- Written by Chris Heinonen
- Published on 01 May 2013
The BenQ W1070 Projector On the Bench
The BenQ W1070 was calibrated using CalMAN 5.1 and i1Pro and C6 meters. ISF and User modes were both tested, and both produced incredibly similar results, so you can calibrate the BenQ without needing an ISF technician.
The best mode uncalibrated was the default User mode. It had the best grayscale numbers, and the best color numbers as well. When using the HDMI input, color and tint controls are disabled, so unless you have calibration equipment, all that you will be configuring on the BenQ are the brightness, contrast, and sharpness controls. Setting those is much easier when Whiter-than-White and Blacker-than-Black are visible, but those are clipped by default. They can be made visible by setting the HDMI Mode to PC from Auto, which I would recommend doing.
I'd also recommend sending all your content in the YCbCr 4:4:4 colorspace. Using 4:2:2, the BenQ used lower quality filtering on chroma detail, and the timing was misaligned by half a pixel. Using RGB produced similar results, but 4:4:4 was handled properly and should be used if possible. Once these are set correctly, the default User mode provides very respectable out-of-the-box performance, though one that calibration certainly improves. The grayscale is really good in User mode, and the gamma was close to 2.2 using the 2.4 setting on my screen.
Colors are decent, with a high amount of red error that leads to a bit of redness in some images before calibration. There are no Color or Tint controls to use with an HDMI signal, so this is the best you can do unless you do a full calibration with the proper equipment. The color checker chart is also very good for a non-calibrated display.
After calibration, I found I got the best results with Brilliant Color enabled, but enabling this without calibrating is a bad idea, as it over-saturates everything. After we enable it and calibrate we can get a pretty good grayscale, though the 2.2 gamma setting tracks a little low at 2.1 overall but I found the 2.4 setting to be too dark. You can't get the calibration perfect, as the Hue and Saturation controls interact with each other, but it still results in a much better image than before calibration. The blue error was acceptable as this is the color where errors are least visible, and having a higher error there with smaller errors in every other color still looked better to my eyes than the initial settings. Calibrating with Brilliant Color off shifts that error to red and green instead.
Measuring off the screen, I had contrast ratios of right around 1000:1 from the BenQ, with white output up close to 50 fL on my screen. Using a different method for measuring contrast, which removes the screen as a variable and tries to account for ambient light (which exists even in a darker room), I measured 1530:1 for the BenQ. I believe the latter method is more accurate and so I would say it has a calibrated contrast ratio right around there, and I will use this method for future projectors as well.
With Brilliant Color enabled, the greens are a little over-saturated, but without it they and red are both under-saturated. Looking at the two back and forth led me to just slightly prefer Brilliant Color, as there wasn't a noticeable green push to it. After calibration, I managed to have 1126 lumens in Economic mode, and 1683 in Normal lamp mode. In the preset Dynamic mode, with the lamp on normal I measured an incredible 2244 lumens. This mode is very blue, so I wouldn't want to watch it, but it is very bright. Almost 1700 lumens calibrated should make everyone happy, as the BenQ is really bright.
One new testing method I'm introducing here with the BenQ is the uniformity of the light output. Almost all projector measurements are made at the center of the screen, where the lens is ideal, but measuring around the screen really tests the quality of the lens as well, which most people take for granted. The upcoming new Spears and Munsil disc has a 13-point uniformity pattern, modeled after EBU Tech 3235. Here I will measure 13 points across the screen using a light meter, so I'm reading from the projector and not off the screen. Then I can input these into Excel and calculate the overall uniformity.
Here we see that one side of the BenQ is really much darker than the other. Light fall-off goes down to as little as 54% of the center at the corners, a noticeable drop-off. Compared to the median measurements, the high and low values vary by +33% and -28%, a fairly large spread. I also measured this last year on the Sony VPL-HW50ES projector, but as it was the first to have it done, I didn't report it to have some comparable data first.
The Sony variance is only 9%, in comparison to 20% on the BenQ, with the lowest output falling to 75% of the center. Good glass is expensive, as anyone that has an SLR camera knows, and cheaper glass often has light fall-off on the edges as we see here with the BenQ. This isn't unexpected, it's just something that we haven't measured before, but I will going forward to help differentiate the lens quality on projectors as well.
On test patterns, the BenQ failed to handle 2:2 cadences correctly, but handled 3:2 and 24p for deinterlacing just fine. Jaggies were handled only fairly, with the synthetic tests being OK but video tests showing very noticeable issues with jagged lines. There was no pixel cropping, and BTB and WTW were handled correctly if you used PC mode as I mentioned earlier. My advice would be to have your Blu-ray player send 1080p signals in the 4:4:4 colorspace to get the best performance out of the BenQ W1070.
Using the Leo Bodnar lag tester, I measured 33.65ms of lag for a 1080p signal, or almost exactly two frames at 60 frames per second. For most gamers this will probably be quick enough.