LRGB filters

[Spongey]

Hi all,

I am more seriously looking at filters now that the QHY268M / ASI 2600MM have been officially announced and are imminent. I have narrowband decided (Antlia 3nm PRO), but am still debating LRGB. I will be going with 36mm unmounted format. 

My main considerations are as follows: 

• Imaging scope - Esprit 100, but I will occasionally use the camera with some canon lenses too;
  • These lenses will be the Samyang 135mm (already owned), and potentially a Sigma Art 40mm should I delve into super widefield imaging at some point.
• Imaging location - My backyard is bortle 4/5, with a mixture of Hg/Na and LED based light pollution (mainly in the west). I will also travel (once we are allowed to) to dark sites a few times a year for holidays;
• Budget - I am looking around the £300 price point (i.e. anything except Astrodon and Chroma);
• Parfocality - This is between the filters themselves and my narrowband filters; and
• Backfocus - My narrowband filters will be 2mm thick, so ideally the LRGB will be the same to avoid any spacing issues. 

There are obviously a lot of options to choose from, such as Antlia, Baader, Astronomik, ZWO, Optolong to name a few.

The current toss-up is between Antlia and Baader, but I am happy to be convinced otherwise. These filters are very similar in spec (2mm thickness, price), but their passbands are fairly different. I have copied each of the filter response curves below for convenience. 

My main concern with choosing the Antlia filters is the gaps in the spectrum around 400-420nm and 575-625nm. These are reasoned to combat blue bloat and light pollution respectively, but I can't help but feel that this a significant portion of the spectrum is lost, even if no major emission lines are in these bands. Is this likely to significantly affect colour balance / reproduction? The Lum filter looks good here.

The Baader filters are established, have a small(er) gap for Hg/Na, but have a much earlier cut-up in the blue filter at around 380nm, which is also not included in the Lum filter. I doubt this would make a significant difference, but I am mildly concerned about blue bloat in my optics (mainly the canon lenses). The Lum filter also looks to have a slightly poorer response in the blue channel overall, and a significant peak around the 360nm point.

Can anyone offer any advice / musings on any of the above? I realise I am likely to be happy with any of the listed offerings but lockdown is getting to me and this is how I am keeping my brain active! 

Baader: 

Antlia:

Edited January 5, 2021 by Spongey

[Martin Meredith]

Hi

I have the Baader set and am very happy with them, although I have not compared them to anything else, and my only use is in an EEVA setting rather than AP. One comment: I don't know if the Antlia set has a C filter, but I find the C filter very useful in the Baader set. The L filter cuts off quite a bit of energy in the IR range , and this can contribute useful photons to galactic imaging. Of course, how much extra energy is available to be picked depends on your sensor characteristics in this region too.

cheers

Martin

[Spongey]

Hi Martin, 

Thanks for your response!

I will be using a 7 position filter wheel so no room for a clear filter I'm afraid. Furthermore, with my refractive optics, IR light won't be focussed to the same point as the visible spectrum so I'm not sure how useful a clear filter would be (probably less of an issue for EEVA).

Cheers

[Martin Meredith]

Ah, ok. I'm using a Newt so am spared that issue. I used to use an achromat and the star bloat (particularly at the UV end) was terrible even in an EEVA context. My 7 position EFW currently has RGBC, H-alpha, dark and a spectral grating. I've kicked out the OIII and SII for now 😉.

Martin

[vlaiv]

If you are really worried about color reproduction - neither of the two.

You want something like this:

Although people praise filters with clear cut off points - such filters are not good for good color reproduction (true color).

Take for example OIII nebula - and observe above Baader filters. They overlap in 500nm part - just barely.

Imagine for a moment that they don't overlap but are perfectly continuous. OIII light would fall either in blue filter or green filter - and you would never be able to get teal color of OIII nebula - it would always render as blue or green. How much of all possible colors can system distinguish is called gamut. Filters that don't overlap have rather small gamut.

They are good for astronomical work in sense that you can easily split wavelengths if you have sharp cut offs and want to know how much photons fall in 500-600nm range for example - but are not good for color reproduction. Even with color balance.

