[dfkt]

Quote:

one has to wonder where those other numbers came from?
 

Maybe the specs of the 'naked' D/A chip under ideal conditions?

 

[nick_charles]

 
Quote:

Maybe the specs of the 'naked' D/A chip under ideal conditions?

Ker-ching , we have a winner !
 
Many small manufacturers popular round here do this routinely, one in particular manages to take a fantastic 24 bit chip,  and when formally measured in circuit 8 bits have somehow gone missing , genius really !
 
 

 

[faileas]

There's one thing that i'm wondering about. While i don't consider myself an 'audio engineer', i have had some training in sound recording techniques and theory and there's a few things that confuse me
 
There's no such thing as 0 db - decibels have to be relative to a . There IS such a thing as 0dbfs - or 0 db of full scale , which is effectively being as loud as possible in a digital level, or an acoustic equivilent, as a measurement of sound pressure value. Common practice is to record as close to 0 DBFS it as possible - so recordings should, in theory, never actually BE at 0dbfs.
 
So, which are we talking about?
 
If if the source is well mastered, 0DBFS shouldn't be an issue, but how close to that the source goes MIGHT be. I haven't heard it yet, and it does, subjectively sound better than the maudio gear i was using before, and i don't notice any clipping with what i throw at it. I believe the e mail that nuforce sent mentioned testing it at -1 dbfs, which sounds plausible, since decibels are logarithmic.
 
Also, i'd assume someone would have *heard* the clipping by now, its a fairly obvious annoyance. There haven't been any reports of it - unless of course, no one's listened to music with DBFS levels. There's also a load of speculation over *where* the clipping is occuring. input of the USB sounds.. bizzare. headphone stage is plausible, but unless you have a VERY demanding pair, you'd never have to hike levels that high - loud for me is about 1/8th the way up, with a pair of bog standard audiotechnica M50 headphones - so i doubt that too.
 
I assume its not SPL. SPL has nothing to do with clipping.
 
Citations:
http://en.wikipedia.org/wiki/DBFS
 
http://en.wikipedia.org/wiki/Decibel#Acoustics

 

[nwavguy]

The uDAC-2 starts clipping around -1 dBFS and gets gradually worse up to 0 dBFS. And, while I agree it would be nice if recordings would only briefly, or never, touch 0 dBFS that's not reality.
 
You can pop a CD into your computer's drive, click play, and what gets sent to the DAC from a typical "loud" pop CD will include lots of clipped 0 dBFS "peaks". It's the sad reality of music mastering today. If you don't believe it, please check it out yourself. Audacity is free and you can zoom in and look at the tracks sample-by-sample even and see the signal slammed up against 0 dBFS. Here's an example, as others have said in this thread, this example is not a-typical (she's a best selling pop artist who's sold millions of CDs):
 

 
And many people use various normalization or volume leveling schemes and these often push levels right up to 0 dBFS on many (or even all if they do plain 100% normalization) tracks in a person's music collection.
 
But, as I said above, the levels don't even have to actually hit 0 dBFS before the distortion starts rising rapidly with the uDAC-2. So even a properly mastered recording will have some increased distortion with the uDAC-2.
 
I've made several high quality digital recordings of the uDAC playing 3 different songs from 3 different artists and CD's compared to other DACs and I'm currently collecting preferences from others to see just how audible the uDAC-2's clipping problem is, or isn't, in real world use. See:
 
http://www.head-fi.org/forum/thread/542968/nuforce-udac-2-listening-challenge
 
And:
 
http://www.head-fi.org/forum/thread/543279/3-headphone-dacs-listening-challenge-sequel
 
But, regardless, I can safely say I've never seen this behavior before in the 100+ digital devices I've tested. Because 0 dBFS is used as the reference level nearly all other test are referenced to (as you point out) it's something that I've run with every digital device I've ever run measurements on. And it's abnormal to find higher THD on 0 dBFS signals. There's also no valid reason for it and it's easily preventable.

