|
|
Post by lukewayne on Mar 22, 2020 13:55:21 GMT -5
Can someone explain what this black box is doing that the laptop running the Dirac software can’t do? If it’s just a network device, the laptop is sending all the data to this black box. what could it possibly be doing that the laptop couldn’t do even if it was using software emulation of a hardware device ... just seems like it’d be simpler / cheaper.
|
|
|
|
Post by geebo on Mar 22, 2020 14:27:46 GMT -5
Can someone explain what this black box is doing that the laptop running the Dirac software can’t do? If it’s just a network device, the laptop is sending all the data to this black box. what could it possibly be doing that the laptop couldn’t do even if it was using software emulation of a hardware device ... just seems like it’d be simpler / cheaper. I'm curious too but I believe that if it was something as simple as doing it with software then that's what they would have done. There are some pretty smart folks at Emotiva. It could even be that the hardware device can do whatever it does better and/or faster than software on an unknown computer with different OS's. |
|
|
|
Post by jonstatt on Mar 22, 2020 15:11:28 GMT -5
Indeed. But in these difficult times, I do wonder how long this dongle will take to get to me in the UK considering that it was bought from a dealer and Emotiva have no idea who or where I am. I expect that a number of these modules will have to work their way to the UK distributor, to the dealership, to me which the way things are will probably take months. So the frustration over the dongle is more that it creates a logistical issue and time delay than if all it needed was a firmware update. Hi John, We don't have all the logistics sorted out mainly because of the global pandemic, and you're right, there are a lot of unknowns at the moment regarding timing which is out of our control. I will say we've done hardware upgrades for our XMC-1 processors, and we dealt with our international customers directly for that, even the ones that purchased from a distributor. So we'll likely do something similar. Please reach out to us to register your product at sales@emotiva.com, and also elect into our email marketing for when we have news to share. Thank you so much for the direct response. I realise we are in uncharted waters with the pandemic and I am still getting my head around how this affects everything in our daily lives. I will register my RMC-1L as you suggested! Wishing you and the Emotiva team to stay safe and healthy. |
|
KeithL
Administrator 
Posts: 10,261
|
Post by KeithL on Mar 22, 2020 15:33:11 GMT -5
Geebo has the right answer....
The "little black box" is acting as a network interface and making some adjustments to the format of the data that are easier to do with a computer than with a DSP. (I wasn't directly involved in the programming so I don't know the exact details there.)
The Dirac software itself is now standardized... and it will be up to Dirac to make sure that it continues to work reliably on a wide variety of different computers.
If we were to create a software module to do this, it would have to work on all of those computers, and keep working after both Dirac and operating system updates..... (At a minimum this would mean new versions, and new issues, every time Microsoft or Apple issued an update...)
From our perspective it's easier to do the entire thing as a separate piece of hardware... that is compatible with but separate from all that complexity.
And, since you already have to install Dirac software on your computer, and plug in a microphone, plugging in one more box doesn't change the user experience very much.
Can someone explain what this black box is doing that the laptop running the Dirac software can’t do? If it’s just a network device, the laptop is sending all the data to this black box. what could it possibly be doing that the laptop couldn’t do even if it was using software emulation of a hardware device ... just seems like it’d be simpler / cheaper. I'm curious too but I believe that if it was something as simple as doing it with software then that's what they would have done. There are some pretty smart folks at Emotiva. It could even be that the hardware device can do whatever it does better and/or faster than software on an unknown computer with different OS's. |
|
KeithL
Administrator
Posts: 10,261
|
Post by KeithL on Mar 22, 2020 15:57:14 GMT -5
Let me answer a few more specific questions here in one place....
*
Yes, it will be "a little black box" that "just has to be connected somewhere on your network. If your RMC-1 and computer are already plugged into a single network then you will just have to connect this box into that same network. We're providing the switch for those of you who don't have enough network ports handy. (We are talking about a local network here... it won't work over the Internet.)
