DWARFLAB Draco is an unannounced smart telescope profile that surfaced inside the DWARFLAB Atlas app, showing a telephoto field of view of 1.87 by 1.05 degrees and a wide-angle field of view of 70.1 by 43.2 degrees, with no official spec sheet, product page, or press release behind it. Lined up against a matching patent filing and a NEAF 2026 teaser, those two numbers are the anchor for everything else in this piece.
This started at NEAF 2026, where DWARFLAB publicly hinted that something bigger than the DWARF 3 was in development and said nothing about what it actually was. That kind of teaser is easy to forget about. It sat unexplained for weeks.
Then a patent turned up. Chinese filing CN121679866A, from DWARFLAB’s own research arm, describing a folded telescope lens compact enough to fit a phone sized housing, with no confirmed connection to any specific product.
Then, in the latest Atlas app update, a new device profile appeared: Draco, attached to nothing DWARFLAB has ever shipped, with both a telephoto and a wide-angle field of view listed for it. Another user in the community spotted it first and posted about it. Pulling up the same listing afterward confirmed it, rectangle and all, the app framing Draco’s sky coverage the same way it does for every real telescope: 1.87 by 1.05 degrees on the telephoto camera, 70.1 by 43.2 degrees on the wide-angle.
None of these three pieces of evidence would be worth a full article alone. A stray app entry could be a placeholder. A patent could describe a design that never ships. A single field of view number could come from almost any combination of sensor and focal length. What changes the picture is having six things line up in the same direction at once, in the order they actually surfaced: the NEAF teaser, the patent, the Draco profile itself, its telephoto field of view, its wide-angle field of view, and the DWARF 3’s existing architecture as a baseline to measure all of it against. The rest of this piece works through all six and shows the math behind each one.
Everything past the existence of the Draco profile and its two field of view numbers is inference, not confirmation. DWARFLAB has not announced this device, named it, or published a spec sheet. Where this piece states something as fact, it is either a number read directly off the app or a figure pulled straight from a patent or sensor datasheet. Everything else is labeled as a calculation, a scenario, or a guess, and confidence is stated in plain terms rather than invented statistics.
Why six weak signals matter more than one
A single unconfirmed spec is easy to dismiss, and should be. It could be a typo, a placeholder, a test build that never ships. But several independent observations, each inconclusive on its own, are harder to dismiss once they all happen to be consistent with the same specific, fairly narrow explanation.
Each piece of evidence below could have pointed somewhere else and did not. The wide-angle field of view could have matched an assumption that DWARFLAB reused an existing camera. It did not. The patent’s three example focal lengths could all have predicted similar fields of view. Only one landed close. Where the evidence does not fit cleanly, that is noted in its own section rather than smoothed over.
The first clue: telephoto field of view
A field of view is not just a number sitting inside a settings screen. It is the output of two physical decisions a lens designer already made, sensor size and focal length, related by one formula.
FOV ≈ 2 × arctan(sensor dimension / (2 × focal length))
Once the field of view is known, that formula narrows the space of physically possible hardware quite a bit, even without knowing either input on its own.
The DWARF 3’s telephoto camera has a field of view of 2.93 by 1.65 degrees, confirmed both by working the formula backward from its published 150mm focal length and 7.68 by 4.32mm sensor, and by matching third party reviews. It also matches the Key Specifications table on this site’s own DWARF 3 page.
| Telescope | Telephoto field of view |
|---|---|
| DWARF 3 | 2.93° x 1.65° |
| Draco | 1.87° x 1.05° |
Draco’s telephoto field is about 1.57 times narrower along each axis, and it covers only around 41 percent of the sky area the DWARF 3 captures in one exposure. That reads like a telescope built around image scale rather than wide framing.
That box is not a one-off. The same 1.87 by 1.05 degree rectangle shows up under the Draco Tele profile no matter which target the app is pointed at, which is exactly what you would expect from a fixed sensor and focal length rather than a per-object placeholder.
