Jump to content


  • Content Count

  • Joined

  • Last visited

Everything posted by PeterSt

  1. FYI: A bit depending on the software version (hence the year we look at), encrypted files (feeds) can't even be stored. They need decoding first and then can be stored (and that goes by A/D). Versions that could store encrypted files have existed, but then they had to be decoded within 24 hours, or you'd lose them. The decoding was a separate process (a "run"). All sh*tty stuff. Why is all on YouTube free ? (at least I think it is) ... because it is all as lossy as hell. I know, this is not A/D stuff, but still the answer to your quest.
  2. Yea, that's what you would logically say. But it isn't so. Again this depends on what actually happens and additionally whether encryption is in order or not (and do note that the encryption is dedicated to our places over here (Canal Digital)). I am fine with the thought that all is recorded digitally and no A/D - D/A is in order, but then they compress more on the D-D job. One way or the other, it is about the number of stages under way. And YouTube is special anyway (and not for the better).
  3. This is what XXHighEnd would show you, when it is so. It is my understanding that the current "standard" (a convention) depicts that there's always an upsampling step in order when the file is under 88.2 sampling rate. I (XXHighEnd) don't comply to that ... (and you (XXHE) would upsample yourself anyway, per your own means and up to the rate your DAC can handle).
  4. PS: Encryption is also a deteriorating process, but it depends a bit on where it happens (we have the decoding in the set-top boxes, which is a pain in general).
  5. Maybe it is about the number of A/D - D/A stages the feed goes through (which includes your home appliance). I have already mucho difficulty with recording the program say 20 minutes ahead in order to skip advertisements during playback. This should imply one additional A/D and D/A stage (the one in my SAT receive which can also record). Picture is poorer, but sound is so much more worse that I have difficulty to follow conversations. Having said that, to me it is totally clear that the means which has been used at the other end, is crucial to the playback quality. Over here we by now have several "missed the program" services, and they all bear another quality. Btw, because this is so, I don't pay attention to YouTube recordings. And oh, many real-time programmes these days, "time shift" the recording already, in case something bad happens (like Trump suddenly being shot which is not allowed to end up in the living room). These too already bear a poorer quality. All 'n all, no surprise to me. It's how it is and it won't get better.
  6. All right Paul, fair enough. Then : Could you please take distance from your own remark that some random ADC requires its time to settle and (level) trigger the recording, so that it is perfectly allowed to chop off the irregular beginnings in order to next have comparable files. I think you can. Next up are the recordings already present. So what can your software do with them ? The whole point is: if that brings nothing, it most certainly does not testify of Mani cheating or anything of the sort. It would tell, however, that the test means are still too weak. FYI, and just in case you didn't know it yet: I can show you and everyone the most easily discernible differences between whatever dozens of means, all leading to bit identical looped back data, those means including USB cables. It goes from, way too wide sound stages, to way to flat ones, compared to narrow and deep. Don't ask me why you and so many others have difficulties with getting that. All I know myself by guarantee, is that somehow we all (!) are not able to find test means showing it. So ... let's have it puuulease ? I am sure Mani will hand you the files, if you don't already have them. All not meant as teasing or anything, but I feel we are in a deadly loop.
  7. Mani, we must assume that what we can incur for minute differences (and let's say that they emerge by accident, hence out of our control really), which is ever so small but audible, and that because we could not control it explicitly, we can not measure it. It is almost a physics law; Why wouldn't we be able to perceive noise-like modulations, especially if they oscillate (re-occur). And mind you, we already know that they oscillate because that is inherent to the processes (at least the ones I apply in XXHighEnd - all of them). Maybe take the analogy of a piano out of tune (or any string instrument for that matter). Would you be able to hear the two strings out of tune (but the ever so slightest !) when individually played ? maybe a very best musician can. But I wonder. Suppose this level of distortion is sow low (the deviation to the absolute A key etc.) that no analyser would be able to measure it. This could be so already because of present system noise. But for sure you'd hear the out of tune when two strings are played (one so slightly out of tune) ! Maybe the audible oscillation would be once per minute only, but you'd hear it still. I think this compares to IMD ... We talked about it before : with a sufficient number of test tones we should be able to dig out some general distortion figure. I think Paul was on the same track (well over a year ago, I think). Mind you, in the end (sufficient number of test tones in parallel), it will be equal to music - with the difference we can measure it.