 

[jasonl]

uDAC-2 is primarily target for headphone use and for mass market consumer to have the best overall listening experience. We will eventually publish a technical doc about our design criteria and approach. We plan to release a 24bit/192kHz DAC (uDAC-192 sounds like a nice model name) with RCA output only.

 

[jasonl]

This is an interesting offer, but if I understand correctly, NuForce is asking people to buy two DAC's, when you may not be able to return the other one or have a shorter time to do so.  If you have only 30 days (or no chance at all) to return the other DAC you must purchase to qualify for this offer, wouldn't most people return the one they can still get a refund on--the uDAC-2?
 
So if you really only have 30 days to return the other DAC you must buy, what good is 6 months if you have to make a choice in 30 days? I'm sorry, but like several other things NuForce has said, and done, this doesn't make any sense to me.
 

Well, it is pretty easy to understand.  There are many DACs on the market.
We standby our product's performance. This is a vote of confident, different from damage control.
You can buy uDAC-2 or you already own uDAC-2.  Go buy the <whatever-dac>, compare with all your favorite musics, not just a few test tracks. listen to it for 30 days, compare with uDAC-2.
Return this <whatever-dac> within its 30 day policy. Well, uDAC-2 win if you do that. If not, return uDAC-2.
Buy another <whatever-dac> and repeat the cycle again in the next month.  Eventually you will find a DAC that you like. 
We received many unsolicited feedback from customers telling us uDAC-2 sounded better than their existing <fill-in-the-blank> dac and headphone amps. We are just a small outfit operations but we are proud of our product's sound quality.
You are too arrogant to be able to accept other people's point of view that's why you can't see this simple logic. It is still our opinion that uDAC-2 is the best sounding headphone amp and DAC on the market at this price range. 
Perhaps it is time for you to disclose who you work for in your day job as an audio engineer. You seems to have an agenda specifically targeting us, from your very first email to us that you seem rather hostile. We have always be truthful and polite.  We even told you that we have to go borrow an AP from another company to do the testing on a Friday evening and that our test lab is in Taiwan. But instead you started to present a different picture that we have no proper test equipment.  Perhaps we are not as resourceful as many big companies with multiple $20,000 AP lying around, at least we never give you the bs.

 

[faileas]

I'm familiar with audacity - i used to use it (and in many ways prefer it to) protools . The lady gaga track looks a lot like my early attempts to convert tape to mp3 using it (which is frankly terrifying) - i'm not convinced ANYTHING could make that sound good, and its a example of how not to mix things, least from the school of thought that whoever wrote the course i did was from. My questions are heavily to do with methodology - while i work out how to do DBT with my gear - I've not heard any distortion with mine, and high computer volume + low DAC knob volume and vice versa ,  dosen't reveal clipping or anything i'd consider out of place - at identical subjective volumes, they sound pretty much the same to me. Maybe its my ears at fault, but i'm having trouble hearing the issue.
 
I'm not convinced playing music that's been recorded through another bit of gear, then played back through what i have would be a clear guide to sound quality though - would some test with an oscilloscope - even a cheap and cheerful DSO nano should be able to to have enough range for basic tests - be a clearer guide? No exotic equipment needed, and its a cheap and common enough bit of kit that anyone can check the results for themselves. Given a methodology, i'd be glad to do a test run on my udac 2 (which i admittedly love to bits) as well as my two mboxen. Anyone who thinks they want to try the same can do it themselves too, in a fair and even way, with gear that anyone can test and share.
 
 

 

[nwavguy]

@faileas We're in agreement about recordings and an oscilloscope would not be any help here. Most have only 8 bit A/D converters in them which gives them very poor resolution and dynamic range. Even a 10 bit or 12 bit scope can't begin to capture subtle audio details. And the input amplifiers in 'scopes are optimized for wide bandwidth well into the tens or hundreds of Megahertz rather than for low noise--it's difficult to have both. That's why using the 24 bit A/D converters in a professional device like the Benchmark ADC1 is a much better strategy. If you really want to look at a 'scope like waveform, you can use Audacity to look at the recordings I've made with the ADC1. Audacity plots voltage over time just like a 'scope does but it's captured with much higher resolution than virtually any 'scope can manage.
 
