Posted by el`Ol on
05-25-2008
|
Romy the Cat wrote: |
tuga wrote: |
I am now more inclined to describe that sound as a portrait taken in broad daylight with some overpowering fill-in flash (I believe nowadays this technique is quite fashionable)... But the model wasn't wearing any makeup. | | Wonderful, wonderful, wonderful. I even experience an epistolary envy that I did not come up with it myself. However, I would very much argue the concept of “system specifically tailored for absolute faithfulness to the recording”. I feel that “faithfulness to recording” is very fealty concept. I would not develop this idea about it in this Stahl –Tek’s thread, but I think that it was worth to state my disagreement.
The caT |
|
Hello Tuga, hello Romy!
I was about to stuble across an other language deficit, when my Google toolbar told me "faithfulness" can be used both in the sense of honesty or credibility. Even if the recording chain is perfect, honesty of the playback chain means revealing the faults of the format, like filter artefacts, quantization noise, or aliasing rests due to the slow reaction of long FIR filters. Most of the time I tend towards getting the maximum information out of a recording, including some unwanted parts. Sometimes I remember my encounter with the Linn CD12 many years ago, that seemed to hide everything that could sound electronic or un-natural behind some fine mist. In my ears Burmester tends in this direction, but from my limited experience with highend gear I would say this philosophy has not become very popular since then. I am currently in the mood of thinking this is the real thing because it makes the given information more credible.
Regards, Oliver
|
|
|
Posted by tuga on
05-25-2008
|
Hello Oliver and Romy,
English is not my mother language so let me clarify that I meant faithfulness to the 0s and 1s and not the performance.
Like I said, that system was designed to expose the recording/register in all it's crudeness, like a window with no glass.
Best,
Tuga
|
|
|
Posted by Romy the Cat on
05-28-2008
|
Oliver, I do not think you understand "faithfulness" in the same way as I do. Ironically out differences of views are not linguistics but rather in the realms of objectives and intentions.
You are right and there are a number of companies who hide the formats faults: Linn CD12 or Forsell or BAT VK5D or most of the CEC models and many would be as illustrations. Others do opposite and they throw in your face very sharply everything that is possible to get out of disk: CEC TL0, Burmester, in a degree some Mark Levinson transports, Spectral and others.
However, the "faithfulness" to retrieve everything from the format or even from a recording (or as people love to say “whatever microphone heard”) is not the "faithfulness" that I care. A recording is just a snapshot of live events, the events that reflect the idea of musical program, the composing and performing intentions. The faithfulness and credibility that I look is the faithfulness and credibility to the cores of musical super meaning. Would it be hard-reading CD transport of soft-reading is not important if the result is not viewed in context of everything else and in context of interaction with the main musical idea. I think to assess "faithfulness" of digital with the faithfulness of musical idea is like trying to assess a performance of aircraft with consideration where and how it will be used for.
Generally I also tend to use hard-reading transports but I try as much as possible to get rid the idea of CD. A raw digital file is fine but being placed in CD/DVD or any other “disk” format is become garbage. I think the inters problem with digital is not in the digital itself bit in industry intention to control distribution that made then to invent the CD, DVD, CASD and any other formats of “disks”. If we have even 44.1kHz at 20 bit of properly implemented raw WAW files then it is already so ahead of anything imaginable of any “disk” format that it might be truly “perfect sound forever”. The Cat
|
|
|
Posted by el`Ol on
05-29-2008
|
Romy,
I have problems in believing that a noisy clock (the lower-priced CEC drives probably have in contrast to the other drives you mentioned) can do more than covering the faults of the actual implementation (mostly regular time deviations at other parts of the data stream). I have no experience in combining highend DACs with different drives, but my experience with lower priced gear is that that the difference between drives is clearly audible, but not that huge. Is this different with real highend, or is it more because you are trained to discover minimal effects? At this year´s Highend Burmester showed a Hollywood movie with their surround receiver. I didn´t see what video-DVD-player they were using, but it didn´t have the ultra-precise clock of the Burmester devices, for sure, not to mention that probably DTS data reduction was involved. In my ears it still sounded natural and un-electronic, just with an overall reduction of sound quality compared to CDs. I guess this is mainly an effect of the used upsampling algorithm. I can´t say much about the dCS upsampler, because I only heard it in combination with a Lamm/Wilson combination that in my ears showed some surrealistic saccharine sweetness I found quite annoying. You mentioned some time that the sound quality "goes down the toilet" when it is actived. What is it that you don´t like about it?
Regards, Oliver
|
|
|
Posted by coops on
05-29-2008
|
Oliver/Romy have either of you heard the HRx discs produced by 'reference recordings' yet? ( or the berkeley audio design Alpha dac ), I don't sell them Romy !
|
|
|
Posted by Romy the Cat on
05-29-2008
|
el`Ol wrote: | ….I can´t say much about the dCS upsampler, because I only heard it in combination with a Lamm/Wilson combination that in my ears showed some surrealistic saccharine sweetness I found quite annoying. You mentioned some time that the sound quality "goes down the toilet" when it is actived. What is it that you don´t like about it? |
|
Oliver,
It is hard to talk about those things with any degree of certainty. In past I made a number of attempts to evaluate if upsampling makes senses. However, I have no idea if the result I experienced was the clean outcome of the upsampling itself or it was due to any other issues, not to mention the specific implementation of upsampling. Generally it we look at DACs that do explicit upsampling then they do less interesting then DACs that output the same that impute. However there is a great dose of simplicity and even hypocrisy in it as most of the DACs (not all) do internal ?X oversampling and they anyhow introduce the assumed, or non-existing bits between the actual bits.
Probably the best would be to look at the DACs that allow changing the upsampling on fly. I play with a number of them and I always preferred the straight 44/16 in and 44/16 out (in that time the source was CD), though the DAC internally might run the 352.8kHz and I do not even knew about it.
What I do not like with sound quality in upsampling DACs? Well, many things. If audio sound changes during upsampling then obviously it is not good and it indicates the fault of upsampling (and most of them done in this way). What I do not like if the auditable sound was not changes? Here we arrive to the conversation hwy we need sound to begin with. As an applied illustration of audio sound juts for sake of sound itself the upsampling has no problems. However, I feel that upsampling might ruins a very fine and very delicate texture of sound, making our interest in sound less naturally curios and physically attractive. There is a certain sense of relaxed and not violent invite-ness in straight sound (analog has it even more) but upsampled sound has that “inviteness” more rough and less sophisticated. The straight sound I found is always more soft not in sonic terms but rather in humain terms… Perhaps it is some kind of self-hypnoses on my part as intellectually I always ask myself why we ever need upsampling as it anyhow introduces non existing events in recorded sound.
Sure, they are juts generalization and I think it all might be just upon the implementation. For instance the 2X upsampling on dCS upsampler do screw up aggressively even the auditable sound. The 2X upsampling on Pacific Microsonics is absolutely non-auditable in realms of auditable sound and to found the faults you need to drill deep into placebo consciousness.
I would like this post not to be my declaration of being pro or against of upsampling. Nevertheless, I think that for us, the people who try to do audio in the shorted and the most natural why possible it is necessary to prove to us, at least to me, that upsampling is truly necessary. So far I recognize the explicit upsampling as a tribute to something that contrived and my listening experience agrees with it. In my book any DSP activation ruins sound. Who decides what would be the value of the none-existing sample? A mechanical algorithm? The caT
|
|