Shutter speeds & long lenses - DPChallenge Forums

Nikon AF Nikkor 50mm f/1.8

02/16/2004 09:19:03 AM · #1

Hi all! A quick quiery:

I recently (last week) picked up the Sigma 70-300mm lens for my DR (wanted the get the action to come a little closer to me). Didn't have a chance to play with it until this weekend.

I went out and took a number of shots, and although they looked okay, there always seemed to be a little softness to the pictures. At times they were outright blurry. I never got the nice, sharp shot I was hoping for.

Now, I know that a *lot* of this has to do with the fact that I was shaking in -10C in a forest without a tripod, but I thought that it being a nice, sunny day, shutter speeds of up to 1/600 sec (8.0 aperture)would be good enough for me to hand-hold the camera and still get good results. Either this isn't true, or the lens has some issues.

My question is, about how fast does the shutter need to be to hand hold a shot at 300mm on a lens with no IS?

At this point, I need to know if it's me, or the lens.

Thanks for any input!

pcody

Canon PowerShot G6

02/16/2004 09:26:58 AM · #2

You might want to do some control shots inside on a tripod and with mirror lockup if you have it. If they don't come out sharp, exchange the lens.

Adrian Tung

Canon EOS-10D

02/16/2004 09:28:46 AM · #3

The general rule of thumb is that a shutter speed of 1/focal length is needed to get good shots without camera shake. Remember that the x1.6 crop affects it, so at 300mm you actually need at least 1/480s.

However, remember that this is just a general rule of thumb - you can still get camera shake depending on your style and the situation. This guideline usually works for "normal range" focal lengths (e.g. 28mm-105mm) but can quickly degenerate at both extreme ends.

No matter how steady you think you are, it is still always better (if possible) to use a tripod or monopod, or make the best use of available support like a nearby lamp post or fence.

:)atwl

jonpink

Panasonic Lumix DMC-LX2

02/16/2004 09:29:04 AM · #4

Aim for 1 more than the focal length.

I would say that either your focusing is out or you were shaking a lot

Professionals can handhold a 300mm at about 1/60, practice practise practise.

lwkimages

Canon EOS-5D Mark II

02/16/2004 09:32:22 AM · #5

I agree with pcody's suggestion. In general, you "should" be able to hand hold a lens at a shutter speed equal to the reciprocal of the focal length. That is, if the focal length is 300mm, use no slower than 1/300 of a second. Considering you were close to that speed (1/600) and were shaking, it's not surprising that you have some unwanted softness.

Use a tripod and make sure the ficus is right on.

Canon EF 100mm f/2.8 USM Macro

02/16/2004 09:40:53 AM · #6

Thanks for all the input. I appreciate it!

I'll take some shots this evening indoors using a tripod to make sure things are kosher. (It's tough to get anything indoors at 300mm though *grin*). The 30sec shutter should be enough.

I figure it's probably me, but I'd hate it if it were the lens...

Spork99

OM System OM-1

02/16/2004 10:27:46 AM · #7

Originally posted by space amoeba:

It would be better to take the tripod outside and shoot the same scene handheld and w/ tripod so that you can compare.

ellamay

Canon EOS-1D Mark II

Canon EF 100-400mm f/4.5-5.6L IS

02/16/2004 10:28:12 AM · #8

I too can hand hold shots with this lens, tho of course using a tripod and self timer will get you much more equisite shots IMO.

Konador

Canon EOS-500D Rebel T1i

Canon EF-S 18-55mm f/3.5-5.6 IS

02/16/2004 10:29:47 AM · #9

1/30 is about the longest I can hand-hold shots without shaking at 300mm. It all depends on situation tho.

W.R.Miller

Canon EOS-50D

Tamron SP AF 28-75mm f/2.8 XR Di for Canon

02/16/2004 10:31:58 AM · #10

Originally posted by Adrian Tung:

Remember that the x1.6 crop affects it, so at 300mm you actually need at least 1/480s.

Are we sure about this? I can see if this were an actual magnification why this would be true but with it just being a crop factor doesn't that make this statement incorrect? Wouldn't this be just like taking a image from the 1Ds (no crop factor) and cropping the center out? Shouldn't the image still be just as clear even though I cropped the center out?

Just wondering because I can't understand how the crop factor would make this true. Magnification I could understand.

Thanks!

02/16/2004 10:43:08 AM · #11

To be actually sharp, I'd say at least 1/500s

I think people who are claiming to be able to handhold a 300mm lens at under 1/125s or so are slightly deluded (only kidding) or have a very different idea on what 'acceptably' sharp is.

