danielleetaylor

<p>If I'm correct there's an 11.25x conversion factor. Might be useful for Holga style spy shots.</p>

<p><em>Some of the people who claim otherwise have owned both, at the same time, and shot extensively with both side by side. </em></p>

<p>If you know any of those people, have them post test photos along with all shot and post processing details for analysis. Thank you.</p>

<p><em>Maybe other people have personal experience that trumps online opinions. </em></p>

<p>Personal experience does not trump tests which yield quantifiable, repeatable results. The 7D out resolves the 5D (see http://www.dpreview.com/reviews/canoneos7d/page28.asp and http://www.dpreview.com/reviews/canoneos5d/page31.asp), and has less noise at ISO 3200 (see http://www.dpreview.com/reviews/canoneos7d/page15.asp vs http://www.dpreview.com/reviews/canoneos5d/page21.asp). As for DR, they're close in JPEG, but the 7D is better with HTP on. Properly processed RAW files in ACR are actually about the same, a change from the original 5D review with older software. Still, the 7D is certainly the winner on this as well if you ever need to shoot JPEGs.</p>

<p><em>Perhaps you should read some of the recapped experience, as opposed to simply repeating some online opinion that agree with your own. </em></p>

<p>Just wondering before we continue this little spat: have you ever shot with either camera? Ever make 24" landscape prints from either? I can answer yes to both.</p>

<p><em>Besides, a crappy lens on the 7D isn't going to yield 'superior fine detail' vs. an excellent lens on a 5D. duh.</em></p>

<p>Who said anything about a "crappy" lens? The Tokina 11-16 f/2.8 is cheaper than a 17-40L yet yields superior IQ. It's an easy match for the Canon 16-35L II, even a bit better.</p>

<p><em>But the OP was referring to using the camera in the following manner:</em><br /><em> "Keep in mind my application: wide shots in tight areas, landscapes and low-light settings in the city."</em><br /><em> None of these applications appeal to the strengths of a crop sensor camera. </em></p>

<p>OP is comparing the 7D and 5D. The 7D is the superior camera for both landscapes and low light shooting. Period. As for wide shots in tight areas, there are lenses available to go as wide as 13mm FF equivalent.</p>

<p><em>The larger sensor (in case you've forgotten) allows less bending (and therefore distortion) of the light to achieve the same FOV. </em></p>

<p>A) You have this backwards. For a given lens registration distance, the larger the sensor the more the lens has to bend light to illuminate the entire sensor.</p>

<p>B) Increased bending of light does not translate to increased distortion. Quality of the optical design and lens manufacturing determine how well the image is focused onto the sensor. In general, it's easier to manufacture smaller pieces of glass than larger ones, though even that general principle cannot be used to prove that lens A is better than lens B. You have to compare the lenses themselves.</p>

<p><em>As optically 'good' as the best UWA crop sensor zoom on the market is, it gets blown away in regards to sharpness, distortion, color, and contrast by equiv zooms that give the same FOV on a FF sensor.</em></p>

<p>You don't have a clue what you're talking about here. I will pit the Tokina 11-16 f/2.8 against any FF UWA EF lens out there.</p>

<p><em>Sorry. For the application and limitations the OP specified, it's pretty clear that the older FF 5D (paired w/ an L UWA lens) would yield superior performance to the newer 7D.</em></p>

<p>Spoken by someone who has never tried such a comparison. I can say for certain that a 7D + Tokina will beat the 5D even if you put a Canon 16-35L II on it. The lenses will be toe-to-toe, but the 7D sensor still resolves more detail.</p>

<p>The 5D does have good IQ to this day, a testament to its sensor design. But it is not the better choice unless it can be purchased at a steep discount, especially once you factor in features, UI, used vs new with warranty, etc.</p>

<p><em>Am I crazy to go for a used 5Dmk1 with L-series lens over a 7D with the widest lens available?</em></p>

<p>Yes. The 5D <strong>does not</strong> have better high ISO. The 7D edges out the 5D there. (Though as Dave points out, both can easily do ISO 3200. In print there's no difference between the two in noise at ISO 3200.) The 7D also yields superior fine detail (quite noticeable on large landscape prints) and superior dynamic range. You can see this for yourself looking at test shots at DP Review and Imaging Resource, though the resolution difference will stand out even more in a large landscape print than in their tests.</p>

