No.2: Concertino d'ete (for solo viola & nine instruments) Ernst Wallfisch, viola
No.3: Concertino d'automne (for two pianos & eight instruments) Genevieve Joy & Jacqueline Bonneau, pianos
No.4: Concertino d'hiver (for trombone & string orchestra) Maurice Suzan, trombone
4 files zip FLAC Mega Download
Ensemble de Solistes des Concerts Lamoureux conducted by Darius Milhaud
Philips 6527 221 1984 Stereo reissue. Recorded: June 1958 Re-edited Nov.2012
You can compare this to 'Mr Pristine's' Fizzed-Up drubbing ('Printemps' mp3) MusicWeb Review (Herr Barnett didn't 'notice' the jacked-up treble...) ...
Thank you for your intestimabile work, I hope you'll publish more English music and Delius in particular, would be nice to listen to his sacred music in some historical interpretation.
ReplyDeleteGreetings from Italy ;)
I cannot guarantee I will continue this (I genuinely HATE editing LP's).
ReplyDeleteBTW - this Milhaud is the first I've edited solely on the laptop (thus missing-out a 'save' - ie: just an EAC 'rip' + the FLAC saved from the Audacity 'live' edit).
Re: 'sacred music'that's (in terms of Delius) the 'Mass of Life'..
Yes, well, I've brand new Fontana copies (also of Appalachia) + others on Philips/etc: and was thinking of adding the 5LP World Records box set of the (mainly) pre-war Beecham discs - but they are on CD from the same transfers.
There is the very interesting WHITTAKER 'Viking LP', mentioned on the R.R.Bennett thread: but I've made 5 dubs already - the remaining files reside on a faulty Hard Drive - but do have the majority of 'English Music' LP's issued (not all having been on CD).
See how things go; but I quickly get 'jaded' after editing an LP or two in a week...
Yes please to Whittaker! Many thanks!
ReplyDeleteYes, in the case of Delius would be more appropriate to speak generally of choral music, anyway thanks again for your work, the choices are always interesting.
ReplyDeleteGreetings ;)
I'll take a listen to my 3 LP transfers of Beecham's RPO 'Appalachia' (Fontana/Philips/CBS Classics) and may well do an upload.
ReplyDeleteThe later EMI RPO choral's (Sargent/M.Davies/Groves) are now close to being 'Historical'?
Even though I only occasionally listen to Delius it appears I've most of the LP's issued (not to mention the 78's..)!
Speaking of Sargent, his interpretation of Elgar's 1st Symphony never appeared commercially but was recorded live at the Proms. That would make a wonderful upload (a bit of it appeared on the 'Music Maker' LP devoted to Sargent. Thanks!
ReplyDeleteWell, I find that the music of Delius has many valuable qualities, his lightness of inspiration, beautiful orchestration, his sense of poetry in music combined with a strong formal structure that rescues him from the mark of an easy impressionist.
ReplyDeleteTo me his major works are the Violin Concerto, Piano Concerto, 'A Village Romeo and Juliet' and 'A Mass of Life'.
You're lucky to have all of these discs of Delius, now my curiosity is beyond the bearable! :D
(apologies for my basic english)
Sargent's Elgar Sym.1 I certainly don't have..
ReplyDeleteIf you know of 'British Music' LP's (or even 'parts of') which have not appeared on CD (or have been 'out of print' for many years) then maybe you could make a list?
There is, on LP, BBC SO/Groves Holst 'Perfect Fool' (+ other English items) from the 1974 'Last Night of The Proms' which I've wanted to upload - also BBC SO/Sargent on HMV ASD 536 - a 'fake' Proms concert - which might not have been on CD - though the performances are not 'too hot'..
You did say a while back that the Aeolian Quartet playing Elgar/RVW quartets on Summit was for consideration.
ReplyDeleteI have the same quartet with Cassini playing Elgar Piano quartet on Delta...a LP I like a lot.
Looking for some time,so far unsuccessfully for the quartets(and the violin sonata from the same source.
So if upload possible.....
I 'never forget' intended uploads!
ReplyDeleteThe Summit Elgar/RVW:1 (TLS 6052 -stereo) is 'new' - but not sure it would turn-out 'noise-free'.
That nice Elgar Pf Qnt/Bax is the stereo Revolution reissue - RCB.8 - CBS mastered/pressed.
(Am 'missing' the Elgar Vln Sonata - c/w Martinu..)
There's also John McCabe's Elgar piano recital on Prelude PRS 2503 (+ have one of his two Decca SDD 'English' recitals) - to join the queue....
Argo ZRG 5475 'Instrumental Music' is anyway next, this week.
Thank you very much for uploading!
