geebo
Emo VIPs
"Too bad that all the people who know how to run the country are driving taxicabs and cutting hair"
Posts: 24,211
|
Post by geebo on Jan 28, 2020 16:05:31 GMT -5
I'm going to offer you some off-the-cuff reality here.... The portion of the firmware code dedicated to processing the Dirac correction filters takes up a significant amount of memory and uses a significant amount of processing power...
Therefore, the sample rate at which it will be able to operate depends on how powerful the processor itself is, and on how the code is written.
The predictions we made early on were based largely on the information we received from Dirac - about the code we then expected to receive from them. Since then, a lot has changed, and the code we finally received from them was quite different than what was originally promised or expected. Because of this, some of our early expectations, based on information that is now outdated, may also have to be changed.
Early on, we had the choice of offering you the best information we had at the time, or of saying nothing, and hearing endless complaints that we don't share enough information with our customers. As usual, we chose to share what we knew at the time... and, as sometimes happens, previously unexpected changes may force us to revise those early expectations.
We will provide you with the final details as soon as we finalize the Dirac code and the interface that connects it to the RMC-1. (We are finally very close to this... so please bear with us just a little bit longer.)
I will again remind everyone that the purpose of Dirac is to make significant changes to your audio signal - all aimed toward improving the sound of your system. If you really want to listen to your original high-res audio sources, with the utmost accuracy, you will use Reference Stereo mode, and bypass Dirac along with all the other available types of processing.
And, if you do choose to run Dirac, because you expect it to deliver a significant improvement, then you should focus on the results, rather than on the details about how they are achieved. On this topic is there a stereo mode that can be used with Subs enabled and no Dirac. From what I'm hearing this is going to be what I'm looking for at least for music playback... Direct mode should do what you want. Minimal processing with bass management. From the manual: Direct Mode Direct Mode provides a relatively pure listening experience, and eliminates most processing, but still retains bass management. In Direct Mode: • The output contains the same number of channels as the input signal. • No extra channels are synthesized; if the output channel that corresponds to a given input channel is not present, that channel signal is mixed into the appropriate available channels. • Bass management IS available and the subwoofer IS active. • You CANNOT use the Loudness Control, or the Tone Presets, or the Tone Trims • You CANNOT use Dirac Live or the manual Parametric EQ Presets. • Level Trims, which operate in the analog domain, are still available. • Speaker Distance adjustments are available for both analog and digital inputs. • Stereo digital signals are presented directly to the processor. • Stereo analog input signals are converted to digital and passed to the processor. • Surround sound digital signals are decoded, and passed to the processor. • Multi-channel discrete digital signals (like multi-PCM) are passed directly to the processor. • All signals receive bass management - but no other processing. • If an LFE signal is present, and subwoofers are also present, the LFE content will be routed to the subs along with low-frequency content from the main channels (as determined by the bass management settings).
|
|
|
Post by andersmi on Jan 28, 2020 16:09:11 GMT -5
I'm going to offer you some off-the-cuff reality here....
The portion of the firmware code dedicated to processing the Dirac correction filters takes up a significant amount of memory and uses a significant amount of processing power...
Therefore, the sample rate at which it will be able to operate depends on how powerful the processor itself is, and on how the code is written.
The predictions we made early on were based largely on the information we received from Dirac - about the code we then expected to receive from them. Since then, a lot has changed, and the code we finally received from them was quite different than what was originally promised or expected. Because of this, some of our early expectations, based on information that is now outdated, may also have to be changed.
Early on, we had the choice of offering you the best information we had at the time, or of saying nothing, and hearing endless complaints that we don't share enough information with our customers. As usual, we chose to share what we knew at the time... and, as sometimes happens, previously unexpected changes may force us to revise those early expectations.
We will provide you with the final details as soon as we finalize the Dirac code and the interface that connects it to the RMC-1. (We are finally very close to this... so please bear with us just a little bit longer.)
I will again remind everyone that the purpose of Dirac is to make significant changes to your audio signal - all aimed toward improving the sound of your system. If you really want to listen to your original high-res audio sources, with the utmost accuracy, you will use Reference Stereo mode, and bypass Dirac along with all the other available types of processing.