I'm not crazy about Lady Gaga recordings either, but some must be. She's sold millions of CD's, is in the Top 100 on Amazon, and had over a billion video views on YouTube.
 
Tests have been done where a Benchmark ADC1 and DAC1 pair have been inserted into fully analog and digital sourced high-end audio signal paths. And without going into too much detail, even hardcore audiophiles couldn't tell when they were in the signal chain or not regardless of the source material. And we're talking about $20,000+ systems. So if the ADC1 is transparent enough to pass that test, I think it's likely transparent enough to expose any audible differences between a $30 DAC and a $1600 DAC. But, I expected some will argue, as you have, it's somehow masking something important despite the fact (I'm not exaggerating here) it has better measured performance than most studio A/D's used to make the recordings we're listening to. If you'd like, I can point you to some professional audio reviews.

 

[nwavguy]

Quote:

You are too arrogant to be able to accept other people's point of view that's why you can't see this simple logic. It is still our opinion that uDAC-2 is the best sounding headphone amp and DAC on the market at this price range. 
Perhaps it is time for you to disclose who you work for in your day job as an audio engineer. You seems to have an agenda specifically targeting us, from your very first email to us that you seem rather hostile. We have always be truthful and polite.  We even told you that we have to go borrow an AP from another company to do the testing on a Friday evening and that our test lab is in Taiwan. But instead you started to present a different picture that we have no proper test equipment.  Perhaps we are not as resourceful as many big companies with multiple $20,000 AP lying around, at least we never give you the bs.

 
I take offense to being called "too arrogant" when I've been trying to stick to the facts--i.e. the measurements, what NuForce has said about the measurements, etc. My issues are with how the product performs, and trying to make sense as an engineer, of NuForce's response. I have, in fact, complimented both you and Casey for being helpful and prompt. I've also said multiple times NuForce has an excellent reputation for customer service. And I believe that.
 
I don't have any agenda specifically targeting NuForce in some malicious way. If I did, I would not have complimented the uDAC-2 in several areas and said good things about NuForce's reputation and customer service. I've also tried to be clear that some of the flaws I've found may or may not be audible. I have complimented the uDAC-2 for having impressively low noise, flat frequency response and good square wave performance. I also said it seems very well made and is nicely packaged.
 
I also would not have put aside my usual objective measurements in favor of listening tests so anyone can judge the sound of the uDAC-2 for themselves against two other DACs--without having to buy, or return, anything. You say above:
 
"It is still our opinion that uDAC-2 is the best sounding headphone amp and DAC on the market at this price range"
 
If the above is true, perhaps those listening to the accurately recorded output of each DAC will come to a similar conclusion? I even went so far as letting people compare the DAC's to each other rather than the original music so they were free to pick the DAC that sounded most pleasing.
 
If I were biased against NuForce, I would have set up the listening tests to compare each DAC against the original music not each other. That way the most accurate (and likely best measuring) DAC would have the best chance. But the way I set up the test (especially the second one), even a DAC with "euphonic" distortion could end up being preferred over a more accurate DAC. And I have done the tests in such a way they can be reproduced and verified by others if anyone doubts their validity.
 
I'm curious myself what the listening tests may reveal. I honestly don't know what the final outcome will be. So far the results have been rather mixed and I don't think there's any strong preference for or against the NuForce uDAC-2.
 
I have also said, that if NuForce comes out with a uDAC-3 (or whatever) and it measures great, I'll be happy to post a great review on my blog and recommend the product. I would like nothing more than to see the problems with the uDAC-2 get fixed as I think that would be better for everyone--including NuForce. I like the idea of an audiophile-grade DAC for $129 and the uDAC-2 could be such a product with four relatively easy, minor, inexpensive fixes.
 
I have said this before, but I do not currently work for, or even have a connection to, any DAC manufacture or competitor of NuForce in any way. If I did, I wouldn't be suggesting people check out FiiO, Firestone, Leckerton, HRT, Behringer, etc. And I certainly don't work for Behringer in Germany. So if I work for some other DAC or audiophile manufacture, why would I give a $29 Behringer product such a good review? I really don't care what company makes the products I evaluate. I just favor the best products for the price.
 
