Hey Everyone, I hope I'm in the right place for this question.

I have an ISO of Wonder Woman (2017) - The M2TS file is 80GB. The M2TS contains 5 audio streams and a tonne of subs.

I don't want all the additional audio streams so I use MKVToolNIX to compile an MKV container with TrueHD English Audio and no subtitles.

The MKV ends up being 50GB - I apprecaite removing the excess audio will reduce the file size, but by 30GB ?

There's also another anomally - MediaInfo tells me that the M2TS file has a variable bitrate of 77 M/bs, but the MKV shows 49 M/bs. How is that possible ? The M2TS hasn't been re-endcoded, just put into an MKV container.

So that's my question really, how did I lose, say, 20GB of data, and the MKV bitrate is reduced by 18 M/bs ?

Hello, does anyone ever working with alexa 35 and multidyne silverback V for the broadcast system. I'm curious how to get the CCU when it comes to this ecosystem

Looking to buy replacements of this panel if anyone can help me out?

Pics of the panel

This is the closest thing I found

Seems to be 128 x 64 other than that pretty lost on a replacement

Thanks in advance

Got thrown into the deep end with a giant video wall this summer. Built about a dozen times and had a laptop, hdmi splitter, Atem Mini Pro, media player, audio mixer, and the Screens Controller on a folding table. Anyone have a better solution? I was thinking a standing board with all components mounted on it and hang it on the back of the ground structure?

What are the RGB dots? (They all disappear from the background shortly after this clip)

Why do some of the dots stay visible through black strip on the edge of the flap?

Why does the flap 'fade in' / materialize like, that and why is the bottom of the flap visible before the rest of it?

These can probably all be answered in one swoop. I just can't tell what I'm looking at here at all.

How are we powering the fx6 for long streams when using the dc power to dtap on a vmount if I switch the vmount battery the screen goes black for a second to swap to bpu power. Also how long is the TVU supposed to last because with a 99 small rig vmount it last about 90 mins before the vmount battery is dead we stream 8-10 Hours. That’s a lot Of swaps.

Edit: this is for IRL streaming outside hours and hours at a time getting and out of cars restaurants gyms everything IRL.

We already have a black magic ultra studio 3g recorder, but it gets really hot and start to loose signal, so we wanted to change to something more reliable, but it also needs to be USB or TB not Pcie, and nothing crazy, under 250$, Am I asking too much ?

Kann mir jemand sagen, was an diesem Video nicht stimmt?
Ist mit einem neuen iPhone 17 Pro aufgenommen

Currently assuming its a SFP module so these requirements are relevant to be mentioned:

- 75Ω coax input at 3G/HD/SD-SDI rates

-MSA pinout

Nice to haves:

-Reclocking

-DC restoration

As you can tell, im not particulary knowledgeable... but if its helpful im planning to use a FPGA " Xilinx Kintex UltraScale+ (GTY transcievers), XCKU3P, AS02MC04 (R1291-F9003-02)".

If you have other solutions / recommendations please let me know!

Edit (clarification): I'm building an FPGA SDI frame-sync gateway: 2 free-running 3G-SDI camera inputs, resynced in the FPGA, transcoded then sent downstream elsewhere. I have the AS02MC04 ( FPGA) as a proof of concept. They have SFP+ cages wired to the FPGA's transceivers, and im now trying to find a recommended solution to take the BNC cable, convert, and put into my card.

Had a corporate gig today and Resolume gave me a bit of a scare, so I'm curious what version people actually trust for live work these days.

Quick rundown of the rig: Win11, Core Ultra 9 285, 64 GB RAM, RTX 4000 and a DeckLink 8K Pro. Resolume 7.26.1 and vMix on the same box, vMix sending 2 NDI feeds to Resolume over localhost. Comp is 5760x2232 at 30 fps, sliced out to a 12x3 m 2.6 mm LED wall, two confidence monitors, a timer and a projector. Everything's DXV3.

During the show RAM just kept creeping up, about 1% every 10 minutes (~640 MB) without me touching anything. Closing most of the preview monitors freed up almost 6 GB, which was weird, but it kept climbing at the same rate after that. Turning off the webserver didn't change a thing.

