48 replies to this topic

[tarbat]

Agreed, but subjectively, the resolution and field of view are huge equalizers, and at least for me, the D3 images are amazing for the dinky toy-looking little thing they come from.

Yes, for some targets the D3's larger field of view is an advantage over the Seestar. For other targets, I prefer the sharper detail that the Seestar can achieve.

 

[tarbat]

That's why my biggest wish for the S30 is a fast processor to speed up it's mosaic stacking.

It's not a faster processor that's needed, but more accuracy in the geartrain to allow the Seestar to settle quicker after a mosaic position change. The significant amount of backlash in the geartrain means that any mosaic position change that goes against the sidereal rate is going to be problematic.

Edited by tarbat, 19 October 2024 - 03:49 AM.

[bradhaak]

You're probably right. That's why Celestron used better gears and spring loaded them in the Origin mount. It looks like the old Evolution mount but is evidently much better.

[LDW47]

The image resolution on the Dwarf 3 at 2.8 arcsecs/pixel is actually worse than that on the Seestar at 2.4 arcsecs/pixel. This is also apparent in the images posted so far from the D3, with the average FWHM being around twice that of the Seestar. And of course the D3's smaller aperture means it's Dawes limit is over 40% higher than the Seestar.

 

And yes, the D3's speed measured by signal per arcsec² is 65% slower than the Seestar, so you'll likely need to image for 3x longer with the D3 to get a similar SNR to that achieved with a Seestar. And the lack of HCG mode usage on the D3's sensor really pushes the sensor read noise to a very high level.

 

The main technical advantage the Dwarf 3 has is it's field of view area, being over 5x greater. That is a significant advantage which in my mind means owning both a D3 and Seestar might be a sensible choice until ZWO perfects the Seestar's mosaic function.

 

2.jpg

Kind of my thoughts save for owning both because they aren't available to own (D3). But there is the option of prepaying but for what and when. Its a great post but you have also gone against others thoughts, lol. I think ' a new champion ' is very premature until ZWO has finished their upcoming kick at the cat.  PS:  And I don't believe they are doing this just to catch up with the D3, its apparently the opposite, maybe. After all they are the steady, most prominent industry leader, they cater to the hundreds of potential, new Smart Scopers in every possible way and its all above board from what I have seen, right. Future progress is still going and I know where my $'s are, thank god.  IMveryHO.

[carver2011]

A friend of mine Beta tested the Dwarf III, and the ZWO update with the mosaic mode. He now owns a Dwarf III. He has owned a SeeStar s50 since it came out. He is very good at processing. His image of M31 with the Dwarf III IN EQ mode, is far better than his image of M31 with the SeeStar in mosaic mode. With the Dwarf's field of view, no need for mosaic mode. JMHO

[LDW47]

A friend of mine Beta tested the Dwarf III, and the ZWO update with the mosaic mode. He now owns a Dwarf III. He has owned a SeeStar s50 since it came out. He is very good at processing. His image of M31 with the Dwarf III IN EQ mode, is far better than his image of M31 with the SeeStar in mosaic mode. With the Dwarf's field of view, no need for mosaic mode. JMHO

Can you get him to let you post some examples, to show a comparison, eh.

[GSBass]

Can’t argue with numbers but in real world use it’s certainly accumulating the photons fast enough that it does not feel slow, my average exposure so far has been about 75 minutes and although longer is always better I have been pleased with the data so far… it’s enough to do full post on without falling apart and although I agree you can detect higher resolution with 50mm, I am not seeing any decreases of what was captured… probably due to the longer subs

Agreed, but subjectively, the resolution and field of view are huge equalizers, and at least for me, the D3 images are amazing for the dinky toy-looking little thing they come from.

 

Mosaic mode is great and hopefully ZWO can get the performance improved over the early dev version that you and I have. Even if they can get the stacking up to the speed of normal live stacking it will be a huge deficit in performance. A mosaic that is twice the standard size in both dimensions will take at least four times as long to get an equivalent capture. But then again, getting rid of walking noise and field rotation artifacts are an added bonus.

 

That's why my biggest wish for the S30 is a fast processor to speed up it's mosaic stacking.

[bradhaak]

Can you get him to let you post some examples, to show a comparison, eh.

