DWARF 3 vs Draco on the Heart Nebula: What’s Not to Love?

Two FITS files sat in the same Siril folder on the same October evening, and the two finished images were edited in Siril twelve minutes apart. One stack held 468 frames from the 35mm DWARF 3 I have been shooting with for fourteen months; the other held 36 frames from DWARFLAB’s 90mm Draco, captured under Bortle 3 skies by the company that built it. Both were pointed at the same target: IC 1805, the Heart Nebula in Cassiopeia. A big thank you to the Dwarflab team for sharing these. If anyone is interested the links can be found in the Draco Facebook group.

Let’s get into it. On paper this was never a fair fight. The Draco brings more than twice the aperture, a cooled sensor and a far darker sky, so the DWARF 3 should have walked in to lose. Then I put the two images side by side. The Draco image left me in awe, the DWARF 3’s wide view was the one I loved, and I still can’t say I love one over the other. Both are mesmerizing.

Fourteen months with the little one

My attachment to the Dwarf 3 has a history. M51 was my first target with the DWARF 3 back in August 2025, and by the one year mark this mighty little scope had given me 27 deep sky targets and 157 hours 23 minutes of integration. Every one of those hours came from a Bortle 6 driveway where a target has to climb past 40 degrees before it clears the trees.

It is also the scope I push hardest, and the Ou4 Squid Nebula shows how far. Nearly 22 hours through the built in Duo Band filter turned up no detection at all, and a 2 inch duo band filter with a flaw in its glass had to go back. Its replacement, an Optolong L-eXtreme 7nm dual band filter mounted on the front of the scope, has now carried the Ou4 data past 51 hours, and the stack is starting to show something. More on that soon.

The Heart nebula, IC 1805, is not new to this scope either. My earlier DWARF 3 pass, in November 2025, was 3 hours 52 minutes through the built in Duo Band filter. This time the 7nm filter went on the front: 468 subs at 60 seconds and gain 90, 7 hours 48 minutes in total, ending in the early hours of October 19. It is the best data of this target the DWARF 3 has given me, and it was about to be measured against its bigger sibling.

Enter the Draco

That Draco started out as a rumor. Before DWARFLAB published a spec sheet, I pieced the Draco together from six clues, including a NEAF 2026 teaser, patent CN121679866A and an unannounced telescope profile inside the Atlas app. The Dragon and the Dwarf then set it against the DWARF 3 on paper. My own Draco, the SHO edition, is due in November.

The launch spec sheet backed up what that post had inferred: 90mm of aperture, 340mm of focal length at f/3.8, a cooled 1/1.3 inch 50 megapixel sensor and subs out to 300 seconds. The filters are part of the design, with 13nm dual band pairs built in (Ha with OIII, plus SII with OIII on the SHO edition).

Until mine arrives, I work with sample data DWARFLAB provides from Draco test units. In September that meant a first look at the Eastern Veil and now this time it is the Heart Nebula stack of 36 subs at 300 seconds and gain 120 with the sensor held at 1°C under Bortle 3 skies. The FITS header spells out how it was made: matched master darks, no flats and a label that reads “offline experiment only.”

The tale of the tape

Put the two datasets next to each other and the mismatch is plain. The capture numbers come straight from the two FITS headers.

	DWARF 3	Draco
Aperture	35mm	90mm
Focal length	150mm, f/4.3	340mm, f/3.8
Filter	7nm, front	13nm, built in
Subs	468 × 60 s	36 × 300 s
Gain	90	120
Sensor	19°C	1°C, cooled
Total	7 h 48 m	3 h
Sky	Bortle 6	Bortle 3
Date (UTC)	Oct 19	Oct 16

Read it like a weigh in and the Draco is heavier on nearly every line: a lens more than twice as wide, longer subs, a colder sensor and a darker sky. The DWARF 3 arrives with exactly one edge, the filter. My front mounted Optolong L-eXtreme passes 7nm, against 13nm for the Draco’s built in pair and that detail comes back in round four.

Round one: light

The heavier side of that table turns directly into physics, and the physics is brutal for the small scope. Collecting area grows with the square of the aperture, so 90mm against 35mm is 6.6 times the area. To match what the Draco gathered in 3 hours, the DWARF 3 would need about 20; it got just under 8, so it entered this comparison with roughly 40 percent of the photons. Treat that as a rough figure. The sensors and filters differ, and if the Draco uses the folded optical design described in its patent, a central obstruction takes a slice out of its collecting area.

