ASI585MC Gain Settings
How HCG mode works—and the gain strategy I'm testing for the Bortle-9 Index
When I started imaging, I thought gain was like a volume knob—higher meant louder signal, so why not crank it? But gain on a modern CMOS sensor is more nuanced than that. The ZWO ASI585MC has a trick up its sleeve that changes how to think about this setting.
A note on where I am: I'm currently working through my Bortle-9 Index—a 25-target experiment to see what's actually possible from extreme light pollution. So far I've imaged three targets with the ASI585MC (Veil Nebula, Pleiades, M33), all at gain 200. The settings in this article represent my planned approach for the remaining 22 targets, not proven results. I'll update this as I collect more data.
The Magic Number: 200
The ASI585MC uses Sony's IMX585 sensor with a feature called High Conversion Gain (HCG) mode. When you set the gain to 200 or higher, HCG automatically engages. The result: read noise drops to around 0.7 electrons while dynamic range reaches approximately 12 bits—roughly the same as at gain 0.
This is unusual. On most cameras, higher gain means lower dynamic range. But HCG mode gives you the best of both worlds—if you know where the threshold is.
For the ASI585MC, that threshold is gain 200.
Note: ZWO originally specified gain 252 as the HCG threshold, but updated this to gain 200 in late 2025. Make sure your ASIAIR app is version 2.4.2 or later (or ASIStudio 1.61.1+) to take advantage of this change.
Why Not Just Use 200 for Everything?
You could. Many imagers shoot at the HCG threshold exclusively, and it's a safe default. That's what I've done for my first three targets. But the theory suggests there are reasons to push higher for narrowband work:
- Dual-band filters block most light pollution. The sky background stays dark even at higher gain, so you can push sensitivity without saturating.
- Faint emission signal benefits from extra sensitivity. When the target barely rises above the noise floor, a bit more gain can help.
- You're trading dynamic range you don't need. A dim planetary nebula doesn't have bright highlights to protect—you can afford to sacrifice some DR for better faint-signal performance.
Conversely, broadband targets (galaxies, reflection nebulae) should benefit from staying at 200. Bright galaxy cores, colorful star fields, and subtle dust lanes all need that dynamic range to avoid clipping. My M33 and Pleiades results at gain 200 support this—both came out well.
The Settings I'm Testing
Based on my equipment (Askar FRA300 Pro, ZWO ASI585MC, ASIAIR) and one-hour workflow from Bortle 9, here's the gain strategy I plan to test across the Bortle-9 Index:
| Scenario | Filter | Gain | Sub Length | Histogram | Why |
|---|---|---|---|---|---|
| Bright emission nebula M42, NGC 7000, Lagoon |
SV220 | 250 | 30–45s | 25–30% | High surface brightness; shorter subs prevent core blowout |
| Standard emission nebula Rosette, Veil, Horsehead |
SV220 | 250 | 45–60s | 25–30% | Best faint-signal performance for Ha/OIII |
| Faint emission / Tier 1 Barnard's Loop, Spaghetti |
SV220 | 300 | 60s | 30–35% | Pushes sensitivity for ultra-faint targets |
| Bright galaxy M31, M81/82 |
L-QEF | 200 | 30–45s | 20–25% | HCG mode; preserves core dynamic range |
| Dim galaxy / low SB M33, M74, M101 |
L-QEF | 200 | 45–60s | 20–25% | Max sensitivity while keeping DR for faint arms |
| Reflection nebula M45 Pleiades |
L-QEF | 200 | 30–45s | 20–25% | Broadband needed; moon-sensitive |
| Mixed emission M1 Crab Nebula |
SV220 or L-QEF | 250 / 200 | 45–60s | 25–30% | Experiment with both; compare results |
| Planetary nebula M27, M97 |
SV220 | 250 | 45–60s | 25–30% | OIII-dominant; dual-band essential |
Quick Reference
| Gain | When to Use | Read Noise | Dynamic Range |
|---|---|---|---|
| 200 | Broadband / L-QEF targets (galaxies, reflection nebulae) | ~0.7e | ~12 bit |
| 250 | Dual-band emission targets (most nebulae) | ~0.7e | ~11 bit |
| 300 | Ultra-faint Tier 1 targets only | ~0.7e | ~10.5 bit |
The Logic (My Hypothesis)
The pattern I'm testing is simple:
- Filter drives gain. Dual-band (SV220) → higher gain is safe because light pollution is blocked. Broadband (L-QEF) → stay at 200 to preserve dynamic range for gradients and bright areas.
- Target brightness drives sub length. Bright cores (M42, M31) → shorter subs to avoid blowout. Faint or low surface brightness → push to 60s to gather signal.
- Histogram target varies by filter. Lower for broadband (20–25%) to handle LP gradients. Higher for dual-band (25–35%) because the filter does the heavy lifting.
A Note on Darks
Different gain settings require matching dark frames. If you switch between 200 and 250, you'll need separate dark libraries for each. The ASI585MC's near-zero amp glow makes this less painful than it used to be, but it's still worth planning your sessions to minimize gain changes.
My planned approach: batch targets by filter type. Galaxy nights will be all L-QEF at gain 200. Nebula nights will be all SV220 at gain 250. One set of darks per session.
The Bottom Line
Gain isn't just a number to set and forget. On the ASI585MC, understanding the HCG threshold at 200 and knowing when to push higher (or stay put) can make a real difference in your final images—especially from light-polluted skies where every photon counts.
These are the settings I'll be testing throughout my Bortle-9 Index project. The logic is sound, but the proof will be in the images. I'll report back as I work through the remaining 22 targets.
The principles should translate regardless: match your gain to your filter, your filter to your target, and your expectations to your sky.
Clear skies,
Pete
See also — Dithering and Binning for One-Hour Sessions