For that same reason, OSC cameras have overlapping filter responses:

That helps to better reproduce actual color.

But there are other things to consider with filters as well - like reflections, being parfocal and similar. Many people use baader filters and produce nice images with nice looking colors regardless of the fact that they might be a bit off in comparison to the real thing.

[Spongey]

Hi Vlaiv,

Thanks for your response.

This is of interest to me as I've recently been reading up on colour and its application in astrophotography. 

The cones in the human eye have spectral sensitivities that overlap (and even have negative responses), somewhat similarly to those in an OSC camera. Therefore, the only way to capture 'true' colour from deep sky objects would be to take images through spectral filters all across the visible spectrum, and subsequently apply colour matching functions to them to approximate the relative response of the human eye.

If my understanding is correct, this is similar to how a chromaticity diagram is computed for all points not on the spectral locus. 

In practical means this is infeasible, and so how to best produce colour with three filters comes into question. I therefore logically come to the same conclusion as you; that filters such as the Astronomik LRGB type 2-c, or those used in OSC cameras are superior for creating true-colour images of deep sky objects. For true colour (or as close as practically possible) using these filters with a mono camera, colour calibration would need to be performed based on the filter response curves, sensor QE etc. As far as I'm aware, this isn't commonly done.

My question therefore becomes why are these the only filters in a sea of LRGB filters to be designed in this way? Why is no one seemingly concerned with this issue? I don't doubt that good colour can be produced from filters with steep cutoffs such as Baader, Chroma, Astrodon etc., but if we astrophotographers were truly worried about producing true colour images then shouldn't we all be using ones similar to the Astronomiks?

Cheers

Edited January 5, 2021 by Spongey

[vlaiv]

Main issue as far as I see it is that people don't really understand all that well color correction and tools available don't really go that deep to be useful in astrophotography.

I'll probably do example at some point, as I have Baader filters and ASI1600 - which I bought before I bothered to learn all this color related stuff. I'll take DSLR type image of color checker chart and one with ASI1600 and I'll try to calculate color correction matrix. That way it will be easy to see how accurate are colors produced with such RGB filters.

I already did this for my OSC camera - ASI178mc, so process is not that difficult. There is also other way to do it - if there was QE graph of ASI1600 (or any other sensor) available that was reliable. I have managed to find two graphs of QE of ASI1600, one published by ZWO and another measured by Christian Buil (very experienced amateur spectroscopist). They are different and not by small margin.

If I had such accurate data - then it would be easy to write software to calculate both conversion matrix and resulting gamut.

In any case - there is quite a bit of debate going on - on topic of actual color of astronomical objects, and if you are interested, please join in:

 

[DaveS]

This IDAS type 4 filters look interesting

This page gives the background.

[Martin Meredith]

Remember that we are not generating an entire spectral profile for each pixel but instead creating just a single intensity value per filter that is a multiplicative function of the stellar energy spectrum, the filter spectral transfer function and the sensor's spectral sensitivity. A single value. So, for instance, differences in the peak region of the passband (which are very clear in the OSC figure above and the IDAS figure) will also have an impact -- perhaps even a larger impact -- than differences in the amount of overlap at the edges. Baader make it pretty clear why they leave a gap in the G/R region. Other manufacturers will make other design choices.

Ultimately, whether any of this makes much of a difference (after correct calibration) remains to be demonstrated by a proper visual psychophysics experiment -- don't hold your breath. Personally, I very much doubt it does make a noticeable difference.

Martin 

[Spongey]

9 minutes ago, Martin Meredith said:

Ultimately, whether any of this makes much of a difference (after correct calibration) remains to be demonstrated by a proper visual psychophysics experiment -- don't hold your breath. Personally, I very much doubt it does make a noticeable difference.

This is a very good point and I would not be surprised if the outcome of such an experiment is that the difference is small / inperceptable.

Back to the topic at hand, another consideration for filter selection is filter bandpass and subsequent SNR.

