Adaptive LED lighting turned on-wow

[dmnc02]

Originally Posted by kraeburn

Is everybody else seeing the lights tunnel around signs when it reflects light back? That was the only situation where I saw it getting tripped up although I understand why the camera would have difficulty distinguishing between the reflecting sign and a headlight.

I am seeing the same. In fact, the Owner's Manual contains the following explicit warning about High Beam Assistant (even in the limited US version):

System limits
The system is not fully functional in situations such as the following, and driver intervention may be necessary:
▷ In very unfavorable weather conditions, such as fog or heavy precipitation.
▷ When detecting poorly-lit road users such as pedestrians, cyclists, horseback riders and wagons; when driving close to train or ship traffic; and at animal crossings.
▷ In tight curves, on hilltops or in depressions, in cross traffic or half-obscured oncoming traffic on freeways.
▷ In poorly-lit towns and cities and in the presence of highly reflective signs.
▷ At low speeds.
▷ When the windshield behind the interior rearview mirror is fogged over, dirty or covered with stickers, etc.

[Bluex]

Yes, I can also confirm a little "confused" behavior with highly reflective road signs as well.

I didn't count that as faulty behavior, however, in my previous reviews.....much like hilltops and valleys. I expected the lights to get a little tripped up by them. At the end of the day, minor stuff.....

It's great to have our $1900 lights fully working!

[oslermarines]

bluex and dmnc, thank you both for working out the kinks! I have a F80 M3 with the lighting package but no DA package ordered. When it gets here can one of you code the headlights for me? I live near philly and can drive to you guys to meet. I don't own a PC

[dmnc02]

Originally Posted by oslermarines

bluex and dmnc, thank you both for working out the kinks! I have a F80 M3 with the lighting package but no DA package ordered. When it gets here can one of you code the headlights for me? I live near philly and can drive to you guys to meet. I don't own a PC

No worries

[acudaf82]

Hello Dmnc: my name is Vic and I will be picking up my new f82 in a week or so. I have LED pkg, exec. pkg, drivers assist, I am not sure I have HBA. I can't find it on my paperwork anywhere. I am VERY interested in getting my car coded for this awesome feature. I have never done coding before and quite frankly would rather have someone who is more experienced do this. I don't want to ruin my baby ! I live in so cal. Do you know of anyone out here that I can get this done?
Does anyone know someone out here in so cal that can do this correctly?
Also, I am leasing..... Do I have to worry about that?

[dmnc02]

Originally Posted by acudaf82

Hello Dmnc: my name is Vic and I will be picking up my new f82 in a week or so. I have LED pkg, exec. pkg, drivers assist, I am not sure I have HBA. I can't find it on my paperwork anywhere. I am VERY interested in getting my car coded for this awesome feature. I have never done coding before and quite frankly would rather have someone who is more experienced do this. I don't want to ruin my baby ! I live in so cal. Do you know of anyone out here that I can get this done?
Does anyone know someone out here in so cal that can do this correctly?
Also, I am leasing..... Do I have to worry about that?

If you know your VIN, you can easily find out if you have HBA using a VIN decoder (such as http://www.bmwarchive.org/vin/bmw-vin-decoder.html): if you have it, it will show up as "S5ACA Fernlichtassistent / High-beam assistant".

Hopefully, somebody in SoCal can point you to a reliable coder if you want to go that way, but doing it yourself is really not difficult if you have a laptop (a cable is all you need to buy). Somebody has even mentioned in another thread that he will be putting out a video showing how to do this. And there might be more things down the road you might want to code.

As many other people here, I had never coded a BMW myself until I got my current car.

I do not see any way that coding your lights could negatively impact your lease (and all the changes are easily reversible).

[Bluex]

Originally Posted by oslermarines

bluex and dmnc, thank you both for working out the kinks! I have a F80 M3 with the lighting package but no DA package ordered. When it gets here can one of you code the headlights for me? I live near philly and can drive to you guys to meet. I don't own a PC

Would be glad to help, but dmnc02 is your man in this regard....if he's able to help you out. If he can't do it, i'd be more than happy to help out.

