Ampli gấu của DENON

Ampli gấu của DENON

Hic các bác ơi em hơi bị choáng. Vừa nãy lên Audiogon.com thấy có thằng nó bán 1 bộ gồm Rề và Pow S1 mà giá là 15.000 USD. Giá mới là 50.000 USD. Không ngờ DENON lại có sản phẩm cao cấp đến như thế mà lại chỉ bán ở NHật thôi. Chứng tỏ tại thị trường Nhật lắm đồ gấu lắm. Nhìn bộ này của nó mà muốn khóc thét. Cục Pow của nó như 2 khối sắt ý. CÒn con Rề lại cò có nguồn riêng nữa chứ. Hic hic

Bác nào biết về con này post lên cho anh em thưởng thức cái.

Có bài viết bằng tiếng anh hay hay. Mời bác xem qua.
Soundscape Articles of Interest
Premiere Designs
No matter how good-looking you are, you''re Don Knotts next to Denon''s S-1 series.
Home Theater, May 1996
Hi-fi gear has earned a reputation for being far ***ier than its fans; Denon''s SS1 series components will keep that rep alive even if Sharon Stone starts reading this magazine.
Load a CD into the DP-S1 transport, top it with their full-sized CD clamp, and batten down the hatch for a totally vibration- and resonance-free ride. Digital-to-analog conversion comes courtesy of the DA-S1 digital processor, which accepts five sources delivered via every industry standard type of digital inter connect. A fiber optic cable links both units, assuring synchronized digital clock speed and minimal jitter.
Possessing the king of all volume knobs, the PMA-S1 integrated amplifier puts out 50 watts per channel, but we''d call that conservative. It''s basically a full-featured preamp (including phono stage) mated to a beefy two channel amplifier. In this design, vibration, ever on the minds of audiophile engineers, has been vanquished&emdash;Denon bolted the PMA-51''s massive transformer to a cast chassis. Not recommended for the audiophile with a weak back. Or a skinny pocket-book&emdash;this shiny trio goes out the door for $20,200.
--------------------------------------------------------------------------------
Denon ALPHA
EUROPEAN INNOVATION OF THE YEAR
Hi-Fi Choice October, 1994
Denon''s ALPHA digital processing system is an important step Forward in the search for perfect digital sound. ALPHA is an acronym for Adaptive Line Pattern Harmonized Algorithm, which is the first processor that attempts to create an analogue signal of approaching 20-bit quality from a digital source with only 16-bit resolution.
Although many companies are experimenting with recapturing information beyond 20kHz, Denon''s ALPHA system attempts to solve the problems affecting sound quality within the audio band. The ALPHA processor''s aim is to overcome quantisation noise and improve low-level resolution. A 16-bit digital sample produces a wave-form that is made up of very fine steps, instead of smooth analogue undulations. A 20-bit sample produces a far smoother waveform, so the ALPHA system generates ad***ional data to simulate four extra bits. Thus, the ALPHA processor attempts to simulate ultrahigh quality 20-bit recording and replay, without having to change the compact disc format at all ALPHA processing also incorporates an intelligent circuit that attempts to recognize if a signal is made up of pulses or actual musical information. The ALPHA system modifies the digital stop band, in an attempt to reproduce very high frequency information while eliminating the deleterious ringing associated with conventional digital filters.
At the moment, Denon''s ALPHA system is only available in the state-of-the-art DA-S1 digital-to-analogue converter. This exquisitely built DAC is a part of the extremely rare and expensive S1 range, which also includes the elegant DP-S1 CD transport, POA-S1 power amplifiers, DL-S1 phono cartridge and matching AU-S step-up transformer. Many of these top-of-the-range components are not currently imported into the UK, but it is expected that the ALPHA technology will filter down the Denon range, eventually appearing in lower cost players as well.

--------------------------------------------------------------------------------
DENON DP-S1 CD TRANSPORT &
DA-S1 D/A CONVERTER
By: Thomas J. Norton
A REPRINT FROM MAY I994, VOL. 17 NO. 5
DP-SI: CD transport with remote control. Outputs: two S/PDIF (RCA-type) coaxial, One BNC coaxial, one XLR (balanced), one ST optical, two TosLink optical. Also, one ST-type fiber-optic input for dock synchronization with DA-S1 converter Dimensions: 17.1 " (434mm) W by 5.6" (143mm) H by 15.75" (400mm) D. Weight: 36.9 Ibs (16.7kg). Price: $8000.
DA-SI: Digital/analog converter Frequency response: 2Hz-20kHz, +0.2dB S/N ratio: 120dB. Harmonic distortion: 0.0015%. Dynamic range: 100dB. Channel separation: 120dB (1Hz). Output voltage: 2.5V RMS into 10k ohm load, balanced or unbalanced. Inputs: one ST optical, two switchable between S/PDIF (RCA-type) and TosLink optical, one BNC coaxial, and one XLR (balanced). Also S/PDIF coaxial and TosLink optical digital outputs, plus ST optical digital output for clock synchronization with DPSI transport. Analog outputs: unbalanced on RCA jacks, balanced on XLR jacks. D/A conversion and filtering: Denon''s 20 bit Advanced Super Linear Converter with Alpha processing and 8x over-sampling NPC digital filter. Dimensions: 17.1 " (434mm) W by 5.6" (141mm) H by 16.5" (419mm) D. Weight: 44.1 Ibs (20kg). Price: $7000.
