I usually go by sound but based just on these specs I wouldn't put this unit on my long list for a listen.
26 responses Add your response
This is one of those times ignorance may be just a little blissful :-) I was one of those kids who was decent at writing in school, but terrible at math. Now, I see frequency graphs and charts and specs and it all looks like advanced nuclear physics to me LOL. I just don't understand any of it. Because of my limited cranial capacity, I have to judge equipment based on how it sounds and if it makes me happy, not based on specs. I'm jealous of all you engineers who understand all that stuff, really I am.
That's the Stadard we accept...Good sound is whatever someone "likes"
And ...well...math..it remembers me to a meeting some time ago..."hey Paul, you are very wealthy...how did you made it? You've been terrible at math when we were at school together?"
Paul answered "Well, you know, I buy something for 1$ and sell it for 4$. And from these 3% I can llive pretty well ..."
The beauty of judging audio products is you can listen and form an impression. Syntax I'll assume you chose your systems components based on hearing them as the final determinant. They met your standard for listening quality and enhanced your enjoyment/satisfaction If I'm wrong my apology. If a component sounds good to you and you later learn it measured poorly would you then change your opinion of it?
Excellent point. -I was recently given a stack of Stereophile magazines to read. I'm about half way through them now. I read each review but haven't bothered to read the accompanying measurements section. Why, because it seems they really don't tell one much about how the component sounds. If the component sings, it sings - measurements are just simply that, measurements. If they told the story, we'd just have to buy the component with the best overall specs.
Art Dudley has reviewed three different D/As in 2014, a Luxman, Lector and Allnic. As measured by John Atkinson 2 of the 3 components are the worst measuring digital devices he has encountered (the Luxman actually measured very well). Art Dudley liked or even loved the performance of all 3 D/As. It should also be noted that Dudley's reference converter is a $500 device. Rather than offering any insight into the measurement vs. subjective debate, I think the review is more about the peculiarities of Art Dudley's sonic preferences. He's a fervent vinyl, triode (DHT) type guy. He is what he is.
That's exactly what I like about Art Dudley, he fits the term music lover. If a component makes music he'll tell you,and if not he'll say so. That's all I want to know. I think he and I hear in a similar manner and our priorities seem the same. Does the music sing with a component or not. If it can't stir the emotion and allow involvement I would have no interest in it at all.
Syntax has a very impressive system, it takes much time,experience and patience to do this. All of his components he proudly describes their "soiund" characteristics. He could have only reached these conclusions and kept them based on listening to them carefully. Lamm products for example are in his system because they sound good!
All I can say, for me are technical specs (very good ones) essential. And the next step is a sonic layout which can deliver a real thing sound. That combination is rare. We find a lot of units with top specs which sound dead like a piece of wood and you find a lot of good sounding units which have no technical standard, which can be rated in the wide, endless "I like it Section. This "I like it" has nothing to do with High End. It is a personal opinion. You can like your car stereo, or MP3, or something expensive which is cheap internally. In a way, I would say, modern High End is loaded with low end "solutions". The term High End is stretched too much in the last 15 years, it is time to create a complete new chapter.
How about "Fun End"?
Some time ago, when the Allnic Hype flooded Audiogon I went to an owner who had the top line Pre and Phono (I think it was 3000 or something like that at that time). He had a similar electronics, Turntable, Arm, cabling and a Lyra Olympos.
We let it run and to be honest I never say something when I am invited. It is my experience and I respect every owner. It is his money, his time, his Fun. Or, when someone is a Fan of Dudley's Finest...I have no problem with it. He asked me several times what I think about it and after a while I said "You don't want to know that".
Anyway, he insisted. Fact was (or is), those Allnics blow up the Tone but that's it. The Phono was not able to deliver the abilities from the Olympos. I need some time to believe that but no matter what setting we used, 70% was not existing AND it was in a way wrong. I had no explanation for that and the last chance for us was to bypass the Phono section via MM with an external SUT (that one had top specs and was a custom made one) and from the first second all in the room said "THAT is MUCH much better"....To finish that story, I was only allowed to leave him when he could keep the SUT.
I deleted Allnic from my memory and when I did read the comment from JA it came back. did it astonish me? Well, no.
