A Very Dirty EOS-1D Mark III!

One of the many features of Reikan FoCal is the Dust Analysis test, and we wanted to share a great example of the results you can get from a truly filthy Canon EOS-1D Mark III!

Our Dirty old Canon EOS-1D Mark III

Have a look at this…

Each of those coloured circles is a detected dust spot!  

The size of the circle is proportional to the dust spot size – it’s not the actual spot size as they’re mostly too small to see on a whole-sensor view like this, so the software has made them a bit bigger for clarity.

The colour shows the range of apertures that are affected – green circles only affect a few narrow apertures, whereas the red ones affect a wider range of apertures.  The ones with “HP?” in are potential hot-pixels.

Running the Dust Analysis Test

The test is quick to run – just focus to infinity and cover the front of the lens with a piece of white paper and hit Start. It takes a few minutes to take shots across the aperture range and you can watch the results happening during the test.

Typical quick setup for the Dust Analysis Test

This is what it looks like in the the FoCal software:

That’s a dirty sensor!  Over 200 dust spots showing up at narrower apertures, and still 10 plainly visible wide open at f/5.6

Choosing Aperture Range from the chart list shows us the image above.  I clicked on a spot to show the crops too (the selected spot is the one in the top image with the affected aperture range shown underneath it – f/5.6 to f/32).

At f/5.6 the crops look a little messy:

At f/32… eeek!

How much will this really actually affect any shots I make with this camera? That’s where the Dust Perception Factor (DPF) comes in – choose it from chart the list and you can see:

The DPF has a threshold – the blue line on the chart – which is the approximate level that you’d notice the dust spots against a clear blue sky.  As you can see in the chart above, the spots on this sensor are far more noticeable than the blue line.

Incidentally, if you want to add any charts from FoCal to your own document/web pages etc, you can hit CTRL+C to copy to the clipboard and then paste into whatever you like – here’s an example of the copied chart from the above screenshot:

Another view we can look at is the Dust Spot Opacity.  This shows how dark the spots appear on the images:

And FoCal even detects potential hot pixels – so those spots that appear bright instead of dark:

And a screenshot view from the software, with the crop showing a clearly bright spot:

The report is here if you’re interested in having a detailed look through the results.

Have you used the FoCal Dust Analysis Test and been surprised by the results? Let us know your feedback!


2 comments on “A Very Dirty EOS-1D Mark III!

  • Rich/Dave

    Is there any way this can differentiate dust from grease / oil ?
    If you could do this you would have a killer feature on your hands.
    We are seeing roughly 1 in 20 cameras that we sensor clean with excessive oil / grease on the shutter / mirror mechanism and this is not just the 1DX Mk1 & 2’s we are talking about (which are notorious for this) . We have had a Nikon D810 and D500 in the last 2 days with this problem.

    If this could be done you / us / the user would have “hard”evidence in which to go back to the camera manufacturers with. As is stands you have to fight your case with Canon, Nikon is not so bad as i now has a really good contact so it is a bit easier to get sorted.

    Regards

    Anthony

    Reply
    • Hi Anthony,

      Thank you for your interest in the beta releases and for providing feedback and ideas!

      I’ve been having a think about this, I do remember the case with one of the Nikon camera models that would spray lubricant on the sensor (I think Nikon ended up doing a recall in that case).

      In terms of FoCal being able to work out the difference between dust and lubricant, it’s likely to be very tricky. At the moment FoCal tries to determine between a hot/stuck pixel and dust and in that case the difference is a little tricky but the determination is reasonably accurate.

      My initial feeling is one way to ‘prove’ a lubricant issue is by designed experiment, clean the sensor, take evidence dust image. Put the shutter on high continuous and fire for 5 seconds (or something!) then take another image, repeat for further 5 seconds type thing etc.

      Dust Analysis test within FoCal is one of the few areas that hasn’t been significantly modified in the 2.9 release. It is an area for which we have some interesting ideas… watch this space 🙂

      Many Thanks,
      Dave

      Reply

Leave a Reply

Your email address will not be published. Required fields are marked *