Question about rip/burn Cd's

[audio amateur]

You would use WMA or other reduced bit rate formats only if you're making compressed files, such as for a portable player. Otherwise, you would want to simply copy the CD in Nero...

I don't agree. Many do and would copy their cd collection to their hard drive in WMA, especially if they can't hear any difference when comparing to lossless or WAV.

[noddin0ff]

I don't agree. Many do and would copy their cd collection to their hard drive in WMA, especially if they can't hear any difference when comparing to lossless or WAV.

...and there's both WMA lossless and WMA lossy.

[noddin0ff]

What point is there in debating what is appropriate to say in this thread when the first reply said this...

If I post something confusing, misleading and irrelevant I hope someone will call me on it.

I'm in no way trying to be a thread nazi or an arbiter of appropriateness. People can say what ever ridiculous things they want on any thread. It's a free forum. And, we're all free to call things as we see them.

E-stat has many insightful things to say throughout many forums. But, here he made a long post about a whole bunch of wonky stuff and then closed with the conclusion

Lossless can be great, but the playback device must be well implemented.

You can't draw that conclusion from what he posted. Is there a problem with me pointing that out civilly, as I believe I did?

[Feanor]

...
I highly doubt whether processor speed has any bearing on transcoding fidelity; it isn't a processor intensive task. The SB is way more than adequate. Similarly, I don't understand what you mean by 'lossless is great, but...'. I mean...lossless is lossless; it is what it is. You don't have to 'well implement' it to get more lossless. Any difference you hear is downstream and subject to variations there.
...

Understand that there are a lot of audiophools out there who insist that they can hear a difference between WAV and compressed lossless such as FLAC. Well there are a lot of variables with compter sound so maybe they do if they aren't "well implement", but you have to be "well implemented" whether WAV, MP3, or whatever.

Noddin, you are perfectly correct that decoding FLAC is a totally trivial task for any CPU made in the last decade. In general sound differences, (assuming they are real rather than imagined), are rarely if ever the result of the decoding process per se.

[E-Stat]

Do you use it with the main system or with garage only? I bet any differences would be more noticeable in the main system.

Presently, I use it only in the vintage system. Indeed, differences would be more noticeable with the Sound Lab based system. I thought that I was agreeing with you in that given a suitable player, I believe that a lossless format is capable of sounding every bit as good as an uncompressed format.

Would you disagree fundamentally if we included 320kbps lossy?

With a good enough system and suitable content, I find that lossy truly sounds lossy.

rw

[E-Stat]

I highly doubt whether processor speed has any bearing on transcoding fidelity; it isn't a processor intensive task.

Try a Touch and get back with us.

You don't have to 'well implement' it to get more lossless. Any difference you hear is downstream and subject to variations there.

In the real world, one must use downstream components that can affect the result.

Isn't the basic thrust of this extended thread, all things being equal, to ask what you sacrifice with various codecs? Why confound the issue by bringing in the perceived effects of wall warts versus $200 power supplies? etc.

They are examples of the variables responsible for the differences you hear downstream. I replaced a CD transport with a network transport that provided different audible results when:

1. A different power supply was used
* and *
2. Transcoding was performed at the server level

Conclude what you may. I'm not convinced that inherently using a FLAC source vs. a WAV source was responsible for either change.

rw

[Dawnrazor]

Noddin, you are perfectly correct that decoding FLAC is a totally trivial task for any CPU made in the last decade. In general sound differences, (assuming they are real rather than imagined), are rarely if ever the result of the decoding process per se.

Hey Feanor!

I kind of agree with this statement for most people. Though for those of us who have gone kind of crazy with the minimalist approach of the cics memory player concept that flac processing is not trivial even though the hardware could normally handle things very easily.

I for one cant really play flac. It causes a hang that .wav doesnt when loading from ram. I accidentally imported some flac and the volume was much higher than normal from a previous song, and the flac was blaring and the mouse hung for a long while and I couldnt do anything. This never happens on .wav

Here is an explanation from the author of the software and overall computer transport concept:

-There's no difference in SQ as FLAC is decoded entirely before playback. The issue that FLAC causes is potential dropouts when cPlay does background RAM loading. Although this takes just a few seconds, FLAC decoding is insanely CPU intensive and will cause a dropout. This gets worse when output rate is at 192k as CPU headroom reduces significantly (at 96k, CPU0 load is ~40% vs ~78%).-

Anyhow I dont really see what the big deal is. Hard drive space is cheap these days.

