Rerurn to Romy the Cat's Site

Analog Playback
Topic: AMR PH-77: just another phonostage or more?

Page 1 of 1 (12 items)


Posted by Romy the Cat on 07-05-2009
fiogf49gjkf0d

The news just come in

http://www.amr-audio.co.uk/html/pr_ph77.html

Thorsten use to tell me that his ambitions phonostage project will be coming through in fall of 2008. Well, it is late but it is there: the AMR announced their new PH-77 Phono Equalizer.

http://www.amr-audio.co.uk/html/ph_individual.html

I think it shall be a very interesting phonostage and I feel kind of sad that I am off the phonostage market.  The PH-77 has provision for 3 arms, adjustable setting for each arm, 72dB gain, tunable impedance for each arm – exactly what a doctor ordered! I presume that this thing shall also sound good – Thorsten was the person who was very close to my EAR-834PT project and the letter “T” in my “End of the Life phonostage” stands from his name. He also run for a while 7788-7721 two-stages phonostage (not the Dima’s design but his own) – so he knew something how good phonostage might be.

Reading the specifications AMR PH-77 there are a few moments that I would criticize. The presence of 24 bit/96kHz Analogue to Digital converter on-board is wonderful but why the 96kHz? Only God in his intimate wisdom knows why people still use that idiotic 48x sampling rate. Like we or not but 44kHz is still the standard digital format and I presume it will be foe many years. Want higher sampling rate and more bits depth – wonderful – go 88kHz or 176kHz but do not introduce the 48kHz base. It is imposable to properly convert stream from 44X base clock to 49X base clock. Even with Pacific – the best rate converter ever was built - it is very auditable if you down-convert from 88K vs. from 96K. Also, when you get 44K you can play it on CDs (cars, work, give to friends that you hate, etc….) however, what the hell people do with 48kHz? Play is from DAWs? Sure, but why they do not do it at 96kHz. Anyhow, if people desire a higher rate then 16/44 than it must be 88kHz or 176kHz. Read my leaps – no 48X clock, ever!

Another subject of my criticism is presence of 22 different EQ curves. It is not a mistake to do it and AMR did it almost right – with real-time change via a remote control. However, I question the practical usability of this thing in dally use. If the objective is to get the best sound from each LP then I would say that the 22 different EQ curves is God sent. However, in dally life if you want to spin a few records in a row then I find that mingling with curve for each record is kind of too much for me. I am for instance do not fine-adjust VTA for each vinyl thinness. Sure, it affects sound but I would like me to driver records not the records to drive me. It is not to mention that having 22 different EQ curves do not help you to get right response but help you to get the response that you like in context of you playback.   How many people use wrong VTA just because they have bad TT and ported speakers and what to compensate the bad bass from TT by “tight-sounding” VTA. So, I would consider do not use 22 different EQ curves but to call it “EQ curves real-time adjuster” and I do not care how fun my current seething from Decca FFSS of from Columbia EQ.

Ok, do I disagree with the naming? Nope, I disagree with application. Let me to expanding. Using my EAT 834PT with variable air capacitors I did experimented with turning the caps and shaping the custom curves for some records. It worked wonderfully and I even bought the 10:1 transmissions for the turning axis to be able to fine-set the exact capacitance and to calibrate the scale. The problem was that I was forgetting to put the caps in the default EQ setting after I was playing the “off the wall” records. I found that it was more annoying that the play the “off the wall” records with the default EQ curve….

Anyhow, here is the solution that I would like AMR to go and it will be the right way to do. The PH-77 shall have a provision for an “active” AMR turntable mat, let call it eMat. The eMat would optically scan the record label and send info to PH-77. The PH-77 would automatically select a different EQ curves for given label or to pull from own memory a custom setting that you specified the last time (you would need to do some marks on label). The eMat might be driven from 2 or 4 watch-size re-chargeable batteries and the PH-77 might act as a changing base, there are other solutions to power the eMat. The point is that with the contemporary technologies it is very simple to automate the process of label recognition. The PH-77 costs around $8.5K – very sensible price if you ask me. If I was at phonostage market, like how the PH-77 sound then would I consider paying extra let say $1k-$2K for an option to have automated labels recognition. I would pay more….

The Cat

Posted by Romy the Cat on 07-05-2009
fiogf49gjkf0d

Actually I just paid attention to an interesting detail that probably would make AMR to be banned from next CES manufactures dinner. The AMR violated the most holly rule of hi-fi. They said:

“* The PH-77 is currently shipping with a reference 16 bit/48kHz Analogue to Digital converter; customers will receive a no-cost upgrade to the 24 bit/96kHz Analogue to Digital converter once it is commercially available from the semiconductor manufacturer.”

