It would be hugely expensive for the home hobbyis to accomplish the professional levels of sophistication that the commercial units bring. It's fairly easy to slap together a high lumen emitter. It's a bit tougher to focus that light pattern to something useful. It's a bit tougher still to give a perfect spill for use on a bicycle. It's darned near impossible to offer it in a package that can be mounted easily to just about any bike, aimed easily, doesn't come apart under vibration and abuse and severe environmental conditions (how many DIY systems would survive long term riding in salt fog from ocean spray?), helmet mounts, flexible cabling with connectors that don't come apart, modular so you can easily replace certain parts of the system, LEDs to tell you when your batteries are nearing end of life, brighness and flash settings, etc. Then consider a design that can be truely mistreated (left plugged in for days, short circuited, modded, run with worn bare or frayed wires) without burning down people's houses, electricuting anyone, etc.

Commercial products cary a price but you get a TON of sophistication for that money.

I have a feeling it's going to take more than 2 years for prices to become affordable at the lumens we want.

Look at the Fenix flashlights, they are a prime example of what I'm talking about. If they come out with 300 or 400 lumen light it would be wonderful. And relatively cheap too.

There's that catch again... Everything can always be brighter, however at the expense of run time in a constrained design :lol:

I think we need to amend that Bontrager quote for now.

Runtime, bright, cheap, (?)...... Pick any two

Nice...but w/ lights it kinda works.... you need to add design to the mix...

I can get 600L for less than 400 bucks with ok run time...but the design is not as simple...

For Example: for 300 bucks or so I can get 700 lum with 4.4 hours of run time, but I would have to get two systems: http://aebike.com/page.cfm?PageID=30&action=details&sku=LT0521

But, I would have a cluttered bar, no helmet mount, and the beam pattern my not be what I need.

I could also get a 200L set up for about 60 bucks or so (see example http://www.dealextreme.com/details.dx/sku.3786) but then there is mounting and beam pattern issues...

The ? of cost to make 600L LED lights is a bad question, at least to me...value or $$ that I am willing to spend is a better question and the answer seems to be driven by specificity of design...

It would be hugely expensive for the home hobbyis to accomplish the professional levels of sophistication that the commercial units bring. It's fairly easy to slap together a high lumen emitter. It's a bit tougher to focus that light pattern to something useful. It's a bit tougher still to give a perfect spill for use on a bicycle. It's darned near impossible to offer it in a package that can be mounted easily to just about any bike, aimed easily, doesn't come apart under vibration and abuse and severe environmental conditions (how many DIY systems would survive long term riding in salt fog from ocean spray?), helmet mounts, flexible cabling with connectors that don't come apart, modular so you can easily replace certain parts of the system, LEDs to tell you when your batteries are nearing end of life, brighness and flash settings, etc. Then consider a design that can be truely mistreated (left plugged in for days, short circuited, modded, run with worn bare or frayed wires) without burning down people's houses, electricuting anyone, etc.

Commercial products cary a price but you get a TON of sophistication for that money.

Agreed, lots of valid points :beer:

Nice...but w/ lights it kinda works.... you need to add design to the mix...

I can get 600L for less than 400 bucks with ok run time...but the design is not as simple...

For Example: for 300 bucks or so I can get 700 lum with 4.4 hours of run time, but I would have to get two systems: http://aebike.com/page.cfm?PageID=30&action=details&sku=LT0521

But, I would have a cluttered bar, no helmet mount, and the beam pattern my not be what I need.

I could also get a 200L set up for about 60 bucks or so (see example http://www.dealextreme.com/details.dx/sku.3786) but then there is mounting and beam pattern issues...

The ? of cost to make 600L LED lights is a bad question, at least to me...value or $$ that I am willing to spend is a better question and the answer seems to be driven by specificity of design...

Feel free to revise the quote anytime :)

Agreed, lots of valid points :beer:

That makes four of us on the planet! :)

A 3rd option in the build vs buy question is to take an older, less expensive model, and simply upgrade the emitters to get more light. That's the route that I took - I have a Cygolite 200 - an older light that came with the Luxeon emitters. There was a thread here that gave instructions for how to upgrade to Cree, so for about 20$ more, I got about 3 times the light.

The charger, batteries, handlebar mount and all the rest are the same as before.