Both: Approximate number of dealers: 5 actually carry the combo, but it can also be ordered through any of Denon''s 400 retailers. Manufacturer: Denon America, 222 New Road, Parsippany, NJ 07054. Tel: (201) 575-7810. Fax: (201) 808-1608.
Visit the top floor of any of the many bazaar-like electronics shops crowding Tokyo''s Aki-Habara district, and you''ll find a high-end audio department that''s loaded to the rafters with esoteric gear from around the world. Most of this equipment is the usual stuff from the US, England, and Europe. But you might be surprised to also find products from some of the largest Japanese companies, whose US-distributed products more often than not run to the rack systems, CD changers, and AV receivers found in your local stack-''em-up, move- ''em-out, audio dry-goods superstore.
Japan''s electronics manufacturers have long produced high-end audio products. But these products arc seldom seen outside Japan and a few other Far-Eastern Countries, because the market for such products is minuscule compared with that of consumer electronics at large. I would venture to guess that, in the US, a single weekend of sales of one-piece CD players vastly outstrips a whole year''s movement in separate transports and processors. Volume is the name of the game; side bets on limited-interest items just don''t pay off.
The huge markup in Japan on American-made high-end equipment, l however, makes the home-grown high-end goodies very competitive. The reverse situation&emdash;Price penalties on Japanese high-end products imported into the US&emdash;though not nearly as severe still has an impact. American high-end equipment is very popular in Japan and other points East, but it''s still out of the reach of many well-off audiophiles there The net result is loads of interesting Japanese audio gear which is sold at home but never reaches us. But every now and then a Japanese high-end audio product does make it into the US. Denon has built a strong reputation in America&emdash;it''s one of the first names to come to mind beyond the niche of specialty, high-end audio. With the DP-S1 CD transport and the DA-S I D/A processor, they''re taking a serious crack at that specialty market, with several products aimed at the price stratosphere where even the most dedicated audiophile runs out of breath while clutching his or her wallet.
I first saw Denon''s DP-S1 transport and DA-S1 D/A converters at the 1991 Tokyo Audio Show, where it was on static display&emdash;a real head-turning combination. When some Denon people came to visit us in Santa Fe after last year''s WCES, they brought with them a pair of Denon''s huge, gorgeous, POA-Sl monoblock amplifiers (over 200 Ibs/side), and pre-production samples of the DP-S1 and DA-S1 But when they left, their new products went with them. Some months later, Santa''s UPS elves delivered to Stereophile two large, heavy shipping cartons The DP-S1 and DA-S1 production samples had arrived.
DP-SI
The Denon DP-S1 belongs to that class of transports that first impresses by their look of sheer mass. With its rounded central structure surrounded on three sides by a squared-off, sculpted block comprising the remainder of the chassis, it almost appears milled from a solid block of metal This, coupled with its gold-and-tan finish, makes the DP-S1 look very distinctive and, well, down-right ***y It exudes class, from its solid, die-cast, top-loading cover which opens silently and smoothly to receive the CD, to its bulky, solid look and feel.
The DP-S1''s massive appearance goes beyond the surface&emdash;Denon has made considerable effort to minimize the impact of the outside environment on its performance. They''ve fashioned what they call a "triple floating suspension" for the DP-S1. The pickup assembly is isolated from its traverse base, and the latter is further isolated from the mechanism base by springs and rubber dampers This base is separated from the bottom chassis by its own suspension in ad***ion, the chassis feet, made of a sintered metal alloy, arc said to provide still further vibration damping The main chassis itself is also sandcast of non-magnetic material.
After loading, the CD is held down by a 200gm stabilizer. An extra-thick spindle shaft and ruby shaft base minimize the wear that this added mass might cause. Once closed, the lid of the loading well is scaled by an airtight rubber flange designed to keep out vibrations&emdash;though this seems to be gilding the lily when added to the effect of the stabilizer The lid and shield also keep out external light, which is believed by some to interfere in subtle ways with proper CD tracking by reducing the carrier/noise ratio in the recovered signal.
Jitter is said to be kept in check by the above mentioned stabilizer, a high-torque, non-cogging motor, and an extra-large (32k) RAM readout buffer. In ad***ion, a provision has been made to operate the DP-S1 from the master clock in the DA-S1 D/A converter by means of an ST optical link between the two. This linking system which Denon refers to as Gen-Lock, supports all of the possible optical or coaxial digital signal links between transport and processor, not just ST optical. Digital outputs on the back of the DP-S1 provide for the full alphabet soup of formats&emdash;coaxial (RCA, XLR, BNC) and optical (ST and TosLink). Note that the XLR outputs (and the equivalent inputs on the DA-SI) have a 250 ohm characteristic impedance&emdash;not the 110 ohms of the AES/ EBU standard.