Before the patent turned up, there were two mathematically sound explanations for that number and no way to prefer one over the other. If Draco reuses the DWARF 3’s Sony IMX678 sensor, the field of view implies a focal length of roughly 235mm. If it instead uses the larger Sony IMX585, the same field of view implies roughly 340mm. Both numbers come out of the same formula, both are physically valid, and field of view alone could not tell you which sensor sits behind the glass.
The second clue: wide-angle field of view
The wide-angle number is newer, and it deserves the same treatment rather than a passing mention.
Atlas lists Draco’s wide-angle field of view at 70.1 by 43.2 degrees, an aspect ratio of about 1.62 to 1, noticeably different from the telephoto camera’s 16:9 shape. The DWARF 3’s own wide-angle lens has a published real focal length of 6.7mm and a 3.4mm aperture, marketed with a 45mm 35mm-equivalent focal length, but DWARFLAB has never published an exact degree figure for it, which makes this comparison harder than the telephoto one and worth being honest about.
Two separate pointings, the same box both times: that is the same consistency check the telephoto screenshots above pass, and it is why this piece treats 70.1 by 43.2 as a real, repeatable app value rather than a one-time render glitch.
Two rough methods bracket an estimate. Converting the published 45mm equivalent using the standard full-frame reference gives a horizontal field of roughly 44 degrees, assuming the conventional 3:2 aspect used in equivalence math. Assuming instead that the wide camera uses a 16:9 crop similar in style to the telephoto camera, run through the same field of view formula at 6.7mm, gives something closer to 60 degrees horizontal, depending heavily on a sensor size that is not published anywhere. Call the honest range 44 to 60 degrees. Either way, Draco’s reported 70.1 degrees sits above the top of that range.
That matters for a specific hypothesis worth naming directly: maybe DWARFLAB kept its existing wide-angle, plate-solving, and panorama camera untouched and only redesigned the telephoto optics. It is a reasonable guess. Reusing a working navigation camera while overhauling the main imaging lens is a sensible way to control both cost and risk on a new product. But the numbers do not support it cleanly. If the DWARF 3’s wide sensor were carried over unchanged into Draco, the real focal length would need to shrink from 6.7mm down to somewhere around 3.8mm to produce a 70.1 degree horizontal field, which is not “the same camera,” that is a materially different lens.
A few explanations survive that gap. DWARFLAB may have built a new wider finder camera specifically because a narrower telephoto benefits from more margin on the wide side. A narrower zoomed-in main camera makes target acquisition and plate solving harder, unless the companion camera gives the algorithm more sky to work with. So widening the finder as the telephoto narrows is a coherent design choice, not a contradiction. It is also possible the 70.1 by 43.2 figure is a provisional value in a beta build that has not been tuned to match final hardware, which is exactly the kind of number that should not be trusted too far. And it is possible the DWARF 3 wide-angle baseline used in this analysis is simply off since no official source states it in degrees and both of the estimates above are indirect. Of those three, the new-finder-camera explanation fits an engineer’s incentives best, but it is the weakest of the two data-grounded numbers gathered here, and it should carry that weight in the final tally.
What the patent says
Chinese patent CN121679866A was filed on February 2, 2026 and published March 17, 2026, weeks before DWARFLAB teased a bigger telescope at NEAF 2026. The applicant is Xiaophoton (Wuhan) Technology Co., Ltd. The French site Astronotrip, which first reported the filing, describes Xiaophoton as DWARFLAB’s research and development subsidiary. The patent itself never mentions DWARFLAB or Draco, and it was written before either the app listing or the NEAF teaser were public knowledge.
Reading the full filing on Google Patents rather than relying on secondhand reporting, it describes an optical system that folds its light path across two mirrors and routes it through a small negative power lens sitting inside a hole drilled through the primary mirror. That lens spreads the converging rays slightly before they reach the sensor, the same trick a Barlow lens performs, stretching the effective focal length without physically moving the sensor further back. The patent’s own background section states the reason for all of this directly: conventional folded lenses top out around f/10 to f/15, which means much longer exposures for the same brightness, and astrophotography specifically needs better than that.
