| How long will it take until we see the next generation of astro cameras? | |
|---|---|
| < 1 year | |
| > 1 year | |
| Login to vote and view results. |
3.81
#...
·
 |
|---|
|
Title says it all. obviously nobody can know that (except maybe the guys who develop the cameras). The reason i ask is boredom and the feeling that the imx571 is already out for quite some time. Sony released the 571 in 11/2018. |
1.20
#...
·
·
3
likes
|
|---|
| they already are but at a price point that no one can really afford - the SBIG and FLI models based on the GSense 4040 sensors are really a world apart...... |
7.75
#...
·
·
1
like
|
|---|
|
One of the questions is how big is the pressure on the manufacturers to release new cameras with new sensors in a niche market. But innovation will move on. What you consider "next generation" is of course debatable. Clear skies Wolfgang |
7.68
#...
·
|
|---|
| The other problem is that it is not like there are any telescopes that can actually support a sensor any larger or with any smaller pixels than the IMX455. |
|
7.75
#...
·
|
|---|
Bray Falls: Just curious why you think so. There are sensors like the 183 with 2.4 micrometer pixels and ZWO is about to release the ASI 678mc with 2 micrometer pixels. Clear skies Wolfgang |
3.27
#...
·
·
4
likes
|
|---|
Bray Falls: Most (but not all) existing telescopes struggle to get good optical performance across a 42mm field with 3.8 micron pixels. Diffraction limited for 2 microns? That's a tall order. And even if the optics are up to it, most scopes and mounts would struggle to achieve better resolution even in the center of the frame just because of seeing and guiding. My moderate focal length scope (1,124mm) with 3.8 micron pixels is already sampling at 0.69"/pixel. I'm not sure I'd actually see any additional details at 0.44"/pixel. Maybe on a really good night? Probably not. The trend is still towards ever smaller pixels. That's fine. It really doesn't hurt anything other than storage needs and computing power for processing. But I think we have already reached the point where smaller no longer results in a better image for the vast majority of imaging rigs out there. |
6.77
#...
·
|
|---|
|
Well.. depends on what you consider to be "next generation".. what does it need to do? what do you want it to do in order to be considered "next gen"? One realistic question is probably more "When will new/better camera's become affordable enough to be available for the public" Or we'd all be parking a James Webb telescope in our backyard right now. - the "next generation" is probably out there already, it is just not ready for the consumer market due to cost. The other: why should a manufacturer rush development on new stuff when the old stuff is still selling. Especially in a bit of a niche market, especially when the price points are already quite a bit on the high side. And like Bray Falls mentions, it's a hobby where certain things matter. I'm sure they can throw in a sensor with the tiniest and tiniest pixels possible (I think Samsung made a 200MP sensor with 0.5 micrometers... But that won't necessarily work with astrophotography. So I don't see that trickling down towards astrophotography. Sony did release a 2 layer stacked CMOS technology recently, with 2 layer transistor pixels.. which massively increases sensitivity to light and increased dynamic range. Doubling the light gathering power of a pixel. - So perhaps that will make its way towards astrophotography once the cost comes down As for other things, I only really expect to see elements of the higher end camera's trailing down into the affordable packages. Sensors of the latest affordable camera's tend to be 1" already. And already doing 14 bit and pretty good full-well capacity. - The mid range is an APS-C, 16bit sensor.. - top range full frame sensors (at high cost) I expect that to trickle down into the lower range cameras. Where, at some point, you can get an APS-C 16 bit camera for the $1300-1500 range.. I doubt we'll go bigger than full frame, like medium format, at the moment. Besides the cost. There also isn't a market yet for that filter wise.. because even a 2" can't do that.. I only expect small improvements in sensitivity, cooling, noise, etc. And a very long time before any big improvements become affordable. |
|
3.27
#...
·
·
2
likes
|
|---|
Title says it all. I'm not certain what I would even want in a "next generation astro camera." I suppose I always want lower and lower read noise, but aside from that? Quantum efficiency is already very good indeed, within striking range of the theoretical maximum for a passive sensor. Thermal noise doesn't require particularly deep cooling any longer. For most of us it is already a trivial problem to take sub exposures that are shot noise limited for broadband. Pixel density already exceeds what most of us can benefit from given typical seeing and guiding results. The physical dimensions of the sensor in (relatively) affordable astronomy cameras already exceed what most telescopes can practically support. Really, the only area where we still have much wiggle room is read noise. Get that low enough, and lucky imaging for improved resolution starts to become practical for more than just bright objects. That would be really nice, though I am dreading the storage and computer requirements for managing 30 hours of data using 1s subs... |
|
1.20
#...
·
·
2
likes
|
|---|
|
I think we'll see APS-C and FF cameras with Sony's latest pixel architectures soon. I don't know whether that will mean 2-layer photodiodes or the deep bucket type photodiodes they're using in the Starvis2 sensors: https://www.sony-semicon.co.jp/e/feature/2022012801.html The 2-layer FF sensor seems to already be out in the Alpha 1, and the deep bucket sensors are out as smaller sensors now, so APS-C sensors with the new pixels in cooled astro cameras seem likely to come in the next year. |
22.86
#...