That makes four of us on the planet! :)

5 counting Bailey :D

5 counting Bailey :D


:roflmao::roflmao:

He told me It's important to have active lighting and reflective material as well. :lol:

http://img225.imageshack.us/img225/8582/hpim8749gv2.jpg
.



OK Back on topic now...............



.

It's darned near impossible to offer it in a package that can be mounted easily to just about any bike, aimed easily, doesn't come apart under vibration and abuse and severe environmental conditions (how many DIY systems would survive long term riding in salt fog from ocean spray?), helmet mounts, flexible cabling with connectors that don't come apart, modular so you can easily replace certain parts of the system, LEDs to tell you when your batteries are nearing end of life, brighness and flash settings, etc. Then consider a design that can be truely mistreated (left plugged in for days, short circuited, modded, run with worn bare or frayed wires) without burning down people's houses, electricuting anyone, etc.

Commercial products cary a price but you get a TON of sophistication for that money.'

And where is the proof that commercial systems incoprorate any of this? This is a lot of words with no concrete examples or evaluations of any commercial system.

'

And where is the proof that commercial systems incoprorate any of this? This is a lot of words with no concrete examples or evaluations of any commercial system.

Isn't the Dinotte mounting system proof enough of the months of design and prototypes that go into such a system? Shucks, with multiaxis cnc milling, they even figured out how to hold a single led and a little circuit board with only twelve little tiny screws! Mine that I made on an old manual lathe in 15 minutes is held together with one dumb old internal snap ring and two o-rings (click! and it's assembled), but at least I was able to recreate the Dinotte mounting system. But no matter how hard I worked at designing it on a lunch break with a pencil, mine isn't as hot. Hang the Dinotte in still, open air, leave it on, and it'll smoke after a while. Now that's hot! And that Dinotte 4AA battery holder, just load it up with batteries and drop it loose in your pannier anytime you need a roaring fire. I never woulda thunk of using a battery holder with exposed 9-volt terminals like that, but that's the kind of sophistication you get for the premium price. [/sarcasm]

Labor.
Most of the cost comes from the cost of Labor, to assemble the product.

As evidence of this, I would point out that there are Cigarette Lighters with LED's, That cost less than $3.00 (three dollars). "Made in China" where the labor costs less.

I wasn't impressed with the mounting system that came with the 200L. My battery pack slides all over the top tube with the included 2x velcro straps. The small one and the big one are equally bad. I've accidentally hit the battery pouch once and spilled it all over the road with my new while getting out to sprint :( How hard would've it have been to have something that would securely attach the AA case to your bike?


Why put it on the top tube? I mount mine on the headtube. And I can easily crank the velcro strap down so tight that the pouch won't even slide far enough to the side to get the batteries out of the top.

As to cost, in addition to R&D and machining. You have overhead costs, support, shipping, salaries. Big difference between a homebrew and a business.

Also add to the engineering a good design. Almost all of the Homebrew 3 LED lights I have seen are huge ugly monstrocities. Not that I care about the looks per se, but they were all too large and clunky to effectively mount on my bike. Also consider, for example with the 600L, that it is custom designed with each LED on a separate circuit. That way any one can fail and you don't lose the entire light. Important if you are relying on the light to get you home.

Add it all up? I will buy the commercial products.

-D

Also add to the engineering a good design. Almost all of the Homebrew 3 LED lights I have seen are huge ugly monstrocities. Not that I care about the looks per se, but they were all too large and clunky to effectively mount on my bike. Also consider, for example with the 600L, that it is custom designed with each LED on a separate circuit. That way any one can fail and you don't lose the entire light. Important if you are relying on the light to get you home.

By this, you mean they're wired in parallel? Custom designed parallel? Well, I guess they're trying to save a little money there... The controllers are getting cheaper than good batteries. Cheaper and lighter (but less efficient) to have three controllers instead of adding cells to account for series voltage drop. Yes, there are step-up current controllers, but they are more expensive and even less efficient than wiring up three cheapos side by side. Unless they designed a proprietary three-amp, three-output controller in house, but that would seem like an extravagant waste of money in a market flooded with so many varieties of ready made controllers selling for peanuts... Nah, Dinotte wouldn't throw money away like that! :D

Add it all up? I will buy the commercial products.

