[colonel00]

Just looking for some input from folks that have owned/used the Samyang/Rokinon 24mm f1.4. I recently picked up this lens, mostly for nightscapes/astrophotography but this copy has me a little concerned. I've read a bit that it could be a little soft wide open but some quick, preliminary shots seem to be worse than I expected. I will get some more examples tomorrow but this is what I see with just a few quick shots of stars. f1.4 looks really, really soft to the point of out of focus. Stopping down obviously gets better but if I have to go to f2.8 to get decent result, that kinda negates the usefulness of this lens as I have a 14mm f2.8 that has been great for me. I just wanted something faster.

I guess I'm just looking for opinions on whether this is just the limitation of the lens or if it's a bad copy.

In the examples below cropped pretty much 1:1, f1.4 and even f2.0 to a point seem way too soft and blurry. The little halo/bubble thing that is formed just adds to the mess. These were shot at 13s so there is a little star trailing which I'm fine with on a non-cropped image.

Edit: I should also note that nothing was changed between these shots aside from aperture. Focus seems to be pretty good and on the infinity mark.

These aren't fantastic images by any means and there is a lot of light pollution around here. I just wanted to get a quick test after getting the lens.

Examples (untouched aside from cropping):

f1.4



f2.0



f2.8



---------- Post added 05-20-17 at 02:35 AM ----------

Here are the full images so you can get some context of the effect of the issue. I used Lightroom's Auto white balance adjust these. It's not as noticeable but still not something that I'm happy with I don't think.

f1.4- Not horrible from afar but still "fuzzy" and just not quite right.



f2.0 - Better, almost tolerable but not there yet.



f2.8 - I can accept this but again, this voids the speed of the f1.4 lens.

---

Last edited by colonel00; 05-20-2017 at 12:40 AM.

[clackers]

My 24 has always been sharp, Colonel. Like every lens you should have tested it against a brick wall as soon as you got it instead of waiting for astro shots.

Is it too late to exchange for a good copy?

[colonel00]

Nope, not too late. Brick wall test is coming up soon.

Though, I've always done the brick wall test for decentering, not so much for aperture softness.

The issue here isn't decentering but what is acceptable "softness" wide open for this lens. It's not just long distance stars that show this issue. I can take some shots from across the room and f1.4 is really quite blurry in fine details at a full crop but it's better at f2.8. I'll take a few photos here and post them.

Also, I forgot to mention that the cropped shots above are from close to the center of the image. If you look closely, you can match the star pattern of the crops to the full sized image just up and to the left.

[clackers]

Yeah, I think people would be happier seeing the results of a conventional sharpness test before introducing the extra variables of astro.

[Na Horuk]

A couple of things to consider.
a) The focus ring / distance scales might be miscalibrated. It is possible that you thought you were focusing at infinity, but actually focused past it or too close.
b) I'm not sure people actually use this lens at f1.4 for astrophoto. Lot of people stop down any lens just a bit for astrophoto
c) Decentering might be a problem. I had it on my Samyang 14mm. The second copy is slightly less decentred, but still not perfect. I would recommend you test it against a newspaper tilted at 45 degrees, near the minimum focus distance. Brick wall tests can be surprisingly difficult, especially with lenses that have a curved DoF (most lenses have some degree of this. Macro primes have the least). Anyway, make sure you do the test in good light, wide open and then at slightly closed aperture, use 2 sec timer and a low ISO and try a couple different focusing distances. Using live view with focus peaking and digital magnification might be the best way to find focus. If one half of the frame is blurred, or if the CA is different in each corner, or if the DoF is diagonal, then there is probably decentering.


Also, check the Samyang lens club thread or search online (500px, Flickr,..) for samples of the Samyang 24mm at f1.4. That way you can compare your results to a wide range of photos taken with all kinds of cameras. That will help you gauge the quality of your lens. Just keep in mind that most photos you see online are resized (This masks lack of sharpness in general) and also post-processed (can have lots of added contrast, sharpness, saturation)
If its in warranty, you should write a nice letter to the dealer. Ask for a replacement. I prefer replacement over repairs, as repairs can take a long time and can even be suboptimal, not to mention the shipping risks and costs.

