Jump to content
pvanstry

Future of Timecode system

Recommended Posts

3 hours ago, Constantin said:

if all cameras and recorders had a generic gps tc device built in and were cross-compatible

 

Mind you, a LOT of cams already have GPS build in. (They write location metadata in the clips.)
So, for some reason (And probably the speed of / reliability of reception) the GPS clock is not used.
I disagree on 'not being compatible', if all devices write their own time based on GPS, it must be compatible.
But, as stated before, it is 'not' that obvious, as GPS is ahead of the 'normal' time by18 seconds today.
NTP could be an option, since internet is available almost anywhere nowadays.
But as stated before, it is not 'that' hard to do it the 'traditional' way. Only if all devices would run on the same clock it could be easier / less wires / equipment, but I predict a gazillion screw-ups before we get there. (Remember the early days of Plural Eyes where  people thought 'everything would sync up automagically.)

 


 

Share this post


Link to post
Share on other sites

TV studios and OB vans have been synchronizing their global TC by GPS for ages. They have rack mounted GPS receivers. Not because they're so innovative, but because it's the easiest way.

With products like these: http://www.ese-web.com/102.htm

Share this post


Link to post
Share on other sites
2 hours ago, Mungo said:

OB vans

 

OB vans normally don't boldly go where no OB van has gone before...

Share this post


Link to post
Share on other sites
3 hours ago, Mungo said:

TV studios and OB vans have been synchronizing their global TC by GPS for ages. They have rack mounted GPS receivers. Not because they're so innovative, but because it's the easiest way.

With products like these: http://www.ese-web.com/102.htm

Ha yes, forgot about that! 10nS accuracy they claim. "Dish" at 60 nanoseconds. Still close enough.

Share this post


Link to post
Share on other sites

I had a lengthy conversation with Ari Krupnik, the guy behid  "Dish" on our booth at NAB about GPS as your sole source of time and clock reference. And while he is clearly aware of the challenges and drawbacks, his system is clearly not meant for us, the pro-media people but rather the system camera filmer.

 

Just to remind, we already offered a GPS solution since the mid 90s (funnily never sold too many of them) until we had to discontinue the ACC controller. So the idea is far from new to us.

 

But, while GPS is a good reference to calibrate a system the idea of having this as your only set reference to rely on is not only ruled out as a majority of locations will leave you without reception, but also half of the world (read every region that shoots on 23.98/29.97) is left in the cold as GPS counts in real time (more or less, as per @Bouke) while NTSC lags behind, so there is no such ting as realtime over GPS in these regions and powering on equipment atr a different time of day will inevitably lead to a TC offset. As a result, "Dish" (at least at the time speaking)  is concetped to output fixed time of day @ 24FPS, btw.



 

Share this post


Link to post
Share on other sites

completely agree that "Dish" doesn't look to be aimed (or very useful) for dedicated audio teams.

 

I don't quite see why you couldn't use a GPS signal to synch the internal TC clock to 23.98/29.98 thought, I'm sure there are some gotchas (like Bouke mentioned) but it doesn't seem to be rocket science either.

 

As mentioned the real benefit would only come once all the cameras/recorders would have this built-in, so we don't even have to think about it anymore. for this to happen there would have to be a big player setting up a standard though. I guess Ambient would be in a pretty good position to make this happen (or at least attempt it) - but on the other hand they would probably kill one of their core products if they succeed, so I can see why this is not too tempting ; )

 

 

 

Share this post


Link to post
Share on other sites
13 hours ago, chrismedr said:

I don't quite see why you couldn't use a GPS signal to synch the internal TC clock to 23.98/29.98

Well, the momant you sync, a 23/29 TC will slowly drift behind the continuously running real world time with a speed of - .1% roughly, which means after 1000 frames (or 42 seconds) you are already lagging 1 frame. So, say such super easy camera set to 23 (which, btw, is every Sony set to "24") comes on at 9:00 AM to for the day, and a second device fires up 10 minutes later, both units will already be 14 frames out of sync. This gets worse with every battery swap, scene rebuilt or similar.

 

Well, one could possibly compensate for the timebase diversion of 1000ms to 1001.001001.... via PLL (what we did) and additionally try to approximate "NTSC real time" by recalculating the current lazy saturn time from 00:00 real world on reception, but if one unit has reception the other one doesn't and you just rely on this things will get hairy very rapidly. All I'm trying to say is the benefit of "not having to think about it" would not last longer than the first day you work which such a setup (if ever) until you get a phone call from post.

