What can I use to determine each rooms Video Delay Offset?

[kd_special]

I remember reading a post of an Item called SYncII or something like that.

It was a box that would measure the video offset for video playback. We have a bunch of studios and I would love to nail it down on all of em. If the box is affordable. mostly use Mojos and canopus boxes but every TV/projector is slightly different it seems.

thanks

kd

[dopplah]

http://www.syncheck.com/

[kd_special]

thats it thanks mang...

kd

[garnoil]

I am probably going to get killed for saying this but.... eyes and ears are just as good. I own syncheck, and yes, it is good to have BUT....if you do not own it, trust your eyes and your ears. Myself and surprisingly a lot of my clients (directors of film for the most part) can easily spot a 1/2 frame sync error (sometimes even as little as a 1/4 frame make things feel odd). So, since I do have syncheck and calibrate< i often find out that I can do just as well by eye. This is the key: you have to make sure that you have an audio and video (tone and flash) correctly lined up in PT. If you are 100% sure that video and audio are playing exactly at the same time in PT, you can easily do it by eye. Use up to 10 seconds of tone/flash, then just play it and adjust as needed. You will be surprised how accurate you can be, in fact, many time in film, one has to match sound to video without a reference at all and if you can do that...syncheck is just a luxury.

NOTE: PT (if not fully resolved) always has a time sync error, up to a frame but often less, this is because PT does not align to the frame edge without clock resolution so....you are starting with a sync error anyway, as much as one frame.

[kd_special]

thanks for the info. I really just want syncheck because we have close to 15 rooms including editsuites that all need to have decent accuracy. Syncheck will do two things for me.

Save me some time. And make sure each room is video offset sync wise identical. No matter what boxes/Tv's/Projector etc.

At least I hope it will

But for now I will set up a 2 beep and try by eye and see how it goes.

kd

[Vedat]

Originally Posted by garnoil

Myself and surprisingly a lot of my clients (directors of film for the most part) can easily spot a 1/2 frame sync error (sometimes even as little as a 1/4 frame make things feel odd)

1/4 frame is roughly 8ms in NTSC.
That's how long it takes sound waves to travel 8-9ft.
So, you can spot this difference if you move 8ft away from your mix position?

Just curious.

[MIKEROPHONICS]

Personally a frame is all I reckon most people can see - members of the public not withstanding who seem remarkably tolerant of any delay, even video delay which is really hideous.

I have never understood either why my SKY HD box in my lounge is late (via phonos) compared to the HDMI LCD sound, Na dbefore you say, I think the box in built delay is only for the optical stuff (which sadly due not in use due to my wife banning all those speakers from the lounge apart from my BBC LS35a Left and right). NO I do not intend on a wife upgrade!

cheers

[garnoil]

Originally Posted by Mundoz

1/4 frame is roughly 8ms in NTSC.
That's how long it takes sound waves to travel 8-9ft.
So, you can spot this difference if you move 8ft away from your mix position?

Just curious.

You can certainly hear the difference between 8 or 9 feet away compared to zero feet away. In fact, in most films where the boom is that far away but also have a lav, you absolutely hear the difference and even see the time difference on the Pro-Tools (or an RTA) waveform display. A 1/4 frame, is indeed a small amount of time and it is hard to see but it may not "feel" right. A frame is an eternity in audio. Of course, you have to have the right conditions. For example if you are trying to see a frame out error when the actor says a word like "analog" it will be difficult, but if you have the word "popper" or "pantomime" the plosives will give it away. And that is what most of the psychoacoustic research shows. The time integration of the ear is somewhere around 10 ms for percussive sounds but can be up to about 30 seconds for non-percussive sounds with a varying rising rising envelope. Stereophonic recording uses this same principle to produce "stereo", although stereo recording also utilizes other characteristics of sound behavior such as, how humans interpret sound source localization based on frequency, volume, Doppler, Has, phase coherence, and others and NOT only on time arrival of wave fronts. Another thing to consider is how humans interpret distance in relationship to both sound and image. For example, on a long shot, say...20 feet in the distance, where the actors shoots a gun it is perfectly "normal" to have the sound arriving the listener "late-er", not necessarily the 20 milliseconds it would take the sound to reach the listener in real life (because movies are not real) but if the sound is 10 milliseconds off (has a 1/4 frame error), as long as the sound is not earlier than the visual, the shot will feel right, the scene will work. This is why sometimes perfect sound-picture synchronism may not "feel" actually correct. On the other hand, the sound of a single definite event such as a clash of swords, right on screen, can not be 8 milliseconds late or it will "feel" off.

So, in my opinion, although 1/4 frame is indeed a very small amount to actually see "as being off-sync", it can be enough to make an audio event "feel" not quite right. A full frame off sync is plain obvious especially on a shot like a ball striking a bat (right on screen) especially if the sound is earlier that the video frame. Consider this: no one knows where exactly -in the video frame- the bat strikes the ball. If the ball strikes the bat at the later part of the frame (meaning at the ending edge of the frame just before the video frame changes to the next frame) and you place the sound 1 frame early (in the previous video frame from where the bat strikes the ball), you actually have almost a 2 frame sync error. The error is: 1 frame that we know we are early + another "almost" full frame because the bat strikes the ball at the very last part of its video frame. This would be a 2 frame error right on screen, it will look and feel way off.

[nucelar]

Oh dear....
Ok, so how do we calibrate sync from now on? at a certain nominal viewing distance or at zero distance?
I don't know if I'm joking or not..

[garnoil]

It is never really about anything other than what "feels" and "looks" right. Years ago, this sort of thing was not really an issue because most systems used for audio post, were synchronized via tc code with address and fully resolved. Also, latency issues were all but non-existent. Monitors as in "multi res pro monitors" had extremely accurate timing responses and so did the clocking used to derive sync for all machines involved. It was also a fact that video clock -derived from a common house- usually, if not always, drove "was the ultimate master" audio clock, so it the video changed or drifted the audio would follow. A simple mechanical system consisting of clacker, crystal sync, and eventually the sprockets kept tight sync and with none of the latency of DAW or LCDs or other digital equipment.

When this technology was ported to digital, there were many issues that were not either fully resolved or people do not know how to deal with. How many times does one get a music sound track for a film that has been overly compressed "squashed with minimal dynamic range" and mastered to peak at 0dbfs? This is because a lot of people simply do not understand the relationship and conversion between analog and digital levels. They think that 0dbfs is "good for music, nice and hot", without realizing that they are actually producing material that is 20dbfs louder than the nominal level corresponding to the 0VU standard of a +4dbm aligned system. And then they expect this to "mix nicely UNDER the dialogue of the film"? This, for example, could not have happened in the analog days...one could never record a level of 20 db above the nominal 0VU meter (steady state tone)...the VU would probably disintegrate and the tape would just be a mass of distortion and become permanently damage as the oxide would never re-align at this level.

In my opinion: one needs a benchmark to compare to and therefore one has to know or trust the latency response/issue of the visual monitor as well as the latency PCI chain. So in this sense Syncheck is good but, if one plugged an analog video monitor (a good one like the old pro Sonny), and Pro-Tools was fully clocked resolved to align to the video edge every single time it plays, one could easily calibrate/determine "by eye" the PCI video delay required by the DAW to be in "perfect human sync" (what normal human see/detect).