Jump to content

borjam

Members
  • Posts

    269
  • Joined

  • Last visited

  • Days Won

    3

Everything posted by borjam

  1. It has improved a lot these last days. Now the Cloudflare checking is working as it should, verifying that I am running a real browser and redirecting to the forum in short order.
  2. Yes, same problem from Spain. I have tried several providers just in case it was provider dependent. It seems to affect especially Apple's Safari, I have just reached the forum using Firefox while Safari was stuck. @Jeff Wexlerif you need help let me know. I would guess you tightened the denial of service protection settings too much, which, by the way, are mostly unrelated to "hacking".
  3. Great to know it's working for you! Just in case I would keep a couple of cheap attenuators in hand. If you see too much interference it might be the front end overloading due to strong signlas out of the displayed frequencies and attenuators help in that case.
  4. I am not affiliated in any way to RF Explorer, but there is some important aspect that would make me avoid the TinySA. Fortunately they are honest (a rare quality these days!) https://tinysa.org/wiki/pmwiki.php?n=Main.Limitations Specifically: The high input (240MHz to 960MHz) has very limited image suppression and only one level optional built in attenuator which makes it difficult to interpret complex signals. Which means: unless you have a couple of attenuators in hand and maybe a filter or two it might give very misleading readings and lead you to a goose chase.
  5. Exactly. License exemption doesn’t mean absolute lack of regulation. For example, you don’t need a license to deploy IOT sensors using LPWAN technologies around 867 MHz in Europe. Yet you are subject to a maximum 5% duty cycle which is appropriate for stuff like temperature sensors but it makes it illegal to, say, stream audio. And of course there are transmission power regulations. Sound should read the relevant regulations. Even in the best effort ISM bands there are as a bare minimum power limits.
  6. just in case someone misunderstood what I meant with "idiot proof", in this case I was pointing at manufacturers of crappy equipment. I found this example. A friend had WiFi problems in his bar. Turns out he had both a crappy audio/video transmitter and a 2.4 GHz motion detector. The spectrum analyzer is a very simple one (an Oscium spectrum analyzer connected to an iPhone) but it worked.
  7. The problem with the 2.4 GHz band is that it can be wildly unpredictable. Apart from the usual WiFi signals there are other potential sources of interference. Except for standards based equipment, each manufacturer can take a completely different approach. So it is really difficult. On the 2.4 GHz band you can find sources of interference such as motion sensors for alarm systems or wireless video feeds taking a lot of bandwidth. I remember some household units that rendered half of the band unusable with a constant carrier. Remember that the 2.4 GHz band is not so wide. The 5 MHz channels are not enough for a WiFi signal, each one takes 20 MHz. So, three different networks, properly configured, will take channels 1. 6 and 11 effectively taking all of the band. Even worse, some careless manufacturers have implemented 40 MHz "wide" channels so a single WiFi network takes most of the band. While the traditional UHF bands are regulated, so there is a limited variety of interference sources you cand find, the 2.4 GHz ISM band is mostly the Wild West. And nowadays you can find a lot of el cheapo junk creating interference. You won't see much of that on the traditional UHF bands. (I am saying "traditional UHF bands" because 2.4 GHz is still UHF (300 MHz to 3 GHz). I read the little information Deity published about their techniques to avoid interference and I am really impressed. I don't know the exact technical detals but I would respectfully suggest to explore spread spectrum and FEC techniques so that the delay can be at least bounded and predictable. Nevertheless I am writing from my armchair, I don't know whether it would be feasible given the power consumption, CPU power and size+weight constraints. And still, depending on the particular location and junk mix shit will happen. Be aware that sometimes, no matter how clever you are overcoming problems, equipment perpetrated by some idiot can still ruin your day. Remember that idiot-proof designs don't exist because idiots can be so effective! At the office we have a Rode Link system used by the marketing department and so far it has worked very well. But most of our WiFi traffic is on the 5 GHz band, so it doesn't cause issues for it. I am sure they will eventually tell me one day "hey, that wireless failed miserably, we couldn't get a reliable link". The 2.4 GHz is an unregulated part of the spectrum. That means that at the end of the day any service working on that band can just guarantee a best effort.
  8. The Internet community published an interesting development several years ago. https://datatracker.ietf.org/doc/html/rfc5984 That should help to overcome the delay due to retransmissions
  9. Me neither It depends on what you want. You can also try combining a shotgun with a figure-eight microphone (or two small cardioids back to back, perpendicular to the shotgun, and one of them phase inverted) in order to get some ambience. The key here is, you can decide how much ambient you want. The two small figure-eight capsules I know (AKG CK94, probably culled by Samsung with most of the AKG microphones) and an Ambient one are quite pricey. But the hack with two small cardioids can do the trick, Careful at least with the Stereo Videomic X from Rode. Looks like a great choice but at least mine is very susceptible to hummidity. Luckily I got it second hand and it wasn't expensive. I don't understand why they didn't make it RF biased.
