The Future of Imaging - Is everything heading OSC

[Northernlight]

Hi All,

My beloved QSI 683 recently died and is being sent away to see if it can be repaired, but expecting the worst i started researching new camera to replace the QSI.  Looking at the camera market with a modest budget of around £2000 - i noted that the choices for mono camera were a bit limited.  The most obvious choice for a chip with near identical size to the QSI 683 was the ASI1600MM Pro - which i know is a very popular choice, but i'm weary of buying this camera as i know it's end of life.

I spoke to Astro Reseller Ianking about the choices, and he recommended the SX Trius 694 which looks like a great mono chip with very low noise - but it's a much smaller chip than the QSI. With no direct replacement in sight for the ASI 1600 - we talked about the future of camera and Ian said the in the future he expected to see less & less mono camera and that most new camera would most probably be OSC.

So this got me wondering about the future of Narowband imaging - Where does this leave us if most new sensors will be OSC ?

[JamesF]

4 minutes ago, Northernlight said:

So this got me wondering about the future of Narowband imaging - Where does this leave us if most new sensors will be OSC ?

It's a very good question and one I've been pondering on myself for quite some time without an answer.  It feels as though there's very little interest in making mono sensors any more.

James

[Alien 13]

I expect some mono cameras are just a by-product of general compact/DSLR camera manufacture, a way of selling off the over manufacture/scrap of unfinished chips.

Alan

[Northernlight]

So where does this leave us with Narrowband Imaging ? - If mainstream camera manufacturers are mainly focusing on OSC - then are we likely to only see mono sensors in very high end specialised cameras.  I hope this doesn't happen as i think it would be a big step backwards for Astro Imaging. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

[knobby]

Interesting debate ! As someone who has joined the mono party very late I can already see the benefits in astro imaging so it would be a shame if mono becomes elitist. Maybe the future is the new Tri band filters ? Still loses out on using the whole chip but a compromise.

[Alien 13]

It might seem daft but if a large manufacturing plant is producing normal camera chips at a few pence/pounds each the extra cost of having a bloke removing a batch before the process is finished so you could have mono chips and the extra costs involved with handling this "batch" could increase the manufacturing costs per item by 1000% or more.

Alan

[MarkAR]

Well if mono CCDs are going to die out, then better get them whilst you can.

I have the Trius 694 and although I haven't used any other and can't compare, I can say that it's been good to me. Cools down fast and made in the UK. I also hear SXCCD customer service is very good.

I would definitely get another.

[CCD-Freak]

Therein lies the problem......Astronomy use of mono sensors is only a very  very small percentage compared to all the OSC sensors made for phones, cameras and such.  Things like machine vision for manufacturing are going the OSC route due to low cost.   The sensor manufacturers are developing products for OSC market were there is a large demand and fewer mono sensor are coming to the market.  Less volume = higher prices to produce.  This is not a pretty picture for astronomy and low cost mono cameras.  😔  

My $.02

 

John
CCD-Freak
WD5IK

[Northernlight]

And this is my concern - where mono is easily accessible to everyone at the moment at a very reasonable price (e.g. ASI1600MM-Pro)  the future of Mono looks very bleak  and i think this transition may already be starting to happen , and if we stick with ZWO as an example and look and their new camera line up. We have the 6200, 2600 & 533 cameras - with all of them being colour, with the exception of the expensive 6200 which is available as a mono full frame option.

If this is a sign of future trends then Mono cameras could become out of reach for a lot of Astro consumers as Full frame cameras are expensive, then factor in  a 2" filter wheel and the cost of the associated 2" filters then you are entering into the realms of big bucks with Mono setups potentially costing upwards of £6000 - £7000. Also you may need to factor in the cost of a new PC as trying to process the files from full-frame camera that are over 60 mega-pixels  will take a lot of storage and a fairly powerfull pc with a decent amount of ram.

If the industry does in fact go this way it will be a sad future.

Edited May 30, 2020 by Northernlight
update

[Adam J]

With all respect to Ian King he is wrong. There are still many application for which OSC makes no sence for comercial use cases. What about near IR imaging, what about UV imaging? What about extream low light levels?

Sorry but mono is not going any place yet.

If you actually look at the number of mono sensors that have come onto the market recently its actually higher than in the previous 10 years.

