Really interesting dust removal test

cmason

This is a great article: http://pixinfo.com/en/articles/ccd-dust-removal/

In this they test 4 new camera systems dust removal systems, and include photos of the results. Very very interesting, especially considering how big a factor this dust removal plays in purchasing these new cameras. I know a number of folks swayed to the new Pentax or Sony because of this feature alone.

Bottomline ranking according to effectiveness:

1. Olympus: good
2. Canon: poor (we are disappointed)
3. Pentax and Sony: useless (we are very disappointed)

http://pixinfo.com/en/articles/ccd-dust-removal/

Mike Lane

I wonder how the 1dIII will do. It supposedly has a completely different cleaning system if I remember correctly.

dogwood

Mike Lane wrote:

I wonder how the 1dIII will do. It supposedly has a completely different cleaning system if I remember correctly.

I could be wrong about this-- but I think the 1DIII plan is to let the camera map where the dust is, and then eliminate it from the image using hardware.

But an interesting test-- none of my camera bodies have dust removal-- but it's sure a feature I want! But only if it works...

Olympus was the first to introduce this technology (if I remember correctly) so maybe the others just have to catch up?

Mike Lane

From the 1dIII whitepaper...

The new, compact Self-Cleaning Sensor Unit for the APS-H size imaging sensor has been designed for the EOS-1D MarkIII and is different from the EOS Digital RebelXTi/400D Digital's unit. (The Dust Delete Data specifications are the same as with the EOS Digital Rebel XTi/400D Digital.) On the front infrared-absorption glass, two thin, single-layer piezo-electric elements are attached. By applying ultrasonic vibration to the infrared-absorption glass, the adhering dust is shaken off. The removed dust particles are stuck onto absorbent material around the infrared-absorption glass. Also, to prevent dust from entering the sensor unit, the assembly is secured with sealing material around the perimeter. Another camera maker vibrates an extra glass plate dedicated to dust removal. Because the EOS-1D MarkIII vibrates the infrared-absorption glass directly, the optical performance is not degraded by an extra layer of glass and the unit can be kept compact. The Self-Cleaning Sensor Unit can therefore be incorporated in a conventional size body.

It also mentions the dust delete data that can be imported into DPP 3.0.

Pezpix

Of course, part of the problem for Canon is the way certain bodies seal. Lord knows, the 5D should have had the rubber seal in the first place to help stop dust.

Even on my 1DsMII body, I still get a ton of dust, even with both the body, and lens having weather sealing though...

gtc

dust
I exclusively use primes and change lenses a fair bit during the day-most of the dust comes from inside the rear element cavity of my telephotos.

Some time is now spent keeping these clean and now my dust problems have reduced to an acceptable level.

I wonder what effect these piezo vibrations will have on sensor life over the long term ,in the real world?

Harryb

The results are not surprising. Sensor dust is so overstated its getting funny. The simple fact is that sooner or later your sensor will get dirty. How often depends on your style of shooting and the conditions you shoot under. I shoot in very dusty, pollen filled areas so my camera sensors are getting dirty quite frequently. Cleaning the sensor takes a few minutes and is quite easy.

An anti-dust system in the camera is nice but it would never be a reason for me purchasing a camera.

Jeffro

Harryb wrote:

The results are not surprising. Sensor dust is so overstated its getting funny. The simple fact is that sooner or later your sensor will get dirty. How often depends on your style of shooting and the conditions you shoot under. I shoot in very dusty, pollen filled areas so my camera sensors are getting dirty quite frequently. Cleaning the sensor takes a few minutes and is quite easy.

An anti-dust system in the camera is nice but it would never be a reason for me purchasing a camera.

In fact if a camera had the anti dust feature, and I couldn't shut it off, that might just make me look elsewhere. Cleaning my sensor once in awhile, when it needs it, seems more reliable than some vibrating sensor...which is just another thing that can fail. I refer to it as a gimmick.

