Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
11-23-2013, 02:27 PM   #16
Pentaxian




Join Date: Apr 2007
Location: Romania
Posts: 8,414
QuoteOriginally posted by Nicolas06 Quote
Maybe I miss the point, but if the camera and sensor move by 1mm, horizontally, this mean the whole scene is moved by 1/23.6 = 4.2% horizontally. This is huge and sure to be noticed.
Nope.
There's a huge difference if you move the camera (including the lens and the sensor), or if you only move the sensor - camera position being fixed. If you move the camera horizontally 1mm, you're also moving the whole scene 1mm.

11-24-2013, 09:19 AM   #17
Pentaxian




Join Date: Jan 2009
Location: Nevada, USA
Posts: 465
QuoteOriginally posted by Nicolas06 Quote
Maybe I miss the point, but if the camera and sensor move by 1mm, horizontally, this mean the whole scene is moved by 1/23.6 = 4.2% horizontally. This is huge and sure to be noticed.

A rotation of 1 impact depend mostly of what is your field of view. The wider angle you have, the less visible. But still 1 shake is more than enough to be visible.
Try this thought experiment:

Take a shot of a 24m wide house so it fills the frame, with the house centered. Then move left 12 m, keeping the camera parallel to the house and shoot againg with the left edge of the house in the center of the frame. The camera shifted 12m but the image on the sensor shifted only 12 mm. So the camera moves 1000 times farther (1000mm / m) than the image on the sensor moves in this case (the 1:1000 ratio is the magnification ratio in this case).

To translate to sensor shift shake reduction, for this focal length and distance, a 1mm camera shift requires 1/1000mm = 0.001mm = 1 micron sensor shift.

Now do the same for shooting a 24mm wide flower with a 1:1 macro lens. A 12 mm shift in camera position here results in a 12 mm shift in image position on the sensor. And a 1mm camera movement results in 1mm image movement on the sensor. Not so good! So your point noting the angle of view dependency is correct. And that's why Pentax cameras ask for focal length when you mount a manual lens.

Last edited by cfraz; 11-24-2013 at 09:25 AM. Reason: Fixed unclear language
11-24-2013, 10:32 AM   #18
Pentaxian




Join Date: Nov 2013
Posts: 4,532
Yes I got it... It because it not the sensor alone that move 1mm, but the whole camera! Thanks.
11-27-2013, 01:19 AM   #19
Pentaxian




Join Date: Sep 2013
Location: Pasadena, CA
Photos: Albums
Posts: 1,103
I can think of two related engineering problems with a full frame sensor SR: increased inertia of the sensor and the thickness (size) of the device.
The SR is ran by electromagnets, and a bigger sensor means a longer pitch to correct for and stronger magnets to stop down the sensor right where it needs to be, when it needs to be stopped. These two values mean a lot more energy and a lot bigger magnets.

More energy is obviously a problem. Higher inertia will reduce effectiveness of SR and stronger magnets at could also mean electromagnetic interference to the sensor. Such interference could be subtracted by software, but they will not be gone completely. This is just a guess.

The size could also be a problem, since they are trying to make a small dSLR. In fact pentax has filed a patent for a slimmer SR system (US20130194442 A1) - but my guess would be that it's for a K-02. I do see a lens in the recent filings, that looks like a long tele zoom.

Interestingly, there is a patent for an in-lens image stabilization and it is meant for a compact lens (US20120219275 A1).

11-27-2013, 09:52 AM   #20
Pentaxian
ChristianRock's Avatar

Join Date: Jul 2013
Location: Marietta, GA
Photos: Gallery | Albums
Posts: 4,048
I would be OK with a full frame that doesn't have SR, or with a less effective SR that doesn't move as much, as long as you can really crank up the ISO (even more than our APS-C cameras). Then you can shoot with higher shutter speeds and SR doesn't become as big an issue. And the bigger/longer lenses can have SR on the lens.