-D

I thought about sending in my old 5-watt Dinotte for the upgrade. But, instead, I made two cree lights for the same cost and keep the outdated ten-month-old Dinotte as an emergency backup/relic. Maybe I'll knock out a triple just for fun sometime, but only if I can round up all the parts for under $40. Shouldn't be that hard.

'
And where is the proof that commercial systems incoprorate any of this? This is a lot of words with no concrete examples or evaluations of any commercial system.

Sorry about the long post, but sophisticated stuff is complicated.

As with any generalization there can be exceptions to the norm. But this is mostly true.

To find proof you would need to know the list of all parts and the costs of parts and labor and to read all the operating instructions for the lights.

I don't imagine a manufacturer would give you a bill or materials if you asked. I wouldn't. They don't want everyone to know. It's not worth the trouble. But I think others that design and develop new products, and start them into production, would agree with me.

However if you read all the features of what a Lupine light, controller, and the charger will do you will probably be amazed (if you read the whole thing). The amount of sophisitcation is amazing. My light came with three small booklets of a few pages each. One for the charger, one for the, light controller, and one kind of general operating and set up. How about resetable battery protection interuption? How about a charger that will charge your battery to 40% for optimum Li -ion storage? Or any other amount. How about the lights tell you the battery voltage when you plug it in? How about a reserve tank and a few different storage level indicators that are adjustable?
That's just the tip of the iceberg.

I'm not trying to be a wise guy, seriously read all this, or scan the begining of the paragraphs just to see the subjects addressed, or just scan the pages. It is what came with my light, all three links.

http://www.lupine.de/content/en/manuals//2006/Edison%2010.pdf

http://www.lupine.de/content/en/manuals//2006/Charger%20One.pdf

http://www.lupine.de/content/en/manuals//2006/PCS%20V4.1.pdf

This is the inside of my headlight..........................

http://img136.imageshack.us/img136/3039/phpthumb2ve5.jpg
http://img136.imageshack.us/img136/6503/phpthumbbz3.jpg

This is about the new 1400 Lumen LED from Lupine. Just some specs, not the manuals, they are more complicated.

Betty

The PCS switch is more than a simple turn on/off switch. The PCS controls the high and low beam, protects your rechargeable battery against deep discharge and includes a low battery indicator. It gives optimal dimming with variable frequency which allows even for a weak battery to generate a bright white light. Out of the box the PCS V7 is set to standard factory settings. There is no need to program the switch as the factory settings already allow full use of the lighting system. However some individuals have their own requirements which can easily be programmed.

PCS V7 Features
Switching on/off
Optimal dimming with variable frequency
Protection from total discharging
Voltage count – protects you not to go on a ride with an empty battery
Electronic capacity display: monitors your battery, red + yellow LED warning
3 additional settings for max. output – Power mode – normal mode – economy mode
SOS signal - Alpin emergency signal - Superflash
Capacity control – possible way to adjust the burn time monitor
2 steps dimming: High beam 100% power. Low beam choose from 1%, 5%, 40% and 60% for low beam
3 steps dimming: 100%, 40%, 1%
3 steps dimming: 100%, 40%, weak Superflash
Stepless dimming: 1% to 100%
0.25W Emergency mode for ultra long burntimes

The back of a circuit board for the Lupine "Wilma" four LED light. Notice all the chips inside the translucent blue controller "switch" too.

http://img512.imageshack.us/img512/2057/w7fm2.jpg

By this, you mean they're wired in parallel? Custom designed parallel? Well, I guess they're trying to save a little money there... The controllers are getting cheaper than good batteries. Cheaper and lighter (but less efficient) to have three controllers instead of adding cells to account for series voltage drop. Yes, there are step-up current controllers, but they are more expensive and even less efficient than wiring up three cheapos side by side. Unless they designed a proprietary three-amp, three-output controller in house, but that would seem like an extravagant waste of money in a market flooded with so many varieties of ready made controllers selling for peanuts... Nah, Dinotte wouldn't throw money away like that! :D

I honestly don't know. I think I read it on a review site. I'm not an electronics kinda guy (prolly another reason I don't DIY a light). All I know is that they touted it as a feature so that a failure of one light (either the bulb or circuit) wouldn't render the other 2 LED's a brick. So you could have a partial failure of the light and not have it go dark. Prolly important for those really counting on the light for a long ride.


