TonyB wrote:

If the PC is a master sending the data (and generating the clock), I would not use it. That is why I would go for one of the PRO (expensive, e.g. Lynx AES 16) cards which can accept master clock when sending data. Can your Bidat generate or accept master clock?

My Larvy ADC can send out clock reference data via a different interconnect. I might get one of these Lynx AES 16 and synchronize the Larvy internal clock with the sound card clock. How important it is and if it would be generally necessary I do not know. As far as publishing data from PC into Bidat if might be different that might not need any synchronizing. Bidat runs it own cClok algorithm that recognizes the clack mark right inside the main digital stream. After Bidat get it is tune own clock to the clock of the source. Once agene, how it all actually affects Sound I have no slightest idea.

TonyB wrote:

but it re-samples the data received from PC in its internal DSPs.

I begin to panic what I hear the DSP word… :-)

TonyB wrote:

If you're sure you're not going to do the DA or AD conversion in the card, then the AES16 is all you need. However, if you might want to try just using the AD conversion on the cards (which might be just fine, hard to say without trying), then you'll want the L22 or Two model which have converters on board. They won't get in the way of digital use, and will let you experiment with using the card itself as ADC. Though I'd be surprised if it were as good as an external box, I would say it's worth a try. The AES16 basically just gives you more channels of digital for your dollar - but you only need two channels.

Still, the AES16 is even more expansive then L22 but the question still reminds if I would need a “good” card of I do juts digital in and out. Hantra uses SPDIF output right on motherboard and he believes that it is not bad. So, if I use any simple bypass supported soundcard, for instance the one with SPDIF that I have now, and terminate the cable form Larvy DAC with SPDIF connector instead of the AES then will it be the same. The reason why I am interested about it is not that would like to save myself $700 (that would not hurt) but because I would like to found out some logic or reasoning behind all of it.

Rgs,
Romy the Cat

jtavan wrote:

I don't know of that many motherboards with on-board SPDIF _input_, most just have output (and not terribly good output at that). The AES16 and the L22 are within $50 of each other, with the AES16 usually being cheaper. The logic behind buying the professional card is that you get AES/EBU balanced digital rather than SPDIF, and you get the ability to use an external wordclock. They do seem pretty expensive for what they are, though. You don't need 16 channels, but I don't know of a good and cheaper digital-only card. All DACs without wordclock input use some sort of clock-recovery circuitry, as AES/EBU and SPDIF are self-clocked protocols and the DAC chip wants to receive a clocked I2S signal. It's easier to do this well if you have clock sync between the sending and receiving devices.

The things are clearing up now, thanks everyone for your help. If after I build this FM grabbing machine I get Sound that I will be pleased then I promise to send some “interesting” recording of my FM broadcasts. Or perhaps I will put then on my file server and let you guys to grab the uncompressed files….

Rgs,
The caT

Guys, a few more questions that I can’t understand: (or understand it wrong)

The Lynx AES16 card has only AES/EBU inputs and output and no analog in or out. So presumably, the made a good pro card that would be only AES/EBU infract for PC and despite that it has an excessive amount of channels for my propose still it looks like what I need. However, there is catch in this card.

The specification for this card said that all channels are transformer coupled…. probably it is OK, I do not know.  Also, the data routing in this card take place by digital integrated digital console that they clams is “hardware based”… Now the interesting thing… “The AES16 model offers eight channels of mastering quality sample rate conversion”. Does the sample rate conversion presumes the signal processing up or down, or actually the DSP? Is it possible to bypass the sample rate conversion on a given input bay?  

Hantra, a question to you and with your L22 card… Dos your card allows to bypass, or to hard-patch the inputs and to avoid to perform any manipulations? I have seen the screen-shots on the Lynx L22 and then in tier parching interface has volume control for each channel. Those thighs really scare me because as soon a digital volume control introduced to the signal then the signal and this card defeat any reasonable purpose. So, could L22 route the inputs without exposing the stream to any bytes screwing functionalities...

