LCD + gaming = okay?

[Kingofcomputer]

Wonderful articles. A shame have yet to prove you understand a damn one of them.

Shame on you. You don't understand a damn one of them.

 

[JKing76]

Shame on you. You don't understand a damn one of them.

No, you have no idea what you're talking about. That has been demostrated several times in these FFD threads.

 

[Mr.IncrediblyBored]

Rearranging the wording of my post doesn't somehow make your comments any less bullshit. Let's try this again.
From the FFD article:

A new 15-in. XGA TFT-LCD with fast response time is developed. We adopt Feedforward Driving (FFD) technology in which acceleration voltage is applied according to the terget of response time improvement. The acceleration voltage is selected so as to reach necessary luminance in one frame period. The response time between grayscale transition is reduced to less than 20ms by FFD. The advantage of the technology compared with the other fast response LC mode is easy to apply to the current product without any disadvantage related to new process and design. High brightness, high color saturation, and wide viewing angle characteristics are also achieved, which are significant for multimedia applications.

From the MVA article:

In the MVA process, LC molecules are angled in more than one direction in a single cell. The cell is divided into two or more regions, called domains, and uses protrusions on the glass surfaces to pre-tilt the molecules in the desired direction. By combining areas of molecules oriented in opposing directions, and by making the areas very small, the brightness of the cells appear uniform over a wide range of viewing angles.

From the PVA article:

Using the PVA technology, we have developed 24 inch diagonal WUXGa TFT-LCD monitors for the first time. The electro-optical properties of the panel after optimizing the cell parameters, ITO pattern shape, and the compensation film demontrate that this mode is a viable technology for improving viewing angle characteristics without sacrificing other electro-optical properties such as panel transmittance, contrast ratio and the response time.

 

[Kingofcomputer]

Let's see how's this FFD hype soon.

 

[NFS4]

I have a Planar 18.1" LCD screen running DVI on my 128MB GeForce4 Ti 4200 overclocked 275/500. I've been using it for a year now and I absolutely LOVE IT!!!!

- I have no problems with ghosting with the games I play (RTCW, Battlefield 1942, UT2003, Need for Speed: Hot Pursuit 2 demo, etc).
- I run all my games at 1280x1024 either with or without FSAA...depends on the game. No problems, and it looks great
- I have a LOT more desk space now. My 21" CRT monitor took up all of my desktop and I had hardly any room for anything. Now I can push my LCD back and have room for tons of stuff (PDA cradle, USB hub, USB card reader, gamepad, telephone, etc)
- My LCD has 4 USB 1.1 ports on the back...nice touch for my many USB devices
- It is EXCELLENT with DVD's. No ghosting.
- I now have NO MORE eye strain. That was the BIGGEST problem with my CRT monitor. My eyes would start burning after about an hour. Not anymore
- Refresh rates really don't matter as much (if at all) with LCD's so I don't see that as being a "plus" for CRT's.

For all of you LCD bashers that say that it is no good for gamers; get off your high horse. Stop quoting specs or what you THINK you know. Don't knock it til you've tried it.

 

[Mr.IncrediblyBored]

Originally posted by: Kingofcomputer
Let's see how's this FFD hype soon.

Of course, when asked to show proof, he pulls the "let's wait and see" bit.

 

[Kingofcomputer]

FFD has been introduced 1 year ago, as it claims just need to do slight modification, then how come today there is still no such product?

If this is Mitsubishi's killer weapon, how come NEC-Mitsubishi still hasn't released this product and let Envision such low-end cheap monitor manufacturer to have it first?

 

[Maverick]

what about running in non-native resolution? I don't want to be stuck at 1280x1024 when the world moves on to 1600x1200 and beyond. My CRT has been able to scale very nicely all the way from 640 to 1600...with LCD are you pretty much chained to the native resolution? I know the monitors are capable of higher res than their native, but it looks like crap, doesn't it? Or does this vary between different monitors?

 

[Mr.IncrediblyBored]

Originally posted by: Kingofcomputer
FFD has been introduced 1 year ago, as it claims just need to do slight modification, then how come today there is still no such product?

If this is Mitsubishi's killer weapon, how come NEC-Mitsubishi still hasn't released this product and let Envision such low-end cheap monitor manufacturer to have it first?

Economies of scale, my friend. Introducing FFD on high-volume panels will help lower the cost of the added circuitry. Generally speaking, less expensive models have much higher volumes.