Three worked examples appear in the patent text.
| Example | Front element diameter | Effective aperture from stated f ratio | Focal length | f ratio | Housing length |
|---|---|---|---|---|---|
| 1, fast | 98mm | ~90mm | 339.92mm | f/3.78 | 145mm |
| 2, variant | 97.8mm | ~90mm | 350mm | f/3.89 | 143.89mm |
| 3, long | 91.9mm | ~90mm | 540mm | f/6.0 | 149.96mm |
Effective aperture here is the focal length divided by the patent’s own stated f ratio, which comes out a little smaller than the physical front element. That gap is normal; lens elements are usually cut wider than their working clear aperture to leave room for the mount.
Example 1’s 339.92mm focal length is the number that matters. Paired with an IMX585 sensor’s 3840 by 2160 output and run back through the field of view formula, it predicts a telephoto field of 1.88 by 1.06 degrees against Draco’s actual 1.87 by 1.05. That is a match within two hundredths of a degree, close enough to fall inside the rounding you would expect from reading a number off a drawn box rather than a printed spec sheet.
Example 2’s 350mm focal length predicts a visibly wider 1.82 by 1.02 degrees, a worse fit.
Example 3’s 540mm rules itself out entirely, predicting a field of only 1.18 by 0.67 degrees, roughly a third the size of what Atlas shows. Run the same 339.92mm focal length against the DWARF 3’s smaller IMX678 sensor instead, and it predicts 1.29 by 0.73 degrees, ruling that sensor out at this particular focal length too.
None of that confirms DWARFLAB is shipping Example 1 inside a telescope called Draco. A patent defines a family of possible designs, not a single locked configuration, and it is worth being direct about what this evidence is and is not.
Example 1 is currently the closest match to the measured field of view. The agreement is genuinely remarkable given how tightly it lands. It remains supporting evidence, not confirmation, and a company can patent more than it ships, file broadly on purpose, or change a design’s details substantially in the eight months between a filing and a shipping product. One commenter on the original Astronotrip post made this exact point, and it deserves to be taken seriously rather than argued past.
What moves this from coincidence toward a real lead is that three separate facts point the same direction at once. The patent names astrophotography as its stated use case in its own words, not our interpretation. The filing date sits right before DWARFLAB’s own NEAF teaser. And the one focal length in the set that was ever going to match Draco’s field does, to within measurement noise, while the other two clearly do not.
Fitting 340 millimeters into 145
The interesting part is not the 340mm focal length by itself. It is fitting that much optical path into a housing less than half as long.
A straight refractor needs a physical length close to its focal length, which is why a conventional 340mm lens does not shrink into a phone sized body. This design bounces light backward off the primary mirror, forward again off a small secondary mirror plated onto the back of the front element, and finally through the hole in that primary mirror to reach the sensor, cutting the physical path by more than half before the diverging lens does the rest of the work.
The tradeoff is a central obstruction from that secondary mirror, sized in the patent at 30 to 45 percent of the front element’s diameter. That reduces effective light gathering and softens contrast a little compared with an unobstructed refractor of the same aperture, the same tradeoff every compact catadioptric design has made since long before smart telescopes existed, just applied here to something that fits in a backpack.
Folded optics also carry engineering costs beyond the obstruction itself: the primary and secondary mirrors have to stay aligned to a fraction of a millimeter across a temperature range that can swing forty degrees between a warm garage and a cold field, and a folded system with three reflective and refractive surfaces accumulates manufacturing tolerance stack-up faster than a simple two-element refractor does. None of that is a reason to doubt the design works. It is a reason the patent spends real text describing how it manages those tradeoffs rather than ignoring them.