And, if you do choose to run Dirac, because you expect it to deliver a significant improvement, then you should focus on the results, rather than on the details about how they are achieved. MiniDSP is able to run 24/96 so Dirac is able to do this. www.minidsp.com/dirac-seriesPlease explain why the Rmc-1 suddenly can't do this when you first expected it to be able to run 192 then 96 and now 48? Can we expected it to be able to do 96 in stereo like the MiniDSP and NAD 758v3?
|
|
|
Post by mikoz on Jan 28, 2020 16:41:46 GMT -5
I agree.... a DAC with a 140 dB S/N is technically superior to one with a 130 dB S/N.... in terms of S/N. And a DAC that has a better DNR/SNR/THD is superior.... in terms of DNR, SNR, and THD.
(As long as some other important performance characteristic wasn't sacrificed to get those numbers.)
However, there are still the twin questions of context and the point of diminishing returns.
It's also obvious to me that, from a technical perspective, a DAC that is flat to 10 mHz is superior to one that is only flat to 50 kHz... Yet, for some reason, you didn't mention that at all. Could the reason be that, in that case, you agree that the difference would be totally pointless in an audio application?
Unfortunately, when it comes to DACs, we still lack a complete understanding of how all the measurements correlate to what we hear. You can compare two DACs, both of which have excellent frequency response, noise performance, and THD... and find that they sound quite different. Then, when you compare their response to transients, you find that they have very different filter characteristics. This is NOT "some esoteric thing that some people imagine is there". The differences are quite obvious and easy to observe on an oscilloscope trace. And, yes, when people listen to them, they hear differences... (So far this only suggests that the differences we're hearing correlate to the filter differences we're measuring.)
You seem to have chosen to focus on a few very specific measurements... But, when it comes to DACs, those few don't seem to be quite sufficient to completely describe the situation of what we hear... So perhaps it's not "safe to set all the other stuff aside"..... Hmmmmm.....
It's very simple to me. all other factors aside, a DAC that has a better DNR/SNR/THD is a superior DAC... it really is that simple. Now the rest of the picture is a function of the quality of the analog design, any "tricks" played in filtering, and a purely subjective "feeling". We then get into differences that people describe in esoteric terns, but things you can measure on a BENCH with a spectrum analyzer... those are real differences. Noise multiples throughout a system, so, things that start out small can get bigger.
That's because DAC flatness isn't commonly specified. It requires either a multitone or a swept source and is a time consuming test. Neither are things like INL or DNL specified commonly. But as to flatness, most people like exaggerated highs and lows anyway... this sort of thing is easily controllable later with filtering.
|
|
|
Post by thxultra on Jan 28, 2020 16:55:33 GMT -5
On this topic is there a stereo mode that can be used with Subs enabled and no Dirac. From what I'm hearing this is going to be what I'm looking for at least for music playback... Direct mode should do what you want. Minimal processing with bass management. From the manual: Direct Mode Direct Mode provides a relatively pure listening experience, and eliminates most processing, but still retains bass management. In Direct Mode: • The output contains the same number of channels as the input signal. • No extra channels are synthesized; if the output channel that corresponds to a given input channel is not present, that channel signal is mixed into the appropriate available channels. • Bass management IS available and the subwoofer IS active. • You CANNOT use the Loudness Control, or the Tone Presets, or the Tone Trims • You CANNOT use Dirac Live or the manual Parametric EQ Presets. • Level Trims, which operate in the analog domain, are still available. • Speaker Distance adjustments are available for both analog and digital inputs. • Stereo digital signals are presented directly to the processor. • Stereo analog input signals are converted to digital and passed to the processor. • Surround sound digital signals are decoded, and passed to the processor. • Multi-channel discrete digital signals (like multi-PCM) are passed directly to the processor. • All signals receive bass management - but no other processing. • If an LFE signal is present, and subwoofers are also present, the LFE content will be routed to the subs along with low-frequency content from the main channels (as determined by the bass management settings).Ya that is what I'm looking for thanks!!
