Best reference song is a song that you know by heart and has decent mastering (nothing crazy). I like Metallica - Nothing else matters, because I have been listening to it forever and know exactly how it should sound. Useful when triamping with three volume knobs. 😧Jazz songs and solo guitar can sound very clear, but can hide problems with midbass or treble. Great if you have problems, but want to impress your friends or customers, for instance with Chet Baker - But Not For Me or Nils Lofgren - Keith Don't Go.
Last edited:
Tom Danley is another advocate of this technique if a person wants a lead on what to search for here. Not that you can't achieve your goals without it but it's potential is compelling.
^
https://www.diyaudio.com/community/...ce-back-into-loudspeakers.105136/post-3143083
Plasnu,
I don't think a list of tracks used as reference from members would give meaningful info except from which generation ( how old) the member is.
I use too tracks which contains 'flaws' in them ( stereo micing techniques introduced) as i discovered the better the system/loudspeakers are the less annoyable this flaws are: you ear them as they are, not more or less, technical issues/ esthetic choices.
On lesser quality system their own flaws can often make the recording unlistenable by additive effect.
https://www.diyaudio.com/community/...ce-back-into-loudspeakers.105136/post-3143083
Plasnu,
I don't think a list of tracks used as reference from members would give meaningful info except from which generation ( how old) the member is.
I use too tracks which contains 'flaws' in them ( stereo micing techniques introduced) as i discovered the better the system/loudspeakers are the less annoyable this flaws are: you ear them as they are, not more or less, technical issues/ esthetic choices.
On lesser quality system their own flaws can often make the recording unlistenable by additive effect.
Last edited by a moderator:
Clickable ones include the London Symphony Orchestra and Danish National Symphony Orchestra on Youtube. These would be excellent recordings if not for Youtube's lossy compression. The microphones are often visible.
I like records put out by Musical Heritage Society (70s/80s).
Recordings of small acoustic ensembles, solo piano, acoustic guitar, and nature sounds are also good for assessing realism.
Ed
In general, as has been said - any music you love, know well, and is well recorded. Having said that...
The good old HFN/RR test disk has "the dynamic range of real life" -- a large garage door slamming. That's a good test! Musically -- Frank Zappa is good, anything from say, One size fits all. Nigel Stanford's Solae Echoes too - Cymatics is a good speaker tester...
The good old HFN/RR test disk has "the dynamic range of real life" -- a large garage door slamming. That's a good test! Musically -- Frank Zappa is good, anything from say, One size fits all. Nigel Stanford's Solae Echoes too - Cymatics is a good speaker tester...
There's the odd label specialising in seriously good recording of what's there (rather than what some producer wants).
John Marks Records in the 90's being a good example.
The various national broadcasters can also do a good job from time to time. The Australian ABC has a good track record.
Likewise, Deutsche Gramaphone vinyl records are pretty reliable. Although DG and ABC/BBC occasionally stray into multimicrophone land. Usually making lovely music but not necessarily "reference material".
Speaking of material, it's easy to record a solo instrument or small group with just two microphones. However the really difficult material to reproduce is also tricky to record: e.g. symphony + massed choir + soloists + pipe organ (cannon optional).
As a reference for evaluating gear, you need the best recording possible. That is the limit of what you can determine with your ears. I agree that originals are always better than re-masters which always have an attempt at improvement but always fall short instead.
From a non-audiofile, whatever you’re going to be listening to. Don’t use something you’re never going listen to again.
For those who said WTF? here is some background
Tom Danley apparently* said the following:
This is a valid technique but it also records the sound of your room and microphone.
Back in the 60 or 70s someone set up this up as an art installation, starting with some spoken words and ending up with the "boing" of the room later in the day.
*I found this attributed to him in another thread, not a directly traceable quote back to Tom.
Some re-masters are good, others not. That's re-master, not re-mix; that's a whole different issue/can of worms.
In our collection, this applies to LPs as well as CD: my first LP copy of 'Who's Next' sounded very good but it got trodden on at a party; the second LP copy was an Australian 're-master' and re-cut which sounded thin and miserable. The second, expanded CD issue sounds great, easily the best of the lot, so I'd choose that as a 'reference' recording. There was a Who double best of - I can't recall the name - but the 'remastered' tracks on it sounded quite thin, perhaps because they tried to fit too much on each side.
The latest CD remastering of Electric Ladyland sounds excellent; the first was full of hiss and the 'No Noise' remaster was bleah. I've had over time three LP and CD issues, love it and know it backwards, but I don't use it as a 'reference' album as the bass and drums are a bit thin and slightly buried in the mix.
Geoff
Your ears can hear when it's right, that's important, but your ears can't reliably identify all the issues. I(/we've) spent years trying.
So I don't feel you should be dependent on specific songs for this because if you have to analyse what you hear so much that you need it to come from the same song over and over then it's probably beyond your ears to identify the cause.
Instead you should rely on a partially academic process, and spend your listening test time aiming for naturalness and equalising. Some songs are poor, but many are able to average out to the point you can tell, once your design is reconcilable.
