GPS Pentax DSLR's

sjwaldron · 06-03-2011, 08:50 AM

Originally posted by DragonLord

A table indicating maximum Astrotracer tracking times is provided at the end of the PENTAX Japan press release for the O-GPS1. All times are in seconds.

Awesome, thanks.

The K-5 must have some physical benefits in SR or in-camera sensors because it can track longer than the K-r. My calculations lead me to believe that I can use my 300mm f4 lens for around 80 seconds at a 45 degree declination. If that's true it is pretty amazing. Without tracking I think the exposure time would only be like 1.3 seconds before trails appear.

Originally posted by mattdm

Looks like the pop-up flash can't be used with it mounted. Shame. Too bad it doesn't have a built-in IR-only controller flash.

I wonder if there are any wireless triggers out there that pass-through all of the data connections.

Originally posted by v5planet

I'm a bit skeptical of that... it would need to track the stars in both right-ascension and declination -- you could simulate this if the sensor could both rotate and tilt, but I don't know if it can do the latter and in any event would stretch the image like in a tilt-shift lens. Hm.

I thought the K-5 added another axis of movement to the SR ability? It does say on their site "Sensor-Shift Shake Reduction with rotational compensation." Don't know about tilt though.

Entropy · 06-03-2011, 09:45 AM

Originally posted by Lowell Goudge

gps don't need calibration, they know where north is all the time, the issue will be whether there is a compas display on it or not

Not true. No GPS receiver knows which direction it is pointed in, if GPS is the sole sensor source. A GPS receiver can only know in which direction it is moving.

To know which direction you are pointed, a heading sensor is required. In most hiking-oriented GPS units, this is achieved using a magnetic sensor (electronic compass) - electronic compass units usually DO need to be calibrated. (Every one in a handheld hiking GPS I know of has.)

As to the usefulness of heading - it's usually possible with a topo map or aerial imagery to know which direction the camera is pointing in. If there are no distinguishing features in the terrain or aerial imagery, the scenery probably wasn't very interesting to begin with.

I can immediately tell in which direction the camera was pointing for the majority of my geotagged images - the few cases where I can't don't justify $150 extra and the reduction in portability.

Gimbal · 06-03-2011, 09:58 AM

Found some pictures

???GPS UNIT O-GPS1?PENTAX

and

??????????GPS UNIT O-GPS1?PENTAX

ihasa · 06-03-2011, 10:51 AM

This unit is already causing a big stir in the astrophotography world. Pentax have posted samples of how the feature works on long exposures (up to 5mins on a 50mm lens). It's fair to say that people are pretty amazed by this, and it will translate into more astronomically minded people upgrading to a Kr or K5 or buying into the Pentax system for the first time. With the great noise control and now this feature, Pentax is now the go-to brand for night sky shots.

pxpaulx · 06-03-2011, 12:34 PM

Wow those are some great examples Gimbal - I wasn't able to view it, but were there exposure times attached in the exif at all? The land shot is kind of interesting, since it is keeping track of the sky, the land is blurred - I suppose one static shot could be combined to alleviate that. I am anxious to see what can be done with a straight mounted telescope for some deeper sky viewing.

Cannikin · 06-03-2011, 12:41 PM

Originally posted by pxpaulx

Wow those are some great examples Gimbal - I wasn't able to view it, but were there exposure times attached in the exif at all?

If you're talking about this page, in the captions for the photos, the number right after ISO (___秒) is seconds.

Quicksand · 06-03-2011, 01:09 PM

This is awesome -- count me in for one.

I wonder whether it will work with my old lenses. I don't see why it couldn't, provided the focal length is input correctly.

pxpaulx · 06-03-2011, 01:12 PM

Originally posted by Cannikin

If you're talking about this page, in the captions for the photos, the number right after ISO (___秒) is seconds.

Figures it had to make sense and be relatively obvious for me to miss, haha.

bwDraco · 06-03-2011, 04:13 PM

Originally posted by sjwaldron

The K-5 must have some physical benefits in SR or in-camera sensors because it can track longer than the K-r.

