Home  |  News  |  Reviews  | About Search :  HardWare.fr 

MiscellaneousStorageGraphics CardsMotherboardsProcessors
Advertise on BeHardware.com
Review index:
LCD, David vs. Goliath: Iolair vs. Dell
by Vincent Alzieu
Published on April 28, 2008

Games: reactivity, input lag
It’s something we’ve never done before: for once we will quickly go over the reactivity of these two screens (in fact, it’s excellent and much better than what announced response times suggest) so that we can spend more time on a more negative characteristic of the Dell, its input lag.

After having tried a few response time measurement methods with an oscilloscope that often gave contradictory results with our visual impressions in games, since 2005 we have opted for a more concrete test using screen shots of afterglow with a reflex in burst mode.

Here’s the concept in this test: A car moves from right to left at high speed. The movement isn’t perfectly fluid and depending on its speed, the car is shown in several successive positions. When this process is sped up, the car goes very fast, positions are very close and the eye perceives a fluid movement.

The perfect screen
A screen with two afterglow images

A monitor without ghosting effects would have previous images completely fading away when a new one appears. This is ideal, however in practice, it's often not the case as images progressively fade. Sometimes up to 5 afterglow artifacts remain on the screen representing the visible white trail behind objects.

We capture this LCD defect with a camera at a shutter speed of 1/1000 s by taking 50 pictures per test. We then can see a monitor’s ghosting effects, or the car’s position in the entire process from the moment when afterglow is at its maximum, up until when the next image is about to be created and the previous image is the least visible.

Here are the two extreme states between which each monitor’s afterglow can oscillate.

PVA 6 ms : Dell 2408WFP

MVA 6 ms : Iolair MB24W

The above test already gives us a good idea of ghosting effects on the two screens. We can go into even more detail by looking at the frequency of minimum and maximum states of afterglow.

In dark blue : the average number of colored afterglow images (the most bothersome type).
In light blue : transparent afterglow images which in practice are barely noticeable.

PVA 6 ms : Dell 2408WFP

MVA 6 ms : Iolair MB24W

What do these figures imply ?
1 : More or less that differences are so small that they are imperceptible and the two monitors are equivalent. Moreover, this is consistent with our visual evaluation.
2 : Also that they have the same behavior as the fastest 24 inch TNs equipped with panels that have a given response time of 2 ms. In reality, the MB24W and 2408WFP are much more reactive than 5 ms TNs.

So does this make them gaming monitors? The input lag test is the other determining factor.
Input lag in games
If you didn’t already know, almost all LCD monitors have a small delay in display. To measure this, we photograph a chronometer which is precise to 1/1000th of a second displayed in clone mode on our reference CRT and the LCD we are testing. We take 12 consecutive differences, eliminating the two extremes and then find the average delay.

For this test, we deactivate graphic card / screen synchronization to capture a more precise result instead a value that is rounded to the nearest image. For this reason, the resulting measurements are not dependent upon the fps value. An LCD functions at 60 Hz (even those that claim 75 Hz) and so 17 ms equals an average delay of a single image. In the same way, 33 ms equals 2 images. This may not seem like much, however, to lan gamer it could make a world of difference. For example, an adversary with an LCD or CRT will see his character two images before him. This can be compounded by the fact that the mouse can add another 1 to 8 ms (except very poor ones), and the graphic card adds a half image delay (at best and actually this can be 5 to 50 ms). More specifically in our research, we found that this can be a cumulative delay of 110 ms or 6.5 images. In this case, it is indeed noticeable and even bothersome for some gamers. For more details on this subject, see our article, LCD images delayed compared to those of CRTs ? Yes ! .

PVA 6 ms : Dell 2408WFP

MVA 6 ms : Iolair MB24W

No, results are indeed correct and we have two screens with diametrically opposed behaviors. On the one side is the Dell with an average delay of 4.1 images, and on the other, the Iolair with zero input lag – even having the luxury of occasionally being slightly ahead of the CRT.

As you can see on the Dell, the worst is that the delay varies from 60 (3.6 images) to 89 ms (5.3 images) once we discard the two extremes values. If we include the minimum, nothing changes. On the other hand, if we add the maximum, the 2408WFP is 10 images behind!

Even with ‘’only’’ 4 images, the difference is perceptible in games and movies because you can hear sound before the image. More concretely in games, this translates into:

The two screens are in clone mode; however, with the Dell the game has just started while on the CRT we can see the character has started to advance with an enemy appearing in front of him. If the two screens were used by opposing players, guess who would win? Even in games versus the computer, on such a monitor we start out with a handicap.

This result was verified on the two 2408WFPs we tested as well as with several American colleagues that use our test methods.

<< Previous page
Contrast, grays, homogeneity

Page index
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9
Next page >>
Viewing angles, movies  

Copyright © 1997- Hardware.fr SARL. All rights reserved.
Read our privacy guidelines.