The SNR (assuming equal exposure lengths and everything else) of a sub in the blue channel for Baader and Antlia filters as shown above will be different, with the Baader filters having a higher value due to letting more light through. The same will be true for each of the colour channels actually, the Antlia filters have thinner bandpasses across the board.

While I appreciate that these are to account for blue bloating and light pollution respectively, I think I would prefer the higher SNR provided by the Baader filters, as these still have a notch for spectral light pollution, and I would like to think that my optics are well corrected down to 400nm (I have no issues with blue bloat when using my Canon 6D with any of the above optics). With the way that light pollution is going, I doubt a wider gap here will retain its value.

Cheers 

[vlaiv]

22 minutes ago, Martin Meredith said:

Remember that we are not generating an entire spectral profile for each pixel but instead creating just a single intensity value per filter that is a multiplicative function of the stellar energy spectrum, the filter spectral transfer function and the sensor's spectral sensitivity. A single value. So, for instance, differences in the peak region of the passband (which are very clear in the OSC figure above and the IDAS figure) will also have an impact -- perhaps even a larger impact -- than differences in the amount of overlap at the edges. Baader make it pretty clear why they leave a gap in the G/R region. Other manufacturers will make other design choices.

Ultimately, whether any of this makes much of a difference (after correct calibration) remains to be demonstrated by a proper visual psychophysics experiment -- don't hold your breath. Personally, I very much doubt it does make a noticeable difference.

Martin 

You don't really have to do any special experiment, you can just wonder what will rainbow look like when recorded in Baader RGB model.

Here - look at this image:

I have marked 4 different emission lines in B part of the spectrum. Let's suppose that they are of equal intensity. What will RGB filters record for each one of them:

G and R will simply be 0 as each of those lie outside of G and R parts of spectrum, right? Only B will have some value. And B values will be fairly close in intensity - they will be QE of sensor multiplied with transmission of Baader B filter - which is mostly 100% or small variations.

So first line will be 0, 0, 0.55 second will be 0, 0, 0.58 third will be 0, 0, 0.62 and fourth will be 0, 0, 0.62 (for example). These are not rainbow colors - these are all blues with different intensity. Instead of recording this:

you'll end up with something like this:

I'd say that's a few missing colors in that image, right?

 

[Martin Meredith]

A few comments

1. We are recording pseudo-continuous spectra in the main

2. The same argument can be made of the Astronomik filters by judicious choice of emission lines, right?

3. This assumes that the filter responses we see are actually accurate or just approximate. I imagine it is quite hard to design a filter with exactly zero transmissivity.

4. The instrument spectral response may well not be uniform.

In the domain of AP and to address whether filter sets make much or any perceivable different, we still need to do a proper experiment. 

[vlaiv]

1 minute ago, Martin Meredith said:

2. The same argument can be made of the Astronomik filters by judicious choice of emission lines, right?

Yes, but there will be areas where such filters can distinguish more than straight edged interference filters.

If you look at this diagram - left pair of wavelengths are indistinguishable. So is far right pair of lines - but lines in the middle - will have different ratio of R and G, and G and B - and we will be able to tell them as separate colors.

In fact - in regions where filters overlap - we can distinguish them as separate colors. If you look at CieXYZ that is standard absolute colorimetric space:

"filters" (here matching functions but really the same thing) overlap all over the place and sometimes even all three.

CieXYZ covers whole gamut of human vision and therefore even single wavelengths of light will have different XYZ coordinates and be distinguishable as are to human eye.

There are filters that are better suited to color reproduction than both baader and astronomik or any other interference filters out there.

Look at absorption filters. Problem is that they often have low QE or rather transmission - sometimes even below 50% and for this reason people don't want to use them - but they provide best filters in terms of color accuracy.

You can combine for example above filters to get good mix / coverage. You just need to perform color calibration, but as you see, blue and green are only 67-68% peak transmission.

13 minutes ago, Martin Meredith said:

3. This assumes that the filter responses we see are actually accurate or just approximate. I imagine it is quite hard to design a filter with exactly zero transmissivity.