Pay attention, they do not release their unit “as is” and then in a year or so when their 24/96kHz converter become available introduce the MKII model of $1000 update. Instead they promise the “no-cost upgrade” in future. That is VERY commendable behavior, something that I never seen in hi-fi.

The Cat

Posted by Paul S on 07-05-2009
fiogf49gjkf0d

Well, I guess that about covers it, all right!

But how do they get the 12AT7 input noise down 147 dB?!?

They do mention something about using "quantum particle research", which might mean Bybee filters, or op amps, or both.

So is this thing 2-stage or (as seems more likely) 3 stage?

I am always paranoid of features and adaptability, which generally imply switching, which I distrust.

But perhaps they use relays for switching, to keep all the switches out of the signal path?

Kudos for the features, in any case, and right-on for the up-front offer to upgrade the betas gratis!

IMO, it would not have to sound any better than other top phonostages in order to be better, in practical terms.

I hope some whack-o runs it head-to-head with the Boulder 2008...

Best regards,
Paul S


Posted by Romy the Cat on 07-05-2009
fiogf49gjkf0d

Yes, I have paid attention to the ridicules noise specification.  - 147dB is ever higher then 24bit dynamic range – good luck with those numbers in tube phonostage! They use 'A' weighted measurement that is not kosher from my perspective and juts a way to manufactures to publish better marketing numbers. 'A' weighted scale is emphasizing human most sensitive zone, while rolling off high and low frequencies. It reportedly measures subjectively perceived loudness but in my view it is very not objective characteristic to talk about S/N. It is like you say that that it is 27420053 degrees of temperature outside but “forgot” to mention that in your scale the water freezes at 2711053 degrees…

Anyhow, they do report the ultra low noise accomplishment and I do feel that the noise in this phonostage might be very low. Ironically ANY manufactures I know when they make phonostages report the “ultra low noise” but the phonostage noise never bothered me. Any more or less properly build phonostage has 80dB S/N and it is absolutely enough not to worry. I believe the record in phono S/N ration was achieved by John Curl in his Vendetta phonostage where he had 96dB or something like this. I think anything after 80dB-85dB is irrelevant. The key is not the amount of noise but the harmonics of noise and how the noise talks with the normal LP noise (surface noise etc…)

I think at any given playback, within a context any specific amplification gain and speakers sensitively if you max out all you attenuators and have at max gain no noise from loudspeakers then any further worry about noise is irrelevant.  Not to mention that most playbacks the majority of analog playbacks nosier does not come from phonostages but from externals cables and cartridges. In case of AMR when they use 3 arms it might be tricky… I always prefer to have grounding flexibility and be able with given cartridge to stick grounds in different locations. I am sure the AMR guys took care about all possible grounding configuration, particularly combining the MM and MC cartridges and mono and stereo cartridges …

The  12AT7’s own noise?  Well, pay $8K and it will be an opportunity for you to learn how it might be. My only concern with the ultra low reported noise number is that they might defeat the noise intentionally. In their power amp they run feed-forward injection and they might come up with some kind intelligent back-face noise injection. It would certainly lower the noise numbers but I am not a big fun of those wide-bandwidth injections. I do not claim that they did anything like this but to operate 12AT7 at 147dB S/N (even ‘A’ weighted) … I do not know, it would be interesting to hear further comments about it ether from  AMR or from anyone who will write about it.

The Cat

Posted by Romy the Cat on 07-09-2009
fiogf49gjkf0d

Thinking further about what AMR PH-77 is trying to do some ideas come to me…

At my photography life time when we use to shot on color reversal films we included a gray scale.  I personally never was huge fun of this practice as the E6 process has some discrepancies at maximum and minimum densities and I have other ways to “read” colors. Still it was an OK way to set the color balance more or less neutral.

I wonder if it is possible to find a premise under which an RIAA corrector of the PH-77 class would read sound (obviously on digital domain) and detect the right EQ curve? Well, know the right frequencies of the right pitches and I think if we have a slow-reacting processor that would calculate the error then it can suggest a right EQ curve. Years back I was proposed the same for real time “reading” the correct speed of the TT’s platter, but I think the idea has some potential.

It is not a concrete proposal but an opportunity to think further about it.

The Cat

Posted by Paul S on 07-09-2009
fiogf49gjkf0d

This would be a fine feature, indeed, if it could really avoid second-guessing.

The key for an auto-reader/adjust would be the ability to  know if a given embedded range of frequencies is supposed to be louder or quieter.  And to be of any real value, it would have to be able to do this with old records, which is to say without any special prior encoding.

It looks like the AMR folks do offer up a list of record labels and related manufacture dates to try to connect with the correct curves.

