Digital Audio Renovation (DAR) is a radical new audio software plugin which results in a dramatic improvement in sound quality from CD, MP3, Flac and other digital formats.

• The plugin integrates easily with free computer audio players such as Winamp and Foobar2000.
• It can be used for both real-time playback and the production of permanently improved tracks (where copyright laws permit).
• The website includes a range of playable DAR-adjusted tracks, and there is also a free time-limited demo available for download.

http://www.digitalaudiorenovation.co.uk

So, what is the radical innovation of this software? Sound degradation?

Is it planned to release a new, enhanced version, which produces "AM radio" like sound quality from hi-quality audio sources?

I tried the demo on Foobar 2000 on a pc running XP (didn't work on a Vista pc) to convert an old style jazz song recorded with today's audio standards. The effect was very pronounced; bass almost completely disappeared. Closer to a shellac 78 rpm than a vinyl 33⅓ rpm, to my mind, in terms of apparent frequency response. A nostalgic sound.

Emphasis mine. While it may make the music sound better for some people, i think improving the sound quality is a bogus claim. If it changes the music, it degrades it.

Moved out of News Submissions, as there is no proof that this component truly results in "dramatic sound improvement" as per our Terms of Service, item 8. Also, please read the wiki regarding vinyl audio quality myths.

I would tend to agree. The only way to realistically improve the "digital sound quality" of a recording would be to re-record the original source, be that a live performance or a 1/2" master, with superior equipment/a higher bit depth/a higher sampling rate (and the latter two would not likely yield any audible improvement in quality anyway). Any post-recording process will always be destructive to audio quality in the purest sense of the term.

I also thought this quote was amusing:

None of these "flaws" are characteristic of what is by and large a perfectly neutral, linear and accurate medium we call digital audio. I don't even know precisely what "blurry cymbals" means, if anything.

Now, I certainly don't have any objection at all to these types of plug-ins. If you want to listen to music in any particular way, even if that's reducing the bit depth to three or four bits per sample (which is sometimes sonically interesting), I say "enjoy!". What I most certainly object to are these ridiculous claims about a plug-in's so-called ability to "improve quality".

iZotope has a FREE plug-in called Vinyl that can degrade your audio:

foobar2k has a similar realtime dsp, with similar results

First of all many thanks for all the interest in this new product.

Re Post #5:



8. All members that put forth a statement concerning subjective sound quality, must -- to the best of their ability -- provide objective support for their claims. Acceptable means of support are double blind listening tests (ABX or ABC/HR) demonstrating that the member can discern a difference perceptually, together with a test sample to allow others to reproduce their findings. Graphs, non-blind listening tests, waveform difference comparisons, and so on, are not acceptable means of providing support.


Apologies for the oversight.

Unfortunately UK laws do not allow the provision of test samples of copyrighted works (under fair use exemptions or otherwise) so fulfilling the above to the best of our abilities can only include the ABX testing aspects.

We have now carried out ABX tests on five different pairs of tracks, each one with and without the Digital Audio Renovation (DAR) adjustment pre-applied respectively.

25 trials were used with each pairing, and scores of 25 out of 25 correct (i.e. 0% probability of guessing) were obtained in every case.

Here is a representative example:

foo_abx 1.3.1 report
foobar2000 v0.9.5 beta 2
2009/02/03 12:10:46

File A: C:\3 Feb ABX Tests\ORIGINAL Goldfrapp - Ooh La La.wav
File B: C:\3 Feb ABX Tests\DAR-ADJUSTED Goldfrapp - Ooh La La.wav

12:10:46 : Test started.
12:11:11 : 01/01 50.0%
12:11:16 : 02/02 25.0%
12:11:19 : 03/03 12.5%
12:11:22 : 04/04 6.3%
12:11:35 : 05/05 3.1%
12:11:39 : 06/06 1.6%
12:11:42 : 07/07 0.8%
12:11:44 : 08/08 0.4%
12:11:48 : 09/09 0.2%
12:11:51 : 10/10 0.1%
12:11:54 : 11/11 0.0%
12:11:57 : 12/12 0.0%
12:12:00 : 13/13 0.0%
12:12:12 : 14/14 0.0%
12:12:15 : 15/15 0.0%
12:12:18 : 16/16 0.0%
12:12:31 : 17/17 0.0%
12:12:34 : 18/18 0.0%
12:12:54 : 19/19 0.0%
12:13:07 : 20/20 0.0%
12:13:10 : 21/21 0.0%
12:13:13 : 22/22 0.0%
12:13:17 : 23/23 0.0%
12:13:30 : 24/24 0.0%
12:13:34 : 25/25 0.0%
12:13:51 : Test finished.
----------
Total: 25/25 (0.0%)