I just bought a uDAC-2 to use with my new HP PC, measured it, found some rather surprising problems, and was even more surprised when you explained it was intentionally designed that way. Many others here have shared my view it's hard to understand how NuForce considers the flaws justified when they're so easily preventable.
 
I am all about the facts. Like the fact NuForce had misleading specifications for the uDAC-2 on their website, and after my review, has since downgraded those specifications by a large amount. If pointing out such things to those shopping for a DAC makes me seem "arrogant" I'm sorry. But facts are facts.
 

 

[SLCanhead]

 
Quote:

 
I take offense to being called "too arrogant" when I've been trying to stick to the facts--i.e. the measurements, what NuForce has said about the measurements, etc. My issues are with how the product performs, and trying to make sense as an engineer, of NuForce's response. I have, in fact, complimented both you and Casey for being helpful and prompt. I've also said multiple times NuForce has an excellent reputation for customer service. And I believe that.
 
I don't have any agenda specifically targeting NuForce in some malicious way. If I did, I would not have complimented the uDAC-2 in several areas and said good things about NuForce's reputation and customer service. I've also tried to be clear that some of the flaws I've found may or may not be audible. I have complimented the uDAC-2 for having impressively low noise, flat frequency response and good square wave performance. I also said it seems very well made and is nicely packaged.
 
I also would not have put aside my usual objective measurements in favor of listening tests so anyone can judge the sound of the uDAC-2 for themselves against two other DACs--without having to buy, or return, anything. You say above:
 
"It is still our opinion that uDAC-2 is the best sounding headphone amp and DAC on the market at this price range"
 
If the above is true, perhaps those listening to the accurately recorded output of each DAC will come to a similar conclusion? I even went so far as letting people compare the DAC's to each other rather than the original music so they were free to pick the DAC that sounded most pleasing.
 
If I were biased against NuForce, I would have set up the listening tests to compare each DAC against the original music not each other. That way the most accurate (and likely best measuring) DAC would have the best chance. But the way I set up the test (especially the second one), even a DAC with "euphonic" distortion could end up being preferred over a more accurate DAC. And I have done the tests in such a way they can be reproduced and verified by others if anyone doubts their validity.
 
I'm curious myself what the listening tests may reveal. I honestly don't know what the final outcome will be. So far the results have been rather mixed and I don't think there's any strong preference for or against the NuForce uDAC-2.
 
I have also said, that if NuForce comes out with a uDAC-3 (or whatever) and it measures great, I'll be happy to post a great review on my blog and recommend the product. I would like nothing more than to see the problems with the uDAC-2 get fixed as I think that would be better for everyone--including NuForce. I like the idea of an audiophile-grade DAC for $129 and the uDAC-2 could be such a product with four relatively easy, minor, inexpensive fixes.
 
I have said this before, but I do not currently work for, or even have a connection to, any DAC manufacture or competitor of NuForce in any way. If I did, I wouldn't be suggesting people check out FiiO, Firestone, Leckerton, HRT, Behringer, etc. And I certainly don't work for Behringer in Germany. So if I work for some other DAC or audiophile manufacture, why would I give a $29 Behringer product such a good review? I really don't care what company makes the products I evaluate. I just favor the best products for the price.
 
I just bought a uDAC-2 to use with my new HP PC, measured it, found some rather surprising problems, and was even more surprised when you explained it was intentionally designed that way. Many others here have shared my view it's hard to understand how NuForce considers the flaws justified when they're so easily preventable.
 
I am all about the facts. Like the fact NuForce had misleading specifications for the uDAC-2 on their website, and after my review, has since downgraded those specifications by a large amount. If pointing out such things to those shopping for a DAC makes me seem "arrogant" I'm sorry. But facts are facts.
 