<p>I'm sorry, but people who claim otherwise have obviously not tested both, or at least reviewed public tests of both, and are simply giving a knee-jerk "full frame is always better" type response.</p>

<p>Not to mention there is a dramatic difference between the two in terms of feature set, responsiveness, user interface, and weather sealing.</p>

<p>If you won't be shooting any action, consider the 60D and put the money saved towards glass. Also, I highly recommend the Tokina 11-16 f/2.8, though the Canon 10-22 is also very good and I've heard good things about the Sigma 8-16.</p>

<p><em>I would prefer a prime as they will always be sharper than zooms.</em></p>

<p>That's not true. Sometimes professional zooms are as sharp or sharper than primes in their range.</p>

<p>That said, Eric is correct that you should experiment to determine which FL will best meet your needs in zoos. But if it's in your budget, I wouldn't ignore Canon's 70-200 zooms.</p>

<p><em>I am targeting this kit: 7D + Canon 17-55 f/2.8 IS + 70-200 f/4 IS (money saved here against f/2.8 could be used for wide angle) + Wide Angle lens Three choices for wide angle:</em></p>

<p>Tokina 11-16 f/2.8, if you can find it in stock.</p>

<p><em>The OP specified a 17-35mm. The FoV of a 17mm TS is the same as the 17-35mm that he's considering. So if the FoV is the same and the IQ is roughly equivalent when processed competently, then where do primes enter the discussion?</em></p>

<p>If you want a 17mm T/S, in 35mm terms, you have to get the 5D2 because on crop it will be a 27mm T/S.</p>

<p><em>Why limit yourself to low to mid ISO? That's one of the big advantages of the 5D2 is its high-ISO performance, lower sensitivity to under-exposure and slightly better dynamic range (vs. the 7D). Anyone doing any low-light street photography and dusk and pre-dawn scenic shooting can really take advantage of the 5D2's strengths.</em></p>

<p>The vast majority of shots are not made at ISO 3200 and then printed to 24". If you're not doing that then there's little to choose between the two sensors even at high ISO. If Gunjan is budget limited then a Sigma 30 f/1.4, which he can buy with the money saved by going crop, will give him low light capability superior to the f/4 zooms he's considering matching up with a 5D2.</p>

<p>It's just not as simple as the knee-jerk "yeah team full frame!" responses make it seem, unless of course you have bottomless financial resources.</p>

<p>marco - in the third pair the D90 shot fairs better because of better light. The Velvia shot still suffers from the color balance issue, and I would pick the color/contrast of the D90 in this case. (Again though, different scenes under different light. Try shooting same scene on both for a while.)</p>

<p>Neither is obviously over sharpened for the screen size.</p>

<p><em>Do you suggest me some settings to keep my images sharp and crisp but not oversharpened ? I like to add some hi radius low amount usm, to add some mid-tonal contrast and clarity, at amount 20, radius 50. I like it.</em></p>

<p>That's known as local contrast enhancement, and I do the same. I use Smart Sharpen (PS CS4) for detail sharpening, usually a radius of 0.5-1px and an amount somewhere between 50% and 100%.</p>

<p>The important thing is that optimum sharpening varies from image to image. On some images I've actually done two Smart Sharpen passes, one at 1px and one at 0.5. On others, I barely touch the image with 0.5px and a very low amount. It all depends on subject, light, lens, previous post processing, and target print size.</p>

<p>Do you have an example pair where you prefer the film sample? Or did you prefer the film sample in all 3 cases here? So far only the first film scan was obviously over sharpened. Perhaps sharpening is not what's turning you off. What don't you like about any of the D90 shots here?</p>

<p>marco - regarding the second pair, neither looks unnatural. The Velvia shot has more pleasing contrast but the color balance is off. The D90 shot is somewhat flat in terms of color and contrast, but that's primarily a lighting issue. Post processing could help. Velvia 50 can be magic under shaded or subdued lighting. Direct, harsh sunlight requires careful choices with digital or film IMHO.</p>

<p>You really should try shooting both side-by-side for a while so that you have comparisons under identical conditions. Literally carry both cameras and if you shoot a scene on one, shoot it with the other right away and match the settings as closely as possible. This will reveal a lot to you and help you better understand what you like and don't like and how to achieve what you like digitally.</p>

<p><em>The kashbah image is a cheap flatbed scan from 35mm velvia. The train image is taken by d90, ps edited</em></p>