Originally Posted by acudaf82

Hello Dmnc: my name is Vic and I will be picking up my new f82 in a week or so. I have LED pkg, exec. pkg, drivers assist, I am not sure I have HBA. I can't find it on my paperwork anywhere. I am VERY interested in getting my car coded for this awesome feature. I have never done coding before and quite frankly would rather have someone who is more experienced do this. I don't want to ruin my baby ! I live in so cal. Do you know of anyone out here that I can get this done?
Does anyone know someone out here in so cal that can do this correctly?
Also, I am leasing..... Do I have to worry about that?

If you have the LED Lighting package on a US bound car, you have HBA and can have the coding done for a vast improvement. Btw, coding like this is an awful lot like taking that first dive off the high board. Terrifying at first, but then much much easier as you become more comfortable doing it. There is logic to it and as long as you have good equipment (that won't crash), good backups, and a plan to vo code and revert back to your initial settings should problems arise, it's really not hard to do at all. I almost had someone do the coding for me at first too.....but then I got impatient, and didn't feel like paying $200 for what I should be able to figure out with all the tutorials out there.

[TonySCV]

Some moderator pruning of this thread would be great (or just start a new thread with the correct info) as the first several pages of this thread contain outdated (and wrong) info.

I've gone back and edited my old posts to remove old info and instead reference dmnc02's How-to guide, but a new thread would be best.

[dmnc02]

There is one additional (and, I think, final as far as I am concerned) coding update on NGHB.

The initial procedure summarized in the "How To" in post #269 was based on a comparison between the US and ECE VO values for the LED modules (TMS and LHM), in order to identify what was needed in order to activate dynamic shadowing (aka, the "tunnel"). A comparison of the US and ECE VO values for the KAFAS camera did not show any relevant difference. I knew that a similar comparison needed to be carried on for FEM_BODY, but I wanted to get fully familiar with the operation of NGHB with the coding mentioned in post #269 before trying anything else. Since everything appeared to work as it should, I had half-forgotten of this.

So yesterday I finally compared the US and ECE VO values for FEM_BODY. This led to the identification of 4 additional parameters in FEM_BODY that tweak the horizontal and vertical offset of the high beam projectors when NGHB is active. I did a test drive last night and, while the previous coding changes had a very noticeable effect (with the elusive tunnel finally making itself clearly visible), these additional changes are not easy to detect. My preliminary impression is that they just result in slightly better center fill of the lighting pattern when the high beams are on but the tunnel is not active.

In any case, I have revised the "How To" in order to reflect these additional changes (they are referred to as "Step 3" in the last page.). If anybody notices any clear difference in the operation of NGHB after Step 3, I would be very interested in knowing.

Edit: There is an additional update to the "How To": see post #429.

[USAM3]

Originally Posted by dmnc02

There is one additional (and, I think, final as far as I am concerned) coding update on NGHB.

The initial procedure summarized in the "How To" in post #269 was based on a comparison between the US and ECE VO values for the LED modules (TMS and LHM), in order to identify what was needed in order to activate dynamic shadowing (aka, the "tunnel"). A comparison of the US and ECE VO values for the KAFAS camera did not show any relevant difference. I knew that a similar comparison needed to be carried on for FEM_BODY, but I wanted to get fully familiar with the operation of NGHB with the coding mentioned in post #269 before trying anything else. Since everything appeared to work as it should, I had half-forgotten of this.

So yesterday I finally compared the US and ECE VO values for FEM_BODY. This led to the identification of 4 additional parameters in FEM_BODY that tweak the horizontal and vertical offset of the high beam projectors when NGHB is active. I did a test drive last night and, while the previous coding changes had a very noticeable effect (with the elusive tunnel finally making itself clearly visible), these additional changes are not easy to detect. My preliminary impression is that they just result in slightly better center fill of the lighting pattern when the high beams are on but the tunnel is not active.

In any case, I have revised the "How To" in order to reflect these additional changes (they are referred to as "Step 3" in the last page.). If anybody notices any clear difference in the operation of NGHB after Step 3, I would be very interested in knowing.

It doesn't sound like it's worth coding if you can't tell the difference. I'm a little weary of coding other parameters if there's no real/noticeable gain. I'll wait for other feedback before trying this.