Adam

Edited May 30, 2020 by Adam J

[discardedastro]

I'd tend to agree - machine vision is a huge market and still quite heavily monochrome for good reasons. Full frame is a bit of a weirder niche and that may indeed become harder to get sensors for, but there's plenty of large format mono users in industry. I suspect the number of sensors and variety may come down but mono sensors geared up for HDR/low-light aren't going to go away. CCTV is another industry where mono reigns supreme at the high end (and that's a lot of market).

[Grant]

56 minutes ago, Adam J said:

With all respect to Ian King he is wrong. There are still many application for which OSC makes no sence for comercial use cases. What about near IR imaging, what about UV imaging? What about extream low light levels?

Sorry but mono is not going any place yet.

If you actually look at the number of mono sensors that have come onto the market recently its actually higher than in the previous 10 years.

Adam

With equal respect I’m pretty certain Ian knows more about the Astro imaging market which is what is being discussed than you Adam - he wasn’t referring to machine vision or anything else but the amateur astronomy market and he is 100% correct - most new sensors for our market have been colour in recent years and most cameras we are aware of in the pipeline are colour (not all, most).

That doesn’t mean we think that’s how it should be, just the reality of how things are heading.

[Northernlight]

I agree with Grant -  Nobody is saying that there wont be any new Mono camera's in the future - just less of them in favour of OSC cameras. and i think there is evidence of that starting to happen already.

But getting back to the topic at hand - "If" this does turn out to be the future trend - Then where does this leave us in regards to Narrowband Imaging  ? So far from what i've seen from OSC  cameras using these quad band filters  - it aint a pretty site. - who knows maybe i haven't seen any good examples yet - and another thing £1200 for the Triad Ultra Quad band filter- are they having a laugh - that's the price of the ASI1600MM-Pro

If there was a way to get half decent Narrowband from a OSC then i would buy the ASi2600MC-Pro tomorrow to replace my beloved QSI 683, but i just dont think it's a reality.

Rich.

Edited May 30, 2020 by Northernlight

[Adam J]

I believed the question was in the underlying context of avaliability of mono sensors to replace the "end of life" ASI1600mm pro. Also the title is "is everything heading OSC"

I am simply saying that there is not going to be an lack of mono sensor avaliability because there are applications (many many applications) that simply cant be performed at all with OSC sensors and many of those are high volume applications leading to cheap sensors.

So mono sensors will always be avaliable to put in astronomy cameras at not too overly inflated rates.  This disparity is inevitable for a couple of reasons and more so with the move to CMOS (ill cover that later).

Its inevitable because almost all mono sensors have an OSC version, but not all OSC sensors have a mono counterpart. A big factor being that if you want a APS-C then it will be OSC becuase very few want a mono DSLR but crutially its not because no one wants a mono APS-c sensor in the astronomy comunity. This fact is also advantagious as it means that the camera manufacturers are able to easily make mono cameras with almost zero effort as if they have already made a camera with the OSC version of the same chip, its a firmware change to make a mono copy. Looking at that the other way around then its almost zero effort to make a OSC version of a camera if you made a mono, probably why some questionable OSC sensor choices have appeared in the context of there already being higher performing sensors in other cameras by the same maker.

51 minutes ago, Northernlight said:

If there was a way to get half decent Narrowband from a OSC then i would buy the ASi2600MC-Pro tomorrow.

Rich.

The performance gap between mono and OSC will not be closing any time soon, more so the performance head room avaliable to OSC is not sufficient to enable it to be closed, we are already at read noise figures aproaching 1e- and QE values of over 80%, almost photon counters, yet as you say if you took a old KAF8300 it will still run rings around the latest ASI2600mc pro from what I have seen so far perticually in heavy LP and its what 12 years old?

If they stopped making mono chips today it will be around a decade before OSC is able to compete in anything aproaching real terms with the current crop of  new CMOS mono sensors.

So thats avaliability of sensors covered.

Now in terms of the market I would easily believe it is i was told there are now more OSC sales, its cheaper for beginners and many people find mono imaging too much of a pain and want to stick with OSC and I understand why sometimes. 

However, as I say there will not be a lack of new mono sensors and people who want performance over ease of use will still want a mono camera. I recently commented that I am shocked that people are willing to pay £1700 for a set of Astrodon 3nm filters and butsome astro imagers are willing to pay those sorts of figures for high performance equiptment and while that is the case someone is going to make a mono camera for them.

