ed_avis2

<p>Recently someone asked about the difference between the old 17-35L wide angle zoom and the even older 20-35. I compared the two at various focal lengths and apertures at <a href="https://www.flickr.com/photos/7500206@N08/albums/72157636215933256">Lens comparison: 20-35L vs 17-35L</a>. This is a rough and ready test photographing several objects at varying (but fairly close) distance, not a scientifically controlled test chart. Nonetheless it may be useful, if only to indicate that there is no clear winner and that the two lenses have different field curvature.<br>

I also made some test photographs comparing these two L zooms at 24mm focal length and a range of apertures against a third party wide angle zoom, the Tamron 19-35 f/3.5-4.5 (which was also marketed as Vivitar Series 1 among other brands) and a manual fix-focal 24mm lens which can be adapted to Canon, the Olympus Zuiko 24/2. You can see these at <a href="https://www.flickr.com/photos/7500206@N08/sets/72157655861904413">24mm old lens comparison</a>.<br>

Finally I tested the three zooms mentioned above at 28mm, alongside another old L zoom, the 28-80L, and two more adapted fix-focal lenses, the Olympus Zuiko 28/2 and the Contax Zeiss 28/2.8. These are at <a href="https://www.flickr.com/photos/7500206@N08/sets/72157655851982394">28mm old lens comparison</a>.<br>

Again this is only a rough test with obvious flaws (the light changes, and I even moved the tripod part way through) but I found it interesting. It seems to show the two wide angle Canon L lenses beaten by the much cheaper third party zoom, which in turn doesn't match the sharpness of the Olympus lenses. That said, the smudged look of the outer frame with the Canons may be due to field curvature. The 28-80 zoom at 28mm looks better than either of its Canon stablemates - surprising, since they are in the middle of their zoom range and it is at one extreme. The Zeiss 28mm does OK but did not quite match the sharpness of the Zuiko in this test.<br>

I would have liked to do some more rigorous testing - perhaps of more distant objects at infinity focus. Sadly, all of the zoom lenses tested were stolen in a burglary. I still have the adapted manual lenses and have bought a replacement 19-35 Vivitar zoom, so I might be able to test those if someone is interested.</p>

<p>Sigma's 50mm f/2.8 macro lens is good, wider than the macro lens you have now, and should leave enough spare cash to get a separate wide angle lens if you need it.</p>

<p>Ryan J, I'm not certain the bee photograph you posted demonstrates an underexposure problem. It depends on the metering mode; if evaluative metering or some other method that takes into account the whole image is used, then it's possible it was trying for a correct exposure of the blue flowers or even the white petals in the corner. You may need to photograph a flatter, more uniform subject to demonstrate the problem.</p>

<p>Does it happen also in Av mode with a fixed aperture of f/2.8?</p>

<p>Yes; I used to run the free image processing program Raw Therapee. It is slow but capable. To transfer the images it is easiest to remove the memory card from the camera and put it in a card reader attached to the PC. You can also install Canon's DPP if you prefer that.<br>

For still images the quality of the 1Ds II is more than adequate - even for pixel peeping on the big 5k monitors now available. It doesn't have Live View or video recording, or the smooth high-ISO performance of newer bodies.</p>

<p>There's also Sigma's new 24-35 f/2 zoom, if that can fit in your budget. Or their 18-35 f/1.8 zoom for crop-sensor cameras, which nonetheless can be made to work on full frame (see reviews for details of how big the image circle is).</p>

<p>In any photo editing program you can correct for barrel distortion manually, without a lens profile. Just adjust a slider manually until the straight lines look straight. The same is true for chromatic aberration (usually just tick a box for auto-correct) and light falloff (apply a negative vignette filter). This manual fiddling is not as good as having a full lens profile but it's not the case that you cannot correct the image without a profile.</p>

<p>...and if you want wide angle zoom on a budget, consider the Canon 20-35 f/3.5-4.5 USM.<br>