This advantage exists because the K-5's SR system is capable of roll correction, +/- 2 degrees. The K-r does not have this function.

**interested_observer** · 06-03-2011, 05:56 PM

One advantage that Pentax has over Canon and Nikon is in the body sensor stabilization system. With Canon and Nikon having lens based image stabilization, their GPS capability is only that of being able to location/position tag the image in the EXIF meta data. Pentax, with the its ability to move the sensor is essentially able to perform limited star tracking, and that is the difference. Up to a 300 second track is going to be very nice, especially when stacked. This coupled with the K5's ability to disable dark frame subtraction (DFS), will be a big plus - for back to back tracks. Now if they had tethering fully integrated, this would be a full function unit. (yes - I know about the open source capability provided by the Pentax community. Now, if only Pentax would get on board....)

That brings the K5/Kr into a very small club. Most astronomy uses the movement of the mounts for tracking - i.e., moving the entire mirror structure to track the object under observation. Pentax's approach is very similar to that used by the Arecibo Radio Telescope. For imaging telescopes, the HET (just a little scope out in the hinder lands of Texas) uses this design approach, where the object under observation is tracked across the stationary fixed face of the primary mirror elements.

I can say this in that I am a bit partial to HET, since I helped design the star tracking system....

With respect to the compass - here is some information on ELECTRONIC compasses (no - I do not know Pentax's actual implementation of the capability)...

v5planet · 06-03-2011, 07:05 PM

Originally posted by interested_observer

One advantage that Pentax has over Canon and Nikon is in the body sensor stabilization system. With Canon and Nikon having lens based image stabilization, their GPS capability is only that of being able to location/position tag the image in the EXIF meta data. Pentax, with the its ability to move the sensor is essentially able to perform limited star tracking, and that is the difference. Up to a 300 second track is going to be very nice, especially when stacked. This coupled with the K5's ability to disable dark frame subtraction (DFS), will be a big plus - for back to back tracks. Now if they had tethering fully integrated, this would be a full function unit. (yes - I know about the open source capability provided by the Pentax community. Now, if only Pentax would get on board....)

That brings the K5/Kr into a very small club. Most astronomy uses the movement of the mounts for tracking - i.e., moving the entire mirror structure to track the object under observation. Pentax's approach is very similar to that used by the Arecibo Radio Telescope. For imaging telescopes, the HET (just a little scope out in the hinder lands of Texas) uses this design approach, where the object under observation is tracked across the stationary fixed face of the primary mirror elements.

I can say this in that I am a bit partial to HET, since I helped design the star tracking system....

With respect to the compass - here is some information on ELECTRONIC compasses (no - I do not know Pentax's actual implementation of the capability)...

interested_observer -- I'd be curious to hear your thoughts on an issue I raised earlier since you have experience in designing conceptually similar star tracking systems. In order to properly track the sky in any situation aside from pointing directly at a celestial pole, won't the sensor need to be able to move in two axes, analogous to the altitude and azimuth tracking of many hobbyist telescopes? If it does, isn't there a concern of the sensor tilting out of alignment with the focal plane?

**interested_observer** · 06-03-2011, 07:33 PM

Evening, Well from my perspective, I would agree with you that 2 axis movement would be needed in order to able to track in a rotational environment. That said, and I don't read Japanese, it appears that the K5 does have some limited rotational capability on the sensor. Gimbal posted this link...

??????????GPS UNIT O-GPS1?PENTAX

... that tends to indicate that the K5 does have the necessary 2D capability. I wish Pentax would strike while the fire is hot and release the manual, and as much engineering data that would help the astro folks understand the capabilities along with the limitations of the unit. Obviously, there is a hard limit to the shutter time which is linked to the range of movement of the sensor. Up to 300 seconds is a pretty long duration. Running back to back frames and stacking, provides a capability that currently really does not exist out side of having a tracking mount.

The k5 body at say ISO 80 (or even maxing it out to what? 52,000), with a really sharp lens like say the 31 ltd or maybe the 14, for a wide field view, should potentially be able to produce some pretty stunning imagery. Using the sensor's dynamic range, and ability to pull detail out of shadows, should put astro work in just about anyone's hands (with out going overboard with some special tracking mounts and setup).

