Q Someone told me that I should spend 10% of my A/V budget on cables. I have about $10,000, so should I actually look for $1,000 cables for this system?
A If
you go into a store and tell them you have $10,000 for you system,
and that $1,000 of it is available for cables, I practically guarantee
you that the checkout counter will be ringing up $10,000, including
$1,000 worth of cables. It is always best to go into the store
with some idea of where you want to start. Look around at the
available components and begin by asking questions about the lower
priced items. Work your way up from there. Lots of good cables
are low priced. Kimber PBJ interconnects are very good, and very
inexpensive. Nordost 2-Flat speaker cable is great for home theater,
and it is just a couple bucks per foot. Tandy Gold Patch interconnects
are good too, and are available just about everywhere. Set the
$10,000 as your maximum, but don't force yourself to spend it
all. You could actually end up with something that you like less
than a lower priced system.
Q What
are the differences between THX certified speakers and audiophile
type speakers? By differences I am talking about specifications,
design, and in general how they sound. What recommendations do
you have for systems designed for both music and theater?
A THX
certification means that the component, whether it is a processor,
amplifier, speaker, or whatever, has been tested by an independent
company (THX) to see that it meets certain specifications. These
specs include noise levels, maximum output levels, headroom, and
many others. It's a seal of approval, rather than just a cool
logo. In general, THX products are designed for playing loud,
but clean. For speakers, sometimes this will result in the speakers
sounding better when played loud than when played softly, due
to tightness of the drivers. If you like action movies, then get
either small THX speakers, or large speakers that are not THX
certified. This way, the small speakers will give you the volume
you like, but the large ones will too, even though they are not
THX. I have a set of small THX speakers in our Seattle office,
and large non-THX speakers in our Redwood City office. If I had
to choose between the two for one system, I would prefer large
speakers, whether they are THX or not. I like having all channels
capable of achieving low frequencies (about 30 Hz), now that DD
is routine. THX-Certified speakers have limited vertical dispersion,
while maintaining a good lateral dispersion, since they are designed
for reaching a larger audience that might be spread out in the
listening room. As a result, they sound a little different than
non-THX speakers, so you should listen to them before buying.
You may or may not like the different dispersion pattern.
Q For the last couple weeks I have been demoing a Martin-Logan system for my home. This system consists of reQuests for the front, Cinema center, and SL3 surrounds. The system has excellent reproduction in the mids and highs, but it lacks in the low end. To remedy this I demoed the Paradigm Servo-15, Velodyne FSR-15, B&W ASW3000, and the Rel Q100E subs. The Paradigm lacked the "kick" needed for rock music, the Velodyne was bland in the 20 Hz - 50 Hz range, the B&W had a nice low end but sounded a bit one note-ish down there, and the Q100E lacked kick like the Paradigm. When the subs were run with the low pass set to 90 Hz, at a certain frequency they started getting boomy no matter what listening room they were in. This disheartned me and I moved on to a Linn AV-51 system (four AV-5140s and one AV-5120 center) because of their excellent bass response. However the Linns weren't as well defined in the mid-range as the Martin-Logan system. So now I'm back to the Martin-Logans. What do you recommend I do about the subwoofer issue? Should I run them with a low pass set to 50 Hz and below because that is where they all sounded precise and not boomy? What crossover frequency would you suggest for accurate bass reproduction in the ML system? Any brand/model recommendations?
Secondly, I have a Yamaha DSP-A1 that I bought a couple months back. Would the DSP-A1 be able to power the ML system above? I have seen the DSP-A1 power a system with SL3 fronts, Cinema center, and Stylos rears. If I need to get a power amp, what would be a good/powerful/precise amp to power the system I mentioned above? At a minimum, I'd want to power the reQuests with an outboard power amp, if necessary that is. Remember, the reQuests can drop to 1 ohm at 20 kHz.