It is probably just me, but I have had pretty bad luck with flac on a wide variety of computers from normal pcs to audio ones.

Here is a vid I put together of one of those issues where the same file played as a flac stutters but the .wav doesnt:

http://www.basaudio.net/diyhosting/asylum/MVI_1602.avi

[Feanor]

Hey Feanor!

I kind of agree with this statement for most people. Though for those of us who have gone kind of crazy with the minimalist approach of the cics memory player concept that flac processing is not trivial even though the hardware could normally handle things very easily.

I for one cant really play flac. It causes a hang that .wav doesnt when loading from ram. I accidentally imported some flac and the volume was much higher than normal from a previous song, and the flac was blaring and the mouse hung for a long while and I couldnt do anything. This never happens on .wav
...

Anyhow I dont really see what the big deal is [with WAV]. Hard drive space is cheap these days.

It is probably just me, but I have had pretty bad luck with flac on a wide variety of computers from normal pcs to audio ones.
....

Dawnrazor, good to hear from you.

I'm glad you concede that getting into CICS is "going crazy".

I'm also glad you mention that decode FLAC is exactly the same as WAV so that any sound diffference that there are, are the result of decoding -- part of "implementation" as E-Stat refers to it.

Personally I've never actually had a "stutter", "hiccup", or dropout that I could attribute to FLAC or other compressed format vs. WAV -- to be sure, I have had these probems. However in my case these anomalies have been the result of (a) bad sound drivers, e.g. the ASIO driver for my M-Audio card, or (b) computer startup programs or Windows services that interrupt processing. These are other aspects of what E-Stat calls "implementation", and by experimentation I have eliminated these problems except for the following ...

Presently about once every ~3 hours of listening, on average I have a network dropout that causes Foobar2000 to kick up a "file not found" message. These dropouts seem to have begun after I got my newest network router. They have nothing to do with the sound quality before or after the interruption. I'll enlarge Foobar's input buffer to see if that makes any difference -- it might not. In any case at the current frequency of these dropouts they aren't really a problem .

If you can do without metadata tags, and if you have the pockets for twice the hard disk capacity, sure, go for WAV. Personally I couldn't live without the tags and, now that I'm retired, the extra cost of HD space is still relevant.

[E-Stat]

Although this takes just a few seconds, FLAC decoding is insanely CPU intensive and will cause a dropout.

I didn't think so many folks were hallucinating. I wanted to pass along the advice that I found to be true when using a non-computer based solution.

Hard drive space is cheap these days.

The chief advantage to FLAC is tagging support, not space preservation. For a couple of years, I had most of my library ripped to WAV and played it back on my computer alone. I confess that the additional convenience of tagging was worth all the hours spent converting to FLAC.

rw

[E-Stat]

Personally I've never actually had a "stutter", "hiccup", or dropout that I could attribute to FLAC or other compressed format

Nor have I when played back on a computer - which is where the processing in my arrangement now occurs. The compact Touch player, however, uses a comparatively low powered imbedded processor that is comparable to the Pentium II used in my 1998 desktop. You end up pushing it pretty hard when it also runs the Linux based server software playing from a local USB or SD card.

rw

[noddin0ff]

Conclude what you may. I'm not convinced that inherently using a FLAC source vs. a WAV source was responsible for either change.

Which is what I said. You're not hearing differences due to codec. Thus, we reached the same conclusion and we agree.

Try a Touch and get back with us.

No need. I can accept that your Touch might be a lemon. Your comment about the added overhead of Linux and local USB is duly noted. I'm no computer expert, no argument. I've fed digital signals to a 10 year old Mac G3 processor (which I guarentee is slower than the Touch), a 7 year old Yamaha receiver, 2 different DACC (MD-10, and Zhaolu), and a current mac mini. I've done this wirelessly and optically and a combination, and now via wireless to an AppleTV and HDMI. Hardwired they all work fine, wireless they used to have dropouts but the networks are better now. I could turn this back to you and suggest that you should now run out and try every system I've implemented and get back to us, but that would be childish wouldn't it. Sonically, yeah there are differences, however, as we agreed above, its due to downstream components.

