I often get the question that with so many switchers offering frame syncs on every input, why would you still need to genlock an entire system? Can you just lock everything using the frame sync’s on the switcher? And the answer is: Yes, you can lock everything at the switcher using internal frame syncs, but as your system expands beyond your switcher, you may not want to.
In the last post I explained how genlock works and what it does to your system, but before I answer the question of if it is still relevant I want to go deeper into locking and syncing your system.
Every destination that accepts a video signal has to get it’s timing or sync from somewhere, including monitors, waveform vector scopes, record decks, projectors, Graphics machines, and everything in your system. There are 2 ways that these pieces of gear get their timing. The first is through Genlock, if the destination has a reference input, not all destinations do. The second is through the inputted video signal it self. Because all video signals carry Sync and timing in them, destinations can inherently time themselves from the source that feeds them.
In the last post I compared genlock to a conductor of an orchestra, setting the beat and the rhythm of your system. If you have ever been to an Elementary school orchestra concert you know that as hard as the conductor tries, the musicians vary in how they interpret the beat. They are still playing together but each musician isn’t as tightly in sync with the next musician as they could be. But if you go and hear a professional orchestra, each musician is very tightly in sync with each other as the conductor directs the orchestra. Your video system is very much the same way. Each piece of gear may be hitting the beat, but within what window? We call this the “timing window.” Because genlock is still an analog signal, there can be variations to the signal from device to device. This can be caused by: cable distance, cable integrity, distribution amps and simply how different devices are built. So as one device may be perfectly timed, another device may be off by a little bit.
Switchers utilize a timing window that as long as the video signal falls within the widow, it can mix seamlessly between sources without any issues, but if it falls outside of that timing window, you can get timing errors and may be required to use a frame sync anyway. Other pieces of gear though, don’t usually have any type of timing window, instead they just have to accept the video signal as it is and when that signal changes it has to physically re-adjust it’s timing to match the incoming video signal even if it is only off by a pixel. In fact switchers are the only device that I know of that has a timing window to allow for differences in timing.
Ideally we want a tightly sync system, much like we want a tightly sync orchestra. Most devices allow us to not only sync up frames of video at the same time, but we can get as detailed to align the exact pixels so all sources scan at the exact same time. We call this “Phasing” a system. There is “H phase” (horizontal phase), “V phase” (vertical phase) and “SC phase” (sub carrier) phase. H phase aligns your signal left and right, Vertical up and down, and SC phase aligns the hue of the video. In the world of SDI we have H phase and V phase, SC is no longer needed thanks to digital video. Most of the time when we genlock a source it will fall in our switchers timing window, but in the old days when composite systems ruled supreme, phasing was much more of a necessity and could take hours to accomplish.
Side note: If you have any sources that shift down when you take them or add blank lines on the top, this is usually due to a vertical phase issue and can be quite common. The devices I most commonly have vertical phase issues with are scan converters. By simply adjusting vertical timing will fix this issue.
The only way to phase out your system properly is with a tool such as a waveform monitor (not all scopes will analyze timing in todays world). I tend to use a rasterizer brand called Phabrix which provide detailed timing information, but Techtronics, Leader, and Harris all have great products that can analyze your signal and gives you timing information. I am not going to tell you how to phase out your system as it can be pretty complex. Find a good local broadcast engineer or a system integrator to help with this.
If you are tracking with me at this point you may be thinking that this is all well and good, but the question still remains, why should I genlock all my sources when I can just use a frame sync in my switcher? The reason is that all of the destinations in your system receiving video signal need to get their sync from somewhere too. In your system you probably have projectors, monitors, record decks, shading stations, converters, streaming systems and many other destinations receiving some type of video signal, hopefully from some type of production router. If your switcher is the only device where all sources are synced together, then any device that gets routed a new signal needs to re-sync itself every time your route a new source to that destination. Every device handles this differently but if you don’t genlock all your sources then you will more then likely see a flash or jump of some sort when you change the source going to a destination. But if everything is genlocked then you can switch any timed sources to any destination with little to no flash, and the better timed or phased out your system, the less jump and flash you will see.
For example: I always setup my shading station as a router destination, that way I can see all sources in their organic state so I can analyze timing, colors, brightness and more without seeing it go through a switcher first where it could be manipulated in some way. If I am comparing 2 cameras I may be flipping back and forth quickly, if my cameras are not genlocked and phased, it can take seconds before a camera or sources syncs up on my monitor and scope, but if they are locked and phased then I can instantly compare sources with no waiting. Or, lets say you are ISO recording cameras, but because you don’t have a dedicated recording deck for each camera, you cut to a certain camera when your speaker walks to the lefts side of the stage and a different camera when they walk to the right. If things are not synced then you will get a nasty, long cut between cameras. And depending on the record deck it may even stop recording on you. But if everything is lock and perfectly in phase you will have a seamless cut. Or, you may want to route a source to your projection screens before service while people are coming in so you can run through stuff on your switcher without it going out for everyone to see. Once again, non-locked signals will go crazy and flash the screens, where a locked signal will be seamless.
Now with all this said, I still use internal switcher frame syncs if I am adding in a quick POV camera or running a scan converter from a far away place. They are convenient and easy and although they usually add a frame of delay I am ok to deal with it on certain sources. But every piece that is permanently hardwired and considered a pivotal source in my system are all locked and phased together for as tightly a timed system as I can get it. And sometimes you can perfectly phase your system and devices will still jump as they can be overly sensitive, you just can’t win them all. (this is usually due to a line drop caused by routers and the device not allowing this in it’s timing window)
I could go on for pages on this topic giving different examples; but I hope that you see that as valuable as internal frame syncs are to a switcher, it is far better to genlock your system everywhere you can so that every piece of gear on it’s own starts off in perfect sync to one another, so that no matter what destination it ends up to, everything stays locked. It is far better to sync sources on the front end then the back end. Your system will be much happier because of it.