Input Lag of various projectors

One of Panasonic's new features is "game mode", and as I was thinking that I'd like this kind of frame rate prioritization available in my next projector (I haven't decided which yet, but I'd like to purchase in 2008) and wondered how much input lag current projectors actually have.
It got me wondering about my own projector, and I figured out an easy to test it.:

1. There's an input lag stopwatch at http://tft.vanity.dk/. Run that on a laptop/PC while outputting the same screen to your projector.
2. Take photos of the two screens. (multiple photos, since a 60 Hz refresh rate means we want multiple images to see fractions of a frame difference. Maximum ISO and shutter speed, since we don't care about grain as long as we capture a clear frame from each display.)

In my case, I used my old Sanyo Z3 at 800x600 through VGA, displaying the image in "true" mode which means no scaling. It was just a quick and dirty test and doesn't mean much, but some of the images are below.

If you can, please run this test with the same PC using different projectors, and post results here! Even if you only have one projector, the results would be helpful. If you have "game mode" or other settings, please try with different settings and see if lag changes.

Here are some results from my quick and dirty test with my Z3:



More photos, if anyone needs them, but I think those show a pretty minimal lag. (Those were the first five I took...two of the remaining four showed some level of difference between the two, but primarily the same numbers, so maybe the frame delay difference is roughly 1/3rd of a frame, while the frames are (strangly) showing a 30ms pattern, so a 1/90th of a second difference for the laptop LCD screen and the projector LCD screen?)

To be clear, since a few people have touched on it below, this does not show the image lag of the projector, only the image lag vs. the screen it's being tested on. Results with other people using different LCD/CRT/Plasma/etc. displays to test their projectors are not cross compatible. Even two people using two similar laptops may show a different image lag due to different setups. That said, I still think this is a good idea to see more details with...and I'd really like to see people with multiple projectors or different settings on the same projector post their results, as those should be useful in direct comparison with each other.

Edit:
Visited Yodobashi Camera and tested a few other projectors:
AE200 - Game mode
AE200 - Normal Mode
TH-AE2000
TW-1000 (Epson 1080)

My thoughts: This is still apples and oranges in a sense. I'm interested in trying to do the same thing with the AE3000 and TW4000 (Epson 6500/7500) when they come out. What surprises me is that the Panasonics clearly seemed to have the worst image lag, which may be why they're the one marketing their new "game mode". The Sanyo and Epson both seem to have pretty similar performance to the AE200's game mode performance.

If it the Reon processor and/or other image processing options in the TW4000 can be turned off, it will be interesting to see whether or not the TW4000 has can reach the same minimal image lag as game mode on the AE3000. Nothing too super scientific, but interesting and fun, and I'm looking forward to hopefully doing the same with the new models as soon as possible.

(While unrelated to image lag, I was very unimpressed with how much dust the AE2000 had. When focusing to get the screen as close as possible, I was kind of shocked with the amount of dust in a 10 month old projector, and unlike the Sanyo, there's no way for a user to mitigate this...)

Probably worth mentioning, two things I'm very interested in:

• What is the lag of the AE3000 with "game mode" on, and with game mode off?

• Will the cheaper 1080p Epson (TW3000/6100/7100) have less lag than the more expensive one (TW4000/TW5000/6500/7500)? The cheaper model has inorganic panels, while the more expensive one has inorganic panels and a Silicon Optix HQV Reon-VX coupled with PixelWorks 390 video processor. Will the panels and/or amount of processing done by the Reon-VX slow things down?

I would say that I do not notice any significant input lag with this...but it's been my primary gaming display for nearly four years, and I've gotten quite used to it. Hooking a second Xbox 360 up to my 40" Sony Bravia I bought less than a year ago for some local network gaming with friends, I did not notice any difference in lag, although both are LCD based displays.

As I mentioned in my post, this is meant to be "quick and dirty". I don't have a CRT display, and I'm not about to lug a PC into that room just to use one of my two LCD displays instead of the LCD monitor on the laptop. If anyone is seriously considering buying a Sanyo Z3, I hope they don't take these tests too seriously.

The main reason I wanted to get these photos done now is that when I do buy my new projector, I can use the exact same program and laptop to test the other projector, and see if there's a discernable difference or not. (Comparing two people's tests could easily be apples and oranges, especially if people are using CRT displays)

I'll post more photos of the same type of setup with my new projector when I get it (still not sure if it's the Panny or Epson) but I'm looking forward to hopefully seeing some other people post some results of their own setups between now and then.

You don't/can't. You can exchange hardware specs and try to figure out how differing systems compare, but basically, it's only really useful when using the same setup to test two different projectors or two different settings on the same projector.

That said, I'll be using the same laptop and settings to test my new projector, at which point comparing the results with my first post may be interesting for some. Surely I'm not the only person on these boards to have more than one projector in their house at some point.

