A Test of the Canon EF 300mm 4.0 IS USM


By Per Inge Oestmoen, Norway


Here is a test of the 300mm 4.0 IS with and without 1.4x Extender, from May 03, 2005.

Among other things you can see the results with three different Canon EOS digital camera bodies, with three different sensor sizes. This should be of interest for those who wonder how the "crop factor" influences the final image with this particular focal length.

These three cameras were used:

* The EOS 1Ds II, with a sensor size of 24 x 36 mm
* The EOS 1D II, with a sensor size of 19.1 x 28.7 mm
* The EOS 20D, with a sensor size of 15.0 x 22.5 mm

To make that absolutely clear, all three cameras are highly competent photographic tools, which will deliver excellent results when used by a skilled photographer and combined with similarly capable lenses. Here, an important purpose was to give an impression of what kind of performance can be expected from the aforementioned 300mm telephoto lens by Canon.

Some might say that one cannot possibly see differences in quality from pictures of the size of the smaller images that are 900 pixels horizontally, but as you can see for yourself that is not the case. If there are significant differences, they will almost invariably be apparent at any size, on screen as well as in print. That should be demonstrated by the 100% crops also provided for each image. If some of you for some reason want the full images, contact me at pioe@coldsiberia.net.

Contents:

The first test, with the aforementioned lens on three different Canon digital SLR bodies
The significant comparison, between the IS and non-IS versions of this lens
My final conclusions
Image optimization
A final note

The first test 

The Canon EOS 1Ds II depicted with the EF 300mm 4.0 IS:

http://www.coldsiberia.net/public/_MG_1992_900.jpg


EF 300mm 4.0 IS, aperture 4.0, on Canon EOS 1Ds II:

http://www.coldsiberia.net/public/_V2O0987_900.jpg

A 100% crop of this image


EF 300mm 4.0 IS, aperture 5.6, on Canon EOS 1Ds II:

http://www.coldsiberia.net/public/_V2O0988_900.jpg

A 100% crop of this image


EOS 300mm 4.0 IS + 1.4x Extender, aperture 5.6 on EOS 1Ds II:

http://www.coldsiberia.net/public/_V2O0990_900.jpg

A 100% crop of this image


EF 300mm 4.0 IS + 1.4x Extender, aperture 8.0 on EOS 1Ds II:

http://www.coldsiberia.net/public/_V2O0991_900.jpg

A 100% crop of this image


EF 300mm 4.0 IS, aperture 4.0, on Canon EOS 1D II:

http://www.coldsiberia.net/public/_I7X8822_900.jpg

A 100% crop of this image


EF 300mm 4.0 IS, aperture 5.6, on Canon EOS 1D II:

http://www.coldsiberia.net/public/_I7X8824_900.jpg

A 100% crop of this image


EF 300mm 4.0 IS + 1.4x Extender, aperture 5.6, on Canon EOS 1D II:

http://www.coldsiberia.net/public/_I7X8826_900.jpg

A 100% crop of this image


EF 300mm 4.0 IS + 1.4x Extender, aperture 8.0, on Canon EOS 1D II:

http://www.coldsiberia.net/public/_I7X8828_900.jpg

A 100% crop of this image


EF 300mm 4.0 IS, aperture 4.0, on Canon EOS 20D:

http://www.coldsiberia.net/public/_MG_1994_900.jpg

A 100% crop of this image


EF 300mm 4.0 IS, aperture 5.6, on Canon EOS 20D:

http://www.coldsiberia.net/public/_MG_1996_900.jpg

A 100% crop of this image


EF 300mm 4.0 IS + Extender 1.4x, aperture 5.6, on Canon EOS 20D:

http://www.coldsiberia.net/public/_MG_1998_900.jpg

A 100% crop of this image


EF 300mm 4.0 IS + Extender 1.4x, aperture 5.6, on Canon EOS 20D:

http://www.coldsiberia.net/public/_MG_2003_900.jpg

A 100% crop of this image


EF 300mm 4.0 IS + Extender 1.4x, aperture 8.0, on Canon EOS 20D:

http://www.coldsiberia.net/public/_MG_2000_900.jpg

A 100% crop of this image

Portrait of a human, with the 300mm 4.0 IS at full aperture 4.0. As you can see, this picture is not critically sharp:

http://www.coldsiberia.net/public/_MG_2011_800.jpg

The significant comparison

Lastly, here is a direct comparison between the IS and non-IS versions of the Canon EF 300mm 4.0, and also a comparison to the EF 300mm 2.8 IS:

The 300mm 4.0 IS (At full aperture 4.0)

The 300mm 4.0 non-IS (At full aperture 4.0)

The 300mm 4.0 IS with 1.4x Extender (At full aperture 5.6)

The 300mm 4.0 non-IS with 1.4x Extender (At full aperture 5.6)

The 300mm 2.8 IS (At aperture 4.0)

The 300mm 2.8 IS with 1.4x Extender (At full aperture 4.0)

My final conclusions

The differences between the EF 300mm 4.0 non-IS and the EF 300mm 2.8 IS are rather negligible, whereas the differences between the IS and non-IS versions of the 300mm 4.0 are very significant. This disparity in optical performance between the two latter telephoto lenses is readily apparent both without and with the use of the Canon 1.4x Extender.

Based on these results, I am of the opinion that the EF 300mm 4.0 IS could, and should, have been better at its maximum aperture. It delivers useful, but not critically sharp results at 4.0. However, the lens is after all capable of sharp results, provided it is stopped down one step from full aperture. The same is also the case when used with the 1.4x Canon Extender. Thus, this lens is effectively a 300mm 5.6 IS and a 420mm 8.0 IS with the 1.4x Canon Extender.

Therefore, I justifiably suggest that the well known and hotly debated MTF result published at photodo.com around the year 2000, where the new 300mm 4.0 IS got a grade of 3.4 as compared to the grade 4.3 given to the older non-IS version of the same lens, reflects a demonstrable fact. The 300mm 4.0 IS lens by Canon is not capable of critical sharpness at full aperture, but improves visibly when stopped down and becomes a sharp lens at 5.6. The non-IS version of 300mm 4.0 achieves critical sharpness already at 4.0.

Image optimization

All images with 900x600 pixels have been taken into my copy of Adobe Photoshop 7.01 . This Photoshop version is fortunately without the Product Activation which no photographer should accept because dependence on an activation service in order to install and use the working tools we need to access and manipulate our own images undermines our control over our own property, and ultimately jeopardizes the safety of our own data.

The images were then treated with Unsharp Mask with the values of Amount: 240-250, Radius: 0.3, Threshold: 0.

The larger crops are not modified in any way.

Except for the human portrait a tripod was used in all cases, as was mirror-lockup and self-timer. IS always turned OFF when a tripod was used.

A final note 

Often, people who are interested in photography ask me: "What is a sharp picture?"

I have given a little thought to that question, and tentatively I would answer something like this:

If a photographic image shows a significantly inferior amount of detail as compared to what the photographer experienced at the moment of exposure, there is more than likely a weak link in the imaging chain. Of course, there is not and will never be a common consensus about what is "a significantly inferior amount of detail," in the same way as some people think that durable camera bodies and lenses are unnecessary for most. I am of the opinion that these things are important whether or not one is a "professional," and likewise I also believe that the ability of a photographic lens to deliver (very) good sharpness, color and contrast is indispensable for any photographer who wants to have optimum possibilities of growth and development within his or her chosen activity. Whether or not he or she is an amateur, semi-pro or pro is not important, what matters is what you want and where you want to go.

It is trivial that no equipment can make an excellent photographer and that content, message and composition are the other crucial elements in all photographic expression. Nevertheless the photographer who sincerely aims to represent quality is best served by capable and dependable quality tools at every step.

Therefore, never let anyone talk you into settling for less, always save patiently and buy the best you can possibly afford.