I am just completing an Akido as the gain stage for a reworked Chinese EL34 SE amp.
Hopefully it will work well with my CD collection and Dac7 CD player (Kenwood DP7060).
However poor quality recordings are not done any favours by using the best amplification and speakers.
It might be better to insert a graphic equaliser into the sigal path or even a processor such as:
https://www.thomann.de/gb/bbe_sonic_maximizer_482i.htm
or
https://www.ebay.co.uk/itm/395154297758
I listen to a lot of those old close-miked jazz recordings, and they can be challenging, but I feel like I've made advances with my system when, instead of being pushed right up against the speakers, I can still detect the air between the mic and the instrument, no matter how short that distance may be.
I would not adhere to such advice. That inevitably leads to listening to a lot of bad music that's well recorded. Great recordings are not always great music. I'll take poorly recorded good music over well recorded bad music any day. I'll take a primitive 20's recording of Armstrong over a modern recording by Marsalis.
But it's far more likely that the flaws of the vintage 60s recordings were all analog-not during any post mastering and digitizing of the analog tape masters for CD release. So even though I will likely be spending seriously on the best neutral sounding DAC I can afford, I'd hardly expect it to fix flaws like tape saturation, waveform clipping (which I can probably fix with https://www.izotope.com/en/products/rx/features/de-clip.html ), and poor miking techniques-nor by however well it may also do upsampling.
I heard as much with headphones, even on some of my not so crappy recordings. I expect to hear that better with my new speakers, though hopefully without penalty.
My new speakers will be at least 94db/w/m and will never fall much below 8 ohms, but this amp will probably still t require at least 5v to sound good. https://www.firstwatt.com/pdf/prod_f4_man.pdf Otherwise, I suppose my DAC's remote volume control would suffice.
I feel that upgrading your existing components is not going to help with your drappy recordings, if anything it will show up the poor recordings even more.
This is why, for these recordings, a different approach could be needed, such as processing, with de-clipper, decompressor, equaliser.
directdriver,
My statement about it is better to purchase CDs that sound good might have been misinterpreted.
Think of a recording as a complete product: musicians performance, recording techniques, post production, etc.
I gave an example:
A great vocalist, an excellent orchestra, and good director.
Very musical.
Billy Holiday, check her out if you have never heard her (not only the 50s, but the 40s and 30s too).
That essentially agrees with your 20s recordings.
The greatest recording of Edvard Grieg's piano concerto, was Arthur Rubinsteins first recording of it, reproduced on 78 RPM media.
There are 2 recordings of Rudolf Serkin playing Beethoven's 3rd piano concerto.
His first recording of that work is the only one to have.
And forget about any other pianist versus that recording.
Enjoy the music.
My statement about it is better to purchase CDs that sound good might have been misinterpreted.
Think of a recording as a complete product: musicians performance, recording techniques, post production, etc.
I gave an example:
A great vocalist, an excellent orchestra, and good director.
Very musical.
Billy Holiday, check her out if you have never heard her (not only the 50s, but the 40s and 30s too).
That essentially agrees with your 20s recordings.
The greatest recording of Edvard Grieg's piano concerto, was Arthur Rubinsteins first recording of it, reproduced on 78 RPM media.
There are 2 recordings of Rudolf Serkin playing Beethoven's 3rd piano concerto.
His first recording of that work is the only one to have.
And forget about any other pianist versus that recording.
Enjoy the music.
Last edited:
Intersample overshoot issues can be fixed. You can either build or buy a DAC with headroom for intersample overshoots, or digitally attenuate the signal by a couple of decibels before anything else is done with it.
This is distortion due to tape saturation.
My solution to bad recordings is not to play them loudly. At least they won't sound loud and bad.
Ed
You could add an inverse distortion knob to the preamplifier. Tape saturation is a compressive third-order non-linearity, so add a knob that adds expansive third-order non-linearity. I haven't a clue how well this works given all the filters that could be in between the tape and the preamplifier, but maybe it helps a bit.
One way would be to build something with a differential pair in a feedback loop, another option is an inverter-like structure: two complementary common-emitter stages in parallel. With translinear circuits, you could make just about any non-linear curve you like. It could also be done digitally, of course.