*
However, just to be clear, the box will have to be connected to the same network segment as your computer and your processor. Having one or more switches between the processor, computer, and Dirac Live Network Interface box shouldn't be a problem. HOWEVER, having more complex devices, like routers, firewalls, network "extenders", or "bridges" in between may be more problematic.
(We're providing a switch so that, no matter what else happens, you will always be able to connect things the way we know will work.)
* The version of Dirac you'll be getting is Dirac Live 2.xx. (It will be "the current version" on the day you download it and you will be able to download updates.)
Dirac has decided to make all of the software the same... for everyone... This is basically equivalent to the original "Dirac Full"... with user editable Target Curves... and support for third party microphones...
Some of the "advanced bass management features" are going to be "cost extra options" that must be purchased through Dirac. (Again, Dirac is now the same for everyone, so that is all up to their roadmap.)
* As far as I know all Dirac Live Room Correction is referred to as that... and the current version is 2.xx . From what I know of "Dirac Unison" it is currently only used in cars (where it is custom integrated with specific speakers installed in a specific car model). However, many of the originally described benefits of Dirac Unison, including being able to calibrate multiple subwoofers to work together, are now part of Dirac Live. (I would refer you to Dirac's website and latest white papers for details there.)
* Everybody gets the little Ethernet switch... whether you need it or not. (Ethernet switches are very cheap these days - and you might need it.)
|
|
KeithL
Administrator
Posts: 10,261
|
Post by KeithL on Mar 22, 2020 15:58:03 GMT -5
Yup. OK, the most obvious question! Or did I miss the answer? When will Emotive be shipping Dirac and its little black box ?The simple answer is that Emotiv e will never ship these. Emotiva will, though.  But seriously... KeithL, let me make sure I understand. As with the old version of DIRAC with the XMC-1, DIRAC software will reside on a PC and we'll plug a mic into that. On the same network as the PC, the PC and the new "black box" will be in communication with DIRAC's systems to send data to/from the PC and/or black box - then end up with filters and relay those to the RMC-1/RMC-1L/XMC-2 (where the filters are stored for use) Is that basically it? Mark |
|
KeithL
Administrator
Posts: 10,261
|
Post by KeithL on Mar 22, 2020 16:02:13 GMT -5
We don't anticipate ever needing firmware updates. (However, all of the firmware is contained on an internal micro-SD card, which could potentially be replaced in a few seconds.)
The interface is necessary for the computer to talk to the processor..... but only when you're taking measurements or downloading filters. (The interface box is just a standard commercial processor, actually a Raspberry Pi, in a standard box... what we've been waiting for was the completion of the code for it.)
Will this Emotiva Dirac Live Network Interface (EDLNI) require firmware updates and if so how would that be done? Do we need it send it back and have it reflashed? I’m coming from the perspective of how EDLNI would work when Dirac Live Bass Management is rolled out for $249 or $399? Maybe I don’t understand the functionality and what I’m asking isn’t making sense and EDLNI can handle Dirac Live Bass Management as is. Also, I’m assuming this now needs to be manufactured, packaged and shipped. Well this explains why we weren’t offered to beta test Dirac - wouldn’t make sense and wouldn’t be possible without this interface. |
|
|
|
Post by megash0n on Mar 22, 2020 16:07:14 GMT -5
Let me answer a few more specific questions here in one place....
*
Yes, it will be "a little black box" that "just has to be connected somewhere on your network. If your RMC-1 and computer are already plugged into a single network then you will just have to connect this box into that same network. We're providing the switch for those of you who don't have enough network ports handy. (We are talking about a local network here... it won't work over the Internet.)
*
However, just to be clear, the box will have to be connected to the same network segment as your computer and your processor. Having one or more switches between the processor, computer, and Dirac Live Network Interface box shouldn't be a problem. HOWEVER, having more complex devices, like routers, firewalls, network "extenders", or "bridges" in between may be more problematic.
(We're providing a switch so that, no matter what else happens, you will always be able to connect things the way we know will work.)
* The version of Dirac you'll be getting is Dirac Live 2.xx. (It will be "the current version" on the day you download it and you will be able to download updates.)