I don't like getting involved in these somewhat contentious discussions, but in this case I feel inclined to - so I will. I have read of  enough your posts, up to now (not through every specific detail about readings and such, but enough to get the gist of the overall points) to see that your positive comments are getting buried amongst some abrasive comments. Why wouldn't Jason go on the defensive, when you take some of the tone, and use certain choices of words that you do? 
 
Your first thread is titled: "UDAC-2 drama". If that isn't inititating negative confrontation, then I don't know what is.
 
I will say Jason is not all in the right. This seems to be a flawed product that could use some changes. He could temper his defensiveness (with your cooperation) and try (in turn) to work with you in reviewing these findings, to better develop his product for the future. Hopefully, the UDAC-3 (if there ever is one) could be a launching point for some of these findings. I am sure it's not easy for him to just "jump" and redesign/manufacture/release this product, on a whim, though. Also, I don't expect him to sell his product down the river ("I admit my product is not the best...".) Let's be realistic.
 
The product isn't bad (I do have one for the record) - and it's not that expensive. No, it's not 29 dollars, but I don't feel ripped off, in the least. It has the foundational elements of what I consider to be good sound, at this pricepoint. And yes, I have owned better gear in the past...and listened to very expensive gear, as well. From my memory, there has been more expensive gear that has sounded worse than this device (once again, IMO.) I do see some of the flaws, though, and I can see that your evidence thoroughly points those out.
 
Bottom line: In my opinion, you have sounded a bit arrogant and abrasive at times (I think abrasive is probably more appropriate.) Be offended if you wish. An example is when you gave Jason a hard time when he tried to provide a "6 month satisfaction" offer. He was trying to put his best foot forward, whether you liked the solution, or otherwise. I am not going to pick apart every detail at your request - the point is look at the whole of the threads (and their overall tone, at times.) I think you can be just as productive/more productive, if you take your tone down a notch, while still providing what I see to be productive/healthy discussion between a versed Engineer (yourself) and the manufaturer's representative. You are providing what could be a great service to potential customers and manufacturer.
 
You can respond to me how you like (shoot away), but I am calling it like I see it. 

 

[nwavguy]

@SLCanhead I appreciate your input. I'd probably use the word "confrontational" rather than abrasive as I keep trying to focus the discussion on the disconnect between what NuForce has said, and the factual reality. But I can agree with much of what you said. Sometimes I can be too blunt for some tastes.
 
If I've been unreasonably abrasive, or arrogant, I apologize. I honestly was done posting in this particular thread for now as I don't have much more to say. But NuForce came along and decided to attack me personally and try and accuse me of working for a competitor. I felt obligated to at least make it clear that's not the case.
 
To me the "drama" centers around how NuForce responded to all this. If they had responded in a way that made sense from a technical perspective (and there could have been a few ways they might have done so), I would have just reported on the numbers and left it at that. But, in my opinion, they've staked out a rather shaky position that made things even worse. And when I try to point out why, they attack me, or fall back on various versions of "we favor sound over measurements", instead of providing a response that makes sense from a technical perspective (which I admit is my main perspective and also the root of my original complaints).
 
With the NuForce 6 month trial, I was only trying to point out the competing product you must buy likely has a 30 day (or possibly no) return policy. So the 6 months isn't of much use to most people. For the vast majority of people shopping for a DAC, it's essentially the same as the 30 day return policy NuForce, and their dealers, already offer. So I don't see a lot of value in it.

 

[SLCanhead]

Your points are well taken and (your) humility is appreciated. I think confrontational is probably a better word to describe it, as well. I am not one to throw stones, as I am sure I have done the same. You are also (certainly) allowed to disagree/debate, including the 6 mo "guarantee" offer, but it was just getting a bit heated IMO (the thread in general.) It's a touchy arena to begin with (analyzing a piece of equipment to this degree), so I am not hoping for peaches and cream, just a little more understanding, from all sides.
 
I want to emphasize the fact that the time/effort/expertise you have put into this situation is really appreciated - I am sure by many. This is supposed to be a forum where we make informed decisions and should be able to analyze/compare the equipment we are purchasing, in order to decide what is best for our own needs. I do tend to be from the group that is more along the lines of "what sounds good to me...", to a point (I do pay some attention to parts used, specs, etc), but this info does open my eyes.
 