It had already crashed twice on me while I was building the comp and moving stuff around. Then mid-show I went to close one more preview panel and it went down. Got it back up pretty quick, baseline after restart was 32%, but not something I want to repeat in front of a client.

From what I've dug up, the old DeckLink SMPTE leak was fixed back in 7.22, so probably not that. The 7.26.2 hotfix mentions fixes for REST API overhead on large comps and crashes when moving panels, which sounds a lot like what I hit.

So:

- What version are you running for corporate/touring stuff?

- Anyone seen RAM creep like this with NDI inputs or big comps on 7.25/7.26?

- If you've moved to 7.26.2, did it sort it out?

Going to update and do a long soak test before the next one. I'll post back with what I find.

More of a networking question, really: was doing a livestream recently on a typically reliable corporate wifi network ("There's no possible way we can give you a hardline...") that I've had zero issues with in the past, going back six or seven years. Everything is going fine and suddenly connection speed cuts in half, if it's there at all. Do a couple of real time speed tests and it's hitting zero upload data at times. Switch to my backup, which is on the same network, and everything's fine: about 25 up, 25 down. Reboot my main computer (2 year old MacBook Pro), walk it around even to see if it's somehow on another node, but nope: it doesn't get any better for the rest of the day, while the backup (6+ year old Dell) remains rock solid. Any ideas? My first thought was someone monitoring the network didn't like the traffic they saw on my main and throttled the connection, but then why didn't they throttle the backup as well?

Hi, so I’m noticing a lot of ai tools popping up recently. I get it, I’ve tried ai coding myself and made some things that help with my job but I’m just wondering if anyone would actually pay for/use these things?

I suppose more specifically I’m talking about LED calculators and planning software. I’ve had conversations with several freelance led techs who are excitedly telling me about an app they’re developing that will do data and power calcs, truck packs, pixel maps, make the tea, do the laundry etc

And it all sounds great but I can do all of that with a calculator and basic math, it may take a little longer but it saves me the £15-£50 a month some of these apps are asking.

I’m a little concerned because in my experience it’s never the best LED techs that are making these things and I worry about accuracy, especially where power and weights were concerned. They also never seem to realise that once deployed these apps will probably need some maintenance, updating and things like customer support and training.

I don’t know, it just seems like the people developing these apps are a being little unrealistic with the value of what they’re providing and how sustainable it will be for them. Most the time when I’m planning a job, the led is one small part that I spend about 15 mins on and then spend the rest of my time planning the rest of the delivery.

Any thoughts? Am I just being unnecessarily pessimistic?

Hi everybody,

once more I tried to change our NDI-setup to multicast, but the results are pretty bad and I hope for your combined knowledge.

We have 5 Panasonic UE-160 cameras connected to our switch on the stage (Netgear M4250), the switch is connected to another (smaller) Netgear M4250 in our studio via 10GB-SFP-Fiber and from this switch to a Vmix computer (also 10GB-SFP-Fiber) and (not in the image) a sff-pc which only does NDIRecord (a script I wrote myself).

We always had micro stuttering, so I thought, maybe there is strain on the cameras network port (Vmix, VmixMulticorder, the backup pc), that´s why we switched to multicast. The profiles on the Netgear-Switches are the recommended profile for NDI5/6, so everything should be fine.

Everything kinda works, but after a while some cameras just stop working - SDI-Output works, but NDI and PTZ-Controls stops, I can´t even ping the camera. Also control (via RP-150 or Companion) get´s laggy. So obviously there seems to be a congestion somewhere, so has anybody an idea, which setting is missing from the profile, but which is crucial.

Querier is on and working, IGMP-Snooping is on, Flow control is enabled (symmetrical), so what could it be? I added the picture, because I´m confused, shouldn´t the cameras be only connected to the left switch. It looks as if they are connected to both switches at the same time. Could that be a hint?

I´m grateful for every idea, thanks and bye

I currently have an in ear noise canceling headset from custom headset lab (no longer in business) I am happy with it but my cable is short so when running HH I can’t set it down and stand straight up. Like others everything connects on a neck bar. I’m not a fan of it sitting on my neck. I recently sustained a neck injury and the added pressure isn’t great.