As was pointed out below, and I meant to say...This is from an unreleased version of v2.1.0 of the Seestar software. It is what I would call pre-beta because based on the development commits from this build, features were still being added. It can be viewed as an indication of the minimal performance of the software that will soon be released, but performance and specific implementation features may change in the final version. This version was leaked onto the Internet without the permission of ZWO, is completely unsupported, and should not be used unless you are willing to risk blowing up your Seestar. I'm not kidding.

 

Also, using it will lock your Seestar into this version. You can't go back to the release build because of firmware incompatibilities.

 

Here is a Seestar Mosaic that I captured at a Bortle 5 site, You can see from the out-of-scope image that it is fifty-minutes of ten-second subs, but it took at least twice that long to get it. 

 

The first pic is a screen capture to show the mosaic being captured. The second is completely unaltered from the scope. I think I adjusted the contrast during the capture to get a darker background, though. The third is edited from the live-stacked FITS. It's overdone because I wanted to see how far I could push the data, and it turns out that I could push it pretty far.

 

 

 

Edited by bradhaak, 19 October 2024 - 05:26 PM.

[bradhaak]

And here is a capture with the Dwarf 3. It's 122 45-second subs in polar alignment mode. zero were rejected in either the live-stack or in the PixInsight stack. It's about 90 minutes of data. But the actual time to get it was about 2 hours. 

 

I would have rather used the stack from the scope to make the test more even, but had to stack it in PixInsight because of an issue with the D3. Currently, they don't provide an unstretched TIFF or FITS file. An unstretched file has been requested from multiple people including me, and they seem receptive. But until it happens, any editing of an unstretched image requires stacking in the app of your choice.

 

 

[bradhaak]

I'll try to get some decently long captures of the same target  with the Seestar in mosaic mode and the D3 in polar mode, as soon as that big round spotlight is out of the sky for a few nights.

 

I'm going to a club outreach event tonight, but I don't expect anything useful to come of it. For the Seestar, I'm just going to point it at the moon and run it over to a portable monitor. I'll use another scope, probably the Origin, and point it at various emission nebulae so I can use the dual-band filter and get rid of as much ambient light as I can.

 

I'm not using or even considering the Dwarf 3 tonight because it takes too long to generate a decent onscreen image, and these outreach sessions are mostly EAA. The Origin is the king of fast results, at least for now.

 

What's the old racing saying? "Speed is money. How fast do you want to go?" It seems to apply to our hobby, too.

[tarbat]

 

Here is a Seestar Mosaic that I captured at a Bortle 5 site

Note that this is beta software, not what's available to most owners. Best not to compare to an early release of the Seestar app that hasn't been optimised yet.

[bradhaak]

It was requested, so I posted it. I meant to state that it is from an unreleased dev build, and will edit my post to say so. But, as you can see, while slow, the results are impressive. However, I have turned off the feature in my Seestar because of slow performance.

 

I agree that it is not final (or even close to final), but it is available from multiple locations. 

Edited by bradhaak, 19 October 2024 - 05:26 PM.

[Podiadoc]

I know the Seestar s50 is an Alt/Az mount whereas the DWARF 3 is an equatorial one. Does that mean the DWARF 3 does not need to have Alt/Az and will function better in terms of tracking? Thanks

[bradhaak]

Yes and no. DwarfLab has announced a mosaic mode that would remove most of the need for equatorial mode. It will be interesting, if they ever ship it, to see which works better. But in both cases, you're stuck with the slow acquisition times of the Dwarf 3. So, take your pick. Personally, I've decided that a lower resolution scope with mosaic mode is better for my purposes (S30, S50).

 

The EQ mode lets you have longer exposures. But, even with less than one degree from perfect polar alignment with the Dwarf 3, I haven't had any luck with exposures longer than 45 seconds. Even that is a huge improvement over ten or fifteen seconds for alt-az mode, but it isn't amazing, and you probably need a guided scope to get much better. The other advantage of polar alignment is pointing up at altitudes higher than 80 or 85 degrees. This is real and is a real advantage...sometimes. Most of the sky isn't going to going to move higher than 80 degrees. It hit a lot of people when the S50 received it's mosaic mode, but that was because everybody was trying to capture M31, which does happen to pass close the the zenith for a lot of folks, myself included. But for any targets a little further north or south, this isn't an issue. Keep it in mind, but don't get worked up about it.

