Expecting too much from a lens?

[Journeyman]

Hello,
A question related to some shots taken at a distance of 110 metres and one taken at 85 metres with a focal length of 400mm. The result looks great at zero magnification.

In general would it be unreasonable to expect a clear view at zero crop and with a tight crop over the mentioned distances. What would be a size could be expected before the grain becomes very obvious. In percentage terms an increase to the image of 50%, 75% etc.?

My main concern is that I maybe expecting far too much from my lenses, or that my focus and hand tremors are taking more of a toll than I would like.

Thank you Journeyman, (Dennis)

[ameerat42]

Hello,
A question related to some shots taken at a distance of 110 metres and one taken at 85 metres with a focal length of 400mm. The result looks great at zero magnification.

In general would it be unreasonable to expect a clear view at zero crop and with a tight crop over the mentioned distances. What would be a size >Ed: "that"<could be expected before the grain becomes very obvious. In percentage terms an increase to the image of 50%, 75% etc.?

My main concern is that I maybe expecting far too much from my lenses, or that my focus and hand tremors are taking more of a toll than I would like.

Thank you Journeyman, (Dennis)


Hmm! - If I understand this right... particularly the crop size parameter. I take 50% as meaning "doubled" in size and 75% as 1.3X size (ie, < 2).

If the exposure conditions and the seeing* conditions are pretty good, then you should easily be able to double your image size.
(Of course, a "decent camera" is assumed). And to generalise further, I'd say that about a 4X image increase would start to show
some "grain", ie, noise.

To a certain extent, a good noise reduction program would help this, but too much would start to change the image into fiction.

That you are finding this with your telephoto lens is partly due to the greater intervening distance of air turbulence that you are
recording in the image. Image stabilisation should help a good bit, and of course there's always focus, but these are other issues
separate from noise or grain.

If you are expecting "too much" from your lenses, then you should soon enough find their limitations. Try some controlled subjects
in fairly good conditions to test your lens performance and to assess the optimal IQ they can deliver.

Sometimes, the hype implies the "have gun will travel" notion (gun<>lens, travel<>good results), but you are proving that adage.

*Seeing (q.v.) is borrowed from astro-nomy/photography.

[Tannin]

Hi Dennis.

Let's clarify the terms:

UNCROPPED: the whole image, what you are calling "zero magnification".

100% CROP: cropped such that the number of pixels you see on the screen (or paper) is the same as the number of pixels in the image. You can't crop any further without making the image on-screen smaller. (Yes, the term makes no sense. Sadly, it is the one we are stuck with.)

50% CROP: cropped such that we see half as many linear pixels on the screen as there are in the image.

Some examples:

Uncropped = 6960 x 4640 on a 90D. You might look at it at something like 1600 x 1200 on a typical screen. Each pixel you see on-screen is an average of about 16 pixels in the image (4 x 4).

100% crop = take a small portion of the image (say 1600 x 1200) and discard the rest. Each pixel you see on-screen is one pixel from the original image.

50% crop = take half the image, crop out the rest. Each on-screen pixel is the average of roughly 4 original pixels .

With a really good lens and excellent technique, you can get sharp pictures even at 100% crop - but it is difficult. You need to do everything exactly right. The higher the pixel density of your camera, the more difficult it becomes. I can get every-pixel-sharp pictures regularly with the 1D IV (30,795 pixels per square millimetre on the sensor), but it takes extra care with the 5D IV 5D IV (34,844 px/mm2) and is very difficult with the 7D II (59,410 px/mm2) and would be much harder again with a 90D (97,193 px/mm2) - quite likely impossible under real-world conditions.

Your two lenses are both excellent, almost as good as it gets.

If you need more resolution than you have now, you can:

(a) Get closer to the subject. With wildlife, that takes skill and patience. With astro subjects, it is more difficult.

(b) Use more optical magnification. You could try a teleconverter but it won't help. Teleconverters work by trading quality for magnification. You are already at or beyond the quality limit so you have nothing to trade off for extra reach. Longer lenses do work, but only if they are good ones. Don't bother thinking about the various cheap "600mm" zooms - they don't achieve anything you can't do better with the 100-400 II your already have and a slight crop. If you really want to go longer than the 100-400 II or the 400/5.6 you need to look at made-for-purpose lenses, such as the 500/4, 600/4 and 800/5.6. THese are very expensive and very heavy but they are the *only* way to improve on the equipment you already have.

(The "only" way? Well nearly. Personally I'd swap the 90D for a 5D IV - losing some detail resolution but gaining much lower noise and better handling - but I do wildlife, never astro. For astro a 90D might be the better bet.)

EDIT: if you are getting grain, it's time to ask what ISO you are using. At 100 through to about 400 ISO you should have no visible grain at all, and not much at 800.

[Journeyman]

Good day Tanin,

Thank you, my mind has let me down here.
I understand the 100% size, I did not think clearly before I posted. Where I have mentioned 50% etc. I really mean 150%, 100+ magnification.

I have attached a couple of pics taken with a Canon EF300 f4 + 1.4x converter. Also a blow up from a Tamron 100-400.
The reality could be that I am missing something in the focus part of the job, my eye not being too sure. When I viewed the photos the reality is they are better than I gave them credit for.

Again thank you for your help, if there were a photography/processing school on the coast I would front up 'toot suite'.
I was waiting for this old mate to "drop a line" in the water. He sat in the same spot for at least and hour and a half. Certainly out waited me, it was dusk when I left.
Kind regards JM

147136

147135

147137

[Tannin]

Sorry Dennis, I was misled by your sig. I assumed that we were talking about the 100-400 II or the 400/4.6. The 300/4 is a wonderful lens, but once you add a teleconverter, it naturally drops back to merely good, possibly very good. The Tamron 100-400 is good for a cheapie but not in the same class as the Canon 100-400 II, the 400/5.6, or indeed the 300/4 bare lens.

Looking at those three pictures, they appear to be about as good as you can reasonably expect from the lenses you used. The Tamron 100-400 is a little surprising to me, I'd not have expected it to be clearly superior to a 300/4 and 1.4 converter. (But I'm assuming (a) that these shots are representative of the two lenses usual output (not just lucky or unlucky flukes) and (b) that the teleconverter is a good one, well matched to the 300/4. I.e., the Canon one, which is designed specifically to go with the 300/2.8, 300/4, 400/2.8, 500/4, and 600/4. Not so sure about the other teleconverter brands without testing them one by one.)

So: can you improve on these shots? (Without getting any closer, I mean.)

Yes: you'll get a small but consistent and significant improvement by going to one of the very best 400mm lenses (Canon 100-400 II, 400/5.6 except that it lacks IS which is a deal breaker for most people - rightly so in my view), or 400/4 DO. Better again a 500/4 or a 600/4. All of the last three mentioned will take a teleconverter happily. But don't expect miracles - with the extremely high pixel density of the 90D, teleconverters really won't help a lot. As a general rule, I put a 1.4 converter on my 600/4 OR swap the 5D IV for a (higher density) 7D II, not both. And although I own a 2X converter, I hardly ever use it, not even with the 1D IV or the 5D IV.

Hope this helps - Tony