Cayin 265Ai Integrated Class A Amplifier March, 2004 Arvind Kohli


Output: 40 WPC into 8/4/2 Ohms
MFR: 2 Hz - 30 kHz
Input Impedance: 95 kOhms
Slew Rate: > 100V/microsecond
Sensitivity: 500 mV
THD: >0.05% at 1 kHz; 0.1% Full Bandwidth.
Size: 6"H x 17"W x 16.75"D
MSRP $1,400 USA

Zhuhai SPARK Electronic Equipment Co., Ltd;
Zhuhai City, Guangdong Province, China.

North American Distributor:
Allied TV & Sound


Until recently, I have exclusively reviewed solid-state designs with an A/B output class. So, I felt it was high time I tried a design rated at full Class A operation.

Well, let me tell you not a lot of these are available in an integrated configuration. However, there are several integrateds rated as Class A/B that have a moderate to high bias into Class A, the intent being that for low to moderate listening levels operation would be in the Class A mode. The previously reviewed Simaudio I-5 was one such model. Long story short, I came across the Cayin 265Ai, and hence this review.

What is the big deal about Class A, you ask? Well, I am not qualified to do more than repeat what I have learned from others on this topic, and there are several excellent papers on the web dedicated to explaining Class A amplification. The appeal to me is a design that is simple and does not compromise the current available to the speaker for the sake of efficiency. Basically, Class A means that the current is already flowing in the output devices, and is diverted to the speakers when the demand is there. With Class A/B, when the Class A power runs out, it is then in Class B, which means that current has to be switched in when the demand is there. This can slow things down a bit, and cause distortion.

The tradeoff is much higher power consumption and heat dissipation (the current in Class A is dissipated as heat when it is not flowing to the speakers), neither of which bothered me with this review subject. The amp was placed on the top shelf of my rack and unencumbered on all sides, so the heat sinks were given plenty of circulation. The power consumption was about 200 watts regardless of output (rated output is 40 watts per channel). In contrast, my old Bryston 4B consumed about the same at idle and went up to over 1900 watts at maximum output. I can actually plug the Cayin into my P300 Power Plant along with my source components, now that is worth quite a bit.

Company Profile

Zhuhai Spark Electronic Equipment Co., Ltd. was founded in 1993 to design and manufacture audio products. It is a subsidiary of China National Aero-technology Imp. & Exp. Corporation (CATIC), which was founded in 1981. Their website claims Revenues of nearly USD $1 Billion for 2002. There is a very elaborate structure of subsidiaries that links Spark to CATIC, the details of which are completely irrelevant here. The only relevant piece of information is that Spark seems to be backed by a large and well-established Chinese conglomerate of high-tech products. That kind of technical infrastructure and talent pool is bound to pay tangible dividends. Their products are distributed in more than 20 countries, and claim to have participated in audio shows in Tokyo, London, Milan, Frankfurt, Copenhagen, Athens, Hong Kong and Taipei. They also supply OEM audio products to un-named companies in Japan, Europe, USA, and Taiwan. There are 15 engineers on staff. A statement from the company emphasizes their background with output and power transformers. If they have a weakness, it seems to be in the marketing, web design, and translation skills. As far as I am concerned, I am only eager to forgive many shortcomings in those areas if the product is something to behold.

The Cayin brand produces a variety of components, including two CD players, a tuner, a mini-monitor, three other solid state integrateds, one solid state amp, and a whole host of tubed amps, preamps and integrateds. The 265Ai is the only solid-state Class A amp they make.


The Cayin was plugged into my PS Audio P300 Power Plant at all times, and seemed to draw about 200 watts, which sounds right for an amp that delivers 40 wpc in pure Class A. Of course, the power consumption did not change with the volume level, since the input and output of energy in a Class A amp is always at the maximum value. The difference is that at higher output, more of the energy is delivered to the speakers and less is dissipated as heat.

The amp is intended to be left on standby at all times and takes about 10 seconds to power up when turned on. About 10 to 15 minutes of idle seems to get it fairly warm, and ready for use. Upon powering up the volume control turns all the way to zero, which turned out to be a very handy feature for me, since I have two very curious toddlers.

The Outside

I would have to give this amp very high marks for its stunning looks and clean layout. It is only available in a brushed aluminum finish, and I found it very attractive as such. The front panel has a large volume dial that provided perhaps the best control I have ever experienced, as getting to the desired level took only one try (I hope other manufacturers begin to realize how important this is). The left side has the power switch that brings the unit in and out of standby. Above it, in a recess are two lights; I could not get confirmation on what those lights indicated. My guess is the lower light indicates standby (green), transition (red), and power on (green), and the light above it probably indicated the amp is being over driven, but I was not able to get that to activate for the levels that I tested. On the right side a single button toggles through the various inputs, and the recess above it has six lights to indicate the active input. Done. No other bells, whistles, knobs, switches, meters, dials, sliders, buttons, LEDs, displays, etc. My definition of simplicity would not allow any more than this, and my need for functionality would not allow any less. A perfect balance.

