3085 replies to this topic

[Psion]

Okay, I get it. After all, anyone interested in this issue can calculate the SNR.

 

https://www.mirametr...culator_mvn.php

[smiller]

So, I'm going to do my best to spot check Tarbat's table for an example in Bortle 5 skies with and without the dual narrowband filter.

Here is Tarbat's table with those two cases outlined:

 

 

You can see for the No filter case at the bottom at 10 seconds exposure you get ~5% read noise from B5.

 

For the top case with the 50nm bandpass dual narrowband filter you get about 25-30% read noise at 10 seconds and ~12.5-15% at 20 seconds.   Not on the chart, but at a hypothetical 50-60 seconds it would back down to ~5% read noise again.

 

So using the tables that Robin Glover produced and showed in his video, and adjusting the results, per Glover's instructions, for the SeeStar S50 with the IMX462 sensor and Bortle 5 skies, I get this:

 

 

What I got for broadband is 10 seconds for 5% read noise and about 60 seconds to get to 5% read noise with the dual narrowband.  I get the same results!!!!!  Math works!

 

For the dual narrowband, at 10 seconds exposure, read noise will be roughly 6x of the 60 second exposure result, or about 30%, very similar to Tarbat's results.  And we didn't ensure to equalize all our assumptions that I made around quantum efficiency, sky background rate, etc...  So these comparisons should only be in the ballpark.

 

Nicely done Tarbat!    You need to do this for the Origin and publish your spreadsheet for others to play with!

Edited by smiller, 16 January 2024 - 11:35 AM.

[tarbat]

Nicely done Tarbat!    You need to do this for the Origin!

What do you think I've just been doing   For the Origin I've used:

• F-ratio = 1/2
• Pixel Size = 2.4
• QE = 81%
• Read Noise @ 100 gain = 1.72

 

Happy for someone to check my calculations!

Edited by tarbat, 16 January 2024 - 11:51 AM.

[smiller]

What do you think I've just been doing   For the Origin I've used:

• F-ratio = 1/2
• Pixel Size = 2.4
• QE = 81%
• Read Noise @ 100 gain = 1.72

 

 

Happy for someone to check my calculations!

Of course I should have known.... lol.

 

Your results look like what I would expect from an F2.2 system, but I haven’t checked your numbers.

 

Since the results are so dependent on read noise, actually read noise squared, and on some of these cameras the read noise continues to drop significantly with gain, I'm curious if there's a way to know what gain they actually shoot at.  Can we tell from the .FITS headers in the downloaded raw subs?

 

As an example of a real optimization case, I shoot short exposures on my 12" F4.9 (F3.7 with reducer) Alt/Az Goto Dob.  So it's just a big fat version of one of these scopes.   I shoot 5-10 seconds and I have both broadband and two different dual narrowband filters.

 

What I found is I push the gain up until i just start to get saturated stars and often this is much higher than the base high-gain mode.  And so my read noise may be 20-30% lower due to that and that cuts down total noise by a fair bit.  This is especially true with narrrowband.  So there are optimizations these scopes can do, now or in the future.

Edited by smiller, 16 January 2024 - 01:10 PM.

[tarbat]

Can we tell from the .FITS headers in the downloaded raw subs?

I know from the Seestar FITS header that it always uses gain = 80 (the HCG cutoff point).  The sensor in the Celestron Origin doesn't appear to have an HCG point, so I used a gain of 100, but who knows what gain Celestron will be using.

[GSBass]

I’m sold on results derived from Robin, gifted…. But so much depends on implementation by the robot makers, I’m not sure you can take Sony specs at face value…. The 462c as an example performs very different from  different companies implementing it, whether it be zwo, player one, QHY or Vaonis…. 

What do you think I've just been doing   For the Origin I've used:

• F-ratio = 1/2
• Pixel Size = 2.4
• QE = 81%
• Read Noise @ 100 gain = 1.72

 

 

Happy for someone to check my calculations!

[Psion]

I agree that calculations are okay to play with; practice tends to be quite different as it depends on the manufacturer's implementation of the sensor, as a colleague writes above.

[eyeoftexas]

"In theory, there is no difference between theory and practice.  But, in practice, there is."

– Jan L.A. van de Snepscheut

[jprideaux]

One of the beta-testers shared on Facebook that the software allows control of both exposure time and gain. He said that exposure times run from 1 second to more the 30 seconds, and the gain settings are 100, 200, and 400, characterized as ISO. The defaults seem to be 10 seconds and 100 gain.

Of course it is easy for the software developers to add in menus to allow changing these things but it is nice that they put these controls in the software at the start.

[jprideaux]

The following shows a comparison of the FOV between the origin and the SeeStar.  
very similar.  The big question is whether (and if so, when) will either of these provide a mosaic mode.  
 

 

[jprideaux]

And a FoV comparison between the Origin and the various Vespera models.  The Passenger isn’t shown here but it would have an even larger FOV (but less rectangular).  
 

 

smallest to largest is the following:

 

Origin (red)

Vespera-classic (green)

Vespera-pro (blue)

Vespera-2 (orange)

Vespera-Passenger (not shown)

 

 

 

 

 

[Psion]

Vespera Passengers has native field of view 2.4 by 1.8 degrees, in CovalENS mode has 17 degrees square, the largest of all the Vespera and Stellina instruments. Native resolution is not essential with CovalENS. Plus, you have a software rotator, so you can rotate the object in the field of view as needed.