|
|
|
Post by davidl81 on Jan 28, 2020 16:57:03 GMT -5
I posted it a lot more times. Here are the official promises of Emotiva at that time. Auro 3D, Dirac with 16 channels 192 Khz. So don't put it on Dirac and stick to your commitments! I would say what did Emotiva have on their website about the RMC-1 the day it was available for sale? To me that is what they promised. I get that they had a press release about a product two years prior to being on sale, but to say you bought that processor based on that information would be naive IMO. Now we can say that Emotiva got ahead of their skis and put out press releases about things they couldn't fulfill and that is accurate, but when someone went to buy the RMC-1 all that was promised was 16 channel (with future upgrades to 32), Atmos, DTS:X, and Dirac (coming soon). So to me if they deliver Dirac (which is sorta questionable honestly) and expansion modules (very questionable) then they fulfilled the promises they made at the time of sale.
|
|
|
Post by jagman on Jan 28, 2020 16:58:46 GMT -5
I don’t think you will see anything like additional processing power on an expansion board. I think when the time comes that these units need more power Emotiva will just release a RMC-2. The expansions are mostly going to be things like extra inputs, streaming module, and extra output channels. Also who knows if or when these things will actually come out. I did buy the RMC-1 myself, but part of the reason was I like the height of the RMC as it is flush with my XPA amps. I also had a 40% off card so the cost difference was only $1200. To me $2000 is a pretty tough jump to make from the XMC-2 to the RMC-1. Thank you. For me, if you're correct, the only benefit of the RMC-1 is the extra channels afforded by a 4 channel output card, and, as you mentioned, the aesthetics. I don't see myself going beyond 9.5.6 or 11.3.6 (I'm currently 9.3.6). If they add Dirac's bass management software then having all the subs under the control of the RMC-1 would probably help, too. I don't think it will happen, but if they add Auro-3D then it would be easier to add a VOG speaker as well as front and side heights (I'm actually wired for those channels even though I don't have speakers there). The more I think about it that flexibility would be nice 😁. The main drawback is not being able to apply BEQ files (if that even makes a difference). Emotiva replied to my question about adding additional processing via an expansion bay... "Yes, the architecture of the RMC-1 is designed in such a way that we have the ability to add DSPs as needed to increase the processing horsepower of the unit for the future. Right now it has two of the fastest DSPs Analog Devices makes, but it can be expanded to virtually any number is needed via the expansion bays. While the architecture is not limited to any specific number of DSPs, I would think fourteen would be a practical limit. Right now, I can't imagine anything that would need more then maybe two additional DSPs, but the capabilities are certainly there to do go crazy should we need too. I hope this helps and best regards, Lonnie Vaughn"
|
|
|
Post by jagman on Jan 28, 2020 17:00:43 GMT -5
Thank you. For me, if you're correct, the only benefit of the RMC-1 is the extra channels afforded by a 4 channel output card, and, as you mentioned, the aesthetics. I don't see myself going beyond 9.5.6 or 11.3.6 (I'm currently 9.3.6). If they add Dirac's bass management software then having all the subs under the control of the RMC-1 would probably help, too. I don't think it will happen, but if they add Auro-3D then it would be easier to add a VOG speaker as well as front and side heights (I'm actually wired for those channels even though I don't have speakers there). The more I think about it that flexibility would be nice 😁. The main drawback is not being able to apply BEQ files (if that even makes a difference). Emotiva replied to my question about adding additional processing via an expansion bay... "Yes, the architecture of the RMC-1 is designed in such a way that we have the ability to add DSPs as needed to increase the processing horsepower of the unit for the future. Right now it has two of the fastest DSPs Analog Devices makes, but it can be expanded to virtually any number is needed via the expansion bays. While the architecture is not limited to any specific number of DSPs, I would think fourteen would be a practical limit. Right now, I can't imagine anything that would need more then maybe two additional DSPs, but the capabilities are certainly there to do go crazy should we need too. I hope this helps and best regards, Lonnie Vaughn" Which to me says processing ability doesn't necessarily need to be a limitation to doing 96khz Dirac.
