I guess the D7000 is almost mainstream now

[woolly1]

<p><em>If you were to look at that photo at the same printed ( or displayed size ) this extra pixel density should allow you to capture more detail</em></p>

<p>John, how would you manage to print to the same physical print size without downsizing the image of one of them to say 300 dpi? Otherwise to print at their native size would produce one print 16/12ths larger area than the other. Wouldn't that defeat the purpose of printing to compare detail. You would have to print the 12mp image at 300dpi and the other at 350dpi to do what you say. This is on the basis of Shun's post<br>

<a href="../photodb/user?user_id=24372">Shun Cheung</a> <a href="../member-status-icons"><img title="Moderator" src="../v3graphics/member-status-icons/mod.gif" alt="" /><img title="Subscriber" src="../v3graphics/member-status-icons/sub10plus.gif" alt="" /><img title="Frequent poster" src="../v3graphics/member-status-icons/3rolls.gif" alt="" /></a>, Mar 28, 2011; 02:27 a.m</p>

[ilkka_nissila]

<p><em>Then please show us a pixel-level JPEG either of the entire image or at least a crop of those three people in the middle to demonstrate your point.</em></p>

<p>Fine. The depth of field 0.8m was estimated using a circle of confusion 0.03mm which corresponds to 3.5 pixels in 100% crops. Person 1 on the floor, close-up.</p><div></div>

[ilkka_nissila]

<p>Neck on person 2 on the left.</p><div></div>

[ilkka_nissila]

<p>Foot of person 1; the out-of-focus boundary on the bottom left of this crop belongs to a fourth person behind the three in the foreground.</p><div></div>

[ilkka_nissila]

<p>Face of person 3 on the right in the foreground. Transition bands of a couple of pixels are considered in focus by the 0.03mm CoC.</p>

<p>Though this is an ISO 1600 image and would not be helped by the use of a higher resolution camera available today, given a bit more light, many key areas of these images I am sure would show more detail if shot with a D3X. However, typical indoor lighting is what it is. I think if shot under more controlled conditions, I would clear the background of people and use flash. But to me, images shot in real-world conditions are more interesting, even if they usually are less "perfect".</p>

<p> </p><div></div>

[ShunCheung]

<blockquote>

<p>Then please show us a pixel-level JPEG either of the entire image or at least a crop of those three people in the middle to demonstrate your point.</p>

</blockquote>

<p>Ilkka, I would like to see either a pixel-level JPEG of the whole image or at least the entire area covering those three people, as show in the red rectangle below, not some tiny areas you select.</p>

<p>Of course, using a D3X at ISO 1600 for the purpose of more resolution is silly, but that is another issue.</p><div></div>

[ilkka_nissila]

<p>Ok, please delete these four posts (12.41pm 12.52pm), I will send you a jpg in e-mail.</p>

[johnw63]

<p>Clive,</p>

<p>"John, how would you manage to print to the same physical print size without downsizing the image of one of them to say 300 dpi?"</p>

<p>Simple. Don't print them both at the same DPI. If the higher resolution camera can use more dots, let it print more dots. It seems like a simple correlation between sensor pixel density and printed dot density.</p>

[szrimaging]

<p>John, that theory only works as long as your printer can print the higher DPI. After that, down sizing will occur, either from the printer or the computer.</p>

[ilkka_nissila]

<p>Typical Espon photo inkjets have a native resolution much higher than 300ppi. Michael Reichmann reported noting visual improvements in print quality up to about 1000 ppi (resolution of the input file) in an article a few years ago. I have observed improvements going from 360 to 720 ppi myself but it was with an earlier model, the R800, that's a glossy preferred model; haven't checked my current Pro 3800 for that.</p>

[johnw63]

<p>It seems odd, that people always seem to want MORE mega pixels , but then continue to down size the output for web, or print, and then harp about lens quality issues.</p>

<p>I'm new to the digital scene, but I would think the goal would be to keep as many of those expensive pixals as you could.</p>

<p> </p>

[paul_g9]

<p>John, I only wanted more megapixels because I do bird photography and sometimes I just have to crop more than I'd like to. I can afford a D7000, but a 600mm lens is out of the question!</p>

 

[woolly1]

<p>@ John<br>

As I said in my third sentence.<br>

<em>You would have to print the 12mp image at 300dpi and the other at 350dpi to do what you say. </em></p>

<p>As for the rest of this thread I have difficulty in understanding why ultimate lens resolution is getting soooo important a hair to split. These abstract threads .... "I guess the D7000 is mainstream now".... for example, always seem to end up in discussions near scientific. Most photographers and amatuer shooters leave the obsessed tech-heads to argue among themselves.</p>

<p>I'm off to the galleries now.</p>

[szrimaging]

<p>Ilkka, the resolution on a R2880 is 5760x1440 for 13x19. This works to be roughly 300dpi. Printing higher than the maximum does not "technically" yield better results. Now, as you send more than this to the printer it will be able to better figure out the exact breakdown as it down samples the image, but ultimately, you are not gaining resolution in your print.</p>

[ilkka_nissila]