[dmnc02]

Originally Posted by USAM3

It doesn't sound like it's worth coding if you can't tell the difference. I'm a little weary of coding other parameters if there's no real/noticeable gain. I'll wait for other feedback before trying this.

That is a reasonable approach. I only tried it last night and the changes are definitely subtle. If my opinion changes with additional test driving, I will post an update.

[copulat3]

Originally Posted by dmnc02

That is a reasonable approach. I only tried it last night and the changes are definitely subtle. If my opinion changes with additional test driving, I will post an update.

dmnc02 - i did notice more active high beam changes/tunneling with step 3 vs without. i willl update with more night time driving.

[dmnc02]

Originally Posted by copulat3

dmnc02 - i did notice more active high beam changes/tunneling with step 3 vs without. i willl update with more night time driving.

Thanks: It is nice to read other people's impressions.

I agree that the tunnel appears to work a bit better after Step 3, although I am having a hard time identifying exactly in what way. I feel that there is perhaps a bit more variability both in the position and in the width of the tunnel, although it is difficult to be sure without a camera. This might be what you meant when you mentioned that the tunnel appears "more active".

In any case, tonight I undid Step 3 and went for a drive. This allowed me to confirm the initial impression, mentioned in post #317, that there is "slightly better center fill of the lighting pattern when the high beams are on but the tunnel is not active.”

The issue is clearly shown in the video that USAM3 posted earlier after coding Steps 1 and 2 (reproduced below). In his video, the system switches the high beams on at about 0:11 and then activates the tunnel at 0:18 (tricked by the street lights). Just before that, he goes by some trees. If you look at the light beam projected on the trees, you can see that the center area is darker (watch it slowly: this is especially obvious around 0:15). You can see the same darker area when he is turning onto the freeway with the high beams on between 1:06 and 1:12. This is something I also noticed on my car after Step 2, but it is gone after Step 3: the high beams appear to point slightly lower and more inward, providing better center fill. To me, this remains the most obvious difference and by itself makes Step 3 worthwhile.

[TonySCV]

Originally Posted by dmnc02

The issue is clearly shown in the video that USAM3 posted earlier after coding Steps 1 and 2 (reproduced below). In his video, the system switches the high beams on at about 0:11 and then activates the tunnel at 0:18 (tricked by the street lights). Just before that, he goes by some trees. If you look at the light beam projected on the trees, you can see that the center area is darker (watch it slowly: this is especially obvious around 0:15). You can see the same darker area when he is turning onto the freeway with the high beams on between 1:06 and 1:12. This is something I also noticed on my car after Step 2, but it is gone after Step 3: the high beams appear to point slightly lower and more inward, providing better center fill. To me, this remains the most obvious difference and by itself makes Step 3 worthwhile.

I've seen exactly what you're talking about. I've often wished BMW had a way to manually adjust the left headlight toward the right and the right headlight toward the left. There is a bit of a vertical "dead zone" between the two lights when the high beams are on. If this solves it, that would be terrific.

I'll give your updated coding a try tonight.

You may want to start adding a version # and date stamp to your How-to guide, as people need to know v2 of the guide is more up to date.

- T

[dmnc02]

Originally Posted by TonySCV

I've seen exactly what you're talking about. I've often wished BMW had a way to manually adjust the left headlight toward the right and the right headlight toward the left. There is a bit of a vertical "dead zone" between the two lights when the high beams are on. If this solves it, that would be terrific.

I'll give your updated coding a try tonight.

You may want to start adding a version # and date stamp to your How-to guide, as people need to know v2 of the guide is more up to date.

- T

Please do post your impressions after giving Step 3 a try. And I did put a revision date in the "How To"

[USAM3]

I went ahead and coded step #3 in your PDF how-to and I noticed that the auto high beams appear, best way I can describe them, fuller. Previously, when the auto high beams were engaged there were times when, although the high beam indicator were on, you couldn't really tell they were from the degree of illumination in certain scenarios. With step #3 coded, I noticed the high beam is much fuller and prominent on the right-hand side of the road when trailing behind a car. When there is no oncoming or preceding cars the beam is more concentrated directly in front of me and seems to extend further forward. I'll try to post a video when I have time.