|
|
|
Post by davidl81 on Jan 28, 2020 17:02:22 GMT -5
I don't understand what has people thinking Emotiva promised 96k w/Dirac. There's nothing to that effect on the product page, and at the last Emofest I recall the discussion being that Emotiva would like to have 96k but it wasn't final and was up to Dirac. I believe that was in my notes that were published here. Maybe there was some later interview somewhere with Dan commenting on it, but to me - if it's not on the product page as a feature when I place my order - I would not count on it or be disappointed if it's not 96k. I doubt it would make any difference in the sound anyway. Mark Let's say they never release any expansion slot modules... You shouldn't be upset over that either then based on your point of view. If they dont have channel expansion modules in two years time. I will sell everything Emotiva I own. I would say the expansion modules are different (specifically the 28 channel upgrade) since that has been mentioned Specifically by Emotiva well after the RMC-1 was released (as recently as the Jan 2020 youtube video) . They 100% promised that. But I have doubts that it will be delivered.
|
|
|
Post by davidl81 on Jan 28, 2020 17:14:33 GMT -5
Emotiva replied to my question about adding additional processing via an expansion bay... "Yes, the architecture of the RMC-1 is designed in such a way that we have the ability to add DSPs as needed to increase the processing horsepower of the unit for the future. Right now it has two of the fastest DSPs Analog Devices makes, but it can be expanded to virtually any number is needed via the expansion bays. While the architecture is not limited to any specific number of DSPs, I would think fourteen would be a practical limit. Right now, I can't imagine anything that would need more then maybe two additional DSPs, but the capabilities are certainly there to do go crazy should we need too. I hope this helps and best regards, Lonnie Vaughn" Which to me says processing ability doesn't necessarily need to be a limitation to doing 96khz Dirac. But would people be willing to pay for the hardware to get to that level (assuming that the current system does not have the power to get to that 96khz level)? I think that if the current RMC-1 can only do 48khz Emotiva would be better off just offering that (and taking the PR hit) then saying "hey we can do 96khz, but you have to buy this $400 expansion card".
|
|
|
Post by mikoz on Jan 28, 2020 17:18:50 GMT -5
I posted it a lot more times. Here are the official promises of Emotiva at that time. Auro 3D, Dirac with 16 channels 192 Khz. So don't put it on Dirac and stick to your commitments!
There again seems to be a fundamental mis-understanding of the customer-vendor relationship. For some reason, Emotiva is soliciting sympathy for the "Dirac issues" from people who paid thousands who continue to wait for the core promises to be delivered. Emotiva should instead direct any problems they have with Dirac at Dirac (but in reality it sounds like the RMC-1 is just underpowered to work with the higher sampling rates, but that's a different story) and thank the people that are still waiting for Dirac, still waiting for HDMI stability and update its users on what and when things will happen. This is the same thing we saw with the HDMI issues... call Directv and complain and we get the "we [Emotiva] can't fix the problems from other vendors", yet in reality this magical FW. if it lives up to expectations seems to suggest that the problems were wither on the Emotiva side all along or they just finally stopped refuting their users and will "workaround" the non-compliant devices like every other vendor. Probably the former, not the latter. Either way, it's the same tale of woes... meanwhile we're the ones who have shelled out the money.
To me, this experience feels like I'm backing a kickstarter project, only with those you tend to get more information and updates, at least generally speaking, oh, and you get the product cheaper than those who buy it later.
If Emotiva cannot deliver 96khz with the first Dirac release and/or cannot do it at all, this should now just be clearly communicated as delaying it will just cause more people to get upset. No discussions of why it's "not important" or whatever, just say yes or no. Of course, who am I to suggest that... no one... but it's the right thing to do. Since other vendors can do it, it's a problem that Emotiva owns. And if they can't deliver on the promises that were made over a year ago, then the users can just vote with their return labels if they don't like it... there seems to be no other outcome.
