Like these:
https://www.amazon.com/gp/product/B0B6LF1712
and these:
https://www.amazon.com/dp/B0BCXGLT1H
Hue lights have a reputation for being rock solid, and that’s been my experience so far. And they work great with the Blue 2-1’s. The pricing is pretty brutal though, about 10x the cost of those Amico lights.
However, the Amico lights are really nothing more than dumb leds in a retrofit configuration. Really no comparison to the Hues.
Well two things is that I’m pretty I’m pretty much all in zwave. Could get ZigBee lights but would prefer to keep it all zwave at this point. Second, as you mention, 185 for a 4 pack. No way. I know hue has a good reputation but that’s way too pricey. The amicos are dumb lights, I just control them with my zwave switches, but would love something like the hues, with that functionality, for a reasonable price.
THIS! I think there is a lot of demand for alternatives to hue can retrofits. A19 bulbs are offered by so many different companies that it will be hard to differentiate yourselves. Bring a retrofit with CCT or Color at a lower price point than hue and that is a compelling product.
I just replaced 10 downlights in our new home with the Hue white retrofits, and have another 11 to do, not sure if I’m going to go with white or color for those. Yeah, it’s painful in the wallet for sure.
I put some cheapo retrofits into a rental property we have a couple of years ago, and one of them went bad after two years. Went back to Lowes where I bought them, and of course that model isn’t made anymore, and nothing they had matches the trim ring or light output. Found something close and it’s ok, nobody but me will probably notice.
But I don’t really want to make an investment in no-name lights for my home that might not be around in a few years when they start to go bad. I have more trust in Hue, although I’m not sure exactly why. Between the Hues and the 40+ switches I’m replacing with Blue 2-1s, I’m seeing a second mortgage in my future 
This is another argument against smart bulbs… they are EXPENSIVE and there is no guarantee that company will be around 5 years from now when a bulb burns out or that their aesthetic won’t change even if they are. So, now you have to replace a whole rooms worth of smart bulbs when one burns out? No thanks.
With dumb bulbs, I usually buy contractors packs and keep some spares for just this reason.
You could apply this same logic to smart bulbs as well. While they don’t have contractor packs, I usually will buy 1 or 2 extras to keep as spares.
Not all smartbulbs are crazy expensive. The last ones I bought were something like $90CAD for 6. Costco sells 4 packs for $29.99. More than a dumb bulb sure, but not crazy expensive. Unless we’re talking Phillips Hue and their $65 bulbs which is just ridiculous.
I have four Innovelli RGBW bulbs in front of my house. They are programmed for a different color each day of the week. But in the house, I use CT strips and bulbs with white circadian control. CRI is important to me and I feel that CT bulbs and strips offer better CRI than RGBW.
I love those CT led strips with warm white and cool white channel. Hard to find controllers for those.
WLED supports CCT strips. You won’t find a better controller for LED strips of any type than WLED. And all you need is an ESP32 that costs about $5 to run it.
I have always bought RGBW, even though it’s usually white I do use the color regularly and why limit yourself to a single color 100% of the time when you don’t have to. Of course I like Hue’s because I can do cool scenes in a room with 4-5 of them. The only CCT I have is not really even white, it’s the Hue “Edison” because I wanted that for a specific purpose.
I prefer ESPhome, because I’m using 24VDC non-addressable strips.
A19 I get, but I don’t understand why BR30 vs PAR30.
The only smart bulbs I have installed are either PAR30 or PAR38. RGBW.
I wish could find RGBCCT PAR bulbs, but I can’t.
Design-wise, I’m leaning toward only using RGB bulbs in strips and doing all point lights as CCT. Though I wish the designs for CCT bulbs used three phosphors, something like 5000K, 3000K, and 1600K, as most CCT bulbs only have a tiny range, like 3000-1800.
ESPHome is an amazing project! For non-addressable strips there’d be little benefit in switching to WLED unless you wanted to treat multiple strips on different controllers as 1. WLED really shines with the addressable strips. My next LED project is going to be Hyperion with WLED for tv ambient lighting
I have our bulbs set to change temperature depending on what time of day it is, but then I also throw in some color (green and gold) when a Packers game is on or turn all the lights blue if a water leak was detected. I have plans to do something with them if the AC or heat is running and a window or door is left open too long. Then we’ll play with some colors on holidays or kids’ birthdays sometimes too.
I have a few Philips RGBW Wiz bulbs, and I really only use them for CCW. The WiZ bulbs are interesting because they use the Red LEDs to create a warmer white than the Warm White Leds can create on their own, but the bulbs are wonky and the different drivers take longer to warm up so your bulbs go red first and then warm white. I personally think those bulbs fit my needs at the time, but I will not buy another WiFi bulb now that i’ve seen the superiority of Zigbee. Zwave is cool too but expensive. Tradfri bulbs are like $12 Canadian.
I have a bunch of Halo brand Zigbee recessed modules, and they’re great except for not having a fade-on/fade-off that I got used to with my other bulbs and smart switches.
What about Par-20 bulbs? I am surprised there’s little in the way of bulbs in this size. Every recessed light in my house has a full can behind it so I can choose to change out to a regular bulb or smart bulb depending on needs. I was just amazed I couldn’t find CCW bulbs in PAR20
Just musing out loud… Does Inovelli see a niche for A19/BR30 RGBW or CCT bulbs? They seem to be the thing that the market already has met.