I have the K20 and not the K5, so the sensor's range of movement described within the K5's manual would be important here.

One item that I noticed, when you image something with stationary objects (landscape), its the landscape that gets smeared. Take a look at the images here that has some landscape elements in them... (again supplied by gimbal)..

??????????GPS UNIT O-GPS1?PENTAX

Raybo · 06-03-2011, 07:51 PM

Originally posted by v5planet

interested_observer -- I'd be curious to hear your thoughts on an issue I raised earlier since you have experience in designing conceptually similar star tracking systems. In order to properly track the sky in any situation aside from pointing directly at a celestial pole, won't the sensor need to be able to move in two axes, analogous to the altitude and azimuth tracking of many hobbyist telescopes? If it does, isn't there a concern of the sensor tilting out of alignment with the focal plane?

You will be "tracking" a small part of the sky, and yes the sensor can do both RA and DE but not like you are thinking, we are not talking 1000mm or 5000mm here, it's a 5 min max with a 300mm lens (think about it).

In this case the sensor would not need to "rotate".

Raybo · 06-03-2011, 08:02 PM

The sensor does what it's supposed to do for the first five minutes, after that you would need a clock drive, the sensor can only move so far without help.

v5planet · 06-03-2011, 08:18 PM

Originally posted by interested_observer

Evening, Well from my perspective, I would agree with you that 2 axis movement would be needed in order to able to track in a rotational environment. That said, and I don't read Japanese, it appears that the K5 does have some limited rotational capability on the sensor. Gimbal posted this link...

Sensor movement freely in an XY plane would be sufficient, or it would only be sufficient if the sensor could rotate around both an X and Y axis? (To wit, the latter is how tracking is achieved with alt-az mounts on telescopes).

Quote:

Obviously, there is a hard limit to the shutter time which is linked to the range of movement of the sensor. Up to 300 seconds is a pretty long duration. Running back to back frames and stacking, provides a capability that currently really does not exist out side of having a tracking mount.

I wonder what the battery life is like. As I understand it, the SR magnets only engage right before a photo is taken -- I had an old Canon S2 IS which had it active constantly and it drained batteries like a mother. If the magnets are engaged for 300" continuously I wonder how many frames you could get before depleting the batts. The other thing I wonder is just how much leeway the sensor has to move around because unless you rotate the camera on the tripod, stars will eventually start to drift out of frame -- does the 300" limitation simply reflect how far the sensor can move before it can't move any further? Then of course there is always field rotation to worry about...

06-03-2011, 05:56 PM	#55
interested_observer Loyal Site Supporter Join Date: Jun 2009 Location: Tumbleweed, Arizona Photos: Gallery \| Albums Posts: 5,707	One advantage that Pentax has over Canon and Nikon is in the body sensor stabilization system. With Canon and Nikon having lens based image stabilization, their GPS capability is only that of being able to location/position tag the image in the EXIF meta data. Pentax, with the its ability to move the sensor is essentially able to perform limited star tracking, and that is the difference. Up to a 300 second track is going to be very nice, especially when stacked. This coupled with the K5's ability to disable dark frame subtraction (DFS), will be a big plus - for back to back tracks. Now if they had tethering fully integrated, this would be a full function unit. (yes - I know about the open source capability provided by the Pentax community. Now, if only Pentax would get on board....) That brings the K5/Kr into a very small club. Most astronomy uses the movement of the mounts for tracking - i.e., moving the entire mirror structure to track the object under observation. Pentax's approach is very similar to that used by the Arecibo Radio Telescope. For imaging telescopes, the HET (just a little scope out in the hinder lands of Texas) uses this design approach, where the object under observation is tracked across the stationary fixed face of the primary mirror elements. Hobby-Eberly Telescope Hobby-Eberly Telescope - Gallery I can say this in that I am a bit partial to HET, since I helped design the star tracking system.... With respect to the compass - here is some information on ELECTRONIC compasses (no - I do not know Pentax's actual implementation of the capability)... http://www.afcea.org/europe/html/KazubekJakubPresentation.pps Digital Compass Sensor - Digital Compass Sensor - Magnetic Field Sensor - Tilt Sensor - magnetoresistive sensors - Electronic Compass Design - Cypress Semiconductor Magnetic sensor makes electronic compass design easy The Embedded Beat: Freescale Blog Community Last edited by interested_observer; 06-03-2011 at 06:55 PM.