Edit: Also, what number did you input as focal length? You have to input the exact same number as on the lens (24mm) or the nearest wider. You should also turn SR off for tripod use. 2 sec timer disables SR automatically

[colonel00]

A couple of things to consider.
a) The focus ring / distance scales might be miscalibrated. It is possible that you thought you were focusing at infinity, but actually focused past it or too close.
b) I'm not sure people actually use this lens at f1.4 for astrophoto. Lot of people stop down any lens just a bit for astrophoto
c) Decentering might be a problem. I had it on my Samyang 14mm. The second copy is slightly less decentred, but still not perfect. I would recommend you test it against a newspaper tilted at 45 degrees, near the minimum focus distance. Brick wall tests can be surprisingly difficult, especially with lenses that have a curved DoF (most lenses have some degree of this. Macro primes have the least). Anyway, make sure you do the test in good light, wide open and then at slightly closed aperture, use 2 sec timer and a low ISO and try a couple different focusing distances. Using live view with focus peaking and digital magnification might be the best way to find focus. If one half of the frame is blurred, or if the CA is different in each corner, or if the DoF is diagonal, then there is probably decentering.


Also, check the Samyang lens club thread or search online (500px, Flickr,..) for samples of the Samyang 24mm at f1.4. That way you can compare your results to a wide range of photos taken with all kinds of cameras. That will help you gauge the quality of your lens. Just keep in mind that most photos you see online are resized (This masks lack of sharpness in general) and also post-processed (can have lots of added contrast, sharpness, saturation)
If its in warranty, you should write a nice letter to the dealer. Ask for a replacement. I prefer replacement over repairs, as repairs can take a long time and can even be suboptimal, not to mention the shipping risks and costs.

Edit: Also, what number did you input as focal length? You have to input the exact same number as on the lens (24mm) or the nearest wider. You should also turn SR off for tripod use. 2 sec timer disables SR automatically

Thanks for the reply.

a) I checked the focus and the infinity mark is pretty much dead on. I'll get some other example shots uploaded here soon.
b)I can understand a little softness at f1.4 but if it's this bad then what's the point of having that option aside from doing some artsy portraits? I don't have to stop down my 14mm as it is sharp wide open at f2.8.
c)Yeah, I'll check for decentering more. However, in the example photos above, the issue was consistent across the frame


Also, 24mm was entered as the focal length. All shots were with 2 second timer.

I tried looking through Flickr. The main problem is this lens for most cameras doesn't register the aperture so the info isn't in the EXIF data.

I'll take some more example photos and post there here in a second.

[Na Horuk]

Ok!
You were not using any filters, right?

Maybe you can check the astrophotography thread if anyone posted Samyang 24mm photos there

b)I can understand a little softness at f1.4 but if it's this bad then what's the point of having that option aside from doing some artsy portraits?

Yeah, but f1.4 is really superfast. The DoF is thin, and it gets 'stretched' a lot when you focus to infinity.

[colonel00]

Nope, no filters. Also, I should mention that this is on a K-3 II.

Yeah, I think that's what I'm trying to get to. Maybe I'm expecting more from f1.4 than this lens can deliver. If it's just a limitation of the lens that there's nothing that can be done and I'll stick with my 14mm f2.8. However, I was really hoping that some others that have this lens and especially use it for astro stuff can chime in with their experiences and examples. I'm actually heading out the door to go to Yosemite so we'll see how this does in the real world.

Here are some more quick tests. All RAW images straight out of the camera with only cropping done.

f1.4 with 1:1 crop of the tree in the center





f2.0





f2.8





f5.6





---------- Post added 05-20-17 at 03:47 PM ----------

Here is a close focus test. It was too windy outside to get really good light so I had to do this indoors. This was pretty close to the minimum focus distance.

f1.4





f2.0





f2.8





f5.6

[Na Horuk]

That does seem a bit soft. At close focus its decent, but nearer to infinity it is not perfect.
Can you check the back lens element? Or the front, for smudges? It seems almost like there is some sort of filter in the way. I had similar results when using a cheap UV filter - something like a glow over even the focused areas
I think a couple people on this forum have this lens, but they are not as active. Thats why I recommend you actively search for them

[nocturnal]

The 24mm is softer at f1.4 according to tests and hardly surprising as most lenses are and with coma with a few exceptions...

It's 14mm f2.8 UMC brother is excellent wide open and with only a little coma.

Very few lenses are excellent wide open with little/no coma or astigmatism. Maybe the new Sigma Art ones are.

All of this is only important if you print really big and stare at it close up. All of it is internet pixel peeping mania, if it is for facebook and wanting a few likes, no-one will see it or care. A much bigger factor is one's photography skill, dark sky access, post processing skill etc. Much better to enjoy learning all this first than fretting about sharpness, and I had to tell myself this 😉

I only slightly care as I do make large astro prints, if it was only to post on the internet then no way. 😁

[colonel00]

Thanks again. No filters and the elements were clean Na Horuk.

nocturnal, It's not necessarily for online sharing. I can't remember the last time I was on Facebook. However, I do share images with folks from time to time. I would to have prints at some point as well. What size do you think would still be acceptable?

