A few questions about LCDs

[Mojo]

#1 - People say that 16ms is the equivalent of 75hz. Others say it doesn't work like that, how exactly does it work, and are 16ms monitors, or lower, hard on the eyes? They don't flicker do they?

#2 - How much of a contrast ratio is needed? The monitor I'm looking at is a 450:1 19 inch LCD, is that enough? I'm not a professional graphics designer or anything.

#3 - Does a DVI connection really look better than a normal VGA connector?

 

[Synful Serenity]

There's no flicker on an LCD because the pixels are in an "always on" state, and there is only a change when the image on the screen changes. Only the part of the screen that changes will be affected, such as moving a cursor....the rest of the screen will remain static since there is no need to update the image. For the parts of the screen that are not changing, like the WIndows desktop., the refresh rate would be effectively infinite. On a CRT, the image is constantly being refreshed as a whole, so there will be flicker, and of course higher rates will cut down on this.

This makes LCD the better choice as far as text work goes, but there is a problem if the screen is being constantly updated, such as in games. The problem mainly has to do with how manufacturers rate the speeds of the LCDs. There are differing standards on how to measure them, and in most cases the rated ms is a best-case scenario estimate, and not indicative of the overall performance of the screen. Also, because there is no uniform standard for testing, some slower rated displays can outperform others that are rated higher.

In the case of 16ms, you might reason that dividing 1000ms by 16 would equate 62.5 frames a second (which would be more like 60hz, not 75hz), but that's wrong. The rated ms is not an exact number, and the response will vary according to the colors involved. Usually, the maximum response time can equal 3 to 4 times the rated ms. In the case of the Benq 8ms, it was tested and found to achieve 8ms in under 10% of the cases, with the rest of the time performing closer to 24ms, for an average around 20ms. So in an average case, you would only be able to reach an equivalent of 50fps without blurring. As I've said before, this practice misleads consumers such as yourself, and makes people think they're getting more than they are getting. Notice this will occur regardless what refresh rate you set your LCD to. If you set it to 60Hz, and the given display is only capable of running at 50Hz without blurring, if you play a game where the images exceed 50fps, they will begin to blur. You could set your LCD to 100Hz and run a game at 100fps, yet the motion would still be limited by the physical performance of the display. The result would be the same whether you ran the game at 100fps, 90fps, 80fps, or any other rate above what the LCD is capable of displaying distinctly (meaning no trace of the previous frame into the next). This also means that as long as the game stays below 50fps, regardless of the refresh rate the LCD is set to, there will not be any blurring.

The bright side to this though is that people's thresholds for noticing this varies. Some people's eyes will catch this blurring, and others won't. You might be able to run the screen fast enough to be quite higher than what it is capable of displaying without blurring, and still not notice any of the blurring. Of course you might be unlucky and notice it immediately as soon as it occurs. That's why it's good to repeat it over and over that you should look at the displays yourself to get an idea.

 

[DaLurker]

From my understanding, the 16ms stands for the time it takes for a pixel to fully turn on and fully turn off (or as they call it, rise and fall time). Generally speaking 16ms is an average, since going from a grey state to a grey state (by that I mean something between fully on and off) takes longer. What the ms second rating does help do is set a good set of benchmarks to base your buying decision on.

Now I have a 16ms panel LCD, and a 25ms panel LCD, and I notice almost no ghosting on a 16ms (if I really look for it, I can see it very faintly).

In terms of contrast ratio, generaly the higher the number, the better, but I think it's fine at 450:1.

I just went to a DVI LCD last weekend and I notice a big difference (it's the same panel LCD, just went to one with a DVI whereas previously, I had to use the d-sub). The text is sharper, the black's are blacker, and the colour's generally sharper.

 

[Synful Serenity]

While you're right, I have to disagree that it is a good set of benchmarks. 25ms displays have outperformed 16ms displays, and 16ms displays have outperformed 12ms displays. It's not a very accurate measure for comparison because it's not standard across the board. While I would tend to treat the number as more of an average than an exact number, it's not really much of an average since the rest of the performance of the display when it is not running at rated ms can vary widely. Some displays like the Benq I just wrote about have a max response time of 3 times the rated ms for 9 out of 10 cases. Other displays go as high as 4 times, and even higher. But some other displays will only have the max response time for half the cases, with the rest of the time performing better. You see what I'm getting at? The actual computed average of the screens performance has little relation to the rated ms because of all the variation. Unfortunately, until the LCD makers either adopt a uniform standard or a method of ascertaining overall relative performance, we have no choice but to go by these overly liberal estimates. Manufacturers do their best to make the numbers seem as fast as possible. I have suggested using a standard formula, such as measurements at multiple sample points computed in a way like the quarterback rating in football, or a true average like the 20ms the Benq 8ms LCD actually displays. An average such as this would be much more useful but this is not likely to happen because ratings would decrease, and people would not adjust to that easily, even though the screen's performance would be the same, with only the ratings changed.