<p>LOL! The over sharpening in the Kashbah image had me convinced that it was the digital one. Sorry, the train image is better. The Kashbah image is over sharpened, which is typically the problem when people complain that foreground and background don't melt together, or that their subject looks like it was cut and pasted onto the background. Yet you say that's the film scan? My guess is that workflow, particularly sharpening, is still your problem on the digital side. The comments that DoF is the issue are off base IMHO as the difference is simply not that great between 35mm and APS-C. It's something in your workflow, though the train image looks fine to me.</p>

<p>Another vote for the EOS 3. It makes a good pair with the 7D. There are even a couple EOS 3 features I wish were on the 7D (i.e. side exposure meter in the VF; multi spot metering).</p>

<p>The Tokina is an excellent lens all around, and the f/2.8 aperture is quite useful in astrophotography. Especially when trying to capture the Milky Way without using a motorized mount. f/2.8 means a lower ISO with less noise, or more stars and detail captured. I highly recommend this lens for astrophotography.</p>

<p><em>Primes and T/S lenses are not required to take advantage of a FF camera. The two zoooms listed by the OP are excellent, when proper post processing is applied. </em></p>

<p>And they offer nothing in terms of IQ over their equivalent crop lenses, yet there's a $900 premium just on the body for going FF vs. the 7D, and a $1,500 premium vs. the 60D. The sensors in those bodies are capable of prints which are just as good at low to mid ISO. And actually, as I'm fond of pointing out since I own both and have tested the 17-40L on both formats, while good the 17-40L cannot match the best crop equivalent for the price (Tokina 11-16, which is a match for the 16-35L II on FF).</p>

<p>My point is not that you need primes or T/S lenses to take advantage of FF, but that FF doesn't offer much advantage right now over crop unless you have those lenses. And I don't say that for the IQ, but for the FoV.</p>

<p><em>That old saw about print size is only true if you never change your mind and decide to have a large print made of a favorite image. Who knows what you may decide to do with a great image five years from now?</em></p>

<p>That "old saw" comes from someone who has tested both formats and who regularly makes large prints. You seem to have misinterpreted my statement as meaning the 5D2 is suddenly much better past 24"-30", and that you can't print past those sizes from a crop camera. Neither statement is true. While I can sometimes see a small improvement at 30" and above in 7D vs 5D2 test prints, there is no dramatic difference in favor of the 5D2, and most viewers would never notice.</p>

<p>Both the 18 MP APS-C and 21 MP 35mm sensors are roughly 24-30" print sensors when dealing with landscape prints viewed under critical conditions. Other subject matter is less demanding and you can go larger. Whatever your combination of subject, critical or non-critical viewer, and viewing distance, if you can acceptably reach a print size with the 5D2 you can do the same with an 18 MP crop body so long as the shot was made at low to mid ISO and properly processed.</p>

<p>Looking forward to the 5D3 can be a valid justification for a budget limited consumer building out a FF system now. If it yields >30 MP with even better noise and DR characteristics, as rumored, then it will restore some advantage to FF. Of course it's a waste of money unless you plan to actually purchase a 5D3 and use it to its full ability.</p>

<p>I respond to these threads because the typical knee jerk recommendations ignore that a) right now there isn't much of an IQ advantage to FF; b) FF comes at a considerable premium in cost; and c) that premium might be put to better uses. No one else seems to have noticed that Gunjan can get a 60D and a Tokina 11-16 and have greater effective focal range than with his FF kit (equal on the wide side; greater on the tele), and still have money left over for a nice macro lens. If he buys a couple lenses used he could even get a nice macro and a fast prime such as the Sigma 30 f/1.4. A 60D with a macro is going to beat the pants off a 5D2 without one any day when it comes to flowers. The Tokina 11-16 is going to have better edge to edge sharpness than the 17-40L for large landscape prints. And a fast normal prime is a fun lens for street and city. So which is really the better kit for him? He has to decide that based on future plans, but I certainly couldn't fault him for putting his money towards glass rather than a slightly bigger sensor.</p>

<p>Going against the grain here (as usual)...FF will not bring any real value to the table for you right now. You're not making large prints from high ISO shots, considering fast wide primes, or considering T/S lenses. And with proper post processing you will not see a difference between the 7D and 5D2 at 24" and even 30" when shooting low to mid ISO. (The 7D requires a bit more sharpening and NR, and is less forgiving of exposure error.)</p>

<p>That said, your typical subjects don't really take advantage of the 7D's unique features either. You could accomplish your goals with the 60D.</p>

