Epson LS10000 Projector Review

Technical Bench Tests

Objective bench tests of the Epson LS10000 were carried out using CalMAN 5.4 software from SpectraCal, X-Rite i1Pro2 and C6 meters, a DVDO AVLabTPG, and APL 18% test patterns. YCbCr was used as a colorspace as it provided the best chroma alignment on the Epson LS10000. The Epson stores individual memories for YCbCr and RGB sources, so make sure to calibrate both modes if you are using one or some sources will look worse.

Pre-calibration the THX and Natural modes provide similar performance. Both are relatively accurate but excel in different areas. The THX mode has more accurate grayscale and color checker while the Natural mode did best with gamma and saturations. I used THX for the pre-calibration measurements in the end as I felt they were better balanced. Without a calibration the Epson LS10000 is semi-accurate but not nearly as good as it can be.

This slideshow requires JavaScript.

Post-Calibration the LS10000 is virtually perfect. The Gamma tracks the BT. 1886 standard almost perfectly, the grayscale error is minuscule, and colors are very accurate. Every aspect of the image improves after a calibration and the Epson CMS does a good job of not introducing banding or artifacts during the calibration. There wasn’t much that stands out from the post-calibration aside from being a very accurate image.

This slideshow requires JavaScript.

With Game Mode enabled, the LS10000 has 57ms of input lag. With it disabled this rises up to 107ms. If you enable the 4K interpolation modes the input lag was no longer measurable. The Epson LS10000 has enough lag that I had to adjust my processor to deal with it, but once that was done it works fine.

After the calibration there were 508 lumens in Eco mode with the Iris open all the way. Moving to Normal lamp power bumps that up to 711 lumens, and Extra Bright was 970 lumens. Fully calibrated this is a very good lumens number and even if it falls 10% over the 20,000 hours you still have almost 900 lumens to drive a screen. Because of this the Epson LS10000 will pair nicely with a very accurate screen like the Stewart Snomatte 100 or even with an acoustically transparent one.

Finally, while content that uses the DCI/P3 gamut won’t be here until Winter, the Epson LS10000 can reproduce 96.8% of this color gamut. So when you get an UltraHD Blu-ray player, the LS10000 will be able to display virtually all the new colors that it is capable of, and more than any TV I have measured at this point.

Calibration Summary
Measurement Pre-Calibration Post-Calibration
Contrast Ratio: 10000:1 10101:1
Gamma Point: 2.89 2.38
Average Grayscale dE2000: 3.80 1.38
Average Saturations dE2000: 2.45 1.17
Average Color Checker dE2000: 1.57 1.74
Post-Calibration Lumens: 970
Maximum Lumens: 1424
Summary: Before calibration the Natural and THX modes offered good performance. THX was better with the color checker and grayscale, while Natural held the edge with gamma and saturation. Using THX as the starting point for the calibration, getting better results out of the LS10000 is easy. It calibrates to be accurate and has a very good contrast ratio. Since the laser should shift very little over time, you can calibrate it when it is new and have the same image years down the line.
  • Roger Brooks

    We currently have an Epson 5010 and are considering upgrading to the LS10000. However, we have a 135″ screen and the 5010 is fine in 2D but we wish it were brighter when watching 3D movies. (We’ve never had the bulb replaced after nine years, but only viewing one movie a week on it). My concern is that the LS10000 is only 1500 lumens. Are we going to be disappointed with the lack of brightness compared to the 5010 which is just adequate? Really appreciate your time!

    • The big difference is that the LS10000 isn’t going to drop off nearly as much. In the extra bright mode it did 970 lumens calibrated, with a 135″ screen with a 1.0 gain, you can expect to see 18 foot lamberts of light. The SMPTE rec is 14-16 so that’s a little extra brightness. It’ll drop off some in 3D, but the 3D mode will be a little brighter at the expense of accuracy. So the LS10000 won’t be dead silent in this mode, but likely similar to the 5010, and it will stay brighter for longer.

      • Roger Brooks

        So, even though the LS10000 is considerably lower rated in terms of lumens (5010 at 2400 vs. LS10000 at 1500) we won’t be disappointed with the brightness given the use of a 135: screen? We think the 5010 is “adequate” but wish it were brighter.