 

On 29/05/2020 at 20:28, Northernlight said:

 Ian said the in the future he expected to see less & less mono camera and that most new camera would most probably be OSC.

So this got me wondering about the future of Narowband imaging - Where does this leave us if most new sensors will be OSC ?

So I think the first part quoated above to be correct, but I would atribute that to a larger number of sensors being viable with moving to consumer level CMOS sensors and improvements in that technology than had been based on CCD technology in the past. But I dont think that means less mono will be avaliable, just more OSC by comparison. 

In the past: Only CCD sensors are viable for high end astro imaging, CMOS are too high noise and too low QE to compete in OSC or Mono. But due to cost and lack of use in consumer terrestrial cameras very few CCD sensors are avaliable. So they are made for scientific reasons and lack veriety. In addition the tech refresh cycle is slower with bigger perfomance jumps between generations.

Now: High investment in CMOS technology has resulted in it overtaking CCD even in consumer products and CCD production is to be halted. But as all sensors will be CMOS you suddenly have a huge choice in OSC sensors and so more OSC cameras are made but this is the bit where i disagree with Ian and ill word it with care. I think Ian is a great bloke, I just recently made a purchase from him and have received excerlent customer service from him in the past, he certainly knows his stuff, but I reserve the right to politely disagree with anyone and will make a reasoned argument as to why I think differently. 

So here is the bottom line, I simply dont think that there will be less and less Mono cameras avaliable I just think that there will be more OSC avaliable by comparision which is not the same thing. I strongly believe that the turnover rate of Mono sensors will remain at historical levels, hence lower than that of OSC as dictated by the relative demand for them and that Astronomy cameras will be avaliable with mono sensors at historical levels and variety for the forseable future irrespective of the increased veriety of OSC cameras. But even then if you think about it there have been more astronomy viable mono sensors introduced in the last 4 years then in the 4 years prior, so its all a matter of perspective. After all the KAF8600 came out in about 2008 and it was not till the ASI1600mm pro was released in 2016 that it had a competitor in its size class, thats 8 years and so its hardly the case that we had huge turn over of mono sensors in the past either thinking about it.

Adam

Edited May 30, 2020 by Adam J

[alexbb]

There are a few sensors on the market, but not yet used in astro cameras. https://canon-cmos-sensors.com/canon-120mxs-cmos-sensor/

[Northernlight]

12 minutes ago, Adam J said:

I am simply saying that there is not going to be an lack of mono sensor avaliability because there are applications (many many applications) that simply cant be performed at all with OSC sensors and many of those are high volume applications leading to cheap sensors.

Hi Adam,

I understand where you are coming with a lot of your rationale - But having worked in high speed manufacturing on the It & Integration side of things for the last 23 years covering many different industries - i can tell you the demand from these sectors are not high at all - with most manufacturing or medical facilities installing either high speed or mono camera solutions into production /logistics or medical equipment which will most likely remain unchanged for at least 10-15 years - with these types of camera being produced in very small quantities at very very high prices any usually with long order times - So i must politely disagree with your comment about high Volume - as i have 23 years of 1st hand experience that says otherwise.

 

[Adam J]

47 minutes ago, alexbb said:

There are a few sensors on the market, but not yet used in astro cameras. https://canon-cmos-sensors.com/canon-120mxs-cmos-sensor/

Probably very very expensive and the pixels are very small.

42 minutes ago, Northernlight said:

Hi Adam,

I understand where you are coming with a lot of your rationale - But having worked in high speed manufacturing on the It & Integration side of things for the last 23 years covering many different industries - i can tell you the demand from these sectors are not high at all - with most manufacturing or medical facilities installing either high speed or mono camera solutions into production /logistics or medical equipment which will most likely remain unchanged for at least 10-15 years - with these types of camera being produced in very small quantities at very very high prices any usually with long order times - So i must politely disagree with your comment about high Volume - as i have 23 years of 1st hand experience that says otherwise.

 

Not sure about those areas, I am thinking about things like self driving cars and security cameras, drone sense and avoid. etc. There are a number of emerging technologies that will require and already require mono sensors.

I also work in a very relevant area involving electro optics but am not going to post that kind of personal detail in this forum. 

Adam

Edited May 30, 2020 by Adam J

[Northernlight]

8 minutes ago, alexbb said:

There are a few sensors on the market, but not yet used in astro cameras. https://canon-cmos-sensors.com/canon-120mxs-cmos-sensor/

Hi Alexbb,  yes you will see new sensors - but like the cannon one you shared a link for - they are very very specific Niche chips for very specific industries and not intended for mainstream consumer market and will be very very expensive.

