Well, I don't know how useful this will be since it is my own internal notes. However, I have written down information about a number of different monitors based on testing done by tomshardware and X-bit Labs (http://xbitlabs.com).
I was looking for good response time (across the board) and good color. Here's the information, if it helps anyone. I know it's hard to read, since the tabs didn't transfer over. But maybe it might help someone somehow.
These first few were taken from tomshardware.com. In regards to the response time, tomshardware does not test it as accurately as http://www.xbitlabs.com/ so there are some significant errors. You can see an example of this in regards to the Samsung 970p and Viewsonic VP930 where xbit labs shows that there is indeed some response time issues that are not revealed by the the way the test is done on tomshardware.com.
Why are some things red?
I basically marked things in bright red that didn't fit what I wanted. I was looking for a monitor with true 16ms across the color gradient with only a few peaks to around 20ms allowed. When overdrive was used, I also wanted the implementation of overdrive to be good. In additon, for the tomshardware monitors, I also took into account color comments, however I didn't for the xbit Labs one.
Why 16ms you ask? Because 16.6666ms corresponds to 60fps (1000/60).
Green = pretty good
Black = maybe
Dark red = probably not
Red = no
sd = slope down
p = a few peaks
Model ... approx min-max response / normal response range .... overdrive/underdrive quality
Acer F20 19-27+ G
FP202W 15-25 / 20+ A
20GX2 10-20 / 10/17 C
8g VX2025WM 13-22 C
8g VP930 14-17 / 14 A 94% of colors rated as perfect
BenQ FP91V+ 7-31 / 10/22 C chronic difficulty LCD monitors have with dark colors.
BenQ FP91V 7-31 / 10/22 C The color rendering wasn't great.
Hyundai Q90U 7-22 / 7/15 C shouldn't serve as an excuse for ignoring color standards.
NEC 1980Fxi 16-33 / \ A 98% of the colors were good, and 95% were perfect -- more like 90
NEC 90GX2 20-25 While 90% of the colors were perfect, the 90GX2 had some difficulty in rendering the darkest colors.
4g 960BF 6-35 / 50+o15 C 90-95 in the white range
6g 970P 16-25 / C 98% The color fidelity of the 970P was absolutely remarkable. Colors were all true by default. The black depth was the best Tom's Hardware Guide has ever recorded on an LCD display.
MFM-HT75 But the result is fairly disappointing.
2ms VX922 5-10 A 96 ViewSonic has set this one a little too cold and saturated by default. bad uniformity
Xerox Xa7-192i 20+ about 90% of the colors were good.
8g VP191b 12-15
The following information was obtained from http://www.xbitlabs.com/ Note that in many cases, the response time measurements were many times more accurate than tomshardware (when the used the 3d graph). You can see an example of this in regards to the Samsung 970p.
Model approx min-max response / normal response range overdrive/underdrive error
Samsung 970p.......20-80 ...... 20......peak is at 17%
Samsung 930BF......5-10 35
Samsung 760BF 4-10-20 ..10-15
TN.Views... VX924 5-12 .. 15 .....136 contrast 100nit
TN.Views... VX724 5-15 ......10 ......172 contrast 100nit
Acer AL1916Ws 10-34 30/20sd
Acer F-19 10-36 35/25--
Asus PW191 20-35 25/35
TN.BenQ FP93GX 1.5-6 15.2% on average w/ 200% peaks from dark to little less dark
PVA.Eizo L778 10-75 15/70p none (black to dark-gray are slow)
TN.NEC 90GX2.....2-20......8......11.7%a 30% spots peaks of 70% (on light tones, when transitioning from the lighter to the darker tone) // (middle to dark&black 20ms)
NEC LCD1990SXI 35
TN.Philips 19066 20
Samsung 940T 20-100
Samsung 940B 15-35 35/22+-
Samsung 940BF 1. - 7 19.1% some 50+; mid to light 100+ peak
Samsung 960BF 1-22 8 huge errors - dark to dark peaks around 16-20 in light to light tones
Viewsonic VG920 15-35 35/20+- x
MVA.View.. VP930 10-85 15 8p ms:huge dark to dark peaks
Viewsonic's use of overdrive and underdrive was absolutly excellent in comparison to say Samsung. They didn't have as much overdrive/underdrive errors, which can basically cause smearing just like bad response times.
Your entry about the Samsung 970P is interesting....
In spite of Tom's Hardware's high marks for color and for it winning in other comparos, there is some concern about Samsung having used "6-bit + FRC" color rendering in its PVA panel.
Therefore: (a) Maybe the 6-bit vs. 8-bit issue isn't as important as some of us fear, and (b) the upgrade to the 970P, the 971P, should be a real champ! It has a supposedly better 8-bit, S-PVA panel.
Hmm... your right. That's interesting. Thanks for the tip.
The reason it was in red overall was because the response wasn't what I was looking for. Basically for it to qualify for green for me was for it to have at or below true 16ms across the color gradient with only a few peaks maybe to 20ms. In addition, the use of overdrive also had to be good. For the tomshardware ones, I also took into account color comments. I didn't for the xbit Labs one.
Why 16ms you ask? Because 16.6666ms corresponds to 60fps (1000/60), which I've heard is the supposed minimum for gaming? (or is 30fps enough?)
A very interesting thing that I did find out was the following:
Viewsonic's use of overdrive and underdrive was absolutly excellent in comparison to say Samsung. They didn't have as much overdrive/underdrive errors, which can basically cause smearing just like bad response times. I'll update the first post with some of this info.
This is actually quite useful to me. I am looking for a monitor mainly for gaming in the range of 20" to 24" and currently leaning for the Nec 20GX2.
If you get any more scores, please update us!
Well, you might have a bit of a conflict....
Nearly all of the 20" widescreens are SXGA with TN panels for speed. I think ALL of the 24" are some form of VA or IPS... slower, but better on colors and viewing angle.
The NEC 20GX is special. I believe it's the only LCD on the market right now with AS-IPS panel. And if you like the glossy, Brite screen and can control glare in the environment it's to be used in, it is SPECTACULAR!... JMO.