Separate components vs. integrated?

[CR250]

I think at least part of esldude's point was that RIAA curves built into phono preamps are known to be inaccurrate, so by using them you are coloring your music- in essence you are adding a tone control which isn't supposed to be there.BTW, with computers you can have a digital RIAA curve which is much more accurate and true to RIAA standard than analog (HW) curves; a known technique for quality vinyl ripping is to rip flat (not thru a phono pre with RIAA in play) and then playback with a digital RIAA curve added in by software. Many feel this gives a superior result. DSP can be many things, it's not a synonym for Room Correction. RC is one use of DSP. In digital playback, whatever you are using it for, it probably gives a more accurate result than trying the equivalent adjustment in analog.

For the first part, that's not true. When a record is made, its run through the RIAA EQ curve. Its done because the lower the frequency, the wider the groove in the record needs to be. At a certain point, the groove gets wide enough so that needle won't be able to touch everything in it. The process isn't about coloring the sound, its about functionality. The EQ curve is needed so the format works as intended. For the rest of it, it is possible to handle this digitally, but it really only makes sense if you're ripping vinyl. If you want to listen to a record, converting the signal to digital defeats the purpose of having it in the first place. Many people make this same argument with DSD that is converted to PCM.

[wgscott]

Let's return to the topic, and if you want to debate the merits of room correction, we can do that in a separate thread in the room correction subforum, please.

 

Thanks!

[esldude]

Hopefully I won't invoke Bill's ire with one more post on the EQ business.

 

Yes, my point was some people eschew EQ yet insist on listening via a medium that requires it be done twice.

 

Yes a good way to rip albums is into a microphone pre-amp then digitizing the result. Then using digital RIAA to get the correct EQ in the final file. You will need to pay attention to impedance values however.

 

Also Redbook had a rarely used provision for preamphasis of the high frequencies to get them out of the noise floor and to avoid low level linearity issues of early DACs. Then demphasis upon playback. You effectively had 18 bit dynamic range that way with a 16 bit medium (at least at higher frequencies). Early on this was done with analog components not unlike RIAA networks. 15 usec/50 usec profile.

 

Apologies for polluting Bill's thread.

[bluesman]

Redbook had a rarely used provision for preamphasis of the high frequencies to get them out of the noise floor and to avoid low level linearity issues of early DACs. Then demphasis upon playback. You effectively had 18 bit dynamic range that way with a 16 bit medium (at least at higher frequencies).

dbx lives!

[gmgraves]

George ... Don't think anyone replied to his question so I will.

 

Naim build a variety of pre-amps which can be powered by a supply in the power amps. These sound good (at least to my ears) however the pre-amp can be upgraded by adding better external power supplies, which raises the overall sound level.

 

Also Naim's sources (CD players, streamers and DACs) can mostly be upgraded with an external supply which lifts the sound quality.

 

As I recall Bryston and Densen also have pre-amps with a basic power supply which can be upgraded with an external supply. I'm sure there are others too but can't think of them off top of my head.

 

 

Eloise. there is no doubt that many components can be improved by improving the power supply. It used to be fairly normal for a company who built power amps and pre-amps to power the latter with the former. MacIntosh did that as did Marantz and even Dynaco (and at one point, Quad). Usually the power amp would have what looked like an octal tube socket on the side of the chassis somewhere, and an umbilical cable was terminated on each end with a plug that liked like the base of an octal tube. This cable carried B+ and filament voltage from the power amp to the pre-amp. Sometimes the power was further "refined" within the pre-amp chassis, but more often than not, it was just the same as the power in the power amp. Many years ago, an older friend of mine had a Heathkit "Williamson" amplifier (mono) and a matching Heathkit pre-amp. The Williamson amp powered the preamp. My friend built his own very robust and well filtered separate power supply for his pre-amp, severing it from the power amp and a definite improvement in sound was noticed - especially in the bass and noise when the pre-amp's volume was advanced. The Williamson amp had a fairly primitive power supply, which was fine for a low voltage gain device such as a power amp (merely a full-wave rectifier tube (5U4 if memory serves) and a simple pi filter with a choke for the B+ and an AC supply off the power transformer for the filaments). My friend's supply that he built for the pre-amp had tube regulators for the B+ in addition to another stage of pi filtering plus a rectified, filtered 12 volts for DC on the filaments of the pre-amp.