  10. Thank you, that explains it I guess bioacousticians don't care much about sound.
  11. I am really curious. Why? Reputable sources such as Cornell University recommend them because of their high sensitivity, relatively low self-noise and reasonable price. Some regard them as the workhorses of field recording. Of course if you are going to record a dialog it's a completely different world. I mean, when recording human speech we can perceive really minimum nuances that set different microphones a world apart. For other sound sources I don't think a beautiful sound matters that much. I would give an opportunity to Rode, especially their RF biased shotguns. I own a NTG8 (the long shotgun) and I am really happy with it. The RF biased shotguns are the NTG3, NTG5 and NTG8. The NTG5 maybe offers the biggest bang for the buck including a suspension and windshield. The NTG8 is a but bulky (very long!) and if you wish to use Rode's own zeppelin you need a special extension and you must mod the suspension with stiffer Rycote lyres.
  12. That's the idea (I don't work for Viviana). In our case we set up a FreeNAS server with 10 or 20 TB of storage (I don't remember) and a clean 1 Gbps connection to our backbone. Had the need arised, we wouldn't have trouble to give it 10 Gbps. That was the trivial part. User management was a bit crazy, however. Also, I've seen a photo of a box with a USB connector on the soundonset.com website. So I guess the box emulates a thumb drive used to copy the files. Looks quite convenient if it works.
  13. Of course you can mail a SSD or a SD card or you can use an off the shelf cloud upload service. The tricky part is doing it in a more secure way. I am sure you have heard terror stories about databases with personal data left exposed on Amazon's S3 service for example. Offering a lot of storage room and bandwidth is not an issue, but a set of configurable access rights in order to accomodate a certain workflow can be very cumbersome to implement with common storage software, requiring a lot of manual work. Imagine for example a fine grained system in which the camera department can upload raw footage, sound can upload sound recordings, none of them can mess with the other department's files, post has access to both but can't destroy them either, still photography can also upload photos that production office will approve so that the publicity agency can download them, dailies are made available to the right parties... I think I mentioned it on another thread. I work for an ISP and a couple of years ago we set up a shared storage system for a Spanish production company. It was quite complex to accomodate all these requirements with camera, sound, fx, sound post, lab, stills, publicity, etc and not having especially designed software to do that. The trivial part was to offer the required TB of storage with good disk redundancy and a lot of bandwidth.
  14. Years ago I was able to clean quite a bit of clipping for a Fender Rhodes piano. Yes, it's a particularly easy target for that treatment. I used a multi band compressor for it, exactly WaveArts MultiDynamics. Also Izotope RX might help. https://www.izotope.com/en/products/rx.html
  15. What is the "garbage" problem, exactly? Distortion? Low signal to noise ratio? Maybe someone can give an idea.
  16. I would have them use an audio interface. Of course that means the remote actor/ess will have to use the audio interface and microphone. Software will really get in the way due to security measures.
  17. Be careful with high temperature soldering. I learned the hard way! I had to solder some female SMA connectors to PCBs and I decided to use lead free solder. I just had got a brand new temperature regulated soldering iron... And I badly botched it. The temperature damaged the connector dielectric material (and the circuits were supposed to work at 1 GHz) and the result was a complete disaster. Lesson learned.
  18. Now imagine that Schoeps clone company that even copies the ads. Will they copy the fart?
  19. You can hide a lav, transmitter and antenna inside that promising radome! Even pros (and I mean two friends who are actual experts on RF, one of them worked for Agilent in Santa Rosa and they are university professors) say that RF can be a bit of voodoo now and then. Now, seriously. No metal in the clothing? I mean, a bodice with some metal wires or something similar? Something must be detuning the antenna. Maybe she is sweating a lot and adding some isolating material between the antenna and the clothes might help?
  20. I forgot: An easy source for radome material can be aquarium stores. They use to stock stuff for making aquarium backgrounds. I remember some of it was thick paper with a rock like texture. It would probably pass.
  21. Apologies in advance if my suggestions sound silly! Is the antenna pointing upwards? If it gets too close to the metal necklaces I would try pointing it downwards instead. As for stone walls, what about hiding an antenna and running some cable? You can build a radome with some RF inert material. Even simple paper (there is paper available imitating stone textures) covering the antenna should work.
  22. I have volunteered to be a Guinea pig in the phase III trials for the Curevac vaccine. I haven't been inoculated yet and of course I may get a placebo instead (50% chance).
