Main > Driving & Racing Cabinets
CRT Vs LCD for driving cabs
twistedsymphony:
--- Quote from: BadMouth on October 01, 2014, 11:35:31 am ---
Have you gathered any notion of how the lag spec reported by the manufacturer measures up against the results they are getting?
The one in my standup cab is supposedly 5ms, not that I checked any spec other than the price when I bought it.
--- End quote ---
Most manufacturers don't ever list "input lag"... they list "response time" which is a completely different measurement.
"response time" is the amount of time it takes for one pixel to change colors, originally it was how long it took to go from black to white but since there was no standard many manufacturers list "gray to gray" time which is the amount of time it takes to change from one shade of gray to the next... obviously easier to do than black to white and thus looks better on a spec sheet.
Response time is important as a large amount of time will result in ghosting, but it completely ignores all the time it takes to perform the digital processing and decryption of the input signal before it actually reaches the display part of the display.
Most Hometheater and AV review outlets have agreed on the Leo Bodnar test as the standard for determining input lag... the way the test works it also takes into account (black to white) response time. it's a COMPLETE view of the lag in a display. It's also important to note that part of the standard is to take an average based on 3 locations on the screen, the top , center and bottom as the lag at the top is generally much shorter than the lag at the bottom. AFAIK there aren't any manufacturers who have ever listed the standard LB tested lag of their display. DisplayLag.com was started by a competitive fighting game player who's taken it upon himself to buy one of these testers and go out to the store and test every display he can get his hands on.
Input lag will ALWAYS be higher than the "response time" listed by the manufacturer. For reference that EVO monitor with 10ms of input lag per the LB test is listed as having a 2ms response time from the manufacturer. Obviously the bulk of the lag isn't in the response time but rather the digital processing of the input signal. 120Hz or 240Hz and other high refresh rate displays generally have really good response time because they have to in order to get those kinds of rates, but again it completely disregards the lag caused by digital signal processing.
I've heard a lot of people claim that using VGA as opposed to DVI/HDMI reduces lag as it tends to bypass a lot of the slower digital processing in LCDs but I've yet to see any tests to confirm this... Unfortunately the LB testers that exist are digital only.
BadMouth:
Thank you for the thorough explanation.
I was just looking at response times (thinking it was lag) vs price and hoping not to end up with a blurry mess.
twistedsymphony:
I've been promoting the ---steaming pile of meadow muffin--- out of that site since I found it. :D Since lag is really only important to gaming, and not to many people seem to know or even realize that no good lag figures exist from the manufacturer.
The guy Running Displaylag.com has done a great job but even with a few hundred displays tested it's still a pretty small list compared to whats on the market.
I've been debating buying a LB tester myself, they're not cheap but they're not unreasonably expensive either: http://www.leobodnar.com/shop/index.php?main_page=product_info&products_id=212
I'm mostly interested to see how the displays in my house stack up, and playing around with settings to see if I can effect the lag for better or worse. but at ~$115 it's right at that price where I'm not sure if it's worth it.
... of course a poor-mans lag tester is to use RockBand 2's built in lag tester, I believe the developers put a sensor in the official guitar to automatically calibrate the video and audio lag. A lot of Fighting game players would use this to determine which LCD displays were worth using.
... the other popular lag-testing technique is to have a PC outputting to VGA then split the VGA signal to both a CRT and an LCD that you're testing. The assumption being that the CRT would be essentially "lagless" then setup a timer app on the PC that display miliseconds and take a series of photographs of both displays at the same time... this would give you an image with different numbers on them so you can see how far behind the LCD was lagging compared to the CRT.
dkersten:
Interesting site, but a couple comments/opinions:
"Input Lag" as he describes it (from what I read) is the combined latency between when the image is sent and when it is displayed, so it would be the sum of the actual input lag (the time it takes for the signal to get processed and start changing pixels) plus the response time of the panel itself. And if he is actually measuring ONLY the input lag and not the overall lag, then response time of the panel will ADD to his measurements, and since each panel type can vary GREATLY in response time (which is not just affecting ghosting, it affects how fast the pixel comes on, which matters a lot if you are actually aiming at something on the screen)
Panel type: He classifies the screens as LED, IPS LED, Plasma, or 3D LED.. I am not sure why since, as I said, certain panels are going to be way faster and make a huge difference in overall latency. "LED" is the backlighting, which has nothing whatsoever to do with the panel response or the input lag. 3D screens simply are higher refresh with more processing involved, so perhaps input lag will be affected if running in a 3D mode, but you can have a number of different panel types that are "3D", and other than using the 3D capability to enable Lightboost (not available on TV's anyway), the fact that the panel can do 3D has no bearing on computer gaming latency. In general, you mostly get TN and IPS panels. TN panels have poorer color quality and worse off angle viewing, but have far faster response times than IPS panels. This is one of the many reasons "gaming" monitors are TN panels at this point, despite the TN panel's lower picture quality.