 

So yeah, improving the power supply always helps. Unfortunately, most modern components offer no facility for doing so. Looks like Naim (not well distributed or even well known here in the US) is an exception.

 

[ATTACH=CONFIG]23481[/ATTACH] [ATTACH=CONFIG]23482[/ATTACH]

 

My friend's old Heathkit mono preamp showing the umbilical cord with the octal plug on the end. Next to it is the Heath Williamson amplifier, notice the tube-socket like connector next to the speaker strip on the back. This where the the preamp docks to "share" the amp's power supply.

[gmgraves]

dbx lives!

 

 

Funny, you should mention that! I was wondering just the other day if a DBX unit might be able (if used judiciously) to reverse some of the damage wrought by the "loudness wars" compression employed in many of these CD reissues that I keep hearing people complaining about. I'll bet it could help. The problem would be matching the attack and release of the studio compression used to master these reissues, so I doubt if the DBX expansion would be perfect. Ut might be close enough to do some good, though.

[sandyk]

Funny, you should mention that! I was wondering just the other day if a DBX unit might be able (if used judiciously) to reverse some of the damage wrought by the "loudness wars" compression employed in many of these CD reissues that I keep hearing people complaining about. I'll bet it could help. The problem would be matching the attack and release of the studio compression used to master these reissues, so I doubt if the DBX expansion would be perfect. Ut might be close enough to do some good, though.

 

George

Nothing helps much with recent digitally compressed music files. Even See DeClip Duo Pro which used to work very well with earlier compression and limiting methods, as well as music video clips, can't do much with these files.

Yes, I have also tried using varying degrees of Expansion in Sound Forge 9 .

 

Regards

Alex

[gmgraves]

George

Nothing helps much with recent digitally compressed music files. Even See DeClip Duo Pro which used to work very well with earlier compression and limiting methods, as well as music video clips, can't do much with these files.

Yes, I have also tried using varying degrees of Expansion in Sound Forge 9 .

 

Regards

Alex

 

 

Thanks. It figures. Even if the compression was analog, getting all the parameters to line up for perfect reciprocal expansion would be a hit-or-miss affair at best. Good thing that for the kind of music that I collect the record companies don't do that to reissues.

[DigiPete]

What is the logic behind having separate components, apart from the ability to upgrade with modularity?

 

It seems like interconnects are a significant source of concern, a problem that doesn't really arise in a unit that has a DAC and a pre and an amp.

 

 

History

 

Separates used to be the only way to an audiophile experience.

At least that was how it seemed at the time.

 

Sound is such an elusive thing and the physics so complicated, that it's easy to just follow the herd or one of the high priests.

I admit: I used to be a linear analogue TT purist. It was the only was I found to audio bliss.

CD's had low noise, but they were evil and hurt my ears ;-(

 

But technology and markets has evolved and it turns out CD's were not evil at all!

The same goes for integrated audio equipment and particularly active monitors.

But audiophiles are mere humans and most stick with what they know, even if they allow a DAC to sneak into their pile of separates.

So we are left with an industry that caters to the usual crowd and refuses to evolve.

 

There's your logic!

 

An alien would see this scenario:

The high priests has taught the (audiophile) herd that they are especially chosen, and that they shall go forth and mix and match equipment chosen by the priests, and all will be good.

 

 

What would be logic?

 

I say we should insist on trimming the fat and always remove losses and waste when possible.

The best deal is removing all the weak links, and what remains must be the solution.

So where is the waste (of signal quality & materials)?

 

 

Waste.

 

- Audiophile branding

- 'House sound'

- Distorted and noisy input signal (analogue sources)

- Mis-matched source and pre-amp

- Mis-matched pre-amp and amp

- Mis-matched amp and speakers

- Passive X-overs

- Mis-matched speakers and room

- Square speaker cabinets (diffraction and resonances)

- Misaligned midt and high speaker drivers (not concentric)

- Less than full frequency response (20-20kHz)

- Analogue interconnects

 

 

What remains must be the solution?