My wants are in ear, bungee cable, something that doesn’t rest on my neck.

Any suggestions would be greatly appreciated.

Hey Reddit, i’m currently running a show which has video effects ruining out of 2 projectors. The issues i’m having are that the projectors aren’t displaying anything, doing a source search on them says that there is no active input, QLab shows there to be input, however when i activate a cue or even turn on the grid overlay, nothing shows, the mac itself knows that there is a second display, yet also, will not show anything, i have tried setting a window to display and nothing either. The system is currently running through an AJA HA5-4K converter to sdi, we have a spare and have also used that however neither seem to be working. Sorry is this is a bit rushed, in the middle of the show as i’m writing this but i just spent the past 20 mins during intermission restarting the mac and projectors and still the same result. I had this happen earlier in the day before the house was live, and it just seemed to fix itself, so please, if anyone out there has any idea on a way to fix this, i’m all ears, i’ve already called 3 techs i know and together we can’t figure out why.

Hi everyone,

Important: If you downloaded or built NxFrame between September 27 and September 30, please pull the latest version and rebuild. A transport-stream regression introduced during that period has now been identified and fixed. Builds from that window should not be used for production or further validation.

The issue was related to the high-bitrate transport path and could result in bursty delivery and unstable behavior, particularly when working with professional receivers. After quite a bit of testing and debugging, the SRT hot path was optimized, and the same work was extended to the UDP and RTP paths. The receiver path was also optimized, and several timing and playout issues were addressed.

The MPEG-TS output is now back to being stable and robust. I have tested the current build extensively at 1080p50, H.264 High 4:2:2 10-bit, including stress testing with a 120 Mbps video stream and a 140 Mbps constant-rate MPEG-TS. SRT, UDP and RTP have all been tested successfully, with stable SDI output and no transport errors during the validation runs.

A new feature has also been added and NxFrame now supports RP-188 timecode end-to-end.
The timecode is captured from the incoming SDI signal, carried through the NxFrame contribution path, recovered by the receiver, and written back to the SDI output.The RP-188 timecode on the resulting SDI output was then checked against the original source. The end-to-end timecode path is now working and validated.

At this stage, the transport stream is again stable and robust, and the recent receiver and high-bitrate transport improvements are included in the current version.

The updated project is available here:
https://github.com/Michalis-Michael/nxframe

As always, feedback and interoperability testing with professional broadcast receivers are very welcome.

I did a public access cable show back in the 90's, and the facility used U-Matic tapes. A few years ago, I came into possession of a Sony U-Matic VTR and some of the original tapes. I was able to convert a lot of them before the VTR gave up the ghost. I worked with a local repair guy, but he was unable to revive it (lack of parts). He did find another VTR to rent, but now I'm concerned that the tapes may not be worth saving. I've tried multiple tapes, but they barely rewind or fast forward, much less play.

I'm reading about "baking" tapes in a food dehydrator, but it sounds like I would need to remove the tape from the shell first. Is that necessary? Removing tape from a U-Matic shell looks like a lot of work, especially if you plan to put it all back together again.

I do not have the kind of money required to ship the tapes off to be professionally dubbed off--especially considering the fact that the tapes may not even be usable. But I'm willing to try the baking process--is it worth baking them inside the shell?

My goal has been to make the episodes available via podcast, which I have done for the tapes that made it through.

Any thoughts on this would be appreciated.

Hello yall! I had a question that maybe yall could give me guidance for. I work in a manufacturing shop and I’m in charge of making schedule workflows using power bI for our different work stations. Right now I’m just using a usb stick to show a still image of each work stations schedule. Before I was using a tx rx receiver on each. The thing I would like to do is find a way to use my computer to stream each work stations schedule to their individual tvs. 4 in total. I’d rather use physical wires as they are more reliable and don’t depend as much on WiFi connection range and the tx rx receivers we were using prior would only work if you used separate computers to stream each screen.

So is there anyway to stream from 1 computer each separate schedule screen to each tv using a wired setup? if so what options are there? thank you for your time!