|
Post by igorzep on Aug 10, 2021 9:14:26 GMT -5
Yes, Emotiva responded extensively and got right onto it with astonishing speed. You are reading a 14 month old article then spreading mis-information! Those issues were completely resolved by Emotiva within just a few days of being raised. Can I find a confirmation of your statement? After 8 more months... The issue of muting around digital silence (which I feel closely related to the one in the article) is still on my XMC-1. And the last firmware for it dates back few moths more before the article was written... Nothing in release notes of "gen-2" firmwares either!
|
|
|
Post by igorzep on Oct 12, 2017 17:42:56 GMT -5
Seems like yet another case of how the 'strings of digital zero samples' could be exhibited through XMC-1...
|
|
|
Post by igorzep on Aug 19, 2017 20:49:05 GMT -5
A house curve is exactly what you'd want. Yes, you "can" apply a house curve (as long as you have the full version of Dirac). However, the problem is that it's applied during filter creation and then downloaded directly from Dirac's server to the Dirac preset of your XMC-1. As it is, you can only store one and a bit (okay, a lot) inconvenient to repeat just to build a preferred curve for music. Just use tone controls to get the desired amount of bass.
|
|
|
Post by igorzep on Feb 4, 2017 14:10:58 GMT -5
Audio today is an IT.
|
|
|
Post by igorzep on Jan 8, 2017 19:20:29 GMT -5
So why spend the extra dollars; better to spend it someplace else. Or the best one of them all - nowhere?
|
|
|
Post by igorzep on Dec 28, 2016 17:11:03 GMT -5
The good news is that at least we now have the option of fixing things with firmware updates. Unfortunately there is only an illusion of it as the history of XMC-1 shows. And those things are not just interoperability issues with some h/w.
|
|
|
Post by igorzep on Dec 28, 2016 17:00:51 GMT -5
Considering the XMC-1 have ASRC in it any incoming content, including the same (seemingly, but actually a little bit different) sample rate will be resampled by ASRC anyway. But if you do 192kHz (or probably even better 188.2 kHz) quality upsampling externally then the job of quality resampling it to 48kHz for the DSP will be easier task for the ASRC chip, I think. But it might depend on on the ASRC chip. Anyway some odd, not directly related to the input frequency instead of 48 kHz would be better for internal processing, but it's done how it's done.
|
|
|
Post by igorzep on Nov 29, 2016 17:03:42 GMT -5
The biggest problem with Dirac on XMC-1 is that it doesn't show you by how much the levels are different on different speakers. If one of the speakers is significantly different in level than others then it determines the limits. If XMC-1 starts limiting volume then usually it means the gain-structure with Subwoofer is the problem. Try to change gain on the subwoofer so it is more in line on sensitivity with other amps/speakers and recalibrate.
|
|
|
Post by igorzep on Oct 20, 2016 2:07:04 GMT -5
^ No way it will be fixed - the flaw is in the hardware. The default setting when you power the board is 'ON', so when you cold start the unit it is ON until the software have chance to access the triggers hardware, then it have to set it to 'OFF' until it boots up completely to be able to set the DACs and other electronics to predictable state. They should have been designed triggers 'OFF' by default, then smooth operation would be possible. But unless the control board (or whatever board they use for triggers) is changed we are stuck with this XMC-1 startup behaviour.
|
|
|
Post by igorzep on Sept 19, 2016 12:25:00 GMT -5
Once you cut out p.1: $0 - not worth considering.
|
|
|
Post by igorzep on Sept 3, 2016 7:41:45 GMT -5
Or is the difference only in the frequency domain where the group just targets the same curve? Group let you just target the same curve for several speakers (subs). They are still equalized separately one by one.
|
|
|
Post by igorzep on May 9, 2016 17:30:03 GMT -5
Sometimes yes, sometimes no. Depends on usage patterns... bugs? anything like the UMC? (terrible) Strings of zero samples inside the content (as between tracks on a CD) cause clicking and muting of of surrounding content. I cannot call this "great sound quality". Actually totally unacceptable for a unit of such price.