ReplyDeleteFrank, I know you are spending a huge amount of time to 'manually' (i. e., one at a time) edit out the little clicks that are inevitable on even mint LPs. Most admirable!
ReplyDeleteBut, rather than burn out from this monumental amount of work, and give up the blog, might I suggest something that could help. There is a Windows or Mac program by a guy named Brian Davies, called ClickRepair:
http://www.clickrepair.net/noise/
Several friends of mine use it and I've tried the limited-time demo. The minimal settings don't do harm to most recordings.
Here's what I do. I employ Audacity, and "highlight" the sections of a movement that are not the loudest, brightest parts; and run a plugin I got years ago (when I used the earlier shareware version of Audacity, called CoolEdit). This plugin is called ClickFix:
http://www.jdklein.com/clickfix/
I have found that with the default "normal" setting, Lp rips -- except the very loudest, brightest passages -- have a huge amount of the clicks removed with no audible change in the tone quality, transparency, or HF brilliance. But...if I run the VERY loudest, brightest parts, I sometimes DO hear a slight extra "IM fuzz" when the software gets confused by the dense highs, "thinking" that the waveforms are anomalous noise transients.
Just running an entire movement all at once, in default, is OK if the record is quite old (say, a pre-1950 recording) though one might afterward have to tweak one or two really LONG scratch noises or thumps that don't have a big HF component.
But it's a hundred times faster than doing the noise reduction 'by hand' (as I had to do until I got ClickFix.)
I've been customarily running your uploads thru ClickFix, merely "excepting" the brightest, loudest passages on something like a Decca ffrr Lp. Takes a few minutes but the end result is almost CD-quiet background.
There are exceptions, of course. For example: the Oiseau-Lyre stereo LP of the brass music from King James' court was ALL too bright for ClickFix. Even the slightest touch caused 'fuzz'. So I left that alone; but your copy of the LP--after your laying on of hands--was very good.
But, I've been able to knock back quite a bit the residual "HMV crackle" in the old EMI Lp transfers. No sane human being could possibly edit that out "by hand" as there are billions of tiny transients, way down in volume. I used to listen to these records years ago without the slightest annoynance, but now years of "CD conditioning" has made me more sensitive to this stuff!
Whereas, I think that, for instance, the typical processing of Rose, EMI, or Naxos, etc., is WAY too invasive. Sure--they get a silky smooth background (changing the little transient scruff into a sort of constant soft hissing)-- but somehow it's just not REALISTIC that way.
Your rips are remarkably realistic! But they are merely nicely un-altered: competent rips, with some human editing, of decent LP jobs. I like to take that one small step further. If I were to quantify this, I'd say: if EMI/Rose/Naxos represent "100%" noise reduction, your rips are "10%" noise reduction. I like to bump that up to, say, "30%" by a light application of ClickFix. So far you have used great original sources and the much more efficient ClickRepair -- which REALLY goes deep! -- isn't necessary.
But, if somebody is trying to resuscitate a horrible record, that program can do the trick (with care in its ajustment.)
SO: please please please DO NOT GIVE UP!
Bertie
Unfortunately, there is a noticeable 'change' if you use de-click software (although I've not tried Click-Repair).
DeleteTo take an example of another Philips '80s Dutch re-pressing: Hanson Sym.4.
The same matrix-pressing - Andrew Rose/mine - http://www.pristinestorage.com/samples/force_download.php?file=PASC292.mp3
Do you believe that his 'blurred/low-definition' transfer is acceptable??
Barnett, in MusicWeb, describes it as 'listenable': so is the music piped through a Shopping-Mall....
I'm not interested in low-definition 'CD quiet' transfers (even if folks LP record/offer '24/96' - or '32 bit' based - as they all are 'cleaning the LP recording' and removing lots of recorded information...and the '12-bits' on the LP will be waaay below that by the time they have processed-it-to-death ...and this is what Rose uses: http://www.soundonsound.com/sos/apr11/articles/andrew-rose.htm
..and using the same 'de-clicker' at the lowest '1' setting begins to remove fine-detail/'smooth-over' the sound....
Email him and ask what settings He uses...
When I finally 'give up' (next month?) and return to spending the time listening to LP's 'for myself' (and as opposed to the broadcast BBC aura-wallpaper 'CD quiet' dross - via any of 14 FM/2 DAB tuners...) then, as you imply, there will be nowhere-else to get such 'flawed' transfers as these...which clearly are a 'waste of my time' offering...
Still..when you have no option but to Pay for a Rose-Job - then 'all the work' will have been done for you! (and you are welcome to 'enjoy' such substandard rubbish!!)
I think, Frank, that we have a slight difference in the way we each describe what we hear and that, unfortunately, I seem to have suggested, in your perception, that I want a different kind of sound than you do.