[TonySCV]

Originally Posted by dmnc02

Please do post your impressions after giving Step 3 a try. And I did put a revision date in the "How To"

Definitely lessened the "gap" between the left and right high beams, which I appreciate because it was quite noticeable before. Now it's subtle enough where I really have to be looking for it.

On the downside, the high beam when split (boxing out cars) seems much brighter and more pronounced, and I'm not sure I like that. It seems less nuanced and more blatant, for lack of a better explanation.

It'll be interesting to fiddle with those settings a bit more.

[dmnc02]

Originally Posted by USAM3

I went ahead and coded step #3 in your PDF how-to and I noticed that the high beams appear, best way I can describe them, fuller. Previously, when the auto high beams were engaged there were times when, although the high beam indicator were on, you couldn't really tell they were from the degree of illumination in certain scenarios. With step #3 coded, I noticed the high beam is much fuller and prominent on the right-hand side of the road when trailing behind a car. When there is no oncoming or preceding cars the beam is more concentrated directly in front of me and seems to extend further forward. I'll try to post a video when I have time.

Originally Posted by TonySCV

Definitely lessened the "gap" between the left and right high beams, which I appreciate because it was quite noticeable before. Now it's subtle enough where I really have to be looking for it.

On the downside, the high beam when split (boxing out cars) seems much brighter and more pronounced, and I'm not sure I like that. It seems less nuanced and more blatant, for lack of a better explanation.

It'll be interesting to fiddle with those settings a bit more.

Thank you both for the feedback.

Personally, I do not get the impression of the tunnel working any worse after Step 3 (if at all I feel the opposite). As mentioned earlier (post #321), I do feel that there is a bit more flexibility in the position and width of the tunnel after Step 3, possibly resulting in better illumination on the sides of the tunnel. Hopefully, we will get another video from USAM3 at some point, so that we can more clearly see it in action.

In any case, just to be clear about the process, the "How To" (including Step 3) is not the result of my fiddling with the parameters in any way, but instead of making sure that all the parameters relevant for NGHB are set exactly as they would be set at the factory on a similar car sold on the ECE market. Hence, these are settings chosen and optimized by BMW, not by myself.

Admittedly, there is a possibility that BMW might have designed a different reflector for the US high beams on account of the fact that it was not supposed to support NGHB, in which case matching the ECE settings is not necessarily the best thing to do. However, at this point I strongly doubt that this is the case, since they appear to have used exactly the same LED array on US and ECE cars.

[SaMaster14]

Just got this coded on.

One question, do I need to push the 'A' button on the turn signal whenever I get in the car at night to activate it, or will it go on automatically?

Thanks!

[dmnc02]

Originally Posted by SaMaster14

Just got this coded on.

One question, do I need to push the 'A' button on the turn signal whenever I get in the car at night to activate it, or will it go on automatically?

Thanks!

Unfortunately there seems to be no way to get HBA to switch on automatically. You have to use the button on the stalk.

[SaMaster14]

Originally Posted by dmnc02

Unfortunately there seems to be no way to get HBA to switch on automatically. You have to use the button on the stalk.

That's no problem. I just wanted to make sure.

[CrazyT]

I coded using step #3 and have mixed feelings initially. I've only taken it on a short 8 mile drive so far, so experience is limited.

1) It definitely got rid of the "blind spot" right ahead of the car with full high beams. A big plus.
2) It seemed to affect the size of the tunnel. At step #2, the tunnel was centered on the vehicle ahead with a little margin on each side. Now the tunnel seems wider and shifted more to the right (as in the vehicle ahead is not centered in the tunnel). This also seems to move the tunnel for on coming vehicles left a little too late.
3) Seems more twitchy. It was adjusting on signs or lights it hasn't done before.
4) It always felt with step 1 and step2 that the left highbeam would turn off too late on on-coming cars. Now it's turning off a little earlier, which is a good thing.

I'm still uncertain on #2 and #3. I'll have to drive the car more to figure it out better. Could just be me, my dirty windshield, or the environmental conditions. Anyway, thanks again dmnc02 for figuring this all out.