06-03-2011, 07:05 PM	#56
v5planet Veteran Member Join Date: Jan 2010 Location: Seattle Photos: Albums Posts: 1,915	Originally posted by interested_observer One advantage that Pentax has over Canon and Nikon is in the body sensor stabilization system. With Canon and Nikon having lens based image stabilization, their GPS capability is only that of being able to location/position tag the image in the EXIF meta data. Pentax, with the its ability to move the sensor is essentially able to perform limited star tracking, and that is the difference. Up to a 300 second track is going to be very nice, especially when stacked. This coupled with the K5's ability to disable dark frame subtraction (DFS), will be a big plus - for back to back tracks. Now if they had tethering fully integrated, this would be a full function unit. (yes - I know about the open source capability provided by the Pentax community. Now, if only Pentax would get on board....) That brings the K5/Kr into a very small club. Most astronomy uses the movement of the mounts for tracking - i.e., moving the entire mirror structure to track the object under observation. Pentax's approach is very similar to that used by the Arecibo Radio Telescope. For imaging telescopes, the HET (just a little scope out in the hinder lands of Texas) uses this design approach, where the object under observation is tracked across the stationary fixed face of the primary mirror elements. Hobby-Eberly Telescope Hobby-Eberly Telescope - Gallery I can say this in that I am a bit partial to HET, since I helped design the star tracking system.... With respect to the compass - here is some information on ELECTRONIC compasses (no - I do not know Pentax's actual implementation of the capability)... http://www.afcea.org/europe/html/KazubekJakubPresentation.pps Digital Compass Sensor - Digital Compass Sensor - Magnetic Field Sensor - Tilt Sensor - magnetoresistive sensors - Electronic Compass Design - Cypress Semiconductor Magnetic sensor makes electronic compass design easy The Embedded Beat: Freescale Blog Community interested_observer -- I'd be curious to hear your thoughts on an issue I raised earlier since you have experience in designing conceptually similar star tracking systems. In order to properly track the sky in any situation aside from pointing directly at a celestial pole, won't the sensor need to be able to move in two axes, analogous to the altitude and azimuth tracking of many hobbyist telescopes? If it does, isn't there a concern of the sensor tilting out of alignment with the focal plane?

06-03-2011, 07:33 PM	#57
interested_observer Loyal Site Supporter Join Date: Jun 2009 Location: Tumbleweed, Arizona Photos: Gallery \| Albums Posts: 5,707	Evening, Well from my perspective, I would agree with you that 2 axis movement would be needed in order to able to track in a rotational environment. That said, and I don't read Japanese, it appears that the K5 does have some limited rotational capability on the sensor. Gimbal posted this link... ??????????GPS UNIT O-GPS1?PENTAX ... that tends to indicate that the K5 does have the necessary 2D capability. I wish Pentax would strike while the fire is hot and release the manual, and as much engineering data that would help the astro folks understand the capabilities along with the limitations of the unit. Obviously, there is a hard limit to the shutter time which is linked to the range of movement of the sensor. Up to 300 seconds is a pretty long duration. Running back to back frames and stacking, provides a capability that currently really does not exist out side of having a tracking mount. The k5 body at say ISO 80 (or even maxing it out to what? 52,000), with a really sharp lens like say the 31 ltd or maybe the 14, for a wide field view, should potentially be able to produce some pretty stunning imagery. Using the sensor's dynamic range, and ability to pull detail out of shadows, should put astro work in just about anyone's hands (with out going overboard with some special tracking mounts and setup). I have the K20 and not the K5, so the sensor's range of movement described within the K5's manual would be important here. One item that I noticed, when you image something with stationary objects (landscape), its the landscape that gets smeared. Take a look at the images here that has some landscape elements in them... (again supplied by gimbal).. ??????????GPS UNIT O-GPS1?PENTAX Last edited by interested_observer; 06-03-2011 at 07:42 PM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
For Sale - Sold: Genuine Pentax Leather Hand Strap for all Pentax DSLR's (Worldwide)	kcmadr	Sold Items	8	02-16-2011 09:02 AM
Pentax K-7 sales rank #9 at Amazon.com for DSLR's	Mr Hyde	Pentax News and Rumors	15	05-10-2010 11:42 PM
What do you think of Touchscreen for future Pentax DSLR's	LeDave	Pentax DSLR Discussion	34	03-25-2010 08:31 AM
Pentax K-7 GPS module	dealrocker	Pentax DSLR Discussion	3	02-06-2010 12:48 PM