The patent’s own numbers suggest the tradeoff was managed carefully. The design keeps its RMS spot radius under 1.5 times the airy disk radius across the full field, and reports 98.29 percent relative illumination at the edge for Example 1, which in plain terms is almost no dimming in the corners. A genuinely tight target for a folded system this fast.
How much of that resolution the atmosphere actually lets through
Aperture and pixel count alone do not determine what a telescope can actually resolve. Three separate limits stack on top of each other, and the tightest one wins.
The diffraction limit is the theoretical floor set by the aperture itself, following the Rayleigh criterion, theta equals 1.22 times wavelength divided by aperture diameter, both in the same units. At 550 nanometers, a 35mm aperture like the DWARF 3’s works out to about 3.9 arcseconds at Rayleigh, and a 90mm aperture like the patent’s Example 1 works out to about 1.5 arcseconds. Amateur astronomy has historically used a slightly different figure for splitting double stars, the Dawes limit. Roughly 116 divided by aperture in millimeters, which gives about 3.3 arcseconds for 35mm and about 1.3 arcseconds for 90mm. Dawes and Rayleigh disagree by about 15 to 20 percent because Dawes was calibrated empirically against how well a human eye actually splits two similarly bright stars, not derived purely from the physics of a single point source, and it does not transfer cleanly to extended, faint deep-sky targets the way it is often used to.
Pixel scale is a separate number entirely, set by focal length and pixel size, not by aperture at all. The DWARF 3 samples at 2.75 arcseconds per pixel. Draco, on the patent’s numbers with an IMX585 sensor, would sample at roughly 1.76 arcseconds per pixel. Nyquist sampling theory says you need at least two pixels across the width of the smallest resolvable feature to record it without losing information to aliasing. So ideally the pixel scale should sit at half the FWHM of whatever is actually blurring the image, or finer.
Atmospheric seeing sits on top of both. From a typical backyard site, seeing FWHM commonly runs somewhere in the 2 to 4 arcsecond range on an average night, and that number has nothing to do with the telescope at all. For the DWARF 3, a 3.9 arcsecond diffraction limit and roughly 3 arcsecond typical seeing are close enough that aperture and atmosphere are fighting for the same ceiling, while its 2.75 arcsecond pixel scale sits comfortably under both, meaning the DWARF 3 is not leaving meaningful resolution on the table by pixel count alone.
For a 90mm Draco, the diffraction limit drops to around 1.3 to 1.5 arcseconds, well below typical seeing, meaning the atmosphere becomes the binding constraint on resolved detail rather than the optics, and the tighter 1.76 arcsecond pixel scale exists mostly to keep pace with what the sky allows on a good night rather than to outrun the optics. The aperture increase still helps enormously, just not primarily through sharper detail. It helps through collecting more photons per second, which is a light gathering argument rather than a resolution argument. The two are worth keeping separate.
The sensor question, in more depth
The sensor remains formally unconfirmed, but there is now enough published data on both candidates to compare them properly rather than gesture at “bigger is better.”
The DWARF 3’s Sony IMX678 is a 2 micron pixel, back-illuminated STARVIS 2 sensor with a single-exposure dynamic range cited at 83 decibels and, per Sony’s own technical materials, at least an 8 decibel dynamic range improvement over first-generation STARVIS sensors at the same pixel size. What is missing from public sources is a clean full well capacity or read noise figure in electrons, the kind of number astronomy camera vendors publish routinely for sensors they sell into that market. IMX678 is primarily marketed for security and dash cam applications, not astrophotography, and that absence of astronomy-specific characterization is itself informative.
