Top bin COB comparison 2
- Thread starter robincnn
- Start date
sixstring2112
Well-Known Member
I like pot ...and vinyl fencing 
it's as dead as the diesel engine in inner city use. commonly used yes, but without a future
ReddEyez
Well-Known Member
Me too
Out of all of his ropes and chew toys he likes that damn milk jug the best.
pop22
Well-Known Member
a mongo frog
Well-Known Member
Double the price of a lot of DE set ups roughly. Sounds like a great deal with power savings i guess. Honestly i haven grown with the cobs yet. I have a bunch but i haven't fucked with them yet. Actually I'm not even qualified to talk about them yet.
CobKits
Well-Known Member
only because most cob users are suckers for efficiency.
watt for watt you could run cobs hard at 45% efficiency and be under a buck a watt installed and still give hps a run for its money. esp some of the larger citis that scale well out to 150-250W each on a $40-$60 chip. remember with HPS youre in for $50-$100 in bulbs per year so a gavita is coming up on $1000 over 5 years
heres a $60 3618 at 210 watts at about 45% efficiency: https://www.rollitup.org/t/top-bin-cob-comparison-2.897765/page-6#post-12764721
4 of those:$60 chip+ budget say $40 each for a fat active heatsink = $400
(4) 200W drivers = $200ish
_____________________________
$600 and that 840W should beat a gavita. and unlike HPS when you dim it efficiency goes up dramatically so you can replace a 600HPS at like 400W or crush a 315 at like 200, etc.
watt for watt you could run cobs hard at 45% efficiency and be under a buck a watt installed and still give hps a run for its money. esp some of the larger citis that scale well out to 150-250W each on a $40-$60 chip. remember with HPS youre in for $50-$100 in bulbs per year so a gavita is coming up on $1000 over 5 years
heres a $60 3618 at 210 watts at about 45% efficiency: https://www.rollitup.org/t/top-bin-cob-comparison-2.897765/page-6#post-12764721
4 of those:$60 chip+ budget say $40 each for a fat active heatsink = $400
(4) 200W drivers = $200ish
_____________________________
$600 and that 840W should beat a gavita. and unlike HPS when you dim it efficiency goes up dramatically so you can replace a 600HPS at like 400W or crush a 315 at like 200, etc.
sixstring2112
Well-Known Member
Gavita lamps last a year so 500 not 1000.i have one thats at least 14 months old and still shows respectable # compared to new bulbs
a mongo frog
Well-Known Member
Yea but your forgetting you can turn a DE fixture up side down and cook a meal. That has to count for some sort of efficiency.
CobKits
Well-Known Member
so $750 then over 5 years with a $50 bulb every year.....and they also recommend replacing the reflectors which oxidize, so say another $40 or so after year 3
SoOLED
Well-Known Member
while all this may be true, I think people who use cobs wont use them for five years. they will stay on top of the newest best gear they can. so long term testing funny to me. I don't think I ran a light system with the exception of my SE HPS's for more then two years. .
so yeah will it stay lit that long, but in 5 years there will be something much more efficient, brighter, spec's that you will want to buy anyway.
if there is original owner running 3590's in five years its will be a miracle, to withhold the temptation to upgrade.
Last edited:
CobKits
Well-Known Member
yes and if youre into the cob infrastructure you just flip your chips out and reuse your sinks and drivers. just a few more years and chips will be cheaper than HID bulbs. i mean at equivalent efficiency of HID theyre under 15 cents a watt right now, and falling. HID is 5-10 cents/watt..... per year. and the bulb you'll buy next year will be no more efficient than the one you used last year.
Another angle for old cobs is to just turn them down to bump the efficiency up to 60% or more and run them next to your newer, more efficient cobs.
SoOLED
Well-Known Member
it has only been 5 years since the 580gtx came out. Fermi's oafish 40nm: w/a mere 3 billion: Transistors. Over heating, power hungry, with awful cooling solutions. that card cost 600.00 bucks just 5yrs ago( well 6 if you count its was released early in the quarter) now look at the new 1080 its 16nm(finfet) core is like 15 billion transistors, a astounding 13 billion increase. w/better: power use, cooling and performance.
that's where lighting is going to be, no amount of tweaking is going to get a 580, on same planet as Pascal 16nm and stacked memory, in parallax to hydroponic lighting solutions.
So it will be for chip on board, you might get away with skipping a tic or toc, but at some point there is just not going to be anyway you can compete with old hardware anymore. even thought it still works perfectly.
TBH, I don't even know if we will still be talking about COBs in five years, we could be talking: stacked COB's or particle ignition(OT: the PBC cannon closer to being a reality) or laser over FO, electromagnetic radiation emitting carbon: light on a nano scale contained by a magnetic field, no wires just a ball of light in thin air. Ok so getting wayyy ahead here.
technology is moving faster then, the hardware fails or a better solution is available. we may keep the case and power supply but, most of us will upgrade, be it video cards, or cob chips or jump to whatever new lighting solutions that work. over clocking or bumping efficiency only goes so far, and it seems to keep getting shorter and shorter.
Uberknot
Well-Known Member
Processors are at a wall right now......Video cards are about to take the next gen leap.
JorgeGonzales
Well-Known Member
Light speed is a constant, unlike CPUs. When your new light does the exact same thing as your old light, but saves a penny an hour, are you rushing to upgrade? If semiconductors did nothing but use less energy as time went on, only the greenest of hippies would replace anything until it died of natural causes.
Cars are a better analogy, and how many people upgrade their car every couple of years for extra MPG? You will never make up for in gas money what you spent on another car.
I disagree. Its possible to build a cob light with efficiency well over 60% that grow really well. We already have 90% plus efficient drivers. Cooling options are already well advanced thanks to the computer industry pushing the envelope.
Getting real close to point of diminishing returns on both efficiency and spectrum increases for the upgrade cost.
CobKits
Well-Known Member
yeah but its not a matter of the most cutting edge technology. all the high bay and stadium lighting is moving towards cob. the bottleneck is in chip efficiency and cost and thermal dissipation. they can improve all of those in the same package size. drivers push upwards of 95% efficiency and have a 20 year mtbf and all our aluminum heatsinks will still be operable when were in the ground.
in 5 years when 35mm LES cobs are 225 lm/W at 200W and are $10 nobody will be regretting their cob infrastructure
hell you can still buy CFLs 30 years later and those should be illegal not only because they are less than 100 lm/W they are toxic mercury nightmares
in 5 years when 35mm LES cobs are 225 lm/W at 200W and are $10 nobody will be regretting their cob infrastructure
hell you can still buy CFLs 30 years later and those should be illegal not only because they are less than 100 lm/W they are toxic mercury nightmares