Only the basic operational functions &emdash;Open/Close, Play, Stop, and Track Forward and Reverse&emdash;are provided for on the transport itself; the remainder are on the full-function remote control. The most important functions, including Index Search, are provided. I do miss having a Pause control on the transport, and wish there were an A-B repeat somewhere (handy for a reviewer, but probably of little value to anyone else), but otherwise, the DP-S1''s ergonomics cannot be faulted.
I did encounter two faults in our DP-S1 review sample, neither of which interfered with its operation, but one of which was mildly annoying. First, the lid developed an occasional "twang" on closing which sounded like a momentary snag of a spring, Jarrmg only in contrast to the lid''s otherwise silky operation. Second, the display would sometimes zero out, and the transport controls would refuse to operate immediately after loading a disc and closing the lid. This could always be cleared by reopening and shutting the lid (which could fortunately be done from the remote). This happened about 20 /O of the time&emdash;I didn''t keep track. It will, incidentally, happen all the time if you neglect to place the stabilizer on the disc&emdash;this is probably intentional.
Both remote and stabilizer come in hinged wooden cases&emdash;a nice touch. There''s even an opening cut into the remote''s case for its transmitter; you can operate the remote without needing to remove it, but you do have to open the case''s lid to push the buttons.
I had one small problem with the stabilizer: Because of its mass and the shape of its hand-grip, I dropped it twice into the CD well and a third time outside of it. The well was fine (aside from an almost inconspicuous scuff mark), but one CD was damaged (part of a 2-CD set, naturally), and the stabilizer itself was nicked in two places. The stabilizer can be bent (denon warns of this in the owner''s manual), but fortunately ours was not.

DA-SI
The more conventionally shaped DA-S1 converter is finished in the same jewel-like fashion as the DP-S1, and has six inputs that correspond to the same output formats found on the DP-S1. Switches on the rear panel toggle two of the inputs between RCA coaxial and TosLink optical. Two digital outputs are also provided&emdash;TosLink and coaxial&emdash;to feed other components, such as the digital input of a DAT recorder.
A three-position toggle switch on the rear panel selects between the Gen-Lock mode (when the DA-S1 is connected to its companion DP-S1 transport), or Level I or Level II&emdash;the two levels of precision for the standard digital audio interface. Put simply, Level I is a higher-precision mode, requiring tighter tolerances from the source in order to operate properly. Such a choice of levels is unusual&emdash;in fact, it''s the first time I''ve ever seen it. I did not encounter the need to switch into the lower-precision Level II mode, though it may be useful in rare circumstances.
The DA-S1''s analog output stages have FET inputs and a discrete, class-A circuit design with no negative feedback. They are DC-coupled; servo circuits are used to keep DC from reaching the outputs. The balanced outputs are generated at the output of the digital filter. The interior layout isolates the left- and right-channel analog stages on the chassis'' left and right sides, respectively, with the power supply and digital stages located in the center. The digital and analog circuitry have independent power supplies, including separate power transformers. In ad***ion to photo-couplers which prevent digital noise from entering the analog signal through the ground line, two copper shield walls inside the chassis separate the analog and digital circuitry.
Denon''s headline feature for the DA-S1, however, is their ALPHA (Adaptive Line Pattern Harmonized Algorithm) processor. The stated purpose of this circuitry&emdash;which precedes the digital filter and Denon''s 20-bit D/A converters&emdash; is to minimize quantization noise and improve low-level resolution. The sampling process, by its nature, results in waveforms which arc a series of very fine stair steps rather than smooth curves. The basic function of the ALPHA processor is to smooth these steps by generating ad***ional data to simulate four extra last-significant bits, 17-20, effectively adding 16 ad***ional steps to each existing amplitude step in the waveform.
These ad***ional bits can provide, of Course, only a "best guess," based on the clues provided by the existing data. They do not exist in the original, 16-bit source material. Denon is not the only manufacturer to attempt increasing low-level resolution by "recovering" ad***ional bits, though most of the other manufacturers use some form of DSP.
ALPHA is not DSP-based. Denon engineers developed ALPHA by studying known analog signals and performing both 20- and 16-bit conversions on them, using the most linear converters available to them. In comparing the results, particularly the change in the least significant bit, they determined that a look-up table could predict what ad***ional 4 bits would be required to closely approximate a 20-bit signal from a 16-bit source.
This, in essence, is how the ALPHA System operates. After these lower-order bits arc generated, they''re added to the original 16 bits, and the resulting 20-bit data stream is passed on to the DA-S1 ''s NPC digital filter and Advanced Super Linear 20-bit D/A converter.
