RE: how does image stabilisation work?
One problem with Canon IS that it must be turned off when the camera is on a tripod. Otherwise it create its own problems. Vibration is actually introduced as it is predictive of motion! Sincerely, C.Brendemuehl Politics is supposed to be the second oldest profession. I have come to realize that it bears a very close resemblance to the first. Ronald Reagan ___ Join Excite! - http://www.excite.com The most personalized portal on the Web!
Re: how does image stabilisation work?
On 24/9/04, CRB, discombobulated, unleashed: One problem with Canon IS that it must be turned off when the camera is on a tripod. Otherwise it create its own problems. Vibration is actually introduced as it is predictive of motion! This is not true of all IS lenses. Certain lenses have an awareness of tripod use and compensate accordingly. IS can be turned on or off with a simple switch, and there are several modes for different types of movement encountered during photography. This page explains more: http://www.dlcphotography.net/TripodAndIS.htm Cheers, Cotty ___/\__ || (O) | People, Places, Pastiche ||=|www.macads.co.uk/snaps _
Re: how does image stabilisation work?
On 2004-09-23 17:14, Martin Trautmann wrote: But how does it actually work? - how many millimeters is the sensor shifted for stabilisation? as a side note: is it shifting horizontally and vertically only? Overlaying these two movements you can move diagonal as well. But you can't compensate for rotations. Rotations are very typical at the moment you press the release button - and thus rotate the camera clockwise. Regards Martin
Re: how does image stabilisation work?
They do that at the expense of image smear. That would not be good for stills. rg Graywolf wrote: Tell that to the video camera makers, they apparently don't know that. -- Gonz wrote: Do you mean electronically? That would not work. Image stabilization can only be done mechanically, either at the lens or on the film/sensor plane. [EMAIL PROTECTED] wrote: That seems unlikely, it would be far cheaper to just do it digitally. -- Martin Trautmann wrote: Hi all, I just had a look at the Konica Maxxum 7D / Minolta Dynax 7D. I'd wish Pentax would provide some kind of image stabilisation as these models do. But how does it actually work? - how many millimeters is the sensor shifted for stabilisation? - what's the typical mass of a sensor and the acceleration values of this stablilisation (e.g. compared to the stabilisation within the lens) - how does it operate when you are already 'at the edge'? Does it fail to work or is it always operation on a virtually centered object? are all those systems working reasonably well, claming about 2-3 aperture values gain? How much is the current price of this function - and is it something that can be expected for every future camera to come after e.g. the two next years? Regards Martin
Re: how does image stabilisation work?
I read the article about the new Minolta in pop photo, but thinking back now I am confused by something. I don't have the magazine here to check, but I swear they said they put a 500 mirror lens on the camera and that the image stabilized through the viewfinder when they engaged IS. How could they see this? If it's an SLR, they are looking through the optics only, and the sensor is not involved with the viewfinder image. If IS is done by the sensor, how could they have seen any live stabilization. I may have missed the obvious here (wouldn't be the first time) or maybe I'm remembering the article incorrectly. Steven Desjardins Department of Chemistry Washington and Lee University Lexington, VA 24450 (540) 458-8873 FAX: (540) 458-8878 [EMAIL PROTECTED]
Re: how does image stabilisation work?
I would have thought that a maximum sensor shift of 1-2mm would be sensible to give 1-2 stops improvement. It's about 10% of the linear image dimension. Anything more than this wouldn't be compensating for camera shake, it would be used for earthquake stabilisation. Nick -Original Message- From: Martin Trautmann[EMAIL PROTECTED] On 2004-09-23 18:45, Alin Flaider wrote: MT - how many millimeters is the sensor shifted for stabilisation? Anyway not more than 4 mm vertically (the gaps to full frame). Giving it some short computation, the max. image circle of 43 mm would permit ± 10 mm upwards, ± 8 mm sidewards.
Re: how does image stabilisation work?
Martin wrote: MT I just had a look at the Konica Maxxum 7D / Minolta Dynax 7D. MT I'd wish Pentax would provide some kind of image stabilisation as these MT models do. Hoepfully you are young enough to see your wish come true. ;o) MT But how does it actually work? MT - how many millimeters is the sensor shifted for stabilisation? Anyway not more than 4 mm vertically (the gaps to full frame). And because of the sensor having to cover more than the APS image circle, I expect this IS breed won't work with full frame sensors. MT - what's the typical mass of a sensor and the acceleration values MT of this stablilisation (e.g. compared to the stabilisation MT within the lens) No idea, but shaking the sensor certainly poses more problems of reliability, correct registration, etc. MT - how does it operate when you are already 'at the edge'? MT Does it fail to work or is it always operation on a virtually MT centered object? MT are all those systems working reasonably well, claming about 2-3 aperture MT values gain? It depends on the focal length, but most A1/A2 owners agree on 2 stops gain. MT How much is the current price of this function - and is it something that MT can be expected for every future camera to come after e.g. the two next MT years? If successful, every APS midrange camera but Pentax. We will wait patiently until technology matures. :oT Servus, Alin
Re: how does image stabilisation work?