Rgs,
Romy the caT

OK, it now looks like my dive in the world of PC is over and I pretty much know where I am staying. A well of so ago I was completely ignorant about the subject but now I could at least to understand what the common buzz-worlds means and able to ask more or less related questions. I still have no practical experience with the field and do not necessary know what the relation between those buzz-worlds and actual Sound now, based on collective “somebody’s else experience” I kind of understand what I am trying to do. Since I was bothering people with all of it I would like to compile my thought that I end up with.

First all of I would like to express gratitude to all whoa assist me and advised me: the contributors of this site, the numerous people who send me direct emails, the recording engineers and technicians, the application engineers designers at some manufacturing houses and some dealers who believe me or not turned out be quite knowledgeable.

The very interesting aspect of my PC’s Audio exploration was a reassuring from very many people (with different credentials) that the recording sound to PC might yield very-very good result if it done properly. I still do not know how their definition of “very-very good result” might correlate with my own definition of “good result” but you and myself will learn about it in a few weeks. Also, I would like to point out that the my objectives of “success” of this entire experiment is not just get of PC “very-very good Sound” but to be able to preserve this “magic” of FM broadcasts that I fell might be more completed…. Anyhow, I have finish the “collect data” stage regarding this subject; the entire PC system is done and being assembled and anticipate it will take a few weeks (or more) to learn how to get out of it Sound that it might be capable of. Then it will be assessed and reported.

Here are some summarizing observations…

The PC:  Low processor speed, SCSI hard drives at 10,000 RPM… still looking for an old CD burner that will be able to writhe 1X speed. I have some people who telling me that it IS possible to burn a superb quality CDs and they share some their experiences how it should be done. I am very skeptical that I will be able to do it BUT I have heard their custom CD and I do confirm that their results were way ahead of any other custom CDs that I heard. Some radical people suggest building own non-switching PS for the recording/storing PC but I do not go for it at this point. The PC has WaveStudio, EAC and the newest Nero Recording Programs. The Sound Card… All my expirations ended up with practically unanimous recommendations of Lynx products.  The chose went for Lynx AES16 card as a better AES/EBU interface. As it turn out that Lynx 22 does not have as sophisticated output stage as AES16 and runs a previous, jitterier generation of Lynx’s outs. Also, the AES16 has transformers coupled outputs the make the entire card with more menacing ground to the rest of my systems. (Still I glad that I did not go optical and I initial though I would need). Also, the AES16 has a lot of advantages over the AES16-SRC as it has no module that allows perform the sample rate conversion.  In the end there are people who would build custom XLR cable sets (replacement for the stock CBL-AES1604) extending it as long as I wish. The AES16 can perfectly fine to run a quite long cubes. Use the SPIDIF input with this card is perfectly kosher thing to do but many people explicitly rejected ideas to use anything then AES/EBU at out stage of this card. Thanks God the BiDat runs AES/EBU and … I hope it will be up to the task…. The DAC as I told was Lavry’s Gold AD122 MKII. People pretty much shared between Prism, Pacific and Lavry but I have personal experiences with Lavrys and his concept that using his Gold line ADC and DAC should produce no other resent then a simple straight wire do sound very impressive, although quite bogus. Also, and it is very important for me… it would be possible to get rid of that miserable gold Lavry’s front plate… ;-)

In the end it would be VERY VERY cool if the PC recoding turn out to be sufficient enough to record the FM broadcast. In addition, it tunes out to be way less expensive then I thought it would be. I was thinking that the project might cost around $15K but it turn out to be way less expansive. Anyhow, I do not what to sound that the project is over, as I have no reference points about the Sound I will get out of this. However, I have no further frustrations of questions at this point and all that I need is to make it to work. I will report the results….

