macular hole
super member
Reged: 08/12/09
Posts: 143
|
|
Is anyone coming out with a true, high def, 1920x1080p astrovideo camera?
There are several 1080P ccd video imagers available now.
|
ccs_hello
Post Laureate
Reged: 07/03/04
Posts: 3298
|
|
First, there is no formal definition over the term: astrovideo.
NTSC or PAL video are Standard Definition video systems and they are analog systems.
One well-known Analog HD Video system is the long-defunct Japanese MUSE system.
HD Video, be it 720p, 1080i, or 1080p, the video transmission/delivery methods are pure digital system.
In such arrangement, they are more like the traditional astroCCD arrangement, such as described in CN "Solar System Imaging & Processing" forum, like this post. These are just the high frame-rate, better than VGA resolution, "imaging head" while the processing and storage done in the tethered PC.
Standalone consumer grade HD Video capture devices are available such as HD camcorders (compressed down and stored to digital tape, recordable DVD, or flash memory) and movie-mode capable DSLRs (compressed and stored to flash memory or delivered to a tethered PC.)
Clear Skies!
ccs_hello
|
macular hole
super member
Reged: 08/12/09
Posts: 143
|
|
I'm wondering why someone has not adapted an HD camera block like the Sony FCDH10 for use in an HD Astrovideo scenerio?
|
DFEmery
super member
Reged: 05/09/08
Posts: 118
Loc: Maine
|
|
As it was explained to me, the real answer lies in the availability of a suitable CCD chip. The MallinCam uses a chip with high sensitivity (pixel dimensions are 8.4um x 9.8um, active array is 768 x 494), and as I understand it, there is no ~1980 x ~1080 CCD chip with comparable sensitivity (at least, at an affordable price). Remember that pixel size is also important... large pixels are more sensitive than small pixels, which is why pixel binning is sometimes used with CCD cameras. So there is a tradeoff between pixel density vs. sensitivity at any given array dimension.
There are plenty of larger arrays available, but they are physically larger and not as sensitive. So the design of an astrovideo camera is a balancing act between pixel sensitivity, pixel size, and CCD array size. So the use of a higher resolution chip (according to today's technology) would almost certainly require much longer integration times to achieve the light grasp of a MallinCam.
The 'name of the game' is pixel sensitivity. Hopefully someone will develop an affordable CCD technology that will greatly increase the sensitivity of small pixels, allowing much denser arrays.
|
highfnum
super member
Reged: 09/06/06
Posts: 232
Loc: NY
|
|
(DFEmery) well written reply - in future 5 (yrs?) bigger chip with same capability should be around at a cost that is withen reach
|
macular hole
super member
Reged: 08/12/09
Posts: 143
|
|
Kodak KAI-2093 HD CCD pixel size is 7.4x7.4. Pretty close. No specs on sensitivity though.
Who else make ccd chips?
Let's find one that works.
|
CCD-Labs
Vendor (CCD-Labs)
Reged: 12/16/06
Posts: 109
Loc: Wichita Ks.
|
|
The problem is the cost vs. performace issue. The digital back end needed to tranmit the signal is basically a Laptop by the time you get everything put together and at 2x the cost. The other issue has been stated that the CCD needed to do this at video rates does not exist at any reasonable cost. If you want to do this you could cobble something together using a:
Decent CCD camera (around 2500$)
Track and Stack software (? not written yet)
Laptop (500$ minimum)
VGA to HDMI converter (around 300$)
Even at these costs it would not be "video" in the sense that most on this forum think about. You could not just plug into a monitor.
The specs for sensitivity on a CCD camera is based on a lot of things....CCD Qe, uV/-e conversion, readout noise, ADU/-e digital conversion...ect ect...ect. Even the latest major video systems using E2V CCD's are not "HD" in the sense of the word. At best you would have a HD system that could view planets or the moon.
Now for a question......would a system that allowed you to capture via a stadard Mintron type camera (which Mallincam is based on) that was able to transmit via Ethernet or diplay on a XVGA monitor be someting that folks would be interested in? You would basically be able to tote a VGA/LCD monitor in the field as opposed to a std. video monitor or transmit the video via Ethernet network to stream over the internet. This would be a purpose built piece of equipment.
William
--------------------
CCD-Labs
Affordable Astro Imaging gear
|
macular hole
super member
Reged: 08/12/09
Posts: 143
|
|
This just in-
Nikon's top of the line, soon to be released, D3s has 8.45um pixels-12.1 million of them- and an unheard of ISO capability of 102,000. That's 10 times the sensitivity of previous gens. Moreover, it has the now standard HDMI output for use with any typical HD 16:9 monitor. The only bad news is that the HD video output is "only" 720p rather than 1080P. Still, a big advance in resolution over 480i standard res TV.
