyang/Rokinon 14mm or Tamron 15-30mm zoom models. Spherical aberrations are noticeable when shooting wide-open, and fully mitigating coma and astigmatic aberration requires shooting at apertures smaller than f/2.8. However, the comatic and astimatic aberrations in the corners of a full-frame sensor is similar if not a little better than the Zeiss Distagon T 15mm f/2.8, and on par, if not a little better than the Nikkor 14-14mm f/2.8 zoom.
This lens isn’t cheap, coming in at $1,599, about $400 more than the Tamron zoom, is much more expensive than the significantly better performing Samyang XP/Rokinon SP premium lens ($799), but is considerably cheaper than the poorer performing Zeiss Classic or Milvus 15mm lenses at $2,000 and $2,699, respectively. But, if you require a prime lens in the ultra-wide angle class with AF capability, then this is probably the best option on the market as of mid-2017.
Zeiss Wide-Angle Primes
Zeiss is known for making some of the best photographic glass in the world. But, no lens is perfect and many of the Zeiss wide-angles rate lower in some areas than other cheaper alternatives. But, the best-performing Zeiss wide-angle lenses are marvels of optical engineering.
Zeiss has stood atop the wide-angle lens market in terms of quality for decades with their Classic Distagon T models. With the advances in sensor technology, Zeiss is beginning to replace models in their classic line with the new Otus and Milvus models (wide-angle models are still based on the Distagon design) that will surpass the original Distagon T’s performance, especially in terms of sharpness. Unfortunately, few new Zeiss wide-angle models have been released as of this posting, so these are line-ups worth keeping an eye on in the future.
In addition to unsurpassed build quality, Zeiss lenses are known for their sharpness, as well as lack of distortions and aberrations. Importantly for star photographers, their top models are also relatively good coma performers, enabling them to be used wide-open without much concern. They also control astigmatism aberrations fairly well, although not as well as coma. Not coincidentally, and due largely to the Distagon design, Zeiss wide-angle glass is expensive, but in the end you tend to get what you pay for. However, except for the latest high-end Otus line, most of the Zeiss offerings are no more expensive than the top-rated Canon and Nikon alternatives.
Unfortunately, when used wide-open many Zeiss lenses are also known for lower transmission and higher vignetting compared to other brands (and is a big reason why I’ve not used many Zeiss lenses yet). This isn’t necessarily a problem if shooting the best full-frame sensors, but otherwise makes capturing & processing star photos a bit more difficult.
Depending on your camera body, a Zeiss lens may not be available with electronic aperture control, although all new Zeiss offerings seem to be employing electronic apertures. While that’s not necessarily a huge issue for this genre of photography, considering the cost, you might think twice since it may limit the usability or convenience during other times. But, fortunately, the lens mount does integrate with the electronics of the camera body to transmit EXIF exposure and lens data, as well as provide focus confirmation.
Unfortunately, as of this posting relatively few new Zeiss lenses are faster than f/2.8, and not many at all faster than f/2.0. But, as other manufacturers continue to improve the quality of their faster wide-angle lenses (e.g. Samyang XP/Rokinon SP lines) you can bet that Zeiss will have to adapt and offer faster alternatives of uncompromising quality.
One final aspect of many Zeiss wide-angle lenses that you should be aware of if you’re serious about getting one, especially the most recent models, is that the focus ring has a relatively wide angle of rotation. That could be a big deal if you want to achieve manual focus quickly on moving subjects during the day. But, if you prefer very precise, accurate focus, the longer focus throw is an advantage. This is especially important when focusing objects at night.
Below are my recommendations of the best Zeiss wide-angle models for capturing a starry sky as of early-2016.
Zeiss Classic Distagon T 15mm f/2.8 (Figure 28)
There simply aren’t a lot of good performing lenses within the sub-20mm focal range, and until the new Samyang XP/Rokinon SP 14mm f2.4 and Sigma 14mm f/1.8 DG HSM Art models were announced, this lens, released in 2012, was about as good as it got for ultra wide-angle lengths.
