Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - totallymaxed

Pages: 1 2 3 [4] 5 6 ... 296
46
Users / Why Do Media Directors Exist?
« on: August 21, 2014, 12:36:41 pm »
First off I know this post will be contentious for many of you...but read on and then respond;

Back in the day when Pluto Inc were first developing Pluto (of which LinuxMCE & Dianemo are direct descendants) the Tech world was very different;

- Crestron were the Elephant in the room (still are?)
- Everything was expensive (I mean everything)
- There were no inexpensive Media Players or SmartTV's
- No inexpensive Tablets and SmartPhones hadn't been invented
- Building custom hardware was difficult & expensive

Its in this context that PXE booted Media Directors were born. They solved several problems in one hit; They enabled a '10 foot' TV based custom UI, they played local CD/DVD discs and could stream video from the Core, they provided local in-room control I/O at essentially zero marginal cost. At the time all these capabilities made incredible sense even though they added complexity. But today they largely don't.

Today the UI has migrated to your tablet or smartphone and is touch driven, we have low cost IP controllable media players and Blueray/DVD players that can play disc media and/or stream video content from anywhere, independent small IP controlled I/O is cheap and easy to install where ever you need it. The Media Director is a hangover from a previous age. It adds cost, complexity and adds considerably to the amount of energy a system consumes. Its another point of failure and a source of many problems both for those building and installing their own systems and for those using systems too.

How many threads here in the forum relate to problems with MD's? I don't know the number but its a big one. And they continue to cause reliability problems in use too. Yes some installations have MD's that don't cause any problems - but most do experience problems at some point. Of course upgrading from one OS release to another is also problematic too in terms getting displays configured and working properly again.

Removing MD's, using Orbiters on tablets or Smartphones and using IP controlled media players/SmartTV's seems to me to be a much better approach overall and its the one we've taken with Dianemo. It improves overall reliability, lowers energy usage, reduces the physical size of your system while improving video playback quality & performance.

For some of us its the challenge of getting our systems working that is the payback and in that sense MD's improve  systems! But I think for most of us we'd like fewer challenges and more smooth and rewarding usage of our systems. After all there are an infinite number of other pieces of hardware you can get a challenge from when integrating that won't stop your family from watching Game of Thrones ;-)

Comments?

All the best

Andy

47
Did you get a chance to do any testing Andrew - looking for an Android solution for Dianemo, and see the same issue as reported in this thread with Dianemo and the touch orbiter.

Yes I did. But I obviously then forgot to post an update here. Sorry!

There are some basic issues with this version of the Android Touch Orbiter that are not now compatible with Dianemo and would require a reasonable amount of effort to fix.

However we are developing a new Orbiter that is built in HTML5 and renders its in-screen UI in HTML5 too. See this thread for some screen captures and some details; http://forum.linuxmce.org/index.php?topic=8880.msg99799#msg99799

There are several major advantages to our new HTML5 Orbiter;
  • It can be rendered by any modern web browser or by an embedded HTML5 view in a wrapper App on any device and any OS.
  • It renders to a live HTML5 canvas and therefore the pre-rendered screen bitmaps are not required.
  • No Orbiter regens are required - ever. As we are not pre rendering screens as all Orbiters have required us to do previously
  • It can render dynamic UI elements like sliders or any other control you can imagine that can be coded in HTML5
  • Orbiter screens/UI's can be designed & layed out by anyone with ordinary HTML5 design skills - No HADesigner required!
  • New Skins can be designed without any programming or low level skills - just HTML5
We hope to ship the new HTML5 Orbiter as a Dianemo update in mid Q4 2014.

48
Users / Re: Power saving topic
« on: August 20, 2014, 12:55:04 pm »
The whisperer device is in svn but is not being build. I am not aware of any corresponding UPnP code in our svn. Could you (or Radu) point me to the appropriate place, so I can incorporate it into the build?

Thanks.

Radu the best person to do that. One of the problems that may have caused this (based on a vague recollection of a discussion) is that our whisperer based device templates have a bunch of dependencies in terms of screens and our different pipes system etc. All of which would need to be resolved to some extent to get our devices environment working acceptably outside Dianemo.