[Adam J]

7 minutes ago, Northernlight said:

 So i must politely disagree with your comment about high Volume:

 

https://www.sony-semicon.co.jp/e/products/IS/industry/product.html

Thats a list of sony products for industrial use. Almost all those sensors are avaliable in mono and RGB.

Most of them are very very new too.

Adam

 

[Adam J]

Of that list the following are already currently avaliable for purchase in cooled astronomy cameras:

IMX 455 - 43.3mm - (likely to be the best anature astronomy sensor ever released)

IMX183 -  15.9mm

IMX 304 - 17.5mm

IMX 253 - 17.6mm

IMX 178

IMX 174

IMX 290

IMX 273

IMX 252 - 8.9mm

IMX 250 - 11.1mm

IMX 249

IMX 428 - 17.6mm

You will likely see an IMX 542 based mono and OSC astronomy camera in the near future and IMX540 too.

8 of the sensors on Sonys list where added in the last 12 months so its likely that they are for pre order only. So I would say that evidence is that Sony at least are still investing in Mono sensors.

The IMX183 mono was listed as avaliable for 12 months before we saw a Astronomy camera with it in.

More new mono cameras will come within the next 12-24 months, I would bet on it.

The thing is thats just sony! If you start thinking about other makes there are more.

If you look at the 1 inch + sensors on that list then we have in effect had on average more than 1 new mono sensor type of useful size for DSO imaging per year for the last 4 years. I just dont follow how this can be considered a reduction when the last generation of CCD camera sensors where with us for between 8-12 years. Its only this year that ATIK have anounced that the 383L+ is going to be discontinued ad the sony CCD chips are still being sold with no end currently anounced. CMOS change over has if anything at all increased the rate of introduction of new Mono astronomy cameras significantly.

Adam

 

Edited June 1, 2020 by Adam J
added 5 x more sensors in current mono camera to the list

[Northernlight]

Well i certainly hope you are right, but i think the future of Astro camera will be driven by consumer grade cmos digital cameras and not buy industrial applications .

Looking at IMX 540/542 - i dont see theses being useful to be used in Astro imaging as they 2.7 micron pixels and a well depth  of 10,000e so they will saturate very quickly.

 

[MarkAR]

A CMOS sensor is basically a mono sensor with an RGGB bayer matrix added on top, why not just leave off the matrix and lets have cheaper mono cameras.

[Adam J]

6 minutes ago, MarkAR said:

A CMOS sensor is basically a mono sensor with an RGGB bayer matrix added on top, why not just leave off the matrix and lets have cheaper mono cameras.

They already do , but only bother when they think they can sell sufficient numbers to make a batch as they cant change the whole production line over for only 1000 chips. 

As a interesting fact there was not IMX183 mono until ASI / QHY persuaded sony to do a run.

Adam

[Northernlight]

I have wondered this myself - but usually in the manufacturing industry - productions lines are setup to run in a very specific way and making changes or deviating from the process can very time consuming & costly - i.e. you need maintain 2 separate production processes as well as 2 different testing & QC processes and changing between them drastically reduces your operating effectiveness (OEE) which gets very expensive really fast.   

So to my mind - this would be the main reason

Edited May 30, 2020 by Northernlight
update

[Adam J]

4 hours ago, Northernlight said:

Well i certainly hope you are right, but i think the future of Astro camera will be driven by consumer grade cmos digital cameras and not buy industrial applications .

Looking at IMX 540/542 - i dont see theses being useful to be used in Astro imaging as they 2.7 micron pixels and a well depth  of 10,000e so they will saturate very quickly.

 

I think you can HW bin them

Also by the time you use the ASI1600mm pro at unity gain its only got 4012 bits of depth anyway and I find that very acceptable.

Low depth is a side effect of the global shutter. But the Moravian C2-12000a uses the IMX253 that has only 9000e with bigger pixels (worse in terms of time to saturation) and it produces some cracking images from what I have seen so far. Its very new, but looks like the 540/542 are its replacement lol. In effect it will probably have more dynamic range than the IMX183 in use.

An important measure is well depth / pixel area. In those terms its not doing bad at all.

Adam

 

Edited May 31, 2020 by Adam J