Dirac has decided to make all of the software the same... for everyone... This is basically equivalent to the original "Dirac Full"... with user editable Target Curves... and support for third party microphones...
Some of the "advanced bass management features" are going to be "cost extra options" that must be purchased through Dirac. (Again, Dirac is now the same for everyone, so that is all up to their roadmap.)
* As far as I know all Dirac Live Room Correction is referred to as that... and the current version is 2.xx . From what I know of "Dirac Unison" it is currently only used in cars (where it is custom integrated with specific speakers installed in a specific car model). However, many of the originally described benefits of Dirac Unison, including being able to calibrate multiple subwoofers to work together, are now part of Dirac Live. (I would refer you to Dirac's website and latest white papers for details there.)
* Everybody gets the little Ethernet switch... whether you need it or not. (Ethernet switches are very cheap these days - and you might need it.)
Thank you for addressing all those questions Keith. I'm sure we all are just anxious to get new toys. Hopefully in the coming days, your team will release some expectations on shipping. |
|
|
|
Post by megash0n on Mar 22, 2020 16:09:10 GMT -5
We don't anticipate ever needing firmware updates. (However, all of the firmware is contained on an internal micro-SD card, which could potentially be replaced in a few seconds.)
The interface is necessary for the computer to talk to the processor..... but only when you're taking measurements or downloading filters. (The interface box is just a standard commercial processor, actually a Raspberry Pi, in a standard box... what we've been waiting for was the completion of the code for it.)
Will this Emotiva Dirac Live Network Interface (EDLNI) require firmware updates and if so how would that be done? Do we need it send it back and have it reflashed? I’m coming from the perspective of how EDLNI would work when Dirac Live Bass Management is rolled out for $249 or $399? Maybe I don’t understand the functionality and what I’m asking isn’t making sense and EDLNI can handle Dirac Live Bass Management as is. Also, I’m assuming this now needs to be manufactured, packaged and shipped. Well this explains why we weren’t offered to beta test Dirac - wouldn’t make sense and wouldn’t be possible without this interface. I had a suspicion it was a Pi. 😉 If I'm not mistaken, the chips are similar to what others use in their processors, correct? |
|
KeithL
Administrator
Posts: 10,261
|
Post by KeithL on Mar 22, 2020 16:16:19 GMT -5
OK, to be fair, the bit about "sending the data to Sweden where it would be processed by secret servers" was a bit of marketing hyperbole...(not ours) The Dirac software on the computer takes all the measurements and performs all the calculations to create the filters.
(It connects to the server to validate your license.) It then downloads the filters it has created into the "DSP engine" in the processor. Then the processor runs those filters, using its internal "Dirac engine" to perform the actual real-time corrections.
With the new one, the measurements and calculations are still done by the software in your computer. And an Internet connection is still required so the software can validate your license with the license server. As before, after creating the filters, the computer software downloads the filters it has created into the "DSP engine" in the processor.
And the processor then runs those filters on its internal "Dirac engine".
The interface box is strictly there to receive the filters from the Dirac software, convert them into the format we need, and pass them on to the processor.
And we used a separate box because..... - the programming involved is just not convenient to do on a DSP (and the DSPs on the RMC-1 don't have a lot of extra real estate for that sort of thing anyway) - if we did it on the computer we'd have to have all different versions for different computers, and we'd have to keep them updated, so doing it this way is easier
I have a guess but only a guess. If I recall there was a statement about the processor used in the devices causing some hiccups. Could it be the black box is a way to offload the measurements into a easier processor solution for Dirac then passes the info in a way that the rmc can handle it? This would seem logical if that’s the case and a smart move by emotiva in order to get Dirac working within a realistic timeframe No complaints from me on the implementation. Looking forward to getting Dirac back since my XMC-1 tradein. I'm only puzzled because as I understood the XMC-1 Dirac process, the Dirac PC software communicated with the processor to run the test sweeps but data collected stayed in the PC and was uploaded to Dirac in Sweden. In order to maintain secrecy of the filter design algorithms, Dirac analyzes the data on their server and downloads filter parameters to the PC which in turn sends the filter parameters to the processor ... once. There is no ongoing/complex processing between the PC and the RMC/XMC. |
|
KeithL
Administrator
Posts: 10,261
|
Post by KeithL on Mar 22, 2020 16:28:20 GMT -5
That depends on what processors you mean....