The Ms ratings game in LCDs is a marketing gimmick. We know that CRT phosphors individually have a response time in billionths of a second, yet if a CRT was rated 16ms, it would do just that - 16ms. But now we have 25 and 40ms LCDs masquerading like 12 and 8ms displays making people think they are getting something they are not. Of course as long as people are satisfied, that's all that matters.

 

[Mojo]

I was specifically looking at two models:

-BENQ FP757 17", 12ms, 450:1, 260cd/m^2, 0.264mm pixel pitch, 1280x1024, Dsub, 140/140degrees viewing angle (horizontal/vertical), $359 Canadian.

-BENQ FP931 19", 16ms, 450:1, 250cd/m^2, 0.294mm pixel pitch, 1280x1024, DVI and Dsub, 130/130degrees viewing angle, $518 Canadian.

I'll likely end up getting the 17" model because it's cheaper, but it also has better specs. I've heard that the 16ms BENQ's are 6bit panels, meaning the color range isn't exactly 16.2 million colors and the image quality isn't too great. Don't really know if that's true or not. But the FP757's specs are better, better response time, better pixel pitch, better viewing angle, and $150 cheaper. The only downfall is it doesn't have a DVI connector. Does that really matter? Which one should I get? I probably won't buy this until February/March so there's lots of time to think.

If you can give me another monitor in the same price range that is better, on either NCIX.com, Vibecomputers.com, Canadacomputers.com, Tigerdirect.ca, or Futureshop.com I will consider them also (ie: Is the Samsung 710N any good?).

 

[DaLurker]

I was specifically looking at two models:

-BENQ FP757 17", 12ms, 450:1, 260cd/m^2, 0.264mm pixel pitch, 1280x1024, Dsub, 140/140degrees viewing angle (horizontal/vertical), $359 Canadian.

-BENQ FP931 19", 16ms, 450:1, 250cd/m^2, 0.294mm pixel pitch, 1280x1024, DVI and Dsub, 130/130degrees viewing angle, $518 Canadian.

I'll likely end up getting the 17" model because it's cheaper, but it also has better specs. I've heard that the 16ms BENQ's are 6bit panels, meaning the color range isn't exactly 16.2 million colors and the image quality isn't too great. Don't really know if that's true or not. But the FP757's specs are better, better response time, better pixel pitch, better viewing angle, and $150 cheaper. The only downfall is it doesn't have a DVI connector. Does that really matter? Which one should I get? I probably won't buy this until February/March so there's lots of time to think.

If you can give me another monitor in the same price range that is better, on either NCIX.com, Vibecomputers.com, Canadacomputers.com, Tigerdirect.ca, or Futureshop.com I will consider them also (ie: Is the Samsung 710N any good?).

Yep, 16ms BenQ panels are actually 6-bit panels, which means it uses dithering in order to obtain the 16.2 million colours, whereas in actuality, it can only genera 256bit colour... so it's simulating the other 98% of the colours ! However, I can tell you owning two of those panels that the colours are just fine. If you've read http://www.anandtech.com/displays/showdoc.aspx?i=2289 you can get a pretty good idea what's going on in the LCD world.

DVI does help, but I've been using a D-Sub Mag 776 (same panel as the FP737s) and I found that the difference wasn't that big. Basically if you went with a D-Sub, you wouldn't notice you're missing anything, but when you go DVI, you do notice the colours and the text do sharpen up a bit. If you were getting it in march/april though, I say don't worry about it for now. Prices and products will change a lot by then.

Btw too bad you didn't get in on it last weekend, but at Canada Computers (Richmond Hill Location) had a grand opening and had BenQ FP737's for $319 - $50 MIR = $269 Hehe, but if you get a 737, they're still $319 - $50 MIR = $315, which isn't a bad price at all. I remember when I first got my Mag 776 last November, I paid $499 for it!

btw while Canada Computers is great for price, their return policy isn't the greatest, so do you research before you buy from them.

While you're right, I have to disagree that it is a good set of benchmarks. 25ms displays have outperformed 16ms displays, and 16ms displays have outperformed 12ms displays.

Perhaps I phrased my response incorrectly. While I think the ms rating is a great way of measuring the performance of a LCD panels, with all the different "ways" to measure the LCD's rise and fall time, we get a lot of situations where the LCD manufacturers are fudging their numbers in order to make them look good. (Ie I think there was a situation where one manufacturer in their infinite wisdom decided to have a ersponse time of 10ms when everybody elses was 25ms. Turns out 10ms was the fall time, and not the complete cycle, so those numbers were obviously false).

So what i mean when saying measuring the rise and fall time is a good way of measuring the performance of a LCD panel (for ghosting anyways) is when everybody is on the same board, it's a good benchmark. With so many manufacturers off in their own little world though, it's hard to compare. I say the best way is to try it for yourself and go take a look.

Sorry about the confusion there.