Feel free to help me, Davehibes. I don't have an issue with installing the latest version. Dead easy, and I was merely asking the author about the helper file.

But maybe you know something about the multi-monitor observation. If so, please share. That's what this thread is all about.

Duped. Ignore.

Edited April 28Apr 28 by Brian Mackie

6 hours ago, Brian Mackie said:

For the avoidance of doubt, is the Helper exe required to reinstall?

Yes, always re-install the lua script and helper with each new release.

6 hours ago, Brian Mackie said:

One curious thing I observe is that when enabling two smaller monitors for pop-out G1000 displays, when I go to FWL Macros>AutoHaze Settings, there is no AutoHaze settings box appearing on any of the monitors.

Ah, I only have one monitor so didn't come across this.  Maybe the AutoHaze window is opening but off screen.  I'll have a test version for you shortly to see if I can fix this.

Thanks, Davehibes. Our author came back and clarified that existing users should install both files (helper and lua) when they upgrade, so the question was useful. The other issue - multi-screen related - is being studied by MrBitstflyer, too.

The number of pages in this thread illustrates 1) the level of users' interest in AutoHaze and 2) the author's unusual focus on making it even better for us. I'm hugely impressed. What he has achieved will, I hope, be noticed by the folks at Laminar Research.

8 hours ago, Brian Mackie said:

Thanks, Davehibes. Our author came back and clarified that existing users should install both files (helper and lua) when they upgrade, so the question was useful. The other issue - multi-screen related - is being studied by MrBitstflyer, too.

The number of pages in this thread illustrates 1) the level of users' interest in AutoHaze and 2) the author's unusual focus on making it even better for us. I'm hugely impressed. What he has achieved will, I hope, be noticed by the folks at Laminar Research.

 

A few of us have (including @Maya2 which I'm sure you've seen earlier), however there's a few problems here

a) There's a few incorrect assumptions about how X-Plane's weather engine actually works
b) Some of the outputs seem arbitrary?

c) (Me personally) still see folks claiming there should be no haze, but not actually providing scientific examples for it. The feedback I get from users seems more qualitative than quantitative

Being candid, I also believe there may be some AI involved here (with all due respect). Which is going to feed back in points A & B here.

All I'm going to say is, if the script makes you happy, great! But there's quite a few queries here that doesn't make this viable for us atm.

2 hours ago, DeltaWho said:

 

A few of us have (including @Maya2 which I'm sure you've seen earlier), however there's a few problems here

a) There's a few incorrect assumptions about how X-Plane's weather engine actually works
b) Some of the outputs seem arbitrary?

c) (Me personally) still see folks claiming there should be no haze, but not actually providing scientific examples for it. The feedback I get from users seems more qualitative than quantitative

Being candid, I also believe there may be some AI involved here (with all due respect). Which is going to feed back in points A & B here.

All I'm going to say is, if the script makes you happy, great! But there's quite a few queries here that doesn't make this viable for us atm.

Thank you for the candid feedback. I'm happy to address each point directly.

On incorrect assumptions about X-Plane's weather engine: I'd genuinely welcome specifics here. AutoHaze writes exclusively to sim/private/controls/scattering/override_turbidity_t. If there are aspects of how this interacts with X-Plane's rendering pipeline that I've misunderstood, I'd be very grateful for the correction.

On arbitrary outputs: The turbidity values are derived from a documented physical chain: CAMS satellite AOD at 550nm → Koschmieder extinction formula (vis = 3.9/AOD) → Linke turbidity scaling (turb = 1.0 + AOD/0.1 × 0.5) → boundary layer altitude correction via Open-Meteo BLH. These are the same equations used in atmospheric optics research. The precipitation effect uses an exponential curve matching published rainfall-visibility relationships. Nothing is arbitrary — every value traces back to a physical measurement or established formula.