·
|
|---|
Most (but not all) existing telescopes struggle to get good optical performance across a 42mm field with 3.8 micron pixels. Diffraction limited for 2 microns? That's a tall order. And even if the optics are up to it, most scopes and mounts would struggle to achieve better resolution even in the center of the frame just because of seeing and guiding. My moderate focal length scope (1,124mm) with 3.8 micron pixels is already sampling at 0.69"/pixel. I'm not sure I'd actually see any additional details at 0.44"/pixel. Maybe on a really good night? Probably not. The trend toward smaller pixels is largely to support the video market using very fast optics (~F/1.2) There are a fair number of high-end scopes that have a flat, well corrected field out to a 50 mm image circle and beyond. The IMX455 is an excellent sensor but I personally miss the field of the KAF16803 sensor with it's 36 mm x 36 mm dimensions. I'd be first in line for an IMX455 type system with the same dimensions. I just wish that it had 4.5 micron pixels (to limit the file size). John |
|
22.86
#...
·
·
4
likes
|
|---|
Paul Ecclestone-Brown: Yeah, they are a world apart. They have horrible dark noise and they are very difficult to calibrate. It can be done and a few folks have been successful with them but I wouldn't touch one of those sensors. Not only are they expensive but they are a real pain to use. As far as I'm concerned the 4040 sensors were a filler between the early, small CMOS sensors and the larger, monolithic CMOS sensors like the IMX455 that are available today. John |
0.90
#...
·
|
|---|
|
Just go look at the specs on the new zwo planetary cameras released a few weeks ago. They almost seem to defy physics. High qe, low read, fwc and dr almost double the predecessor sensor on similar or smaller pixels. I assume the new sensor layering tech allows you to have a larger fwc, but dont quote me on that, Im just staring at a list of stats on the zwo page. https://astronomy-imaging-camera.com/product-category/planetary-cameras So if these are the cheap low end sensors, you can imagine the ones they pick for fullframe and aps-c are going to trigger some severe cases of gear acquisition syndrome. |
|
7.91
#...
·
|
|---|
Wes Schwarz: 10% more QE (at the very best, I think they are overly optimitistic about it) and higher read noise (in some instances) but deeper well depth isn't going to break any sweat I'm afraid as far as planetary imaging is concerned. |
|
0.90
#...
·
·
1
like
|
|---|
andrea tasselli:Wes Schwarz: You think they are overly optimistic? In this field the specs are all data driven, thinking is not going to cut it. You are right though, some of these new cameras have high read noise, I do wonder why that is. Deeper well depth on a planet cam means you can turn up the gain even higher and achieve higher frame rates without clipping data. Compare the Gain vs DR relationship of the 662 to the 462. But my point was that these sensors are an indicator of things to come. Who knows what we are going to see? The specs on the low end sensor indicate that the new large sensors are going to be capable of great things. |
#...
·
·
2
likes
|
|---|
|
It is indeed interesting to speculate. I for one could never have imagined the enormous advances that we have seen in amateur astronomy over the past 40 years. But for me the question around camera tech really raises another question. Why assume that the camera is the limiting technology in imaging? It depends what one wants to achieve. I suppose that chips could be made ever bigger to achieve wider and wider fields - but then you can get there now anyway (albeit more slowly) by stitching montages together and without the optical problems raised by trying to achieve very wide flat fields, Then as far as SNR goes the limitation is normally sky glow -- so the best answer there perhaps lies in portability and improved mobility to darker skies - and/or the use of narrowband filters? Then with respect to improving resolution cameras already exist with not only > 80% quantum efficiencies but also in a range of pixel sizes that can be matched to a variety of optical trains to sample at rates where the seeing rather than anything else is limiting. For me - if I am looking for advances - then it would be great to see proper adaptive optics start to become feasible in the amateur sphere (-i.e beyond the SBIG tip and tilt system). Tim |
|
7.91
#...
·
|
|---|
Wes Schwarz: Because they very much sound like marketing chat. Sometimes they say "we think" or "we measured" without further details. And look at the curves. They are far too smooth for be real test data. Point in question is they declare a very high QE (say 91% for the 2600MM) when actual tests show something different, e.g.: http://www.astrosurf.com/buil/asi6200mm/ They did the same for the their camera equipped with IMX571. Overly optmistic whilst they should have been cautiously pessimistic Wes Schwarz: Or reduce the exposure. And increase the frame rate. It always been so for the last 25 years and I don't see it changing now. Besides, why should I increase the gain beyond a certain measure at any given frame rate. As what matters is the SNR anyway? |
|
0.90
#...
·
|
|---|
andrea tasselli:Wes Schwarz: i can't read that article. How many 2600 sensors did he test, how did he calibrate his equipment, what did the confidence interval indicate in his regression analysis involving his samples against the entire population on manufactured monochrome 2600 sensors? What did the manufacturer analyze sensors with, how do they calibrate their equipment, how do the two calibrations compare? It seems like a fun experiment but we don't have enough information Available to make a proper assessment. Im not sure there is a point discussing the planet imaging any further, The point of my example was not to teach the theories of lucky imaging, that is better suited in another thread |
|
3.81
#...
·
|
|---|
Tim Hawkes: Lot of great answers in here. Adaptive optics would be nice but i think there is still a lot if room for cameras to evolve. QE is, if i am not completely wrong, only measured for specific wavelengths. So while camera x could have a good qe for a given wavelength, the qe could be completely different for other wavelengths. |
|
0.00
#...
·
|
|---|
Paul Ecclestone-Brown: FLI and SBIG are working on new IMX455 based cameras. Hope price target will be closer from what my pocket $$ can affords. You are right...GSense sensors may be top industrial quality but it cost is far from what many can afford. Dont expect too much from a new sensor launch.....cameras with internal stacking capabilities or add-on spec using the current IMX sensors would still lead the market for long. We are a small market for the sensor manufacturers against the cell phone cameras market. Shortage and covid did not help to invest money on new sensor developments from recent years. |