<p>If you go crop, I would say get the 60D and put the money saved into glass. Get the Tokina 11-16 f/2.8 for ultra wide and a macro lens and put whatever is left over into a trip or a printer or something.</p>

<p>But if you think you're going to really want the next 5D, then getting the 5D2 kit is probably the best option so that you start building a FF lens collection now.</p>

<p>While I'm certain there will be another 5D, no one knows what it will be like. (Or at least those that do know can't talk.) I'm hoping the 5D3 will be >30 MP. But this will admittedly only be truly beneficial for those making large prints of demanding subject matter, i.e. landscapes at >24".</p>

<p><em>Does the Sigma provide enough of an improvement that one would want to use it at f/1.4, would be the 64$ question. If not, then the Canon would be fine.</em></p>

<p>I do not hesitate to use my Sigma at f/1.4. And the Sigma has significantly better bokeh at any of the larger apertures. It's not that the Canon 50mm is a bad lens. It's a very good lens. But the Sigma, being a newer design, is better.</p>

<p>I doubt there will be another Sigma lens incompatibility incident, but that's always a risk I supposed. It didn't stop me from buying.</p>

<p><em>Perhaps if the faults of their tests were better known it would be obvious to more people Mr. Taylor. Will you please enumerate and explain what you see as their flaws?</em></p>

<p>I'm not privy to the inner workings of their analytical software or why it's so often off. The closed nature of their testing would of course be considered the first flaw by many. I can replicate DP Review's tests. I cannot truly replicate DxO's tests as I would be using their software which is a black box.</p>

<p>I can only tell you that their results on specific tests are regularly at odds with sites that use more traditional, and well understood, tests. This is clearly evident in their dynamic range results for example. I've said this before, but I just had to laugh when I looked up the Canon bodies I'm most familiar with on DxO and saw the DR results. My personal testing and experience is very close to DP Review's reports, and DxO isn't even in the ball park.</p>

<p>Others have complained that DxO's test results are easily thrown off by small variances in AA filters or internal pre-processing. Something is throwing them off to produce rankings where DSLRs can supposedly out gun MF backs.</p>

<p>Here's one essay that discusses the issue further, but doesn't really have any technical explanation as to why since, again, DxO testing involves a big black box in the middle of the process: http://www.luminous-landscape.com/essays/eyes-vs-numbers.shtml</p>

<p>DxO also ranks some DSLR sensors as better than medium format digital backs that cost more than an average sedan. And if you believe them, they also have some Arizona beach front property for sale on their site.</p>

<p>Their tests are fatally flawed. I can't believe that's not obvious to more people.</p>

<p>Great shots Keith!</p>

<p>Yep. Keep your 17-55 IS and get a 7D. You'll have plenty of money left over to add glass. And I would suggest adding some fast primes to the line up for your indoor roller derby work. I haven't shot this but I'm assuming from your lens choices that you don't need a lot of reach. Look at the Sigma 30 f/1.4, Sigma 50 f/1.4, and Canon 85 f/1.8. The two stop light gain on the primes more than offsets any high ISO advantage of the 5D mkII.</p>

<p><em>I am not sure why the 5DII images look slightly sharper at low ISO but I think there are a number of possible explanations. these include the anti-alias filter, the fact that the lens does not have to work as hard and possible differences in the Canon signal processing.</em></p>

<p>It comes down to lens MTF, or as you put it, the lens not having to work as hard on the 5D mkII. For the same composition any given detail is going to occupy more space on the 35mm sensor, meaning it is rendered at a lower lpmm/higher MTF point.</p>

<p>However, the difference is small enough that the 7D can be made equal in post processing. Same for noise for the most part up to about ISO 800. (You might still see a bit more while pixel peeping, but you won't see it in print.)</p>

<p>While I agree the 5D mkII is the better high ISO camera, the 7D should not be dismissed and is capable of very good prints at ISO 1600 and 3200. Just be careful with exposure as post work as it's less forgiving of each.</p>

<p>The 70-300 does not offer any improvement over your 70-200 + 1.4x.</p>

<p>If low light is a consideration (it usually is for school sports) then grab a 7D. Your 70-200 becomes a 112-320 f/2.8. Add the 1.4x and your reach extends to 448mm at f/4. For sports I often find I can crop down to 8-10 MP out of the 18 MP and still make very good 16x20" prints. Not to mention the great AF and 8 fps.</p>