On TV's, HDMI is the typical connection, and it plugs into a processor, which is where most of the lag comes from. The "gaming" mode bypasses a lot of this, but particularly with HDMI and PC gaming, there is added processing time when dealing with HDMI. On an old CRT monitor, you already have analog RGB info in the VGA connection, a few analog adjustments, and not much else before it hits the tube, so processing is minimal in comparison. But as stated earlier in the thread, CRT tech is dead and parts are becoming scarce. LCD monitors (as opposed to TVs), still have less processing, usually using native VGA connections (or DVI, which bypasses a TON of processing). Most TV's with VGA or DVI inputs are using processors to convert it so it can go through the same processing as an HDMI goes through, then on to post processing, all before hitting the panel. So even in gaming modes there is a lot more going on. The irony is, the panels are virtually the same between a monitor and a TV, but the overall lag will be higher in most cases just because of design. A TV <> Computer monitor, but then we knew that from your first post about this.
Input lag in video/theater applications is countered easily by adjusting the audio delay that almost every decent device has. Who cares if there is even a few seconds of lag when watching a video, as long as you sync the audio and video, it doesn't matter because there is no human input for you to compare to. But for computers, audio sync isn't the factor that matters. If I were to buy the tester you are talking about, I would do it so I could measure my theater display and adjust the audio latency to match, not so I could benchmark my monitor. But that is just me I guess.
For "gaming" that is actually affected by input lag (primarily FPS games), monitors specifically for gaming are superior. The recent Nvidia based G-Sync monitors are geared toward both super low input lag (pre-processing is minimal and designed to be super fast) and super low response times (e.g. TN panels as opposed to IPS). But for quite a while now, Lightboost (which uses 3D processors in monitors to add extra frames of white in between frames to lower the ghosting effect of LCD's) has been around, and it also uses TN panels because of the superior response time (and hence less overall latency). BUT, I see that this website reviews almost no monitors, and from what I saw, no "gaming" monitors at all. This is probably because his testing equipment can't handle displayport or VGA. (This is still a good site for determining the best TV to buy for gaming with I guess, if you are looking to game on a big screen)
To me, focusing on input lag for anything other than competitive FPS gaming is like worrying about the octane rating of your gasoline in a regular car. Sure, there are some benefits that almost every person in the world will never notice unless you are using some device to measure it, but when it comes down to it, the difference is academic, and not really useful in the "real world". Even if you are playing an FPS game at home with some competitive players, you still have a 50-250 ms latency to the server, so an extra 10 ms doesn't "multiply" the effect of lag as that website says, it simply adds to it, in a negligible percentage.
BUT, if it is important to you, then spend the money to buy the Asus 27" ROG G-Sync monitor which will give you a TOTAL latency around 5-7ms (input lag plus panel response time) AND it will match the frequency to your frame rate. And if you aren't so concerned with the occasional screen tear, switch from G-Sync to LightBoost and enjoy ghost-free "CRT-Like" performance.
Personally though, $800 to play an arcade game that was originally played on a crappy CRT with phosphor glare and scan lines and lens distortion (which we like to recreate with HLSL) is overkill, but who am I to judge what is best for other people? Bottom line for me is I just finished a cab with a cheap 32" TV as the monitor, and in MAME games, I didn't see anything different than any other LCD, and it played just as well. Also, not one single game I play will stretch and distort. If it is a 4:3 game, it plays in a 4:3 window and just doesn't use the extra screen.
I guess my overall opinion is this: Do what works best for YOU. But if you are focused on one small performance aspect of a system, like input lag of the monitor, then you better focus on every other aspect of performance or you simply come across as a "snob". Get the fastest monitor you can find, but don't forget to load up the computer with an overclocked core i7, titan GPU, 512gb SSD, ultra low latency DDR3 ram, and shorter throw buttons and joysticks so you have faster response time. THEN take the time to train your body, eyes, mind, etc. to be quick enough to actually take advantage of the extra 5-10ms you gained by reducing lag. OR, just don't worry so much about a fraction of a fraction of a second when playing Pac Man. I go for not worrying about it, but again that is just me.
twistedsymphony:
--- Quote from: dkersten on October 01, 2014, 03:31:12 pm ---"Input Lag" as he describes it (from what I read) is the combined latency between when the image is sent and when it is displayed, so it would be the sum of the actual input lag (the time it takes for the signal to get processed and start changing pixels) plus the response time of the panel itself. And if he is actually measuring ONLY the input lag and not the overall lag, then response time of the panel will ADD to his measurements, and since each panel type can vary GREATLY in response time (which is not just affecting ghosting, it affects how fast the pixel comes on, which matters a lot if you are actually aiming at something on the screen)
--- End quote ---
That's not entirely correct. while you are indeed correct in how he technically defines "input lag" the listings on the site are LB Lag numbers which DO INCLUDE response time. if you look at the link to the LB tester it clearly says that it measures "input lag + response time". Considering how the lag tester works it'd be impossible for it to NOT include response time because it can only see what the dispaly outputs and the output is limited by the response time.