The Sony IMX585, by contrast, has been sold directly into the astrophotography market for several years by both ZWO and QHYCCD, and both vendors publish detailed numbers. Peak quantum efficiency is cited at 91 to 92 percent. Full well capacity runs 40,000 electrons in standard mode and up to roughly 46,000 electrons in QHY’s extended linearity mode. Read noise runs as low as 0.7 to 1.0 electrons at high gain. QHY has published quantum efficiency specifically at the OIII line, 500.7 nanometers, at 91.2 percent, and at the Hα line, 656.3 nanometers, at 80.9 percent, which happen to be exactly the two wavelengths the DWARF 3’s own Duo-Band filter targets. Both sensors are STARVIS 2 generation, so neither is obsolete technology, but IMX585’s larger 2.9 micron pixels and larger overall sensor area give it a real full well and dynamic range advantage on top of the field of view math that already favored it.
None of this confirms Draco uses the IMX585. It is what the available evidence points toward, and now there is a reason beyond geometry to hope it is right: better documented low-light performance at exactly the wavelengths narrowband deep-sky imaging depends on.
What Draco would be good for
A narrower telephoto field at a finer pixel scale favors compact, high surface brightness targets over sprawling ones, and a genuinely bigger aperture adds real sensitivity on top of that framing change. Planetary nebulae like M27 and globular clusters like M13 would benefit most, since both are small enough that the DWARF 3’s wider field currently frames a lot of empty sky around them. The Crescent Nebula and the Bubble Nebula, both around 20 arcminutes across, would fill noticeably more of the frame. M51 and its companion, together spanning roughly 11 arcminutes, would sit comfortably inside Draco’s field with room to spare and show more resolved structure per pixel than they do now, which matters here specifically since M51 was the very first target this site ever imaged with a stock DWARF 3 back on August 12, 2025
Larger targets do not fare as well. M16’s full pillar and cluster region, already tight in the DWARF 3’s field after the workflow refinements built for this site, would need more deliberate framing on Draco, and something the size of Andromeda’s major axis would need a mosaic where the DWARF 3 could at least attempt one tighter frame. If the wide-angle math above holds, though, Draco’s finder camera would actually make locating and framing those smaller targets easier than it is today, even while the telephoto camera itself is doing less of the wide-field work.
Ou4 as an engineering case study
Ou4, the Squid Nebula, is not a hypothetical target for this site, and it is a useful place to check whether bigger aperture alone explains the outcome, since several other variables are involved too.
Twenty five hours of DWARF 3 Duo Band integration produced a linear FITS result of 0.49 sigma, statistically indistinguishable from no detection at all, after tearing down the Hα leakage bleeding into the OIII channel. A follow up 21.9 hour restack with per temperature dark calibration held that null result, with an upper limit around 1.4 percent of sky level. Ou4’s status in the professional literature is unsettled too. It was originally proposed as a real bipolar outflow from the massive star system at its center, but later analysis raised the possibility that it is a foreground alignment with something unrelated rather than a genuine 3D structure.
Aperture is one lever, but not the only one. Going from the DWARF 3’s 35mm aperture to the patent’s roughly 90mm effective aperture increases light gathering area by something in the range of five and a half to six and a half times, depending on how much the folding mirrors obstruct the beam, and in a purely photon limited regime that alone would cut integration time for equivalent signal to noise by a similar factor. But narrowband OIII imaging at these light levels is not purely photon limited on a sensor like the IMX678; read noise and quantum efficiency at the specific wavelength matter just as much once each sub-exposure is already faint.
If Draco does use an IMX585 class sensor, its 91.2 percent QE at the OIII line against whatever IMX678 achieves at that same wavelength, which is not publicly documented, would compound with the aperture gain rather than simply add to it, and its sub-1-electron read noise would matter more on the DWARF 3’s shorter, noisier subs than on longer ones. Faster focal ratio helps too, independent of aperture, since surface brightness per unit time scales with f-ratio for extended sources the way it does not for point sources. Thermal noise, filter bandwidth, and total integration time all still sit on top of every one of these factors and none of them individually decide the outcome.
