Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
06-30-2015, 07:43 AM   #1
Veteran Member




Join Date: Aug 2013
Location: Chicago, IL
Posts: 439
How much larger is a stabilized FF circle vs non-stabilized?

I assume that the new FF lenses have to cover a larger image circle because the sensor moves due to stabilization. How much larger is this image circle compared to a non stabilized FF image circle? I wonder because I assume the larger the circle the greater the vignetting from a FF compatible APS-C lens.

06-30-2015, 07:56 AM   #2
Pentaxian
Fogel70's Avatar

Join Date: Sep 2009
Location: Stockholm, Sweden
Photos: Albums
Posts: 3,062
A few tenth of millimeters extra is enough to cover the movement of the sensor for SR.
But most lenses ever produced have a over sized image circle as to handle the tolerances in manufacturing, so I don't believe vignetting is much of an issue even on old FF lenses.

And the sensor will be centered on the image circle when the image is captured.
06-30-2015, 07:57 AM   #3
Site Supporter
Site Supporter
jatrax's Avatar

Join Date: May 2010
Location: Washington Cascades
Photos: Gallery | Albums
Posts: 12,991
This has been asked and debated endlessly in posts going back years. Many years.

The end result of hours of debate and thousands of posts is: No one knows except Ricoh. And they are not saying.
06-30-2015, 08:12 AM   #4
Administrator
Site Webmaster
Adam's Avatar

Join Date: Sep 2006
Location: Arizona
Photos: Gallery | Albums
Posts: 51,584
I also think than another end result of all the debating is that it doesn't really matter. Vignetting can easily be corrected in post or in camera.


Adam
PentaxForums.com Webmaster (Site Usage Guide | Site Help | My Photography)



PentaxForums.com server and development costs are user-supported. You can help cover these costs by donating or purchasing one of our Pentax eBooks. Or, buy your photo gear from our affiliates, Adorama, B&H Photo, KEH, or Topaz Labs, and get FREE Marketplace access - click here to see how! Trusted Pentax retailers:
06-30-2015, 08:31 AM   #5
Otis Memorial Pentaxian
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 42,007
QuoteOriginally posted by Fogel70 Quote
A few tenth of millimeters extra is enough to cover the movement of the sensor for SR.
I used to think that until I did (just for kicks) a 4 second exposure Live View with lens off while looking at the sensor. To be honest, I was alarmed.

QuoteOriginally posted by jatrax Quote
This has been asked and debated endlessly in posts going back years. Many years.
...with no definitive answers. I suppose some enterprising test engineer could devise a dolly with controlled frequency and amplitude to actually get some data. The huge variable, of course is the lens itself. Traditionally, most lenses designed for 35mm film have had fairly generous image circles, much more so than those designed for APS-C, so I would not expect any problems.

I suppose we could see if any of the Sony A7II users out there have had any problems.

As for correction...yes, it is easy...unless the vignette is asymmetrical.


Steve

Last edited by stevebrot; 06-30-2015 at 10:36 PM. Reason: Maded factual
06-30-2015, 08:56 AM   #6
Senior Member




Join Date: Nov 2014
Posts: 113
QuoteOriginally posted by jatrax Quote
This has been asked and debated endlessly in posts going back years. Many years.

The end result of hours of debate and thousands of posts is: No one knows except Ricoh. And they are not saying.
That's remarkable, because it's rather trivial to find out:
Take any stabilized body + manual lens, dial in a long focal length, take off the lens and see how far the sensor moves when you tilt the body. Scale this up to FF and you pretty much know how big your image circle needs to be.
I tried it and it's much much more than fracions of a milimeters. Actually, it's quite impressive how quick and far that thing can move.

Also, you can try the composition adjust feature of the later bodies to get an idea how far off-center that sensor can shift.
06-30-2015, 09:02 AM   #7
Pentaxian




Join Date: Oct 2011
Location: Albuquerque, NM
Posts: 6,027
QuoteOriginally posted by Fogel70 Quote
And the sensor will be centered on the image circle when the image is captured.
No it isn't (necessarily). The sensor is moving around in the stationary (relative to the inside of the camera) image circle, that's how it works. If it was guaranteed to be in the center, that's the same as not moving at all, i.e. no SR.

06-30-2015, 09:09 AM   #8
Pentaxian
normhead's Avatar

Join Date: Jun 2007
Location: Near Algonquin Park
Photos: Gallery | Albums
Posts: 40,442
The most reliable guess I've seen are somewhere in the neigborhood if 1/4 inch larger, room to move 1/8 inch, (3mm) in any direction. Since the patent for the 50-200 is online, we know it's images circle is that much bigger than it needs to be for a non-SR camera, the sketchy part being , we don't know how much bigger it has to be, to avoid vignetting.