06-03-2011, 08:50 AM	#46
sjwaldron Veteran Member Join Date: Aug 2009 Photos: Gallery Posts: 1,235	Originally posted by DragonLord A table indicating maximum Astrotracer tracking times is provided at the end of the PENTAX Japan press release for the O-GPS1. All times are in seconds. Awesome, thanks. The K-5 must have some physical benefits in SR or in-camera sensors because it can track longer than the K-r. My calculations lead me to believe that I can use my 300mm f4 lens for around 80 seconds at a 45 degree declination. If that's true it is pretty amazing. Without tracking I think the exposure time would only be like 1.3 seconds before trails appear. Originally posted by mattdm Looks like the pop-up flash can't be used with it mounted. Shame. Too bad it doesn't have a built-in IR-only controller flash. I wonder if there are any wireless triggers out there that pass-through all of the data connections. Originally posted by v5planet I'm a bit skeptical of that... it would need to track the stars in both right-ascension and declination -- you could simulate this if the sensor could both rotate and tilt, but I don't know if it can do the latter and in any event would stretch the image like in a tilt-shift lens. Hm. I thought the K-5 added another axis of movement to the SR ability? It does say on their site "Sensor-Shift Shake Reduction with rotational compensation." Don't know about tilt though.

06-03-2011, 09:45 AM	#47
Entropy Veteran Member Join Date: Jun 2007 Location: Owego, NY Posts: 976	Originally posted by Lowell Goudge gps don't need calibration, they know where north is all the time, the issue will be whether there is a compas display on it or not Not true. No GPS receiver knows which direction it is pointed in, if GPS is the sole sensor source. A GPS receiver can only know in which direction it is moving. To know which direction you are pointed, a heading sensor is required. In most hiking-oriented GPS units, this is achieved using a magnetic sensor (electronic compass) - electronic compass units usually DO need to be calibrated. (Every one in a handheld hiking GPS I know of has.) As to the usefulness of heading - it's usually possible with a topo map or aerial imagery to know which direction the camera is pointing in. If there are no distinguishing features in the terrain or aerial imagery, the scenery probably wasn't very interesting to begin with. I can immediately tell in which direction the camera was pointing for the majority of my geotagged images - the few cases where I can't don't justify $150 extra and the reduction in portability.

06-03-2011, 09:58 AM	#48
Gimbal Pentaxian Join Date: Apr 2007 Location: Sweden Posts: 2,106	Found some pictures ???GPS UNIT O-GPS1?PENTAX and ??????????GPS UNIT O-GPS1?PENTAX

06-03-2011, 10:51 AM	#49
ihasa Veteran Member Join Date: Jul 2009 Location: West Midlands Photos: Gallery Posts: 2,066	This unit is already causing a big stir in the astrophotography world. Pentax have posted samples of how the feature works on long exposures (up to 5mins on a 50mm lens). It's fair to say that people are pretty amazed by this, and it will translate into more astronomically minded people upgrading to a Kr or K5 or buying into the Pentax system for the first time. With the great noise control and now this feature, Pentax is now the go-to brand for night sky shots.