Overall its performance is similar to the old Samyang/Rokinon/Bower 14mm f/2.8, but is of notably better build quality, gets better chromatic aberration ratings, and is much sharper. Aberrations are well controlled and better than almost all other lenses at this focal length, but isn’t quite as good at controlling sagittal astigmatism as the various Samyang/Rokinon/Bower 14mm options in my opinion. You’ll also have to weigh the slightly better overall performance against its high cost (about $2,297). That’s why relatively few night sky photographers choose this lens over the other prime or zoom alternatives. However, as of September 2016 this lens has been discontinued and replaced with a comparable, more expensive ($2,699) model in the Milvus line-up so you might be able to find this classic version for a better price before supplies run out.
In addition, this lens doesn’t transmit light very well (but is similar to the Samyang/Rokinon/Bower 14mm f/2.8 in that regard), requiring the use of either longer exposures or higher ISO settings. The lens data on the new Milvus version doesn’t appear to perform any better in this regard. But, other alternatives that transmit more light wide-open (e.g. Canon EF 14mm f/2.8L II USM) don’t exhibit the low coma and low astigmatism that this Zeiss lens does. Finally, with large 95mm threads, using filters will take a big bite out of your piggy-bank if you need to use them for other work.
Within the sub-24mm full-frame sensor prime lens class, as of early-2018 I rank this as the 4th or 5th best performing ultra-wide angle lens for star photography, but without doubt the best in terms of build quality.
Once the new Milvus 15mm and 18mm f/2.8 models have been more thoroughly tested I’ll update this post, but in all likelihood I do expect they’ll perform as well or better than this now discontinued model.
Zeiss Milvus 21mm f/2.8 (Figures 29 a-c)
Figures 29 a-c. The Milvus 21mm f/2.8 lens replaces the previous Classic Distagon T version, and is a good contender as one of the better star photography lenses on the market.
This recently released lens replaces the Classic Distagon T 21mm f/2.8 lens which was a good lens in its own right, and performed similarly to the 25mm version listed below and demonstrated relatively low levels of aberration, although some anomalies are still noticeable. As with other Zeiss lenses, light transmission was fairly poor. Also, the sharpness of the classic lens was poorer than many of its competitors. Still it did a respectable job capturing the stars and as with all Zeiss lenses you simply couldn’t beat the build quality. As of 2016 you can still get the Classic version for about $1,508 to add this to your arsenal.
The newer Milvus version retails for nearly the same at $1,495. As of this posting no ratings were available on DxOMark, but when tested this lens will most likely rate much better than its predecessor. I did find a relatively good review of the Milvus version at the-digital-picture.com, and the coma is definitely acceptable, although not as low or of the type that I prefer. To fully remove the coma you need to stop this lens down 1 EV when it won’t be nearly as usable for star photography, but I wouldn’t have any qualms using the lens wide-open, especially on a crop sensor or when making a stitched image on a full-frame sensor.
Zeiss Batis 25mm f/2.0 (Figure 30)
Released in April of 2015 and as of March 2018 selling for around $1165, this model belongs to the Batis line-up designed for Sony full-frame E-mount mirrorless cameras (excellent sensors for shooting stars). As such, and unlike most Zeiss lenses, it is a fully automatic lens, including auto focus. I’ve been unable to find detailed reviews by star photographers, but DXOMark rated this lens in late-Feruary 2016, and it obtained some impressive ratings.
Since I don’t shoot with Sony mirrorless bodies (yet) I have no experience with it for shooting stars. But, if you shoot Sony A7-series or A9 bodies, you should keep an eye out for reviews or online star photos taken with this lens, especially since it proves to be versatile. I have seen some impressive photos taken with it from a Sony A7II, and in mid-March 2016 Zeiss posted some images on it’s Lenspire site taken with a Sony A7S that demonstrated very low levels of aberration. If you shoot Sony E-mount mirrorless cameras, you should definitely keep an eye out for future focal length offerings in this line-up.
Zeiss Classic Distagon T 25mm f/2.0 (Figure 31)
Available since 2011, this particular lens is relatively fast for a Zeiss offering, and unlike many wide-angle lenses can definitely be shot wide-open due to the well controlled aberrations it exhibits. Some aberration is still evident though, even at f/2.8, but while there are faster 24mm alternatives with better transmission, none of them can be shot wide-open without generating considerably more coma and/or spherical aberration than this lens exhibits. Although, some of the other faster competitors (Samyang/Rokinion/Bower 24mm in particular) do generate similar or less aberration at f/2 and f/2.8 than does this Zeiss model. Nevertheless, the very acceptable performance, in combination with its superb build quality makes this lens an excellent option when needing slightly more magnification, or if you plan on stitching multiple images.