I certainly agree that specs and measurements will generally provide little if any insight into how a component sounds. And I certainly agree that buying based on specs and measurements is almost certain to result in expensive mistakes.
However, I doubt that there are very many, if any, serious audiophiles having even a little bit of experience who buy based on specs and measurements. Although those who express disdain for specs and measurements often seem to make the IMO false assumption that those who consider specs and measurements to provide value do buy in that manner. In other words, a straw man argument.
The very considerable value specs and measurements can provide, as I see it and as I've mentioned in a number of past threads, is that when purchase decisions are being made they allow candidates to be RULED OUT, on the basis that they would be poor matches with either other components in the system (e.g. impedance incompatibilities, gain and sensitivity mismatches, etc.), or with the listener's requirements (e.g., dynamic range and maximum volume capability).
Without taking advantage of those kinds of benefits that can be provided by specs and measurements (and a good understanding of them), the randomness of the component selection process increases greatly. As does the likelihood of expensive mistakes.
And I suppose also that now and then there will be instances where poor measurements will be indicative of design problems that may be sonically significant, but which the reviewer may not have perceived for one reason or another. Perhaps because either his ancillary equipment or the recordings he used did not bring it out, or perhaps because he simply overlooked it.
If measurements are so poor as to suggest that kind of possibility, my question would be why buy that component, or even include it on one's short list for audition, when most likely others are available at a comparable price which sound at least as good and also don't have those measurement issues?
Apparently the Allnic review is in the December issue, which I haven't received yet. But I've read through the Lector review which Onhwy61 mentioned. With respect to the "dreadful" measurements which JA cited I can't say in general how likely and to what degree those may be sonically significant. What particularly strikes me, though, is his statement that "USB Prober appeared to indicate that the Digitube operated in isochronous adaptive mode rather than the preferred isochronous asynchronous mode." If in fact that indication of the "USB Prober" was accurate, use of the older and now largely superseded adaptive USB mode, rather than asynchronous USB, makes the device a non-starter in my book.
And perhaps most significantly, I would point out that in general adaptive USB sonics can be expected to be a good deal more dependent on the particular computer that is used than would be the case with asynchronous USB. And note that Dudley's sonic assessment was done "primarily" via USB. Which means that even if his assessment was 100% accurate, that assessment stands a good chance of not having much relevance to use of the Lector device in a different setup.
Your examples still make the obvious point that listening was required to make the ultimate decision. No escaping that fact. It all subjective, you didn't care for Allnic yet others love them. Some listeners don't care for Lamm, but you do, see what I mean? Measurements or technical specs can be good,bad or indifferent, you're still going to have to actually hear the product.
Technical specs serve a purpose just as Al has pointed out numerous times and he's right. They will tell you next to nothing in terms of the sound quality. You can't buy any audio component and based solely on its specs have any idea if you'll enjoy it or not. If there's are some where this is possible someone please let me know. Thanks.
Before I read the review, I thought maybe Allnic took the Audio Note approach of not filtering the digital datastream, which would lead to some poor measurements. I was frankly surprised that the Allnic employed upsampling and other digital manipulations and measure that badly.
So the reviewer said it was the best-sounding digital he ever heard, and JA said it was the worst-measuring digital he ever saw. Go figure. IMO, that reviewer was not qualified to review this product, as his experience with digital was extremely limited. His preference is analog and his digital consists of a $450 Halide dac. Also, as Lula pointed out, it appears that the Allnic was only reviewed via the USB input, and there could potentially be large differences, for better or worse, using a transport through the SPDIF input or whatever.
Having owned the D5000 for the last 5 months, I encourage everyone reading this thread not to dismiss this dac based on the Stereophile review.
After reviewing the article and researching some background regarding the physical dac used by Stereophile, I am somewhat perplexed myself.
The D5000 was first bench tested by John Atkinson, then reviewed by Art Dudley. As we have read, there is a significant discordance between the actual bench test and the reviewer's cerebral perception. I have no desire to be critical of either the approach or perception. This unit may simply demonstrate that the bench tests performed by Mr. Atkinson, while producing accurate measurement, do not translate to poor sonic qualities.
Could DHT tubes by nature test poorly?