What should we aspire to?

 

 

Great strategies

 

- Professional equipment (not amature/audiophile)

- Professional equipment - again ;-)

- Digital inputs only

- Factory matched source and pre-amp

- Factory matched pre-amp and amp

- Factory matched amp and speaker drivers (active = 1 amp -> one driver)

- Digital or analogue active X-overs

- DSP room correction or alternative matching speakers and room

- Room chosen, treated and dampened

- Rounded speaker cabinets

- Concentric midt and high speaker drivers

- Full frequency response (20-20kHz)

- Digital interconnects only (wired or not)

 

 

Proof

 

Equipment that punch way above it's price tag:

Following most of the above points

(Increasing price)

 

- Genelec 8330 / 7350 sub

- Dynaudio Air 6 / Air Base-1

- Lyngdorf TDAI-2170 / DP-1 / BW-1

- Genelec 8351

- Genelec 8260

- Steinway Lyngdorf Model S

- Meridian DSP8000

- Steinway Lyngdorf Model D

 

OK, so the lower two defies all financial logic unless you are an Abu Dhabi sheik.

But they are examples of sound quality unreachable by usual stacks of ultra high-end separate components.

 

 

QED

 

 

 

Lindberg Lyd / 2L for great digital recordings of great music up to 9.1 and 24/192

[4est]

Your links don't work, but I assume you are referring to the W4M or W5M amps. Gosh that brings me back... I remember using those types of sockets and/or the rectifier tube sockets to repower pre amps and amps using monster Lambda, HP or Kepco tube based regulated power supplies. It is AMAZING what can happen when you refine/augment say the driver circuits of a power amp. I think I still have a 1000# worth of those regulators around here.

 

 

Eloise. there is no doubt that many components can be improved by improving the power supply. It used to be fairly normal for a company who built power amps and pre-amps to power the latter with the former. MacIntosh did that as did Marantz and even Dynaco (and at one point, Quad). Usually the power amp would have what looked like an octal tube socket on the side of the chassis somewhere, and an umbilical cable was terminated on each end with a plug that liked like the base of an octal tube. This cable carried B+ and filament voltage from the power amp to the pre-amp. Sometimes the power was further "refined" within the pre-amp chassis, but more often than not, it was just the same as the power in the power amp. Many years ago, an older friend of mine had a Heathkit "Williamson" amplifier (mono) and a matching Heathkit pre-amp. The Williamson amp powered the preamp. My friend built his own very robust and well filtered separate power supply for his pre-amp, severing it from the power amp and a definite improvement in sound was noticed - especially in the bass and noise when the pre-amp's volume was advanced. The Williamson amp had a fairly primitive power supply, which was fine for a low voltage gain device such as a power amp (merely a full-wave rectifier tube (5U4 if memory serves) and a simple pi filter with a choke for the B+ and an AC supply off the power transformer for the filaments). My friend's supply that he built for the pre-amp had tube regulators for the B+ in addition to another stage of pi filtering plus a rectified, filtered 12 volts for DC on the filaments of the pre-amp.

 

So yeah, improving the power supply always helps. Unfortunately, most modern components offer no facility for doing so. Looks like Naim (not well distributed or even well known here in the US) is an exception.

 

[ATTACH=CONFIG]23481[/ATTACH] [ATTACH=CONFIG]23482[/ATTACH]

 

My friend's old Heathkit mono preamp showing the umbilical cord with the octal plug on the end. Next to it is the Heath Williamson amplifier, notice the tube-socket like connector next to the speaker strip on the back. This where the the preamp docks to "share" the amp's power supply.

[CR250]

"Proof

 

Equipment that punch way above it's price tag:

Following most of the above points

(Increasing price)

 

- Genelec 8330 / 7350 sub

- Dynaudio Air 6 / Air Base-1

- Lyngdorf TDAI-2170 / DP-1 / BW-1

- Genelec 8351

- Genelec 8260

- Steinway Lyngdorf Model S

- Meridian DSP8000

- Steinway Lyngdorf Model D

 

OK, so the lower two defies all financial logic unless you are an Abu Dhabi sheik.