On qualitative vs quantitative feedback: Agreed — and that's exactly why AutoHaze uses CAMS satellite data rather than subjective judgement. The Copernicus Atmosphere Monitoring Service provides validated hourly aerosol measurements used by ECMWF, national met services and the aviation industry. Flying over the Indo-Gangetic Plain with AOD 1.65 producing 2km visibility isn't a subjective choice — it's what the physics says.

On AI involvement: Correct — I use Claude as a development assistant, which I'm entirely transparent about. The physics, the data sources, the validation against real conditions are mine (and the grateful help given here on the forum). The AI accelerates implementation but the science is independently verifiable.

The weather engine in X-Plane 12 is spectacular, which is why I moved over from MSFS 🙂, but there were just a couple of things I thought could be improved - climbing out of and descending into the boundary layer to better simulate haze, and Improved rain visibility effects where the visibility will increase and decrease based on rainfall rate hitting the aircraft.

1,700+ users downloading it suggests the community finds real value in the enhancements I have made.

 

 

 

Version 2.2 is available with much better rain effects and increased haze levels for the areas in the 'red' band on the CAMS charts.

Thank you for the superb work !

Your method looks perfectly "scientific", well founded and sound to me. Thanks for the above post where you take the time to - again - explain the approach/model you're using. Not that different from work I've seen at the metoffice I work at since more than 31 yrs, and also has as main feed ECMWF...

Using AI to code? What's the big deal? Used a lot, with all sorts of advantages, provided one knows / understands what is being generated... Saves a LOT of useless time, and minimizes code errors when you're in a hurry...

On a side note and even if not, I guess, with your level of detail, MSFS uses revolutionary approaches to visibility based on aerossol data, as well as other aspects of weather modelling like convection based on an interesting "heat map" / albedo /... model, even though their thermals leave a lot to be desired in terms of their structure, compared to RL thermals...

What is not acceptable is the way most scenes look like when I load Xp12 in a crystal clear day and have that unrealistic reduced visibility that is truncated by the way Xp appears to incorporates METAR data into it's weather rendering, not to mention the lack of a proper visual representation of the boundary layer / inversion layer visual effects. The Grib/Grib2(?) data they gather doesn't appear to be used for that matter although it can have, depending on the source model, parameters like those for aerosols / atmospheric chemical constituents.

With your plugin you are at least introducing a "new discipline" in the modeling of weather parameters important for a better simulation of meteorological visibility effects.

Edited April 30Apr 30 by jcomm

1 hour ago, jcomm said:

Thank you for the superb work !

Your method looks perfectly "scientific", well founded and sound to me. Thanks for the above post where you take the time to - again - explain the approach/model you're using. Not that different from work I've seen at the metoffice I work at since more than 31 yrs, and also has as main feed ECMWF...

Thank you so much for your incredibly kind words, and what a validation coming from someone with 31 years at the Met Office working with ECMWF data. That means a great deal.  You've understood exactly what AutoHaze set out to do. My goals were straightforward and modest:

• To better simulate the atmospheric boundary layer, keeping haze consistent from the surface up to the BLH top, then clearing realistically into cleaner air above it — something X-Plane doesn't currently represent.
• To provide more physically accurate visibility using real satellite aerosol data from CAMS, rather than relying on METAR reports that are capped at 9999m regardless of actual conditions.
• To simulate the dramatically reduced visibility when flying into heavy rain, and the equally rapid clearing when flying out of it.

The science behind the values written there such as Koschmieder's extinction formula, Linke turbidity scaling, CAMS AOD at 550nm, Open-Meteo boundary layer height, is the same physics used by professional meteorological services.

Laminar have always said they would only consider suggestions backed by science. I have genuinely tried to do that, so I was, I'll admit, a little surprised by the response. I have enormous respect for what they have built in X-Plane 12, it is a remarkable simulator, and I mean no criticism of the team. Perhaps in time the approach AutoHaze demonstrates will find its way into the simulator itself in a supported form. That would be the ideal outcome for everyone.

On the subject of AI involvement, yes, I use Claude as my coding assistant, and I'm entirely open about that.