06-03-2011, 12:34 PM	#50
pxpaulx Veteran Member Join Date: May 2009 Location: Twin Cities, MN Posts: 2,867	Wow those are some great examples Gimbal - I wasn't able to view it, but were there exposure times attached in the exif at all? The land shot is kind of interesting, since it is keeping track of the sky, the land is blurred - I suppose one static shot could be combined to alleviate that. I am anxious to see what can be done with a straight mounted telescope for some deeper sky viewing.

06-03-2011, 12:41 PM	#51
Cannikin Veteran Member Join Date: Mar 2010 Location: SW Washington Posts: 833	Originally posted by pxpaulx Wow those are some great examples Gimbal - I wasn't able to view it, but were there exposure times attached in the exif at all? If you're talking about this page, in the captions for the photos, the number right after ISO (___秒) is seconds.

06-03-2011, 01:09 PM	#52
Quicksand Pentaxian Join Date: Apr 2008 Location: Central Ohio (formerly SF Bay Area) Posts: 1,519	This is awesome -- count me in for one. I wonder whether it will work with my old lenses. I don't see why it couldn't, provided the focal length is input correctly.

06-03-2011, 01:12 PM	#53
pxpaulx Veteran Member Join Date: May 2009 Location: Twin Cities, MN Posts: 2,867	Originally posted by Cannikin If you're talking about this page, in the captions for the photos, the number right after ISO (___秒) is seconds. Figures it had to make sense and be relatively obvious for me to miss, haha.

06-03-2011, 04:13 PM	#54
bwDraco Veteran Member Join Date: Jul 2010 Location: New York Photos: Albums Posts: 1,071	Originally posted by sjwaldron The K-5 must have some physical benefits in SR or in-camera sensors because it can track longer than the K-r. This advantage exists because the K-5's SR system is capable of roll correction, +/- 2 degrees. The K-r does not have this function.

06-03-2011, 07:51 PM	#58
Raybo Veteran Member Join Date: Dec 2007 Posts: 872	Originally posted by v5planet interested_observer -- I'd be curious to hear your thoughts on an issue I raised earlier since you have experience in designing conceptually similar star tracking systems. In order to properly track the sky in any situation aside from pointing directly at a celestial pole, won't the sensor need to be able to move in two axes, analogous to the altitude and azimuth tracking of many hobbyist telescopes? If it does, isn't there a concern of the sensor tilting out of alignment with the focal plane? You will be "tracking" a small part of the sky, and yes the sensor can do both RA and DE but not like you are thinking, we are not talking 1000mm or 5000mm here, it's a 5 min max with a 300mm lens (think about it). In this case the sensor would not need to "rotate".

06-03-2011, 08:02 PM	#59
Raybo Veteran Member Join Date: Dec 2007 Posts: 872	The sensor does what it's supposed to do for the first five minutes, after that you would need a clock drive, the sensor can only move so far without help.

06-03-2011, 08:18 PM	#60
v5planet Veteran Member Join Date: Jan 2010 Location: Seattle Photos: Albums Posts: 1,915	Originally posted by interested_observer Evening, Well from my perspective, I would agree with you that 2 axis movement would be needed in order to able to track in a rotational environment. That said, and I don't read Japanese, it appears that the K5 does have some limited rotational capability on the sensor. Gimbal posted this link... Sensor movement freely in an XY plane would be sufficient, or it would only be sufficient if the sensor could rotate around both an X and Y axis? (To wit, the latter is how tracking is achieved with alt-az mounts on telescopes). Quote: Obviously, there is a hard limit to the shutter time which is linked to the range of movement of the sensor. Up to 300 seconds is a pretty long duration. Running back to back frames and stacking, provides a capability that currently really does not exist out side of having a tracking mount. I wonder what the battery life is like. As I understand it, the SR magnets only engage right before a photo is taken -- I had an old Canon S2 IS which had it active constantly and it drained batteries like a mother. If the magnets are engaged for 300" continuously I wonder how many frames you could get before depleting the batts. The other thing I wonder is just how much leeway the sensor has to move around because unless you rotate the camera on the tripod, stars will eventually start to drift out of frame -- does the 300" limitation simply reflect how far the sensor can move before it can't move any further? Then of course there is always field rotation to worry about...