Coding Anti-Dazzle and Variable Light Distribution

BMW i3 Forum

Help Support BMW i3 Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
The only adaptation the "adaptive" low-beam LED headlights on our former 2014 U.S. i3 performed was panning vertically to adjust to different loads being carried. I had assumed that the adaptive LED headlights on our 2019 U.S. i3 would do the same but I wasn't aware that they would pan vertically when switching between low and high beams.

Thanks for staring directly into the bright headlights to try to see the segments. Who knew that a welding mask could be so handy? Maybe it would work to watch the solar eclipse in April.
 
‘06 330i with sport pac factory adaptive headlights swivel R & L with wheel turns & up & down with road elev & depress. Also big load adjusts tilt up & down. $960 repl cost post accident.
 
Could I trouble a driver of a European i3 with Tech Package (and therefore ECE headlights) to try to replicate the photos below and post them to this thread?

1low_beam.jpeg

2high_beam.jpeg

Above is my i3's light footprint with low/dip beam (top photo) and with high/main beam — both shot with an iPhone 15 Pro:
• With the front of the headlights 15-20 feet away from the white garage door
• With the headlight switch set to auto
• With High-Beam Assist disabled
• With the headlight stalk on the left pushed forward to manually engage high/main beam lighting
• In landscape mode
• With the phone resting atop the steering wheel
• On a level road

Mine is a 2020 North American i3 s with the Tech Package and other options, including:
• 5AP Decoding Anti-Glare High-Beam Assistant
• 5A4 LED Headlight With Extended Scopes
• 5AC High-Beam Assistant
• 5AT Driving Assistant Plus

Before taking the photos, I modified the Body Domain Controller (BDC) to:
Code:
Enable Enable Variable Light Distribution (VLD)
    3073 LaMaster1->C_AFS_ENA: set to I001_enable (0x01)
    3073 LaMaster1->LUT_FLC_FORWARDLIGHTING_Y: set to I001_mit_AFS (0x9C 0x9C 0x9C)

Enable Glare-Free High Beams (GFHB)
    3073 LaMaster1->C_HBA_ENA: set to I001_enable (0x01)

I suspect the headlight modules of North American i3 models lack the necessary hardware — perhaps optic shutters — that allow matrix lighting as conceived by BMW. If one of our European friends could help, we might see a black box "tunnel" in a similar high/main headlight photo of a 2019+ i3 with Tech Package.
 
As SSi3 says - the headlight system will be seeing nothing to react to in the conditions shown in that photo. Auto headlights (Euro spec) should have all of the main beam elements lit in those conditions.

With the headlight stalk on the left pushed forward to manually engage high/main beam lighting

I confess I've never checked this, but doesn't this completely override auto/adaptive headlights?

The problem is that you can't really test these things in an artificial situation, because the camera system reacts to many different triggers - so (for example) on my Euro-spec car, if it sees street lighting, the system will turn off main (high) beams altogether. It's quite a sophisticated system.

I'm pretty sure there are no optical shutters. I've compared the i3 system with the older systems used by (e.g.) Mercedes Benz - which does have mechanical shutters - and it doesn't look anything like the same from the driver's seat. I'm pretty sure that all elements are switched electrically.
 
I would have done this already, but my driveway is completely different (heavily sloping) and I have a street light next to it which never turns off (I don't complain to my local authority because it makes the front of my house more secure). So my results probably won't look like your results, even if the headlights work in exactly the same way.
 
I would have done this already, but my driveway is completely different (heavily sloping) and I have a street light next to it which never turns off (I don't complain to my local authority because it makes the front of my house more secure). So my results probably won't look like your results, even if the headlights work in exactly the same way.
OK, thanks much. I'll cast a wider net ;)
 
I think the video that you post is of a completely different BMW system. You can clearly see the headlamps panning left and right in the video - a Euro-spec i3 headlight doesn't do that (or at least, my Euro-spec adaptive lights don't do that).

The "tunnel" is (I think) simply there because the cutoff at the inner edge of the main beam is quite sharp. If the person in that video had reversed back a little further, the two would have joined up and you would see no "tunnel".
 
Clearly someone has activated a service diagnostic mode to make the lights go through their movements... the average owner can't do this or replicate it.
Oh - I didn't listen the commentary so I didn't know that - I just assumed that this was the standard start-up sequence for those specific lights!
 
How do you know this is a service diagnostic mode? It's news to me!
I don’t know it for sure, but I’ve seen this sort of thing before with automotive systems… and that guy with the YT channel and the 6-series is clearly very proud of his anti-dazzle lights and other coding/modding.
 
These are the same adaptive matrix lights as the i3... Martin describes exactly how they function, and this is exactly my experience with them after coding (his weather conditions were not great for really using these lights):
 
The other night, on a divided 4 lane highway, my newly coded headlights stayed on high beam (blue light on) the entire trip. I could clearly see the left light dip as a car came the other way. So the i3 lights acted similar to the Mini lights. You could clearly see segments turn off. No one flashed their lights at me.

As I slowed for my turn onto a two lane country road, the lights dipped. So maybe they a speed dependent also. Then the lights started to play games like they didn't know what to do. Sometimes when a car approached, they dipped. Sometimes they stayed on high beam and the segment darkened.

There is a whole lot more that the controller or computer for these headlights are doing than I have figured out yet. But ever since coding, they sure are much nicer, and I like them better than the older automatic setup.
 
So maybe they a speed dependent also. Then the lights started to play games like they didn't know what to do. Sometimes when a car approached, they dipped. Sometimes they stayed on high beam and the segment darkened.
Is it possible that your speed was around 35MPH? The owner's manual for my 2021 car says "the system does not switch on high beams at low speed" - but the manual itself doesn't say exactly what speed that refers to. Others have suggested the threshold is around 35MPH.

I know that mine does similar things to yours (i.e. the behaviour changes) when I get into a built-up area but I haven't worked out whether it is speed, or ambient light levels from street lamps, or both (or something else!).
 
These are the same adaptive matrix lights as the i3... Martin describes exactly how they function, and this is exactly my experience with them after coding (his weather conditions were not great for really using these lights):

I agree that Martin's experience was similar to my experience after activating the adaptive headlights on our U.S. i3.

Thanks for posting this video which shows how these lights are supposed to work. I now feel confident that the coding changes that I made fully activated the adaptive features which aren't as advanced as some headlights.
 
I agree that Martin's experience was similar to my experience after activating the adaptive headlights on our U.S. i3.

Thanks for posting this video which shows how these lights are supposed to work. I now feel confident that the coding changes that I made fully activated the adaptive features which aren't as advanced as some headlights.
Alooha Art; Could you try and explain, in detail, what coding steps you performed in order to enable this feature? I have only done a couple of basic coding with my 2019 i3 (hold state of charge, passenger side mirror tilt, etc.), I'd like to do this but don't want to go down the "rabbit hole" and find myself screwing things up.

Thanks,

John
 
Back
Top