But they are examples of sound quality unreachable by usual stacks of ultra high-end separate components."

 

Of the speakers you list, have you ever heard the Meridians 8000? They're the only ones I've owned myself, so I can't compare them to some of your other picks. I can tell you that, aside from playing extremely loud, I can't give you any other qualities that set them apart from any other speakers, or systems I've had, at a fraction of their selling price. When you go with a full Meridian system, you throw a lot of money in the garbage to achieve, at best, mid-fi sound. If you happen to own them and are happy, please don't take offense. If they work for you, that's all that matters. You made the right choice. But with the experience I had, I can't in good conscious recommend them.

[DigiPete]

Of the speakers you list, have you ever heard the Meridians 8000? They're the only ones I've owned myself, so I can't compare them to some of your other picks. I can tell you that, aside from playing extremely loud, I can't give you any other qualities that set them apart from any other speakers, or systems I've had, at a fraction of their selling price. When you go with a full Meridian system, you throw a lot of money in the garbage to achieve, at best, mid-fi sound. If you happen to own them and are happy, please don't take offense. If they work for you, that's all that matters. You made the right choice. But with the experience I had, I can't in good conscious recommend them.

 

Are you sure, that you prefer a linear frequency response???

I have not heard the largest Meridian, but they have a good topology and the smaller ones sound promising.

 

I have not heard the Steinway Lyngdorf Model D either, they are way too expensive for my wallet and too large for my habitat.

They did fool the experts at Steinway into thinking that they were listening to real Steinway concert pianos Model D.

That's a deed not easily done, and the smaller ones are extremely transparent IMO.

 

My daily rig is a surround setup based on the Genelec 8260's driven by a Weiss AFI-1, and I have demoed my rig at other venues with great results.

[gmgraves]

Your links don't work, but I assume you are referring to the W4M or W5M amps. Gosh that brings me back... I remember using those types of sockets and/or the rectifier tube sockets to repower pre amps and amps using monster Lambda, HP or Kepco tube based regulated power supplies. It is AMAZING what can happen when you refine/augment say the driver circuits of a power amp. I think I still have a 1000# worth of those regulators around here.

 

Sorry about that. They weren't links, though, they were attachments -JPEGs, in fact! I've never had them do that before, where they don't show up as pictures. I have no idea why. Perhaps we'll have better luck with these:

 

The preamp was a WA-P2

 

 

And the Power amp was a W6M

 

[CR250]

Are you sure, that you prefer a linear frequency response???

I have not heard the largest Meridian, but they have a good topology and the smaller ones sound promising.

 

I have not heard the Steinway Lyngdorf Model D either, they are way too expensive for my wallet and too large for my habitat.

They did fool the experts at Steinway into thinking that they were listening to real Steinway concert pianos Model D.

That's a deed not easily done, and the smaller ones are extremely transparent IMO.

 

My daily rig is a surround setup based on the Genelec 8260's driven by a Weiss AFI-1, and I have demoed my rig at other venues with great results.

 

I find your reply a bit disappointing. When you make statements and give examples like this, "But they are examples of sound quality unreachable by usual stacks of ultra high-end separate components.", as proof to the point you are trying to make, I think you would need to listen to them at the very least. Otherwise, you're just assuming and guessing. You're doing the same thing in your last quote when you ask me if I'm sure that I prefer a linear frequency response. I don't see how you could take the Meridian's superiority in this area as a given. Especially if you don't know anything about the other components I used for the comparison. Not only that, the vast majority of speakers are measured in an anechoic chamber. There's no telling how good the FR will be in actual setups because every room is different. Either way, you don't need to buy a 50k pair of Meridian speakers to achieve a linear response.