I spent last night (all night) in the dark skies at Yosemite and have some 2400 photos go to through. Lots of timelapses obviously. On the bright side, I think I was able to some more positive results but I'm waiting to see how it comes out after PP. I found that when checking focus in live view, if I tweaked the focus ever so slightly, the purple aberration would go away it seemed to help. I'll post up some examples once finished.

[nocturnal]

Wow you were in Yosemite!!! Respect to you and if you go here then you do want your equipment to perform!

As for print size I keep a ruler beside my monitor, zoom in and measure the distance across reference points so I can have a real life scale in front of me, check the quality before I upload the files to the printing lab.

A bit of dust on the elements or inside is to be ignored and you will learn this with experience, it is only dust or other particles stuck to the sensor that are a problem with small apertures but not with the lens wide or nearly wide open.

Checking the close focus is a waste of time for an astro lens and you don't need these fancy charts either, your eyes and brain will immediately tell you if a lens is good or not. Even looking through the viewfinder will tell you this also so learn to trust your body's ability! 😉

Do you know how to focus at infinity with this lens? You can use LiveView and zoom in to a reference star and slowly turn the focus ring to and fro until the star is minimised then make a reference of exactly where it is for future use. The infinity dead stop is useless on Samyang and on the 14mm they purposely provide a range of roughly where infinity is at ( a sort of horizontal L-shape). I also use the Pentax refconvertor angle finder 2x to aid critical focussing using my F*300 f4.5 lens, much of this is to avoid using LiveView which destroys your night vision and also heats the sensor which in turn causes more noise.

You may have got a bad copy of that lens which doesn't focus at infinity well, but try what I said and if it is still poor then send it back.

With the new moon this Wednesday I'm trying out a darkish sky site, never tried it but only 20 miles from me and I'm teaching an astrophotography (nightscape!) course at the site this winter. It is obviously the worst time of year for astro in the Northern hemisphere (Ireland here) but I'll be able to gauge it for the winter.

Yosemite! I am so so jealous, the home of the master... Ansel Adams! I've been to California too and some of the most remote parts of America, even been to Roswell, sadly I wasn't into proper photography then. 😔

Try other lenses you have besides the 24mm and then you will know. 😃

Oh and DO NOT EVER use auto white balance... Use Daylight 5400K on Pentax, and the same throughout your processing. I keep the default 'Natural' setting too on the camera... So Daylight and Natural 😉

[colonel00]

Yeah, it's about 3-4 hours away. The forecast was good for last night and a good moon for a dark sky so I drove up and pulled an all nighter. Sadly, even Yosemite isn't the darkest of skies depending on the direction you face but it's still loads better than most other places around here (Bay area).

It's funny you mention it because my 14mm is actually dead on at infinity at the stop. I got lucky apparently with that one.

I'm still torn on this lens after going through a bunch of the photos. Even once I was able to make some tweaks to the focus, it's still producing some bubble distortion things and some aberration (fixable) but you really aren't going to see that unless you are really peeping at full crop.

f1.4 13s ISO1600


File here: https://goo.gl/photos/MvE1zLHyxMB2necY9

[Na Horuk]

f1.4 13s ISO1600

Great shot!

I'm still torn on this lens after going through a bunch of the photos.

I think you answered your own question, mate. This will bug you forever. Contact the shop, see if you can get it replaced fast

[nocturnal]

but you really aren't going to see that unless you are really peeping at full crop

Right... if you are peeping close up... measure what size this would print at as I described before.

Who prints out this size then stares at a massive poster close up? That is the real life mania of pixel-peeping.

If it is faulty or under-performs then send it back but and a big but is... you used the lens fully open wide and it is soft open wide according to measurement sites like lens-tip etc..

Good astro-photographers know with the majority of lenses including the 24mm 1.4 need to be stopped down to f2 or f2.8 or even f4. The Samyang 14mm f2.8 can be used wide open as it measures sharp wide open, that is why it has the huge bulbous front element, the 24mm doesn't have a huge bulbous front element to make it sharp at f1.4.

I think some Sigma Art lenses are good wide open but they are wider angle, maybe 35mm lenses, you would need to look at the measure tests.

I'd love to use my Pentax 50mm f1.4 wide open for astrophotography but no way ho-se!!!! I need to stop it until f2.8 to look half decent, then f4 to get rid of most coma, and f5.6 to get where the lens is superb!!!!!!!!!

By the way (for me!) coma is the biggest barrier to printing large astro, this shows up more than anything else.

Get the picture? Excuse the pun!

Oh really nice photo by the way, I was expecting a blue milky way :P