My background is as a software engineer, but that was over 20 years ago before redundancy took me in a different direction. I'm in my mid-60s now, and while the logical thinking is still there, the idea of getting back up to speed with modern Lua, Python, async architecture and X-Plane's SDK from scratch felt genuinely daunting. I simply wouldn't have attempted this project without that assistance.

What Claude has allowed me to do is concentrate entirely on what I do have, the research, the atmospheric physics, the data sources, the ideas about how boundary layer behaviour should actually look in a simulator. I describe what I want physically and scientifically, and Claude translates that into working code. Every decision about which formula to use, which data source to trust, what the boundary layer transition should feel like, what precipitation visibility should look like in a thunderstorm — those are mine.

I think that's actually a healthy model for this kind of project. The physics doesn't care whether the code implementing it was written by a 25-year-old developer or assembled with AI assistance by someone in their 60s who used to write software two decades ago. What matters is whether the output is correct, and a Met Office professional with 31 years of experience suggesting the methodology is sound is a rather better validation than the tools used to implement it.

AutoHaze wouldn't exist without Claude. But it equally wouldn't exist without the research, the testing, the user feedback, help on this forum and the persistence to get the atmospheric physics right. Both parts were necessary.

Thank you again.

I'm not a meteorologist - I'm an IT guy 🙂 with a background in applied math and AI, working at the metoffice... But! as I am directly involved in the support area of Aviation Weather Forecast (and not only), I am a glider pilot, I really love meteorology and climatology, and the amazing area of numerical forecast, and I am happy to be surrounded by meteorologists, and numerical forecast research colleagues, I can satisfy my "curiosity" and learn from them a worth of good information I used to use for flight simulation, long ago, cooperating with some of the early weather injection applications, and information I use and abuse from everytime I plan a glider task...

I am also very picky with the way desktop flight simulators I use, and while I feel confortable with the way X-Plane models weather, I surely want more, specially if that "more" can be achieved within the limits of efficiency and cost...

I have been following and trying your move to X-Plane 12, and recently this great plugin you're dynamically supporting and offering for free to the XP user community, which I find remarkable!

Keep the good work !!! 

 

Edited April 30Apr 30 by jcomm

On 4/29/2026 at 5:41 PM, MrBitstFlyer said:

Version 2.2 is available with much better rain effects and increased haze levels for the areas in the 'red' band on the CAMS charts.

Hi, i have just downloaded this version but i am getting a Windows script host error on launch: Windows Script Host
Script:
C:X-Plane
12\Resources\plugins\FlyWithLuaScripts\AutoHaze-helper.vb
Line:1
Char:1
Error:The operation was canceled by the user.
Code: 800704C7
Source:：(null)

Any ideas on the way forward. 

Thanks in advance.

same! I just wanted to write it here. I'm having the same error-message...

3 hours ago, Davehibes said:

Hi, i have just downloaded this version but i am getting a Windows script host error on launch: Windows Script Host
Script:
C:X-Plane
12\Resources\plugins\FlyWithLuaScripts\AutoHaze-helper.vb
Line:1
Char:1
Error:The operation was canceled by the user.
Code: 800704C7
Source:：(null)

Could you check...

1..  Right click AHHelper.exe/Properties/Unblock/OK
2..  Launch XP12 as administrator for the first launch only
3..  Add the Scripts folder to your anti-virus exclusions.

Let me know if that fixes it because I do not see this.

I'm not sure if I am breaking forum rules by answering tech issues here, maybe better if these queries are on the AutoHaze page?

6 hours ago, MrBitstFlyer said:

1..  Right click AHHelper.exe/Properties/Unblock/OK

Thank you for your patience and assistance. The "unblock" fix worked, and all is perfect again! Next time i will read the manual first, instead of trying to fault find on the forum. Your product is absolutely paywear quality...for the first time in a year of flying XP12, i can actually see the beautiful scenery that XP has to offer! Many thanks.