The processor in the Raspberry Pi is quite similar to the processor found in a powerful modern smart phone. It also has a lot of useful built-in features, including Ethernet, WiFi, USB, and 4k video.
You can build one up into a very nice video client using something like Plex... And my music player at home is a Raspberry Pi running Volumio...
However, they're more of a general purpose than a dedicated processor, so you won't see one powering an A/V processor like the RMC-1. You could design a version of the RMC-1 that used a Raspberry Pi and its management processor but you wouldn't use it for the audio DSP. (They're a little light on the raw processing power for that...)
We don't anticipate ever needing firmware updates. (However, all of the firmware is contained on an internal micro-SD card, which could potentially be replaced in a few seconds.)
The interface is necessary for the computer to talk to the processor..... but only when you're taking measurements or downloading filters. (The interface box is just a standard commercial processor, actually a Raspberry Pi, in a standard box... what we've been waiting for was the completion of the code for it.)
I had a suspicion it was a Pi. 😉 If I'm not mistaken, the chips are similar to what others use in their processors, correct? |
|
|
|
Post by makutaku on Mar 22, 2020 16:36:58 GMT -5
if we did it on the computer we'd have to have all different versions for different computers, and we'd have to keep them updated, so doing it this way is easier Could Emotiva release the source code for the community to port it for the PC/Mac/Linux/Docker, and maintain it for you ? |
|
|
|
Post by marcl on Mar 22, 2020 16:46:53 GMT -5
OK, to be fair, the bit about "sending the data to Sweden where it would be processed by secret servers" was a bit of marketing hyperbole...(not ours) The Dirac software on the computer takes all the measurements and performs all the calculations to create the filters.
(It connects to the server to validate your license.) It then downloads the filters it has created into the "DSP engine" in the processor. Then the processor runs those filters, using its internal "Dirac engine" to perform the actual real-time corrections.
With the new one, the measurements and calculations are still done by the software in your computer. And an Internet connection is still required so the software can validate your license with the license server. As before, after creating the filters, the computer software downloads the filters it has created into the "DSP engine" in the processor.
And the processor then runs those filters on its internal "Dirac engine".
The interface box is strictly there to receive the filters from the Dirac software, convert them into the format we need, and pass them on to the processor.
And we used a separate box because..... - the programming involved is just not convenient to do on a DSP (and the DSPs on the RMC-1 don't have a lot of extra real estate for that sort of thing anyway) - if we did it on the computer we'd have to have all different versions for different computers, and we'd have to keep them updated, so doing it this way is easier
No complaints from me on the implementation. Looking forward to getting Dirac back since my XMC-1 tradein. I'm only puzzled because as I understood the XMC-1 Dirac process, the Dirac PC software communicated with the processor to run the test sweeps but data collected stayed in the PC and was uploaded to Dirac in Sweden. In order to maintain secrecy of the filter design algorithms, Dirac analyzes the data on their server and downloads filter parameters to the PC which in turn sends the filter parameters to the processor ... once. There is no ongoing/complex processing between the PC and the RMC/XMC. Aha I was sucked in by marketing hyperbole! Not the first time and likely not the last :-) Thanks for the details Keith! Looking forward to the solution as soon as the world doesn't end. |
|
|
|
Post by jonstatt on Mar 22, 2020 16:51:51 GMT -5
What I am still curious to understand is what is different about the new architecture that required the off boarding of this processing but the XMC1 didnt.