I have noticed that my 18-135 vignettes on one side only, usually the top in images with SR turned on, which would lead me to believe, it's image circle should have been a little bigger. The reason for that would be the camera barrel moving lightly downward when I push the shutter release and the Shake Reduction moving the sensor to the top of the image circle.
06-30-2015, 10:15 AM   #9
Pentaxian
Fogel70's Avatar

Join Date: Sep 2009
Location: Stockholm, Sweden
Photos: Albums
Posts: 3,062
QuoteOriginally posted by stevebrot Quote
I used to think that until I did (just for kicks) a 4 second exposure with lens off while looking at the sensor. To be honest, I was alarmed.


Steve
I bet that shutter speed is way beyond the specification of the SR system. Maybe something like 10 stops of SR.

Most of the sensor movement is before and after the image is captured, But then it does not matter if the movement is outside the image circle as no image is captured.

---------- Post added 30-06-15 at 19:17 ----------

QuoteOriginally posted by vonBaloney Quote
No it isn't (necessarily). The sensor is moving around in the stationary (relative to the inside of the camera) image circle, that's how it works. If it was guaranteed to be in the center, that's the same as not moving at all, i.e. no SR.
The sensor start moving when the shutter is pressed and during the mirror is moving it accelerates and is positioned to be able to capture the image centered over the image circle. If it moves 4/10 mm during exposure it will be moving +- 2/10 mm outside of the image circle.

---------- Post added 30-06-15 at 19:24 ----------

QuoteOriginally posted by romay Quote
That's remarkable, because it's rather trivial to find out:
Take any stabilized body + manual lens, dial in a long focal length, take off the lens and see how far the sensor moves when you tilt the body. Scale this up to FF and you pretty much know how big your image circle needs to be.
I tried it and it's much much more than fracions of a milimeters. Actually, it's quite impressive how quick and far that thing can move.

Also, you can try the composition adjust feature of the later bodies to get an idea how far off-center that sensor can shift.
This is not really relevant as the sensor is moving during a much longer time than the SR system support.

Let say the focal length is 400 mm and you need 1/400s shutter speed to get a pixel sharp image without SR, the 3-4 stops of SR would be during 1/25 - 1/50s. How much is the sensor moving during that time?

The composition adjust feature might not be possible on a FF DSLR. Or maybe not support as much movement as on APS-C

Last edited by Fogel70; 06-30-2015 at 10:26 AM.
06-30-2015, 10:30 AM   #10
Pentaxian




Join Date: Oct 2011
Location: Albuquerque, NM
Posts: 6,027
QuoteOriginally posted by Fogel70 Quote
The sensor start moving when the shutter is pressed and during the mirror is moving it accelerates and is positioned to be able to capture the image centered over the image circle. If it moves 4/10 mm during exposure it will be moving +- 2/10 mm outside of the image circle.
Well obviously it can only go so far, but it must move as the exposure is being made (or it would have no effect), and how much it must move depends on the focal length and amount of shake it is trying to compensate for. So it can't always be in the center even if it is "centered in the center" because of course is it.

One other way to test it would be to mask off a lens to make a visible circle vignette and then see how much asymmetry you can cause by shaking...
06-30-2015, 12:35 PM   #11
Otis Memorial Pentaxian
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 42,007
QuoteOriginally posted by Fogel70 Quote
I bet that shutter speed is way beyond the specification of the SR system.
Wrong question. The SR system does not tune to shutter speed. The SR system does track for motion and to the best of my knowledge does not differentiate as to whether the shutter is open or closed while it does so. The key is what it is doing when the green rectangle is displayed.

My statement was simply to show that the sensor is capable of compensating for significant displacement well over a few tenths of a millimeter over the length of a typical exposure. The case for movement at 10mm/s over the span of a 1/10s exposure is not far-fetched with lenses of moderate focal length. Ditto for movement at 100mm/s over 1/50s exposure. Note that the latter case would generate 2mm displacement.


Steve
06-30-2015, 12:39 PM   #12
Otis Memorial Pentaxian
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 42,007
QuoteOriginally posted by Fogel70 Quote
Let say the focal length is 400 mm and you need 1/400s shutter speed to get a pixel sharp image without SR
Good luck on that one. You may get acceptably sharp for display at moderate magnification with that shutter speed hand-held on a moderate resolution APS-C sensor (say 10 Mpickles), but not pixel sharp. Not even Heie is able to pull that one off at 16Mpx or 24Mpx APS-C. Now for FF 35mm format the situation is a little more loose, but even so, the old 1/focal length rule will still result in blur at the pixel level. I can't say for sure where the boundaries are since I don't own a FF digital camera. I do still shoot 35mm film and while I will shoot as slow as 1/15s with a 50mm lens, if the results must be critically sharp, I am on tripod at any exposure longer than 1/250s with that same lens.

