Jump to content

ian_hobday

Members
  • Posts

    97
  • Joined

  • Last visited

Posts posted by ian_hobday

  1. The 17-40 4L is still listed on Canon Japan's site:<br>

    <br>

    <a href="http://cweb.canon.jp/camera/ef/catalog/category/wide_zoom.html" target="_blank">Canon Japan Wide Zooms</a> (Opens in a new window.)<br>

    <br>

    Nothing is impossible of course, but it does not seem likely that Canon will drop this lens.<br>

    <br>

    Ian<br>

    --<br>

    Ian Hobday<br>

    Osaka, Japan<br>

    <a href="http://hobday.net/photos" target="_blank">http://hobday.net/photos</a> (Opens in a new window.)

  2. I don't have specific knowledge of the N series lenses, but generally speaking modern AF lenses can not be adapted from one mount to another.<br>

    <br>

    Not only would you need to build a physical mount adapter, but also a electronic signal "translator". You'd need some sort of battery power for it and a signal processing chip to do the actual translation. You would also need to reverse engineer the signal systems used by both the EOS body and the AF lens you wish to adapt.<br>

    <br>

    Impossible? Not totally. Easy? Definitely not!<br>

    <br>

    Ian<br>

    --<br>

    Ian Hobday<br>

    Osaka, Japan<br>

    <a href="http://hobday.net/photos" target="_blank">http://hobday.net/photos</a> (Opens in a new window.)

  3. I find the autofocus of my D60 to be about the same as what my EOS 5 was. Obviously <b>way</b> below the performance I get with my EOS 3.<br>

    <br>

    Going to upgrade the D60 later this year and autofocus performance, especially in low light, is one of the things I will be considering carefully.<br>

    <br>

    Ian<br>

    --<br>

    Ian Hobday<br>

    Osaka, Japan<br>

    <a href="http://hobday.net/photos" target="_blank">http://hobday.net/photos</a> (Opens in a new window.)

  4. <i>So what is it in the EOS 3 and higher cameras that allows AF at f/8?</i><br>

    <br>

    Smarter/better focusing system electronic...<br>

    <br>

    I have an EOS 3, and the low-light focusing is <b>way</b> better than the EOS 5 was or my D60 is. Not cheapo lenses either -- 17-40 4L, 28-70 2.8L, 70-200 2.8L.<br>

    <br>

    Low light focusing performance is one of the things I will consider when I replace my D60 later this year...<br>

    <br>

    Ian<br>

    --<br>

    Ian Hobday<br>

    Osaka, Japan<br>

    <a href="http://hobday.net/photos" target="_blank">http://hobday.net/photos</a> (Opens in a new window.)

  5. I have the 2.8L. Bought it when I was shooting film with an EOS 3, now use it with my D60.<br>

    <br>

    With both the 3 and the D60 the extra stop is worth the extra weight -- for me.<br>

    <br>

    As other posters have mentioned, whether or not the extra $$$ and extra weight are worth it <i>for you</i> is something that only you can decide.<br>

    <br>

    If you are really unable to decide, check around to see how much it costs to rent the lenses for one day each. Get them on different days and walk around shooting with them for several hours each... You'll soon know if you want the 2.8 or the 4. :)<br>

    <br>

    Good luck!<br>

    <br>

    Ian<br>

    --<br>

    Ian Hobday<br>

    Osaka, Japan<br>

    <a href="http://hobday.net/photos" target="_blank">http://hobday.net/photos</a> (Opens in a new window.)

  6. I went from EOS 5 --> EOS 3 --> D60. I still have the 3 but <i>very</i> rarely use it. So much so that I am going to sell it when I replace the D60 later this year.<br>

    <br>

    The EOS 3 low light focusing absolutely wipes the floor with both the 5 and the D60. Even in good light I can feel a difference, but in low light it is vastly different.<br>

    <br>

    Focusing speed and accuracy will be one of the things I consider when I replace the D60. Not the only thing of course, but it is one of the things I really miss from my EOS 3 days.<br>

    <br>

    Oh, for reference my main lenses are:<br>

    <br>

    17-40 4L<br>

    28-70 2.8L<br>

    70-200 2.8L<br>

    Sigma 180mm 3.5 Macro<br>

    <br>

    Ian<br>

    --<br>

    Ian Hobday<br>

    Osaka, Japan<br>

    <a href="http://hobday.net/photos" target="_blank">http://hobday.net/photos</a> (Opens in a new window.)