The unit was returned to Korea and found to have 3 faulty DHT tubes, is suspected to have been damaged in shipping. This obviously can not be confirmed. Allnic has requested Mr. Atkinson to re-test with a different unit.
Interestingly, I had one DHT tube fail myself after 350-450 hours and observed no sonic degradation. Replacing the tube was easy and took only a few minutes.
On the first paragraph of Mr. Atkinson's review states "Apple's USB prober utility identified the processor as the 'D-5000' from 'ALLNIC SOUND' with the serial-number string 'USB' and confirmed that the USB input operated in the isochronous asynchronous mode." I did not read in the review: isochronous adaptive mode.
I spoke with the head engineer who also confirmed USB asynchronous mode, with similar traffic behavior to the XMOS chipset, although the standard UAC 2.0 is their own proprietary design with galvanic isolation.
I will post my thoughts of the dac itself soon. To my ears and those around me, it best EMM DAC2x. I also had less issues with the XMOS driver, which is used in both units.
Both the D5000 and the Empirical Overdrive SE loaded are wonderful dacs, each with their own qualities. If you desire DSD playback, I recommend auditioning this dac.
Sunil at Careaudio has a Allnic demo program. Mattenship will be auditioning the D5000 on his audiogon blog and I am curious to compare notes with his shootout with Empirical.
I own both dacs. If you are in my geographic region, you are welcome to listen for yourself.
3A5 are tiny DHT tubes. Is that an explanation why they are failing so much?
In the L7 Lampi, I can use 101ds, 2A3, 6A3, 300b, 45,VT52, 5b4g, etc and these seem way more robust (2.5v to 6.3v). It also seems like the bigger the tube the airier the sound. Rectis give even more rolling possibilities, as 5Y3, 5u4g,5R4, 274b, 422a, GZ34/37, 83v and 80 with adapters, U52, etc and they influence the sound enormously.
Tube rolling the Dac is loads of fun and one drawback of the Allnic is that you are stuck with the 3a5 as output tubes (2.8v). Can you at least roll the rectis?
In any case, I dont doubt that it sounds great. I would not pay much attention to the test bench report, as your ears will tell know all you need to know.
Bryston is very proactive r and d division with continuous firmware updates implementing improvements Much sooner than other manufacturers. Depending on the firmware version the dac works. Waversa other times rewrites their software to be compatible. However, this has been short lived with further bryston updates. They are working on a solution and. Will post here.
Btw. I trialled my d5000 with the bryston with the local dealer. I really like the bryston and they would pair well once this issue is resolved.
Waversa who designed and manufactured the digital components for Allnic has investigated Bryston's implementation. This has been an ongoing issue before Allnic was designed and built.
I have not heard from Allnic or Waversa of any issues from other transports or servers. From my personal experience, the impedance of my AES cable was not to the standard spec and this would create clicks while playing tracks. I had no issue with an Audience AU24se cable.
In my experience, the USB implementation is the most robust of both the M2Tech used in Empirical Overdrive SE, which was far more stable than EMM Labs use of XMOS.
If anyone owns or demos a D5000 and has any conflicts, please post your experience.
I sent my personal D5000 south for some bench tests which tested with similar results to John Atkinson. Interestingly the person who was kind enough to take the time study the dac also auditioned the dac and found it quite musical and enjoyable. He believed the abhorrent test results are attributed to the inherent characteristics of the DHT tubes. I would be interested in seeing if other tube dacs like the new Modwright Elyse or Lampi 7 re-produces similiar bench tests.
If you are interested in my impression of the D5000, read on:
After a somewhat long delay, I wish to report my thoughts on the D5000. Since this is a DHT tube dac, I was not able to burn the dac in 24/7 and rapidly achieve what I believe close to its optimal sounding potential. Instead this dac was burned in over longer time period of normal use. Surprisingly this dac sounds wonderful new and although the sound did improve over the next 300 hours, the improvement heard with the SABRE chip was not as pronounced when compared to other dacs. By my experience, I believe you will hear 90% of the dac potential after 40 hours when the tubes have had time to settle. For a short period of time I had two D5000's in my possession. The DHT tubes have a silicone ring positioned midway and I was able to A-to-B the dac with and without these rings. While I prefer not having these rings on either my L5000 DHT linestage (preamp) or the A6000 300b monoblock amps, they do give sonic improvement with a larger, deeper sound stage and better inner detail, particularly clearer attack transient and subsequent decay mostly noted in higher pitched percussion.