|
|
|
Post by igorzep on Mar 29, 2016 7:05:47 GMT -5
|
|
|
Post by igorzep on Jan 28, 2016 8:05:17 GMT -5
I don't know why any experienced engineer would master digital audio without the use of proper dithering, but therein lies the problem. While I agree with your thinking in some way and agree that dithering of at least 2LMB (TPDF) must be used on any audio content, I know a lot of reasons for digital zero strings to be present in the audio stream. After all the non-dithered areas doesn't really contain any sound, so there is no reason to dither it unless this silence is part of a single continuous event that is recorded, but recorded silence is never absolute, so there is simply no possibility for that to happen. So, while I agree, that dithering whole-disc at the end of mastering it in high-resolution is a preferred way there could be a reasons not to do it, for example if the disc is composed from several pre-mastered pieces without wanting to touch them more, preserving the original quality (that is well dithered per-piece already). There are also synthesized sounds composed from pre-recorded samples as with menu navigation audio-feedback, games, etc. So, the thing is - this is valid contend and it practically exists, so, rephrasing you - I don't know why any experienced engineer would code the processing so that it distorts given such valid and trivial content. In the instances where I find a CD that is affected, what I have done is import it into Audacity or, now, Adobe Audition and export the WAV files with "dithering" applied. At this point, I can use those dithered WAVs to either re-burn an audio CD or re-export to FLAC for streaming. While this is kind of workarounds the problem with XMC-1 and actually doesn't fix the problem with the content in case if reason for those zeroes is lack of dithering in the mastering. It only adds additional bit of noise to the recording, meaning - makes it worse. While this 'worsening' might be negligible it is still there (and given 16 bit resolution - it is not quite negligible actually or else different noise shaping schemes would not be invented). Since I use JRiver on a dedicated music server (and immediately rip any new CD to FLAC anyway), I have found that the "easier" solution is to simply up-sample all 44.1k material to 88.2k (JRiver's DSP engine will apply "dithering") and voila, everything plays perfectly. Yes, it kind of works if your source is a PC, but started to work very recently, the price I've paid for this is to be hard-banned from their support forums (includes reading them). If I wouldn't raise the problem with the way they dither a year ago, no one would even believe it can be wrong, except the XMC-1 owners... that would get no help from JRiver's dithering that being 1LSB in amplitude would dither digital silence back to exactly the same digital silence). I'm sure that, for some, this won't extinguish the skepticism, but, for others, I hope this helpsβ¦.. Unfortunately it won't help with any hardware player that honestly pass content as it is with no processing applied (after all I bought the XMC-1 to do the processing and to be independent of a PC - if all would be possible on a PC I wouldn't need a processor then)... as well it can't help with content that I get over cable-TV.
|
|
|
Post by igorzep on Jan 25, 2016 7:13:59 GMT -5
As Keith stated, with only 1 or 2 people having noticed this at all, wouldn't Emotiva replace those units immediately as defective if for no other reason than to protect their reputation? I am sure they would replace them if they had something to replace to. Keith had tested a number of XMC-1s they have from different batches, and all of them had the issue muting around digital silence in the stream.
|
|
|
Post by igorzep on Jan 25, 2016 7:07:57 GMT -5
So that's useful info. Seems the problem is hardcoded into the DSP and that can't be changed via firmware upgrade? It is a suspect that it is hardcoded into the DSP. But even if it is it usually means the code is inside the chip vendor supplied library. The Emotiva have no ability to change this code themselves, but it is most probably part of the firmware, so, if new version of core library is used then it could still be possible to upgrade. I'm sure that many are accustomed to using an Apple TV to AirPlay content and would find the weird pop/click on the start screen obvious and troubling... Most users turn off audio feedback on such devices very soon after first use. On XBMC, for example, when it is turned on I simply hear nothing instead of audio feedback in 99% of the cases and only hear ticks together with muting artefacts if I scroll really a lot and very fast. So, you could live a lot of time don't even knowing there is some audio feedback from such device, depending on what is configured as feedback sound.
|
|
|
Post by igorzep on Jan 24, 2016 17:10:37 GMT -5
Directing questions to either Keith or tech support has been fruitless. If this is only occurring on some units I would be very pleased if they chose to replace mine with one that didn't pop or click. So far that option hasn't been offered that I am aware of. So far it is like that. But, after my lengthy conversation with Keith they are made aware of the issue exists among many common skips/clicking noises sources. And contrary to other ones it is not that random and purely content dependent. The problem I think many who had the problem was not really specific and even if there was a lot of complaints they were drowned in sea of other similar-symptoms complaints and were attributed, mostly, to a new stream initialization noises (wich is natural to do as a first assumption). Unfortunately most ordinary users cannot directly describe "I have the problem when stream of digital zeroes is passed to XMC-1". They tell 'in between tracks', and that makes confusion about re-establishing audio streams.
|
|
|
Post by igorzep on Jan 24, 2016 16:45:12 GMT -5
Thanks for the info, but I'm not sure if what Keith states is the complete problem. Here is his more detailed response about the digital silence problem: emotivalounge.proboards.com/post/761039/threadSo, it it true that they think the problem is too rare. And it is really great suggestion to call them. The more people made them aware this is an issue for them, the more chances they will re-evaluate the importance and make appropriate actions (either find a solution themselves or use their forces to push their chip/dsp code provider).
|
|
|
Post by igorzep on Jan 23, 2016 20:14:30 GMT -5
You just can't seem to understand that some people do not listen to CDs full of test files. The ridiculous (and rude) "test file" argument is still here. I don't listen to CDs full test files too... As you would like to suggest. But I use them to test/verify problems if they allow to do it quick and easy way. Remember - some of my "test files" are just regular CDs...
|
|
|
Post by igorzep on Jan 23, 2016 18:54:16 GMT -5
Mine does NOT pop, click or tick between tracks. The tracks that have strings of zeroes recorded in between them? Or the limited set of tracks that you are listening? I don't know why you intervene to any discussion when someone mentions the problem just to confuse people and pretend that your unit behaves somewhat differently, if it is known that it isn't...
|
|