|
|
|
|
Post by rbk123 on Mar 22, 2020 16:52:01 GMT -5
Let me answer a few more specific questions here in one place..<snip>... Excellent info - maybe compile this post and the several other questions you answered into a Dirac 2 FAQ. Lot's of good info that people will inevitably re-ask countless times over the next several months. Pretty cool use of a Raspberry Pi, imo. |
|
|
|
Post by rbk123 on Mar 22, 2020 16:57:49 GMT -5
What I am still curious to understand is what is different about the new architecture that required the off boarding of this processing but the XMC1 didnt. Technically they weren't required to off board the processing; it's just a more expedient solution for them on several fronts. However to answer your real question, it's a completely different version of Dirac, coupled with the new architecture has completely different DSP's. I'd wager the RMC wouldn't need this device were it to run the orignal Dirac1 that the XMC-1 runs. |
|
KeithL
Administrator
Posts: 10,261
|
Post by KeithL on Mar 22, 2020 17:00:25 GMT -5
I don't think that is anywhere near a real possibility....
The Dirac software itself is licensed, as is its interface, and our reputation rests on how our RMC-1 sounds when you run it, and on how well it works... If the software in the interface were to fail to function correctly, it has the potential to cause the RMC-1 to crash, or to adversely affect the results of a room calibration... Besides that, we all know that, when that open source program bricks your RMC-1, we're going to be the ones getting the support call...
I really don't see much of an upside there... Avoiding occasionally having to connect a small network interface box - which you're receiving at no cost - hardly seems worth the bother.
I should point out that we DO have an interface which allows you to enter PEQ parameters directly into the RMC-1... (Not only is it openly available but you can see all of the internal workings simply by examining the XML files you get when you "Export Filters".)
If you really want a serious open-source project then you might consider an open source alternative to Dirac itself... Or even a conversion program to allow the "XMC-1" export option on REW to work with the RMC-1. (We've heard that the author of REW is too busy to add a specific entry for our new processors.)
if we did it on the computer we'd have to have all different versions for different computers, and we'd have to keep them updated, so doing it this way is easier Could Emotiva release the source code for the community to port it for the PC/Mac/Linux/Docker, and maintain it for you ? |
|
|
|
Post by deewan on Mar 22, 2020 17:03:42 GMT -5
Will Emotiva share with us an estimate of when they think they may be able to ship use this blackbox? I'm always fearful of when Keith starts sharing information but avoids timeline questions. Usually means we are annual quarters away, not weeks or months.
Maybe it's just me, but if you say Dirac, Success at last, that should mean the feature is ready to be released to the owners. Not just a communication that you are able to get it to work on some test units.
|
|
KeithL
Administrator
Posts: 10,261
|
Post by KeithL on Mar 22, 2020 17:04:43 GMT -5
That would be a question for Lonnie...
All I can tell you is that our new processors use the Griffin Lite processor...
While all of the other pre/pros that support Dirac Live use processors in the same family as the XMC-1...
And it was porting the "Dirac engine" code over to the Griffin Lite that proved to be so difficult...
But, for whatever reason, even when the code was ported successfully, the interface was apparently still an issue...
What I am still curious to understand is what is different about the new architecture that required the off boarding of this processing but the XMC1 didnt. |
|
|
|
Post by megash0n on Mar 22, 2020 17:18:21 GMT -5
That depends on what processors you mean....
The processor in the Raspberry Pi is quite similar to the processor found in a powerful modern smart phone. It also has a lot of useful built-in features, including Ethernet, WiFi, USB, and 4k video.
You can build one up into a very nice video client using something like Plex... And my music player at home is a Raspberry Pi running Volumio...
However, they're more of a general purpose than a dedicated processor, so you won't see one powering an A/V processor like the RMC-1. You could design a version of the RMC-1 that used a Raspberry Pi and its management processor but you wouldn't use it for the audio DSP. (They're a little light on the raw processing power for that...)
I had a suspicion it was a Pi. 😉 If I'm not mistaken, the chips are similar to what others use in their processors, correct? After reading your posts, I was wrong about why you are using the Pi... So, your responses make sense. This opens up more questions though... Will we have the ability to run other software on the Pi, or will this be locked down? |
|