<p>Zach, the 5760 and 1440 are the dots per inch (DPI) values that Espon claims for the printer along the two directions. How do you make 300 dpi out of that?</p>

<p>Here is a comment regarding my printer, the 3800:"http://people.csail.mit.edu/ericchan/dp/Epson3800/faq.html#native_res</p>

<p>If this document is correct, by using another resolution than 360 or 720 ppi the driver will first resize the data to one of those resolutions (depending on whether "Finest detail" setting is on). The glossy-friendly R1900 apparently takes 720ppi files (it has smaller droplets than the "Pro" series) as native or resizes the files to that resolution if given something else, in all photo modes.</p>

 

[szrimaging]

<p>I said roughly. I plugged the numbers into photoshop and let it calculate a DPI.</p>

<p>Either way, you just admitted that 360ppi and 720ppi are max resolutions on your printer, and that the printer is either upsizing or downsizing an image as it sees fit. So you just proved my point.</p>

[ilkka_nissila]

<p><em>I said roughly. I plugged the numbers into photoshop and let it calculate a DPI.</em></p>

<p>I don't understand at all how you derive 300 DPI out of 5760x1440. Please tell me how you calculated that?</p>

<p><em>Either way, you just admitted that 360ppi and 720ppi are max resolutions on your printer, and that the printer is either upsizing or downsizing an image as it sees fit. So you just proved my point.</em></p>

<p>My point is that printers which take and utilize higher input file resolutions than 300ppi are very common (e.g. the R1900) and therefore doing a comparison of two cameras with different file sizes is easy. All you have to do is pick a print size so that the native resolution is not exceeded by the file from either camera. Thus downsampling will not occur. Something like premium glossy paper will show differences more clearly than matte or semiglossy.</p>

[szrimaging]

<p>Ilkka, make a document with those pixel dimensions in PS. Then go into the image size dialog. In there make sure that resample (or whatever it is called) is turned off and then set the longest side to 19". The DPI listed worked out to be something to the effect of 303 (if I recall correctly). Now, that all said, there is a problem with my logic in that the short side (1440) only works out to be about 1/3 the size it should be. That tells me there is a whole different level of math going on in there that is shrinking my vertical resolution while maintaining my horizontal, or that a print dot is a vertical rectangle instead of a square pixel. Follow? So my math is fuzzy, and needs some work. There is a whole world of image alteration going on in that printer that probably both of us have no clue about.</p>

<p>I'm not arguing how common higher resolution printers are, I was arguing based on theoretics. It wasn't to say you were totally incorrect, just that you need to maintain proper controls and cases.</p>

[User_2000406]

<blockquote>

<p>However, some rumored 24MP Sony chip is meaningless to me, and so are those random model numbers such as some "D9000."</p>

</blockquote>

<p>"D9000" is not some random model number. It should be quite obvious by now that after using up most of the two-digit model numbers that end in "0" for its DX cameras in only a few years, Nikon is going to four-digit numbers for these cameras and will use this range of numbers more carefully so it will last longer. There's a D3000 series (the D3000 and D3100 so far), a D5000 series (the D5000 and D5100 so far), and a D7000 series (so far just the D7000). Each of these series will progress to higher numbers over time, e.g. D7100, D7200, etc.</p>

<blockquote>

<p>I wouldn't confuse marketing with a product plan. When the DX lineup is finished, it should look something like this:</p>

<ul>

<li>D3100 (14MP)</li>

<li>D5100 (16MP?)</li>

<li>D7000 (16MP)</li>

<li>D400 (16/18MP?)</li>

</ul>

</blockquote>

<blockquote>Luke, as compared to the lineup I listed above, where would a "D9000" fit in? I don't really understand what it would be.</blockquote>

<p>So why a D400? Nearly everyone assumes that the D300 will be succeeded by a D400, but as I note above Nikon is clearly going for a four-digit number system for DX cameras. It's only a matter of common sense that there will be no D400, but instead a D9000 (and later a D9100, etc.)</p>

<p> </p>

[ShunCheung]

<p>Bill, model numbers are just names; they don't mean anything. Without details about features, capabilities, price, and time of availability, it is all empty talk.</p>

[User_2000406]

<p>But in this case Nikon's gradual change to a new model-numbering system led to questions over whether Nikon would stay with four levels of DX cameras or instead change to five levels. My real point is that there is no indication at this time that Nikon will change to five levels of DX cameras any time soon (if they ever do). </p>

<p>None of this really matters much to me at this time, as I finally have my D7000 which will focus all my AF lenses, meters with my manual focus AI lenses, has MLU, and offers high performance at a reasonable cost, weight, and size. It'll be a long time before I upgrade to another DSLR. My task now is to learn to use my D7000 well and to learn post-processing. Up till now I've used digital cameras for snapshots and film for serious photographic efforts. Most of my images here are digital snapshots shot as JPEGs.</p>

[craigpjenkinsphotography]

<p>I'll tell ya'll this; I just upgraded from the D90 to the D7k and am blown away by it! The D90 was no slouch in its own regard either. I still have my nose in the manual every time I want to change a setting but this will taper off with use. I have the impression the D7k represents all the features of the FX format in a DX package. I expect to be using this camera a very long time.</p>