My eventual goal is have a professionally
designed listening room in a new home I may get built. Because
of this, I don't want to buy anything I'm
going to throw away. In other words, I want quality stuff.
A Martin
Logan electrostatics are fine speakers, but like all electrostatics,
they are power hungry. First, I would suggest getting the same
model of MLs in the front and the back. This will give you a better
matched tonality with digital surround formats. Secondly, most
subwoofers sound boomy up near 90 Hz, because they don't use damping
material in the enclosure. Set the low pass to 50 Hz - 60 Hz and
that will pretty much eliminate the problem. Lastly, the DSP-A1
is a wonderful product, but it won't handle the needs of the Martin
Logans. I would suggest outboard amplifiers with at least 200
watts per channel and capable of dealing with low impedance loads.
The Sunfire Cinema Grand or CinePro both would do the job nicely.
Q In the review of Yamaha's DSP-A1, new writer Scott Evans says "...the more samples per second, the closer you can get to recreating the exact input waveform." In general, this is true, but the Law of Diminishing Returns rears its ugly head -- since humans can only hear frequencies up to about 22 kHz, a sampling rate of 44 kHz is theoretically all you need to recreate the input waveform exactly, 44 kHz being what is known as the Nyquist rate. It is hardly coincidental, then, that CDs sample at 44 kHz.
The operative word, though, is theoretically.
A waveform that contains frequency information throughout the
audible spectrum (that is, from
20 Hz - 22 kHz), when sampled at 44 kHz, will need a D/A converter
that is incredibly accurate, or risk additional information which
was not present in the original waveform. This is part of the
reason why you mentioned in a show report that an inexpensive
DVD transport/DAC at 24/96 kHz sounded as good as their astronomically
priced CD player equivalents at 16/44kHz -- while the expensive
D/As may be extraordinarily accurate, they carry a matching price
tag, but a cheap D/A at 96kHz can afford to be less accurate and
still not change the input waveform.
The point is that, say, 196 kHz would be
overkill. D/As would have to be just as accurate as they are at
96 kHz. This makes Mr. Evan's statement
untrue except in a very small frame of reference.
All this aside, though, thank you for the
excellent review on the DSP-A1. I have been waiting for some time
to see your well-considered opinions on this piece.
A Scott
Evans responds:
The Nyquist theorem was not developed for
digital applications (contrary to what many people think). Nyquist's
theorem deals with
bandlimited signals, modulation, stability, Fourier transforms,
s-domain, etc. The theorem is used in everything from modulation
concepts to analog control system design.
One of the basic principles that I think
you are getting at, is that the Nyquist theorem as applied to
digital audio, requires a perfectly bandlimited system. In theory,
audio can be lowpassed to say 22 kHz, the upper limits of human
hearing, but in reality it is almost impossible to achieve
this. Noise and other information that leaks into the signal (even
at frequencies outside the target band) will get aliased back
into the target
band, resulting in distortion and other nasties. When we try and
create these ideal filters we normally "mess up" the
analog signal in the process
(phase errors are a big problem with passive filters).
I did a net search to find some supporting
docs and found http://www.amtechdisc.com/file/amtech/CDPAPER.HTML
which does an excellent
job of explaining the basics.
Q On
DVDs, is 2 channel stereo downmixed from the 5.1 as it is played,
or is there a separate 2 channel PCM track for stereo playback?
A It
depends on the disc. Most of the Columbia Tristar titles have
audio 1 using a 2 channel PCM track and audio 2 using the 5.1
track. So if you are using the analog outputs and choose audio
1, you get real 2 channel PCM, and if you choose audio 2 you get
the 5.1 downmixed to 2 channel. Other DVDs do not include a true
2 channel PCM, so they are always downmixing. In general, the
discs have a DD track that delivers the 2 channel stereo for use
with Pro Logic receivers. Video Essentials does not have 2 channel
PCM.
.