In the real world, one must use downstream components that can affect the result.

It may surprise you that as a flesh and blood member of the species and the forum, that I do live in the real world. I certainly don't live in the esoteric world and I never argued counter to this statement. Thus, we agree.

I took umbrage to a post that listed peripheral/downstream influences and then suggested the sonic differences were due to implementation of a lossless codec. I saw no connection there. Your posts often comment on how you perceive extremely subtle differences in your system; I felt this case was worth challenging as I believe all lossless codecs are perfectly capable of being transcoded (implemented) to create the same audio signal. I pointed out that you seemed to be arguing you could distinguish between the fidelity of lossless formats when really you were saying that the position of racks and the present of wall warts was affecting your sound. I thought this was confusing and misleading and still do. But, as you see above, we agree on the reasons for the perceived differences.

They are examples of the variables responsible for the differences you hear downstream. I replaced a CD transport with a network transport that provided different audible results when:

1. A different power supply was used
* and *
2. Transcoding was performed at the server level

As to this, its all far to esoteric for me to weigh in on. There's no way, in the world I inhabit, that I would ever convince myself to believe that I could hear any sonic difference induced by a power supply. Well, if it was arcing lightning maybe I could. Perhaps I've been blessed by high quality electricity; I wouldn't know. I'll take your word on it that you do hear a difference.

It does strike me as weird, however, that a company with a solid reputation for providing a robust DACC/server thingy would sell one that couldn't do the one thing it was built for. I'd send it back.

[E-Stat]

Which is what I said.

Hmmm. What you said was:

"I highly doubt whether processor speed has any bearing on transcoding fidelity; it isn't a processor intensive task...Thus, we reached the same conclusion and we agree. "

We (and quite a few others) do not agree on that assertion. I submit that Logitech disagrees as well. Otherwise, why would they provide the configuration option? If you choose to run it stand alone using local storage, it must necessarily transcode locally. On the other hand, if you use a separate server, you can exploit the benefits of offloading the processing task.

I can accept that your Touch might be a lemon.

On the contrary. I think the Touch is a great value for $280 as Wi-Fi receiver, network player, DAC and local server. Better performance from the unit, however, is available for those who wish.

I took umbrage to a post that listed peripheral/downstream influences and then suggested the sonic differences were due to implementation of a lossless codec. I saw no connection there.

It has everything to do with the implementation, i.e. what makes it work - as opposed to the potential capability of the FLAC format.

There's no way, in the world I inhabit, that I would ever convince myself to believe that I could hear any sonic difference induced by a power supply.

That's certainly fine. Others can and you might as well if you knew what to listen for. There is a very good reason why there are zero high end players that use wall wart switching power supplies: added HF noise as seen graphically here. The GamuT CDP I use incorporates separate linear power supplies for the transport and DAC.

It does strike me as weird, however, that a company with a solid reputation for providing a robust DACC/server thingy would sell one that couldn't do the one thing it was built for. I'd send it back.

It certainly works out of the box. I think it is great that it provides a cost effective platform suitable for audible improvements via configuration (transcode at server level) and the use of a replacement power supply (for which there are at least four different manufacturers of aftermarket solutions). Just plug in the new PS and enjoy a smoother, quieter top.

rw

[recoveryone]

I hate to break up the current ongoing debate, but I have a quick observation since I'm about 80% done with ripping my own CD collection. I have notice that Jazz artist or I should say the mixing engineers take more pride in how the finish sound comes out. I have played back CD's from the 70's and 80's from various artist and the quality of the sound is right their with current artist CD's today. But in other musical styles I can tell right off hand how dated the sound quality of the recording is. Which leads me into my next question, when you guys talk about drop off, is this due to your network or the ripped format, cause I have never experience such occurrence or maybe I am not listening close enough.

[noddin0ff]

Well, since the noddin/E-stat sub-thread has now stepped into a confusing muddle of misinterpretation, mis-recontruction and loss of context where we can't seem to agree on the fact that we're not disagreeing...

I'll just ask a question about the link E-Stat provided above regarding the graph of HF noise. In the interest of fair disclosure--although I admit I don't think I have the capacity to hear wall wart induced noice in my system--at heart I'm a skeptic. Even though those with claims have better ears and better systems, I will doubt until convinced otherwise. Doubt is not to be confused with disbelieve. So no personal slight is intended.