 

If you're wondering as to what I didn't like about the 8000's, the main issue was timbre. For example, cymbals didn't sound like cymbals. It was more like what you would here if you were to drop the cymbal on a concrete floor. Completely artificial sounding. I know it wasn't the recordings fault because I played the same albums on everything I listened to. Also, another thing that I didn't like at all, was the waste of money that gets spent on features you can't use if you go with an all Meridian system. It should be the opposite.

[tranz]

Are you sure, that you prefer a linear frequency response???

I have not heard the largest Meridian, but they have a good topology and the smaller ones sound promising.

 

I have not heard the Steinway Lyngdorf Model D either, they are way too expensive for my wallet and too large for my habitat.

They did fool the experts at Steinway into thinking that they were listening to real Steinway concert pianos Model D.

That's a deed not easily done, and the smaller ones are extremely transparent IMO.

 

My daily rig is a surround setup based on the Genelec 8260's driven by a Weiss AFI-1, and I have demoed my rig at other venues with great results.

 

....so....separates are still the way to go?

[DigiPete]

....so....separates are still the way to go?

 

Nope, to the contrary

 

 

The Genelecs are active and digital input (AES3) with full room compensation.

The signal chain is:

 

Computer -> firewire -> Weiss AFI-1 -> AES3 cable -> Monitor

 

Genelec uses a secondary hard wired network to control volume and room compensation DSP.

 

The Genelec 8260 is an audio jewel, check it out.

It is so invisible, and the image so clear, it still give me the shivers after 4 years of ownership.

[tranz]

Ah, I see now. I have always liked the Genelecs but never listened to those with DACs in them. Thanks.

[gmgraves]

For the first part, that's not true. When a record is made, its run through the RIAA EQ curve. Its done because the lower the frequency, the wider the groove in the record needs to be. At a certain point, the groove gets wide enough so that needle won't be able to touch everything in it. The process isn't about coloring the sound, its about functionality. The EQ curve is needed so the format works as intended. For the rest of it, it is possible to handle this digitally, but it really only makes sense if you're ripping vinyl. If you want to listen to a record, converting the signal to digital defeats the purpose of having it in the first place. Many people make this same argument with DSD that is converted to PCM.

 

That's not exatly right. The reality has to do with a combination of factors, some related to the physical reality of "cutting" a disc and others have to do with the actual physical properties of vinyl itself. The RIAA curve does two things: One, it attenuates the bass during cutting and boosts it by the same amount on playback, thus restoring it to the level that was on the master tape before the cutting process. Secondly, the RIAA curve boosts the treble on cutting and attenuates it by the same amount on playback. The reason for this level of manipulation is because, on the bass end, the stylus deflection caused by a heavy bass line can actually cause the cutting stylus to break through the groove wall into an adjacent groove, thus ruining the master. Even short of that, too great a bass excursion can yield a groove that is untrackable by your average playback rig.

 

In the treble region, almost just the opposite can be true. High frequencies have little energy when compared with midrange frequencies. If cut "flat" treble frequencies would be so small that many styli would simply skip over them, while many more would wear them away in just a few playings. Boosting the highs during cutting results in a much more robust treble "groove" that can be more easily traced by an ordinary stylus without being worn away in just a few plays. Also, it turns out that there is an added advantage to boosting the highs on cutting the disc and reducing them to their original level on playback. Turns out that vinyl, even so-called "virgin" vinyl, isn't all that quiet. A s/n ratio in the mid 50dB range is about all one can expect. Vinyl noise is pretty constant, and is worse at high frequencies than it is at low frequencies. So by increasing the level of the highs when cutting the disc, and then attenuating the highs by the same amount on playback, one also attenuates the high frequency vinyl noise already on the pressed record by the same amount thus giving a not insignificant modicum of noise reduction, improving the s/n of the record significantly!

[miguelito]

What is the logic behind having separate components, apart from the ability to upgrade with modularity?

 

It seems like interconnects are a significant source of concern, a problem that doesn't really arise in a unit that has a DAC and a pre and an amp.

 

(For present purposes, let's consider powered/active speakers as an extreme case of integrated.)

It depends... Generally I would say separates have the ability to optimize power supplies to the task (eg preamp requires very low noise, not much power/current delivery, power amps the opposite). Having said that there are some very nice integrateds out there (I use an integrated).