  8. Is there an off-topic button somewhere ? Edit : of course there is. But it seems unethical to use it. But still ...
  9. Can someone please repeat what's on then test bench here ? apart from a nice discussion, I personally lost the subject. Mani, you can't just let yourself drive into a corner and remain to be in the position that you are able to hear differences between "anything", which is what it seems to be about by now. I mean, up to listening to 1KHz sines, as it seems. 😮 You never clearly stated what this listening test will be about, according to you and your capabilities. You seem to allow to let others determine what you will be able to discern. Well, that most certainly isn't going to work ... I am not trying to tell you what you can't and what you can, ... what it is about currently just lacks. And while you're at that anyway, please accompany it with feasible control means. Like you, I too can easily state what I can (and what I can't) discern, but that doesn't mean that I would be able to technically prove it. Scientifically, yes (statistics), but technically not. I'd even give you or anyone the guarantee. Let me remind people that quite in the beginning of the Red and Blue Pill I, I indicated that no technical test means would exist to prove it. People, mansr too, ignored that too easily. This, while it is (and appeared to be) the most crucial. Thus, someone scoring 99% probability is one, but trying to prove by technical means how that was done, is two - and impossible. One simply can-not prove the rightness of it by means of using too weak test means. So better have consensus about that first. And that can't happen by means that I know of, so we should ditch that part. What thus, IMHO, remains, is a fairly simple test depicted by Mani (he knows what his capabilities are and with what means), which definitely needs real time checking for bit perfectness. And that is easy enough. Still a note of warning: whatever the means, it will influence SQ and Mani can't just let it happen that any random means is injected in his chain which will obfuscate. So whatever the means for testing bit perfectness, *or* #1 Mani must have applied that himself for approval by those directing the later test, *or* #2 those who direct the test must apply it, for Mani to check the usability for him. Probably for quite some days of easy listening (not under stress of the day itself). Next he should be allowed to reject the means because it does not work for him. If I may give an advice on this one: try to accept #1 because else the lot will be a too tedious procedure (#2 may even imply several iterations). All of the above implies a fairly simple technical setup, because no analogue recording means will be involved. Why not ? because no means to prove differences in those recordings will exist anyway. Yes, I am repeating myself. If you guys want to prove those differences, then I am happy to provide whatever means of bit perfect playback which sound different to each and everyone of you, in this room over here. No need to try a 10 out of 10 there. Only the technical means to show it. So ... does that exist already ? Paul perhaps ? It would be an unrelated test which can be done an other time. I'd be happy to host that, but others must provide the means. I hope I have simplified the game.
  10. Maybe a small update from my side: Each other day I am again and even more amazed what's dug out of my so very well known albums. If I was allowed to use one word only, I'd say clarity. As if all is suddenly completely without distortion. I am still thinking of the technical merit in order. Oh, I should add that I am now also using an ET^2 towards the EtherREGEN. This definitely brought another change, seemingly for the better because it lessened the effect of what I described earlier on, hence the extra ET^2 made it all more "normal". The technical aspects of that I again can't see (with the EtherREGEN re-generating the data and all). Let's keep in mind that I am not streaming over Ethernet (I use USB) and that the only thing at play is the connection from the audio playing PC to an other (say for RDC reasons). So the EtherREGEN must have some isolating kind of role which should not even be about the transmission of data (data transmission of functional packets is idle during playback, also for (RDC) screen updates).
  11. But what do you do ? smoke weed over them ? I never ever have such issues (and in the end all of my 2000+ CDs were successfully ripped). So how can you have issues with so many of them ? something has to be wrong. One thing : I always took care of buying the most decent optical drives (mostly Plextor). So maybe if that goes with less care, the problems arise more easily (never mind you tried 6 of them) ? Maybe the power supply to yours ain't right ?