So I don't feel you should be dependent on specific songs for this because if you have to analyse what you hear so much that you need it to come from the same song over and over then it's probably beyond your ears to identify the cause.
Instead you should rely on a partially academic process, and spend your listening test time aiming for naturalness and equalising. Some songs are poor, but many are able to average out to the point you can tell, once your design is reconcilable.
The songs we use must include certain requirements:
1. Very high quality recording
2. Digital and high quality vinyl versions available
3. They must exercise various performance limits of the system in a variety of parameters
The minimum set of songs that meet enough requirements has been found to be two. Keeping the number small means the songs can be known in great detail. Also, a small number that are very well known can make it possible to reliably compare how the songs sound on one system to how they sound on another system even if listening to the systems happens at different times and places (not necessarily quick A/B). A friend can memorize system performance well enough to reliably compare the sound of a system at one point in time to how it sounds months later (NP might have a pretty good idea of who that person could be... ).
The songs we use are:
Janis Ian - Breaking Silence 16/44, 16/44->DSD256, 45 RPM Vinyl
Alan Parsons - The Voice 16/44, 24/192, DSD64 (SACD), 24/192->DSD256, 45 RPM Vinyl
Breaking Silence is good for evaluating precise imaging, low IMD, extreme dynamics, very low frequency drum rumble, etc.
The Voice is the more musically dense recording of the two; it includes some other things such as male vocals, a variety of instruments playing in different combinations, useful for assessment of "black level" between instruments, etc.
1. Very high quality recording
2. Digital and high quality vinyl versions available
3. They must exercise various performance limits of the system in a variety of parameters
The minimum set of songs that meet enough requirements has been found to be two. Keeping the number small means the songs can be known in great detail. Also, a small number that are very well known can make it possible to reliably compare how the songs sound on one system to how they sound on another system even if listening to the systems happens at different times and places (not necessarily quick A/B). A friend can memorize system performance well enough to reliably compare the sound of a system at one point in time to how it sounds months later (NP might have a pretty good idea of who that person could be... ).
The songs we use are:
Janis Ian - Breaking Silence 16/44, 16/44->DSD256, 45 RPM Vinyl
Alan Parsons - The Voice 16/44, 24/192, DSD64 (SACD), 24/192->DSD256, 45 RPM Vinyl
Breaking Silence is good for evaluating precise imaging, low IMD, extreme dynamics, very low frequency drum rumble, etc.
The Voice is the more musically dense recording of the two; it includes some other things such as male vocals, a variety of instruments playing in different combinations, useful for assessment of "black level" between instruments, etc.
Last edited:
In my student days I had a six week holiday job on Myer Melbourne rooftop carnival: they had one reel to reel tape: Neil Diamond's Hot August Night, which I didn't like to begin with, played all day every day over the PA, for those six weeks - 40 plus years later I still hate it!
Geoff
Thing is it's tough to find something re-mastered that hasn't been tampered with. I have never seen "re-mixed" stated on a re-master. Even when they're not re-mixed they're at least compressed. I have a box set of ZZ Top's studio collection that was considered a disappointment from reviews I saw but they are absolutely untouched original recordings with all the dynamic range present. IMO they are perfect and sound awesome. When people don't "hear" a difference they think they've been short changed. I have the latest Beatles Sgt. Pepper and Abbey Road that I listened to just once to recognize actual missing information. I have not listened to them again. Very disappointing eye openers. Subtle nuances I'm used to cueing to are gone along with significant low level resolution. With the Stones' Exile on Main St. it's back to the loudness war. All of my early first versions of everything sound best by far. Except of course those that are untampered with, easily recognizable.
Oops, my bad. I just looked and see "New Stereo Mix by Giles Martin" on the Beatles' Sgt. Pepper and Abbey Rd. 🙂
Jump - Van Halen
The guitar intro repeats except for the choruses and instrumental. The level and presence should stay constant throughout, not be overshadowed and hold it's place in the stereo image.
The guitar intro repeats except for the choruses and instrumental. The level and presence should stay constant throughout, not be overshadowed and hold it's place in the stereo image.
Thelma Houston & Pressure Cooker: I've Got the Music in Me
I have 2 copies of the original direct to disk vinyl. These are gone forever.
However, they had a backup tape recorder running during the session and the content was release in various forms.
Note that Bill Schnee, the recording engineer, also did the Steely Dan Aja mentioned previously
Also good: Emmylou Harris - Quarter Moon In A Ten Cent Town.
I have this Mobile Fidelity vinyl, also unobtanium. The content is also available in various forms.
I have 2 copies of the original direct to disk vinyl. These are gone forever.
However, they had a backup tape recorder running during the session and the content was release in various forms.
Note that Bill Schnee, the recording engineer, also did the Steely Dan Aja mentioned previously
Also good: Emmylou Harris - Quarter Moon In A Ten Cent Town.
I have this Mobile Fidelity vinyl, also unobtanium. The content is also available in various forms.