 

[Mr.IncrediblyBored]

Originally posted by: Shiva112
what about running in non-native resolution? I don't want to be stuck at 1280x1024 when the world moves on to 1600x1200 and beyond. My CRT has been able to scale very nicely all the way from 640 to 1600...with LCD are you pretty much chained to the native resolution? I know the monitors are capable of higher res than their native, but it looks like crap, doesn't it? Or does this vary between different monitors?

The native resolution is the max the screen can display.

With LCD's, you're dealing with physical pixels on the screen. Non-native resolutions look bad because you're essentially stretching a lower resolution image overtop a higher resolution screen. This "stretching" process isn't perfect, and never will be.

 

[dszd0g]

Originally posted by: NFS4

For all of you LCD bashers that say that it is no good for gamers; get off your high horse. Stop quoting specs or what you THINK you know. Don't knock it til you've tried it.

NFS4, you are generally above such comments. Please don't start now. As is the case with many things of this nature, what is noticeable to one person may not be noticeable to another. If you say that you do not notice any ghosting, I believe you. Ghosting may not be apparent to you. However, it is quite noticeable to many people. Most good LCD reviews have a section covering ghosting and how bad it is on that LCD, I have yet to see these sections start saying "none" from a site I trust. I have tried it myself and I do notice it.

It has been quite noticeable on any LCD screen I have used, but I do not claim to have looked at the top of the line current LCDs.

While, I agree that the refresh rate of a monitor is not the same thing as the pixel response time of an LCD. There is a reason it is under a different name. However, the calculation I gave does give how many times a pixel on an LCD can change per second (thus the Hz). If the pixels on the LCD can't keep up with the game one is playing, then ghosting is going to occur. If one has a 25ms pixel response time LCD and one's system is only putting out 40fps in the game, then ghosting will most likely not be visible. If on that same system, one plays a game at 90fps then ghosting will be quite apparent to some people. Maybe that is one reason you may not notice it (I am trying to be as polite as possible in this response).
[Edit: However, even with a very slow pixel response time an LCD will not flicker like a CRT does. Thus, most people do not experience eye strain as quickly on an LCD screen. I personally can sit in front of an 85Hz CRT for many hours at a time, but a 60Hz CRT I can't stand for very long. However, I do believe that the eye strain of an LCD is still less than a high refresh rate CRT.]

I am quite positive that pixel response time is one of those things that is more important to some people than others. It is important to me, claiming that I am on a high horse because I care about something that you don't care about is just flaming.

 

[NFS4]

Originally posted by: dszd0g

Originally posted by: NFS4

For all of you LCD bashers that say that it is no good for gamers; get off your high horse. Stop quoting specs or what you THINK you know. Don't knock it til you've tried it.

NFS4, you are generally above such comments. Please don't start now. As is the case with many things of this nature, what is noticeable to one person may not be noticeable to another. If you say that you do not notice any ghosting, I believe you. Ghosting may not be apparent to you. However, it is quite noticeable to many people. Most good LCD reviews have a section covering ghosting and how bad it is on that LCD, I have yet to see these sections start saying "none" from a site I trust. I have tried it myself and I do notice it.

It has been quite noticeable on any LCD screen I have used, but I do not claim to have looked at the top of the line current LCDs.

While, I agree that the refresh rate of a monitor is not the same thing as the pixel response time of an LCD. There is a reason it is under a different name. However, the calculation I gave does give how many times a pixel on an LCD can change per second (thus the Hz). If the pixels on the LCD can't keep up with the game one is playing, then ghosting is going to occur. If one has a 25ms pixel response time LCD and one's system is only putting out 40fps in the game, then ghosting will most likely not be visible. If on that same system, one plays a game at 90fps then ghosting will be quite apparent to some people. Maybe that is one reason you may not notice it (I am trying to be as polite as possible in this response).
[Edit: However, even with a very slow pixel response time an LCD will not flicker like a CRT does. Thus, most people do not experience eye strain as quickly on an LCD screen. I personally can sit in front of an 85Hz CRT for many hours at a time, but a 60Hz CRT I can't stand for very long. However, I do believe that the eye strain of an LCD is still less than a high refresh rate CRT.]

I am quite positive that pixel response time is one of those things that is more important to some people than others. It is important to me, claiming that I am on a high horse because I care about something that you don't care about is just flaming.

