I don’t quite understand ROI!

[nicspenceryork]

Hi all,
I have had a ZWO ASI178MM for a few years. I recently bought a ZWO183MM. Both have the same pixel size 2.4um.
The resolution of the 178 is 6.4MP 3096x2080 and the 183 has a resolution of 20.18MP 5496x3672. If I’m honest I don’t really understand what that means other than the 183 has a higher resolution so that means it picks up more detail (does it?!) Until today I’ve been using the 178 with a 2x Barlow without a specified ROI. Today I didn’t use a Barlow and chose an ROI of 1280x960 as a point to try which was probably a bit small. I was going to try a few different ROIs but I only had one 30s gap in the cloud and was grateful for what I DID catch! I really noticed when I zoomed in on the image after processing that the detail was nowhere near as good as it usually is when I use a Barlow.
I feel like there’s a penny about to drop… but it’s just currently hanging in the air and needs a nudge from one of you… what is it I don’t understand?!
Does the smaller ROI record a lower quality video?
Is it better to use a 2x Barlow with all of the sensor in use to zoom in?
Would the 183 using an ROI without a Barlow give a better zoom in than the 178 with a barlow?
What am I missing?!
I use a Skywatcher Evostar 80 Ed ds pro.
Yours in confusion,
Nic

[GreatAttractor]

Assuming the same sensor, setting a ROI and using a Barlow do two different things. A Barlow increases the image scale (plate scale); e.g., if with a given telescope the naked sensor sees 2"/pixel, than a 2x Barlow will make it 1"/pixel. On the other hand, ROI selects a fragment of the sensor to capture, without affecting the image scale.

So, if your 80ED + 178MM needs a 2x Barlow for optimal sampling, imaging without a Barlow gives you worse-than-optimal details. Enabling ROI still leaves you with the suboptimal resolution, just the resulting image is smaller than 178MM's full 3096x2080.

Is it better to use a 2x Barlow with all of the sensor in use to zoom in?

Possibly, depending on the "critical sampling"; here's an example thread on this.

Would the 183 using an ROI without a Barlow give a better zoom in than the 178 with a barlow?

No, since both have the same pixel size. 183's sensor is just physically bigger and for the same telescope covers a larger FOV.

[KMH]

Nic,

The ROI feature is basically just a crop of the full field - only the data from the selected ROI is read to the computer. This reduces files size, since only a portion of the field of view is being recorded. More importantly, it increases frame rate. It's often used for planetary imaging where the full field would show a planetary disk surrounded by a large sea of black. The ROI eliminates this unwanted (unless you are trying to capture moons as well, far from the disk) background and the higher frame rate aides with lucky imaging. It could of course also be useful for imaging small features on the sun for the same reasons.
But, the ROI does nothing to increase the image scale per pixel, so it is not a substitute for increasing the effective focal length. That's where the barlow comes in. I don't know the details of your scope, but with the pixel size you have you should probably do well with the 2x barlow.

Hope that helps,
Kevin

[rigel123]

I might be totally off base, but if they have the same pixel size, then they would have the same resolution but the 183 has a larger chip so you can fit more into the FOV with a Barlow then you can on the 178, so in that case, imaging a full disk would have a higher resolution with the 183 because you are spreading more of the disk over a larger number of pixels.

[nicspenceryork]

CLINK!
That's the sound of the huge penny dropping. I understand the principles now thanks to kind input from you. I've puzzled over it for a long time. I wish I had asked earlier!
I will read the information on the link about 'critical sampling' now, that'll be a bit trickier to fathom I think.
Your help is hugely appreciated, thank you,
Nic

[petevasey]

Hi, Nic,

One thing that hasn't been mentioned here is Dawes Limit. The limit theoretically describes the finest detail that can be observed through a telescope under ideal seeing conditions, and can be simply calculated by the formula 116/D where D is the diameter of the telescope objective in mm. So for your 80mm telescope, the result is 116/80 = 1.45 arc-secs. Yet another useful formula is (Pixel Size / Telescope Focal Length ) X 206.265 which gives the arc-seconds per pixel. At your focal length of 600 mm and a pixel size of 2.4 um, the image scale is 0.825 arc-secs per pixel.

So in your case as you can see the image scale is much less than the resolution of the telescope, so it's unlikely that a Barlow would improve the final image. On the very rare occasions of perfect seeing, you might just beat the theory, but most of the time leave the Barlow off - it's just adding extra glass and reducing the brightness of the image.

Cheers,
Peter

[marktownley]

Dawes Limit.

Doesn't take into account the wavelength being imaged.

Resolution at CaK wavelengths is 1.6x higher than at Ha wavelengths. Sampling etc therefore varies accordingly.

[petevasey]

Thanks, Mark,

I didn't know that. And I don't suppose the Rev. Dawes considered narrowband wavelengths either ;-)
Cheers,
Peter