One downside in addition to it’s cost ($1,375), is the considerable vignetting and lower transmission when shooting wide-open that limits the lens’ usability, although its scores in these areas are generally better than most other offerings by this manufacturer, and better than many lenses of shorter focal lengths regardless of brand. If you shoot the best full-frame sensors on the market and especially if shooting multiple exposures for blending during post-processing, this isn’t a terribly big issue.
I’m personally interested to see if Zeiss replaces this model with a faster one of similar focal length in the new Milvus or Otus line-ups, and if those offerings will demonstrate significant improvements in coma, transmission and vignetting. If so, I may well consider replacing my Rokinon 24mm f/1.4 with the replacement of this model.
Zeiss Classic Distagon T 28mm f/2.0 (Figure 32)
Released in 2009, this lens performs very similar to the slightly shorter 25mm version. This Zeiss model controls aberrations relatively well, but should be stopped down to f/2.8 for best results. It also generates a bit more chromatic aberration than the 25mm model. However it’s much cheaper at only $978.
Zeiss Otus 28mm f/1.4 (Figures 33-35)
Figures 33-35. The Zeiss Otus 28mm f/1.4, the pinnacle of wide-angle lens quality as of early-2018, and a lens that takes star photos wide-open with very low coma. At 28mm, this lens is particularly well-suited to taking multi-image stitched star photo compositions. Canon mount shown with electronic aperture control capability.
Released in early-2016, this lens is the cream of the crop…..as close as you can get to wide-angle lens perfection. In fact, it is touted as the best performing wide-angle lens in the world, and as of this posting the initial reviews suggest it may live up to the hype. However, that’s not to say that it doesn’t suffer from some aberrations….all wide-angles lenses do. This lens uses a baseline Distagon design, generates as close to no chromatic aberration as is possible when shooting wide-open at f/1.4, far better than any other standard wide-angle lens out there.
When I initially posted this review I had only seen one star photo taken with this lens wide-open, which exhibited impressively low levels of aberration. Because of this I decided to rent a copy for testing, and if I find the time will post either a review of the lens, or some test photos on this tutorial in the future. Thus far I’m extremely impressed with the Otus and it seems to be living up to the hype with some of the lowest comatic and spherical aberrations I’ve seen when shooting at f/1.4…..it’s a real game changer for serious star photographers.
While I’ve not done an objective sharpness test, this lens is by far the sharpest wide-angle I’ve used, and the build quality can’t be beat……it’s built like a tank. The lens does an admirable job controlling astigmatism aberrations, but they’re still present and to completely eliminate them you need to shoot at f/2.8 or smaller. There is some notable spherical aberration throughout the frame on the highest magnitude stars at f/1.4, but impressively this is well controlled by only stopping the lens down to f/1.6, and thus outperforms the Rokinon/Samyang 24mm f/1.4 lens, one of my favorite star photo lenses, which must be stopped down to f/2.0 to sufficiently control spherical aberration.
Of course it’s expensive, like all Zeiss offerings…no I mean OBSCENELY EXPENSIVE! As of the February 2016 release date it will set you back $4,990, approaching the cost of the best pro-grade f/2.8 prime telephotos on the market, but you can get used copies for just over $4,000. Although a wide-angle lens, it’s also a relatively large, heavy beast, with 95mm threads, and weighing in over 3 lbs. As such, if you want to use filters for this lens they won’t come cheap. But, if you already use 100x100mm or 100x150mm filters, Lee Filters makes a push-on adapter that fits this model perfectly. This model is only available in Canon and Nikon mounts as of this post. The Canon mount version has auto-aperture control, while the Nikon mount has a manual aperture ring, but the chip still allows recording of metadata.
As with most Zeiss lenses, this model can only be focused manually, and there is considerable light fall-off when shooting wide-open, although no more than most good performing f/1.4 competitors. This latter disadvantage is significantly mitigated, becoming quite acceptable when stopped down to f/2.0.
