The big, cheap monitor roundup

LCDs have come a long way in the last 18 months, when anything bigger than a matchbox would set you back a grand and image quality was always a wildcard. But we have moved on from these dark times into a new and enlightened era of affordable, big panels that also perform really well.

Introduction and how we test
LCDs have come a long way in the last 18 months, when anything bigger than a matchbox would set you back a grand and image quality was always a wildcard. But we have moved on from these dark times into a new and enlightened era of affordable, big panels that also perform really well.

The market always has a sweet spot, and right now it’s on 22" panels, so we’ve taken the time to compare half a dozen panels from the big brands, all of which are less than $400 and have a resolution of 1680 by 1050.

Unsurprisingly, the supplied facts and figures for each of these monitors were misleading, so the only way we found to choose between any of the offerings was to see them in action.

All the monitors that we looked at had a few common traits. They were all very bright -- even the darkest monitors put out more than enough light for a very bright room and should be dimmed significantly for normal use (we generally run ours at 20-50%).

Response times were all identical, and we couldn’t see any smearing. Interestingly, we found the 2ms monitors were no faster than the 5ms ones. Looking into it a bit further, we noticed that the 2ms monitors were only rated as such because of their grey-to-grey performance. The “typical" response for all them was 5ms, and grey-to-grey response wasn’t measured on panels that quoted 5ms responses, so from what we can tell, monitors that quote faster response times are just fudging figures.

None of the budget LCDs we tested had height adjustment, so keep your old phone books. You’ll probably want to stick at least half, if not the whole alphabet beneath your monitor to get it at a comfortable height.


A note on dead pixels
Every manufacturer has their own policy on what constitutes a defective pixel and what they are prepared to do about it. ISO 13406-2 classifies these different policies for simplicity. Class I monitors with any pixel defects will be replaced with a shiny new one. Class II monitors can have a few defects before they are replaced and often have restrictions on the location of the defects before replacement.

All of the monitors we tested were either Class I or II.


How We Tested
Short of a pistol duel, we fired everything we could at these monitors. The industry standard DisplayMate combined with our very own Mark I Eyeball gave us a lot of insight into our victims test subjects, but in order to properly penetrate their shiny black façade (hot. –Ed), we brought in custom software and pulled a few tricks out of our sleeve.

In DisplayMate we were looking at the range and quality of colour reproduction, and we focused on a couple of tests in particular. The first was a colour and greyscale gradient ramp. Here we looked at three things: whether every shade is distinguishable from its neighbours; whether the overall gradient is smooth or has steps; and whether the gradient gets linearly brighter, without dropping off to black nor blooming out to white too soon.

Next up we had a look at how well it can produce extreme greys. Due to the way TN LCDs work, the darkest and brightest hues are difficult to display. As a result, some panels can flicker violently or simply won’t display those hues at all. This gives us an insight into the true performance of the firmware that drives the monitors.

Measuring response times was a little trickier, and we used two tests before we got a result. The first: a UT3 deathmatch. In theory, the rapidly changing gameplay would illustrate any major issues with the monitor. But they all looked good, and none of them smeared colours.

Another test was in order.

Our second response test saw us writing our own software – a very simple program that rapidly flashes halves of the screen between black and white. By flashing at 62 Hz - just above the refresh rate - we created a page tear, giving us a hard edge to measure against. Next we whipped out a DSLR camera. With the exposure time set ridiculously low (1/400th of a second), we snap the screens and examine the image where the colours swap for gradients and smears that shouldn’t be there.

Our last test was a comparison of a standard image. We were familiar with how the image should look, and we used it to gauge the reproduction on the test monitors.

Finally, our overall impression of the monitor weighed heavily on the overall score. Here we were interested in the little details -- like card readers, multiple inputs and the general overall niceness of the monitor. Attention to those details was rewarded, but if corners were cut so was the score.