I wonder offhand how accurate their information would prove to be, in practical terms.  If very accurate, then it should be easy enough to just set the curve with a switch, which could be on a remote, of course.

Without some kind of guidance, though, you'd go crazy trying to do it by ear!

Best regards,
Paul S


Posted by Paul S on 07-18-2009
fiogf49gjkf0d
I thought of the AMR equalization curve options as I listened to a London FFRR LP (Kubelik/VPO/New World) today using my new technique to optimize stylus tracking.

The FFRRs are well known to have some extra LF, all right, but with optimum VTA this recording was obviously not standard RIAA, so strong and freakish is the deepest "bass" with my standard-RIAA-only phono stage.

As it happens, the AMR offers up several (4) different equalization curve options just for the various London/Decca FFRR!

I hope I don't start morbidly cruising through my old LPs just to check for equalization mis-matches!

How sick would that be!?!

Paul S

Posted by Romy the Cat on 07-18-2009
fiogf49gjkf0d
 Paul S wrote:
I thought of the AMR equalization curve options as I listened to a London FFRR LP (Kubelik/VPO/New World) today using my new technique to optimize stylus tracking.

The FFRRs are well known to have some extra LF, all right, but with optimum VTA this recording was obviously not standard RIAA, so strong and freakish is the deepest "bass" with my standard-RIAA-only phono stage.

As it happens, the AMR offers up several (4) different equalization curve options just for the various London/Decca FFRR!

I hope I don't start morbidly cruising through my old LPs just to check for equalization mis-matches!

How sick would that be!?!
 
Actually I have explored this feature in context of PH-77. It looks like they did it properly: all curves might be recalled individually for each arm from a REMOTE CONTROL and each arm might be returned to its default preset curve by one single click of remote control. It sound very smart and does not sound sick at all. You put a record, you flip on your remote the curves until you get a right one and then you mark the records with info about the right curve. Sure, it would be cool to automate the process and to let the PH-77 to “hear” sound and to figure out the right curve or to scan the labels… Anyhow, since it is manageable from remote then it is not a problem…

The Cat

Posted by Paul S on 07-19-2009
fiogf49gjkf0d
What I became aware of today was the FFRR "special" curve that made the low end especially and the sound generally to be strangely "shaped", if you know what I mean.  And the sound of the wrongly de-coded curve can never really get back its proper shape by means other than proper de-coding; that would be like trying to cure frequency range limitations with "special" IC (and what kind of a dimwit would do that?).

No doubt about it, really; matching the "correct" recording curve at the playback end would be worth... about whatever it cost.

In the remote, hunh?

Next time I hit the Lottery...

Best regards,
Paul S

Posted by Paul S on 07-19-2009
fiogf49gjkf0d
Too hot to expect the best sound today, and besides I have gotten my morbid curiousity worked up about the recording/playback curves, so I fished out a few LPs mostly to hear/see how they sound with respect to the "shape" of their sound, aspects of which I now relate to and target as the result of their native curves being "subjected" to RIAA.

Gray 6-eye Columbia mono, ML5164; Rudolf Serkin, Beethoven: "Hinge" on this record is ~ 2nd C; wonderful tone otherwise, but it is pitched too high.

Decca Gold Label (from original DGG tapes), mono, DL9401; Brahms, Magelone Songs, Dietrich Fischer-Dieskau with Jorge Demus, piano: Pitched a little low; spikey part of HF is greatly attenuated; shape is not too bad; not germane here, but excellent surfaces and "traction", both (so rare to find together!).

Everest mono, SDBR 3377; Rachmaninoff plays Chopin and Rachmaninoff (with a rather lumbering, eliptical cadence!):  I have no info on this LP, but based on other Everest I have I suspect that the "canned" quality of the sound owes to original recordings. Although Everest is generally pretty scrupulous, it's hard to know about how curves shape sound here.

Vanguard Stereolab, SRV-117 SD; Moussorgsky, Pictures at an Exhibition and Night on Bald Mountain; Vienna State Opera Orchestra/Vladimir Golshman: With perfect VTA I finally "get" Mussorgsky!!!  I will certainly re-visit this LP with good electricity!  Maybe because of the "Opera Orchestra", but Mussorgsky's "patterns of speech" are so intelligible!  A GREAT recording/shape/sound, so I suppose it is dead-center RIAA all the way.  Bass fades out just as it "should" based on my system limits.  Everythng is so "clear", with no tonal or textural penalties for it.

Westmisnster (mono) XWN 18380 (1957); The Art of Clara Haskel; Mozart PC 19 & 20: Winterthur Orch., Henry Swoboda:  Great "Bloom", but then this proves to be largely due to tonal "hinge" just below MF.  "Coffee can" sound. Haskel plays with a "woman's touch" (too soft) and Swoboda accomodates her.