Or, it could have the regular SR that the K-3 has, but on some lenses we would know that it would produce vignetting if it was applied strongly. I think most FF primes would be fine, but for zooms it could be an issue. Then we, the intelligent Pentax camera users, would know to turn off SR for those lenses (or there might be a 1/2 SR setting, like there is for AA). And we'd shoot differently prioritizing shutter speed over ISO/aperture, or we'd use a tripod. See, no big deal! Of course, the complication lies in the fact that it's the superzoom users, who buy an expensive camera to use it as a point and shoot, will see the vignetting and having no knowledge of what to do with the camera's buttons, might then freak out. But I always thought that Pentax was a brand more oriented to people who know what they're doing.... I'd be fine with a solution that leaves control in the hands of the users.

Of course all the new full-frame zooms that Pentax would release would not have this problem, and that might actually help them sell some glass

See, I'm a "the glass is half full" kinda guy.
11-27-2013, 11:14 AM   #21
Pentaxian
Fogel70's Avatar

Join Date: Sep 2009
Location: Stockholm, Sweden
Photos: Albums
Posts: 1,737
QuoteOriginally posted by ChristianRock Quote
I would be OK with a full frame that doesn't have SR, or with a less effective SR that doesn't move as much, as long as you can really crank up the ISO (even more than our APS-C cameras). Then you can shoot with higher shutter speeds and SR doesn't become as big an issue. And the bigger/longer lenses can have SR on the lens.

Or, it could have the regular SR that the K-3 has, but on some lenses we would know that it would produce vignetting if it was applied strongly.
Most image stabilization systems are optimized so the sensor or optics are centred in the image circle when the image is captures. AFAIK Pentax SR works this way too, so if SR on a FF sensor only has to take care of up to 4 stops of image stabilization it will not be much of a problem with vignetting on existing or future K-mount lenses.
11-28-2013, 02:21 PM   #22
Veteran Member




Join Date: May 2012
Posts: 698
Let's do some simple math.

Camera has a 50mm lens and the subject is 5 metres away.

Camera moves sideways 1mm without changing direction (no rotation)

Image on the sensor moves 1 x 50 / 50,000 = 0.001mm (approximately but very close)

On a 24MP APS-C sensor each pixel is 23.6mm / 6000 = 0.00393mm

SO the movement of the image (shake) on the sensor is one quarter of a pixel or 0.004% of the image width.

edit : oops sorry I hadn't noticed that there was another whole page of posts after the post I was replying to.
12-01-2013, 08:35 PM   #23
Junior Member




Join Date: Jan 2013
Posts: 37
Wouldn't you need lenses with a larger image circle than traditional FF to use the SR?

I thought that was the whole point of the DA lenses is they were smaller than full frame but large enough image circles to take advantage of the SR.

Best,

-Q

12-02-2013, 10:09 AM   #24
Veteran Member




Join Date: May 2012
Posts: 698
In practice the image circle rarely has a definite cutoff. The image circle of a normal FF lens would still cover more than the full frame. It would just gradually lose sharpness and suffer more vignetting the further you move out of the circle. The SR movement is only about +/- 5% of the sensor width so the excursion out of the image circle will not be too much and will only affect the corners.
Apart from the areas near the corners the rest of the image will be unaffected because all parts some distance from the corners will never get outside the image circle.

There might be the occasional lens that has a very tight and sharply defined image circle, but even so it would likely only cause some dimming at one or more corners that happen to venture outside of the image circle.

In any case if one is relying on SR to keep the image steady the situation will normally require that you do not frame very tightly so there is likely to be margin for removal/correction of darkened corners. If the framing is steady and only using SR to minimize the effect of small vibrations then the sensor would be moiving very little off its central position and will remain well within the image circle.
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!
body, ff, full-frame, lens, lenses, patents, pentax, reduction, ricoh, sensor, stabilization
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Shake Reduction brent-b Pentax K-30 & K-50 6 02-17-2013 07:58 PM
Shake reduction drawbar Troubleshooting and Beginner Help 9 09-30-2012 11:13 PM
Shake Reduction Study Dionisis_K Pentax K-5 8 07-02-2012 08:27 AM
Shake Reduction Problem lukelbd Pentax K-5 4 05-13-2012 11:59 PM
Macro-shake reduction bjake Troubleshooting and Beginner Help 11 04-02-2012 03:50 PM



All times are GMT -7. The time now is 07:25 PM. | See also: NikonForums.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