I thought about sending in my old 5-watt Dinotte for the upgrade. But, instead, I made two cree lights for the same cost and keep the outdated ten-month-old Dinotte as an emergency backup/relic. Maybe I'll knock out a triple just for fun sometime, but only if I can round up all the parts for under $40. Shouldn't be that hard.

I did. Noticable difference and i am happy with the upgrade.

-D

I honestly don't know. I think I read it on a review site. I'm not an electronics kinda guy (prolly another reason I don't DIY a light). All I know is that they touted it as a feature so that a failure of one light (either the bulb or circuit) wouldn't render the other 2 LED's a brick. So you could have a partial failure of the light and not have it go dark. Prolly important for those really counting on the light for a long ride.


Multiple LED lights do not have that feature, they are all direct connected, which means if one wire fails, they all fail but given that actually will happen is highly unlikely as manufacturers do use the appropriate gauge wires to connect their lights. By what you mean, they would have to be connected in parallel, and they don't do it that way because one, soldering points and wire connection rarely fail, and two, the charging time will take forever, and KnhoJ has made rest of the point really clear.

Multiple LED lights do not have that feature, they are all direct connected, which means if one wire fails, they all fail but given that actually will happen is highly unlikely as manufacturers do use the appropriate gauge wires to connect their lights. By what you mean, they would have to be connected in parallel, and they don't do it that way because one, soldering points and wire connection rarely fail, and two, the charging time will take forever, and KnhoJ has made rest of the point really clear.

I'm just going off what I have read:

"They are also making all of the electronics redundant. Re-read that last sentence. Each LED is run off of it's own circuit. If one fails for any reason the other two will be unaffected. In fact, things could really go bad and you could lose two of the three and still be able to see where you are going. How cool is that?"

In this review:

http://acidinmylegs.blogspot.com/2007/02/500l.html

-D

Labor.
Most of the cost comes from the cost of Labor, to assemble the product.

As evidence of this, I would point out that there are Cigarette Lighters with LED's, That cost less than $3.00 (three dollars). "Made in China" where the labor costs less.

Nah, the LEDs those $3 lights are using are far less sophisiticated than the ones in a bike light. Sure you can make stuff with cheap LEDs, but the LEDs used in bike lights are not cheap. Nor are they housed in a simple plastic housing, nor do they use a simple plastic lense molded on to the LED. Parts cost alone for a bike light will be far in excess of that $3 lighter no matter where on the globe you decided to make them. Even if labor were free there's no way they could sell commercial bike lights for $3.

BTW, for your typical manufactured consumer good labor cost for assembly is roughly only 5%-10% of total product cost. Materials usually account for 80%, and the rest is "overhead": capital depreciation, engineering, sales, general, and administrative, etc. So the "overhead" peanut butter is actually more expensive, typically, than the labor to assemble.

I'd bet lots of these bike lights are made in China anyway...

Nah, the LEDs those $3 lights are using are far less sophisiticated than the ones in a bike light. Sure you can make stuff with cheap LEDs, but the LEDs used in bike lights are not cheap. Nor are they housed in a simple plastic housing, nor do they use a simple plastic lense molded on to the LED. Parts cost alone for a bike light will be far in excess of that $3 lighter no matter where on the globe you decided to make them. Even if labor were free there's no way they could sell commercial bike lights for $3.

But knock out a ton of them and talk a manufacturer into making them efficiently automation-compatible, and they could be a lot cheaper. Those lighters might be assembled by hand, they're light, small, and cheap to ship from halfway around the world where the labor cost is too small to measure, but on the other hand pick-and-place machines could pour those things out like water even in an American factory. Efficiency takes some consideration and a little effort, but it might be the only way the boutique bike-light businesses can compete if the less-than-boutique guys (cateye, planet bike, etc.) decide to cut margins and snatch up the bulk of the market.

Fenix is already jumping in on the opportunity with a product that coincidentally competes directly with bike-specific products doing the same thing for 2-3 times the price. If Fenix can justify the development cost, the molds, and substantial quantity initial product run for a Fenix specific, bicycle mount for a flashlight, they've already got a significant portion of the bike light market in their ring. When that many people were willing to go so far out of the way to rig up a flashlight instead of easier to buy and use bike lights, we might take that as a sign that the bike-specific light producers are outpricing the target market...