[eyeoftexas]

The big question is whether (and if so, when) will either of these provide a mosaic mode.  
 

 

Amen.

[Rac19]

The big question is whether (and if so, when) will either of these provide a mosaic mode. 

I think that this is what Vaonis covalENS is. I assume that the others will follow.

 

It looks quite nice. You draw a box to frame the required image area and covalENS does the rest.

 

https://vaonis.com/c...edded-telescope

Edited by Rac19, 17 January 2024 - 04:50 PM.

[Psion]

The question is how they're going to do the mosaic. They certainly can't copy CovalENS, and it's not a simple thing that a programmer can put together in a few months.

[jprideaux]

Maybe Celestron and/or ZWO could give Vaonis a boatload of money to license the CovalENS intellectual property.

 

Or maybe they will do something more simple like have the telescope automatically take the separate stacked images needed for the mosaic but not live-stitch them all together and leave that for the user to do in 3rd-party software.

 

The Dwarf2 is currently doing that for their panoramic mode where once you define that you want a grid of (for example 3x3 images), it will take and save the 9 images (appropriately named) and then you just take those 9 images after the fact and import them into a 3rd-party software package to stitch-together the panorama..  

[Psion]

Yes, there are multiple ways to do a mosaic, but CovalENS has a unique procedure. it's actually not a standard mosaic, i.e. a composite of several windows, but it stacks the images one on top of the other with a certain offset (about 15%). Unfortunately, a mosaic in AZ mode is not at all as simple as an EQ mount. The images keep shifting against each other at different angles, and assembling them into a whole image is a very tricky thing due to the different rotations and gradients.

Edited by Psion, 17 January 2024 - 05:02 PM.

[Bob W4]

I think alot of people are missing the purpose of the Origin.  Personally, I think the package is intended for computer challenged idiots like myself.  Celestron has to know that "astro photographers" can put together a kit of their own liking for equal or less $, but guys like me that are lost in the technical dust, have no other alternative.    Along came Smartscopes to bridge the gap.  I love the results from my Seestat S50, but have developed "apeture/smart scope fever"  However, for $4000 bucks, I need to see a review first.  Cuiv, Galaxy Hunters???  I hope Celestron is "smart" enough to send the YouTube folks beta samples.

Edited by Bob W4, 18 January 2024 - 08:17 PM.

[eyeoftexas]

Cuiv posted a video on the Origin 3 days ago.  Unfortunately, all he could do was go through the specs and components online and comment on the pros and cons.  He ended the video by asking Celestron to send him one for review, so he doesn’t have one.

[Bob W4]

Cuiv posted a video on the Origin 3 days ago.  Unfortunately, all he could do was go through the specs and components online and comment on the pros and cons.  He ended the video by asking Celestron to send him one for review, so he doesn’t have one.

I can only hope the Celestron marketing folks are "smart" enough to get some trusted real world reviews put up online, not some models drinking fancy wine and eating crackers around the pool.  No way will I spend big bucks on an unknown product.  Trust me, I want to buy one, but not until I know it is worth the big $ Celestron wants for it.  I own 3 Celestron telescopes and trust their quality/value, but for 4000 bucks, "show me the money". 

Edited by Bob W4, 18 January 2024 - 08:41 PM.

[Bob W4]

To state in bluntly, Celestron ain't going to sell an unknown $4000 product without any trusted real world reviews. Celestron ... Wake up and smell the photons.

[jprideaux]

If you have previously been tempted to buy the Rasa-8 but were hesitant to try to figure out all the details of putting together your own system, then the origin could be the answer. For such a person, the $4000 would not seem so much (since you had probably already priced out such a system).

On the other hand, if you never before heard of a rasa, then the $4000 would probably seem way too much. The success of the system will be in the software and the long-term Celestron support for providing upgrades (both hardware and software).
No one has a crystal ball to look into the future to see what kind of upgrades Celestron will offer (and when).

At least we are fortunate that they are offering something.

[Psion]

With smart telescopes, it's all about the software. I think the big surprise will be when someone writes smart software for standard telescopes. That is, you buy any GoTo mount, telescope and camera, build it and run the "smart box", and everything will work just like a smart telescope.

[divers]

That will be a game changer for those of us that do not want to go down the rabbit hole of making astrophotography our life's passion or spend $$$$

With smart telescopes, it's all about the software. I think the big surprise will be when someone writes smart software for standard telescopes. That is, you buy any GoTo mount, telescope and camera, build it and run the "smart box", and everything will work just like a smart telescope.

[amitshesh]

I can only hope the Celestron marketing folks are "smart" enough to get some trusted real world reviews put up online, not some models drinking fancy wine and eating crackers around the pool. No way will I spend big bucks on an unknown product. Trust me, I want to buy one, but not until I know it is worth the big $ Celestron wants for it. I own 3 Celestron telescopes and trust their quality/value, but for 4000 bucks, "show me the money".

The single biggest weakness of iOptron. Last month when I was in the market for a mount, it was so difficult to find any reviews for the newer iOptron mounts. Just videos by ioptron, and those are akin to someone seeing a mount for the first time and going "ooh look at my telescope move"!

Hopefully Celestron does not make that mistake. We are in the digital era now. Advertisements will not do much. Reviews by third parties, online, is the way to sell things.

Edited by amitshesh, 19 January 2024 - 08:33 AM.