New 1920s Technology... Today!

On Friday, April 16, 2021 at 8:57:26 AM UTC-7, sms wrote:
On 4/15/2021 5:52 PM, Jeff Liebermann wrote:

snip
The problem is that most of the lumens claimed by headlight
manufacturers and vendors are little better than science fiction.
That is true for most of the no-name lights from Amazon, eBay, or
Aliexpress. But there have been independent tests of some of the more
reputable manufacturers' lights that show pretty accurate lumen claims,
at least when the battery is fully charged. For example, the Magicshine
Monteer 8000s claims 8000 lumens and it was measured at 7800 lumens
https://www.mtbr.com/threads/magicshine-monteer-8000s-bike-light-review.1173672/.

Look at the graphic at
https://www.mtbr.com/attachments/rn1200-ranger-1400-allty1500-jpg.1368263/
where they test three lights at maximum power and lower power. The
Magicshine RN1200 actually exceeded its lumen rating for the first 80
minutes.

Some manufacturers take the rating of each LED, add them up, and claim a
theoretical number of lumens. If you had active nitrogen cooling you
might reach the theoretical number, but you'd never reach it in normal use.

Here's someone that tested some of the Chinese lights
https://youtu.be/m1vFMbSEkdQ?t=957. Not all of them are super-cheap,
some cost $100 or so. Two of the better ones he tests are the Towild
Professional 1800 ($65)
https://www.wild-fires.com/bike-front-light-br1800/ and the Gaciron
V9D-1800 ($100, which I have)
https://www.gaciron.com/V9D-1800bule.php. The Towild is one of the few
LED lights I've seen that have two independent beams, spot and flood.
You used to see this long ago with MR11 and MR16 lights where they'd
have two separate beams for spot and flood.


From being on the inside of making and testing lights, one dirty secret is that the bike headlight "standard" only requires claimed lumen output for the first 30 seconds after being turned on.
The very small mass of heat bearing material in current small single LED lights saturates quickly. Light and Motion was a driving force behind that bike light headlight standard, and their lights are a noticeable offender.
The calibration, metering equipment, integrating sphere size, and mounting of a light makes a huge difference in measured lumens.
Another filthy secret is how MTBR and the related Francis Cebedo websites doctor their photography to make a sponsored product appear brighter in a photographs. Their reviews shouldn't be trusted as far as you can spit.
Before Francis was able to take them down, I downloaded pics he posted where the brighter looking Taz, had metadata showing that the picture was taking later than the picture of my Oculus (first generation 1500), and more exposure than the picture he took of the Oculus beam. But the picture of the Taz' beam looked abnormally brighter, with the blue night sky noticeably lighter blue than the beam picture from my Oculus which was taken earlier. He also aimed my beam downward more to the ground, resulting in smaller and uneven coverage than the beam picture of the Taz.
Still have the pictures with the metadata showing the time and exposure, if anyone wants to see how money and influence have as much effect on what a light gets tested and reviews like vs what a light's true output and beam pattern is.

On 4/16/2021 3:27 PM, Oculus Lights wrote:

snip

From being on the inside of making and testing lights, one dirty secret is that the bike headlight "standard" only requires claimed lumen output for the first 30 seconds after being turned on.

Yes, that's why the test results at
https://www.mtbr.com/attachments/rn1200-ranger-1400-allty1500-jpg.1368263/
are useful. They were done independently, and over a period of hours,
not seconds.