Put together, and treating this as a first order estimate rather than a promise, a season’s worth of Ou4 integration on the DWARF 3 could plausibly compress to something closer to a single strong week of clear nights on a Draco built to the patent’s Example 1 specification. That estimate still leaves out the central obstruction’s contrast penalty and the DWARF 3’s own uncooled thermal behavior on warm nights, both of which cut against Draco in ways this post cannot quantify without real hardware in hand. It does not make Ou4 real. It changes whether trying again is worth the nights, and by how much depends on more than the aperture number alone.
Evidence that would change my mind
Held loosely, in both directions, on purpose.
Evidence that would raise confidence: a beta app teardown or leaked firmware string naming an actual sensor part number for Draco, a production photograph or teardown showing a folded optical assembly matching the patent’s mirror and lens layout, a second patent or trademark filing that explicitly names Draco or ties Xiaophoton’s filing to a specific DWARFLAB SKU, or a second independent field of view sighting from a different beta tester that matches the numbers already gathered here rather than contradicting them.
Evidence that would lower confidence: an official DWARFLAB announcement with a field of view that does not match either number in this piece, a firmware update that quietly removes the Draco profile from Atlas without a product ever shipping, credible reporting that the CN121679866A filing is a generic optics patent unconnected to any specific consumer product roadmap, or a second wide-angle sighting meaningfully different from 70.1 by 43.2 degrees, which would suggest the current figure is a placeholder rather than a real spec.
The engineering case against this hypothesis
Start with the Draco profile itself. Beta software regularly ships with test entries, competitor benchmarks used for internal calibration, or placeholder devices that never see the light of day, or in our case the light of stars. Nothing about a name appearing in a settings menu guarantees a product exists behind it. Then there is the wide-angle number, which this our math suggests does not fit cleanly with either “same camera as the DWARF 3” or a simple crop of it, meaning at least one of the two field of view figures gathered here may not describe final shipping hardware.. Or our baseline estimate for the DWARF 3’s own wide camera, built from an equivalence conversion rather than a direct spec, could simply be wrong.
The patent is the single piece of evidence doing the most work in this analysis, and patents are filed for reasons that have nothing to do with an imminent product: to block competitors from a design space, to establish prior art defensively, or to protect a technology platform that gets used across several future products rather than one specific telescope. Xiaophoton could file exactly this kind of folded optical system without DWARFLAB ever building a consumer product around Example 1 specifically. Firmware could also crop or bin a larger sensor’s output in ways that make the field of view formula misleading, since the formula assumes the full active sensor area is being used, an assumption we cannot verify from outside the device.
Weighed against all of that, what keeps the hypothesis standing is narrower than “the patent exists” or “the app shows a name.” It is that the one focal length in the patent capable of producing Draco’s telephoto field of view does so almost exactly, while the other two examples in the same document do not, and that kind of precision is hard to get from an unrelated coincidence. A vague match across a family of similarly sized lenses would not move the needle much. A specific number landing inside two hundredths of a degree does, and that is the piece of evidence this whole analysis actually rests its weight on, or better said our opinion.
Where this leaves the confidence level
Our confidence levels are editorial judgment, not statistics or facts. They describe how well each scenario explains everything gathered above, nothing more. They are lower across the board than a single clue of this analysis would justify, precisely because the wide-angle number introduced a real complication rather than a clean confirmation.
| Scenario | Confidence | Why |
|---|---|---|
| Patent Example 1 or a close variant, IMX585 sensor, roughly 90mm effective aperture, roughly 340mm telephoto focal length | Moderate to high | Telephoto field of view matches to within measurement noise, the patent names astrophotography as its use case, and the filing lands right before DWARFLAB’s own bigger-telescope teaser |
| A genuinely new, wider wide-angle or finder camera alongside the new telephoto | Moderate | The math does not support reusing the DWARF 3’s existing wide camera unchanged, and a wider finder makes engineering sense next to a narrower telephoto, but this rests on an indirect baseline estimate for the DWARF 3’s own wide field of view |
| Patent Example 2, a tuned variant of the same folded design | Low to moderate | Same architecture and manufacturer link, but a measurably worse telephoto field of view match |
| Conventional IMX678 refractor around 235mm, no folding | Low | Mathematically consistent with the telephoto field of view, reuses existing hardware, but nothing beyond that consistency supports it |
| The Draco profile is a placeholder or unrelated test build | Low, but not negligible | Common in beta software, and would invalidate most of the above regardless of how well the numbers otherwise line up |
Final thoughts
A field of view number cannot describe a whole telescope, a patent filing cannot confirm what ships under a product name, and a second field of view number can complicate a tidy story as easily as it can confirm one, which is exactly what happened here. What six pieces of evidence together can do is narrow the field from anything is possible down to a short list of specific, physically consistent explanations. Some stronger than others, with the weak points named rather than buried.