The rear panel features one set of XLR inputs, five sets of RCA inputs, and one set of RCA tape out. My only wish here is that there also was a pre-out. Speaker connectors were via a heavy-duty set of brass five-way binding posts that are well spaced. There also is an AC input for a detachable cord and the main power switch just above it. Again, the rear is simple, clean, and very thoughtfully laid out. The sides brandish the beefy heat sinks, mandatory for this output class.

Fit and finish are outstanding. Sorry, no headphone jack, phono preamp, balance, mute, mono, fixed level out, or other frill. Frankly, I was glad, since I have no value for any of these features, and don't care to pay even the nominal amount they would add to the cost. More importantly, I am glad they were not included at the cost of cheaper parts elsewhere or needless degradations to the signal path. I do wish there was a pre-out for a subwoofer, but am not terribly heartbroken about it.

The plastic remote is a full function and compact design, which easily fit into the palm of my hand and was easy as pie to use. I am glad it was not some very elaborate piece fashioned out of metal to give the impression of needless quality. Again, I do not care to pay for those frills or have something else compromised to make up for the cost that would have been lavished here.


The manufacturer's website and literature are lacking in detail and very poorly translated. I had to rely on Mark Levine of Allied TV & Sound (North American distributor) to supply the technical data. He in turn, had an engineer on his staff bench test the specs quoted here. Interestingly enough, the engineer decided to buy a unit immediately after completing his tests.

Input impedance is a very high 95 kOhms (all inputs), indicating that sources with a very high output impedance will also be easily accommodated. The input sensitivity is 500mv for full swing output, again this will be sure to accommodate any source device I have heard of. The preamp section is fully discrete, i.e., there are no integrated circuits in the output path.

The Dual mono layout starts with a 500VA Toroidal transformer for each channel. These are EMI shielded and claim to have very low mechanical hum and vibration. Power rating is 40 wpc, and does not change for loads from 8 ohms to 2 ohms. Rail voltage is 100v maximum. Total capacitance is an astounding 40,000 μf per channel. Available energy storage is 200 Joules.

Output devices are Toshiba Mosfets. Bandwidth is rated at 2 kHz - 30 kHz
± 0 dB. Slew rate is better than 100v/μs. THD is rated lower than 0.05% at 1 kHz and 0.1% at full bandwidth.

No wonder the distributor's engineer scooped up a unit for himself after confirming the specs on his bench.

Warranty in North America is 3 years.

The Sound

All tests were done with the previously reviewed Dynaudio Contour 1.3 MkII speakers and Triangle Electroacoustique Titus 202 speakers. I also had the recently reviewed Creek 5350SE integrated amplifier on hand for direct comparison. To disallow bias due to varying volumes, I measured two levels that registered at 60 dB and 80 dB at my listening spot (on each speaker and amp), using a -20 dB pink noise test tone. Here are the detailed highlights of my listening tests.

Coming of the Mandinka (Mumtaz Mahal; VM Bhatt, Taj Mahal, N Ravikiran; Waterlily Acoustics; WLA-CS-46-SACD) has become a constant in my life for the past few months. I am not sure if it is the fusion of Blues and Indian Classical, or the purist recording with audiophile components that make this a great album, but I just cannot stop listening to it. I started with the Creek and was very impressed with the detail in the decay of the primary strings on the three stringed instruments and the resonance of the sympathetic strings of the two Veenas. I was also impressed with what that little integrated could do in terms of staging and the dynamics of Taj Mahal's voice; that man can effortlessly belt out an extra 10 decibels or more.

I slipped in the Cayin, and things moved up to a whole different level. My normal method of critical listening is to listen to about a one-minute segment repeatedly (say 10 to 20 times, allowing my short-term memory to retain as much detail as possible), and then switch in the comparative component and listen for differences. In most cases, I have to listen to the second component several times before I can distinctly pick out differences, and then I have to go back and repeat the process just to make sure I really heard those differences (Whew! I get tired just talking about it).

Well, with the Cayin three differences were evident immediately. First, there was a large increase in what is called macro dynamics. Whenever Taj Mahal would bellow out the extra 10 decibels or so, it would feel like a very effortless and natural increase. I never felt the Creek had a shortcoming in this respect (or any other for that matter), until I heard the Cayin right behind it. The dynamic range was so markedly increased that at times it felt like the volume level was set much higher, but since I had done my homework on establishing volume levels I knew that was not the case. Second, the strings on all three instruments seemed fuller and more lifelike, to the point that I could almost feel the physical impact of the string being plucked. Lastly, there was a marked increase in the warmth and richness of the vocals, again making them seem very natural and lifelike. This comparison was done with the Dynaudio at the 80 dB level.

