Question About the Common "50 mm prime" Recommendation

[astrostu]

In the past several months that I've been on these forums, I've heard the recommendation time and again to get a 50 mm f/1.8 or f/1.4 lens for general use (portraits, landscapes, etc.). Since it's about half the price (for the f/1.4) of the super-duper wide zoom that I also want, I was planning on buying one this summer (since I'm also buying a new computer).

However ... I have a cropped (1.6x) sensor, so a 50 mm is effectively an 80 mm lens, which'll give me around a 24°x16° field of view as opposed to 40°x27° of a true 50 mm. It seems like the closest lens would be a 28 f/1.8 (44.8 mm equivalent).

I don't really do any portraits whatsoever, but I do do a lot of landscape. The other intent for the lens was to finally have a very fast lens to be able to shoot constellations (for example, see some of the photos at the bottom of http://burro.astr.cwru.edu/stu/me/photos/stars.html ). The lens I have now for that range is the kit f/3.5, so an f/1.8 would be about 3.78x faster, but an f/1.4 would be about 6.25x faster. So a previous 15-minute exposure could be done in 4 minutes with an f/1.8 or 2.4 minutes with the f/1.4.

But, the problem is that constellations are generally around 30° tall/wide, so I don't think the 50 mm lens would work.

So, although I think I just talked myself into getting the 28 mm f/1.8, I would like to hear what you guys think about this. Am I going to hate myself later for the slower lens? Does the recommendation usually refer to a true 50 mm? Etc.?

Thanks!


Edit: I was slightly mistaken in that at 24 mm, my zoom lens is f/4.0, so it would be a 4.9x improvement with an f/1.8 vs. 8.2x improvement with f/1.4. So a 15-minute exposure could be done in 3 minutes or 1.8 minutes.

 

[Steph]

I dont think you can buy a f/1.4 28mm lens from Canon or other manufacturers (but I could be wrong). However, with a 1.6 crop factor, the closer to a 50mm equivalent would be a 35mm. Canon makes a f/1.4 35mm and Sigma makes a f/1.4 30mm than may suit your needs.

 

[astrostu]

Sorry, I was attempting to compare the 28 mm f/1.8 vs. the 50 mm f/1.4.

I can't afford L glass at the moment, and I think a 35 -> 56 mm lens might still be too narrow. I'm also looking for a lens that I could eventually use on a full-sensor camera, and that Sigma is designed for the APS-C sensors.

 

[deanimator]

Lenses near to 50mm are easier for manufacturers to make sharp
Hence, not only the price, or the angle of view (near natural size), but picture quality is why they are often recommended.
This means that the 35 and 85 mm lenses are also pretty sharp. Many 28mm lenses hold up well too, but as you get further away from the 50mm "benchmark" it gets harder...meaning the price will go up if you are going for max quality. The speed of the lens becomes more of an issue as well, meaning a 28mm f 1.8 is gonna be a lot less sharp than a 35mm f 2.0

For shooting the stars, I think sharpness and minimal aberration, fringing etc will be very important to you. You need to check lens comparisons published in the internet...

 

[shorty6049]

i can get orion all in one shot with my 50mm. heres an example of how wide i can get . but if you dont think you'll use a 50, maybe get somethign a bit wider , its just nice to have a prime lens.

 

[Iron Flatline]

I'm not the right person to answer all your questions, but let me add some snippets of knowledge here:

Lens speed can be important, esp. when shooting "environmental portraits" (read: people doing stuff) without additional lighting or reflectors. Of course, if you're shooting "wide open" with you aperture set to f/1.4 then you will have a very shallow depth-of-field, so your focus better be spot-on, and your background unimportant. It will be out of focus.

Regarding shooting the stars: I would suggest you do so using a tri-pod (that may be obvious.) NO LENS is consistently sharp, and most lenses are sharpest somewhere between f/5.6 and f/11 - each has there own sweet spot, and there are endless articles about every lens and its maximum sharpness point.

So let's say your going to shoot at f/8 - that's 5 whole stops off f/1.4, so you're going to need to keep the shutter open that much longer. You need the tri-pod, because you'd get camera shake hand-holding that shot. Fortunately shooting the stars doesn't require a lot of attention to the Depth-Of-Field, BUT you can't keep the exposure going for too long either - the earth is rotating and it only takes seconds before stars turn into light trails on your image.

There's a Canon 35mm f/2 lens for about $220. I don't know if it is any good, you'd have to do some reading about it. You are right in calculating that it will give you the effective viewing angle of a 50mm on your cropped sensor.

Personally I can highly recommend the Canon 50mm f/1.4 USM lens.

 

[fmw]

Took the words right out of my mouth. A 35mm lens is about "normal" for a digital sensor. The 35 f2 is an easy lens to correct and all the manufacturers make excellent ones. My own 35 f2 outperformed my 50mm lenses optically.

 

[astrostu]

shorty - That's not all of Orion. You missed Betelgeuse and the three stars that form the head, and you also missed the arm extending with the bow and the other arm extending up with the club. The 50 mm lens on a Canon 350D offers a vertical view of 19.5°, which is not large enough to get the entire constellation, unfortunately. The same holds true for many others, since I have never seen an equatorial camera mount that allows you to tilt the camera so that the sides of the detector are not aligned perfectly with RA/DEC. So a 50 mm would not be able to capture the field I want.

fmw - It looks like a 35 mm lens would offer a 27.4°x35.2° field. Upon further measurements, it looks like I would only miss out on getting the Summer Triangle and the tips of Hercules for the main constellations that I was hoping to image. It's just too bad that the f/1.4 is 4x more expensive than the f/2.0 (why can't this hobby be cheaper???). However, it looks like that's only 23% slower than the f/1.8 28 mm that I was looking at.

deanimator - Going with what I just said, it seems as if the 35 mm would be the better option since it would add the bit of sharpness since it's closer to the 50 mm "sweet spot."

Iron Flatline - It would take about 2.4 seconds tripod-mounted with a 35 mm lens before the star covers 2 pixels, assuming I were to get it focused enough to hit 1 pixel, though since each pixel would cover about 1/98° = 37" and the average seeing in Boulder is 5", that should be doable. Anyway, thanks, for the info about astrophotography, but I know about the techniques already (the link I have in the original post is to my images).

Looks like I have some reading to do on reviews for the 35 mm f/2.0 Canon lens.

 

[Seefutlung]

Have I got a lens for you ... you want a 28 - F/1.8 ... then get a Sigma 28 - 1.8 about $269. It is nosy when you focus and a bit slower than the HSM cuz it's a microdrive ... but it should meet all your expectations and then some. A buddy of mine has one and he loves it because it will focus down to about an inch or so in front of the lens. I'm getting the 20mm - 1.8 version for my 5D. Look at B&H and/or Sigma4Less for pricing.

Gary

 

[astrostu]

Okay, the reviews of the Canon 35 mm f/2.0 lens are not so hot, mainly faulting build and motor noise, as opposed to the 28 mm f/1.8. I think I'll post a question in the equipment section of this site in a bit and ask for some real-life feedback.

Gary - The Sigma 28 mm has a minimum focus distance of 7.8", not 1".

 

[Seefutlung]

Okay, the reviews of the Canon 35 mm f/2.0 lens are not so hot, mainly faulting build and motor noise, as opposed to the 28 mm f/1.8. I think I'll post a question in the equipment section of this site in a bit and ask for some real-life feedback.

Gary - The Sigma 28 mm has a minimum focus distance of 7.8", not 1".

True ... but the minimum focusing distance is measured from the film/sensor plane ... so when the body and lens is added ... it's right there.

Gary