ReplyDeleteI don't!
I agree *100%* with your criticism of Rose's transfer. Absolutely.
What I am talking about is not continuous hissing or the HF presence, but merely individual one or two cycle anomalous record transient noises. Usually when ClickRepair analyzes them, in music that does not have extreme HF musical content, only these very short-duration (milliseconds long, no longer) will be altered. The ear takes much longer than that to quantify "quality" so removal of a one-cycle transient SNAP is not identifiable as "dulled sound".
I enjoy analyzing and discussing this stuff so please don't take my reply as an attempt to debate.
The way I have experimented with ClickFix (which is what I use, as I never paid for a registration for ClickRepair but only tried it out during the thirty-day evaluation period) is to make two identical bit-for-bit copies of a sound file that I felt had too many clicks and pops. I then ran ClickFix on the SOFTER passages only. Following that, I *subtracted* the two files from each other (this is easy to do by inverting the phase on one or the other and then doing a mix.)
If anything CHANGES, that is all you hear--otherwise the two signals, being 180 degrees out of phase, perfectly cancel.
With the proper application of ClickFix in the softer parts of the file (since the very loud ones naturally mask the faint crackles) the subtraction process -- in the tests I have done -- have virtually NO sound at all...except the tiny crackles!
No musical highs that are identifiable as partials, related to fundamentals. No constant signal of any sort. Just isolated snaps and pops! It's really quite amazing, Frank!
FWIW, I've also done the same thing with two identical WAV files, unprocessed, converting one to FLAC and then back to WAV again -- and then subtracting them. If the FLAC process makes an audible change, the subtraction should reveal the difference. IT DOES NOT--in my tests, FLAC was completely transparent. It yields about a 50% reduction in file size, on average; but the compression was totally reversible and did not alter the original samples!!
That one really surprised me...as if you do the same thing with, say, mp3 (even 320 kbs) you can here differences after the subtraction.
I satisfied myself that applying ClickFix to the passages, up to (say) mezzo forte, of most LP records (save ONLY the brightest, punchiest stuff like Mercury or Decca) that I could use ClickFix and get rid of much of the crackling, to the point that the background was quiet--sans the clicking. But not alter the HF *musical* transients, the string partials, the "air".
When ClickFix *does* fail is almost always in very loud passages with brass, and *sometimes* with cymbal crashes at full volume. So I do not run it thru those parts of the file, in my sound editor. I highlight only the quieter parts, below (say) 50% digital modulation level.
END OF FIRST PART OF MY REPLY...
Bertie (to be continued)
CONTINUING...
ReplyDeleteAny music that is 'busy', with lots of frequency content all over the spectrum, and at louder dynamics, tends by definition to cover up the tiny, faint crackles--so that can be left alone.
When I started using ClickFix, it was to try to clean up old WRC pressings of Griffith transfers of wonderful Beecham performances. There is a two LP set of French music that has his 1930s Carmen and L'Arlesienne. I ran ClickFix on my rip but the spots where there was strong brass were, on playback, kind of 'hashy' and vaguely distorted--worse than the raw unmodified rip. I was disconsolate, until I realized that all I had to do was highlight (in the editor) ONLY the other parts, without the problematic highs!
When I got thru, I ended up with a background that was nearly as lacking in crackle as a Dutton transfer--but it also had the original slight acoustical "air" that Dutton always, inevitably wipes out.
All that I got rid of was...crackle.
Yes: some records are SO constantly noisy that if you DO take out just crackle, on playback they SEEM less vivid. This is a psychoacoustical effect! I tried the process with a very old Pearl CD of the Muck Parsifal electrical records, which has countless billions and billions of ticks, as well as obviously boosted highs. Wow. I took out only the crackle, and it seemed LESS BRIGHT. But when I focused my attention closely on the actual musical sounds of the stringed instruments, I realized I was still hearing exactly the same frequencies and harmonics. The crackles just sort of added to a sense of presence, but were not real MUSIC.
So there are sometimes differences but it is important to try to quantify them and make adjustments as necessary. That Pearl transfer was really bad...worst-case. Whereas, the WRC transfers, done (probably) from matrices or maybe vinyls struck from them, are often remarkably quiet to begin with--certainly quieter than the released shellacs. Taking out the last trace of crackle -- but not the actual musical content -- makes them (in my intended analogy) -- sound "CD quiet".
But I don't mean, by that, Dutton-CD quiet; nor EMI-CD quiet. I merely mean "quiet" in the sense that a CD made from an analogue tape master, or a digital master, HAS NO CLICKS AND POPS.