Girl from Ipanema (Stan Getz, Joao Gilberto; Getz/Gilberto; Verve; 314-589-595-2) on SACD. The Cayin had a warmer and fuller sound in the midrange compared to the Creek. Vocals had an obvious increase in what I would call projection, and they seemed more lifelike. I cannot honestly say that I was fooled into thinking that Gilberto was actually in the room, but that illusion was substantially more felt with the Cayin than the Creek, or anything else I can remember listening to. I could hear the smallest changes in Gilberto's pitch or tempo, micro-dynamics is how this is often classified. I have never before heard such subtle changes in a reproduced human voice, with the perception being so immediate and effortless.

The Creek sounded noticeably different in the higher frequencies. I switched back and forth between the amps several times, but could not quite put my finger on what exactly that difference was. I almost wanted to say that with the Creek the soundstage and the instruments within seemed a little more distinct and spread out. Since I could not very clearly pick out the differences, I picked out a few more tests to see if I could isolate and distinguish what was going on. This test was done with the Dynaudios at the 60 dB level.

So I replaced the Dynaudios with the Triangles, which has relatively less bass extension and would allow me to better focus on differences in the mid and higher frequencies. The track I used was "General Image and Resolution Test" (Chesky Jazz and Audiophile tests Vol. 2; Chesky; JD68), which involves a purist recording of four musicians walking from the back of a hall towards the microphone setup, circling the microphones and walking back out of the hall, all while playing a different instrument each. This test helped clarify the ambiguity I had in the previous test. The Cayin actually had a significantly better separation of instruments, sonically and spatially. The maraca sounded extremely lifelike, and I could track its movement around my room with extreme certainty at all times. It felt like I could feel the musician walk in front of me.

I also tried to follow the hand-drum player, and again it was much easier with the Cayin to be able to point him out with confidence. With the Creek, there were moments when I might have hesitated to try and estimate his location, or there were times when I knew where he should be, but my ears would not confirm that.

There were a couple of interesting things I noted with this test. The liner notes on the CD indicate that in near field listening, you should hear the musicians circle you. I could only perceive them walking in I circle in front of me. I suspect that is more due to my setup not being near field, rather than its resolution. Also, I first listened to this track with the Creek and then the Cayin. With the Cayin I immediately noticed that the circle the musicians were walking was farther out on the right side than the left. I quickly discovered the reason: I had accidentally toed in the right speaker a little less than the left. Now I did notice something askew with the Creek, but it only came to light as soon as I started listening to the Cayin.

Of course, this is not a scientific conclusion that the Cayin is better. It is entirely possible that my brain is so slow to recognize a lop-sided soundstage (and my wife would agree about my being slow). Or, it took a long time to realize the slight variation, which by then the Cayin just happened to be the amp playing. Maybe I would have had the realization at the same time even if the Creek was still playing. Personally, I would give the Cayin at least some credit for highlighting the problem. I like that possibility, since it diminishes the accusation of me being a dimwit. This test was done with the Triangles at the 80 dB setting.

My expectations of this amp were snowballing into avalanche proportions. I started to expect enormous differences on every track I listened to, but I was quickly brought back to reality with "All or Nothing at All" (Diana Krall; Love Scenes; Impulse; IMPD233). Now don't get me wrong, there were obvious differences, but they were just not as easy to pick out or as large as I had expected. This could, in part, be due to the fact that this was the first critical listening session with a recording not from an audiophile label or in SACD format. This indicates that the magnitude of differences noted above are so small that they are noticeable only when listening to a high resolution format or a high resolution recording.

On the double bass intro it was easier to delineate the plucks with the Cayin than the Creek. There were several "things" I heard with the Cayin I thought I had never heard before, but when I switched in the Creek they were surely there, but just not as prominent.

When I say "things" I am referring to sounds so subtle and short in duration that I cannot with certainty say what they were; possible examples are a fingernail hitting the body of a guitar, a gasp of air, a slap on a thigh, etc. It is the minutest piece of audible detail, but often too vague or unfamiliar to describe. The more readily I hear these "things", the higher I rate a component for detail.

On this track, there also was a touch more warmth and life in her voice, but what was really obvious was the dynamic range. It was plainly obvious to perceive large and small changes in her voice, the transitions were effortless and immediate. Again, I had never noticed anything lacking with the Creek, till I heard it through the Cayin.