Not IME. Have lot's of sixties and seventies LPs that did not take gracefully to digitizing. A good digital reproduction chain is still worth having for those, but there is lots of material from that era beyond salvation. Recent reissues and remasters are also a hit and miss but generally much better than the original transfers.
The entire point of maintaining an analogue rig, at least for me, is to be able listen to music from that period without cringing.
Fixing things is not always that easy.
Adding 'inverse distortion', perhaps.
But that depends on what type of distortion you are talking about.
OK forms of distortion:
Inverse frequency response curve to flatten things out.
Inverse compression (expansion) to get the dynamic range back.
Not OK forms of distortion:
Just try and eliminate Intermodulation distortion products when they are already in the recording.
Just try and eliminate harmonic distortion products when they are already in the recording.
Just try and eliminate noise, without removing any of the music, when the noise is already in the recording.
Just try and eliminate hum, without removing some bass notes, when the hum is already in the recording.
Adding 'inverse distortion', perhaps.
But that depends on what type of distortion you are talking about.
OK forms of distortion:
Inverse frequency response curve to flatten things out.
Inverse compression (expansion) to get the dynamic range back.
Not OK forms of distortion:
Just try and eliminate Intermodulation distortion products when they are already in the recording.
Just try and eliminate harmonic distortion products when they are already in the recording.
Just try and eliminate noise, without removing any of the music, when the noise is already in the recording.
Just try and eliminate hum, without removing some bass notes, when the hum is already in the recording.
When the intermodulation and harmonic distortion products are due to an instantaneous smooth non-linearity, you can in theory get rid of them with an inverse instantaneous non-linearity. It is even possible (but much more complicated) for the non-instantaneous case.
Unfortunately, there are always bandwidth limitations, so you never really have the instantaneous case. What's worse, when the recording is a mix of, for example, a seven-track multitrack recording and only one track is distorted, attempting to apply an inverse distortion to the mix will do more harm than good: you get intermodulation products between all seven tracks.
Still, adding an expansive third-order distortion knob to your preamplifier will do no harm as long as you can turn it off. Maybe it will prove useful in some cases.
Unfortunately, there are always bandwidth limitations, so you never really have the instantaneous case. What's worse, when the recording is a mix of, for example, a seven-track multitrack recording and only one track is distorted, attempting to apply an inverse distortion to the mix will do more harm than good: you get intermodulation products between all seven tracks.
Still, adding an expansive third-order distortion knob to your preamplifier will do no harm as long as you can turn it off. Maybe it will prove useful in some cases.
Dynamic range expansion to correct for dynamic range compression with unknown parameters is also not straightforward, by the way. I once built a receiver for a Hicom DiPort 10 radio reporter transmitter, which is an FM transmitter with built-in 2:1 compressor for noise reduction. I reverse-engineered the compression as well as I could, but still didn't get the parameters quite right. As a result, the sound coming out of my receiver always sounded a bit weird.
I've always known that. The good news is that most of my troubled recordings seem to have no audible clipping distortion. Observing spectrograms in Izotope Rx will show this; and it's a fairly simple matter to use its De-Clip utility if needed.
However, it seems that anyone would be hard pressed to find anything that could decompress a recording without consequence, especially to final stereo or even mono mix. I posted the question at the Roon Labs forum and was told: Eventide makes a compressor that is also an expander and dynamics reversal unit, called the Omnipressor. It can analyze the signal in advance and adapt to a degree. You could try to run every track you own through it one at a time and see if you get any improvement to your ear. I expect you would mostly get weird sonic effects, which is why it is a useful tool for producers.
In any case, except perhaps via an expertly developed AI solution, I don't see much hope for repairing these compressed recordings, and probably even less so for whatever actually makes some of them sound congested.
I was hoping that Roon or Jplay Windows players-or https://www.sonicstudio.com/amarra/, even though I have no Mac-would have developed software-based fixes for this and for the other damage I've discussed. But nothing from them so far. Hopefully these guys.
https://community.audirvana.com/c/studio/55
Last edited:
I suspect that remastering software is very sophisticated: it can correct tape saturation along with wow and flutter, frequency response, and reduce noise. "Let It Be" on "Beatles 1" sounds like it could have been recorded yesterday.
Ed