  12. Probably not. Bespoke but I'd agree with that.
  13. Those who protest the most against the proposed ToS changes, are those who pose the problems most. Why, actually, are they left to be bespoke. If votes would be out to list these names, the subjects would be quite shocked to be on top. It would be half a dozen indeed, probably plus 1. Comfortably numb.
  14. Yes, as does the MQA thread for all intents and purposes Personally I see these kind of vendetta-like posts (both of them) as the worst and most hurtful to keeping threads on-track. These kind of often sneaky and snarky at the same time posts, could imply an immediate ban of, say, one day. This self-indulged posting (which is what it is too, IMO), is killing for the 99% of people who beg to differ with the skills involved. If not 100%, because audio is what it is, and one's own ideas and systems are always the best (and imply the highest degree of subjectivity, but so be it).
  15. The problem is the so-called justification. "The material is not that expensive, it is all the development". Well, generally true, but not to these extents IMHO. Otoh, when I laid out the other day what all constitutes the cost price of a Lush Cable (all bought by the spool) it initially comes across as quite unbelievable. But still that is true. I suppose it depends on what one brings across (or tries to) and what seems plausible and what just for sheer profit. Often way too much in our views. Still (these days and and of quite late(ly)) I wouldn't have any problems with thinking on how analogue cables could be improved. Still, however, I would work with those parameters, though quite over the top. And not with materials as such. Still materials have function and may imply higher cost. Example is the shielding of one size we just ran out of. The bill of the spool shows me 600+ euros (ex VAT). This is about material like tinned copper) but also about the coverage (like 110%). This (I think) can be justified. Gold or anything does not justify anything (again IMO), especially not if one says "it sounds warm". Oh well ...
  16. Galen Gareis makes it easy on himself because he talks about analogue application only. The R, L and C thing do not even come to the same result in my personal (hardly experienced) book, because L and C contribute to shielding aspects and are therefore "absolutes" already. VP would even be my explicit thing because this is all about current spread (no matter how minute for audio and now "not-measurable"). Ricardo, what was your idea with this thread ? too few cable threads ? haha
  17. The USB as we used it (Async) is not error corrected, and in all cases no errors assumed (which can be observed by means of driver GUI programs), USB should just transmit the data which does not imply additional or changing payload at the transmitter (PC). Notice that the "no errors assumed" is definitely the situation for 999 out of 1000 people. So forget about changing SQ because of error. With Ethernet/LAN and SATA (in my previous post I hadn't even read yours - about SATA !), this is very different; Both anticipate error explicitly, and both could even deal with a 50% error rate (of whatever, if it's only not 100%), without you knowing except for slower throughput. One thing: when direct streaming is in order (be it from disk/ssd or be it over the LAN (Ethernet-connected DAC), the errors would need to be solved before the audio samples are clocked out to the amplifier etc. If not, you'd have a glitch or whatever happens, and this latter is not likely at all. All is more than fast enough to a. deal with dozens of errors per second and b. to be in time for that next audio sample to put out. Compare 100Mbit (bandwidth of EtheREGEN) with 44KByte x 2 (stereo) = ~ 880000Kbit = ~ 0,88Mbit (just doing this quickly, so correct me where I'm wrong). But a 100x over-capacity and thus room for 99% error. Because Ethernet (but SATA too) is so much anticipating large levels of error, the errors are allowed to happen just the same. It just is not bad. A manufacturer may use a $1 chip instead of a $10 chip, because the $1 implies an error rate of 5% while the $10 implies error of 0.5%. Depending on the projected environment, both are good enough. Still both may differ for SQ per the way I reason now. And for Ethernet this is enormously more than USB, which latter should incur for a steady data stream (with only relatively small variation because of packet content and length varying). And so I may also have found the reason why my changed connection makes such a vast difference in the first place. And mind you, this could already be the EtherREGEN which does NOT incur (or incurs for less) errors and re-transmissions, which also thus hammers on the transceiver at the other end of its output. Even its input could be important (but not under control, unless by means of ET^2 cable !) because the re-transmits on (towards) the input side (of the outputting PC in my case) ... because lag on the input means lag on the output and thus peaks at the input of the Audio PC. And (current !) peaks are the devil for audio ... I didn't anticipate this thread to be intriguing. But it could be more than any other. For myself at least.