But I could be alone on this one......

PS....and just where does this dust go when it's shook off? Oh I know, it's laying there in front of your sensor waiting to jump back on....don't you think?

cmason

What is most surprising to me is that when you look at the photos, the dust removal feature appears to do NOTHING AT ALL. You are no better off with it. A big 'benefit' of the 400/XTi is shot to hell.

Even with the system, I would think one would have to still do a manual clean from time to time, so there is really no impact, but it is odd that something that sounds so good is so useless. It seems It basically goes along with the 10MP sensor...it does no good, and seems to be simply designed to get a sucker to buy it. Maybe that is cynical, and the systems simply need more development. Btw, the 400/XTi also has the software dust removal, but that was not tested.

Jeffro

cmason wrote:

What is most surprising to me is that when you look at the photos, the dust removal feature appears to do NOTHING AT ALL.

Like I said....gimmick.

Bob&Glennie

Just as a possible point of interest : we've had an Olympus E1 in regular use for over 2 years and it's definitely not babied. We also have an Olympus E500 in regular service for just over 1 year. Both have the Oly dust removal feature and in-camera pixel mapping. Neither camera has ever had the sensor cleaned and there's still no spots. A great deal of the work that we do with these cameras ends up being printed at sizes up to 20x24 inches ... No spots!

The E500 has had to be mapped once because a dead spot showed up after a loooong 45 minute night exposure that overheated the sensor.

In all fairness we are extremely happy with our choise to go with Olympus.

Bob

claudermilk

Harryb wrote:

The results are not surprising. Sensor dust is so overstated its getting funny. The simple fact is that sooner or later your sensor will get dirty. How often depends on your style of shooting and the conditions you shoot under. I shoot in very dusty, pollen filled areas so my camera sensors are getting dirty quite frequently. Cleaning the sensor takes a few minutes and is quite easy.

An anti-dust system in the camera is nice but it would never be a reason for me purchasing a camera.

Just like so many aspect of photography now, many are making mountains out of molehills (and in some cases have to go build the molehill to begin with).

I look at it this way: more and more people are catching on to the fact that the megapixel number isn't the be-all, end-all spec to judge a purchase by, so they need a new, simple gimmick to point at to sell with. Since the dust molehill has gotten built into Mt. Everest, the marketers are taking that ball and running with it.

I think we all here know that a Pec-Pad and drop of Eclipse will work far better and more reliably than any gimmick anti-dust system offered today. Myself, I've only had to go to that extent to clean my snesor twice...in three years. Yup, dust is a pressing, daily problem. And the world will stop spinning if I don't figure out an immediate solution. poke

dogwood

Jeffro wrote:

PS....and just where does this dust go when it's shook off? Oh I know, it's laying there in front of your sensor waiting to jump back on....don't you think?

Well, most camera makers claim the dust gets stuck to a little strip of adhesive. But... the question remains-- what happens if that adhesive gets filled up like a lint roller? Can you change the adhesive strip?

dogwood

Mike Lane wrote:

From the 1dIII whitepaper...


It also mentions the dust delete data that can be imported into DPP 3.0.

Aaahhh-- didn't look at the whitepaper. Thanks for the info Mike. This last part (the dust delete data) was what I was referencing.

MarkR

The most interesting thing about the article to me is how *effective* a simple air blower is, especially compared to all this high-tech vibrational stuff.

Nikolai

The Hobbyist wrote:

The most interesting thing about the article to me is how *effective* a simple air blower is, especially compared to all this high-tech vibrational stuff.

I agree...
OTOH, blowers were used for centuries, while this new "dust techMology" is only a few years old, at least on consumer market...

simmo

The side mirrors on an old Lexus car I had vibrated to clean themselves. They had enough size and surface area to generate enough force to shake off some dust.

The size of a dslr sensor (yes even a full frame canon sensor) is too small IMO for this to be an effective way of removing dust.