Summaries of other tests:

foo_abx 1.3.1 report
foobar2000 v0.9.5 beta 2
2009/02/03 11:29:55

File A: C:\3 Feb ABX Tests\ORIGINAL Police - Roxanne.wav
File B: C:\3 Feb ABX Tests\DAR-ADJUSTED Police - Roxanne.wav
----------
Total: 25/25 (0.0%)

foo_abx 1.3.1 report
foobar2000 v0.9.5 beta 2
2009/02/03 11:36:55

File A: C:\3 Feb ABX Tests\DAR-ADJUSTED Beatles - Help!.wav
File B: C:\3 Feb ABX Tests\ORIGINAL Beatles - Help!.wav
----------
Total: 25/25 (0.0%)

foo_abx 1.3.1 report
foobar2000 v0.9.5 beta 2
2009/02/03 11:40:36

File A: C:\3 Feb ABX Tests\ORIGINAL Elvis Costello - Watching The Detectives.wav
File B: C:\3 Feb ABX Tests\DAR-ADJUSTED Elvis Costello - Watching The Detectives.wav
----------
Total: 25/25 (0.0%)

foo_abx 1.3.1 report
foobar2000 v0.9.5 beta 2
2009/02/03 11:50:43

File A: C:\3 Feb ABX Tests\DAR-ADJUSTED David Bowie - Suffragette City.wav
File B: C:\3 Feb ABX Tests\ORIGINAL David Bowie - Suffragette City.wav
----------
Total: 25/25 (0.0%)


Many thanks again for the interest.

I believe you've misunderstood the basis for his moving the thread based on TOS 8. It's claimed that this product produces, in your own words, a "dramatic improvement in sound quality". Your product wasn't described as producing a "different" output, which ABX testing could/would plainly reveal, but an "improved" output, which ABX testing cannot demonstrate. foobar's ABX comparator is only designed to allow users to attempt to discern one sample from another.

If "sound quality" is defined in the traditionally-accepted manner, there is no way your statement could be viewed as accurate or even potentially accurate. From a technical perspective, no post-recording process with a finite level of processing precision can yield an output that offers an improvement in sound quality to the original input.

You'll notice that Izotope avoids making any claims regarding sound quality with their Vinyl plug-in, only claiming it offers an authentic vinyl simulation.

The only way to somehow "prove" better sound quality would be to let a certain number of people judge whether they prefer the original or the processed version of an audio file. Of course the test persons may not know which one is which. When performed correctly, the claim "xx% of all test persons prefer DAR over original recording" could be made.

NTS, you completely misunderstand what I was getting at. The site makes absurd claims like:andAs per the vinyl myths HA wiki article I linked to, this is pure nonsense. Vinyl represents a decrease in audio fidelity (and therefore quality, unless you consider low-fidelity a quality) from CD and FLAC. There might be a market for this product among subjectivist audiophiles who buy into nonsensical, unscientific claims like these, but Hydrogenaudio is not that market.

Furthermore, many of us prefer to hear the highest-fidelity reproduction of the bits recorded onto a CD. Adding DSPs to the playback chain is kind of anathema to such a view.

Awesome! Now we have proof that a massively distorted copy of a CD track, sounds different from the original.

As the primary author of the Vinyl Myths article, I'd like to point out that some vinyl distortions really are considered euphonic, even though all of them are ultimately degrading to the objective sound quality. Tonearm resonances can cause a bass boost and widen the soundstage; harmonic distortion in the recording/playback chain can sweeten the sound; increased background noise can affect timbral perception, etc. But JJ knows more about that than I do so I'll defer to him on more specifics.

So I don't think it's a TOS8 violation, insofar as euphonic distortion is a generally allowable topic here. But I'm still as skeptical as anyone else is. The OP has got to be a lot more descriptive as to what's going on before I would let this plugin anywhere near my music.

Not providing samples is some seriously weak sauce. SRSLY, go pull some Creative Commons media from archive.org and have at it.

Except... Not all recordings are "perfect", and sometimes the sound can be improved. For example, when I digitize vinyl, I use tools to remove the ticks & pops, and with older recordings I sometimes use EQ to restore some of the highs.

But, this requres human judgement to identify the particular defect or weakness, and human interaction (and trial-and-error) to correct or minimize the defect. I wouldn't blindly process all of my files with a "sound improving" program.

More generally, distortion removal requires either the distortion to be (more or less) linear and accurately estimated, so that the distortion can be cancelled out (ie, it is an EQ problem with a known equalization curve). Or it is a distortion that is easily isolated from the rest of the music (pops/ticks are very temporally focused, no-noise circuits operate by separating the noise floor from the music itself, etc).

