Canon 24 mm f/1.4 L: shortest FL f/1.4 experience?

[Roger Clark]

Hi,
I've been looking at upgrading to a faster short focal length lens for night sky photography. My fast wide lens is currently the 20 mm f/2.8. I find the image quality OK, but not stellar, although no short focal length lens seems to have stellar (pun intended) image quality. According to the MTF charts at
http://www.usa.canon.com/cusa/consum...ef_lens_lineup
the 24 f/1.4L has similar image quality at f/1.4 as the 20 mm f/2.8 does at f/2.8, so I would gain 2 stops.

Does anyone have experience with this lens on night star images? If so, could you point me to some full resolution crops (e.g. center versus corners). I would mostly use the lens with 1.3x crop bodies for the near term. The smaller field of view does not bother me as I would do mosaics using something like 15 second exposures.

Roger

[Alan Lillich]

Roger,

You might try contacting Jennifer Wu. I met her a couple months ago at a Canon event and she mentioned that the 24 1.4 is her favorite night lens. In her case, I got the impression mainly because the extra 2 stops has a big impact on eliminating star motion.

Alan

[arash_hazeghi]

I don't have any experience with star photography but I did try the 24 f/1.4L II for some landscapes on a 5D2. IMO it is not worth the price Canon is asking. The corners are soft at wide apertures. I tried the TS-E 24 MKII and it was amazing. In terms of corner sharpness it is the best wide angle Canon makes. Unfortunately it is only f/4 and you lose AF which may not be a big deal for what you are doing. It is not cheap but you also get Tilt-Shift capability too.

There is also a new 24mm f/2.8 IS (non-L) which has good reviews the IS is the latest generation 4-stop so it can make up for slower f/2.8.

The sample images I have are not controlled just random photos so it's a bit subjective but photozone.de has a a good review that might interest you

http://www.photozone.de/canon_eos_ff...anon24f14mk2ff

[Niklas Hedin]

Hi,

I wouldn’t go for the EF24/1,4II if I were you. Apart from not being particularly sharp at full aperture (not even in the centre) the lens suffers from heavy coma as does the Nikon equivalent. The Nikon 14-24/2,8 (used with a Novoflex adapter) on the other hand is a “stellar” choice for night sky photography and very flexible. If gaining the 2 stops is the primary thing you could go for the Samyang 35/1,4 lens and a full frame camera such as 5DII or Nikon D700 (and still ending up with a cheaper solution than the EF24/1,4II) Of cause the field of view would be even smaller, but as you said – you would be doing mosaics.

Niklas

[Roger Clark]

Well, even after the above warnings I bought this lens. I found some night sky images from an Alaskan photographer who shows crops at the edges of the lens, and how the coma get under control at f/2 (not perfect, but quite reasonable). Remember that when imaging stars with a tripod, the stars move and in a 15 second exposure, movement is around 4 pixels, so if there are a little aberrations, it just adds to the blur. Well, after a night of imaging (hoping for aurora and seeing only a little), I got many wonderful night sky images. The lens is simply stunning for night sky imaging. And pretty cool to take f/1.4 hand held images of people in a very dark room, where the corners are so out of focus from the shallow depth of field and that coma is irrelevant.

Here is a link to a night sky image made with the lens.

http://www.birdphotographers.net/for...280#post821280

and more links on that page. I'm very happy with the lens as it is allowing me to get images never before possible.

Roger

[Niklas Hedin]

Glad it worked out for you Roger.
It might very well have to do with sample variations. The sample I bought was the sharpest one of the five they had in the store, I didn’t check for coma though because I was not yet aware of that potential problem. I bought the lens specifically to shoot the aurora and for that the lens is Ok. (Especially if we are talking video or time-lapse) but corner coma doesn’t decrease one bit (on my sample) when used at f/2 on Full-frame. On APS-H, well, the coma gets better in the corners at f/2, I’ll give you that, but considering I always go for full frame when shooting pictures including the night sky, the lens together with the EF35/1,4 have ended up mostly collecting dust in the wardrobe due to their coma problems. I might fore safety sake mention that I afterwards also bought the Nikon equivalent (24mm) but got rid of it after finding out it displayed the same weaknesses as the Canon lens.
Regarding the Nikon 14-24/2,8. - Yes, I know it’s a bit on the slow side compared with the f/1,4 alternatives but it’s almost unbelievably sharp. At f/2,8 it’s sharper than both my TS-E 24II and TS-E 17 at their maximum apertures, but as I’ve already mentioned we might only be talking sample variations here.

A question for you Roger: I see you obviously prefer your DIV to your 5DII when it comes to night sky photography. I personally prefer my “old” DsIII (and 5DII) to my DIV for night sky photography.The amounts of hot pixels are actually about the same on the two 1D bodies but the 5D have a slight disadvantage. I “combat” the fixed-pattern noise by dark frame subtraction. I know the DIV should have at least a theoretical technical advantage in having lower Read Noise having the newest sensor and supporting hardware and that there is a difference in pixel size (5,7 / 6,4 microns) and that you are doing mosaics. What is the main reason you preferring your DIV to your 5DII for night sky photography?

Regards Niklas

[Roger Clark]

A question for you Roger: I see you obviously prefer your DIV to your 5DII when it comes to night sky photography. I personally prefer my “old” DsIII (and 5DII) to my DIV for night sky photography.The amounts of hot pixels are actually about the same on the two 1D bodies but the 5D have a slight disadvantage. I “combat” the fixed-pattern noise by dark frame subtraction. I know the DIV should have at least a theoretical technical advantage in having lower Read Noise having the newest sensor and supporting hardware and that there is a difference in pixel size (5,7 / 6,4 microns) and that you are doing mosaics. What is the main reason you preferring your DIV to your 5DII for night sky photography?

Regards Niklas

Hi Niklas,

I do more than night sky. I got really annoyed with my 5DII after trying night imaging in Greece (see: http://www.clarkvision.com/galleries/gallery.greece/ ), where I needed to work at low ISO (e.g. 100) for the bright lights in the towns, and finding so much pattern noise in the darker regions. I do take dark frames, but to properly subtract dark frames, one should do a linear conversion which means no standard raw converter (like ACR) with all the wonderful other corrections (e.g. use DCRAW). It is a real problem for work flow. Maybe I should try it again on ACR or DPP output, but my previous experience is less than perfect. Then on star imaging, I find that to avoid fixed pattern noise I need to use at least ISO 3200 (without dark subtraction), which means more saturated stars and lower dynamic range. My 1DIV, in comparison has lower fixed pattern noise at ISO 800 than my 5DII at 3200, fewer hot pixels, and higher sensitivity (about 20%) for each pixel, producing better results. The images at ISO 1600 are so clean I do not need to do any background subtraction as the fixed pattern noise is pretty much nil.

Roger

[John Chardine]

Hi Roger- DPP has a checkbox called "Linear" in the Raw tab in the Tools palette. Is this the Linear conversion you would need?

[Roger Clark]

Hi Roger- DPP has a checkbox called "Linear" in the Raw tab in the Tools palette. Is this the Linear conversion you would need?

HI John,

After not using DPP for a couple of years, I installed it and updated to the latest version. The checkbox does appear to do a linear output. I'll post my other comments in the DPP-DLO thread

http://www.birdphotographers.net/for...-Your-Workflow

Roger