I honestly believe a bike-specific business could produce a bike-specific light using last month's Cree or SSC, maybe a Rebel led to keep the battery count down, in an efficiently produced bike-specific product sold in quantity in bike shops, and come out with a comfortable return for $50 or less. Meet the Fenix flashlight's specs with a competitive price, and they'll get knocked right out; why fiddle around with a flashlight you have to order online when there's this just-as-bright bike light with a quick release and everything at the lbs?

I honestly believe a bike-specific business could produce a bike-specific light using last month's Cree or SSC, maybe a Rebel led to keep the battery count down, in an efficiently produced bike-specific product sold in quantity in bike shops, and come out with a comfortable return for $50 or less. Meet the Fenix flashlight's specs with a competitive price, and they'll get knocked right out; why fiddle around with a flashlight you have to order online when there's this just-as-bright bike light with a quick release and everything at the lbs?

Considering the market size for bicycle lights versus the flashlight market, I am not as sure as you that this is a viable proposal.

Even with the law and plain common sense in line with having at least "be seen" lights when biking at night, I still don't see a majority of my fellow commuters (i.e. likeliest bicycle light customer) having a rudimentary front light; most just get by with a rear blinker and reflectors. If this potential customer population does not see the value in spending that $30 for a Cateye, how are you going to convince them to spend that $50? If you don't get enough volume, how are you going to get cost down?

Like many (including I) had wrote before, the proof is in the pudding. If there is such an opportunity as you seem to see, why aren't the big guys jumping in? More importantly, why aren't you jumping in?

More importantly, why aren't you jumping in?

Among other things, I'm not prepared to commit to a 5,000-10,000 part initial production run to convince a plastics shop to create proprietary molds just for some guy's pet startup business that hasn't turned over a penny of revenue yet. Fenix seems convinced it's worth the risk to invest in a non-native market, though. Maybe they snatched up someone's introductory product line rather than developing from scratch, that's also not something some garage entrepreneur could pull off.

The big bike accessory guys left a hole in the product line, people are rigging up flashlights to fill the gap, and they lost an unrealized portion of the market. Either they'll react with introductory level high-power led product, or we'll have a new brand name in the bike shops filling that niche soon. In other words, the prices are ready and set to race to the bottom, and that's not a smart time for someone like me to jump into the business...

^^^

Fenix is certainly aware of the bike market and has come out with a bike mount for their flashlights.


http://www.bikeforums.net/showthread.php?t=365414

https://www.fenix-store.com/images/fbike6.JPG

Fenix is certainly aware of the bike market and has come out with a bike mount for their flashlights.

Seriuosly, that is like bringing a knife to gun fight.

Fenix is certainly aware of the bike market and has come out with a bike mount for their flashlights.

Seriuosly, that is like bringing a knife to gun fight.

Go get em', cowboy!

200 lumens on high setting. Not very bright.

200 lumens on high setting. Not very bright.

I agree, I ride with no less than 600 lumens at all time, and it's still not enough light.

200 lumens on high setting. Not very bright.

There aren't any other led bike lights that make that much light for $50-60. Not even at twice that price. And it's really weird to see that. Who would have expected Cateye, for example, to completely disregard the cheap end of the product line? They've always worked the full spectrum, from $5 blinkies to pretty decent stuff. But suddenly they've focused on the top level boutique market, and completely disregarded the "upgrade" sales path which they've always maintained with other products. Consumers tend to purchase things bottom up; introducing a product top-down supports competitors. Like Fenix.

I agree, I ride with no less than 600 lumens at all time, and it's still not enough light.

That's disturbing to hear. I was saving up for a 600L. Not to be a knock on this forum but the MTBR forum seems to be much more active:

http://forums.mtbr.com/forumdisplay.php?f=124

Lots of DIY projects etc.

200 lumens on high setting. Not very bright.

Not all of us are light geeks who wish to have the brightest damn thing possible; some of us just wish to commute to work safely. Not everyone needs a 1200 lumen $800 light to pedal to work.......

That's disturbing to hear. I was saving up for a 600L. Not to be a knock on this forum but the MTBR forum seems to be much more active:

http://forums.mtbr.com/forumdisplay.php?f=124

Lots of DIY projects etc.

What is disturbing about it?

That's disturbing to hear. I was saving up for a 600L. Not to be a knock on this forum but the MTBR forum seems to be much more active:

http://forums.mtbr.com/forumdisplay.php?f=124

Lots of DIY projects etc.