From the Nikon brochure-
D movie mode offers 1280x720p at 24fps. Using the new High Sensitivity Movie mode at ISO 120,400, you can shoot movies in light too low for the eye to see well.
Sounds like plug and play, in a scope with a standard SLR adapter on it? It is pretty heavy at 46 oz.
Now you have a High def Astrovideo cam that does triple duty as a high res astrophotography cam as well as a high res terrestrial cam.
|
CCD-Labs
Vendor (CCD-Labs)
Reged: 12/16/06
Posts: 109
Loc: Wichita Ks.
|
|
ISO is not a measure of sensitivity.....only of a relelitive gain of the system compared to some old film standards. I could build a camera with an ISO of a 500K but if the S/N ratio was poor it would not be much better than a standard video camera.
For a reference you can view this write up:
http://www.dpreview.com/articles/compactcamerahighiso/
Basically nothing is for free and the price of the Nikon is going for over 5K USD. I would not beleive any hype till it is proven. I am sure someone will sooner or later try it out. Physics is Physics and you can't cheat it too much...you can play games with post processing in camera but you can't do to much.
William
--------------------
CCD-Labs
Affordable Astro Imaging gear
|
ccs_hello
Post Laureate
Reged: 07/03/04
Posts: 3298
|
|
Re; fake ISO 102400, please read this CN post.
Clear Skies!
ccs_hello
|
ccs_hello
Post Laureate
Reged: 07/03/04
Posts: 3298
|
|
Quote:
...
Now for a question......would a system that allowed you to capture via a stadard Mintron type camera (which Mallincam is based on) that was able to transmit via Ethernet or diplay on a XVGA monitor be someting that folks would be interested in? You would basically be able to tote a VGA/LCD monitor in the field as opposed to a std. video monitor or transmit the video via Ethernet network to stream over the internet. This would be a purpose built piece of equipment.
William
William,
Similar topics had been raised in this forum before. It seems to be most of this forum viewers prefer turn-key, no PC solution (but rather use video camera's/video monitor's picture adj knobs for image processing/enhancement) and use analog RS-170A video transmitting PAL or NTSC baseband signal as the common denominator. This is still true when long exposure mode (2, 4, ..., 56 seconds) is used. I.e., most of the video frames are just the same repeated frame in the previous snapshot capture.
Some people in this forum then use anlog video capture device to PC for record keeping or Internet multicast.
Clear Skies!
ccs_hello
|
macular hole
super member
Reged: 08/12/09
Posts: 143
|
|
Don't shoot the messenger.
It looks like this guy has some real possibilities, hype or no? Pixel size is on par with the above mentioned Mallin cam sensor, but here we have current HD SOTA.
It sure would be great to see live native 720P planet or moon images from the telescope to a 1080p 50 inch flat panel TV. What a boon that would be for an Astronomy class. How about feeding it to a 1080p PJ on a 100 inch screen?
I would much rather run a progressive 720p all digital signal through an HDMI cable to a 1080p flat screen and not have to bother with any conversion processing. Particularly analog.
|
ccs_hello
Post Laureate
Reged: 07/03/04
Posts: 3298
|
|
These movie-mode DSLRs in such mode are not designed for long exposure thus (in such mode) not for deep sky viewing.
Clear Skies!
ccs_hello
|
CCD-Labs
Vendor (CCD-Labs)
Reged: 12/16/06
Posts: 109
Loc: Wichita Ks.
|
|
Quote:
Quote:
...
Now for a question......would a system that allowed you to capture via a stadard Mintron type camera (which Mallincam is based on) that was able to transmit via Ethernet or diplay on a XVGA monitor be someting that folks would be interested in? You would basically be able to tote a VGA/LCD monitor in the field as opposed to a std. video monitor or transmit the video via Ethernet network to stream over the internet. This would be a purpose built piece of equipment.
William
William,
Similar topics had been raised in this forum before. It seems to be most of this forum viewers prefer turn-key, no PC solution (but rather use video camera's/video monitor's picture adj knobs for image processing/enhancement) and use analog RS-170A video transmitting PAL or NTSC baseband signal as the common denominator. This is still true when long exposure mode (2, 4, ..., 56 seconds) is used. I.e., most of the video frames are just the same repeated frame in the previous snapshot capture.
Some people in this forum then use anlog video capture device to PC for record keeping or Internet multicast.
Clear Skies!
ccs_hello
This would be stand alone, no teathered PC, But instead of RS170 it would output RGB or VGA signals.