I don't know how to ask this without causing offence to NTS (who seems quite sincere), but I can't think of any other way of saying it...

Is this is joke?!

Is there a person on the planet who wants their audio to sound like this?

Unless it's faulty on my system, it's exactly as MLXXX described - the frequency response is closer to a telephone than an LP.


I don't know about everyone else, but on a decent system I like to leave the frequency response flat, and on a poor system, I usually want more bass and treble, not less!


EDIT: I suppose it could be a clever hoax. Do something so stupid that lots of audio boards link to the website for a laugh, and then switch it for the real product. The page will now have a much higher google ranking than it would have ever achieved if they had launched with the real product immediately.

Or I suppose a much simpler explanation is that I've just installed a virus or spyware on my PC!

Cheers,
David.

Good afternoon all, and in response to some of the queries about the Digital Audio Renovation plugin we have decided to include a FAQ section in the support page of the website:

http://www.digitalaudiorenovation.co.uk/dar_support.html

For those who have either already downloaded or are about to download the demo, particular attention is drawn to FAQ number one re the potentially deceptive nature of speedy sonic comparisons:

Q: Sometimes when I switch quickly from a non-adjusted to DAR-adjusted version of the same track the DAR-adjusted one sounds a bit flat.

A: Both the eye and ear have a tendency to be temporarily deceived by the 'bigger is better' syndrome (visually for example over brightness and saturation, in the audio sense regarding overall volume, relative bass and treble levels etc). In reality it is quality rather than quantity which counts.

For this reason it is recommended that any evaluation of the DAR plugin involves at least some prolonged and uninterrupted periods of listening with the effect applied, rather than constantly switching back and forward between on and off settings.

On a misunderstanding based around some of the terminology used in relation to the plugin:

Digital Audio Renovation is not in fact a vinyl emulator, but rather a general digital audio reconditioner. The basic aim of the DAR plugin, like all progressive audio tools, is simply to assist in the full enjoyment of the recorded music itself.

Many thanks again for the continuing interest

Calling it an "improver" is still a stretch. That particular claim bothers me and many of the people on this forum. As per Terms of Service, point 8, I must insist that you provide some form of proof that, if nothing else, listeners actually consider it to be an improvement over an unaltered stream. The term "reconditioner" is technically meaningless.

At best, this DSP provides additional distortion that its producer claims to be preferable to the unaltered version. That's about all there is proof of at the moment.

Good afternoon again, and re post #18:



The term 'improver' has now been removed from the posting, hope this covers the issue.

Regards

NTS

As a hint to NTS: to prove that some listeners prefer that "DAR distortion" you have to set up an ABC/HR test (not ABX). Representativeness of such tests highly depends on number of participating people (you are welcom to call for volunteers on this board).

I'll be honest here, even if I found the "DAR distortion" more pleasing to my ears, my mind and personality would be utterly convinced that it is still wrong. My (personal) goal is accurate representation of artistic intent. At some point I must trust that the product as delivered to me is as the artist intended. Any manipulation of the final delivered product, no matter how pleasing, fails to meet this goal. I might as well apply a colorwash to the Monet on my wall, as my eye likes blue whites.

What I'm getting at is I might very well rate a "DAR distorted" track higher in a ABC/HR test if not intimately familiar with the work. There is a disconnect between the mind and the body.

If it could be shown that most people considered the modified version preferable to the original version for most of the music that they listened to, then wouldn't audio engineers start applying it to the music before distributing it?

And if it really does improve the sound, why not apply it a second time, then a third, etc. Does the sound keep getting better and better?

Hello again, and re Post #22:



Without making any claims here about the success or failure of this agenda, it is precisely the goal of the Digital Audio Renovation plugin to deliver the recorded music as intended by the artists, producers etc.; in other words to remove any untreated digital-chain related discolouration, distortion etc.

One quick way in which the user can evaluate the effectiveness of this is to experiment with a track they are familiar with from its original analogue-recording-to-vinyl incarnation (such as an original vinyl version of an Elvis Presley, Beatles or Clash song), and now possess in a digital format.

Listen to the CD or MP3 version first with then without the DAR conditioning applied, then decide which comes closer to the original (hence artistically intended) sound. Incidentally to make this test most effective the proviso again has to be made that relatively prolonged listening with the effect both on and off, and a pause in between (rather than simply quickly switching back and forward) is necessary to avoid confusing the aural senses.

Many thanks again for the interest.

Your claim of the existence of "digital-related discolouration, distortion etc." seems to me to require proof.

I just tried this out and it's so bad I laughed. This has got to be a joke.

Before anyone else posts I suggest you try it out. This can't be for real.