For me, I did various experiments with my lenses to understand what speeds I could use with them. It is simple to do it yourself too.

All I did was, on a bright, sunny day, took some shots of something with a fair bit of fine detail. I did this for bursts of 3 shots at speeds from just over 1/(lens length*1.6) secs to about 1/30seconds.

Then I did the same thing with the camera solidly mounted on a tripod, using mirror lock up and a cable release.

After looking at those results, its really easy to tell when you can get sharp shots hand held or not. (assuming you look at 100% crops of edges in the image)

I'm interested in being able to make 20x30 enlargements of the shots, so what's acceptable for that is certainly a lot higher than what's acceptable for a 640x480 web image or a 4x6 print.

I'd suggest just trying it yourself - it can be quite educational.

02/16/2004 10:44:12 AM · #12

You folk have lost me with this crop-factor talk. I'm not sure what you're getting at. There was no cropping that I'm aware of... the shots I took were straight out of the camera.

And you can hand hold a 300mm lens at 1/30 Konador? Geez, I wanna know what meditation technique you use to get rid of your stress. :)

02/16/2004 10:45:37 AM · #13

Originally posted by space amoeba:

You have a Canon 300D. Your 300mm lens is actually a 480mm lens on that camera.

The 'crop factor' is just that the 300D sensor is actually 1.6 smaller than a 35mm bit of film.

Canon EF 100-400mm f/4.5-5.6L IS

02/16/2004 10:53:26 AM · #14

Originally posted by Gordon:

You have a Canon 300D. Your 300mm lens is actually a 480mm lens on that camera.

The 'crop factor' is just that the 300D sensor is actually 1.6 smaller than a 35mm bit of film.

Oh, I think I understand. Yes, I knew that the length didn't translate perfectly. Didn't know that was a "crop factor" though. Thanks for the enlightenment. :)

So then, (at the risk of going utterly off topic here) if the sensor is 1.6mm smaller than an actual 35mm bit of film, how is it that the length of a lens actually translates into something *longer* thanthe original?

Admittedly, I'm still a bit of a newbie here. I love photography, but I never really thought about the science much before recently. I suppose I really should.

Adrian Tung

Canon EOS-10D

02/16/2004 10:57:18 AM · #15

Originally posted by wackybill:

Originally posted by Adrian Tung:
Remember that the x1.6 crop affects it, so at 300mm you actually need at least 1/480s.

Are we sure about this? I can see if this were an actual magnification why this would be true but with it just being a crop factor doesn't that make this statement incorrect? Wouldn't this be just like taking a image from the 1Ds (no crop factor) and cropping the center out? Shouldn't the image still be just as clear even though I cropped the center out?

Just wondering because I can't understand how the crop factor would make this true. Magnification I could understand.

Thanks!

Remember that sharpness, like DOF, is a very subjective term and often depends on your output or print size as well.

So lets say that you take an 80mm shot with a full frame camera and then print an 8x10 and get a nice and sharp picture.

Now take a shot with a 300D with a 50mm lens (i.e. to obtain the same focal length as the full frame shot). Due to the size of the sensor, you have to enlarge the image by 1.6x more in order to get the same enlargement as the full frame. The additional 1.6x factor will also enhance any sort of shake or blur in the shot. Something about Circle of Confusion....

Hope I made some sense with all that....

Brent_Ward

Canon EOS-5D

Canon EF 85mm f/1.2L USM

02/16/2004 11:10:30 AM · #16

Now this should be one of the challenges. How slow of a shutter speed can you hand hold. The best combination, longest focal length x slowest shutter speed wins! ;D

ellamay

Canon EOS-1D Mark II

02/16/2004 11:29:14 AM · #17

Cropping and sharpness are also significant to me with this lens in that if I shoot an eagle and he is quite far away if tripod used I can safely crop without losing quality to bring him closer, however if handheld I can't crop nearly as much and still have good quality.

soup

Canon EOS-30D

Canon EF 17-40mm f/4.0L USM

02/16/2004 11:32:38 AM · #18

how about 1/6" at about 90mm equiv - f:5.6?

the crop factor has to do with the field of vision - on a 35mm camera a 300mm lense will have 1.6x the field of vision as the same lense on the Rebel - or - the Rebels image will be same same as a cropped version of the image off the film camera. its isnt 1.6mm smaller its 1.6x smaller
or about 22mm - 35/1.6 = 22mm. so you are taking those into consideration - at 300mm its gives the FOV of a 480mm on a 35mm film camera- but the subject isnt actually any larger than the wider FOV of the film camera. i think anyway

Originally posted by :

So then, (at the risk of going utterly off topic here) if the sensor is 1.6mm smaller than an actual 35mm bit of film, how is it that the length of a lens actually translates into something *longer* thanthe original?