SYSTEM
The DP-S1 and DA-Sl were evaluated in a system consisting of the NHT 3.3 loudspeakers bi-wired with Straight Wire Virtuoso loudspeaker cable from a Krell KSA-300S amplifier. The preamplifier was the Rowland Consummate, connected to the Krell by a balanced pair of Cardas Hexlink interconnects. The converter-to-preamp link was TARA Labs RSC Master, unbalanced, except where otherwise noted. The NHTs were set up along the long wall of my new 18'' by 26'' by 11'' (approximately) listening room and treated with a variety of devices, including MSB free-standing absorbers, Ring Diffusers and Absorbers, and ASC Tube Traps and wall panels.
SOUND
With the review of any piece of audio equipment, two questions need to be addressed: How does the product sound on its own merits? And how does it compare with the best of the competition? The second question is the more difficult to answer, because the relevant products to make the comparison are not always available.
Fortunately, the Mark Levinson Nos.31 and 35, and the Krell Digital DT-I0 and Reference 64, were all in-house. The No.31/35 combination''s price is similar to the DP-S1/DA-Sl combination; the Krell system, reviewed by RH last January, is more expensive. In this class, however, a few thousand one way or another is a walk in the park!
Because of the large number of possible ways of connecting the Denon pieces (to say nothing of the complications added by the comparisons with other equipment), I needed to pare down my options. I restricted my listening to ST optical and S/PDIF coaxial links (using a Kimber AGDL digital lead for the latter). I used Denon''s Gen-Lock link for all my listening, except where I wanted to assess the audible effect, if any, of its removal. In order to judge how they might differ from the unbalanced outputs, I briefly au***ioned the balanced outputs near the end of the listening (I would have brought them into play earlier if the results with the unbalanced had been going poorly); the differences were minor and did not alter my conclusions.
I also listened to the Denon transport and processor separately by comparing them to their opposite Levinson numbers (in ad***ion to comparing the full Levinson package with the full Denon one) I suspect, however, that most potential customers in this price category will be looking for the complete Denon (or Levinson or Krell) CD transport/processor package, rather than a mix''n''match alternative.
I began my listening with the ST optical link plus Gen-Lock. I''d been listening for several weeks to the Krell DT-10/ Reference 64 transport/processor combination, and didn''t expect the Denons to knock me off my peg. They didn''t. But, significantly, I was in no way driven to remove them from my system, either On the contrary, I found the Denons'' performance refreshingly clean, sweet, and easy to listen to, yet at the same time in no way lacking in detail. The top end was silky-smooth, depth was convincing, and vocal reproduction was see-through and palpable, with a combination of clarity, warmth, and three-dimensionally&emdash;rarely encountered at the same time in digital playback.
There may have been, in fact, a trace too much warmth on some program material, but that counted for little compared with the overall quality of the presentation. The Denons'' deep bass extension and weight were also absolutely stunning through the reference system, from the pin-you-against-the-wall impact of the low percussion on Vaughan Williams''s Sinfonia Antartica (Koss Classics KC-2214), to the torrent of deep, resonant, and character-laden low pipes on Jean Guillou''s organ transcription of Pictures at an Exhibition (Dorian DOR-90117).
The Denon pair, if not quite the opposite of ruthlessly revealing, did tend to be forgiving of program material that doesn''t sound the best. But you could easily say that it simply presented the material without exaggerating its warts. While I''m fond of a number of the songs on Daniel Lanois''s Acadie (Opal 25969-2), I''ve been bothered in the past by the recording''s rather artificial sound. While the Denons did not transform it into a model of natural balance and perspective, I could clearly hear things in the mix which enhanced my appreciation of the recording and the musical values in the performance&emdash;among them a spatial quality which ranged from the intimate to almost pseudo-surround. The balance remained slightly on the warm side of neutral, but I''ll never think of this recording in quite the same way. This is a very good sign with any piece of equipment.
The same was true for other recordings which can be difficult on many CD players. Leo Kottke''s My Father''s Face (Private Music 2050-2-P) seldom reproduces well in its entirety, though certain cuts travel well on most systems, particularly "Mona Ray." On the Denons, it sounded superb from beginning to end. The toughest Cut of all, "Back in Buffalo," fell into place, from the densely mixed multiple guitar and percussion threads to Kottke''s slightly laid-back and reverb''d vocals. This cut sounds fuzzy on many systems. Not here&emdash;I could almost count the individual vibrations in the guitar strings, yet the sound was neither irritating nor etched. This certainly wasn''t a warm classical-guitar sound, but it was an exiting, first-class sonic performance nonetheless.

COMPARISONS
I spent a great deal of quality time listening to the DP-S1 and DA-S1. But how would they compare to their high-end Krell and Levinson competitors?