--- Quote from: dkersten on October 01, 2014, 03:31:12 pm ---Panel type: He classifies the screens as LED, IPS LED, Plasma, or 3D LED.. I am not sure why since, as I said, certain panels are going to be way faster and make a huge difference in overall latency. "LED" is the backlighting, which has nothing whatsoever to do with the panel response or the input lag. 3D screens simply are higher refresh with more processing involved, so perhaps input lag will be affected if running in a 3D mode, but you can have a number of different panel types that are "3D", and other than using the 3D capability to enable Lightboost (not available on TV's anyway), the fact that the panel can do 3D has no bearing on computer gaming latency. In general, you mostly get TN and IPS panels. TN panels have poorer color quality and worse off angle viewing, but have far faster response times than IPS panels. This is one of the many reasons "gaming" monitors are TN panels at this point, despite the TN panel's lower picture quality.
--- End quote ---
I'm pretty sure he only lists the screen types as a convenience for the user... if so, for instance you're specifically looking for a 3D TV then you can limit the results by that and see which 3D displays has the least lag. You're right that different display technologies generally have different response times but that's inconsequential to the lag listings in the database because, again, the response time is included in the lag numbers.
--- Quote from: dkersten on October 01, 2014, 03:31:12 pm ---On TV's, HDMI is the typical connection, and it plugs into a processor, which is where most of the lag comes from. The "gaming" mode bypasses a lot of this, but particularly with HDMI and PC gaming, there is added processing time when dealing with HDMI. On an old CRT monitor, you already have analog RGB info in the VGA connection, a few analog adjustments, and not much else before it hits the tube, so processing is minimal in comparison. But as stated earlier in the thread, CRT tech is dead and parts are becoming scarce. LCD monitors (as opposed to TVs), still have less processing, usually using native VGA connections (or DVI, which bypasses a TON of processing). Most TV's with VGA or DVI inputs are using processors to convert it so it can go through the same processing as an HDMI goes through, then on to post processing, all before hitting the panel. So even in gaming modes there is a lot more going on. The irony is, the panels are virtually the same between a monitor and a TV, but the overall lag will be higher in most cases just because of design. A TV <> Computer monitor, but then we knew that from your first post about this.
--- End quote ---
HDMI and DVI (D) are the same protocol, you can use a straight through connector to change one plug end to the other and neither the display nor the source device knows the difference. The same can be said about VGA and DVI (A) Most of the lag associated with HDMI processing comes from that fact most HDMI connections even on gaming and PCs are encrypted for copy-protection reasons (you have the MPAA along with Sony and the Blu-Ray committee to thank for that) which adds more crap that needs to be done (decrypting the signal). Technically DVI can be encrypted the same way but it was rarely used until Blu-Ray came about and they started forcing manufacturers to include encryption in order to support that drive type.
The LB Tester only supports HDMI natively, I'm not sure if it includes encryption or not. I'd love to see a VGA variant to determine how different the lag is based on the different input types.
--- Quote from: dkersten on October 01, 2014, 03:31:12 pm ---Input lag in video/theater applications is countered easily by adjusting the audio delay that almost every decent device has. Who cares if there is even a few seconds of lag when watching a video, as long as you sync the audio and video, it doesn't matter because there is no human input for you to compare to. But for computers, audio sync isn't the factor that matters. If I were to buy the tester you are talking about, I would do it so I could measure my theater display and adjust the audio latency to match, not so I could benchmark my monitor. But that is just me I guess.
--- End quote ---
I'm bit into home theater myself. I hadn't considered it's use for audio sync... that's not a bad idea.