[bluesman]

Funny, you should mention that! I was wondering just the other day if a DBX unit might be able (if used judiciously) to reverse some of the damage wrought by the "loudness wars" compression employed in many of these CD reissues that I keep hearing people complaining about. I'll bet it could help. The problem would be matching the attack and release of the studio compression used to master these reissues, so I doubt if the DBX expansion would be perfect. Ut might be close enough to do some good, though.

 

As I recall, the dbx 180 had trim pots for several parameters, so you could play with the compander curves. There may have been other adjustable units as well. I bought a 122 to try with my high speed Crown 724, but it had no adjustments that I remember - and it added very little (if anything) to the quality of my live recordings, so I used it very little and finally sold it.

[gmgraves]

As I recall, the dbx 180 had trim pots for several parameters, so you could play with the compander curves. There may have been other adjustable units as well. I bought a 122 to try with my high speed Crown 724, but it had no adjustments that I remember - and it added very little (if anything) to the quality of my live recordings, so I used it very little and finally sold it.

 

Yes it did, but unless you knew the release and attack parameters of the compressor used by the record company, I'm not sure that you could match them by ear. Plus as Alex reminded me, the record company compressors are digital, not analog, these days and done with DSP technology and therefore the compression parameters are likely not set, but vary with content. A DBX 180, used as an expander, most likely would be of no use at all.

 

I had a DBX unit (actually, it was my recording partner's and I don't recall the model number) that we used with my Sony 850, early on when we were recording the San Jose (CA) Symphony. I too found it less than satisfactory (it "pumped" audibly). So, I bought a TEAC Dolby B unit (the one with the chrome fascia and the huge Vu meters). I used that for a full season and when a used Dolby A unit came up for sale at a local music/recording equipment store, I jumped at it and used that through the rest of the time that I did analog recording. I started with a Sony 850 and a DBX box and ended with an Otari MX5050 and Dolby A.

[CR250]

That's not exatly right. The reality has to do with a combination of factors, some related to the physical reality of "cutting" a disc and others have to do with the actual physical properties of vinyl itself. The RIAA curve does two things: One, it attenuates the bass during cutting and boosts it by the same amount on playback, thus restoring it to the level that was on the master tape before the cutting process. Secondly, the RIAA curve boosts the treble on cutting and attenuates it by the same amount on playback. The reason for this level of manipulation is because, on the bass end, the stylus deflection caused by a heavy bass line can actually cause the cutting stylus to break through the groove wall into an adjacent groove, thus ruining the master. Even short of that, too great a bass excursion can yield a groove that is untrackable by your average playback rig.

 

In the treble region, almost just the opposite can be true. High frequencies have little energy when compared with midrange frequencies. If cut "flat" treble frequencies would be so small that many styli would simply skip over them, while many more would wear them away in just a few playings. Boosting the highs during cutting results in a much more robust treble "groove" that can be more easily traced by an ordinary stylus without being worn away in just a few plays. Also, it turns out that there is an added advantage to boosting the highs on cutting the disc and reducing them to their original level on playback. Turns out that vinyl, even so-called "virgin" vinyl, isn't all that quiet. A s/n ratio in the mid 50dB range is about all one can expect. Vinyl noise is pretty constant, and is worse at high frequencies than it is at low frequencies. So by increasing the level of the highs when cutting the disc, and then attenuating the highs by the same amount on playback, one also attenuates the high frequency vinyl noise already on the pressed record by the same amount thus giving a not insignificant modicum of noise reduction, improving the s/n of the record significantly!

 

I'm pretty sure we're in agreement. You do a much better job of explaining it than I do.

[gmgraves]

I'm pretty sure we're in agreement. You do a much better job of explaining it than I do.

 

 

While in college, I worked for a bit as a cutting engineer at a regional "vanity" record producer. They would press small runs of records (both mono and stereo) for civic orchestras, high-school bands, cut demo discs for aspiring performers, and occasionally release a commercial product (mostly bluegrass). So I know quite a bit about record mastering from that experience.