First up was the Krell DT-10 transport and Reference 64 converter, tied together by their ST optical link and ST optical "Time Sync" connection. "Wow, this is interesting," I wrote in my listening notes. The Denons definitely sounded different from the Krells. The DP-S1/ DA-S1 sounded fuller and richer, with a less clearly defined bottom end. its soundstage was more expansive, with a sense of bloom, particularly with solo instruments and voices, which eluded the tighter, more analytical Krells. The latter had the greater pinpoint focus, was a bit less forward and palpable, and had a tighter, if somewhat smaller, sound. The Denons were more relaxed if, at the same time, less precise. On Potato Radio Justice JR 0802-2), the second album from singer Nancy King and acoustic bassist Glen Moore, the Krells, by any objective measure, gave the more open, transparent window on the performance. Yet the Denons definitely sounded richer and somehow more palpable. For example, with the Krells, it was easy to separate the backup vocalists who sing near the end of the title track. But with the Denons, it was easy to just relax and go with the flow of a sweeter, more relaxed presentation.
It was almost like an analog vs. digital face-off played out on an entirely digital stage, or a tube vs. transistor challenge with nary a glowing bottle. Certainly neither product was guilty of creating serious digital artifacts, nor did either sound like tubes. But the analogy holds up: The Krells had a more sculpted sound; the Denons, without losing vital detail, had a sweeter one. The analogy only broke down in the bass, where the Denons, though less tight, sounded fuller and more powerful. On Pictures at an exhibition for example, the Krells seemed slightly resonant next to the Denons. But which is "right"? Not so fast&emdash;the saga continues. At this point, I was still trying to make up my mind. I acknowledged that what I was feeling was probably a more accurate portrayal from the Krells, but I also felt a powerful pull toward the Denons'' undeniable sonic appeal. As an audiophile and reviewer concerned with what''s actually on the recording, I wanted the grip and control of the Krells. As a listener and budding low-bass freak (the NHT 3.3s will do that to a person), I wanted the upper-end ease, midrange palpability, and weight and power of the Denons. A textbook case of terminal Audiophilia nervosa.
But was I getting the absolute best out of the Denons? I needed to take a side trip here, making two stops along the way. First, I wanted to investigate the effect of Denon''s Gen-Lock link between the transport and processor. The differences were definitely subtle at best&emdash;certainly less than the differences between the Krell and Denon pairs&emdash;but the more I listened, the more I preferred the Gen-Lock mode. The principal difference seemed to be a slightly sweeter sound overall, though with equal or even slightly better detail retrieval, with Gen-Lock on. There was also more subtle bloom&emdash;a more expansive sound&emdash;with the locked connection That said, I wouldn''t want to be asked to pick out the differences in a blind test; it really was one of those improvements which make sense only after everything else has been sorted out.
Next, I replaced the ST optical link with a Kimber coaxial digital cable, once again with the Gen-Lock hooked up. I slipped Acadie into the DP-S1''s CD well, plopped on the clamp, pushed Play, and settled into my new listening chair while the transport lid gently lowered into place. There was definitely a change hardly earthshaking, but clear. The sound of the Denons, while still retaining the qualities I found so appealing, tightened noticeably: more focused in the mid-treble, tighter in the bass. But it still retained its bottom-end extension, top-end sweetness, and midrange richness. In ad***ion, while it didn''t completely cure it, the Kimber cable considerably helped a small problem that I''d noted the Denons had on some program material with the ST link: Their slightly soft balance in the mid-treble gave a bit of added prominence to the upper treble, resulting in a slight dryness. The change to the coaxial link launched a very good transport/converter combination into competition with the very best.
Including the Levinson No.31/35? To determine this, I performed three comparisons: First, using ST optical links with both the Levinsons and the Denons; second, using Kimber coaxial with both; and third, using Kimber coax on the Denon and ST on the Levinson. The Denon vs Levinson with ST resulted he a similar conclusion to that reached when I used the ST link with both the Krell and the Denon: The Levinson pair had a tighter, cooler, more focused sound, while the Denons were warmer, fuller, richer. With the Kimber coax on both, the Levinson became too cool and lean while the Denon sounded more musically right. Readers with good memories will recall my preference for the sound of the Levinson No.35 with coax over ST in my Vol. 16 No. 11 review. The difference here was likely due to a different system in a different room. The WATTs/Puppies driven in the Stereophile listening room sound warmer than the NHT 3.3s in my new, larger listening environment.
In any event, the above led to the third comparison: Denons connected by Kimber coax (with Gen-Lock) vs Levinsons connected by ST. Here the Levinsons were just short of being too sharply focused and precise and ending up crisp and analytical. The Denons were more easy and relaxed, loosening that grip and focus slightly, but not too much. The Denons were also a bit more forward and immediate, bringing solo instrumentals and vocals a shade closer, and they retained their deeper, richer, but marginally less tight bottom end. I have a weakness for full-bodied, palpable vocal reproduction, and it was probably this quality, more than any other in the Denons, that roped me in.
So which Combination&emdash;the Levinson or Denon (or the Krell, for that matter)&emdash;is "right"? Well, when we all have perfect recordings and systems, it''ll be easy to answer that question. Until then, I say that the Denon combination moves easily in the best of company and, in certain respects which are important to me, can show the others a clean set of heels.