--- Quote from: dkersten on October 01, 2014, 03:31:12 pm ---For "gaming" that is actually affected by input lag (primarily FPS games), monitors specifically for gaming are superior. The recent Nvidia based G-Sync monitors are geared toward both super low input lag (pre-processing is minimal and designed to be super fast) and super low response times (e.g. TN panels as opposed to IPS). But for quite a while now, Lightboost (which uses 3D processors in monitors to add extra frames of white in between frames to lower the ghosting effect of LCD's) has been around, and it also uses TN panels because of the superior response time (and hence less overall latency). BUT, I see that this website reviews almost no monitors, and from what I saw, no "gaming" monitors at all. This is probably because his testing equipment can't handle displayport or VGA. (This is still a good site for determining the best TV to buy for gaming with I guess, if you are looking to game on a big screen)
--- End quote ---
I'd say that Fighting Games depend on low lag as much if not more-so than FPS. there are many attack windows in those games that are literally 1-2 frames large and many people who play them competitively often watch footage frame by frame to find these windows and learn where there are openings for improving their game. Both the LB tester and DisplayLag.com were developed by members within the competitive Fighting Game Community.
The LB tester only support HDMI and the displays tested are only what the owner of Displaylag.com has been able to access and test personally. If those G-Sync displays aren't available for in-store testing or don't support HDMI then you wont seem them on that site... at least not until there is a Display port version of the LB tester.
It's also worth noting that while the FPS community prefers PC, consoles are still the preferred platform for fighting games and as of right now there aren't any consoles that support display ports or this g-sync technology. (seems very promising I'd be interested in trying it out myself).
--- Quote from: dkersten on October 01, 2014, 03:31:12 pm ---To me, focusing on input lag for anything other than competitive FPS gaming is like worrying about the octane rating of your gasoline in a regular car. Sure, there are some benefits that almost every person in the world will never notice unless you are using some device to measure it, but when it comes down to it, the difference is academic, and not really useful in the "real world". Even if you are playing an FPS game at home with some competitive players, you still have a 50-250 ms latency to the server, so an extra 10 ms doesn't "multiply" the effect of lag as that website says, it simply adds to it, in a negligible percentage.
--- End quote ---
As I said low lag is vital among competitive fighting game players. Enough such that they're always on the same hardware (not over a network) and they even use wired controllers instead of wireless.
Rythm games are another genre were timing is very important. I'm actually big into Dance Dance Revolution circles and many players avoid using LCDs altogether due to lag reasons. Even some newer arcade machines that come equipped with LCDs from the factory are swapped out for hard to find wide-screen CRTs to help improve display lag and thus improve their ability to perform better in-game.
Modern SHMUPs are another genre where timing is vital.
It seems to me that you're into FPSs and that's cool but that's not the end-all be-all of high precision gaming. There are many many places in the modern gaming world where low lag is very important.
--- Quote from: dkersten on October 01, 2014, 03:31:12 pm ---BUT, if it is important to you, then spend the money to buy the Asus 27" ROG G-Sync monitor which will give you a TOTAL latency around 5-7ms (input lag plus panel response time) AND it will match the frequency to your frame rate. And if you aren't so concerned with the occasional screen tear, switch from G-Sync to LightBoost and enjoy ghost-free "CRT-Like" performance.
Personally though, $800 to play an arcade game that was originally played on a crappy CRT with phosphor glare and scan lines and lens distortion (which we like to recreate with HLSL) is overkill, but who am I to judge what is best for other people? Bottom line for me is I just finished a cab with a cheap 32" TV as the monitor, and in MAME games, I didn't see anything different than any other LCD, and it played just as well. Also, not one single game I play will stretch and distort. If it is a 4:3 game, it plays in a 4:3 window and just doesn't use the extra screen.
--- End quote ---
As I stated in my first post about displaylag.com I found the site because I was interested in using it for Taito Type X games (http://www.system16.com/hardware.php?id=903)... If you're unfamiliar with that gaming hardware it's not a CRT arcade game, it's hardware that was designed to run in 720P on a Wide-screen LCD... and most of the games I plan on playing are Fighting Games (such as Blaz Blue and Street Fighter IV)... as well as SHMUPs (such as GigaWing Generations and Raiden IV) which is why I want a low-lag wide-screen LCD to go with it.... I don't know where you got the idea that I'll be playing "pac-man" on this screen (I'm not, if that was my interest I'd build a CRT machine for that)... I'm interested in the best possible machine to play modern HD fighting games and SHMUPs... and for that I need a low-lag LCD
You don't need to spend $800 on a bleeding etch G-Sync display along with the PC hardware to support. as you can see on the DisplayLag website there are quite a few "pretty good" monitors with less than 1 frame of lag that can be had for under $200. I only shared it because it's a pretty good guide help you select your next LCD... if you were planning on getting just a 32" whatever for your machine would you rather get the 26ms lag monitor for $200 or the 87ms lag monitor for $200? the decision at that point is pretty easy and it's an excellent resource for those kinds of decisions.
I'll probably use the site again when I go to build my Virtual Pinball machine. trying to find a low-lag 42" display is difficult and low display lag is vital when you're emulating the real world.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version