Whether or not it ships this year, or under this name, this exercise is an illustration of how much a handful of overlooked numbers in an app can give away once someone runs them through the optics. The next real spec sheet, whenever it shows up, will settle more of this in one line than everything above can.
Until then, this is what the numbers are consistent with and the wide-angle figure in particular is the one piece of this analysis that still owes readers a follow-up once more data exists.
Clear Skies!
More DWARF 3 target reports from this site
- Whirlpool Galaxy M51 with the DWARF 3: 8h 14m Across Three Sessions
- Crescent Nebula (NGC 6888) with the DWARF 3: 13 Hours on One of the Rarest Light Shows in the Sky
- One Bad Dark Frame Almost Ruined My M13 Hercules Cluster
- DWARF 3 Squid Nebula (Ou4): Why the Duo-Band Filter Can’t Detect It
- The DWARF 3 Smart Telescope — full spec page
What is DWARFLAB Draco?
Draco is an unpublished telescope profile discovered inside the DWARFLAB Atlas app, with no official announcement, product page, or spec sheet yet. Two details are confirmed from the app itself: a telephoto field of view of 1.87 by 1.05 degrees and a wide-angle field of view of 70.1 by 43.2 degrees.
How does Draco’s telephoto field of view compare to the DWARF 3?
Draco’s telephoto field of view is roughly 1.57 times narrower per axis than the DWARF 3’s 2.93 by 1.65 degrees, and covers about 41 percent of the sky area per frame.
Does Draco reuse the DWARF 3’s wide-angle camera?
Probably not unchanged. The DWARF 3’s wide lens has no officially published degree field of view, but estimates built from its stated specifications land in the 44 to 60 degree range, meaningfully narrower than Draco’s reported 70.1 degrees. If the same sensor were retained, the real focal length would need to shrink to roughly 3.8mm from the DWARF 3’s 6.7mm to explain the difference.
Is there evidence beyond the app listings?
Yes. A patent filed by Xiaophoton (Wuhan) Technology Co., Ltd, reported by Astronotrip to be DWARFLAB’s R&D subsidiary, describes a folded optical design with a 339.92mm focal length that predicts Draco’s telephoto field of view to within two hundredths of a degree. The patent does not name DWARFLAB or Draco directly, so this is strong circumstantial evidence rather than confirmation.
What sensor does Draco use?
Unknown. The telephoto field of view math fits a Sony IMX585 sensor paired with the patent’s 339.92mm design more closely than it fits the DWARF 3’s existing IMX678 sensor at any focal length tested here, and the IMX585 has publicly documented astronomy performance, including 91.2 percent quantum efficiency at the OIII line, that IMX678 lacks in public sources.
Will Draco replace the DWARF 3?
Based on the field of view math alone, unlikely. A narrower telephoto field favors compact, high surface brightness targets, while the DWARF 3’s wider field stays better suited to large, sprawling objects. The two would likely serve different jobs rather than one replacing the other.
Is any of this confirmed by DWARFLAB?
No. Everything beyond the two field of view numbers themselves is inference from optical formulas, a third party patent filing, published sensor datasheets, and DWARFLAB’s existing product history, labeled as such throughout this piece.
Leave a Reply