Q I have a question concerning hooking up a subwoofer with full range speakers and an audio/video processor. I have an M&K MX-5000THX MkII sub and a pair of PSB Stratus Gold-i speakers and a B&K AVP-3090 processor. I was using a pair of M&K main speakers but decided to go to a more musical non-THX system.
I have the sub hooked up to the sub output of the 3090 which sends all frequencies below 80 Hz to the sub. I have the front speakers set to the large setting which sends all frequencies to the front speakers and the system was calibrated using a sound pressure meter. The bass response looks like this in the stereo mode with a Stereophile test CD and a Radio Shack sound pressure meter:
Frequency | Golds with Subwoofer | Golds without Subwoofer |
200 Hz | 77 dB | 77 dB |
160 Hz | 77 dB | 77 dB |
125 Hz | 80 dB | 80 dB |
100 Hz | 80 dB | 78 dB |
80 Hz | 79 dB | 76 dB |
63 Hz | 84 dB | 82 dB |
50 Hz | 74 dB | 71 dB |
40 Hz | 77 dB | 67 dB |
31.5 Hz | 88 dB | 79 dB |
25 Hz | 82 dB | 71 dB |
20 Hz | 77 dB | 60 dB |
I thought if I hooked up the sub in this
way I would have a much broader peak in the mid to lower bass
region but I think this response looks OK or am I interpreting
the results wrong? The bass still seems a little loud on certain
recordings using the B&K X-over, but if I use the M&K
X-over I will be using two X-overs in series and this is a bad
idea, right? Also, the M&K sub does not provide for a speaker
line in/out hook up. Besides isn't bass response very dependent
on the room? It seems like there is a peak at 63 Hz of +7 dB and
at 31.5 Hz of +11 dB as compared to the 77 dB baseline set at
200 and 160 Hz. Could this be a function of the room? I don't
think that using an X-Over set at a lower frequency than 80 Hz
would help eliminate the peak at 31.5 Hz. Is it a coincidence
that the peaks are multiples of each other 63 vs. 31.5? Could
this be a resonance frequency of the room? Actually, I think this
bass response is OK and I should leave the X-Overs set the way
they are. I just might need to turn the sub volume down as the
Golds break in. What do you think? Does this look like a good
bass response or should I try to hook up the sub a different way,
i.e., via a second set of pre-outs from the 3090 which send a
full range signal to the sub and use the sub crossover? I don't
want to buy extra cables if they are not needed and in order to
hook up the sub this way I would have to purchase another set
of 3 meter interconnects.
A The
peaks at 31.5 Hz and 63 Hz are indeed, room effects. Changing
your crossover won't modify this. If the peaks really bother you,
put a line-level equalizer, such as Audio Control makes, between
the subwoofer output on your 3090 and the line-level input on
your subwoofer. You can use the 31.5 Hz and 63 Hz sliders to reduce
the peaks at these frequencies. Employing the crossover of the
processor as well as the crossover of the subwoofer is not the
best idea, but it is unavoidable in most situations. Using the
pre-outs for the mains to go to the subwoofer will work, as long
as you have the bass from the center and rear sent to the mains,
which is an option on modern processors. Whether this change results
in audible differences is hard to say. Distortion in bass is harder
to detect than in midrange and high frequencies.
Q With
the new HDTVs, what will the scanning rates be for the two principal
formats?
A 1080i
scans at 32 Khz, and 720p scans at 45 kHz.
Q Does
it take time for preamps and power amps to "break-in"
and if so how long does it take and why?
A Most
components do take time to break-in. No one is really sure why,
but it probably has to do with the chemical structure of the silicon/doping
material of transistors and dielectrics of capacitors changing
as they are subjected to electrical current. When they settle
down, the component is broken-in. The time varies from several
hours to several hundred hours. The unbroken-in sound is usually
one of harshness. Speakers need to be broken-in too, and the unbroken-in
sound is usually a weak bass (woofer is tight) and a harsh treble.