Supraphon DV 6106 (mono), 1964; Beethoven 9:  I can't decipher the Czech, but the sound is decidedly "tubular", to the point where Ii put off listening more at this time.

Supraphon SV 8047 (stereo!!!); Dvorak, 9th; CPO/Ancerl:  Never sounded better, with perfect VTA and obvious center of RIAA.  I wish everything I owned sounded this good!

RCA LM-1893 (early mono/New Orthophonic Shaded Dog); Ravel, Daphnes  & Chloe; BSO/Munch: Tubular sound shape.

Now I'd love to try the same records (and others, like Vox, etc., etc, etc.) using their correct playback curves, to hear/see if/how that affects  the shape  of the sound, as I predict it would/will.

AMR, how about a "review sample"???

Paul S

Posted by Romy the Cat on 07-19-2009
fiogf49gjkf0d

I am not a big fun of Dvorak 9th under Ancerl. I’m not big fun nether the orchestra’s balance nor the recording quality.  Anyhow…

I do not think that that you need to worry PH-77’s accuracy of curves or the comfort to switch them. It looks like the took care about this part. I do not think that any review would help you as well. My only concern is that in this remote control that flip the curves it looks like it will not be an absolute phase switch. One might argue that the phase switch must by on preamp but how many of us have a fully balanced preamps? I do not and I would love to have it in phonostage.

I think the biggest question about the PH-77 shall be how it sounds itself with no respect to curves. Obviously in order to accommodate the curves flexibility specific topological decision were made, the decision that probably would not be made if the corrector survey just one curve.  How it affects the sound and what the corrector own default sound – that is the key. The Boulder 2008 that has adjustable curves sound very-very nice; however Boulder it is not adjustable but has plug-in modules. As far as I know the PH-77 is the first attempt to attach the problem, at which level it is implanted sonically is not known yet. If they did it at the level of other phonostages with price tag of €8K (there are a dozen of them) then it will be good for them and I then foresee a bright future for this phonocorrector. For whatever it worth the PH-77 it has provision for 3 arm and for some people it might be God sent…

I just a bit afraid that marketing-wise to release the PH-77 for €8K was incorrect move. I instead would like to see PH-77 “Regular” and PH-77 “Universal”. The “Regular” version might have one RIAA curve and one arm and it go for €8K. The “Universal” version has a provision for 3 arms, all multi-curve functionality and would go for let say €15K. The reason I say so because I know that audio peoples are idiots and they feel that not expensive price is an evidence of sonic inferiority. I remember 10 years ago the shitty sounding Karma 2.5 was trying to put itself for $6.5K and no one wanted it. People lately trashed this piece of crap and dealers use Karma to cut fish on them in back rooms. As soon the canny dealers juts totally for fun increased the price for this crap from $6.5K to $21K then suddenly the deaf idiots begin to buy it and the industry whores and pimps of all calibers begins to hangs their stupid medals on this poor speaker. I just a bit afraid that for €8K the Moron might not take it seriously.  I think that AMR need to come up with some kind of alibi to handle this issue. If I work for AMR then I would leak some kind of BS story like this: “we can keep out price very low as the UNESKO financed International Sound Archives Foundation give to us a huge grant to develop PH-77 phonostage with objective to preserve and to advance the legacy of recording history.”

The Cat

Posted by Paul S on 07-19-2009
fiogf49gjkf0d
Points about the self-sound of the PH-77 are well taken, not that I'll be spending 8K on anything any time soon, regardless of its "value".

My own K&K phono stage is totally Spartan, just to avoid all the circuitry that typically facilitates the sorts of "options" and  "flexibility" sported by the PH-77.  The K&K does have mono and phase switching, however.

Likewise, my "pre-amp" is totally hair shirt; but it is balanced and it also has a phase switch (on its remote...).

As for BS, AMR came right out of the gate with that noise spec, so I am sure they can come up with creative spiel about pricing, etc., as well, if they need to.  I think the English Tom Evans "Super Groove" phono stage has done OK at near that price, if that is any indication, and people here seem to scoop up the Steelheads and such.

As for how valuable non-RIAA re-shaping/correction of playback might might be to me, obviously I'd have to try it.  Sure, the AMR night be a poor implementation of a good idea, or a good implementation of a poor idea.  And I am generally suspicious of gear that tries to do too much.

It is true that I just want something that fixes the weird-shaped sound, and certainly I am projecting my desire onto this object.

Like I said earlier, I hope someone runs it against the Boulder 2008.

Best regards,
Paul S

Page 1 of 1 (12 items)