 

Now for the disadvantages. These are based on my experience and are also subjective. 

 

The equatorial mode is a pain to get aligned decently. DwarfLabs have done a decent job simplifying it. But, in order to make it easy to get a basic polar alignment without a ton of trouble, they've made it very difficult to get a good polar alignment. They also put a very nice sensor into the D3. But it overpowers the small 35mm aperture of the scope, so that getting the kind of captures the scope is capable of takes a very long time. And don't get me wrong. If you give it enough time, the Dwarf 3 is capable of making extremely pretty pictures. It's guiding is pretty good, and the rejection rate of even 45-second subs, is low. If it weren't, the acquisition time would be even longer and too much for my tastes.

 

But the lower resolution solution of the S50 or S30, with mosaic mode is probably better for most folks. Again, this is my opinion. The mosaic mode, as implemented, is very easy to set up and removes field rotation artifacts and walking noise from images. With this solution, you have a lower resolution sensor that is going to capture an image much faster for a small target. Even for smaller mosaics, you can capture them quickly.

 

Last month, I did a 45-minute capture of the Pleiades that was just slightly larger than normal at 2496x1404 (compared to the standard resolution of 1920x1080) This gave me a nice open area around the cluster, and let me use the mosaic advantages of no field rotation or walking noise. If I'd used the Dwarf 3 to do the capture, it would have taken longer because I would have been capturing a wide-field 6 MP image. Then I would have cropped it down to what I actually care about.

 

You should make up your own mind, and I don't expect you or anyone else to agree with my opinions. But to me the perceived advantages of polar alignment are rapidly disappearing.

[jprideaux]

I agree that the practical exposure length is impacted by both the field-rotation but also the tracking ability of the mount and with the very inexpensive mounts built into these smart-telescopes like the Dwarf2/3 and SeeStar50/30 that will place a limit on how long the sub-exposures can go even it equatorial aligned.

 

Of course just going up to 45 seconds or so would help in darker skies especially if a filter is being used.

 

I do like not having to worry about what gets too close to Zenith so that is a nice feature of the new Dwarf software to natively support equatorial alignment even if the setup does require some extra manual steps (to polar align).  I look forward to the official SeeStar software also eventually supporting equatorial alignment.  I'm sure it will come at some point since some hackers are already doing that to a limited extent with the SeeStar with using some custom software.

 

I'm also looking forward to the Celestron Origin also eventually supporting equatorial alignment.  That should come sometime next year.

 

I don't think the Vaonis scopes with their current physical configuration can support equatorial alignment. 

[bradhaak]

The main advantage of longer exposures is lower noise, but that isn't a real issue any more because of the constantly improving state of noise removal software. That's what I meant, above, when I said that the advantages of polar alignment are rapidly disappearing. 

 

I doubt that I'll even get a wedge for my Origin. Polar alignment isn't really necessary any more and is, relatively speaking, a pain in the butt for minor gains in increasingly limited circumstances. I've spent quite a bit of time playing with polar alignment on my D3, and I'm ready to declare it as an interesting experiment, but not necessary. 

 

But again, this is my subjective opinion and I respect that other folks will disagree and are welcome to do so.

[GSBass]

A lot of my d3 judgements are kind of up in the air until probably several software/firmware releases, people I have discussed longer subs issues with that have access to internal company builds seem pretty convinced that the hardware is not the limiting factor and that we are likely to see improvements across the board… we will see…. I also agree with you that the mosaic modes smart scopes are developing does make eq mode less relevant, however I started an andromeda photo pretty close to zenith last night that worked well and there is no way I could have done that with my vespera… perhaps not that big of a deal… I am use to waiting for targets to track in to the sweet zone… but it was nice to have that capability. Btw, thought it was nice of you to offer your d3 up at cost given the delay… but do wish you were keeping it, I would have appreciated seeing your opinion/results as it evolves

The main advantage of longer exposures is lower noise, but that isn't a real issue any more because of the constantly improving state of noise removal software. That's what I meant, above, when I said that the advantages of polar alignment are rapidly disappearing. 

 

I doubt that I'll even get a wedge for my Origin. Polar alignment isn't really necessary any more and is, relatively speaking, a pain in the butt for minor gains in increasingly limited circumstances. I've spent quite a bit of time playing with polar alignment on my D3, and I'm ready to declare it as an interesting experiment, but not necessary. 