Let me get out of the way of this thread (stepping to the side)...

 

[jasonl]

Ok, lets just stop the drama and I hope that you can revise your review to remove those "drama" related chronological description on your blog.
We sometimes make mistakes and readily admit it, such as the early production of uDAC-2 (about 200 units were affected if I recall and all have been fixed). We don't have a PR firm to shield us and always truthful to customers. I took offense that you use such info to create drama and tarnish our reputation. Every company encounter problems with products. We tell it like it is and customers are very loyal.
 
One of our senior engineer has written a very long technical article about uDAC-2 and I will post it here soon. He is well qualified to discuss the merits of your measurement and our design approach because not only that he has been designing speakers for many years, he was also a senior test engineer for audio and power products for more than 10 years. Many customers here have talked to Bob Smith and you know he is very knowledgeable.  There are many real world practical reasons uDAC-2 is designed that way. If we have to do it again, we won't change it.
Our 24bit/192kHz USB DAC (yes, this will be the purist DAC) will have different design approach as uDAC-2.  Similarly, our higher end Icon HDP took a different design approach than uDAC-2.
 
 

 

[blockhead]

I think the fact jasonl has been willing to respond and discuss this shows NuForce in a really positive light. I can see how this could come across as a bit of a campaign against NuForce but I don't feel that is the case. 
 
@nwavguy Thanks a lot for all the effort you've put into testing, really enjoying your blog. I would really like to see some more DACs around this price range tested. I've seen you making inquiries about Audio-GD's lower models which would be great. The Hotaudio (Ecstasy, Wow) and the Audinst HUD-MX1 are a couple I would love to see some numbers on, free-time and your ability to get your hands on them considered of course.

 

[Currawong]

I just noticed this thread after someone reported a post.  Interesting stuff, but bear in mind one thing: the uDAC 2 is a $100 USB-powered DAC/amp, not a $1000 DAC with extensive power filtering, a dedicated amp and whatever else built in.  There are physical limits to what can be achieve with electronics and keep the price low. Expecting cheap gear to measure like expensive gear and you'll just be disappointed. While there is cheap audio gear that, on the surface (IE: if you use just RMAA) that appears to measure well, it very often sounds terrible.  It's quite funny for me to read months of subjective impressions of a product and even use it myself, then see someone post measurements that just confirm what I and others heard when we used it (though in that statement, I'm not referring to the uDAC, as I've never owned one).
 
My attitude, which I suggest others adopt it: I listen to/with something, gain some impressions, then if I or others measure anything, it's to discover what the cause is of those impressions. With headphones it might be the frequency response and other measurements, such as HeadRoom does. With DACs and amps it will relate to the ability to drive headphones, the quality of the power supply and measurements that demonstrate those things (not just RMAA results from unspecified equipment, again not referring to this thread but a popular trend lately with amateur scientists on the forums).  From the measurements, applied properly, one can match gear better to achieve better results, such as advice often given about amps, output power vs. impedance and headphones.  That I think would be very beneficial to Head-Fi.
 
I would say that, in general, with a proper understanding and realistic outlook, you'll find that as you go higher up the range measuring gear from any manufacturer, you'll find the distortion goes down, and the ability to reproduce signals (such as the square wave test on AMB's Beta 22 pages) improves.  Doing a comprehensive set of measurements, as long as allows for use of that equipment connected to a wide (widely electrically different) variety of gear and, importantly, can be translated to what people perceive when they listen with that gear, you will build something useful.  At least one manufacturer posted that they can tell what a circuit or piece of equipment is going to sound like by the shape of a square wave coming from its analogue outputs, so I don't see why we shouldn't learn to do the same.  Using simple one-off measurements to make arbitrary judgements about whether a piece is "good" or "bad" will just result in rubbish, both in the form of rubbish science and the forums being rubbished with pseudo-knowledge, causing them to degenerate into pointless arguments.