I read the text and looked at the chart. As far as I can tell the sound measurements are taken from the noise the wall wart is making directly to the ambient environment. ie not any noise the wall wart may be introducing into the downstream audio chain that comes out your speaker. I'd like this clarified for me.

If all the past posts about the glories of high end power supplies, how wonderful they are and everyone should by them were a simple referendum on the ambient noise they don't make, then I have completely misunderstood. My take was that the argument on wall warts was that they degrade downstream audio quality within the electronic chain to the speakers--an argument I'm skeptical of.

Also, the differences are presented in dB, which I believe is a signal to noise ratio and not a measure of absolute noise. I presume you can have a huge signal to noise ratio between two values that both fall in a completely inaudible range. As is, the chart doesn't convince me that there is any perceptible difference in wall warts. I don't know what a zero Volt ref is, but I assume it's a very very very quiet reference (the quieter the reference the greater the more exaggerated the dB value). Clearly the chart isn't using the standard SPL scale and that alone would be misleading to the casual reader.

If this is all about cheap power supplies buzzing, specifically at 50 or 60hz. That can certainly annoy a person, even me. My low tech solution is to plug into a more distant outlet and run an extension cord. I can understand the lack of appeal though.

[noddin0ff]

I hate to break up the current ongoing debate, but I have a quick observation since I'm about 80% done with ripping my own CD collection. I have notice that Jazz artist or I should say the mixing engineers take more pride in how the finish sound comes out. I have played back CD's from the 70's and 80's from various artist and the quality of the sound is right their with current artist CD's today. But in other musical styles I can tell right off hand how dated the sound quality of the recording is. Which leads me into my next question, when you guys talk about drop off, is this due to your network or the ripped format, cause I have never experience such occurrence or maybe I am not listening close enough.

My take is that we're talking about audio drop outs (silent gaps) and pops (digital artifacts that make sharp noisy pops). When I first ripped my collection (I've re-ripped it all) I used iTunes. iTunes is not the best ripper and in a number of my rips, particularly from older CDs, there were digital mistakes in the rip. These produce pops at the same point in the song every time it's played. Anecdotally, I believe different DACCs can do a better job of filtering out the pops. Drop outs are due to halts in the smooth transmission and/or buffering of the data. The computer experts can provide specifics and, please, correct me if I'm wrong. You can probably also get pops if the data stream gets interrupted. You can probably also argue that the actual bits in the data can be altered by interference or phasing issues resulting in pops and artifacts. Reflections of data at the ends of cables are oft cited. I tend to think that this latter type of error must be exceptionally rare and that digital signals are very robust these days. The pops I hear have always been in the rip.

[E-Stat]

As far as I can tell the sound measurements are taken from the noise the wall wart is making directly to the ambient environment.

Nope. This is a noise measurement performed on the gear just as you commonly find on any piece of audio electronics. Such gets amplified and as you can see, is frequency dependent. In this case, you can see and hear added false brightness which masks detail.

If this is all about cheap power supplies buzzing, specifically at 50 or 60hz.

It isn't. Look again where the greatest noise peaks exist - two octaves between 2 and 8 kiloherz. It is about clarity throughout the bandwidth and more dynamic punch.

rw

[E-Stat]

...when you guys talk about drop off, is this due to your network or the ripped format, cause I have never experience such occurrence or maybe I am not listening close enough.

With my new experience with inexpensive network players, I've found two separate sources: Wi-Fi transmission and local transcoding. The former manifests itself quite obviously - the music stops for a moment while the device rebuffers the signal. Such can be addressed in a number of ways. The latter results not so much in a dropout per se, but the occasional presence of digital artifacts like clicks or "swishy" sounds manifested at the top end of the frequency range. Choose the SB option to allow the computer server to do the heavy lifting. Turn off the simulated power meter or RTA screen saver.

rw

[recoveryone]

I must be very lucky on both the WIFI transmission and the quality of my Rips (320kps) Mp3 format. I have 3 SB's I use in the house with the one downstairs getting a 89% connect signal strength and the 2 upstairs getting between 75-80% signal strength. I have the music server in the garage with a WIFI card in it, so nothing in my setup is hard wired. It may be the repeater (Linksys Range Expander) I have in the garage to boost the signal to my sons room in the front part of the house. The only time I get any hic ups is when listening to internet radio from time to time (very rare) and then its usually on the sender (weak signal) and not my system. But thanks for clearing that language up, I was wondering if the drop outs were a result of the player or the platform (rip format) used.