49
Users / Re: Can I control a DLNA device (SONY TV) from Linuxmce?
« on: August 19, 2014, 11:49:38 am »
Just a quick update on this topic;

We use a combination of device templates to control SmartTV's. See our Wiki page section for specifics; http://wiki.linuxmce.org/index.php/Dianemo_S#TV_Control_In_Summary. I think this section needs some updates for other brands of TV's - which I will follow up on asap. Most modern SmartTV's are controllable to some extent by using our Universal UPnP template alone but every make/model varies a lot in this respect. One universal limitation is that if you force  SmartTV's into stand-by mode with an 'Off' command they will not respond to an 'On' command to wake them up again (because the network port is powered down in stand-by mode!). So we always use our CEC HDMI over IP device to power up TV's from stand-by.

So depending on the brand of TV we use one or all of the device templates mentioned to deliver control of the TV. Its a little bit messy but until TV manufacturers standardise how they implement this this is not going to change - don't hold your breath for that!

All the best

Andy


50
Users / Re: Power saving topic
« on: August 14, 2014, 01:55:32 pm »
@Andy: do you have a definitive answer? ;-)

As far as I know the various UPnP devices have been committed but they may not have been integrated as all our devices are built on our bash based DCEwhisperer framework (which was also committed but not integrated as far as I know either)

Andy

51
Users / Re: thinking about new av receiver
« on: August 14, 2014, 01:51:19 pm »
Isn't it possible to link the software volume control to the hardware volume control in a sense that when one of the software volume levels is at max level it will increase the hardware volume and decrease the software volume of the other (not being increase) channels so they stay at the same actual audible audio level?

I can't see that working to be honest. But even if it could be made to work the only advantage I can see is that all the signals are transmitted down a single cable. It seems a lot of complexity for no real gain.

52
Users / Re: thinking about new av receiver
« on: August 13, 2014, 06:57:57 pm »
I am using pulseaudio. there you can make a virtual device for each channel and then control the volume of that channel. Off course then you not using the volume of the receiver (maybe with some smart calculation that can be linked together).

...thats the point you can't control the Amp gain levels for discrete channels or pairs of channels especially as the Amp is receiving a 5.1/7.1 signal down the HDMI (I assume?)

Andy

53
Developers / Re: squeezelite as an alternative to squeezeslave
« on: August 13, 2014, 06:27:46 pm »
I would use the existing DT otherwise you will have to change code since lmce needs to know that it is usable for playing audio. Thom pointed me to it and I made the changes but probably did something wrong cause it did not work and I ended up using the existing DT. Since phenigma wants to replace squeezeslave I see even less need to use another DT since otherwise you end up removing the same code afterwards ;-).

If possible please add a optional configuration to be able to specify pulseaudio sinks. Has no influence on the way it works for normal usage and might be useful later on (I use it myself since I run on pulseaudio).

We use a separate DT in Dianemo for SqueezeLite as we need to keep existing installations working with Squeezslaves.

Andy

54
Users / Re: thinking about new av receiver
« on: August 13, 2014, 04:59:16 pm »
cause I don't have a receiver with ethernet yet (mine are still from the RS-232 age). However on my desktop I use a hdmi cable to my amp which has 7.1 and by selecting a virtual output on my desktop I can select which channel of the 7.1 is playing sound. So if an amp can stream 7.1 / 9.2 then the only question is what is the best way to stream that to the amp and to control and automate that. Starts to sound like a nice feature ;-).

The problem is that if you inject a 5.1 or 7.1 signal into your Amp over HDMI (or optical, Coax) you have no volume control over the individual channels that make up that signal.

Andy

55
Users / Re: thinking about new av receiver
« on: August 13, 2014, 12:37:06 pm »
Me neither. I guess one could get the mux part working, however, all receivers that I know, do not have the ability to define multiple zones from a single digital stream. You might be able to output sound to each of the speakers, but wont be able to change the volume using zone volume commands.

I've never seen a receiver that can define a multiple zones from a single stream either and yes volume control would be a problem too.

56
Users / Re: Power saving topic
« on: August 13, 2014, 12:33:37 pm »
Re Motion: http://wiki.linuxmce.org/index.php/Advanced_IP_Camera allows this on LinuxMCE atm.

Re UPnP Controller @LinuxMCE: As far as I know, the code has not been committed into LinuxMCE atm. I might have missed it, but I do not think so. totallymaxed probably has the definitive answer.

Thats a different approach, but an interesting one. In this approach each 'Advanced IP Camera' manages its own events and services and motion is not running centrally at all. Some dedicated IP Camera PVR's (eg Camera Servers) also manage their IP based cameras this way too with the PVR simply getting events from each Camera when something happens local to the camera.

