Pixel density and lens quality

michael_young3 · February 12, 2010

<p>Ahh, fuhget about it. DLA/DCF crashes headlong into output-referred criterion, primarily DoF, and can't be so readily reconciled. It'll take a larger brain than mine that has spent more time thinking about this to make sense of it.<br /> <br /> It is still worth noting, though, that the effect and concerns are real, and a relatively new phenomenon with the very fine pitch, small sensors. Previously, DoF conflicted minimally with DLA. The important apertures of f/11 and f/16 had relatively little loss of detail with diffraction. To that end, the following tables list the equivalent megapixels of detail for various apertures smaller than DLA, for each of several bodies for comparison. I included the 300D, since that is my most recent frame of reference.<br /> <br /> The pixels/mm values come from the same equivalence of DLA = N = 1.6 * pixel_pitch, but reversed to solve for pixels/mm. If this is valid, the equivalent MP values are valid. The short summary version is this: for apertures smaller than DLA, effective resolution depends only on sensor size, not pixel spacing or density. Bigger, of course and as always, is better.</p>

<tbody>

<td width="93">Pixels/mm</td>

<td width="70">Equiv. MP</td>

</tr>

</tr>

</tr>

</tr>

</tr>

</tr>

</tbody>

</table>

<tbody>

</tr>

</tr>

</tr>

</tr>

</tbody>

</table>

<tbody>

</tr>

</tr>

</tr>

</tr>

</tbody>

</table>

<tbody>

</tr>

</tr>

</tr>

</tr>

</tr>

</tbody>

</table>

<tbody>

</tr>

</tr>

</tr>

</tbody>

</table>

<tbody>

</tr>

</tr>

</tr>

</tr>

</tbody>

</table>

g dan mitchell · February 12, 2010

<p>Michael, am I misreading you, or are you suggesting that I'd get less detail at f/11 on a 18 MP cropped sensor camera than I would at f/11 on a 8MP cropped sensor camera if the same lens is used in both cases?</p>

alan_bryant1 · February 12, 2010

<p>If this pissing contest continues much longer, I'm going to start building an ark :-)</p>

michael_young3 · February 12, 2010

<p>Dan, they should be the same. Are they not? Oh. The 300D's sensor is a few square mm larger, 22.3mm for the 7D, 22.7mm for the 300D. In other words, if they are diffraction limited, the resolution depends on area alone.</p>

michael_young3 · February 12, 2010

<p>An 8 MP APS-C (22.3mm wide) sensor is diffraction limited at f/10.3. Its performance at f/11 is degraded by diffraction to the same effective resolution as other diffraction limited APS-C sensors at that aperture: 7 MP .</p>

g dan mitchell · February 12, 2010

<p><em>An 8 MP APS-C (22.3mm wide) sensor is diffraction limited at f/10.3. Its performance at f/11 is degraded by diffraction to the same effective resolution as other diffraction limited APS-C sensors at that aperture: 7 MP .</em></p>

</blockquote>

<p>OK, I think we're on the same page. Neither sensor image is more or less degraded by diffraction at any aperture.</p>

michael_young3 · February 12, 2010

<p>Yup. It also doesn't change with sensor size. The figures above maybe muddied things by including Equiv. MP. The important number is pixels/mm, which is really lines/mm, a more common definition of sharpness. It also translates more readily to print size.</p>

<p>Notice that the 5D2 sensor appears to have been engineered for about 150 lines/mm, or f/11. Maybe a different way to look at the 7D is as the center crop of a (non-existent) 46 MP full frame sensor. That's relevant for birders shooting for very fine plumage details at wide apertures with their super-lens primes. No real need to end the MP wars just yet, and conversely hasn't been compelling need to chase escalating MP's for some time. It just depends on what you need, and if you know how to wring it out of the equipment you own.</p>

g dan mitchell · February 12, 2010

<p>Well, it does change with sensor size. A diffraction blur of a given measured size on the sensor is smaller as a fraction of the overall frame width as the format size increases. This is why you can shoot at smaller apertures on larger formats without giving up image sharpness.</p>

<p>The fact that sensor/film size makes a difference to potential resolution and selection of aperture was a given with film and continues to be with digital capture. (Whether the difference does or does not make a difference to a given photographer is a different question.)</p>

michael_young3 · February 12, 2010

<p>Call it what you want, Dan. The lens is still diffraction limited to 150 lines/mm at f/11, no matter what you stick behind it. It would help me, but I've already given up hope of it, if you'll separate output side references from strictly input side discussions. Doing so might even help you keep from sounding like such a muddled mess. The only thing _I'm_ discussing here is whether the sensor can resolve and capture those 150 lines/mm. If it can, the sensor is diffraction limited at that aperture. If it can't, it isn't.</p>

g dan mitchell · February 12, 2010

<p>"separate output side references from strictly input side discussion?" Pardon me?</p>

<p>And thanks for the advice on how I can avoid "sounding like a muddled mess." Much appreciated.</p>

michael_young3 · February 13, 2010

<p>If you're going to pout, we're not gonna play. I had mixed feelings about including the equivalent MP. They do drive home the point that a 1Ds can be reduced to the same pixel dimensions as my cellphone. Mayhaps you're right. MP numbers reach a gut-level understanding that the more abstract lines/mm figures don't.</p>

<p>As for not losing sharpness at small aperture... What have we been talking about if not losing sharpness?</p>

yog_sothoth · February 13, 2010

<p>Keep in mind that a Bayer sensor (or any real sensor) may need more MP than the theoretical minimum to resolve those line pairs. </p>

Sign In

Pixel density and lens quality

Recommended Posts

Link to comment

Share on other sites

Top Posters In This Topic

Popular Days

Top Posters In This Topic

Popular Days

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Create an account or sign in to comment

Create an account

Sign in