For all intents and purposes, it does not matter much if it is less than 1-2% - it won't have enough "resolution" to make a difference.

15 minutes ago, Martin Meredith said:

4. The instrument spectral response may well not be uniform.

Indeed, you have to perform color calibration for any combination of camera and filter. However, it is 3x3 matrix as linearity must be preserved (addition of light and variation of intensity by exposure length) so we have 3 component vector to 3 component vector linear transform - that is 3x3 matrix.

Two components of vector being 0 means that all possible values are on a straight line - that will be the same when you project it. You can't get better variance of color in projected space - it will still be some shades of single color as it is in primary space.

 

[Martin Meredith]

There is a huge loss of information when going from any given spectrum to the outputs of a small set of filters. All filter systems are going to be doing a pretty bad job of representing carefully selected edge cases like pure monochromatic sources. Look how 'terrible' the Sloan system is with its near lack of overlap and G-R gap... 😉 

https://astrodon.com/products/astrodon-photometrics-sloan-filters/

Given the infinite number of spectra that can, in theory, produce the same output value in any one of these filters, calibration is really the only practical way to use a photometric system.

I bring up the Sloan system (and I could equally bring up the GAIA or any other 'science' system) for a reason, because it raises the question of what is the end goal of AP for any particular practitioner (as this will affect filter set choice). Is the goal to restrict colour information to the range and sensitivity of human vision, or is it to take advantage of the wider range/sensitivity of astronomical sensors, and then remap this information into the visible colour space? If it is the latter, then issues of correct colour balance and colour 'reproduction' are moot.

Put another way, maybe a quantised rainbow is a fair price to pay for detecting L type stars, to give just one example?  We are talking astrophotography (and not spectroscopy and not terrestrial photography) after all.

Martin

 

[vlaiv]

1 hour ago, Martin Meredith said:

Given the infinite number of spectra that can, in theory, produce the same output value in any one of these filters, calibration is really the only practical way to use a photometric system.

Thing is that human visual system behaves in the same way - any color we see will be generated by infinite number of spectra.

If our eyes can do it, why not camera - we also have three different "filters" - why would it be a problem to match our filters to telescope ones.

1 hour ago, Martin Meredith said:

I bring up the Sloan system (and I could equally bring up the GAIA or any other 'science' system) for a reason, because it raises the question of what is the end goal of AP for any particular practitioner (as this will affect filter set choice). Is the goal to restrict colour information to the range and sensitivity of human vision, or is it to take advantage of the wider range/sensitivity of astronomical sensors, and then remap this information into the visible colour space? If it is the latter, then issues of correct colour balance and colour 'reproduction' are moot.

Put another way, maybe a quantised rainbow is a fair price to pay for detecting L type stars, to give just one example?  We are talking astrophotography (and not spectroscopy and not terrestrial photography) after all.

I agree. For those that enjoy that side of astronomy - more information or specific information is better than producing visually equal color, but to quote myself from the start of this discussion:

20 hours ago, vlaiv said:

If you are really worried about color reproduction - neither of the two.

All of this was with above goal in mind - color reproduction. When one wants to reproduce colors that we would see when looking at particular light. Again, as above - that is not bad thing in itself - some enjoy creating realistic images of object out there and realism comes in many forms - one of which is visual color matching.

[andrew s]

The eye had a response like this.

I am sure no set of filters for AP would be so poor!

Regards Andrew. 

[vlaiv]

1 minute ago, andrew s said:

The eye had a response like this.

I am sure no set of filters for AP would be so poor!

Regards Andrew. 

Why is this:

Poor?

You can use normalized response curves and then scale with coefficient ...

[Spongey]

2 minutes ago, vlaiv said:

You can use normalized response curves and then scale with coefficient ...

While this is true and will be the best way to achieve true colour, I assume that the sharp cut on/offs of typical RGB filters are used to maximise signal, whilst producing colour that is 'good enough'.

[andrew s]

1 minute ago, vlaiv said:

Why is this:

Poor?