QuoteOriginally posted by Fogel70 Quote
the 3-4 stops of SR would be during 1/25 - 1/50s. How much is the sensor moving during that time?
Quite a bit, I would suspect, though I can't say for sure beyond my general impressions when attempting to hand-hold at those focal lengths.

It just occurred to me that a quick and dirty estimate might be had by doing the laser pointer trick used on a few of the tripod reviews on this site.


Steve

Last edited by stevebrot; 06-30-2015 at 12:53 PM.
06-30-2015, 01:20 PM   #13
Veteran Member
glasbak's Avatar

Join Date: Apr 2009
Location: Netherlands
Posts: 369
The amount of sensor shift is very easy to calculate, but you need to know how much shake (angular displacement) you want to compensate and the focal length of the lens.
06-30-2015, 02:19 PM   #14
Pentaxian
Fogel70's Avatar

Join Date: Sep 2009
Location: Stockholm, Sweden
Photos: Albums
Posts: 3,062
QuoteOriginally posted by stevebrot Quote
Wrong question. The SR system does not tune to shutter speed. The SR system does track for motion and to the best of my knowledge does not differentiate as to whether the shutter is open or closed while it does so. The key is what it is doing when the green rectangle is displayed.
Of course it is. it is based on what shutter speed you can get acceptable result without SR on. The specification on the SR then tell you how many stops of longer shutter speed you can use. If you get the same blur with 0.25 - 0.5s without SR with the focal length you tested, as you get with 4s with SR on if SR has 3-4 stops specification.

The thing is that for all focal length the sensor move the same distance for the same amount of stops compensated, the sensor just moves faster the longer the focal length get.

QuoteQuote:
My statement was simply to show that the sensor is capable of compensating for significant displacement well over a few tenths of a millimeter over the length of a typical exposure. The case for movement at 10mm/s over the span of a 1/10s exposure is not far-fetched with lenses of moderate focal length. Ditto for movement at 100mm/s over 1/50s exposure. Note that the latter case would generate 2mm displacement.


Steve
I believe the speed and exposure above is too high to be within the specification on the SR. 100mm/s at 1/50s with 4 stops of SR would mean that you accept 1/800s at 100mm/s shake without SR on, which is 0.125 mm blur on the sensor. Which would be a blur of 25 pixel on a 24MP FF. I doubt anyone accept that.
06-30-2015, 03:50 PM   #15
Otis Memorial Pentaxian
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 42,007
QuoteOriginally posted by Fogel70 Quote
The thing is that for all focal length the sensor move the same distance for the same amount of stops compensated, the sensor just moves faster the longer the focal length get.
Ok...this is a leap. Please provide a reference.

QuoteOriginally posted by Fogel70 Quote
100mm/s at 1/50s with 4 stops of SR would mean that you accept 1/800s at 100mm/s shake without SR on, which is 0.125 mm blur on the sensor. Which would be a blur of 25 pixel on a 24MP FF. I doubt anyone accept that.
Nice piece of arithmetic. Now lets calculate back putting a few numbers in place. What should we say is the acceptable amount of blur? How about zero detectable? The pixel pitch for a 24Mpx FF sensor is about 9μm (0.009mm) on the diagonal so undetectable would be 4.5μm or less. That calculates* to a maximum non-SR displacement of 72μm (0.072mm) or less than 0.2% of the sensor diagonal. I doubt anyone would accept that either.

Apparently acceptable motion blur is somewhat less than pixel perfect.


Steve

* A four stop correction (published best performance of Pentax SR) is always 16x maximum acceptable linear blur.

Last edited by stevebrot; 06-30-2015 at 03:59 PM.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
amount, body, circle, ff, ff circle vs, image, k-mount, length, lens, mm, movement, os, pentax, pentax lens, pitch, sensor, shutter, slr lens, sr, stabilization, system, time
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
SR Stabilized,Not Ready? bento2 Pentax K-5 & K-5 II 7 12-10-2012 09:36 AM
How much BF/FF adjustment is too much? JinDesu Troubleshooting and Beginner Help 19 07-11-2012 03:06 AM
Benefit of Pentax DSLR's - Image Stabilized Prime Lenses dugrant153 Pentax DSLR Discussion 59 04-14-2010 11:53 AM
Has anyone used a stabilized Sigma lens with their Pentax yet? ftpaddict Pentax SLR Lens Discussion 17 09-06-2009 12:43 PM
M42 (Super Takumar 50mm f1.4) not stabilized on K100D messthetics Pentax SLR Lens Discussion 6 09-30-2007 07:45 PM



All times are GMT -7. The time now is 09:46 PM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top