William
--------------------
CCD-Labs
Affordable Astro Imaging gear
|
johnnyha
Pooh-Bah
Reged: 11/12/06
Posts: 1105
|
|
The improved color separation as well as increased resolution would be a quantum leap above already spectacular images with the Mallincam.
--------------------
Johnny
Spicewood, TX
Sherman Oaks, CA
|
ccs_hello
Post Laureate
Reged: 07/03/04
Posts: 3298
|
|
Quote:
This would be stand alone, no teathered PC, But instead of RS170 it would output RGB or VGA signals.
William
William,
Is it a going to be a two box solution? I.e., a camera head --- digital data cable --- processor box with knobs and levels for near real-time image processing/display.
I would imagine the big processing box can be done by a laptop motherboard with some levels and control knobs then wrap into a nice build box.
The challenge is the production volume (think beyond astro community) and application specific one-purpose price.
I suspect some will then use a laptop, MIDI control panel (as a multi-level image controlling input device), dedicated OS (Linux), and one application-specific software to do the same.
IMHO, this community should embrace COTS devices such as a PC. People are using laptop/nettop to run EQMOD to control a GOTO mount, do autoguiding, and other image capturing tasks. Skill level will have to be elevated over time. 
Clear Skies!
ccs_hello
|
macular hole
super member
Reged: 08/12/09
Posts: 143
|
|
Quote:
These movie-mode DSLRs in such mode are not designed for long exposure thus (in such mode) not for deep sky viewing.
Clear Skies!
ccs_hello
CCS,
Are you implying that the Nikon D3S may have another mode that could be used to view Deep sky in near real time, on a 1080P flat panel monitor?
|
dVnt
member
Reged: 08/31/09
Posts: 29
|
|
I don't think that's what ccs_hello was saying.
The problem is that it takes x time for enough photons to hit a CCD and create a meaningful image. As such, you'd need a CCD that is sensitive enough to produce live images instead of the stacked long exposures that produce the beautiful astro pictures we all know and love, this is the technical limitation for deep sky objects.
I suppose the moon and planets would be a more reasonable goal.
--------------------
Zhumell Z8 Deluxe Dobsonian Reflector
Zhumell 9mm 1.25" EP, Zhumell 30mm WV 2" EP
GSO 2X Barlow
Zhumell 2" Filter Kit (Skyglow, UHC, O-III, Variable Polarity)
Baader Planetarium 2" to 1.25" EP Adapter
|
DFEmery
super member
Reged: 05/09/08
Posts: 118
Loc: Maine
|
|
DVnt has it right... that's the technical problem. No issue with making a CCD array with large X-by-y pixel dimensions... the QHY-8, for example, uses a chip that is 3032 x 2016. And its one heck of a camera, but not for "live" video of DSOs. None of those 6 million pixels are individually as sensitive to light as are any one of the 400,000 pixels in a MallinCam.
So there you have it: lower resolution with greater light grasp; or higher resolution with less light grasp.
What we really need is a CCD chip with more sensitivity per pixel... not merely more pixels per square centimenter. When technology allows us to have greater sensitivity in the smaller pixels necessary for a denser array, THEN we can have the light grasp of a MallinCam at higher resolution... and hopefully at a price we amateurs can afford.
And I'll bet that day is not too far in the future...??
|
highfnum
super member
Reged: 09/06/06
Posts: 232
Loc: NY
|
|
awhile back I had e-mail dialouge
with Rock mallin about this
There are ccd chips based on GaAs (gallnium Arsnide)
they are more sensitive than silicon based chips
But you can not just drop the chip in a mallincam circuit
they require different voltage
but someday it may happen
|
CCD-Labs
Vendor (CCD-Labs)
Reged: 12/16/06
Posts: 109
Loc: Wichita Ks.
|
|
There are no GaAs based commercial CCD's in the visible light spectrum available....
William
--------------------
CCD-Labs
Affordable Astro Imaging gear
|
highfnum
super member
Reged: 09/06/06
Posts: 232
Loc: NY
|
|
what is the cut off
are u saying its only IR ( whatabout ha-line)
besides it was all just future speculation anyway
Edited by highfnum (10/20/09 07:19 PM)
|
macular hole
super member
Reged: 08/12/09
Posts: 143
|
|
Quote:
These movie-mode DSLRs in such mode are not designed for long exposure thus (in such mode) not for deep sky viewing.