  18. Then a for myself important one - in the end important for all of us. I use the EtherREGEN in an I think more special chain: As the connection between Music (file) Server and the Audio PC, with mostly Ethernet only used for RDC (Music Server PC controls the Audio PC). I say "mostly", because no streaming occurs during playback (loading the albums prior to playback, yes). Well ... Now think of the sheer fact that the EtherREGEN is a bottleneck for the speed of loading the albums. Thus, without switch the net bandwidth is larger than with (the EtherREGEN has a bandwidth of 100Mb/s, according to the specs) and with my Music Server PC also having 100Mb (audio PC has 1G), the EtherREGEN still is a bottleneck, already because of latency and the way I (XXHighEnd) load files (this is quite special for its setup, with as main characteristic that communication is about at the packet level and not at the level of the largest possible burst means). This implies relatively huge overhead, and the EtherREGEN (any hub-like device) adds to that again - I'd say with a factor of 2, compared to "no hub"). But now the sheer gag, assumed I am right on my observations in the first place: A FLAC album loads and decodes in about 1 second, over mentioned Ethernet connection; The same album loads/decodes in 2-3 seconds with the EtherREGEN in between (which I find no problem at all, it's worth it). This is with ET^2 config A:WYR, B:WYR. Then yesterday at first attempts to tame the maybe too dynamical sound, I did this : A:WR, B:WYR. ... And now the time to load/decode an album has been doubled or so (still worth it) ... Again, if my observations are correct. Don't ask me how many re-transmits Ethernet normally experiences, but please do know that Ethernet is 100% made for it, as collisions normally happen all over the place (say two people sending data at the exact same time), and that Ethernet (Internet too) is one big pile of "error", but no worries, when more error, it only means that more re-transmissions need to happen; only if things really are in error, no re-transmission will help, and a whatever presented error message will be your share. Unlike Async USB, any error is detected and corrected (re-transmit), BUT this means extra payload to the transceivers involved (our PC's and connected devices, in my case two PC's with now the EtherREGEN in between). But point of case: the audio PC now too has to deal with extra payload; It requires additional processing (possibly at the hardware level only (a controller) but that doesn't mattter. You know, that processing which is supposed to be audible (no discussions about that pooollllease !). And so what we may be dealing with is a configurable error generator, the ET^2. And for background, any normal Ethernet cable will not be different for errors as such - it is only that we don't experience the errors. They can easily be there, but we don't see them; yeah, by means of unexpected slow connections (if interpretable in the first place). And so I suddenly think I found a quite explicit means why Ethernet cables can matter for SQ (SATA the same, but not with me because I have no SATA means in the Audio PC). I suppose that the best sounding connection will be the one with the least errors. Not sure how to see those by simple means. Alex @Superdad (😜), maybe you know what you guys did to check for proper (acceptable) working ? I just lack the experience (and like to put the question to you instead of giving it to Google). Peter
  19. Mark, please consider the fact that connecting the shield(s) to ground of devices (that's what you would be doing with the black wire and the extension cable for it) on one end, is a just as legit solution as connecting through to ground on both sides (or the one-other side for that matter). So the logic that the shields must connect from end to end (once connected to one end) does not exist. And thus it is a means to influence SQ with 4 additional options (connect none, A, B, A and B). Sadly it will depend rather completely to the "switch" (hub) device you use (with the EtherREGEN mostly as suggested one for this thread). What's in my mind for devastating vs good options, is the consideration whether the ground as such (think chassis ground for where the ET^2's shields to connect to), is tied to signal ground (the 4 white or color-white-dashed Ethernet wires). And that a random switch would itself tie chassis ground to signal ground, well, is in my mind (and I wouldn't like that - pure from my own theory).