 

But there's always proof to the pudding and as per @Bouke other time beases that are supossedly as accurate (NTP/network time etc.) If soemone has the ability to compare 2 timecode sources (e.g. with one of our Lockits with display on6.xx) just download a free app suhc as LTC Timecode generator for Android, launch it on 2 phones and compare the "real time" timecode on the headphone output - you will notice a considerable difference even on straight framerates. So,let me asure, we are not only in the position but we pursue this very actively. Actually we do have a software based solutionalready in the drawer but we definitely won't release anything that raises false expectations and does not live up to the quality standards that we are appreciated for just to compromise the bad reputation tc has already ;)

 

Share this post


Link to post
Share on other sites
51 minutes ago, Klaus said:

Well, one could possibly compensate for the timebase diversion of 1000ms to 1001.001001.... via PLL (what we did) and additionally try to approximate "NTSC real time" by recalculating the current lazy saturn time from 00:00 real world on reception, but if one unit has reception the other one doesn't and you just rely on this things will get hairy very rapidly. 

 

yeah, of course for such a synch method to be truly useful you'd have to have an internal accurate clock as well which holds time code over a few hours (not much use if it drifts if you loose GPS signals for like 20minutes). 

I'd also want some warning indicator if there's no GPS synch possible and possible a submenu where you can see when the last synch has occured.

 

obviously not going to happen in the next Canon 5D MkV, but maybe in 5 years time something like this is becoming a standard at least in professional gear.

 

 

Share this post


Link to post
Share on other sites
1 hour ago, chrismedr said:

yeah, of course for such a synch method to be truly useful you'd have to have an internal accurate clock as well which holds time code over a few hours (not much use if it drifts if you loose GPS signals for like 20minutes).  

Which basically rules out the vast majority of camera manufacturers - unless they start to consider implementing our module ;)

1 hour ago, chrismedr said:

I'd also want some [...] submenu where you can see when the last synch has occured. 

 

Nettime, not GPS but, like this?

nettime.jpg

Share this post


Link to post
Share on other sites
21 hours ago, Klaus said:

is left in the cold as GPS counts in real time (more or less, as per @Bouke)

 

Eer, does the @ means that I have to comment on how tricky this part is?

(Well, it is NOT RT, it has to be compensated with the launch date of the satellite, but then you have to know that, and there is the tricky part, let alone the subscription you need on having an accurate bird database. So, when we get 5G +, there will be 20.000 of them... (But, I'm sure there will be lots of nerd fun to do then, to generate a stupid horrible sounding totally outdated blockwave.)
 

 

 

Share this post


Link to post
Share on other sites

Parts of this discussion remind me of ones taking place on the cusp of the digital revolution. Along the lines of: "Won't it be great once everything goes digital -- we'll all be working with one standard." 

Share this post


Link to post
Share on other sites
20 hours ago, John Blankenship said:

Parts of this discussion remind me of ones taking place on the cusp of the digital revolution. Along the lines of: "Won't it be great once everything goes digital -- we'll all be working with one standard." 

 

I understand what you mean, but there was a point when people developed a standard for the current TC system, and it would definitely be possible to do the same with TC based on GPS (and probably even to keep the two compatible).

 

if there's enough demand and benefits to do that is a different question 

Share this post


Link to post
Share on other sites
6 hours ago, chrismedr said:

 

I understand what you mean, but there was a point when people developed a standard for the current TC system, and it would definitely be possible to do the same with TC based on GPS (and probably even to keep the two compatible).

... 

 

That's my point.  With the current SMPTE time code standards well in place, look at how many different, often sketchy and misunderstood, implementations camera manufacturers have come up with.  Do you really wish to complicate that even further? 

 

Share this post


Link to post
Share on other sites

yeah, I can definitely imagine some manufacturer making this a bit of a painful transition.

 

but still, the topic of this thread is called "Future of Timecode system", and personally I feel that satellite synched devices would make for a strong contender (the other option being, let's keep the current method for the next 100years)

Share this post


Link to post
Share on other sites
1 hour ago, chrismedr said:

yeah, I can definitely imagine some manufacturer making this a bit of a painful transition.

 

but still, the topic of this thread is called "Future of Timecode system", and personally I feel that satellite synched devices would make for a strong contender (the other option being, let's keep the current method for the next 100years)

They (SMPTE) are busy with the successor of the current standard for a while, it is called TLX (Time Label, eXtensible) and most likely the choice of synchronisation will be over IP:

https://www.smpte.org/webcasts/standards-feb-beyond-smpte-time-code-tlx-project

Share this post


Link to post
Share on other sites
2 hours ago, chrismedr said:

I feel that satellite synched devices would make for a strong contender (the other option being, let's keep the current method for the next 100years)

 

I definitely don’t mind a new sync method if someone comes up with one. 

I don’t see it being gps sourced tc, though, unless they somehow address the indoors no service available situation. 

I also don’t mind the current system, as it has progressed quite a bit in the recent years. 

But really, gps tc,  or little boxes, or Bluetooth, or Wifi, or whatever, all of that only concerns the mode of delivery. What might be due for an update is tc itself...