Yeah, I know... go pound sand.
|
|
|
Post by PaulBe on Jan 28, 2020 17:20:05 GMT -5
It's also obvious to me that, from a technical perspective, a DAC that is flat to 10 mHz is superior to one that is only flat to 50 kHz... Yet, for some reason, you didn't mention that at all. Could the reason be that, in that case, you agree that the difference would be totally pointless in an audio application? Unfortunately, when it comes to DACs, we still lack a complete understanding of how all the measurements correlate to what we hear. You can compare two DACs, both of which have excellent frequency response, noise performance, and THD... and find that they sound quite different. Then, when you compare their response to transients, you find that they have very different filter characteristics. This is NOT "some esoteric thing that some people imagine is there". The differences are quite obvious and easy to observe on an oscilloscope trace. And, yes, when people listen to them, they hear differences... (So far this only suggests that the differences we're hearing correlate to the filter differences we're measuring.)
Absolutely true. It is not very difficult to change the filters on the Oppo 205 using the app and hear the difference between say, minimum phase fast and linear phase fast. For some reason, the vast majority of people and reviewers ignore the obvious affect of filters. - Rich Yes. The filters on the oppo 205 sound different. I like linear phase fast. Good article: archimago.blogspot.com/2018/06/measurements-oppo-udp-205-part-1-output.htmlIt's in 3 parts.
|
|
|
Post by harmonium on Jan 28, 2020 17:59:21 GMT -5
How close is Firmware 1.8? If more than 2 weeks out I will install the beta 1.7.6 Thank you in advance. Where do I get the 1.7.6 firmware?
|
|
|
Post by Thunderduck on Jan 28, 2020 18:18:17 GMT -5
How close is Firmware 1.8? If more than 2 weeks out I will install the beta 1.7.6 Thank you in advance. Where do I get the 1.7.6 firmware? You need to request from Emotiva support.
|
|
klinemj
Emo VIPs
Honorary Emofest Scribe
Posts: 14,772
|
Post by klinemj on Jan 28, 2020 18:30:40 GMT -5
I don't understand what has people thinking Emotiva promised 96k w/Dirac. There's nothing to that effect on the product page, and at the last Emofest I recall the discussion being that Emotiva would like to have 96k but it wasn't final and was up to Dirac. I believe that was in my notes that were published here. Maybe there was some later interview somewhere with Dan commenting on it, but to me - if it's not on the product page as a feature when I place my order - I would not count on it or be disappointed if it's not 96k. I doubt it would make any difference in the sound anyway. Mark Let's say they never release any expansion slot modules... You shouldn't be upset over that either then based on your point of view. If they dont have channel expansion modules in two years time. I will sell everything Emotiva I own. Well, given what I know they are thinking of for expansion modules, I won't have a need for them. So, I would be buying either an RMC-1L or XMC-2 (leaning to an XMC-2 at this point). That said, what they say on the product page is this: "Expansion slots for additional optional custom Input/Output modules" Net, their promise is the slots. The "hope" is for something to fill them, but - of course - they have many times said they will have certain ones. Based on what I've seen, I'm pretty sure we'll see them. The only question to me is timing. Do I expect them within 2 years? You bet - and not that far into the future. I might be wrong, but I suspect as soon as we see Dirac launched that expansion modules will be the main priority. Based on what I've read from both sides, I don't think Emotiva expected they would have to pull as much weight on Dirac as they are, so other things are taking a back seat. (My guess...) Mark
|
|
|
Post by mfeust on Jan 28, 2020 18:40:21 GMT -5
I have no dog in this fight and I am very glad I am more than content with my 2.2ch system. But if I was an early adaptor to this Processor there's no f---kn way I want Dirac available until this unit is 100% stable. So again if it was me I would be says nothing about Dirac and everything about the other problems.
Mark
|
|
|
Post by aswiss on Jan 28, 2020 19:28:53 GMT -5
I'm going to offer you some off-the-cuff reality here....
The portion of the firmware code dedicated to processing the Dirac correction filters takes up a significant amount of memory and uses a significant amount of processing power...
Therefore, the sample rate at which it will be able to operate depends on how powerful the processor itself is, and on how the code is written.
The predictions we made early on were based largely on the information we received from Dirac - about the code we then expected to receive from them. Since then, a lot has changed, and the code we finally received from them was quite different than what was originally promised or expected. Because of this, some of our early expectations, based on information that is now outdated, may also have to be changed.
Early on, we had the choice of offering you the best information we had at the time, or of saying nothing, and hearing endless complaints that we don't share enough information with our customers. As usual, we chose to share what we knew at the time... and, as sometimes happens, previously unexpected changes may force us to revise those early expectations.