That seems unlikely, it would be far cheaper to just do it digitally. -- Martin Trautmann wrote: Hi all, I just had a look at the Konica Maxxum 7D / Minolta Dynax 7D. I'd wish Pentax would provide some kind of image stabilisation as these models do. But how does it actually work? - how many millimeters is the sensor shifted for stabilisation? - what's the typical mass of a sensor and the acceleration values of this stablilisation (e.g. compared to the stabilisation within the lens) - how does it operate when you are already 'at the edge'? Does it fail to work or is it always operation on a virtually centered object? are all those systems working reasonably well, claming about 2-3 aperture values gain? How much is the current price of this function - and is it something that can be expected for every future camera to come after e.g. the two next years? Regards Martin -- graywolf http://graywolfphoto.com/graywolf.html
Re: how does image stabilisation work?
Do you mean electronically? That would not work. Image stabilization can only be done mechanically, either at the lens or on the film/sensor plane. [EMAIL PROTECTED] wrote: That seems unlikely, it would be far cheaper to just do it digitally. -- Martin Trautmann wrote: Hi all, I just had a look at the Konica Maxxum 7D / Minolta Dynax 7D. I'd wish Pentax would provide some kind of image stabilisation as these models do. But how does it actually work? - how many millimeters is the sensor shifted for stabilisation? - what's the typical mass of a sensor and the acceleration values of this stablilisation (e.g. compared to the stabilisation within the lens) - how does it operate when you are already 'at the edge'? Does it fail to work or is it always operation on a virtually centered object? are all those systems working reasonably well, claming about 2-3 aperture values gain? How much is the current price of this function - and is it something that can be expected for every future camera to come after e.g. the two next years? Regards Martin
Re: how does image stabilisation work?
On 23/9/04, Bob W, discombobulated, unleashed: Do you mean electronically? That would not work. Image stabilization can only be done mechanically, either at the lens or on the film/sensor plane. or you could stick the world in a clamp... Didn't you know? If you look closely at satellite images of the south pole, you can just make out a humungous tripod bush, it's either 2/3 or 3/4 mile across, I forget. Cheers, Cotty ___/\__ || (O) | People, Places, Pastiche ||=|www.macads.co.uk/snaps _
Re: how does image stabilisation work?
Now, Rob, think about what you just said. No one has claimed that Pentax is going to retrofit IS to the *istD. If they build a camera with IS in the body they can well choose a sensor that can do that. Remember the rumor I reported here that Pentax is working on designing and producing their own sensors. Of course that is one of the advantages to IS in the lens, it works with older cameras. It is apparently also far more precise, going by those video cameras again. --- Rob Studdert wrote: On 23 Sep 2004 at 15:13, Graywolf wrote: Tell that to the video camera makers, they apparently don't know that. The difference is the type of sensor in the *ist D (and I suspect most other cameras of similar capability) reads the full frame and can only read the frame 3 times a second. Any sensor for feedback has to be much quicker than this response time. Rob Studdert HURSTVILLE AUSTRALIA Tel +61-2-9554-4110 UTC(GMT) +10 Hours [EMAIL PROTECTED] http://members.ozemail.com.au/~distudio/publications/ Pentax user since 1986, PDMLer since 1998 -- graywolf http://graywolfphoto.com/graywolf.html
Re: how does image stabilisation work?
On 23 Sep 2004 at 20:39, Graywolf wrote: Now, Rob, think about what you just said. I did, I did, I just didn't lay out all my reasoning. No one has claimed that Pentax is going to retrofit IS to the *istD. If they build a camera with IS in the body they can well choose a sensor that can do that. Remember the rumor I reported here that Pentax is working on designing and producing their own sensors. Firstly I wasn't alluding to a retro-fit just pointing to the technologies currently chosen. Currently high res high quality sensors tend to be mechanically shuttered slow full frame reading (1-10fps) sensors not fast interline (15-30fps+) sensors. The CCD in the *ist D can manage 3fps (but it's not quite that fast in practical terms in the *ist D implementation), this is also apparently the sensor adopted for use in the recently announced Minolta DSLR in which it appear they have successfully implemented anti-shake technology. I would be keen to see a cross section of the camera or a little more discussion to find out how they derive the corrective signal and how it is applied for that matter. A hint might be if the anti-shake is linked to focus point etc? As you know I'm a little sceptical of Pentax becoming involved in sensor production, they just aren't big or experienced enough to do it (for much the same reason Ferrari don't make their own F1 tires :-). In any case even if Pentax did start making their own sensors they would likely be polishing older designs not doing ground breaking research and dev (LOL), so magic new sensors from Pentax aren't likely IMHO :-) Of course that is one of the advantages to IS in the lens, it works with older cameras. It is apparently also far more precise, going by those video cameras again. Optical IS is a reality and is relatively easy to implement but I'm still not sure how real time electronic stabilization could be implemented for still cameras. Electronic video stabilization really can't be compared to still camera stabilization, the techniques and requirements are totally different. Video stabilization only has to be done at an interframe level not per frame unlike still shots. Did I do OK? Rob Studdert HURSTVILLE AUSTRALIA Tel +61-2-9554-4110 UTC(GMT) +10 Hours [EMAIL PROTECTED] http://members.ozemail.com.au/~distudio/publications/ Pentax user since 1986, PDMLer since 1998