For a while now, I have been feeling a little frustrated with my Dynaudio Contour 1.3 MkII speakers. I am absolutely in love with the Triangle speakers, but they have one big limitation: they lose their charm at lower listening levels (specifically, the level I calibrated as 60 db, or anything lower). I have two young kids who go to bed early, and a house with all the rooms on the main floor connected via large openings, allowing sound to travel unimpeded. Thus, I have not had very much time listening to them in the way I would love to, i.e., at the volumes they really come to life at. Well, the Cayin has changed that for reasons unknown to me. I can conclusively say that listening to the 1.3 MkII has become just as pleasurable at lower volumes as at higher ones now.

Curious about the reason behind this, I called Al Fillipelli of Dynaudio. His explanation was that the drivers of the Dynaudio Contour 1.3 MkII have large ferrite components that are highly inductive, i.e., have a tendency to hold a lot of current. That is why the speaker fully comes to life only when supplied with an adequate amount of current. Class A/B amps are designed to supply only the amount of current needed at the time, and there is a time lag between demand and supply. Hence, the perception of 'sluggishness' or 'lack of speed'. According to Al, Class A/B amps delivers full current only at the top of their RMS sine wave. In contrast, a Class A amp has full current on the rails at all times, usually being wastefully dissipated into heat, but it is there and ready to go. It delivers the required current as needed (within its capacity) with virtually no lag, hence the perceptions of the sound being 'alive' and 'immediate'.

Mike Manousellis of Dynaudio also explained that the Cayin is likely a product of high resolution, and that improvement is apparent through the very neutral and transparent Dynaudio. When he learned that it was the Cayin that had caused me to call him, he mentioned that he had recently spoken to an acquaintance that owns the $30,000/pr Dynaudio Temptation and had also paired it with the Cayin. Seems like this gentleman was even more impressed than I was, and found the Cayin to be just as good as a to-remain-unnamed $18,000 amp. His enthusiasm for the Cayin being greater than mine could be for two good reasons. First, the Temptations are very likely to be much more revealing than my Contour 1.3 MkII. Second, I am as miserly with praise as I am with money, my wife will verify on both accounts.

To offset this good fortune, the Cayin also produced another change in my perception. I now do not like my Triangle Titus 202 speakers as unequivocally as I used to. I always felt that the Titus did a marvelous job in the midrange and mid to upper bass. However the high frequencies were not as refined as the presentation of the rest of the spectrum. But I never perceived that discrepancy to be large enough to impede my overall enjoyment of that speaker. As a matter of fact, it has been one of my all time favorite audio products. It still is, but now the caveat attached has grown considerably.

On many and several tracks, the Cayin plainly illustrated how wide the gap between the tweeters on the Titus and Dynaudio Contour 1.3 MkII really is. While with the Contour there was an immense amount of detail (perhaps the most I have ever experienced with any amp), the treble was smooth and natural. In contrast, the Triangle, while also quite detailed, tended to sound relatively harsh. I thus have to alter my verdict on the Titus. I still highly recommend it, but either with amps of moderate resolution or high-resolution amps with rolled off highs. I suspect high-resolution tube amps may be just the ticket. When the Triangles were returned to the second system to be mated with a NAD 317 Integrated, I enjoyed them just as ever. I am not sure if the NAD is of lesser resolution or has the highs rolled off, but I would have to guess at least one of the two must be true.


The clichė used ad nauseum by many reviewers is "This component (on hand) is better than those costing several times as much". Well, if everything is better than the average, then what exactly is average? But, I better get off that soapbox, before I digress too much. I would not be surprised if the 265Ai indeed is better than much pricier gear, but I did not have anything that expensive on-hand to compare to. I did compare it to the slightly pricier Creek 5350SE ($1500) and found that I preferred the Cayin by a wide margin.

At this point, I should write several lines with various superlatives (perhaps even invent a few), and prostrate myself to the virtues of this amp. But, like I said above, I am not lavish with praise. However, let me summarize how I feel about this amp by saying I bought one for myself, and it will be my reference for the foreseeable future. Period.

When I clearly prefer a component to a venerable competitor in its price class, I become very curious about how it would stack up against other highly regarded models. Rest assured, I intend to pursue that train of thought in future reviews.


- Arvind Kohli -

Associated Equipment:
Dynaudio Contour 1.3 MkII Speakers
Triangle Electroacoustique Titus 202 Speakers
ACI Force Subwoofer
Creek 5350SE Integrated Amp
Sony DVP-NS755 CD/DVD/SACD Player
PS Audio P300 Power Conditioner
Cables: Homemade

Our Vault pages may have some display quirks. Let us know if we need to take a look at this page or fix a bug.
Connect with us
  • Instagram
  • Google+
  • YouTube
  • LinkedIn
  • Pinterest
Secrets "Cave"