Reviewer's Note: This is a review of the NiteCore D10 and EX10 lights with Golden Dragon Plus (GDP) emitters, as compared to the standard Cree versions. For a detailed review of the D10/EX10, please see my original review thread. The GDP D10 and EX10 were provided by Edgetac/NiteCore.
UPDATE MARCH 26, 2010: The D10 has been replaced with a new "special edition" that lacks the continuously-variable interface but adds strobe and SOS support. Check out my detailed review of the D10SP for more info.
Part I - Build Differences
Cree Q5s on the left, GDP on the right:
Not much has changed on the surface of the NiteCore D10 and EX10. As you can see, the GDP versions look roughly comparable to the previous Cree versions, although the knurling seems to go a little further up, closer to the end of the heads. I notice the "D10" label has moved from the head to the body (written in smaller script, consistent with the EX10 label style). Labels remain sharp and clear, although I have noticed some variability among all the samples.
My D10-GDP seem to have a better lube applied, but the EX10-GDP had the same viscous blue-lube than the earlier Cree versions came with. In this case, it needed to be cleaned off for proper function (the light was too stiff to use initially).
A more significant difference is that my D10-GDP has a rather loud whine on its highest settings on standard batteries (although not on 14500 on max). All modern multi-stage lights can generate some degree of inductor whine, but my Cree-based NiteCore were all relatively quiet in comparison. This particular D10 sample on max is one of the louder lights in my collection (but virtually quiet on the lower levels).
UPDATE: I bought a second D10-GDP as a gift, which is completely silent at max - and has only the slightest of whines at some of the lower settings. As always, you should expect that YMMV.
One thing that does differ slightly is the head of the EX10. The stainless steel bezel ring on my EX10-GDP is bit thicker than my EX10-Q5. More significantly, the threads seem to be cut to more of a point on my EX10-GDP, making them wider overall. Although well matched for its own body tube, it will not screw into my EX10-Q5 body (and conversely, my EX10-Q5 head is too "loose" to screw into the EX10-GDP body). Not a problem of course, but something to keep in mind if you were ever planning to try and mix and match heads and bodies. The D10s were all completely interchangeable in my testing.
The most obvious change between the lights is the emitter - you can see the much smaller GDP emitter on both samples above (Cree Q5 on the left, GDP on the right).
What isn't as obvious is that there have been some changes to the reflector. First off, they are a little "deeper" on the GDP models, coming to a smaller opening at the emitter end (as you would except given the smaller GDP emitter). Harder to see is the texturing, which seems subtly different to my eye. I've check with NiteCore, and they confirm the GDP reflectors are more finely textured with "slimmer" lines. They are also say the shape has changed slightly (the original reflector was optimized to try and reduce Cree rings).
And that's about it for build differences - there are no changes to the UI, piston design, etc. For all intents and purposes, these are basically the same lights with a different emitter/reflector installed. To learn more about how the lights work, please see my detailed original D10/EX10 review.
For GDP beamshots, I have included the original Cree samples of these NiteCore lights, along with two Fenix Rebel lights and a custom mod of an Edison Opto KLC8 emitter (in an Ultrafire WF601A with a home-made sputtered reflector). I have added the KLC8 into the mix because that emitter was known to produce strong blue/yellow shifts in its spill (something the Golden Dragon has also been accused of).
Unfortunately, I have only had time for indoor "white-wall" beamshots for the moment. For the D10 shots, all lights are on 100% on Sanyo Eneloops. Beamshots are taken at about 0.5 meters from a white wall.
Note: Despite how the pics look, the GDP and Rebel lights are very smooth with no visible rings in real life.
First thing you will notice is the GDP has a secondary blue corona immediately adjacent to the wide yellow primary corona (surrounding a fairly neutral hotspot). The spill also seems less intense than the Cree Q5 version, but the infamous "Cree rings" are no longer present.
In terms of tint shift, you'll note the GDP is still a lot better than the KLC8 emitter, which had a blue hotspot, wide yellow corona, and pronounced blue spill.
In terms of overall beam pattern, the GDP is closer to the Rebels than the Crees (although with less wide spill than the Fenix head). Both my GDP samples are very smooth with minimal rings - the pics make the beam look ringier than they really are.
Here are some extra shots with the EX10, but they don't look appreciably different from above. All lights are AW RCR.
Discussion of Beam Patterns & Tints:
First off, let me be clear: These up-close white-wall beamshots are very misleading when it comes to actual usage!
For what it's worth, here are my subjective impressions:
At close distances (i.e. 1-2 meters), the secondary blue corona of the GDP is present, but it is not as pronounced as shown in the pictures above. It is also only visible if you are shining it on white objects. The overall visual impression is of fairly cool spill with a very warm (yellow) center. In constrast, the Cree has a consistent tint between center and spill - but the dark ring in the corona is noticeable. The Rebels probably have the best pattern at this distance, with a smooth even transition from hotspot to spill, without any tint changes (although you will note my RB080 is rather consistently cool whereas the RB100 is consistently very warm).
At longer distances (i.e. 3-10 meters), thing are very different. The secondary blue corona of the GDP is invisible, and all you see is a very nice wide hotspot/corona which is very warm-tinted. In contrast, the Cree looks a lot like a spotbeam, as the Cree ring produces a sharp demarcation to the dimmer spill area. The Rebels still have a pleasing transition, but the corona around the hotspot is not as bright as the GDP, limiting how much you can see immediate around the hotspot. Frankly, the GDP is probably the most useful beam pattern at this distance, since it lights up the widest area in its warm-tinted hotspot/corona.
At really long distances (i.e. >10 meters), all that matters is the hotspot. In this case, I personally like the Cree pattern, since it gives me fairly uniform illumination of objects in the center area.