An Xbox 360 and PS3 can send 720/1080 feeds directly to the projector through a variety of methods, including HDMI, and I would not use a switch for my gaming console, specifically to avoid image lag.

I definitely get what you're saying about lag varying from scene to scene, but I can't think of a great way to test image lag with a constant video load. I thought about running the stopwatch windowed, with another video running, but that just introduces more variables that would vary when I do the test again with my new projector. If you have a better idea, please share it.

Is there a way to turn off non-critical in-projector video processing with Epsons? I'm assuming that's what the "Selective Frame Response" of the Panasonics ("game mode") do, but I'd like to see something similar to this with and without that option turned on to see if it shows a difference.

So I figured the only meaningful comparison photos to my original post would have to come from an identical setup, so I'd have to be the one to take the photos. I decided to head down to Yodobashi Camera (where photo taking is actually expressly prohibited...) with my laptop and VGA cable and decided to try the same setup with all their 1080p projectors. Unfortunately, the ceiling mounted projector was out of the question, and the Z2000 and TW2000 (1080UB) still both had dead bulbs. (The Z2000 was dead when I first visited a few weeks ago.) That meant I could test the AE2000, the Epson TW1000 (1080), as well as the Panasonic TH-AX200, which, while 720p, has game mode like the up and coming AE-3000.

All the photos were taken in "A" (Aperture mode) with my 50mm f/1.4 wide open at ISO 1600 to capture these images with as little exposure time per frame as possible. These results are not scientific, I am not a professional at this, please consult your physician about what other alternatives to image lag may be appropriate for you, etc. etc.

AE200 - Game mode

AE200 - Normal mode


(Just before switching from Game (bottom) to Normal (top).)

Updating the first post with my thoughts, but quickly here:

• I thought it was surprising that the Epson (1080) seemed as good in normal mode as the Panasonic performed in game mode.
• The two Panasonics clearly had the most image lag of the four projectors I've tested so far. (AE200, AE2000, TW2000 and Sanyo Z3)

(While unrelated to image lag, I was very unimpressed with how much dust the AE2000 had. When focusing to get the screen as close as possible, I was kind of shocked with the amount of dust in a 10 month old projector, and unlike the Sanyo, there's no way for a user to mitigate this...)

The TW2000 is the 1080UB. Unfortunately, that's the projector I couldn't test. (dead bulb)

The TW1000, the 1080 in the US, is the one I was able to take photos of...and while it's quite possible the 1080UB could have the same results, it's also possible that it wouldn't. I'll probably go back with my camera and laptop once I've confirmed that the AE3000 or TW4000 is on display (maybe even the TW3000 [6100/7100]) but not until then, so I don't know when or if I'll get photos of the 1080UB.

Until then, I'll keep my fingers crossed that someone else with a camera and access to more than one projector might be able to share some of their results with us.
Hahaha. Replied just before your edit. Thanks for your complement? (I think.) Like you, I'm really hopeful the TW2000 does have the same, or better, image lag than the TW1000.

Seems to be a limitation of my "ultralight" laptop. It's one of those really light 12.1" Dell ones, and even though it's rather recent, I think the graphics chipset just isn't up to snuff. Hooking it up to another monitor and running power saver mode or high performance mode doesn't make a difference. My older, larger notebook runs the stopwatch much more smoothly. I see the difference very clearly now that it's been pointed out to me.

My small laptop claims 60 Hz, but I think you've pointed out a pretty big discrepency with the 30ms.

Sadly, there's not a lot I can do to change the setup I'm using (not to mention, I don't want to test my luck snapping too many photos in Yodobashi Camera) so there won't be 16ms frame results from me, but I'm hoping someone else can chime in.

At the very least, the results gave me more of a reference point for LCD projector image lag than I had previously...hopefully they're a starting point for others to contribute more to.

No CRT to use. Always using VGA.

As you may already know, scaling 480i material is much more time intensive (more image lag) than 720p/1080p material, for any LCD screen or projector. There's a thread about that here, and that's definitely not a front projector specific issue.

Sending 720p material to my LCD TV or projector, I use component or HDMI, depending on the source, but I have no way of sending a time clock or similar image lag test to two monitors using anything but VGA, which is why I used VGA for my quick and dirty tests.

FWIW, I don't find the Z3 to be noticeably laggy when fed 720p/1080p, but as I mentioned earlier, image lag is one of those things people subconsciously adapt to, and I've been gaming on my Z3 for years.

Edit: If someone could prove that there's less image lag over VGA than both HDMI and component with my next projector, I'd be very happy to hook up my Xbox 360 over VGA, move my HD DVD player into the projector room to play HD DVDs and DVDs over HDMI. (Move the HD DVD drive to a second Xbox for the Bravia LCD TV in the other room for HD DVDs in that room.)