The USB interface is implemented through custom hardware by Waversa Systems and is UAC2 standard compliant. Because of this compliance, the XMOS Windows driver links the D5000 to my custom CAPS server without issue. After a multiple USB cable trial, I found galvanic isolated interface greatly reduces the sonic differences between cables. Although there remains a difference between extremes comparing a USB 2.0 generic USB printer cable to my preferred Vertere Pulse-R 1.5 meter USB cable, the difference is muted demonstrating subtle differences. The Pulse-R removes a slight course texturing or hash bringing the overall soundstage into focus.
Waversa's custom internal processor will perform either upsampling or conversion to DSD in real time. When upsample is selected, repeat pressing of the button will progressively increase upsampling from 44/48 to eventually 352/384 then revert to 44/48. Therefore, the user may inadvertently select a lower sampling rate than native and down sample that particular track. The user selection of a specific sampling rate was intentional as some tracks may sound better upsampled but the sweet spot may not be DXD. I tend to prefer upsample to DXD over conversion to 128 DSD for reasons difficult to qualify. Maybe this simply has to do with the sonic signatures of PCM vs DSD.
Listening to the D5000 right out of the box while playing red book files without any upsampling I noticed the smoothness and fluidity of the music with a natural quality that immediately reminded me with great fondness listening to my older brother's studio reel-to-reel decades ago. Both upsampling to DXD and conversion to DSD options improve openness and transparency.
Unlike the EMM which produced a precise, neutral clinical sound, the D5000 is both involving with improved timing and pace creating a natural liveliness across all frequencies and viscerally engaging triggered through a deeper intricately layered larger soundstage where the speakers disappear into the music often creating phantom imagines that become palpable.
Empirical Audio Overdrive SE (ODSE) is my current reference dac. This unit is fully loaded with all available options offered by Empirical. Like the ODSE, the D5000 naturally envelops me in the music. While they both are excellent dacs, they truly are a different flavor. The D5000 provides more stage presence, as if you were at a nice outdoor venue sitting in a lawn chair somewhere in front of the soundboard. Empirical has more delicacy, revealing the fingertip-friction sounds on an instrumental string, better voice articulation and over all inner detail creating an ambience of a small studio session. I remain to find a dac that re-creates this level of timbre with such realism that naturally differentiates the uniqueness of each individual instrument when simultaneously playing the same note.
My sound system has naturally evolved into an all Allnic system over the years, consisting now of 26 vacuum tubes in total when I add the D5000 into the system. Because of this, I wonder if the DHT output stage of the D5000 would provide sonic benefit in a solid state system that perhaps otherwise becomes lost it an all DHT sound system. The solid state output stage of the ODSE is well designed and includes Paul Hynes voltage regulators that provide greater inner detail. What the ODSE offers pairs very well with Allnic gear.
To me the D5000 is sonically better than the EMM DAC2x and different than the Empirical Audio ODSE. In todays market there are many excellent dacs and I realize the sonic qualities I have come to appreciate are not linked to market price. Dacs if this quality are like wine, each with their own strengths and sound characteristics. Despite the poor bench test demonstrated by John Atkinsons review in Stereophile published in December, 2014 which most certainly has to do with the DHT tube characteristics, the D5000 is very musical, organic and engaging. I recommend anyone interested to audition this dac. Sunil at Care Audio (Care-Audio.com) is offering an in-home demo program. He will send any Allnic model to your home anywhere in the lower 48. This requires a 50% deposit which is 100% refundable. Experience dictates dealers offer some discount off list price, especially with a multi-unit purchase.
I am certain other Allnic Dealers too will offer an in-home demo arrangement, which I believe is wonderful since I am unable to get any local audiophile dealer interested in letting me trial gear in my personal system without full purchase and return restock fee. When Sanjay Patel at Ciamara had no issue sending me any gear I desire and offered a great price on any piece I am interested in, he was my go to person since Ciamara deals with many quality manufacturers. Unfortunately Ciamara does not deal with Allnic, but they have gone virtual and national and if they carry gear you are interested in, they will treat you well.