Rgs,
Romy the caT

PS: Dan Lavry: Truth in Tone Coloring

I brought today my Lavry’s AD122 MKII processor and decided to conduct excrement feeding the ADC with “live” feed from FM broadcast and outputting it directly into my Bidat DAC: WHRB 95.3 FM  8:00 pm San Francisco Symphony orchestra “IN CONCERT” with Michael Tilson Thomas conducting:

Adams: My Father Knew Charles Ives (world premiere)
Stravinsky: Violin Concerto in D; Hilary Hahn, violin
Tchaikovsky: Suite No. 3 in G, Op. 55

I think something is very wrong in here, as the result that I get is too bad to be considered juts unsuccessful. The digital feed is very noise, compressed (probably because of noise) and with huge lost on tonal nuances (probably because of noise). I did use the crappiest balanced cables that I have and the cheapest Monter made digital cable but even with all of this the result is way more horrible then I would expect form the “peripherals”. Something is wrong, as I do not believe that such a world-class level DA converter might perform at such a primitive level. I need to edict he myself what I am doing wrong. It turned out that it will not be juts plug and play….

The caT

Bud wrote:

In that vein..... the only CD's that I have found to be worthy of using, to copy audio to, are made, or resold, by TDK as the Music CD-R (not CD-RW). The important point here is the moderate green cast to the plastic. This reduces the backscatter and probably jitter also. Side by side comparison to the clear or yellow will convince you very quickly.

jtavan wrote:

This might be an overly-obvious suggestion, but have you checked (with an audio editor program) that you're using most of the available recording range? That is, are peaks on the signal up near the maximum level possible given your recording depth? If not, you need to adjust the input sensitivity of the ADC to match the output level of your tuner. This gain matching is critical to getting a decent recording - you're just throwing away bits if your peaks aren't NEAR digimax.

Rgs,
Romy

I have found the problem and was able to resolve the noise quandary.

The Lavry ADC has balance analog input and the Sansui has single-ended output. I use my PAD’s Dominus cable with an RCA to XLR adapter. The adaptor was made by BAT and I used it for another project to have in it the 12dB voltage devider. I opened it up, re-soldered the contacts and reassembled the adapter. I did not mean to use it eventually but for sake of tests I thought to use an adapter should be OK. Here is the catch: the BAT adapter is not grounded by default as it has a Teflon separation between the RCA source and XLR load. Since I do not do anything balanced for years I kind of never cared that an inch of the space inside of the adapter was not shielded and that there was no direct contact between the grounds of the components and the shields of the XLR adapter. In the case when the adapter not juts route the pins but also convert grounds from XLR to RCA the proper grounding become an issues but I stupidly did not use the third pin and continues treat it as a single ended-pair where the twist is enough most of the time …. Also, the Larvy ADC is stunningly sensitive at it’s inputs and require very heavily shielded cables, according to the Larvy application engineers. I end up with buying another cheap temporary adapters form pro shop, explicitly ground them and drop the POD cable as they are essentially not shielded. When the nose is gone… I think I would need to find a pair of XLR-RCA quarto-shielded patches for future…

So, what was then… As before, I connected the tuner into the Lavry ADC and the ADC directly into the Bidat…

In short, it was quite remarkable and frankly speaking I DID NOT ANTICIPATE that it might be so good.  The preamp was deed by FM-direct and FM-ADC-DAC run. I’m so || close to declare that the result is absolutely identical, no mater how unrealistic it might sound. Ironically the sound is virtually identical although I use the crapy cables and without final tuning of ADC (it runs now 16/44). Interesting that the result is virtually identical while I DO NOT USE the entire dynamic range of the Lavry ADC.  In order to have the identical volumes from tuner and DAC I can use only the ADC input adjustments and of course no DAC ether ADC has any volume advisement at digital domain. So, I have very good (identical!!!) result without even using a full ADC’s range. (At the loudest picks the ADC show –12dB) I mean… running it a full dynamic range, take care about the cables   and fine more suitable “modus operandi” for the ADC should give a even better (!)  result?  Hey, and I do appreciate what is going on even right now!