We will provide you with the final details as soon as we finalize the Dirac code and the interface that connects it to the RMC-1. (We are finally very close to this... so please bear with us just a little bit longer.)
I will again remind everyone that the purpose of Dirac is to make significant changes to your audio signal - all aimed toward improving the sound of your system. If you really want to listen to your original high-res audio sources, with the utmost accuracy, you will use Reference Stereo mode, and bypass Dirac along with all the other available types of processing.
And, if you do choose to run Dirac, because you expect it to deliver a significant improvement, then you should focus on the results, rather than on the details about how they are achieved. If the RMC-1 can handle Dirac with 48k on Multichannel (which is enough for movies, because majority of the movies just don't have higher rates) then it should also be able to do 96k for the Stereo part (2-Ch). But maybe it's also a limitation of the license. All complain about the 48 - how about Bass Management? We will find out what we get, once we get it.
I use Roon together with my CocktailAudio X45 - capable to go up to DSD128 (with upconversion if needed) - connected with XLR to Balanced in.
My Room is optimised for acoustics, so except movies, I would not use Dirac.
I wait until I get the new FW. If it's better than now, solving my issues - I'm fine - Rest is Addons for me.
If it's worse (like with 1.7) I really have to contact the german dealer, to check return policy. If its end of this week, or in 4-5 weeks, doesn't matter anymore.
|
|
|
Post by bitzerjdb on Jan 28, 2020 19:51:12 GMT -5
Firmware Update 1.7.6
Good Stuff
My RMC-1 is no longer locking up after sitting in Standby (Wife acceptance factor has increased) No Drop Outs while watching two movies, one streamed via a Roku Ultra/Netflix and one Blu-Ray.
Unchanged
Still dropping the first few seconds of a digital feed. Most noticeable when listening to Music (not a huge deal for me at the moment because I have a second Audio server with an Integrated DAC. The Analog Input on the RMC-1 works great)
Changing inputs still does the double flash
Untested
Still have my Tivo Bolt directly connected to the TV. It has been rock solid directly connected. I need to see some proof that the HDMI handshake issues have been fixed (or addressed) before I switch it back.
Overall - Nice step forward, hopefully, the other fixes will be added shortly.
|
|
|
Post by mgbpuff on Jan 28, 2020 21:08:23 GMT -5
All problems will be solved shortly, have no fear. How do I know? Big Dan sold me his Marantz AV8805.
|
|
|
Post by deewan on Jan 28, 2020 22:04:15 GMT -5
They need to clarify ASAP. If that is the case the RMC-1 is going back. I'll go Storm Audio or the new JBL maybe. Excuse my ignorance, but what’s the need to do ANYTHING above 48khz??? Outside human hearing and.. How many speakers produce any sound above 48khz? I have seen very few actually. I'm not sure you understand the audio terms being used. I believe you are confusing the sample rate of 48khz and the audible frequency of 48,000 Hz. Examples of high sample rates include 48 KHz (used for DVD video), 88.2 KHz (2x the rate of CD audio), and 96 KHz (used for DVD-Audio and other high definition audio formats). In audio production, a sample rate (or "sampling rate") defines how many times per second a sound is sampled. Technically speaking, it is the frequency of samples used in a digital recording. The standard sample rate used for audio CDs is 44.1 kilohertz (44,100 hertz). That means each second of a song on a CD contains 44,100 individual samples. The discussion revolves around if the RMC would be able to apply filters to a sample rate of >48khz, not to the frequency above 48,000 Hz which is above human hearing limits.
|
|
|
Post by mick on Jan 28, 2020 23:04:48 GMT -5
On that note... just FYI...
(From the current manual for the Monoprice HTP-1....)
Engaging ANY surround sound modes results in the sample rate being limited to 48k. The Dirac Room Correction filter limits the sample rates to 48 kHz.
If you dont plan to use dirac it will always be full res. I'm nit sure what res dirac is but I would clarify with the other companies that DIRAC runs at 96khz. Keith, Why not just tell us what the sample rate is? why dance around the forum and avoid the question?
|
|