<p><em>Generally speaking, full frame sensors are recommended for landscapes over crop sensors, given their greater inherent depths of field and ability to render finer detail. </em></p>

<p>There is no significant difference between Canon's 21 MP 35mm sensor and Canon's 18 MP APS-C sensor at low to mid ISO. One will render just as much fine detail as the other, all other factors being equal. Also, 35mm sensors do not yield greater DoF, but less for a given aperture. (Though I consider this inconsequential when shooting for greater DoF. Diffraction is the limiting factor for either, and the impact of diffraction is essentially the same for all formats for a given FoV and DoF.)</p>

<p><em>If you're shooting primarily from a tripod, I'd go for a refurbished original 5D, and a used 17-40.</em></p>

<p>This will yield a moderate improvement as the 17-40L is a sharper lens than the Tamron. A greater improvement could be achieved by moving to a T2i or 60D and the 15-85 or 17-55. (I'm recommending that because Mark doesn't seem to favor the ultra wide end of his current zoom. If he did, a Tokina 11-16 would be the ticket.) And yes, the extra 6 MP matters in landscape work, especially when rendering foliage or printing big.</p>

<p>I'm not sure if that would be in his budget, but it would be the better system. If he can't upgrade both at the same time...given that he does not use the ultra wide end very much I would recommend one of the two lenses above.</p>

<p>Neither the 12 MP sensor nor the Tamron are bad, so huge improvements should not be expected after upgrading either.</p>

<p><em>The main gotcha with the EOS 1N, 3 and 1V is their AF system is only at it's best with F2.8 or faster lenses. if you use a variable aperture zoom, AF can be disappointing. Why? The cross AF point degrades to single axis with slower lenses and have difficulty snagging.</em></p>

<p>From http://photonotes.org/reviews/1-1N-3-1V/<em> - The 3 and 1V have 45 autofocus sensors, all of which can detect horizontal and diagonal lines at normal precision when used with a lens of f/8 or faster. However they also have 7 cross sensors capable of high-precision focussing. Of this group of 7, the central sensor retains high precision down to f/4 or faster and the remaining 6 cross sensors require f/2.8 or faster for high precision.</em></p>

<p>They don't all lose an axis. But 7 have variable "high precision" depending on max aperture. Then again, do you need high precision with an f/4 or slower lens? Isn't the extra precision related to precise focus vs DoF rather than speed?</p>

<p>You would have to send it back if you couldn't get it to focus properly even using Micro Focus Adjustment.</p>

<p>Don't worry about it, just run a few test shots when you get it. If they're in focus you should be good.</p>

<p>I will note that f/1.4 presents razor thin DoF and this is challenging for AF even with a perfect lens. I tend to use spot AF with my Sigma for this reason.</p>

<p>The Sigma is sharper wide open and has better (legendary in fact) bokeh. It also has better build quality and comes with the hood.</p>

<p>Of course there are two risks with the Sigma. One is that you will have to send it back to be adjusted. Two is that it may not work at some future time with a future Canon body.</p>

<p>For me the Sigma was still an easy choice, and I have not been disappointed. It is spectacular for portrait use and has amazing bokeh. I love it.</p>

<p>I went the EOS 3 route.</p>

<p>Pros: fast, sure AF. (My 7D's AF is better, but the EOS 3's is very good.) Relatively modern UI. Great VF with right side exposure scale. Built like a tank and very comfortable to hold, at least for my hands. (The body shape feels slightly better to me than my 7D.) Good CF set. Multi spot metering (love that feature and wish it was on the modern bodies.) Eye control focus if you like that. It's cheaper on eBay than a 1v.</p>

<p>Cons: loud. MLU is buried in the CF set. Missing some of the UI refinements of later bodies, but that's probably true of the film EOS 1v as well. (I'm thinking of how the top left mode buttons work. You have to hold 1-2 while moving the top dial. Modern bodies let you press one button, not hold it, then control two parameters at once with the two separate dials. Also, I miss the separate AF and AE lock buttons on the rear of the 7D.) The hinged right side door sometimes catches on your hand a bit and opens, it's not a slide-then-open design like on modern digital bodies.</p>

<p>I guess I'm comparing it more to my 7D than to a 1v. But I can't think of anything that it lacks or does poorly. I've certainly never shot with it and thought "boy, I wish I had this particular 1v feature." YMMV.</p>