  20. I was fairly sure it was wrong. But suddenly the real nick wouldn't come to me. I was trying and trying and trying, and fund this one. Still to be sure that it was wrong ... I never thought of that, and when reading the user manual it when in one ear, out the other, because at that stage of reading I planned the switcher just because it's advised. And later I did not think of that. However ... Yesterday, at connecting the LPS-1.2 I considered it by myself, and still did not do it, because it would be too much of connecting-through ground, while it is the first things I'd avoid for Ethernet. Otoh, I still have one ET^2 only and I'd like to experiment with the side of the ground connection. I mean, the input side would definitely be OK with me, but that is exactly the side where the ET^2 not is, so I can't even connect ground there. And mind you, the ET^2 has been setup so that it doesn't need to be conform the device in question, and you's just find some ground point, like that could be the chassis. But think Ethernet connections with inherent ground in the (metal) plug, and the then available option to try it or leave it out. Anyway, very very good that you pointed out the floating ground - I just didn't even think of that. More in a next post.
  21. All right. I can't believe what I am hearing ... ... That is what I typed at the 2nd day of listebing to the EtherREGEN with the ET^2 cable from its output to Audio playing PC. And did I announce that I was going to be honest ? I never typed more than the italics above, because I was too astound to even know what I should express about this "combo". I intended to blame the combo indeed, because I never tried either on its own (and don't intend to). The only thing I should do, is also add an ET^2 for the input (Music (file) Server to the EtherREGEN). (building up tension a bit) 5 days or so later ... The sound is so different from what it was, that my brain is actually continuously working on the how-come, instead of listening to the music or value of the SQ now perceived. And mind you please, I regard my system to be of the most "unheard" resolution / resolvement, so I don't even know what others can / will perceive from this, but actually this is too crazy to be true ... And to add to this: I never even wanted to add an ET^2 to my own system at whatever place in whatever chain, because it wasn't going to make a difference anyway. Or maybe I did not want to know ... Even the most recognizable songs have been transferred to something else. Say that even Riders on The Storm is now of an unknown remix. It has been changed so much and from so many angles, that I don't know how to start interpreting. Actually it is all for the better to infinity, weren't it that ... That it is all too "present". All tries to be on the foreground and in the end this becomes tiring. Especially with Rock or any real wilder type of music, it gets crowded. But in such a strange way ... Last night I was focusing on where the "crowded" actually springs from, and I "saw" something like too much dynamic range between nothing and sound (and that at the highest resolution). This would resemble the so much more resolution to infinity. E.g. cymbals sound and extend in a completely transformed way, again, I can't even start thinking where this originates. A digital connection ... (which even is galvanically isolated to begin with, because Ethernet). I am not sure yet what the next step should be, but I tend to use a linear power supply for the EtherREGEN as a first next trial, despite the advice against it (by UpTone). This relates to their remark "we could not detect any real change" (with their own LPS-1.2 I presume). Well, I wonder. So Alex @Uptown Audio, I hope this doesn't turn out into xxx orders for LPS-1.2 modules, because I know you won't have time to pack those all, but don't blame me for being the messenger, might it happen. 😑 This proposed next step is not even the most logical one because the ET^2 still has its 255 configuration possibilities, or a bit less (half, I'd say) if I leave out the ground connection at at least one side (currently that is disconnected at both sides, because: A: WYR, B: WYR (no B at either side) However, because I comes to me that this high perceived resolution could be faked by high on/of (switching) underlying layers (I seem to hear that), I want to eliminate that first. Or at least I want to know the influence which undoubtedly *is* there (bet ?) but the how of it (what's perceived from it) I like to know. And prior to further cable changes. Besides the idea of the switching, the other idea is that my system is now elsewhere lacking. I just hear too many good things (and not little) to ignore that possibility. So, meanwhile, others have opinions by now ? Or, what would be the most useful - how ET^2 config changes influence the sound in absolute sense ? Please try to avoid the term "harsh", but merely try to relate it to the music content. Or more technically, how frequencies are (where) emphasized. Or how that works out (more/less forward (of what ??)). For me harsh or distortion is not in order at all. But I could use some "rounding". The sound now is more towards the Clairixa (USB cable). So much delineated that it becomes digital again (the opposite of what the Lush^2 brought). The fact that I also use the Lush^2 and the Blaxius^2 for analog interlink makes it, well, complicated. I like to leave those be (I know, this could be the wrong approach).
  • Create New...