600 lumens is not enough for city streets and to light up a wet roads, the water just bounces the light back in the air. 200 lumens is useless in seeing the road when it's wet.

In zero light conditions, it's fine.

There is a lot more people at the MTBR forums into night riding, it use to be that night time was the only time you can get solitude on the trail without seeing another human being. With the explosion of North Shore trail riding, it's not the same anymore, but it's still a heck of a lot better than riding during the day. Go to Whistler in spring, even in pouring weather, the line ups for the gondolas are very long.

Uh................200 lumens is still not very bright, and totally useless in some riding conditions. There's plenty of times when even car low beams seem like they are off in the rain and in traffic.

Like the rain, and/or a car aiming it's lights at you. You can't see the ground for a few seconds. If that's OK with you that's OK with me for your bike. I prefer to see the ground in all conditions. I do lots of after dark century, or last half of century riding. I need to see the little pieces of glass and the potholes etc. And I do, and I miss them. I have had this argument with lots of guys with weak lights and It always gets proven when they ride over a bottle. We do side by side tests all the time.

Lots of people don't ride over anything when they can't see the ground and think they never will. They might, they might not. I would like to see the ground all the time. Bike lights are just weak, that has nothing to do with someones personality or interests. It's just a fact of life.

What is disturbing about it?

That 600 is a minimum and that's going to be ~ $400 for a ready-made system with batteries. I'm going to totally agree that 200L is not enough for commuting. Especially for streetlighted streets and where you'll get rain/snow/ice. If you're going through pitch black streets, it might be ok.

I like to go fast so low power lights won't really cut it.

Oh yeah I just picked up a PB Superflash today per some users recommendations. And it is super bright for what it cost me ($13).

That 600 is a minimum and that's going to be ~ $400 for a ready-made system with batteries. I'm going to totally agree that 200L is not enough for commuting. Especially for streetlighted streets and where you'll get rain/snow/ice. If you're going through pitch black streets, it might be ok.

I like to go fast so low power lights won't really cut it.

Oh yeah I just picked up a PB Superflash today per some users recommendations. And it is super bright for what it cost me ($13).

People have been commuting for years with less than 600 lumens. Read some some older posts from BF and you'll notice that as light power increases so does the minimum amount of lumens recommended by users.

Two years ago the Dinotte 3w (120l) was the bees knees, and it was highly recommended; now that seems laughable as LED technology has increased, although street lighting hasn't changed. It's real easy to get caught up in the better and brighter mindset as technology is increasing so fast.

People have been commuting for years with less than 600 lumens. Read some some older posts from BF and you'll notice that as light power increases so does the minimum amount of lumens recommended by users.

Two years ago the Dinotte 3w (120l) was the bees knees, and it was highly recommended; now that seems laughable as LED technology has increased, although street lighting hasn't changed. It's real easy to get caught up in the better and brighter mindset as technology is increasing so fast.

I think more people are just finally realizing that the 120 lumen light they used to think was adequate is not really as good as they used to believe. Once you start using brighter lights that aren't washed out by car lights and that allow you to see details and objects in the road then you realize you might not have been as safe as you thought with the weaker light. Anyway technology advances and we benefit and that makes me happy.

I think more people are just finally realizing that the 120 lumen light they used to think was adequate is not really as good as they used to believe. Once you start using brighter lights that aren't washed out by car lights and that allow you to see details and objects in the road then you realize you might not have been as safe as you thought with the weaker light. Anyway technology advances and we benefit and that makes me happy.

Advances in technology are great, and I'm all for brighter lights, but if lights under 600 lumens are so unsafe most of us longtime commuters would be dead by now.

I hear ya. I don't find I need 600 lumens to be safe while commuting, but I don't think it is overkill either. A 120l I would find unsafe though. I think a couple 200l's is a pretty good combo, but for the price of 2 of those you can probably just get a 600l so that makes more sense. I was riding around the other night semi-ninja style with about 100 lumens and I didn't die either, but I certainly couldn't ride as fast as I would with 500 lumens or more and I wasn't able to see the road surface as well as I would like. I set my Wilma to 12% power (~100 lumens) just to see what it was like and it was doable, but when cars came at me it made it unsafe as 100 lumens wasn't enough for me to see the road safely. 400-500 lumens seems to me like a good amount to safely commute, but more certainly helps in some situations.