Frankly speaking it kind of irrational. The direct analog feed vs the feed coming via DAC-and ADC and practically no worsening of sound! I remember in my youth I records a lot of reel-to-reel and I never was able to make the original tape hiss and the hiss coming form a copy to have the absolutely identical harmonics, it always was very slightly off. Here, the FM nose if absolutely the same form the FM feed and via the digital feed. I mean I flip the preamp freed with a remote control and there is NO EVENT IN SOUND, no event at all!!!!

Anyhow, I’m so pleased that I am VERY strongly consider not to try the Lavry DAC now…

One more question that I would sadly be asking… IF THAT RESULT IS AVAILABLE then why we have our digital so much more horrible then our analog? Anyhow, I need to open an audiophile freak show in my listening room… :-)

In a series note, I would like to look at it more and see at the result deeper. So far, at the level of “the sounds” everything looks very-very attractive

Rgs,
The caT

jtavan wrote:

I'm really pleased to hear that your results were so good. This is what I expected should be possible, but I admit to being surprised that it worked so well just at 44.1/16. Wish I could afford the Lavry ADC - that sounds like a winner.

I am also very much surprised with the result. Regarding the Lavry ADC… I was paying MUCH less then I was expecting and if you wish I might probably to do one more for you. Contact me privately is you are interested.

jtavan wrote:

Just for kicks, maybe you should try plugging your vinyl rig into it and record some nice records at 96kHz/24bit - see how close the playback is to the big setup.

Yes, this would be defiantly the next step, but I will go there when I finalize the cables and few other things… I would be VERY interesting to try.

jtavan wrote:

I've often considered recording my library of favorite vinyl to digital, so I can play them back at will for guests, without vinyl wear. Lots of work, though.

If it would be possible with IDENTICAL quality then I would be very appealing, although I did not target to do LP archiving…. The Lavry ADC stays and soon or later I would defiantly try it…

The
Purring caT

Another question for you the digital savvy people: what software I would need to mange my FM recording. Here are the requirements that I have - the software should:

1) Grabs the digital stream from my sound card and load it to hard drive in windows uncompressed format (any other format would be better?)
2) Alloy me to name, sort, group the files, add the custom annotations, search, move across the drives and permit to crate my own hierarchy of the files clustering.
3) To edit the windows uncompressed format, I would need just cutting the tails and heads, nothing else.
4) Play the windows uncompressed format, permitting no proceeding; no DSP, no mixing, no sampling conversion and no anything like this. It is very critical that the software before dispatching the bytes to the sound card would not do any pre-proceeding
5) Should have a graphic interface that would present the current playing position within the entire file and also allow advancing the playing within the file.
6) Create the new windows uncompressed file form the fractions of another files.

Can anyone suggest anything?

The caT

This weekend I decide to take my FM where it should be. I ran ~125” and installed a dipole on the roof of my building and serviced my tuner. The result was as expected but there was something unexpected. Along with all normal benefits of better antenna the tuner begun to produced HF slightly better and yielded slightly more dynamic (I have no idea why!!!). Now running the tuner direct vs the AD-DA produce very-very-very slightly different result. I can’t characterize it as a compression but rather very-very minuet flattening of sound. The effect indescribably minute and it is absolutely undetectable blindly. If I 10 times switch between the analog and the digitalized feeds than probably 3-4 times I might that “something happens”, although to identify those change would be very difficult only by the “flattening” or compression. Still the difference, if it does exist now with the boundaries of this affect I might attribute to the crappy cables and the lover then necessary recording level of the ADC input. Another problem is that my Bidat perhaps dose not handle the quality of the digital stream? Or perhaps what I am beginning to feeling is the way in which digital should sound on my playback? I do not know the answers but I will continue to collect the observations. The “cable up” the system and trying to got for a full recoding dynamic range should be next step. Then I put the PC, lynx and the file into the game…

The caT