You're missing the point You guys are all basing this on reviews and calculations and all the planets being in alignment or WHATEVER! It means nothing until you get down and look at it for yourself. Go and look at the monitors. Witness it for yourself. Monitor purchasing is not about reading reviews IMHO.

Purchasing a monitor is different from purchasing a processor. Specs are only a part of the equation in monitor selection, b/c there is that "human factor" that comes into play as you look from product to product.

Purchasing something like a processor is more of a strictly by the specs/#'s routine. There's not much "leeway" in the decision. Either it does or it doesn't.

 

[Coherence]

I have a Sharp LL-T1820, and I love it. It boasts a 25ms (40hz/40fps?) pixel response, and lots of other great new technologies in it. Is ghosting a problem. Not for me. Do I notice ghosting? Yes, a little, but not enough to worry about. It can depend on things like contrast of the pixels in motion compared with the pixels around the object.

Does it bother me? Nope. I love this monitor, and think it was well worth the "Perfect-10 Kick Ass" award given to it by Maximum PC magazine. It's EXPENSIVE, though. (It was an impulse buy.)

One thing to be aware of, many companies may boast a low pixel response time, but do your research carefully, as some will only tell you the "rise time" of the pixel, and not the "fall time". Rise time is how long it takes the pixel to draw a value from a zero-state. Fall time is how long it takes to go back to the zero-state. It's the fall time that causes most of the ghosting problems, particularly if the fall time is high.

Good LCD companies list the response time as the average rise-fall time. Not-so-honest ones will only list the rise time, making you think you're getting a 20ms pixel response when it's really more like 40ms. In other words, the bad companies interpret "response" as meaning only how long it takes the pixel to "respond" to the draw signal (rise), and ah heck the fall time.

 

[Kingofcomputer]

response time can be translated to fps, but not Hz.

even in CRT, for example, you're playing a 60 fps game at 100Hz refresh rate.
can you say you're playing the game at 60 Hz?

 

[Kingofcomputer]

ttp://www.hitachidisplays.com/press/CML174_pressrelease.htm
Hitachi's new 17", 12+4 ms response time, is this using FFD?

 

[Mr.IncrediblyBored]

Originally posted by: Kingofcomputer
response time can be translated to fps, but not Hz.

Response time refers to the time it takes for the rise and fall of a color value on the screen. This "rise and fall" process is referred to as a cycle, and cycles are generally measured in hertz, not frames per second.

even in CRT, for example, you're playing a 60 fps game at 100Hz refresh rate.
can you say you're playing the game at 60 Hz?

No, you don't, but the framerate of a video game has little to do with the monitor refresh rate. By saying response time = FPS, you're implying that LCD response time goes up as framerate goes down, which is BS.

Originally posted by: Kingofcomputer
ttp://www.hitachidisplays.com/press/CML174_pressrelease.htm
Hitachi's new 17", 12+4 ms response time, is this using FFD?

Corrected link.

 

[Kingofcomputer]

Ok, the unit for cycle is Hz, I agree, but this is not case for LCD response time.

No, you don't, but the framerate of a video game has little to do with the monitor refresh rate.

Funny, this is what I was telling you.

By saying response time = FPS, you're implying that LCD response time goes up as framerate goes down, which is BS.

I didn't say LCD response time goes up as framerate goes down. This is your own BS.
It's you don't understand the simple 25ms = 40fps concept.
25ms LCD doesn't mean when it's showing a 24 fps movie, the response time will down to 41.6ms.

If you're playing a game fixed at 60fps on a 25ms LCD, you're missing 1 frame in every 3 frames, which is still okay for gaming, as 24 fps is already smooth to human eyes.

Now I've answered the thread's question: LCD + gaming = okay?: YES on current 25ms LCD, you don't need FFD.

 

[adhoc]

Relating response time of LCDs to Hz and fps is not as simple as it may seem. There are a lot of other factors such as separate rise and fall times, latency, and bandwidth; but here is a simpler situation, as I see it...

Given that an LCD has an absolute 25ms response time:

(1 chagne)/(25 ms) * (1000 ms)/(1 s) = (40 changes)/(second)

This value, unfortunately, is an average because the actual response time of a pixel is based on the previous color of that pixel. Therefore, response time not always 25 ms. However, for the purpose of an example, a static value of 25 ms is respectable. It should be known that, for every pixel on the screen, the response time for a pixel that already holds the new color is effectively zero. It should also be known that 40 changes/sec is not the rate at which data is entering the LCD, but rather data enters on a 165MHz (not sure) clocked bus if using DVI...