So that's what I'm getting at. Careful processing of ONLY the clicks and pops can be done automatically -- requiring some degree of hands-on tweaking -- removing a degree of the transient noises that are impossible (due to their huge quantity of individual cycles) for a human editor to take out one at a time!
I've been doing digital editing since c.1998, myself, with restorations; and before that I did live digital recording dating back to the days of the Sony PCM boxes that used U-Matic cassettes (they weren't editable in those days!) and have done analogue tape mastering since c.1962, having transferred about 500 multi-disk 78 classical sets, and several thousand Lps during work I did for a syndicator of programs. I have a positive abhorrence of heavy handed "erratic filters" that leave loud parts bright but dull the soft parts. Drives me CRAZY. The nice thing about ClickFix is that, used judiciously, it simply does not change the tone color, *continual* soft background hiss, or presence.
We are in exact agreement over the blanket that Rose throws on Mercurys and Deccas and other such wonderful, vivid records. It's a false sound that no longer has the immediacy and realism, while at the same time lacking in the true transparency and depth of a good modern 24 to 32 bit classical master. What you end up with is lukewarm dishwater...and what I do with used dishwater is pull the plug and let it run down the drain!
Bertie
One final thing. I have lamented the numerous uploaders who think they "improve" the sound. Usually they are worse than, say, Rose! All they do, in my estimation, is cut the highs and then (stupidly) add more filtering in the softer parts with a "denoising" algorithm. Talk about self-defeating! You end up taking a recording that, in its time, was state-of-the-art high fidelity (such as Beecham's astounding Fair Maid of Perth suite from 1934) and turning it into indistinct mush.
ReplyDeleteNow, tell me: WHO would be the intended audience for such a recording? Why: none other than OLD FOLKS! Those of us who have a whole lifetime of experience with records, back to the days (at least in my case) when I owned ONLY a 78 rpm player! Unless you utterly RIPPED a slice across a disk by some horrible blunder, you did not worry about tiny clicks or shellac hissing; you just enjoyed the music.
The neurosis about those faint noises was starting to appear by the early sixties when analogue variable-threshold filters became very commmonplace. It got even worse when the all-time most obnoxious audio device in history was introduced: the Burwen Dynamic Noise Filter (yucch!) It actually started dumping the highs at 500 Hz! It sounds like you have stuffed a pound of cotton in your mouth and then try to articulate "Suzy sells sea shells by the sea shore". What comes out is: Ozzy ell ee ell eye da ee-ore. I'm not kidding! This was then emulated in the earliest digital sound editors by clueless programmers, who wrote NR applets that caused the same amount of damage (countless old time radio shows are uploaded on the net with this damage. Often you can only get one or two words of an entire sentence! Does it make sense--when the listeners ALREADY, due to age, have diminished HF sensitivity?)
So, once again, Frank: you and I absolutely AGREE. We both hate "filtered" sound, especially FILTERED SOFT PARTS that completely lose definition.
Acoustical "air" is buried in the constant, innocuous background noise of analogue tape, and the somewhat more apparent substrate of the wax and matrix effects. These cretins are oblivious to this necessary "air" and think that if they remove the crackling and hissing, they have "improved" the recording. NOT SO! The best commercial transfer engineers in the business have learned to LEAVE THAT AIR ALONE.
Bertie
I'm a bit too busy to reply at length - but did wonder if 'Bertie' is Mr.Waldee ???? (one 'hint' being 2 mentions of the Beecham/2LP WRC French set [and, specifically, 'Jolie Fille' mentioned on the 'Sound-Fountain site'...] - which I have - also most of those 78's, new..). So apologies for being a Gum-Shoe...
DeleteBroadly, I cannot do '30% Noise-reduction', as even at initial settings the sound becomes debased - in the manner I described: ie: as being heard via Dell Inspiron 6400 laptop/s feeding into Sony active SRS-58 desktop 'editing' speakers (which were bought Sept.2011 being mains powered - which my SRS-55's were not: & both a decade+ old) - and played back @ HIGH Audacity level I am removing damn-near every 'anything' - so, via my HiFi systems, I would not expect to hear any 'defects'.
This, however, cannot apply to headphone listening: so LP's are not at all suitable for them - unless you engage in very substantial Noise Reduction.
If there are 'problems' with an LP (such as various tape defects in the Dart/James I brass) then I will mention that: just as I would not remove from a new 6-Eye Sibelius 4/5 Ormandy the various tape-noises - just to make it 'match' Rose's (German Philips pressing, apparently) 'sound' - as I showed his transfer was defective a few years back..
So I don't see a middle-path to this: hence I may as well stop - as others are doing/have done - especially as it really isn't 'justified' - other than dissuading people not to get 'ripped-off' by the unscrupulous.