  23. I think it's an interesting development and I dare to say it will be more widely used than you imagine! I work for an ISP and a couple of years ago a producer (a real one, I mean, not a short feature amateur) approached us for a solution to set up a distribution facility. They were shooting a feature and they wanted a quick way to distribute takes, dailies, stills, exchanging them with audio post, vfx, lab, even the ad agency to make press kits available. Everything. It was also used for final delivery to distribution for a year or so. We set up a high availability storage system with proper security measures connected to our backbone network, and we assisted the producer to create a folder hierarchy together with the different permissions so that vfx only saw the files intended for them, the ad agency only the stills and press kits intended for them, etc. It also implemented hard measures against threats such as ransomware. Not even the producer could actually destroy files unless he personally ordered it to us and we had an authentication protocol in place (only two persons with their phone numbers registered were authorized to order any change, and we called back to the authorized number for verification before attending the request) The system worked, albeit it wasn't perfect mosltly due to being a makeshift solution without the thorough design process it requires. Anyway the consequence was that it required more manual attention than it should. Nothing serious. The main hurdles were: - Common storage systems are not designed with complicated workflow hierarchies in mind. This wasn't a hyped production like, say, Game of Thrones or Lord of the Rings (which means nobody expected a serious effort to steal media), but security must be top notch nonetheless. And fine grained security is very complicated. It demanded some manual attention in order to tune folder permissions. - In order to work smoothly this needs excellent connectivity on set and at the different facilities involved. Of course the vfx, audio post and lab companies (major ones) had knowledgeable technical staff and I had no problem at all in order to interact with them and fine tune stuff in order to maximize file transfer performance. Yet there are challenges due to the fact that the Internet protocols were designed at a time when networks had less than sixty four kilobits per second of bandwidth and, despite having being thoroughly tuned afterwards, reaching 1 Gbps of file transfer throughput can be problematic. Some of the facilities used for the sets had subpar connectivity for example. - After I have mentioned the throughput problems I am sure someone has thought: Ha! bandwidth control stuff in place. No, we do not use it at all. We do not have the boxes nor we intend to purchase any. Yet there are technical limitations inherent to years old technology! - Lack of standards is a major hurdle. You can set up a firewall in front of the storage facility and use encrypted connections, but do every parties involved have each and every secure access tool available and ready? It is complicated to manage. Firewall manufacturers don't focus much on interoperability. - Manual work involved. It was production staff who after the shots uploaded the files, etc. That was time consuming and inefficient. - More manual work. Because the lack of previous design work I remember I always kept an eye on the network activity of that particular server in order to avoid any security troubles. That is possible when it's the first time you set up the service and you have just one customer. Despite the hurdles I think it worked (I recall I should send an email to the producer in order to ask what he felt was lacking although I am sure it was mostly the weak points I have just mentioned). Now, take a platform that solves several of these. Say: - Files are automagically uploaded right after each shot. During a take all the previously captured material is being uploaded invisibly. - Forget about having the right Cisco, Fortinet, Juniper, whatever VPN software properly configured. An end to end encryption facility is used. - The software already implements the needed TCP/IP tuning mechanisms in order to maximize throughput. Also, it can upload several files in parallel, which is the best solution or it can split a large file into smaller chunks and upload them concurrently. That is complicated when a production assistant is uploading files manually. - The software makes it easy to create a permissions hierarchy so that each actor has access to the required resources and no more. So: I haven't checked that service yet but without reading anything but the name and the first post in this thread I am sure I ticked several of the boxes And I forgot: What happens with fire on the set or any other similar threat? How long beteween a shot and backup media moved to a different location? With such a system very little material will be lost.
  24. I would stick to a known good powerbank. The MixPres are power hungry and USB power is non trivial. I was experimenting with using a MixPre as an audio interface for an iPad (to run SMAART tools on it for live sound) and due to the flaky USB power provided by the USB to lightning adapters I had startup problems (it took several attempts to boot!) and when the AA NiMH ran low the MixPre entered a reboot loop. Interestingly, the reboot loop didn't seem to affect the audio flow from the microphone to the iPad. Turns out that the FPGA keeps pumping samples once it's been configured. That was funny
×
×
  • Create New...