MIXING IT UP
Up to this point, I''ve taken the Denon transport and converter as a package&emdash;all for one, one for all. Though it''s unlikely that anyone shopping in this price category would opt for an unmatched set, I still took the time to make two selected comparisons.
First, I compared the Denon DP-S1 and Levinson No.31 transports driving the Denon DA-S1 converter. Using the Kimber coaxial for both digital links, with the Gen-Lock engaged for the Denon pair, I found the biggest difference to be in the warmth region: The Denon transport gave a warmer, fuller sound, while the Levinson tightened things up noticeably, as it had earlier with its own No.35 stablemate. At the top end of the rangy the Levinson was a shade less dry at the very top and slightly brighter in the mid-treble. While this was a net gain on some program material (especially on closely miked vocals, where the Levinson was marginally smoother on sibilants), the Denon''s added warmth and bloom won out on others. When I switched to the ST link for the Levinson transport, I noted minor changes, but they did not significantly alter the above observations.
Next I used the Denon transport to drive both the Levinson and the Denon D/A converters. This was the most interesting comparison of all, and also provided the closest match. Combined with observations in the above paragraph, this test indicated to me that much of the perceived warmth of the Denon combination is in the transport. When I used the Denon DP-S1 to drive them both, the Levinson and the Denon converters were much more closely matched, with respect to this feeling of warmth, than they were when each was driven by its own transport. (Again, this was the case with either ST optical or Kimber coax linking the Levinson No.35 to the Denon.) In fact, I welcomed the added warmth that the Denon transport added to the sound of the No.35.
The slight dryness noted earlier in the Denon converter''s very top remained; at the end of the day, this remained its only real weakness. While the differences between the No35 and the DA-S1 in this comparison would not be enough to drive me to play mix''n''match if I were buying both transport and processor from scratch, it would be a worthwhile comparison for someone who already owns a No.35. With its attractive yet natural-sounding warmth, the Denon transport would appear to sonically resemble the C.E.C. TL 1, though I didn''t have enough time to make a direct comparison of the two pieces.
MEASUREMENTS
The measurements taken wore primarily of the DA-S1 D/A converter. Unless otherwise noted, the measured results shown arc for the unbalanced outputs.
I ran into one problem with the DA-S1 almost immediately. It refused to lock to an input with a sampling rate of 48kHz or 32kHz. Fortunately, it did lock to the CD standard of 44. 1kHz, else it wouldn''t have gotten very far into the listening tests! Since even the cheapest processor will lock to all three sampling rates, I can only attribute this to a sample defect. I''ll request an ad***ional sample to check this out, and report the results in a follow-up. The Denon DA-S1 has an output of 2.54V (L) and 2.61V (R), unbalanced, when decoding a 1kHz, 0dBFS (full-scale) sinewave. The corresponding balanced output levels are almost identical at 2.63V (L) and 2.66V (R). The output impedance measured 90 ohms. The output impedance at the balanced outputs measured double that. The Denon was non-inverting from the unbalanced outputs, with a positive-going impulse test signal reproduced as positive at its outputs. The balanced output is configured with pin 3 positive, the opposite of the AES standard. DC offset was 1.7mV in the left channel, 1mV in the right, unbalanced.
The DA-S1''s frequency response was very flat, as may be seen in fig. I . The top curves indicate the response at the unbalanced outputs, the middle curves at the balanced ones. The only significant difference is the slightly tighter left-to-night output-level tolerance in the balanced outputs. The bottom curve in fig.1 shows the de-emphasis error in the DA-S1&emdash;as close to a perfect response as I''ve seen. The latter is only of significance on a small percentage of discs, however; most are not pre-emphasized. The cross-talk shown in fig.2 is outstandingly low in either the balanced or the unbalanced mode. The balanced result is slightly better, but the differences are academic.
Fig.3 shows the audio-band spectrum of a decoded -90dB, 1kHz ***hered sine-wave signal. The balanced result (not shown) is virtually identical. Both results are excellent, with superbly low levels of noise. The virtually flat curve here above 2kHz is unusual (it was also noted in the Levinson No.35''s balanced result, but not its unbalanced one). Using the same type of spectral analysis, fig.4 shows the result of the Denon decoding a track of "digital silence" (all data words zero) out to 200kHz, with first-rate results.
Fig.5 shows the error produced with a ***hered fade-to-noise signal, indicating superb linearity down to well below -100 dBFS. Only the unbalanced result is shown; the balanced is very similar, matching the unbalanced to within about 0.2dB down to at least -110dB and to within 1dB down to -120dB.
The plot in fig.6 shows the results of the Denon decoding a 1kHz, un***hered sinewave at -9OdB. This is an unusually good result&emdash;closer to a noisy sinewave than to the stairstep result usually considered optimum for such low-level decoding. The difference here is in the effects of the ALPHA circuit, which is designed to do just what you see: create a better ren***ion of the waveform by interpolating low-level bits. Remember, however, that this is an un***hered sine-wave. (I''ll have more to say on this at the end of this section, when I take a closer look at the effect of the ALPHA circuitry.)