  7. The 550EX has dropped about US$120 in price and the 580EX is priced where the 550EX was. Well, at least here in Japan.<br>

    <br>

    I plan to keep my 420EX and get a 580EX when I get my next DSLR. Waiting to see what comes at Photokina before deciding which DSLR I will get. :)<br>

    <br>

    Ian<br>

    --<br>

    Ian Hobday<br>

    Osaka, Japan<br>

    <a href="http://hobday.net/photos" target="_blank">http://hobday.net/photos</a> (Opens in a new window.)

  8. <i>The recent additions to the EF-S lens line is a very clear indication that 1.6x type small sensors are not going away.</i><br>

    <br>

    No, it is an indication that Canon has smart marketing people.<br>

    <br>

    It is an indication that Canon knows they need such lenses in order to compete in the current high-end consumer (so-called "prosumer") market.<br>

    <br>

    If I buy a 20D (I have a D60 now), I will likely get the 17-85 IS lens too, unless the reviews are very negative. However, I will buy it knowing for sure that I will not be able to use it for more than a few years.<br>

    <br>

    Ian<br>

    --<br>

    <a href="http://hobday.net/photos" target="_blank">http://hobday.net/photos</a> (Opens in a new window.)

  9. <i>When has this ever been an issue to the "birthday party, anniversary, Disney World vacation" point-and-shooters, digital or film? Good points in general tho Ian.</i><br>

    <br>

    Hmm... True. Especially for the low to mid-range P&S, very true.<br>

    <br>

    For the high-end P&S cameras like the Pro1 and the G6 I think it is going to be interesting to watch though.<br>

    <br>

    Ian<br>

    --<br>

    <a href="http://hobday.net/photos" target="_blank">http://hobday.net/photos</a> (Opens in a new window.)

  10. The short answer is "no".<br>

    <br>

    The longer answer is, well, longer. ;)<br>

    <br>

    Yes, you can apply PS blur filters to give the impression of bokeh. It doesn't look quite the same though as the bokeh of a lens is not equal throughout the image, but the bokeh you apply in PS usually is. It also isn't quite the same effect as good lens bokeh.<br>

    <br>

    Further, doing a really good job of blurring a background in PS takes time... A lot of time if you want a <i>really</i> good job. If you have one image that you "goofed" and want to save, spending time to fix the background is probably worthwhile. But it is pretty much impossible to do this for a batch of 100 images...<br>

    <br>

    There are good points to the 1.6x sensor. Sometimes having the FOV of a 320mm lens with the DOF of a 200mm lens is extremely useful. Other times it is a nightmare.<br>

    <br>

    But as I said above, you can crop a full frame image to get the effect of a 1.6x image. It is impossible to "expand" a 1.6x image to get the effect of full frame though.<br>

    <br>

    If CMOS sensors were less expensive to make now you can be sure that we would all be shooting on full-frame digital bodies. For the 1DM2 it is one generation away. For the D20, two generations. For the 300D/Rebel/Kiss, 2-3 generations. Basically within 5 years all Canon bodies will be full frame.<br>

    <br>

    I really wonder what will happen to the P&S cameras though. Being able to have a 10x lens in a $800 P&S is a <i>huge</i> marketing thing. Going to larger sensors will make that impossible to do. But keeping smaller sensors will mean lower image quality and eventual resolution limits. It will be interesting to see how this unfolds as DSLRs move towards full frame.<br>

    <br>

    Ian<br>

    <a href="http://hobday.net/photos" target="_blank">http://hobday.net/photos</a> (Opens in a new window.)

  11. As you have worked in the semiconductor industry for so long, you are also aware of all the times that "impossible" things have happened.<br>

    <br>

    Ah yes, the 1 micron barrier... We'll never get high yield production below 1 micron. The original 8086/8088 chips were 3 micron (1978). It took until the 80486 in '89 to get to 0.80 micron, but it was done.<br>

    <br>

    Since then there have been other big breakthroughs as well, and now we are at 0.13 heading to 0.09. Not only has the "impossible" 1 micron barrier been broken, but we are now 10x smaller than that and getting smaller all the time. 0.065 is not that far away.<br>

    <br>

    As you also surely know, until now shrinking the process has been more desirable than increasing die size for many reasons -- voltage, heat, speed -- just to name three. Yield & cost per chip are two other ones -- more chips per wafer is a big bonus, certainly.<br>

    <br>

    However in image sensor design other things are important, and that changes the priorities. So now we have companies spending large amounts on R&D to find out how to reliably increase die sizes. What we had before was companies spending $$$ on R&D to reduce the process size and thus shrink the die.<br>

    <br>

    Over time we can certainly expect sensor sizes to go up and prices to come down.<br>

    <br>

    Ian<br>

    <a href="http://hobday.net/photos" target="_blank">http://hobday.net/photos</a> (Opens in a new window.)