This means, that if one wants the LCD to display EVERY color change on EVERY pixel, then the maximum frequency that these pixels can change is 40Hz.

Now, if an LCD can handle 40 color changes/second PER PIXEL as above, that means if one wants to convert that number to FPS, we first have to define the meaning of what a frame is for an LCD.

For example (and remembering to take into account an absolute response time of 25ms regardless of color, except when a new pixel color is the same as the old, making the response time 0), if the fastest pixel changing color is 40 changes/sec , yet the computer is outputting 200fps, each and every color change will be accurate from the LCD at "200fps". Say, then, for example that the computer is still running at 200fps but the fastest pixel changing is at 80 changes(colors)/sec, this means that a "frame" is dropped every other pixel change (if we regard a frame output as an exact representation of all the pixels during that frame), we have "100fps". Using this logic, if the fastest pixel changing is changed EVERY FRAME, then the maximum allowable FPS is equal to 40 changes/second.

These definitions are hazy, but I'm trying to prove a point that response time (being dymamic) cannot be directly related to FPS nor Hz, becaues there are other factors involved. If I'm unclear or totally off-base, do not hesitate to let me know!!

Hopefully this sheds some light on the LCD response times...

*EDITED* : A few minor grammatical errors.

 

[Coherence]

Originally posted by: adhoc
For example (and remembering to take into account an absolute response time of 25ms regardless of color, except when a new pixel color is the same as the old, making the response time 0), if the fastest pixel changing color is 40 changes/sec , yet the computer is outputting 200fps, each and every color change will be output of the LCD or "200fps". Say, then, for example that the computer is still running at 200fps but the fastest pixel changing is at 80 pixels/sec, this means that a "frame" is dropped every other pixel change (if we regard a frame output as an exact representation of all the pixels during that frame), we have "100fps". Using this logic, if the fastest pixel changing is changed EVERY FRAME, then the maximum allowable FPS is equal to 40 changes/second.

Adhoc makes a good point. But, this is not just the case with LCD monitors. Keep in mind, just as with LCDs, if you are running a game at 200fps, but your CRT monitor runs at 85hz, you're going to "lose" those extra frames the CPU is drawing, because the CRT can't keep up.

 

[dszd0g]

Adhoc has a more thorough explanation. However, as he said, it isn't that simple. If one really wanted to do it right one would have to draw a timing diagram with rise time, fall times, propagation delays, and all that fun stuff to determine what the cycles really are. One would also have to know what the clock runs at internally to the LCD (I bet there is one, as most electronics these days are synchronous to a clock). However, it is more complicated than that because as has been mentioned, not all the pixels change every cycle.

KingofComputer, I was doing a rough worst case analysis of how many times an LCD can change per second. Changes/Cycles per second are measured in Hz, by definition. Hz is a much better comparison between the two than FPS, because FPS has to factor in what the software can actually do. I would prefer sticking to the hardware side of things for this comparison.

From WordNet (r) 1.7 :

Hertz
n 1: the unit of frequency; one Hertz has a periodic interval of
one second [syn: Hertz, Hz, cycle per second, cycles/second,
cps, cycle]

Kingofcomputer, mind doing some reading before you keep telling people they are wrong?

 

[Ilmater]

Damnit people, your arguements are all circular. First off, you are correct in that Hz would be a better measure of what response time means to an LCD in relation to a CRT. However, if the game you are playing on is running at exactly 40fps and the monitor is running at exactly a 25ms response time (I know that this wouldn't happen, but bear with me), then you should see no ghosting, as the monitor would be albe to shift in time with the screen. However, if the game is running at 60fps, given the same response time, you would see some ghosting. Of course, this won't happen all the time, but this is the reason that it is not completely inaccurate to talk about a relationship between response time and fps. Anyone want to disagree? Is that statement incorrect. No on both instances.