I’m wondering if going more specialized is the way to go. Be that odd shapes like PAR20/PAR38/MR16 or maybe doing RGBALC (red, green, blue, amber, lime, cyan) or RGB+365nm or a wide range 10,000K-1200K white? How about frosted T185 or T30 bulbs? Those are all form factors where you just can’t get a Zigbee or ZW bulb.
Just seems like like RGBW A19 is a market that is just too well met right now. Competing with IKEA isn’t easy.
If the monster companies like Phillips aren’t doing these form factors, there’s a reason… I just had to google what a T185 and T30 bulb look like. I can’t see them having a large enough market to cover the expense of bringing something like this to market, even if capturing 100% of the market for them.
I get the overall point you’re trying to make though. Take the road less travelled and capture a niche market. The problem with Niche is that it’s not often profitable and 1 bad gamble could seriously damage a company financially. Especially a smaller one.
Take a company like Soraa or Tala… I’ve probably spent about $2000 on dumb bulbs between the two of those companies. I wonder how much of the cost of their bulbs would be affected by making them Zigbee or Z-wave. Would it at $5 or $25? When you have a bulb that’s already $100, that’s not too much more to ask. I don’t know where the costs are. Is it optical design or electronics? I’m kind of surprised that modular bulbs aren’t a thing. Like a electronics package that’s either E26 or E12 or GU10 or whatever that has a bayonet connector for interchangeable optical modules. Could be a PAR20 RGB 9 degree spot or a A19 RGBW or a T30 CCT. Again, just thinking out loud about the lack of options in the market other than run-of-the-mill A19 and BR20.
One other thought… the quality of light suffers when you pack more LED colors into a bulb unless you have VERY advanced optics. The best quality LED bulbs are single color, and getting decent grade CCT bulbs is a challenge. I’ve never seen a quality RGBCCT bulb, so while I’d love to get one, I think the cost or some other technical limitation makes them non-viable.
Some of the Hue color bulbs aren’t terrible - not spectacular, but “good enough” for my purposes.
For example, I recently tested one of the latest Hue A19 color bulbs (1100 lumens, 75W equivalent, model LCA007) with a spectrophotometer. Over the white tuning range of 2200k to 6500k, their minimum CRI Ra is 89 and minimum R9 is 38. That’s still mediocre compared to a Soraa bulb, but it isn’t that far off. (If I’m being perfectly honest with myself, I’m not sure I could even tell the difference between the color rendering of this Hue bulb and a Soraa bulb without breaking out a color checker chart… hence my feeling that this may be good enough for me)
The prior generation A19 color bulbs were much worse, though (model LCT019). At 4000k, I measured a CRI Ra of 83 with an R9 of only 12. (this is one of my big complaints about Hue - there is no way to know how a particular bulb will perform without buying it and testing it yourself, and most people aren’t going to have the tools needed to properly test a bulb)
I suspect Hue’s hardware/firmware design is already up to the task of producing an excellent RGBCCT bulb - they “just” need to upgrade their white LED chips to some higher CRI models. But this would have 2 major downsides:
That could be a tough sell - how many people are going to want to spend more money for a dimmer bulb? Especially when Hue bulbs are already a bit disappointing on brightness… and they’re already driving the LEDs harder than they really should be (as evidenced by their underwhelming lifetime specs).
LED tech is continuing to evolve, though. I’m optimistic that high-CRI white LEDs will continue to become cheaper and more efficient, which will hopefully result in increased adoption.
There does already exist a high-CRI drop-in replacement for Hue’s RGBCCT LED strips (Bifrost by Sowilo DS). It can be driven directly by the Hue lightstrip controller. I have one of these strips, but am still waiting on a Hue controller to fully test it. From just testing the LED channels hooked directly to a power supply, it seems like this strip will perform well: naively combining the 2200k and 6500k channels produces a 3800k white with CRI Ra of 92 and R9 of 70 (on their own, each white channel has a CRI Ra of 97+ and R9 of 86+), and I’d expect this to improve once the RGB channels are engaged as well.
BTW, if you like Soraa bulbs, you might be interested in LTF’s Sunlight2 bulbs - they have similar color rendering performance to Soraa bulbs, but have a dim-to-warm option (3000k down to 1800k). I just received some of the GU10 ones, and am impressed with them. They do live up to their specs. I do wish that they started at 4000k, though, and that they were available in A19 (only GU10, PAR20, and PAR30 unfortunately).
I tried some of those (in MR16). The color was great (though I agree, the K range on CCT bulbs is lame, especially since LTF makes tunable COB ). Having a few choices of beam angle was good, too, but 24 degrees wasn’t tight enough to replace my Soraa spots (which are single color, though). What I did NOT like was the way the LEDs blended. There was a significant delay between the dimming and the color change, and the color change was in obvious discrete steps. I want to say there was like 8 or 10 steps. So if you had a bank of say 20 bulbs on a wall, and you dimmed the whole set, you’d get this wacky light show. Maybe it was an issue with my transformer or dimmer, I should try it with their electronics.
Could we also perhaps vote on the importance of CRI?
For me, and I suspect many others, the availability of color or color temperature options are outweighed by the quality/CRI of the actual light. Some measurements of the quality of the light (CRI and CRI R9 in particular) could go a long way in reassuring a large portion of the technically-savvy user base!