congrats on your Opples
I agree they give similar results with High CRI LEDs… but not w Low CRI LEDs
Stock Fenix E01… note the difference in CCT and DUV:
also, the Spectrum Plot in Opple 4 is not functional, no matter what light you test, for example one 3000K and one 5000K, there is no change in the position of the crosshairs.
On Opple 3 the spectrum plot crosshairs Do actually move, and will show 3000K is in a different position than 5000K
Here’s a comparison of a Wuben X1 on High.
Wuben X1 High LM3
CCT 5516K
Ra 67.8
x 0.3321
y 0.3523
duv 0.0058
Results from an older LM3 at a different date:
CCT 5591K
Ra 67.4
x 0.3303
y 0.3519
duv 0.0064
Wuben X1 High LM4
CCT 5602K
Ra 68.9
R9 -
x 0.3301
y 0.3430
duv 0.0021
Your Opple 4 reads duv 0.0043 lower than your Opple 3
on my Fenix E01 test, the difference is bigger, my Opple 4 reads duv 0.0087 lower than my Opple 3
fwiw, your measurements are at very low lux… suggest you try closer to 20,000 lux
So, here are three readings I took with mine that just arrived. Not done this at all before and this was just a quick test in a fairly cluttered cupboard so my methodology is probably sloppy, and I don’t have anything else to compare to at the moment (just bought a 3 as well, will do some comparisons when it gets here). Can also try to take some better ones with more lights when it’s dark.
5000k 519A:
W2:
E17A 1850k:
I’m still not super knowledgeable about what readings to expect, but these seem accurate for the 519A and W2 to me? E17A seems kind of off though and was also jumping around more than the others.
If anyone has a good guide on how to get consistent repeatable readings, I’d appreciate it.
Not the only way and prob not the best way but this is how I measure :
Excellent data collection! One thing you did that I really like, is you got the lux reading up above 20,000…
suggest you also share a photo of the lights, something like this:
notice in my photo that the hosts are visible, that is because I take beam color photos during the day, at my desk, not in the dark. This helps give the auto white balance on my iPhone, a chance to use daylight as a reference.
Here is a single light, with the readings from my two Opples copy pasted together:
I take Opple readings in the dark, on my bathroom counter… light is stationary on the counter, opple is stationary standing against the wall. They are very close, about 6" apart. I usually measure at 100 lumens or less.
Do you prefer to run the lights at the same ramp setting for comparisons, try to hit a specific target lux, or something else?
I try to get a lux reading over 20,000
and for side by side comparisons I sometimes match the output, on my lumen tube…
these lights were set to similar output of 100 lumens, and very close to the same lux… the results seem quite good… matching claimed CCT
Im still hoping for some Spectrometer confirmation for the R9 CRI values from the Opple 4. I dont put much faith in its DUV numbers… pending confirmation
Sounds good. I was doing a bit of experimenting and it looks like that’s the spot where it’s reasonable for LEDs I can find specs for (mostly 519A), I also noticed it tends to overestimate R9 on most stuff but especially cool LEDs as power gets higher, and for warm LEDs, it reads CCT a little on the low side. I guess makes sense the accuracy would be best when not pushing it to the edge of its designed operating conditions.
So at the moment and for the moment LM3 is at first position?
Has anyone tried to connect to the Light Master 4 with a mobile browser BLE API?
It’d be cool if we can use a webpage to connect, generate a chromaticity graph with CCT, CRI, duv, torch name and/or mode, generate a R1-14 graph, and have a copy/paste feature for raw data.
Here are some more measurements:
ZebraLight SC700d H1 2 LM3
CCT 4706K
Ra 92.3
x 0.3551
y 0.3669
duv 0.0037
ZebraLight SC700d H1 2 LM4
CCT 4766K
Ra 89.2
R9 79.0
x 0.3531
y 0.3654
duv 0.0037
Wuben X1 High 2 LM3
CCT 5373K
Ra 65.0
x 0.3358
y 0.3597
duv 0.0078
Wuben X1 High 2 LM4
CCT 5649K
Ra 65.6
R9 12.2
x 0.3291
y 0.3405
duv 0.0013
imo, not necessarily because
the LM4 does things the LM3 does not, for example, LM4 gives R9 info, which I consider valuable… and the LM4 does the DUV calculation…
otoh, the LM3 might be more accurate for DUV (but we have to go to a website to calculate and plot it), and its spectrum plot actually sort of works
the LM4 definitely has problems with DUV accuracy, and when testing Low CRI, it gives incorrect R9 data… Plus the spectrum plot is totally non functional
so, it depends how much you care about R9 data… the LM3 does not have that info… And it depends how hopeful you are that the DUV value might be remotely accurate… on either LM4 and LM3, they might both be wrong, compared to a Sekonic or Xrite…
hopefully McBob will be back with some comparisons to refine our knowledge of how accurate the DUV is on both Opple models… and whether the R9 on the LM4 is in the ballpark, at least for High CRI LEDs…
Which duv is true?
difference from LM3 to LM4 reading the same dedomed 3000K 519a:
Turns out my Sony TV is a flicker monster. Sigh.
Also, my cellphone switches to nasty PWM at very low brightness, but at higher brightness it’s (mostly) DC.
This meter is nice!
Oh and this one:
and… that’s the “flicker free” light from Amazon. CRI is 98 though…
Check the DUV online is not a problem… we use app for read Opple data so we can open another app and finding duv…couple of seconds! Obviously is not immediately and not fast how LM3 but is not a big problem at the moment.
The LM4 is upgradable?
The possible upgrade resolve the problem?
the problem is that after checking online the two Opple meters dont give the same results:
I spent three months forty years ago working with a Perkin Elmer spectrophotometer that had cost more than my house. It had just been serviced and set up properly by Perkin Elmer folks.
Did a bunch of analysis that meant someone was spending three years in jail for pumping highly toxic waste down a residential sewer.
Turned out the spectrophotometer gave readings that only depended on how long it had been powered up.
Don’t believe the hype on lab grade gear. It can be a lot worse than you even imagined.
If you haven’t the (Trusted and calibrated) gear to check it - it is making up numbers. I wasted half a year on the best (wet) chemical analysis I’ve ever done on that POS spectrophotometer. Which cost about 4 years of my pay.
I am sure it can happen. I just think that comparing one cheap uncalibrated device to another cheap uncalibrated device can be counter productive. Just looking for more data.
Of course the increasing divergence of data at different CCTs is very concerning. I am anxious to see what changes they make and what they will fix.
Cheule we need your Sekonic 
I just received a new Sofirn SC21 w 5000K LH351d and tested it on both Opple 4 and Opple 3. As we have seen with other comparisons…
the R9 is not credible, and the DUV dont match:
here is an Xrite test by zeroair, it coincidentally matches the duv of the Opple 3
====
swapped the LED to 3000K 219b:
Opple 4 shows almost identical R9 to the LH351d… which imo is not possible.
===
comparing flicker with Opple 4 and Opple 3, the frequencies do not match