In the future Dianemo we will have a 'External Camera Server' device that will get events from the Camera Server when a motion event is triggered by a Camera. How the camera server manages its cameras is abstracted in our case. The objective is to lessen the load on the Core and therefore allow us to shrink the hardware required at the Core, and lower energy usage. At the same time this improves reliability and is more fault tolerant too.

All the best

Andy

57
Users / Re: thinking about new av receiver
« on: August 13, 2014, 12:22:29 pm »
I assume, espergu wants to mux the output of multiple squeezeslaves on the core into a single digital stream that gets send out to the AV receiver over HDMI or TOSLINK, and wants the receiver to output the received data into the various output zones.

Yep...I got that. But I'm not aware of anyone successfully implementing that...unless espergu plans to give it a try?

58
Users / Re: thinking about new av receiver
« on: August 13, 2014, 11:35:47 am »
So some kind of streaming server with multichannels is required on the core? Or does something like that already exist? That software can then present itself as multiple output devices on the core (to which squeezeslave/lite can output) and then combine these outputs into one (multichannel) stream that is played by the receiver. Then on the receiver it gets played on the relevant channels since each channel is linked to a speaker.

Does that make sense?

...not quite.

You can run multiple Squeezeslaves on your Core. Each Squeezeslave will need to use a dedicated stereo sound output - these can be USB based or in a Expansion slot. You can also use multiple output soundcard, say a 5.1 card, and 'virtualise' it into 3x stereo cards. Once you have a Squeezeslave/Sound Card combo you can use that to play anything a real Squeezebox can play - simply connect its stereo audio output to an stereo audio input on an amplifier + speakers and you have a working mulit-channel sound capability.

All the best

Andy

59
Users / Re: Power saving topic
« on: August 13, 2014, 11:27:03 am »
Very interesting!

Obviously I would never dream of saying "oh how I wish upnp media players were supported by LMCE, when will it be sorted" because that would simultaneously violate rules #1 and #2, but if they were supported I would certainly consider it from what you are saying.

I guess if you run cameras then you will always struggle to downsize your core? Is there anything else that would make down-sizing the core difficult?

Cheers,
Matt.

Yes running motion on your Core (or any system) will increase energy usage. But as UpdateMedia is running continuously essentially its likely that motion has a lesser effect energy wise. In Dianemo we are transitioning away from hosting motion on the Core and will start using small, low energy, dedicated IP Security camera PVR's. These use 5-10 watts each and are Linux based with typically ARM based processors. The Core then interacts with the Camera PVR over IP for control and events/alerts. This is all part of a general trend we have of moving these kinds of services off to dedicated low energy hardware and focusing the Core back to providing the logic & control.

All the best

Andy

60
Users / Re: Power saving topic
« on: August 12, 2014, 05:16:10 pm »
I did a quick back-of-a-fag-packet calculation for a 20W MD left on continuously for a month, and it costs about 65p. Mine switches on/off (OK, hibernates) when I arm/disarm the alarm, so it will be even less than that. Barely seems worth worrying about? Of course I am talking purely from a monetary point of view here, there is always the green argument.

However I appreciate it allows you to downsize the core, I have no idea how much juice my core is using.

Cheers,
Matt.

The overall savings are in fact much greater especially if your system is larger. If you assume about 80-90p per MD, plus say 450p savings on downsized Core, plus less cooling/ventilation in the centralised cupboard/rack of say 120p & less wear & tear on the other equipment in the cupboard/rack too - say another 55p. Add all that up and you have a saving of just over 700p per month or approx £85 UKP per year (these are all based on real numbers from real installations).

Obviously all installations are unique and no two properties are the same but some simple incremental savings can be delivered easily by making the right choices. We have customers who are saving in the region of £185 - £225 UKP per year ($300 to $360 USD per year) by doing this. Small savings can quickly accumulate.

Obviously there are other incremental effects like being able to forget about ventilation when your media player only outputs 0.7w of heat, compactness, simplicity & very importantly reliability - which is a big one in fact. Reducing complexity by removing MD's is effective at improving reliability for sure. When we looked at reliability events the most common issues were related to MD's. So removing them not only saves energy costs (and purchase costs relating to the MD hardware too) but also improves overall system reliability too.

All the best

Andy

Pages: 1 2 3 [4] 5 6 ... 296