I've kept a Cosina 19-35 f/3.5-4.5 (also sold as Vivitar Series One and other brands) since some quick tests seemed to show it optically better than either of the older L zooms you mention. These lenses are dirt cheap, but autofocus grinds a bit.</p>

<p>I used to have both and did a quick comparison: http://www.photo.net/canon-eos-digital-camera-forum/00c2T1<br>

<br />The used prices of the 17-35 seem quite high. I can't imagine why you'd want it a lightly used 16-35 f/4 IS costs only a little more.</p>

<p>The lens mount is different - look at the position of the red dot. Canon refers to this as 'EF mount (cinema lock type)'. The EOS C500 also appears to use this mount. So what's the difference and does it really take unmodified EF lenses?</p>

<p>Marcus Ian, is it really the case that the 1Ds3 and 5D2 have the same sensor? They are of the same resolution but I thought that the 5D2 was better in low light (while the 1Ds3 may suffer less from 'banding' or some other artefact identified on pixel-peeping online forums).</p>

<p>If you have a spare camera body then ruggedness seems less important - after all you have a spare if one breaks. So a second identical 6D would be easiest. Why learn a different menu system and all that?</p>

<p>Do the pink and blue lines also appear when the image is previewed on the camera's rear LCD?</p>

<p>Agreed about removing both batteries first of all. But then shouldn't you first leave it to soak in pure (deionized) water to remove any deposits left by the river water? Change the pure water once or twice. Then dry it out as above.</p>

<p>Olympus made 28mm lenses for their OM system. The f/3.5 is good and very cheap. The f/2 is good but costs a bit more.</p>

<p>It would be interesting to compare the 5Ds, processed with some sharpening to compensate for the low-pass filter, and the 5Ds R, processed to correct moiré. Which of the two is easier to fix?</p>

<p>David S, thanks for posting the link. It is true that the STM lens performs worse at the longer focal lengths. At 24mm and f/4 it seems the STM does better than the L. I was clearly too hasty in judging it 'at least as good' overall.</p>

<p>Do check the reviews of the 24-105 STM before dismissing it as a 'consumer zoom'. From what I have read it is at least as good optically as the 24-105 L. It is a stop slower at the long end.</p>

<p>Also consider the 24-105 STM lens, which weights 0.525kg.</p>

<p>I don't think you will get a small and light full-frame lens wider than 24mm. Unless you go to the extreme of an 8mm fisheye lens (which gives a circle in the middle of the frame).</p>

<p>The 28-70 is supposed to have full-time manual focus according to http://www.photo.net/equipment/canon/28-70 so in principle you should be able to leave it at AF and then still adjust focus manually. But I have to admit I have never quite understood how that is meant to work. The trick Marcus Ian suggests is a way to make sure it does what you want.</p>

<p>If you shoot video the 1D IV might be better than the 5D II. But you already have the 7D II for video.</p>

<p>I think your best bet is postproceessing... but there are lenses which give a 'soft' look which is more appealing than just being out of focus. Canon's 135/2.8 Soft Focus lens is the only native EF-mount one as far as I know. There are those alternative/classic lenses which are less sharp wide open but manage to give a pleasing softness. For example I find my Contax Zeiss 35-70 zoom is not that sharp at its max aperture of f/3.5, but manages to be unsharp in a glossy, liquid kind of way that preserves contrast. Most Canon lenses shot wide open are either soft in an unaesthetic way (eg the 50/1.4) or just so good that they are sharp even wide open (pretty much every EF lens released in the past ten years).</p>

<p>AFAIK, The 1.4x Mark I has the same optical formula as the Mark II, but the II has better coatings and weather sealing.</p>

<p>Since you mention checking the calendar...</p>

<p>"<strong>Today is cancelled.</strong><br>

There is no April 1st, in 2015. This is a leap year."</p>

<p>I thought that was pretty good.</p>