You can use normalized response curves and then scale with coefficient ...

But what about the noise in the blue and the minimal discrimination between red and green. Needs neural processing by the brain to make sense of it. 😏

Regards Andrew 

[vlaiv]

10 minutes ago, Spongey said:

While this is true and will be the best way to achieve true colour, I assume that the sharp cut on/offs of typical RGB filters are used to maximise signal, whilst producing colour that is 'good enough'.

Why not extend their coverage and create overlap - like Astronomik ones?

right shape has more "volume" compared to left - which means more photons captured - higher signal, better SNR.

[vlaiv]

11 minutes ago, andrew s said:

But what about the noise in the blue and the minimal discrimination between red and green. Needs neural processing by the brain to make sense of it. 😏

Regards Andrew 

How about this then:

This is well defined color space - in fact used as absolute colorimetric space. Most if not all of our color matching theory is based on that color space. We don't need to invent anything special for that space.

[Spongey]

1 hour ago, vlaiv said:

Why not extend their coverage and create overlap - like Astronomik ones?

right shape has more "volume" compared to left - which means more photons captured - higher signal, better SNR.

I feel like this is a good question to pose to more experienced imagers than I, and perhaps even to filter manufacturers. A quick search doesn't return many results for photos made with the Astronomik type 2c filters. In fact, the only photos I can find are made with the Astronomik Deep-Sky RGB filters which have steep cutoffs.

[vlaiv]

20 minutes ago, Spongey said:

I feel like this is a good question to pose to more experienced imagers than I, and perhaps even to filter manufacturers. A quick search doesn't return many results for photos made with the Astronomik type 2c filters. In fact, the only photos I can find are made with the Astronomik Deep-Sky RGB filters which have steep cutoffs.

That was more rhetorical question to give example that steep filter sides don't necessarily mean better SNR.

For narrowband filters that is certainly the case - narrower the band and steeper the sides - better the filter but that is not true for filters used in broad band. Steep cutoffs have their uses when you want to just capture certain band and want to do that at best SNR possible.

Again - not the case when you want to get best color reproduction.

[andrew s]

4 hours ago, Martin Meredith said:

There is a huge loss of information when going from any given spectrum to the outputs of a small set of filters. All filter systems are going to be doing a pretty bad job of representing carefully selected edge cases like pure monochromatic sources. Look how 'terrible' the Sloan system is with its near lack of overlap and G-R gap... 😉 

https://astrodon.com/products/astrodon-photometrics-sloan-filters/

Given the infinite number of spectra that can, in theory, produce the same output value in any one of these filters, calibration is really the only practical way to use a photometric system.

I bring up the Sloan system (and I could equally bring up the GAIA or any other 'science' system) for a reason, because it raises the question of what is the end goal of AP for any particular practitioner (as this will affect filter set choice). Is the goal to restrict colour information to the range and sensitivity of human vision, or is it to take advantage of the wider range/sensitivity of astronomical sensors, and then remap this information into the visible colour space? If it is the latter, then issues of correct colour balance and colour 'reproduction' are moot.

Put another way, maybe a quantised rainbow is a fair price to pay for detecting L type stars, to give just one example?  We are talking astrophotography (and not spectroscopy and not terrestrial photography) after all.

Martin

 

The SDSS filters did not have a gap in the in originals used in the survey. If you look at the M Fukgita et. al paper referenced by Astrodon there slight overlaps between adjacent bands. The paper is worth a scan for a feel of the effort needed to put measurements on a standard absolute scale. Of course the SDSS filters are designed for photometery not RGB imaging. 

Regards Andrew

Edited January 6, 2021 by andrew s

[Spongey]

Considering that my main goal with imaging, ultimately, is to make pretty pictures, I have decided to go with the Baaders.

Their reputation within the community, together with the SNR gain caused by the extended gaps present in the Antlia bandpasses were the deciding factors for me.

While the colour of my photos might not be exactly as one would see them with the naked eye, I'm sure they will be good enough for me 

Edited January 7, 2021 by Spongey