[E-Stat]

I must be very lucky on both the WIFI transmission and the quality of my Rips (320kps) Mp3 format. I have 3 SB's I use in the house with the one downstairs getting a 89% connect signal strength and the 2 upstairs getting between 75-80% signal strength. I have the music server in the garage with a WIFI card in it, so nothing in my setup is hard wired.

Indeed, one must begin with a solid ripping solution. I've always used EAC and had good luck.

As for Wi-Fi reception, it depends upon location and bandwidth requirements. I had always gotten good strength in the garage until I moved the steel rack to what turned out to be a direct path between the access point in the office and the garage. It was a simple fix simply to move that out of the way. As for bandwidth, your needs are more modest than mine. FLAC or Apple Lossless typically requires about double that of 320k MP3. Allowing the server to offload the transcoding increases the bandwidth further to 1,440k. I think the sonic benefits are worth the trouble since I am directly comparing the performance of using a CD transport vs. that of a transmitted FLAC source.

rw

[Dawnrazor]

Dawnrazor, good to hear from you.

I'm glad you concede that getting into CICS is "going crazy".


If you can do without metadata tags, and if you have the pockets for twice the hard disk capacity, sure, go for WAV. Personally I couldn't live without the tags and, now that I'm retired, the extra cost of HD space is still relevant.

Hey Bill, it IS crazy in the sense that many people sadly just dont seem to care about quality sound. Maybe it is the knee of the curve thing but I for one care about that extra 10-20%. For one who really cares about the sound cics stuff is not crazy...it works.

And to you and E-stat I just dont get the need for tagging.

It must be a classical thing. Personally my cue sheet/wav method gives me all I need: artist, album, song.

a cmp2 box also adds genre, which is kind of meaningless for me.

(Fwiw it is a long long story but most of my collection doesnt even have song names, though my latest rips do contain that info. I still enjoy the music whether it is "track 4" or "False Faces"....)

Others have mentioned to me in the past that one can do searches through the metadata. But I for one dont want my audio rig to behave like a computer per se requiring a keyboard.

Anyhow I am not saying there is anything WRONG with tags, just that I dont see the benefit especially if there MAY be an issue with sonics.

On the issue of the Squeeze box and the psu, one of the most knowledgeable people is John Swenson, who is an engineer of some renowned has repeatedly said that the psu in the squeeze box needs replacing. The Audio Asylum is down right now (why I am here) so I cant find the posts but they are there if you want to look. He also talks about power cords and has measured many and price point doesnt always seem to correlate with performance.

On a related note there is general agreement within the memory player project that the psu has a big effect on the sound and there are several advanced mods that take the cmp2 recipe to the next level (adding a separate psu for the drives and usb, and cpu is one of the hallmarks of the recipe, though it uses easy to do switchers):

http://www.cicsmemoryplayer.com/inde....ApdxBAdvanced

[E-Stat]

And to you and E-stat I just dont get the need for tagging.

Nor did I for many years - until I started using it.

Others have mentioned to me in the past that one can do searches through the metadata. But I for one dont want my audio rig to behave like a computer per se requiring a keyboard.

What I use contains a touch screen that works like an iPhone. You scroll through choices with your finger. Sometimes, I find it nice to quickly locate a song title via searching for its name.

Anyhow I am not saying there is anything WRONG with tags, just that I dont see the benefit especially if there MAY be an issue with sonics.

I'm not yet convinced what it would take to fully duplicate the performance of the CDP I use in the main system. On the vintage system however, I find that my tweaked Touch works every bit as well as various CD transports and changers I've used in conjunction with the Manley DAC.

On the issue of the Squeeze box and the psu, one of the most knowledgeable people is John Swenson, who is an engineer of some renowned has repeatedly said that the psu in the squeeze box needs replacing.

It was reading his posts and comments of others that led me in the direction of looking for a high quality linear for mine.

rw