At a focal length of 28mm, using the Otus for single shots of the galactic center might be limiting in some circumstances (especially if using a crop sensor), although less so than when using a longer 35mm lens. Regardless, the Otus is fabulous for creating high-resolution stitched images, and no other wide-angle lens can match the sharpness or take better advantage of the latest high-resolution full-frame sensors on the market. Perhaps Zeiss will come out with a 21mm to 25mm version in the Otus line-up that will perform similarly, and if they ever do, it will be most sought-after (although I think the 28mm might well fit that billing).
Other than the hit to your pocketbook (a life-time investment), you can’t go wrong with the Otus. I was so impressed with the copy I rented that I finally decided to make the jump and purchase a copy. Before investing in it though, I recommend renting a copy as I did before taking the plunge, or purchase a copy through a distributor with a consumer-friendly return policy.
As of this posting DXOMark has yet to review this lens, but you can find two reviews at Camera Labs and The-Digital-Picture.com that provide more useful information than most, including addressing coma. If you simply need and can afford the best, and know how to use creative compositing techniques to achieve wider field-of-view compositions, then this may well be “THE” Milky Way landscape lens. It is likely to be one that all other standard wide-angles lenses will be measured against in the coming years.
Some General Purpose Alternatives
The following fixed focal length alternatives should work well for star photos, although don’t perform as well at night as the previously mentioned lenses. I’ve never used them, so am relying on information in other reviews, but will update the information presented here if I get around to testing them in the field.
I present them here as alternatives that come with the advantage of excellent build quality, are fully automatic for added versatility, and will take standard screw-on filters when needed for daytime work. Overall these three lenses rate extremely well for general-purpose photography on DxOMark.
Sigma 20mm f/1.4 DG HSM Art (Figure 36)
This 2015 released lens is similar to the Sigma 24mm f/1.4, but exhibits much less vignetting. It also exhibits less coma and astigmatism than many alternatives, including the Sigma 24mm, but based on the review at LensTips.com appears that it must be stopped down at least 2 EVs to mitigate these aberrations sufficiently which limits it’s usability for star photos unless you plan on making stitched images that will crop-out the aberrations when shooting at the widest apertures.
This lens performs a little worse than the Samyang/Rokinon/Bower 24mm, is more expensive ($900), but is sharper and has AF and electronic aperture capability. Nevertheless, along with the Sigma 24mm f/1.4, it is a legitimate star photo contender when used at f/2.8, probably the 2nd best performing model for star photos that Sigma has offered to date behind the recently released 14mm f/1.8 model.
As with the Samyang 14mm, Nikon 14-24mm, and Tamron 15-30mm options listed above, you’ll need special filters and accessories to use polarizing, ND and grad-ND filters for daylight work.
Sigma 24mm f/1.4 DG HSM Art (Figure 37)
Available since early-2015, and similar to the Sigma 20mm f/1.4 Art, but exhibits much worse vignetting by at least 1 EV, and must be stopped down to f/2.8 to control aberrations to a moderately acceptable level. This lens performs more poorly than the Samyang/Rokinon/Bower alternative, is also more expensive ($850), but renders detail more sharply and has AF and electronic aperture capability. Along with the slightly better performing Sigma 20mm f/1.4 Art and 14mm f1.8 Art, this lens is a legitimate star photo lens if used at f/2.8.
Canon EF 24mm f/1.4L II USM (Figure 38)
This offering (released in 2008) performs close to the Sigma 24mm. It exhibits slightly less distortion, but unfortunately generates more chromatic aberration. Coma and astigmatism in the corners is similar to the Sigma, requiring stopping the lens down by 2 EVs to mitigate. This lens also suffers more from vignetting by about 1 EV than the Sigma 24mm (2 EVs worse than the Sigma 20mm). This lens is far more expensive ($1,450) and performs worse than the Samyang/Rokinon/Bower alternative, although is more sharp throughout the field of view.
I hope you found some of this information useful for deciding what lenses to invest in for capturing the Milky Way. If you have any questions regarding the information provided in this tutorial, please leave a comment or contact me at ImagesByBeaulin@charter.net.
Now, go shoot for the stars!
© Beau Liddell, ImagesByBeaulin.com, All rights reserved.