Who are we to decide how the artists, producers and engineers want their music to be reproduced? I haven't bought a CD in a while -- are they putting this sort of information in the liner notes now? Do the artists recommend purchasing your product and/or other products like it such that listeners can enjoy their recordings as intended?

I'd like to see "untreated digital-chain related discoloration" defined. Such a term is aggravatingly vague.


No, that's typically only a quick way to evaluate the differences between one master (intended for vinyl records) and another (intended for audio CDs). That has very little to do with the technical differences between an analogue recording and a digital one.


Again I take issue with your claim that the vinyl record is universally any closer to what any content producers "artistically intends" than an audio CD/MP3/other high-quality digital format.

Not trusting the artist to correct any "untreated digital-chain related discolouration, distortion, etc." runs counter to my line "At some point I must trust that the product as delivered to me is as the artist intended."

A slight ramble, but still on-topic, I think...

This reminds me of the Yamaha AV amplifier I recently installed and set up for a friend. Despite being perfectly capable of setting up a typical 5.1 surround system without having to read every single word in the manual, I did the good deed and kept it by my side whilst referring to it for anything that didn't seem immediately obvious.

I burst out in fits of uncontrollable laughter when I came to the DSP page where the manual was very insistent that all MP3 files were badly broken and HAD to be processed with their special, custom DSP to make them sound as close to the original source material as possible.

How can an encoding that, when made properly, is completely perceptually transparent to the vast majority of listeners possibly be classified as broken? I did try the DSP and it made all sources sound like massively crushed and distorted FM radio broadcasts to me.

Moral: If it ain't broke (which, generally speaking, it isn't), don't fix it!

Cheers, Slipstreem.

100% agree.

However, there is a glimmer of truth. If you listen to recordings with very little bass and treble for about a month, the "real thing" will sound very artificial when you switch back to it.

Most stereo systems, even expensive ones, have quite artificial sounding treble until you become accustomed to it (which we all have, of course). If you don't hear any treble for about a month, it sounds really bad when you switch it back on again.

If you don't believe me, try it - make sure every reproduced audio you hear for an entire month has the treble turned right down. Of course the real world will still sound as it does, which is why it's such a surprise that turning the treble back up after a month sounds so wrong - the real world should have kept your ears used to normal sounds with normal treble - which it does of course - your ears are objecting to the abnormal treble of a stereo system.

It'll only take a few hours for your stereo to sound normal to you again.

Cheers,
David.

Your claim that all hifi has abnormal treble is interesting but not understood. Your test strongly suggest you mean there is considerably extra or exaggerated higher frequencies. Specifications for most of the equipment involved in music recording and production seems reasonably frequency linear except where deliberate manipulation is involved e.g. equalizers. The same is true of playback equipment.

In fact, my impression is that overcoming limited high frequency response is more a challenge than for most of the audio spectrum. Higher frequencies are generally more easily absorbed, thus harder to record, and harder to distribute evenly on playback. It would seem that for your statements to be true, music would have to be widely mastered with exaggerated higher frequencies. Is this the gist of it?

I am uncertain about precisely what 2Bdecided is trying to say but the treble is indeed one of things that makes a stereo sound like a stereo and not musical instruments in the listening room. If you compare the sound from an omni-directional loudspeaker with that from a traditional loudspeaker the increased treble of the former is striking. This is not caused by deviations from a flat frequency response in the direct sound but by the omni-directional loudspeaker producing a much closer to flat response in the indirect sound. Traditional loudspeakers tend to radiate a flat frequency response within a few tens of degrees from the forward axis with the rest of the radiation containing little treble.

Of course, the directivity of sound radiation varies from musical instruments to musical instrument and so a stereo signal reproduced in a room with a pair of loudspeakers is doomed to sound like a stereo.

I just tried this on some of my wee brothers music(which was digitally recorded to a .wav file then converted to FLAC and tagged.) and is 'as the artist intended' since he is the artist and it sounds 'awesome' to him.

after applying the filter that is meant to play back digital files 'as the artist intended' this utterly failed. sounds like it was recorded in a cheap studion with cheap equipment in the 1930's to me. not entirely sure how that is supposed to be his intentions as it is supposed to be a futuristic/rock/i'm not really sure what he was aiming at but he tells me he got it right thing.

After been following this thread for some days, I believe it has a limited audience (if that wasn't obvious).

That sentence is what more precisely defines what he intended this DSP to do and I don't think it could be considered as quality.


If we read the sentence, it states that "having a digital remaster of an old analog recording, this plugin could conceveably reproduce it as the people heard it at that time, with the equipment of that time".

Of course, that doesn't mean "quality" to any of us.
And saying that it was mastered to be played on the audio hardware of that time doesn't necesarily mean that it was how the artist would have liked it to be.