Consider that audio interface today!

[JerryPH]

This does not exactly sound like arcane wizardry so I wonder why people invent expensive and complicated workarounds involving various amounts of quality loss.

The last time I looked the WAV file output was a 44.1 file on an 8X, so its right at the "bottom" of having an acceptable level of quality. If you are doing ANY kinds of post processing with it, it will start to fall apart fast.

Another reason that I personally don't use this is that don't exclusively make audio files with the FR-8X. I sometimes want to add the arranger or use a completely different acoustic accordion or create a multitrack song, or just want a higher quality of recorded sound... so lots of reasons to have many ways to get to the same location... because we don't all use the same cars... lol

There are dozens of options one can use to capture audio from a source.

My point is to do things that give people ideas, concepts and that using the mics from your cellphone are very much bottom of the barrel in terms of audio quality (though surprisingly they are also at the bottom of the barrel for video quality, but at least its not as low in comparison to the sound, and one can do some surprising things like even GREEN SCREENS which are VERY demanding of a video file, much more so than audio... but you can STILL see those differences easily too).

My entire point is... do it the way that you want, but having choices at least means you took the time to become knowledgeable about your chosen method and knowing the limitations of each helps you decide what you want to use.

 

[dak]

The last time I looked the WAV file output was a 44.1 file on an 8X, so its right at the "bottom" of having an acceptable level of quality. If you are doing ANY kinds of post processing with it, it will start to fall apart fast.

It is exactly the quality that the Roland synthesizes. If you want to upsample for the sake of post processing, it makes more sense to do that digitally than in analog.

There are dozens of options one can use to capture audio from a source.

Sure, but why not just use the cheapest and best? Going through the analog outputs is not magically going to improve the original signal. Even if you then capture with a $2k A/D converter, the $30 D/A converters in the FR8x will limit what you can achieve.

My entire point is... do it the way that you want, but having choices at least means you took the time to become knowledgeable about your chosen method and knowing the limitations of each helps you decide what you want to use.

Well, what I was puzzling about is why the obvious choice was just being passed by. After all, the Fr8x has a button named "REC".

 

[JerryPH]

I'm not sure there is no conversion. Is there direct output to USB before the D/A?

Do you mean like a RAW output of the data to a USB? I don't think so.

The digital audio is passed though the D/A converter, captured likely just before the analog outputs, to an A/D conversion to a WAVE file and saved at 44.1... I don't know of any other way it could be possibly done?

I say this because the complexity to create a WAV file from the left/right hands and all combinations of instruments would have to be tremendously complex. A/D-D/A converters are fast, cheap and plentiful, especially if they are limited to 44.1 CX standards.

But where the video that started this thread was concerned, it was to discuss the advantages between a cellphone audio and a fast, EASY way to get it in to your phone and no need to even sync it... doesn't get much easier and the results were audibly worth it.

 

[dak]

Do you mean like a RAW output of the data to a USB? I don't think so.

The digital audio is passed though the D/A converter, captured likely just before the analog outputs, to an A/D conversion to a WAVE file and saved at 44.1... I don't know of any other way it could be possibly done?

That would be completely nonsensical for a digital instrument. The synthesis is done digitally, and it does not make any sense at all to include anything as complex as an A/D converter in a Roland.

I say this because the complexity to create a WAV file from the left/right hands and all combinations of instruments would have to be tremendously complex.

That doesn't make a iota of sense since a WAV file is just a recording of PCM samples, exactly the form that digital synthesis produces before D/A conversion.

EDIT: The manual also states that the USB player can only playback MPG and WAV files with a sampling frequency of 44.1kHz. That makes it very likely that its whole processing happens at 44.1kHz and all its converters are running at that sample frequency. If it had separate conversion paths from and to the USB sticks, sample frequency would not be constrained as rigidly.

 

[JerryPH]

EDIT: The manual also states that the USB player can only playback MPG and WAV files with a sampling frequency of 44.1kHz. That makes it very likely that its whole processing happens at 44.1kHz and all its converters are running at that sample frequency. If it had separate conversion paths from and to the USB sticks, sample frequency would not be constrained as rigidly.

... I think you meant MP3, and I believe you are correct in thinking that the entire sample rates used/designed for the 8X are in the 44.1 range. Even when recording and playing back longer WAV files the 8X tends to choke and lag/stop/start on WAV files, so they recommend using MP3 files which are smaller.