When it comes to tint, things also change with distance. Up close, you may think my GDP samples look on the cool side overall. At intermediate distances, they definitely look very warm, since the yellowish corona dominates. At longer distances where you don't see much of the corona, the center hotspot actually looks pretty neutral to my eye - but with slight cool tint around its edge.
The take-home message here is really this: There's no such thing as a perfect beam at all distances for all uses.
The Crees suffer from the infamous dark rings around the hotspot (although this can be minimzed with a well designed reflector). Still, some may actually prefer their spotbeam-like appearance - especially at a distance.
Personally, I think SSCs and Rebels have typically have a fairly pleasing patterns for up-close work - but Rebels are hard to find in mainstream lights now (due to Luxeon's QC and quantity production issues). But I also find the typical SSC/Rebel hotspot pattern to be a bit narrow, without much of a corona.
IMO, the GDPs offer one of the best compromises with their nice wide corona and lack of rings. But the blue/yellow tint variation across the spill/corona is noticeable at up-close distances on white or lightly-coloured surfaces. It is important to note that the supposedly "cool" overall GDP actually gives you a decidedly warm perception at intermediate and longer distances, because the yellow corona dominates.
Output, Throw and Runtime Comparison
Testing Method All my output numbers are relative for my home-made light box setup, a la Quickbeam's flashlightreviews.com method. You can directly compare all my relative output values from different reviews - i.e. an output value of "10" in one graph is the same as "10" in another. All runtimes are done under a cooling fan, except for the extended run Lo/Min modes which are done without cooling.
Throw values are the square-root of lux measurements taken at 1 meter from the lens, using a light meter.
Throw/Output Summary Chart
At first blush, it looks like initial max output of the GDP samples are lower than the Cree Q5s. But you need to realize that there can be considerable variation in max output between samples of the same light (likely due to Vf differences). See the runtime graphs below before jumping to any conclusions.
Min output levels seem unchanged.
Note that I have 2 samples of D10s with Cree Q5s (identified as #1 and #2 below). I've also include a NDI Cree Q5 in the runtimes, since the circuit efficiency has not changed signficantly in the D10s.
Li-ion rechargeables (14500/RCR) - D10 & EX10
- Max output of my GDP samples tends to be at the lower end of my Cree Q5 samples, but with similar runtimes overall.
- When matched at Med-ish output levels, runtimes appear roughly similar between my GDP and Cree Q5 samples, although with perhaps just a bit less output or runtime on the GDP samples.
- Taken together, these results suggest that the GDP versions are equal (or close) to the Cree Q5s in output/runtime efficiency on 14500/RCR.
1xAA Standard batteries (NiMH/Alkaline) - D10
- Max output of my one GDP sample was lower than either of my Cree Q5 samples on standard batteries, but with similar runtimes to the higher output Cree Q5.
- When matched at Med-ish output levels, runtime of my one GDP sample was clearly shorter than either of my two Cree Q5s (which were exactly the same to each other, despite having different max outputs).
- Taken together, these results suggest that the GDP versions are slightly less output/runtime efficient overall, with potentially lower max output on standard batteries.
1xCR123A Standard batteries (Primaries) - EX10
- Max output of my GDP sample was definitely lower than my one Cree Q5 sample on 1xCR123A, with increased runtime (but runtime was not increased as much as I would have expected if they were equally efficient).
- When matched at Lo/Med-ish output levels, runtime of my one GDP sample was shorter than my one Cree Q5.
- Taken together, these results suggest that the GDP versions are slightly less output/runtime efficient overall, with potentially lower max output on primaries.
Note that there is considerable danger in drawing conclusions about overall output/runtime efficiency based on only 1-2 samples of each light. But I think a couple of take-home messages are suggested by my testing:
- On Li-ions, the GDP versions appear to have equivalent or nearly as good overall output/runtime efficiency. Max outputs tend to be toward the lower end of the natural variation among Cree Q5s.
- On standard batteries, the GDP versions appear to be less output/runtime efficient - but the performance difference is not huge when you take normal variation among samples into account. Max output seems to be a bit lower than my Cree Q5s.
Take from that what you will. Personally, the difference is not great enough to me to be a deal breaker either way. I would base my decision on the beam characteristics.
I have been EDCing my Cree Q5 D10 for the last month or so, and will now start carrying my GDP D10 instead. I will report back further impressions as time goes by as to my ultimate preferences.
UPDATE 9/10/08: Color Rendition
There has been a lot of discussion about Color Rendition Index (CRI) differences between various emitters. Unfortunately, our perception of color is heavily influence by tint - for example, "cool" tints look brighter, but "warm" tints tend to differentiate greens/browns better outdoors.
I've taken shots of part of my garden that has green, browns, blacks, reds and whites. All lights have been matched for overall output, and are taken at 5 sec exposures at f3.2.
First, a control shot in daylight
Fenix Rebel RB100
NiteCore D10 Cree Q5
Nitecore D10 GDP
I don't know about you, but I don't see any great difference between the lights.
Per request, here is an the animated gif comparing just the Q5 and GDP (reduced in size and optimized for bandwidth). As is quickly apparent, the spill is less bright on the GDP up-close. The GDP also has a much wider corona around the hotspot. Thus any color differences in the flowers seems to have more do with light intensity (i.e. less bright in the spill, over-exposed near the center due to the long exposure times)
Although I know it seems like the brighter spill of the Q5 would be better for most uses (i.e. more floody), this benefit drops rapidly once you shine the light out >2 m. In the 3-10 meter range, corona width is far more important to help you see what you are looking at. This helps reiterate my point that the GDP is not best suited for up-close tasks, but is good for intermediate distances.
Certainly by eye, I could detect no real difference in color rendition between the 3 lights in my comparison. I think everything we are seeing here is just corona/spill intensity differences.