Canon EF 17-40mm f/4.0L USM

02/16/2004 11:35:38 AM · #19

Originally posted by soup:

how about 1/6" at about 90mm equiv - f:5.6?

it looks fine shrunk to 640x480 but how does it look at 100% ? You can get away with a whole lot on a screen and shrinking it.

Circles of confusion are related to output size, as is the concept of 'sharpness' so you get away with a whole lot more for web viewing than you ever do when you try to print something at anything approaching 8x10 or above.

soup

Canon EOS-30D

02/16/2004 11:45:27 AM · #20

if it wasnt sharp full size i wouldnt have considered resizing it...
it isnt my style to post a highly a cropped image for web display - or to trick you to thinking a photo is sharper than it actually is because its displayed on the net in a smaller version - or to post an image that is actually cropped down to a mere 640x427px full size image - my stuff is mostly full frame off the camera. if i cant print it at 10x15 or higher - its doesnt usually get displayed.

its grainy due to 1600iso

i'm at work - will post a crop of the lense from the original when i get home tonight

have a little faith in me ;}

GeneralE

Canon PowerShot S3 IS

Canon 12x IS USM (S3 IS built-in Lens) -- 36-432mm (35mm EQ)

02/16/2004 11:47:15 AM · #21

Originally posted by MeThoS:

Now this should be one of the challenges. How slow of a shutter speed can you hand hold. The best combination, longest focal length x slowest shutter speed wins! ;D

My PWL was a 1/2-second, handheld shot. Focal Length 5.6, f/2.8, ISO 200.

Canon EF 17-40mm f/4.0L USM

02/16/2004 12:35:36 PM · #22

Originally posted by soup:

I think you misunderstand what I'm saying.
I have plenty of hand held shots that look 'sharp enough' until you actually try to print them large. They all look great on the web, and sharp enough on the screen.

The idea of sharp is intimately tied to the viewing size and distance, through the concept of circles of confusion. I found personally I wasn't even happy with the results I get from 1/(focal length) hand held if I was considering a really large version. Certainly 1/focal length was much softer hand held than it was using my best tripod technique.

Also the tests I did were with a reasonably expensive, high quality (70-200 F4L) lens - this is also a factor when considering what we mean by 'sharp'. Some of my cheaper lenses (Canon 24-85 f4.5/5.6) never give me an acceptably sharp enlargement beyond about 8x10 - no matter what technique I use.

soup

Canon EOS-30D

02/16/2004 12:55:55 PM · #23

i suppose - i misunderstood.

though you should be able to get an idea of the printed output when viewing the image on the PC at 100% res at the output dimensions... no?

thats what i have to base my opinion on - as i dont print every photo i take at maximum possible print size.

also - the tripod - in a lot of circumstances just isnt helpful - cumbersome, and a hassle - so striving for a steady hand rather than relying on support is my focus. i can say my entry in the Black challenge would have been impossible to take with any support, and didnt come out as sharp as i would have liked, but better to get the shot - than miss it altogether worrying about potential softness in a 20x30 enlargement, or missing it to set up the tripod. i can see your point, and am sure its valid. at some point i'll run the tests you mention.

02/16/2004 01:39:40 PM · #24

Another way of putting it - at least for what I find acceptably sharp, it is hard to get a 6Mp image that is sharp enough. By default my D60 gives me something that is about 8x10 at 300dpi so I have to enlarge it quite a lot to get an acceptable print - any lack of sharpness, fall off towards the edge or deficiencies in technique really shows up, pretty glaringly.

e.g.

is not sharp enough to print well. nor is

Both were handheld with a 70-200 + 1.4x tele, on a 1.6x crop camera, at or slightly slower than the 1/focal length shutter speed. They look great at web sizes, they even look great in 8x10's but the details are too soft to enlarge very well. Yes I am expecting a lot from my lenses and gear and it is pushing the limits of what a 6Mp camera can do, but as these are selling for $500 or more unframed, I want to have good quality.

Message edited by author 2004-02-16 13:41:15.