Even at 44.1, ALL recordings will sound better than those captured by any cellphone mic on the market.

 

[kep]

This does not exactly sound like arcane wizardry so I wonder why people invent expensive and complicated workarounds involving various amounts of quality loss.

We are talking about the quality, right?
Try a simple trick: record silence to the USB and listen to the noise level. Compare it with the noise level from the line outputs. If DA is not involved in the USB recording, the noise level should be an order of magnitude lower.

 

[dak]

... I think you meant MP3,

Why would I be writing something I don't mean?

and I believe you are correct in thinking that the entire sample rates used/designed for the 8X are in the 44.1 range. Even when recording and playing back longer WAV files the 8X tends to choke and lag/stop/start on WAV files, so they recommend using MP3 files which are smaller.

No, they don't. Quite the opposite:

Please consider the possibility that I know what I am talking about, at least to the degree where you check the manual before telling me I am wrong.

 

[JerryPH]

Why would I be writing something I don't mean?

Please consider the possibility that I know what I am talking about, at least to the degree where you check the manual before telling me I am wrong.

You did say MPG, not MP3... one is a VIDEO FORMAT, one is a sound format.



Oh David, I am so sorry that I ever doubted you... well ok, not all that sorry... lol

I figured that after a year of being on my ignore list that you finally mellowed a little. Seems not, welcome again to that list. It's just better that you and I not exchange... seems that all you want to do is argue. I don't need that in my life.

 

[dak]

We are talking about the quality, right?
Try a simple trick: record silence to the USB and listen to the noise level. Compare it with the noise level from the line outputs. If DA is not involved in the USB recording, the noise level should be an order of magnitude lower.

Since Roland will not include world-class D/A converters in their instruments, that sounds like a solid plan. Even if the 44.1kHz recording from the Roland is not done with D/A-A/D conversion, it would be conceivable that Roland would use higher sample rates for their synthesis on the analog outputs than for their recordings. However, the inability of the USB player to play anything but 44.1kHz sample files makes it very likely that Roland runs everything through a single set of D/A converters and does all of its processing digitally and at 44.1kHz. And that would mean that there just is nothing better in quality you can hope to get other than that WAV file recording on its USB port.

 

[dak]

You did say MPG, not MP3... one is a VIDEO FORMAT, one is a sound format.



Oh David, I am so sorry that I ever doubted you... well ok, not all that sorry... lol

You are correct about that spelling error.

 

[kep]

it would be conceivable that Roland would use higher sample rates for their synthesis on the analog outputs than for their recordings.

I can't remember the source, but I'm positive I read the sampling rate for the FR series is 16bit/44.1kHz. Same with synthesis.

 

[dak]

I can't remember the source, but I'm positive I read the sampling rate for the FR series is 16bit/44.1kHz. Same with synthesis.

That would make the WAV recordings on USB stick right in the FR8x the best you can get, and when you need higher sample rates as output or for postprocessing, using digital upsampling (with good-quality methods) rather than going through the analog path should provide the best results.

 

[dak]

Oh David, I am so sorry that I ever doubted you... well ok, not all that sorry... lol

I figured that after a year of being on my ignore list that you finally mellowed a little. Seems not, welcome again to that list. It's just better that you and I not exchange... seems that all you want to do is argue. I don't need that in my life.

Sadly, the list will likely be better off without the dynamic that results from you considering me unfit for understanding audio electronics and signal processing. Having worked with electronics and audio hardware since child's age (and repairing and modifying audio hardware routinely) including a university degree in signal processing, that gets somewhat tiresome, and I am lacking the stature and character to respond gracefully. So the forum is likely better off without our interaction.

 

[kep]

That would make the WAV recordings on USB stick right in the FR8x the best you can get

Again, if it somehow goes directly into the wav file with no DA/AD.

 

[dak]

Again, if it somehow goes directly into the wav file with no DA/AD.

Yes, I meant once it can be corroborated that the WAV file on the stick indeed records silence without noise, meaning that no AD conversion is involved. I just mentioned that it would still be theoretically possible that the analog output (though noisier) could conceivably use a higher sample rate for the synthesis, but since you state that the available info already points to a 44.1kHz sample/synthesis base and the USB player only wants 44.1kHz WAV files, that would be very very unlikely.