Clear Skies!
ccs_hello
How about just putting the camera in multiexposure mode with a series of timed exposures or multiple exposures that are then transmitted via HDMI to hd flat panel monitor in a nearly live video feed from the OTA?
|
ccs_hello
Post Laureate
Reged: 07/03/04
Posts: 3298
|
|
Yes, repetitive (timed) long exposure can work. However, mirror will be dropped once per exposure. Also without dark subtract, picture will be noisy.
Clear Skies!
ccs_hello
|
macular hole
super member
Reged: 08/12/09
Posts: 143
|
|
I think you can manually lock up the mirror on these pro DSLR's?
|
ccs_hello
Post Laureate
Reged: 07/03/04
Posts: 3298
|
|
No: on repeated shots.
Each shot requires the mirror be dropped once.
Clear Skies!
ccs_hello
|
DFEmery
super member
Reged: 05/09/08
Posts: 118
Loc: Maine
|
|
What do you mean by "Each shot requires the mirror be dropped once" and "(the) mirror will be dropped once per exposure"?
|
ccs_hello
Post Laureate
Reged: 07/03/04
Posts: 3298
|
|
DSLR is its normal and native mode is still capture. It always starts with reflex mirror in down position. This would allow optical visual image being seen (thru pentaprism or pentamirror) on the optical viewfinder. This also allow part of the image being diverted down to the "phase differential" Auto-Focus module. During actual still shooting, reflex mirror is out of way just for the exposure duration. This sequence always gets repeated in the still shooting mode.
The exception is Liveview or movie mode.
There may be few DSLRs have special mode to defeat this standard still capture behavior (but shutter is still doing its job  .) (Mainly for super silent shooting in covert situation.)
Clear Skies!
ccs_hello
|
ragebot
Carpal Tunnel
Reged: 08/26/05
Posts: 1672
Loc: Tallahassee, FL, USA
|
|
Quote:
This just in-
Nikon's top of the line, soon to be released, D3s has 8.45um pixels-12.1 million of them- and an unheard of ISO capability of 102,000. SNIP
I have used the LiveView from my Canon 7d output to a 29 in Sony TV and the image is crispy in dim light. It provides a very nice image of the Moon when attached to a threaded 40mm 2in EP, and one of the best live images I have ever seen of the sun when the 40mm EP is put in my SO .3<A HA filter. Jupiter, Saturn, Venus, are stunning on the Sony and my HD flat screen; M42 blows peeps away. I would guess the new Canon 350 would perform in a similar way; and the same for the Nikons with LiveView. I also get nice views with my Nikon 4500 CoolPix; and suspect most other digicams would provide acceptable views.
That being said the image produced by any of these cameras when viewing M27 pales when compared to what my Mallincam produces. But I can stack and post process output from the 1920X1080 movie clips from my d7 and get very nice results.
The strength of the Mallincam is that it is almost real time with no post processing; something no other device I know of can do, except maybe a Stellacam.
--------------------
Meade ETX 90, Meade AR5, Orion ED80, Atlas GT, 8 in Newt, Coronado DS SM40, Garrett 10.5X70, Sigma SD10, SD14, Canon 1D2, Xti, Nikon CP4500, C-14
|
macular hole
super member
Reged: 08/12/09
Posts: 143
|
|
Rage,
What 2 inch 40mm eyepiece did you attach to your Canon 7D? Pentax XW 40 perhaps? What kind of adapter do you use for this?
Your Canon 7D live view report is very interesting. Thanks for posting it. However, I cannot find any specs from either Canon or Nikon on Live view's actual digital HDMI output resolution. It could be outputting a lower res like 480i, 480p,720p or even 1080i to save on processing power. Thanks for the info that it seems to work well for planets, the moon, etc.
I'm hoping for full 1920x1080P output, transmitted through the HDMI cable directly to the 1080P flat panel tv, in all digital native form, with minimal/no processing or compression.
Edited by macular hole (10/27/09 02:38 AM)
|
macular hole
super member
Reged: 08/12/09
Posts: 143
|
|
Per Canon tech support-
Live view is only 480P. FYI. Not even close to 1080P.
As I suspected.
|
ccs_hello
Post Laureate
Reged: 07/03/04
Posts: 3298
|
|
I assume Canon tech is talking about HDMI output.
Info from the SourceForge EOS Movrec (USB2 interface)
resolution:
450D - 848x560,
40D, 50D - 1024x680,
1000D - 768x512
compression algorithm: M-JPEG encapsulated in AVI
Clear Skies!
ccs_hello
|
macular hole
super member
Reged: 08/12/09
Posts: 143
|
|
Live view HDMI out from just announced $5000 Canon 1d Mark 4
|
Previous
Index
Next
Threaded
Edit
Reply
Quote
[Re: 