  12. I too find the increase of DOF both helpful and frustrating, depending on the situation.<br>

    <br>

    I've always wondered why people look at the FOV "bonus" side of an APS sensor, but never seem to consider the DOF downside. I think a lot of people still don't understand exactly what that 1.6x means. A lot of pros don't seem to "get it" either. Sigh...<br>

    <br>

    As CMOS production techniques improve we will certainly see costs fall dramatically. Full frame 35mm sensors will become standard. APS-sized sensors will become what APS film is now -- a bizarre oddity in an SLR body. I can't tell you how much I am looking forward to that day!<br>

    <br>

    Ian<br>

    <a href="http://hobday.net/photos" target="_blank">http://hobday.net/photos</a>

  13. These lenses are totally different beasts.

     

    What do you want to use it for? Do you need 400mm? Don't forget that your field of view (FOV) is 1.6x narrower than on a 35mm camera. So the 200mm will have the same FOV as a 320mm film. If you get the 1.4x extender you can get a FOV of 448mm on your D20.

     

    Of course with the 400 you can get a lot longer with the 1.4x and the 1.6x FOV.

     

    The answer to your question is therefore another question: What do you want to do with the lens?

     

    Ian

     

    http://hobday.net/photos

  14. Why do people continue to believe that 1.6x will become a permanent standard? It won't.

     

    As we all know, 1.6x has advantages. Like a 320mm field of view (FOV) with a 200mm depth of field (DOF). This can be extremely useful. I get a 448mm FOV from my 70-200 2.8L using a Canon 1.4 Extender on my D60. Even better, the DOF is that of a 280mm lens.

     

    But the other side of the sword cuts too, and it's not good. Want to use your 135mm 2L for portraits on your 1.6x? Uh-oh, now you have a FOV of a 216mm lens. Better back away from your subject 1.6x more. But damn, now you have a lot more DOF and you can't get the background to bokeh properly. Better move the background further from your subject. What? Your studio isn't that big? Too bad. Well, you could go to an 85mm 1.2L to get the same FOV of the 135mm, but you still don't get the DOF of your 135mm.

     

    What if you want to get an 85mm FOV? Then you have to go to a 50mm lens, and you've got more bokeh problems. Plus the issue that 50mm is pretty borderline for portraits unless you want that type of effect.

     

    Beyond all of that though, there are the laws of physics to deal with. As you pack more and more photosites into a given area, the image quality goes down. Yes, noise removal helps, but it is never as good as having less noise to begin with. Better glass helps, but as we all know good lens glass is expensive...Very expensive. And even the best lens can't change the laws of physics.

     

    Finally, if you want to get a 1.6x effect, just crop your full-frame image on the computer! You get exactly the same thing as you would have shooting with a 1.6x camera. (As long as the photosites are roughly the same size on the 1.6x and full frame that is. Otherwise you will get a slightly lower-resolution version of the 1.6x image.) So there is really no need to stay with 1.6x sensors once the cost of the sensors come down.

     

    So you can get a 1.6x effect by cropping a full frame sensor image in post-processing... But you can never get extreme wide angles from a 1.6x. Want a true 24mm T&S image? Too bad, there's no lens that can do it. Or how about getting a real 12mm image out of the Sigma 12-24? Nope, never. Fisheye images? I could go on and on, but I think the point is clear already.

     

    And yes, prices will come down. With every new technology there are always people who believe it will stay expensive forever. And it never does...

     

    Personally I can't wait! Affordable (ie under US$2000) full frame DSLR? BRING IT ON!

     

    Ian

     

    http://hobday.net/photos

  15. Not to nit-pick, but it isn't the larger sensor that reduces noise. It's larger photocells/pixels that reduces noise.

     

    If Canon made a full-frame sensor with photocells the same size and density as those used on their new 8mp P&S we could expect similar noise levels as the P&S suffers from now.

     

    This distinction is important because pixel counts are likely to continue rising. Multilayer sensors will help, but only until they also reach high densities.

     

    Once this happens we may start to see even more layers added for different wavelengths of light. Or maybe another solution will be found. In any case we are only at the very beginning of high-quality digital photography... Things are going to change and improve at a pace the camera industry is not used to at all.

