[selfbuilt]

REVIEWER'S NOTE: This thread has been replaced with a new comparison review of the latest multi-stage 1AA lights:
Multi-stage 1AA Review - Part III: Runtimes, beamshots & more! Please continue the discussion of the new lights on that thread - thanks!

The contenders:

From left to right: Jetbeam C-LE (group buy), DealExtreme X.V, Jetbeam MkIIX, DX X.1, Rexlight 2.0, Fenix L1D-CE


Beamshots:


The relative output can be a bit misleading in the photo (i.e. the cool-tinted lights look brighter than they are). The MkIIX is probably the "purest" white of all the tints. The DX X.V and Fenix L1DCE are bit warm, and the JB C-LE, Rex 2.0 and DX X.1 are a bit cool. Here's a lower exposure to show you the hotspots better:



Summary Chart for 2650mAh NiMH on Hi/100%


All my output and throw numbers are relative for my light meter/box setup. Throw values are calibrated to Quickbeam's flashlightreviews.com, so you can convert to approximate lux at 1m by squaring the Throw values.

Notes:
DX X.V, JB C-LE, and C-LE v1.2: X.V and original C-LE both use PWM at 73 Hz. Twist action is a bit stiff on both, due to battery tension. New v1.2 C-LE uses 317 Hz PWM (i.e. much less "flicker"), has improved threads, is thicker in diameter, and features a "memory mode" that retains last setting used
DX X.1 and JB MkIIX: Both use PWM at 120 Hz, build quality is excellent and roughly equivalent, but slight advantage to MkIIX in my case. Complex user interface, but manageable.
Rex 2.0: I opted for the black Rex 2.0, and mine doesn't retain the 5 sec smart memory feature on regular batteries - only on 14500. PWM freq is 94 Hz. Build quality is pretty good.
L1D-CE: Current-controlled, simple elegant user interface.
14500 Batteries: Although my L1D-CE, JB C-LE, and DX D.V. can all take 14500, they are not recommended and I don't want to run the risk given reports of people blowing their circuits on these lights.

Method: home-made lightbox a la Quickbeam's FR.com method. My relative overall output numbers are typically similar to his, although generally a little lower. You can directly compare the graphs - i.e. an output value of "10" in one graph is the same as "10" in another.

Runtimes:
"Hi" mode on NiMH (Duracell 2650mAh)



Med-Hi and Low modes on an AW Protected 14500 (750mAh) on two separate graphs clarity



A couple of comparable "Low" and "Medium" modes on alkaline Duracells, to allow you to compare relative output/runtime efficiency. Again on two graphs for clarity.




Key observations:

DealExtreme X.V vs. Jetbeam C-LE vs C-LE v1.2

• On Hi, lower output on X.V compared to original C-LE (from group buy), and with WORSE runtime. C-LE v1.2 has slightly increased output overall, and intermediate runtimes to the earlier two versions.
• On Medium, similarly lower output and with lower runtime on X.V compared to C-LEs. C-LE v1.2 has greatly improved output and runtime, but at a cost: now has rapid drop-off to zero output (i.e. lacks the long "moon mode" of the earlier versions that switched into low mode for extended runtimes)
• On Low, the X.V again has lower ouput, but now outperforms the original C-LE in time to 50%. The new C-LE v1.2 has about the same runtime as the X.V, but with twice the output! Note that unlike the original C-LE, both the X.V and C-LE v1.2 lack a "moon mode," and drop out of regulation to zero output very quicky.
• Apparently, X.V uses an early version of the C-LE light engine (notice different head knurling), which has some efficiency problems at higher output levels
• Update to v1.2 C-LE now includes 2sec "memory mode" feature that retains the last setting used - but with intermittent problems (see Conclusions at the end of the post)

DealExtreme X.1 vs Jetbeam MkIIX:

• On Hi (100%), DX X.1 is noticeably dimmer then the equivalent MkIIX, but compensated by extended runtimes
• Generally equivalent in all others modes and other batteries
• Regulated output on 14500, but runtime not as long lasting as Rex 2.0

Rexlight 2.0:

• Brightest light on NiMH initially, but at the expense of short runtime.
• On lower modes, alkaline runtimes are similarly poor compared to DX, JB and Fenix lights.
• Definitely seems to be designed to run best on 14500. Output on Hi on 14500 not as bright as DX X.1 or JB MkIIX, but much better output/runtimes overall (not regulated).

Fenix L1D-CE

• On Hi with NiMH, considerably brighter than DX/JB models. Note quite as bright as Rexlight, but much longer runtime
• On Med/Low, unbeatable output for comparable runtime of other DX/JB/Rex models

Conclusion:

• On regular batteries, L1DCE best choice for max output (on Hi) or best output/runtime (i.e. on Med/Low, L1D-CE is 2-3 times as bright as most of its competitors, but with equivalent runtime). Need to go with one of the other lights if you want a lower low.
• On 14500 batteries, DX X.1 or JB MkIIX are brightest and have regulated output, but Rexlight 2.0 has the best output/runtime ratio in all modes.
• DX X.V is not as good a performer as the JB C-LE from the group buys in terms of either output or runtime in primary or high modes. Signficant improvements on threads of C-LE v1.2, as well as considerably improved output/runtimes in primary and low modes, but at the expense of loss of moon mode on alkaline.
• New C-LE v1.2 memory mode is problematic, as contact problems quickly arise in every day use, resulting in missed modes, out-of-sequence switches, or restricted number of modes available. Threads and contacts need to be kept scrupulously clean if the light is to function properly.

There you have it! Should hopefully give you a pretty good overview for general comparison purposes.