The noise modulation as a function of signal level vs. frequency is plotted in fig.7. The results shown here are for six different, progressively decreasing signal levels, from -50dB to -100dB. The more tightly clustered and the lower in level the results, the better they are; the clustering here is one of the best results I''ve seen, with only the slightest divergence visible.
Feeding a full-scale combined 19kHz - 20kHz signal into the Denon and performing an FFT analysis of the output results in the plot shown in fig.8. The artifacts are very low, with the highest&emdash; at 1kHz and 17kHz&emdash;over 85dB down (less than 0.006''/o IM). The balanced result (not shown) is better yet, with similar high-frequency artifacts&emdash;all approximately 85dB down or more&emdash; but no artifact at 1kHz. I used the Mitner LIM Detector to assess the jitter performance of the DA-S1; FFTs of the LlMD''s output are shown in figs.9-12. The source for figs.9 and 10 was the CBS CD-I test disc played on the PS Audio Lambda transport with a coaxial link to the DA-S1. The results here are not among the best we''ve measured. Fig.9 shows the worst-case result, at a signal level of-50dB (RMS jitter: 352.6 picoseconds); fig.l0 shows the best-case result, at -70dB (142ps jitter). The best-case result was more typical of the values obtained at the other four 1KHz data points measured. Note also that the results here were measured with the Level II setting of the processor''s input (see text of review). The level I setting produced higher jitter, undoubtedly due to a better match between the Lambda and the DA-S I at the Level II setting.
RH discussed the jitter in the DP-S1 transport in his article on transport jitter in Vol.16 No 11 (p.105). Basically, the results were very good&emdash;comparable in magnitude to other top transports&emdash; though some of the latter measured slightly better. In ad***ion to measuring the DA-S1''s jitter when driven by the PS Audio Alpha, I also repeated some of the same measurements using Denon''s own DP-S1&emdash;specifically with -50dB and -70dB 1kHz sinewaves&emdash;to assess the overall jitter of the combination. First I used the coaxial link, without Gen-Lock.
The results were quite comparable to those obtained with the Lambda, at least in magnitude. At -50dB and -70dB, the jitter''s RMS values were 1 78ps and 355ps, respectively. Here, I used Level I precision for the input; with the Denon transport, Levels I and II gave similar results. The plot of a -70dB, 1kHz input, with the two Denons, is shown in fig.ll. There arc fewer spikes than in fig. 10, but those that remain appear Clustered in several frequency bands.
When I hooked up the Gen-Lock between the Denons, I expected to see a reduction in jitter, this being the principal reason for using transport-converter synchronization. Instead, I noted a significant Increased The -70dB level is shown in fig.12 (4783ps RMS jitter). The jitter level was particularly sensitive to the insertion of the ST Gen-Lock connector. Removing it and reinserting it could Change the reading up to 25'':h. 1 was so surprised by the increased jitter with Gen-Lock that I reran the test the following day, and got essentially the same results (the plot in fig.l2 and the RMS value shown are for the best-case result) .
While this is a bit troubling from a technical Standpoint, recall that I found the audible differences between locked and unlocked modes to be very subtle in my system. In fact, as I reread my listening notes on the differences, I noted that, if anything, I found the sound with Gen-Lock to be very slightly sweeter, with a trace more bloom. But was this actually an artifact of the higher jitter? Was I perhaps assuming that if any differences were heard, the Gen-Lock sound should be correct, since it''s supposed to reduce jitter? Again, however, the audible difference was small. I tend to be conservative on the subject of how jitter affects the sound and what its threshold of audibility might be; we''re still accumulating experience and data on this subject.
The DA-S1 has a hidden code which reconfigures it to allow the ALPHA circuit to be toggled on and off (with the Invert button) to compare its effect on the out-put. This code also allows the use of the CHS input to perform a bit-shift function, raising the output of the DA-S1 by 48dB and making the changes in low-level signals more audible (and visible on a ''scope) .
Using the ALPHA on/off mode, I first measured the effect of the ALPHA circuit on a 1kHz squarewave recorded at -6dB. (The source was an ALPHA test disc furnished by Denon&emdash;essentially similar results were obtained with a 0dB square-wave on the CBS test CD, though with somewhat more flipping evident in the wave-top oscillations.) First, in fig. 13 (top trace), we see the response of the DA-S1 to this signal without the ALPHA circuit: It''s identical to responses we''ve seen in the past at the output of a CD player or converter that has linear-phase digital filters. (This response is so common that, for the most part, we''ve Cased publishing it.) But look at the bottom trace in fig. 13, with ALPHA switched on: a nearly perfect square wave There''s both more and less here than meets the eye. one of the features of the ALPHA circuit allows it to differentiate between a music signal and a pulse-type signal&emdash;such as an impulse or a squarewave&emdash;and optimize the digital filter accordingly. While this feature (Denon calls it Automatic Low-Pass Filter Harmonic Adjustment, in a further wrinkle on the ALPHA acronym) would seem to be of little value in normal use, it does result in the best looking square-wave I''ve ever seen from a CD.