Now, as for FFD, how can anyone say that this is just hype. It makes complete sense. The only way that it wouldn't work the way that say it should would be if the technology itself isn't doing what they say it is. Let's breakt he tech down as simple as possible: the liquid crystal changes colors based on the voltage that it is given. That's a fact. Currently, switching from totally on to totally off (or highest voltage to no voltage, however you want to say it) is the fastest change from one color to the next. That's also a fact. The reason for this is that the delta between the voltages is higher. In other words, instead of 2 guys pushing a car up a hill, there are 5, and they move the car faster. That's a fact AND it makes complete sense. My 13-year-old sister can understand that concept. So, if you apply the highest voltage (or no voltage) to get it to switch colors as fast as possible, and then switch to the appropriate voltage (sustaining voltage as it is called) for the desired color when it is reached, then you get the fastest possible times from one color to the next. Not only does everything in that technology make complete, logical sense, but I haven't explained that to a single person yet that hasn't said that it sounds brilliant, simple and understandable at the same time.

Therefore, unless you are trying to say that you don't think that NEC and Mitsubishi have a technology like that - that they are in fact lying - then you must understand that you anyone that says that FFD is only hype is utterly wrong.

 

[adhoc]

Ghosting really doesn't have anything to do with how fast screen data is being updated by the computer. Ghosting is due mostly to the response time of the LCD screen, or the quality of phosphors in a CRT. You see ghosting when your eyes notice a smearing effect, due to the pixel response of the display being low enough for the eyes to see. Ghosting is present on every computer screen, TV, etc. Fortuntaly though for most CRTs, the pixel response time is fast enough that the human eye does not notice the ghosting, or color cycling.

For CRTs, the Hz rating (refresh rate) is the rate at which the electron beam passes over a pixel, NOT the pixel's response time. If you don't believe me, run your monitor at abour 40Hz (if it is possible); all you will notice is that the screen will be visibly blinking at you, because the phosphors are "turning off" before the electron beam gets back to refresh their electron states, or luminescence. Ghosting will not be present, however, because you have not changed the pixel response time...

Your computer can be running at 50, 60, 100, or 200fps (regardless of refresh rate or response time) and all will exhibit the same degree of ghosting (whether or not you can notice it), but if you do it is because the response of the pixels/phosphors are slow and your eyes can notice the changes that are not "instant".

This is how the old oscilliscopes worked; they used phosphors that had a high luminance time interval, which resulted in the effect of ghosting and the presence of a continuous waveform on the screen.

 

[Kingofcomputer]

I said it's hype because I saw so many people got too excited on this new technology.
I don't mean it doesn't exist as I provided the Mitsubishi's FFD doc and I found Mitsubishi had showed off such LCD in the PDP show in Japan last nov.
Other manufacturers also keep doing r&d on their own technology to improve LCD.
Everybody has a 25ms or better LCD using their own technology already.
FFD is not a thing that everybody must use.

For worst case, 25ms (typical) already can produce 40 fps display in average, which is more than enough for movie, tv, gaming.
And you're even getting faster than 25ms in some situations.

Keep in mind, we're talking about a-si TFT-LCD here, using TN+(add FFD or not), MVA, PVA, IPS or ASV, it's still a-si TFT-LCD.
p-si TFT-LCD is another story.

And the new Hitachi 17" 18ms is using Hitachi's own Super-IPS technology.

 

[subhuman]

I totally agree with NFS4 for once - don't let the specs decide it for you - actually spend 8 hours in front of your CRT, then 8 hours in front of an LCD (because an LCD does take a little getting used to), then tell me which one gives you less of a headache...

And I think Kingofcomputer has some very valid points, no need to bash him...

 

[NFS4]

Originally posted by: subhuman
I totally agree with NFS4 for once - don't let the specs decide it for you - actually spend 8 hours in front of your CRT, then 8 hours in front of an LCD (because an LCD does take a little getting used to), then tell me which one gives you less of a headache...

And I think Kingofcomputer has some very valid points, no need to bash him...

It's crazy in here, ain't it? These guys are spouting off numbers back and forth and try to sound "smarter" than the next person by seeing who can throw throw around the most technological terms and numbers. That's not the issue here!!

When you buy clothes, do you care about what percentage of the material is made up of lion's fur or what percentage of blue die is used in the process? Most likely not. But what do you do; you try it on to see how it feels. Do you buy a pair of shoes without first trying them on??

Same thing with a test drive, you can quote all the specs you want, but that test drive makes all the difference.

The point is, stop trying to turn this into a battle of who's smarter...in this case, it's not your booksense or computer knowledge that determines the better product, it's your eyes that determine that.