Aperture also sets the resolution limit. The Dawes limit, 116 divided by the aperture in millimeters, gives 3.3 arcseconds for the DWARF 3 and 1.3 for the Draco, although the atmosphere on a typical night blurs stars beyond 1.3 arcseconds. My DWARF 3 frame plate solves at 2.79 arcseconds per pixel, so its optics and its pixels are closely matched; detail finer than that never reached the sensor, and no deconvolution can invent it.

Cooling adds a third advantage. The Draco’s sensor ran at 1°C while the DWARF 3 sat at 19°C; thermal noise climbs as a sensor warms, and that difference keeps adding up over a five minute sub.

Round one goes to the Draco, and it produced the clearer image. That is the physics of a larger aperture and longer exposures.

Round two: the wide view

The same physics hands the DWARF 3 its counterpunch, because a short focal length buys a wide field. At 150mm the little scope fits the whole heart shape in one frame, and even my crop spans 2.75 by 1.46 degrees. Melotte 15, the young cluster whose hot stars sculpt the nebula with their winds and radiation, sits near the center. The Fish Head Nebula (IC 1795) glows in the upper right corner with a dark lane cutting across it, while a thin red arc traces the top of the Heart and the outer walls curve down both sides.

This is the view I loved when the two images sat side by side. The Draco sample, framed much tighter on the core, is a gorgeous crop on its own though. But you would not necessarily know you were looking at a heart. For a target this size the DWARF 3 tells the whole story in a single look, so round two goes to the little one.

Round three: the core

Zoom into the middle, though, and the numbers from round one start showing up on screen. The Draco’s bigger aperture and longer focal length are why Melotte 15’s stars separate into clean points, and why the dark pillar below the cluster has a crisp edge with a reddish rim glowing along it. The same finer scale breaks the dust on the left wall into ridges and filaments. The extra light is the likeliest reason for the pale gray teal haze across the interior of the Heart, which is oxygen emission caught by the dual band filter.

My DWARF 3 image has every one of those landmarks (the pillar, the cluster, the bright knot of gas beside it), only softer, with more stars crowding the field and little of the oxygen haze. This is the part of the comparison that left me in awe of the Draco, and round three goes to it without a contest.

Round four: the sky and the clock

The last round is about conditions. The Draco had Bortle 3 skies, where the background barely competes with the nebula. I had a Bortle 6 driveway, so the DWARF 3 leaned on its 7nm passband. At roughly half the width of the Draco’s 13nm filters it rejects more of the skyglow sitting around the Ha and OIII lines. A narrow filter also costs some signal, and my 7nm vs 30nm test on the Veil showed that 25 minutes was too short for it to pay off. Nearly eight hours on the Heart gave it far more time.

The clock still favored the Draco, which needed three hours to the DWARF 3’s nearly eight, and that gap is exactly what aperture buys. Call round four a draw. The Draco wins on time, while the DWARF 3 wins on showing what a small telescope with one well chosen filter can do from a compromised sky.

Same hands, same steps

Neither image got special treatment. Both stacked FITS files went through Siril 1.4.4 in the same order, on the same evening:

1. Crop the stacking edges
2. Automatic background removal script
3. Color calibration
4. Denoise
5. Deconvolution
6. Separate stars from the nebula
7. Denoise the starless image
8. Stretch the starless image
9. Open the star mask, denoise it, and give it an asinh stretch
10. Recombine stars and nebula
11. Green noise removal

Months of grinding on the Ou4 data taught me this sequence, and this time it felt easier. Because the steps and their order were identical for both files, whatever differs between the two results came from the telescopes and the skies, not from the processing.

Both of them win

So who takes the Heart? After four rounds, the answer is the same one I had the moment the images sat side by side: both of them.

The Draco is the specialist. Give it a dark sky and three hours and it resolves structure in the core that a 35mm lens cannot reach, which is what left me in awe. The DWARF 3 is the one that shows up night after night. It framed the whole Heart and the Fish Head from my driveway powered by more than a year of practice and experiences from my sky already behind it. Both scopes create amazing images, just at different scales.

Those different scales are also what make them good companions. After the Veil data in September I wrote that running the two side by side looked like a productive setup, and the Heart makes that case more clearly: the DWARF 3 frames the whole object and its neighbors, while the Draco moves in close on the region that rewards aperture. One image tells you where you are, and the other shows you what is there.

Fingers crossed, my own Draco arrives in November, and its first job will be this same target from this same driveway. With Bortle 3 off the table, aperture and cooling will be the only differences left between the two. Two solid companions, one patch of Cassiopeia. What’s not to love?

Stay tuned for more comparisons. Cloudy nights are in the forecast and M42 is at the center of my next write up.

Until then, clear skies!

AK

---

Frequently Asked Questions