Next, I wanted to look a little deeper into the result I obtained in fig.6, the low-level sinewave response. Here I was able to make use of the DA-S1 ''s bit-shift feature to raise the signal high enough to minimize the effect of noise. In the top trace of fig.l4, I repeated the measurement on a -9OdB, 1kHz un***hered sinewave, this time with ALPHA disabled. Note that, except for some oscillation on the tops and bottoms of the steps, due to the expected band-limiting effects of the CD system (the same as that observed in the 1KHz squarewave), it''s almost a perfect stair-step response&emdash;better than we usually obtain, due to the clarifying effects of the hit shift. The lower trace in fig.14 shows what happens when we switch ALPHA the change is dramatic, the result being a far more convincing sinewave.
But what happens when we add ***her? The top trace in fig. 15 shows the same -90dB sinewave, but this time with ***her, without ALPHA. The lower trace shows it with ALPHA. l don''t see any substantive difference This is a bit sobering. Denon makes much of the ALPHA circuit, yet when the low-level signal is ***hered, the result of the added bits is not evident. Since almost all digital recordings these days use ***hering, the real-world effects of ALPHA may not be all they would appear to be from fig.l4. ***her, you recall, was added to increase the CD system''s resolution below the 16-bit floors.

Denon has also recorded some low-level music on their ALPHA test CD, which may he au***ioned with the bit-shift engaged. Listening to this, I noted considerably more noise without the ALPHA circuit, but the remaining noise With ALPHA was more intermittent&emdash;more spiked less hissy. The test disc does not state if this Selection is ***hered or not; I suspect not.
Finally I checked the DA-Sl''s tracking using the Pierre Verany Digital Test CD set (PV.788031/032). Tracking was good up to band 32&emdash;well above the level required by the CD standard&emdash;but we''ve measured a number of CD players that have better test-bench tracking. Though I was a bit disappointed in the DA-S1''s jitter results, they were not Unacceptable. We also need to follow up on its 48kHz and 32kHz locking capability. But these items need not detract from the DA-Sl''s generally first-class measured performance.
CONCLUSION
CD replay is a moving target. Though the pace of development has noticeably slowed over what it was, say, three years ago, the river still runs fast and deep in this corner of the high-end valley. Today''s top CD playback may well be found in tomorrow''s more affordable machines. Nevertheless, both the Denon DP-S1 and DA-S1 can more than hold their own with the best I''ve heard. Whether or not Denon''s ALPHA processing has anything to do with this&emdash; and I tend to doubt it, because I was taken by the transport/converter combination''s natural timbre and balance rather than with any magic way it had with detail, low-level or otherwise&emdash; something worth watching is happening at Denon. I''m intrigued enough by their latest CD technology to think that we should look at their best current one piece CD player with ALPHA processing.
For those who have been looking at the Levinsons and Krells of the CD high end, however, your world has just become a little more complicated. While you may well feel that one of those other options better fits your system/perception of reality, you definitely can''t afford to overlook the flagship CD combination from Denon. It''s really something special.

Con S1 đấy chắc VN không có anh nào chơi nhỉ.
Ở HN thấy có bộ POA-8000 monoblock cũng rất gấu, giá hơn 20 triệu cả rề đời 6000, bác nào máu chơi denon là cũng ngon đấy

Tuần trước vừa thấy trên Ebay có chú bán con DP-S1 transport này giá buy now là $1200. (giá mới $8000).
Đang có chú list con này trên audioweb giá $1050 này:
http://www.audioweb.com/Ad/AdInfo.asp?adid=146190.
Tôi thấy đồ high-end Nhật, cỡ: Sony, Pioneer, Denon, Marantz... mất giá rất nhanh so với các sp cùng loại của Châu Âu, Mĩ. Về transport người ta vẫn nói đến những hãng reference như: CEC, Forsell, Wadia và Mark Levinson thôi.

tui băng thông của ampli công suất nên lớn hơn 20khz. Bởi vì do hiệu ứng băng thông nên băng thông cuối cùng khi nối với pre ampli sẽ nhỏ hơn 20khz

hè hè hè
thường thì băng thông do của các cty là kết quả của đo đạc thật tức là có cắm vô nguồn tín hiệu rồi! chứ không cắm vô nguồn tín hiệu thì lấy gì mà đo
Mô phỏng là đo đạc trên máy vi tính các bác ạ!.Cũng như gõ văn bản trên máy vi tính.Gõ sai? không sao! chả có chuyện gì trước khi máy in hoạt động. Tương tự đó là lợi điểm của mô phỏng. Dòng điện qua con sò quá cao? không sao? điểu chỉnh thiên áp lại là được!Chả có chuyện gì trước khi ráp mạch thật sự
Chỉ có ngồi tính toán khơi khơi mà không mô phỏng hay ráp thì mới nghi vấn thôi