 

But again, this is my subjective opinion and I respect that other folks will disagree and are welcome to do so.

[GSBass]

Btw…I have been accepting my d3 polar alignment at 2 degrees out because I can get that fairly easily now… can’t make definitive statements yet but thinking this may be close enough and may not be affecting tracking… in fact whether I am seeing any drift visually seems to be target location and perhaps dither settlement…. But anyway… main point is polar alignment may not be as critical as I thought a month ago… I think I would advise people to try again if it said they were 4 degrees out but at 2 I think you can accept without worry

The main advantage of longer exposures is lower noise, but that isn't a real issue any more because of the constantly improving state of noise removal software. That's what I meant, above, when I said that the advantages of polar alignment are rapidly disappearing. 

 

I doubt that I'll even get a wedge for my Origin. Polar alignment isn't really necessary any more and is, relatively speaking, a pain in the butt for minor gains in increasingly limited circumstances. I've spent quite a bit of time playing with polar alignment on my D3, and I'm ready to declare it as an interesting experiment, but not necessary. 

 

But again, this is my subjective opinion and I respect that other folks will disagree and are welcome to do so.

[GSBass]

Also… not sure of technicals but after a few nights of eaa this week, I noted stars seemed to be much tighter track on way down from zenith vs on the way up…. Unsubstantiated… just an unverified observation

[Morgi75]

This is not an appropriate comparison. Because of the higher resolution of the D3, the visible object size is not much smaller:

 

  

 

Yes, for some targets the D3's larger field of view is an advantage over the Seestar. For other targets, I prefer the sharper detail that the Seestar can achieve.

 

 

Attached Thumbnails

• 

[MattPenn]

The image resolution on the Dwarf 3 at 2.8 arcsecs/pixel is actually worse than that on the Seestar at 2.4 arcsecs/pixel. This is also apparent in the images posted so far from the D3, with the average FWHM being around twice that of the Seestar. And of course the D3's smaller aperture means it's Dawes limit is over 40% higher than the Seestar.

 

And yes, the D3's speed measured by signal per arcsec² is 65% slower than the Seestar, so you'll likely need to image for 3x longer with the D3 to get a similar SNR to that achieved with a Seestar. And the lack of HCG mode usage on the D3's sensor really pushes the sensor read noise to a very high level.

 

The main technical advantage the Dwarf 3 has is it's field of view area, being over 5x greater. That is a significant advantage which in my mind means owning both a D3 and Seestar might be a sensible choice until ZWO perfects the Seestar's mosaic function.

 

2.jpg

How is 4.8 square degrees divided by 0.9 square degrees equal to 104%???  Might want to check your arithmetic there...

[GSBass]

Current experimentation plans, I’m finally investing in a filter holder for my D3, ordered this adapter I found on thingiverse and having xometry print it in nylon. I’ll use the 2” to test my baader neodymium filter and my l’enhance, I decided to not install magnets and just secure with painters tape so that I have to option to flip it for use on wide angle and telephoto…the magnet pattern would prevent you from doing that otherwise. On the other side I am going to epoxy a 37mm ring so that I can screw in the Raynox teleconverter to run some experiments, I know from experience with the d2 it will work well on the moon but will just have to see if it works with Dso , the main culprit on d2 was centering because you don’t have access to manual movements without stopping tracking. Anyway should be fun… as always I’ll post results

Attached Thumbnails

• 

[GSBass]

The thing about numbers is they don’t really tell you that much about real world results..as comparisons that have been done so far confirm… if your just looking at the target at the same scale/fov…. Ie zooming out the Seestar image instead of zooming in the d3 image, you will be hard pressed to see a difference and at least on some targets the d3 looks better, the differences between 50 mm and 35mm and all the other numbers referenced only become more obvious if you crop/zoom in to match the tiny fov on the Seestar… then you can note the detail differences. I’m not saying it’s not a worthy thing to consider… especially if you’re just going to buy one or the other and not both…. Main take away, they are very different scopes with different purposes and both perform well at what they were designed for

How is 4.8 square degrees divided by 0.9 square degrees equal to 104%???  Might want to check your arithmetic there...

[GSBass]

Plan my night demo https://www.youtube....h?v=0mQcl-vPz_E