you have done the obvious blind test of the original 96k source and 20 kHz low pass filtered version - still at 96k to determine if there is an audible difference introduced there?
looked at the fft of the digital anti-alias filtered output before decimation to see if any leakage frequency content that could fold down to bass exists?
looked at the fft of the digital anti-alias filtered output before decimation to see if any leakage frequency content that could fold down to bass exists?
Last edited:
There may be something more than that going on, Scott.
When I was running thru a Berhinger DEQ2496 and DCX2496 I switched my media server output to 24 bit. Did it sound better? Yes, a little. But not at all what I expected, it sounded better in the bass.
I still haven't figured that out, but it was consistent.
I highly doubt it's the format, but something going on in the electronics or software along the way. I had at first forced it to 96/24 but found that 44.1/24 had the same effect.
When I was running thru a Berhinger DEQ2496 and DCX2496 I switched my media server output to 24 bit. Did it sound better? Yes, a little. But not at all what I expected, it sounded better in the bass.
I still haven't figured that out, but it was consistent.I highly doubt it's the format, but something going on in the electronics or software along the way. I had at first forced it to 96/24 but found that 44.1/24 had the same effect.
There may be something more than that going on, Scott.
When I was running thru a Berhinger DEQ2496 and DCX2496 I switched my media server output to 24 bit. Did it sound better? Yes, a little. But not at all what I expected, it sounded better in the bass.
I highly doubt it's the format, but something going on in the electronics or software along the way. I had at first forced it to 96/24 but found that 44.1/24 had the same effect.
Pretty easy to diffmaker the results?
There may be something more than that going on, Scott.
When I was running thru a Berhinger DEQ2496 and DCX2496 I switched my media server output to 24 bit. Did it sound better? Yes, a little. But not at all what I expected, it sounded better in the bass.
I highly doubt it's the format, but something going on in the electronics or software along the way. I had at first forced it to 96/24 but found that 44.1/24 had the same effect.
How did you increase the bit depth? Starting with 16 bits and zero filling is exactly what the spdif interface does all by itself. (Just hook a scope to the data stream synced to word clock and see for yourself.) Is your software adding the evil dither to make it sound better and give more bass?
If you switch to 24 bit content is it even the same recordings?
The analogy is not very good, but any mid and high frequency mess affects perception of bass tones. The origin of the mess is not necessarily the common non-linear distortion.
Last edited:
I think that a detailed look at the schematics of both the record and reproduce electronics for the digital system would be in order to evaluate the number of electrolytic or ceramic (gasp) coupling caps in series. A talk yesterday with a known modifier of electronics, as well as a digital recording engineer, gave me some insight to even the problems with the Korg 2000, which is a better product than most here use. It would appear to this engineer that it is the ANALOG part of the electronics that is the primary weakness. Digital seems to be working as expected. Perhaps that is a good part of the bass problem.
My software player allows me to choice the output bit depth. Any volume or EQ it does is 32 or 64 bit float, then you choose the output format. Since at the time I was using digital volume control (about 10dB) I figured that a 24 bit output might be a better way to go than 16 bit - as the following electronics are all 24 bit anyway. Why cram it down into 64K levels when you have millions in 24 bit?
So the difference in SQ could have come anywhere along the path, the player software, the DEQ, the DCX. I don't think that 96/24 has inherently better bass, just that some software or hardware (or both) were happier along the way. That's the point. Nothing is perfect, I just happened to find a setting that worked better for me.
While I would agree that the Nakamichi was a pretty good cassette recorder, it was not the very best that could be done. I know, because I designed much more sophisticated electronics, in 1978 for HK. Unfortunately, my design was never put into production, but the essential circuitry that I developed was quickly applicable to designing all discrete 30 ips electronics for a Studer transport, within 6 mo, for Mobile Fidelity.
One factor was the bass coupling caps that I really improved on. However, even the Nak, used very traditional, simple electronics, with cap coupling absolutely necessary.
One factor was the bass coupling caps that I really improved on. However, even the Nak, used very traditional, simple electronics, with cap coupling absolutely necessary.
John, I doubt the analog section was used in this case.
http://www.diyaudio.com/forums/anal...ch-preamplifier-part-ii-1801.html#post2784802
As I read it, it was the down-sampling to 44.1K that caused the difference in the bass. That was likely not done in the analog domain. Joachim?
Nakamichi in its day was accused of moving the goalposts by re-interpreting the original (Philips?) specs for standard playback. Their contention (from memory - add granular NaCl) was that their gap losses were correct and everybody else was out of step. They were probably righter than not, but whadda I know?
Thanks,
Chris
Thanks,
Chris
Recorders as a class have the un-enviable position of being subject to a straight-wire bypass comparison. That's a tough test, depending on good quality source material, environment, yada-yada, but at least the standard of comparison is completely unambiguous: is it different or is it ain't?
I've finally gotten old enough that I can't tell the difference between an A/D/A copy of commercially available source material and a bypass, using a good modern A/D/A of course.
Probably wouldn't apply to everybody, but good enough for me is good enough for me. Leaves me more time to bitch about other stuff!
Thanks,
Chris
I've finally gotten old enough that I can't tell the difference between an A/D/A copy of commercially available source material and a bypass, using a good modern A/D/A of course.
Probably wouldn't apply to everybody, but good enough for me is good enough for me. Leaves me more time to bitch about other stuff!
Thanks,
Chris
No one else ever has, either, assuming levels are set correctly. So enjoy your golden years.
- Status
- Not open for further replies.