     

    Ian

  16. Kirk Darling , jul 06, 2004; 08:27 p.m.<br>

    > They say that about non-Canon lenses, too.<br>

    > They also say that about non-Canon Compact Flash cards.<br>

    <br>

    Sure. But the difference is that faulty lenses and faulty flash cards don't <i>explode</i> if they go bad...<br>

    <br>

    Just a minor point... ;)<br>

    <br>

    Ian

  17. I can't answer your question as to which one you should get -- I think that is basically decided by what you are going to shoot with it.<br>

    <br>

    However, if you go 180mm I strongly recommend the Sigma 180mm F3.5 lens. There were some compatibility problems with earlier versions of this lens, but anything you buy new now will work without trouble.<br>

    <br>

    Optically it is absolutely stunning. A few sample images shot with this lens on a Canon EOS D60:<br>

    <br>

    <a href="http://hobday.net/macro/">http://hobday.net/macro/</a><br>

    <a href="http://hobday.net/parkmacro/">http://hobday.net/parkmacro/</a><br>

    <br>

    Ian

  18. This is a problem of physics. As photocells get smaller noise levels go up and image quality goes down.

     

    To combat this there are two options: 1) Better lens glass 2) Larger photocells.

     

    You can see an example of #1 in Canon's new 8mp P&S that uses L glass. Even with L glass though, reviewers are commenting on the higher noise levels, especially at high ISOs.

     

    Further, lens making technology is a very mature thing. It is not going to suddenly get better in leaps and bounds. It isn't likely to get cheaper, either.

     

    Sensors on the other hand are going to improve very quickly and get cheaper almost as fast.

     

    So, the path to reducing noise levels at all ISOs is by making the photocells larger. This means larger sensors and eventually multi-layer sensors.

     

    As for dynamic range, a good DSLR already exceeds the dynamic range of slide film and approaches that of negative film. Over time we will see 48bit images from pro DSLRs. This would give much wider tonal range per colour and would also make good B&W shooting possible too.

     

    Ian

  19. > Jim Larson , jun 17, 2004; 01:05 p.m.<br>

    > Personally, I don't give ONE WIT regarding full frame.<br>

    > I am more interested in; <br>

    ><br>

    > a) Improved dynamic range<br>

    > <br>

    > b) Improved high ISO performance. <br>

    <br>

    Jim,<br>

    <br>

    One of the ways to improve the areas you suggest is to have a sensor with larger photocells or to have less densely packed photocells.<br>

    <br>

    How can this be done? Well, either go backwards to 4mp or less 1.6x sensors or go to larger sensors.<br>

    <br>

    Full frame is the future. Beyond that will be Foveon-like multilayer sensors.<br>

    <br>

    Ian

  20. Speculation:

     

    I think Canon is going to announce TWO new DSLRs this autumn. Or maybe even three.

     

    Various rumors have circulated on the net, some more believable than others. After reading through everything, this is what I think Canon will do:

     

    10D will finish production.

     

    New "7D" based on the current 10D body (which is in a lot of ways a digital version of the 7 already.) New sensor with better quality high ISO, same 1.6x crop, same 6.3mp. Probably using Digic II and the new flash system. Price a bit lower than the current 10D.

     

    New "3D" based on the EOS 3 body. I think this will have the same 8.2mp sensor as the 1DM2 has, but with slower (perhaps 5fps?) shooting. Autofocus and metering technology based on what is in the EOS 3 now. Magnesium body. EOS 3 style with available vertical grip / battery pack. I feel that this will run between US$2000 and $2500.

     

    The 3rd one? I think that Canon might replace the 300D this year too. The D70 seems to be a better all-round camera, so... My thoughts? Same basic body as the current 300D, same sensor as the new 7D. More features than the 300D has now, but fewer than the 7D of course. I think they'll drop the price by about US$100 too.

     

     

    Why do I think this way? Well, there have been rumors in the Japanese magazines of a new DSLR similar to an EOS 7. And there have been rumors that the 10D will move up higher... Neither really makes sense by itself as both would leave a huge gap in Canon's offerings. Only releasing a "3D" would leave too big a gap between the 300D and the 3D. Likewise only releasing a 7D would leave too big a gap between the 7D and the 1DM2.

     

    As for the replacement of the 300D, I think this makes sense because having two similar cameras (300D & 7D) with different sensors and electronics would be more expensive to manufacture. Also see the comment about the D70...

     

    Back to Canon releasing both a 7D and a 3D, look at Canon's film SLRs:

     

    300; EOS 7; EOS 3; EOS 1V/EOS 1V HS

     

    These nicely split the market into various segments, and it has worked well for Canon... If it works, why change it?

     

    300D; 7D; 3D; 1Ds / 1DM2

     

    Obviously there are differences between the 1V/1VHS and the 1Ds/1DM2, but since you can't use different films in a DSLR, having two distinctly different high-end DSLRs makes a lot of sense...

     

    Anyway, these are just my thoughts based on what I have read and observed.

     

    Currently I have a D60. If the 3D comes out as I suggest above I will buy one immediately.

     

    Ian

×
×
  • Create New...