Samyang 14mm - star image quality assessment

[coatesg]

Have got my second Samyang 14mm (returned the first for decentering) - and last night gave the new one a quick whirl. 

Having not seen a "good" one and what the output is like, I'm finding it hard to quantitatively assess whether these lenses are in spec or not (it's a bit better than previous one but not perfect). Here's the corners of an image at 100% - this is unreduced, 20sec, f2.8 at 1600ISO on a 350d (APS-C not full frame). My gut instinct says that this would be very soft in the top left on a full frame, though it's at full aperture - would you return this and try again, or do I need to shoot at f4? If you have a "good" example, can you post what your corners look like?

Thanks for any guidance...!

[pete_l]

I think there are very few primes that give perfect star images wide open. A lot of users will say that they have to stop-down their primes by 1 or 2 F-stops to get perfect images.

It may just be something we have to live with. Star images are the most stringent test that any lens can be put to.

You might even find that these flaws can be "processed out" if you can find a profile for this lens that works with Photoshop or some other tool.

[coatesg]

So that I understand, but also the point I'm struggling with - I think I'm going to have to repeat the test at stopped down apertures to see if it'll run better at f4, but I'd like to know what a "good" profile looks like here. It also defeats the point a little in having a really quick lens for (eg undriven Milky Way photography) if I have to stop it down once or twice to get stars under control. 

However, I can't see any way of "processing" out the stars in the top left - there is significant coma in that area (compared to the opposite corner), and corrections for barrel distortion etc is not going to make a significant impact on that. 

 

[Zakalwe]

I'm no expert, but that looks very, very good to me. Certainly a lot better than my Samyang.

 

[symmetal]

I tried my Samyang once for widefield and there was severe coma like smearing over the whole image, though worse in the corners, but all in the same direction. It was still present stopped down to f8 so I never tried it again. It's a lot sharper on daytime terrestrial shots though compared with the standard 18-55mm zoom that came with the camera although the standard zoom produced better results on stars.

[CraigT82]

To me that also looks very good for wide open, I'd certainly keep that lens

[geordie85]

Barring your top left corner it looks like you've got a good one.

Much better than mine. I'll upload some recent images from the weekend when I get home 

[Alien 13]

These short fl lenses are normally ok ish wide open provided you dont pixel peep or crop too much, as said a generic lens profile or better still one you make yourself will help a little. 

Alan

[ultranova]

Your never going to get perfection at 14mm, looking at the pics you posted

they are pretty darn good,

Paul

[geordie85]

Here's a single sub from the weekend at a dark site. My stars aren't too bad (compared to my first lens that I returned) apart from the bottom left and there's a weird blue cast on the right hand side of my stars. Plus abit of coma all round the edge

[coatesg]

Thanks for the advice everyone! Geordie85 - your one does look like a good one

I found it kinda hard to know what I should be expecting here - I'll run the lens at f4 for driven stuff so will retest anyhow, but it looks like a good enough example to keep based upon all your feedback. 

Cheers!!

[The Admiral]

Have a look at the Polish review site: http://www.lenstip.com/239.1-Lens_review-Samyang_14_mm_f_2.8_ED_AS_IF_UMC_Introduction.html. As with most lenses resolution improves with stopping down a couple of stops, though even wide open you may be happy enough. They reckon coma and astigmatism in the review model was reasonable with APS-C, and the spot tests look OK. Of course, they may have had a good sample.

Ian

[Wyvern]

If it's of any interest or use to you, here's a compressed shot for a recent "early Perseid watch" from the tail end of July. Captured several satellites, not much else!!!

Image is 30 second single shot, unguided at f2.8 with my Samyang 14mm coupled to the Canon 1100d. Image processed via Photoshop with what little "tweaking" ability I possess. 

Really don't know if it's a "good" or suspect Samyang I've got? That said, I'm more than happy with the results I'm getting and that's got to count for something.

Regards, Les

[geordie85]

1 hour ago, Wyvern said:

If it's of any interest or use to you, here's a compressed shot for a recent "early Perseid watch" from the tail end of July. Captured several satellites, not much else!!!

Image is 30 second single shot, unguided at f2.8 with my Samyang 14mm coupled to the Canon 1100d. Image processed via Photoshop with what little "tweaking" ability I possess. 

Really don't know if it's a "good" or suspect Samyang I've got? That said, I'm more than happy with the results I'm getting and that's got to count for something.

Regards, Les

You've got yourself a very good lens there. If you compare your corner stars to mine you'll see a big difference. 

Wanna swap?