Share this post


Link to post
Share on other sites

What if on a job and some space junk collides with the satellite TX'ing TC/GPS signal? 

 

Googlemaps is the big dog in the GPS business and my phone still regularly makes mistakes. Despite capturing every minute detail and aspect of my life they still cannot tell if I'm walking north up the street or south down the street, or if I am on the frontage road or the freeway. 

 

Unless there is a revolutionary groundbreaking technological breakthrough, I think I'll just stick to Lockit's and JB-1's. I like to think there are people sitting at a desk getting a salary from Ambient and Deneke to have these types of conversations for us. 

Share this post


Link to post
Share on other sites
On 5/5/2019 at 7:29 PM, Dalton Patterson said:

What if on a job and some space junk collides with the satellite TX'ing TC/GPS signal? 

 

Life as you know it will cease to exist, but no worries, the devices will fall back to their internal clock and you will be in sync for at least 16 hours.
After that, you might be experiencing a frame offset, but since Post is no longer alive, nothing to worry about.

 

 

Share this post


Link to post
Share on other sites
On 5/5/2019 at 6:29 PM, Dalton Patterson said:

What if on a job and some space junk collides with the satellite TX'ing TC/GPS signal? 

This nearly happened in 2018, India destroyed a satellite which could have set of a chain reaction of debris taking everything else with it.

Share this post


Link to post
Share on other sites

“Initially, the term space debris referred to the natural debris found in the solar system: asteroids, comets, and meteoroids. However, with the 1979 beginning of the NASA Orbital Debris Program, the term also refers to the debris from the mass of defunct, artificially created objects in space, especially Earth orbit.”

 

I was referring to the artificially created objects in Earth orbit, as opposed to an asteroid colliding with the satellite before eventually causing a catastrophic end of the world. 

 

Theres basically trash everywhere. Oceans, land space was my main point. 

Share this post


Link to post
Share on other sites

Hi. I'm Ari Krupnik, the designer of DISH.TC . It flatters me that my idea came up in a thread titled "Future of Timecode." I'm happy to answer any questions you may have about DISH.

 

As some of you know, we just launched a Kickstarter for DISH. This page has a lot more information, videos and thoughts from people who've used our prototypes: 

https://www.kickstarter.com/projects/ari-krupnik/dish-timecode-sync-by-satellite

 

dish-startup.png

Share this post


Link to post
Share on other sites
2 hours ago, iter said:

Hi. I'm Ari Krupnik, the designer of DISH.TC . It flatters me that my idea came up in a thread titled "Future of Timecode."

 

Hi Ari and welcome,

if you scroll up a bit you'll notice that quite a few people came up with "your" idea before ; )

credits though for actually trying to bring it to market and I hope it'll succeed. the real benefit won't come until the method is built into devices and we can forget about hooking up extra devices though. 

Share this post


Link to post
Share on other sites
33 minutes ago, chrismedr said:

 

Hi Ari and welcome,

if you scroll up a bit you'll notice that quite a few people came up with "your" idea before ; )

credits though for actually trying to bring it to market and I hope it'll succeed. the real benefit won't come until the method is built into devices and we can forget about hooking up extra devices though. 

Thanks, @chrismedr.

 

It was a big surprise to me that this didn't exist yet, and a bigger surprise that camera manufacturers didn't offer this functionality. The irony is that many cameras already have GPS onboard, but they don't use it for time!

 

The idea isn't new at all. I've seen records of hobbyists and prominent manufacturers tinkering with the idea over the years. Like @Bouke and @Klaus say, the details are what's hard. Finding the exact start of a UTC second is hard. Dealing with leap seconds is hard. Translating between UTC and 30/1.001 timebases is hard. So my claim is only that I did the grunt work of engineering and testing the solutions to these issues.

 

I think my angle is different from a lot of other vendors. My goal is to build a TC device that anyone can use. I've met so many people who told me TC was too complex for them. Or they say they tried it once and it didn't work and they were worse off then without it and post was even more awful than usual. I want to build a TC generator that a PA can't screw up when they are trying to be helpful. Using a satellite time signal is simply a way to get a consistent, reliable, replicable way to get all clocks on a set on the same time. And as a side effect, ALL dish clocks in the world are on the same time. All the other design choices in DISH come from the same place--how can I build a device that people who aren't sound engineers can't misconfigure.

Share this post


Link to post
Share on other sites
1 hour ago, iter said:

how can I build a device that people who aren't sound engineers can't misconfigure.

 

well, the concept is certainly appealing, but as long as we have to rely on audio LTC through the mic-input on most cameras and/or frame rate preselection etc, I'm afraid there are still plenty of opportunities for mess ups - but one less is still one step closer : )

chris

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...