|
|
Post by arthurz on Apr 4, 2013 12:21:20 GMT -5
I plotted the THD+N charts from Emotiva's 8Ω test results for the three amps. AFAIK, that should capture all the potential SQ advantages of class A over class AB (am I missing something?). It seems that: - XPA-1 beats XPA-1L. - None of the differences should be audible. There are no XPR-1 test results to be found anywhere. I remember a few months ago Emotiva was waiting for some new test equipment, but that came and we never saw any results. Would be good to know! Attachments:
|
|
|
|
Post by garbulky on Apr 4, 2013 12:30:57 GMT -5
Absolutely! And that would be...listening  Most of the emotiva amps measure so low in THD that you could use the same argument for the u-series amps as well. (Inaudiblity). However, there are some things to note (if you ignore inaudibility). There are SNR figures on the XPA-100's that don't stack up that well on their 4 ohm tests compared to the xpa-1 vs xpa-1 L. I wish they had some way to measure and compare reactance to a continuously changing load. Edit: I have NO idea how you managed to do integrate the graphs. I'm obviously way behind on my computer literacy. Good job. It would be cool if emo followed your steps and did that for all their graphs so people can compare similar amps. |
|
|
|
Post by arthurz on Apr 4, 2013 22:59:17 GMT -5
Listening to the podcast. They just said they expect it to become a cult amp. That MUST mean it should sound better than the XPA-1. So how am I wrong? Someone please explain...
|
|
KeithL
Administrator 
Posts: 10,261
|
Post by KeithL on Apr 5, 2013 1:14:25 GMT -5
You're kind of missing the fact that THD and S/N are NOT the only things that affect the way an amp sounds. Each of us is sensitive to different things; one of us may be really annoyed if they can hear ANY background noise, while another one doesn't mind a little hiss but can't stand distortion, and some people prefer one form of distortion over another. The THD numbers on most of our amps are indeed so low that to most people the THD is totally inaudible. To those people, all the amps will sound the same in that regard. However, those people might notice that an XPR has a bit more power than an XPA-1L, or that a UPA is slightly quieter  Class A amps make different sounding distortion than Class A/B amps (the distortion spectra are different). This means that, to the people who are sensitive enough to hear those tiny amounts of distortion produced by any of our amps, they sound different. Some people may hear the difference, while others don't; likewise, some may hear the difference, but not find it important to them. Pretty much everybody who CAN hear the difference prefers the way Class A amps sound. [But you can't "read" this difference on a THD graph.] We prefer NOT to encourage people to compare the number with a microscope because, well, the numbers don't tell the whole story anyway. (We'd also rather spend the time developing audio equipment instead of writing fancy JAVA apps for overlaying graphs.) I plotted the THD+N charts from Emotiva's 8Ω test results for the three amps. AFAIK, that should capture all the potential SQ advantages of class A over class AB (am I missing something?). It seems that: - XPA-1 beats XPA-1L. - None of the differences should be audible. There are no XPR-1 test results to be found anywhere. I remember a few months ago Emotiva was waiting for some new test equipment, but that came and we never saw any results. Would be good to know! |
|
|
|
Post by bolle on Apr 5, 2013 3:40:15 GMT -5
We'd also rather spend the time developing audio equipment instead of writing fancy JAVA apps Well but as the XPA-1L FW bug shows, or also the UMC-1 and UMC-200 quirks, there might be something to be learned from "fancy JAVA apps" which even helps in "developing audio equipment". It is getting more and more software today... Btw. thanks for the graph! |
|
thor48
Seeker Of Truth
M. The most powerful letter in the world !
Posts: 8
|
Post by thor48 on Apr 5, 2013 4:16:13 GMT -5
Numbers are good..........but a system connected with the right stuff is what counts. One can have a mega low THD/SN/watt/cost Amp/Preamp but with the wrong connections it may as well be a transistor radio (no offense to any transistor-radio-philes). I am very content with my Amps and Preamp (and the connections).
|
|
|
|
Post by dima333a on Apr 5, 2013 7:50:15 GMT -5
One of the problems in audio industry is the fact, that generally accepted and commonly used variables to represent parameters of amplifiers do not have much meaning any more. Almost every amplifier has harmonic distortions below sensible level, with frequency response as flat as it can be, and not appreciable noise level. Yet, amplifiers do sound differently. For some reason, the graph of THD vs. frequency is not published for every amp. The interlobular distortions (which are more annoying) are not given in the specs. The power output cube (Power-phase- impedance) is not shown either.
Essentially, the data about amplifiers, as it is commonly provided, does not allow to make much of judgement.
|
|
|
|
Post by kellys on Apr 5, 2013 12:51:09 GMT -5
Notwithstanding the comments by Kieth and others, I like your graphs. I would be particularly interested in seeing the four ohm version too. The XPA-1L looks like it is even better into a four ohm load.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 5, 2013 13:08:39 GMT -5
It's in your head.
They all sound the same. Some play louder then others.
A flat frequency response is just that, if it sounds different, then this response is not flat anymore.
|
|
|
|
Post by garbulky on Apr 5, 2013 13:29:11 GMT -5
Kraemer: the flat frequency response is usually represented using a fixed load on tests AFAIK. A speaker is not a fixed load and it's resistance changes over the frequency band. Also, music is very rarely in the form of a 20-20khz sweep or fixed hz tones. It usually has multiple peaks over multiple frequencies all at the same time.
To me, what a flat frequency response means is that the amp is able to reach that frequency response and is a GOOD but not perfect indication that it should indeed play a neutral sound through the speakers.
My asus xonar (agreed it;'s not an amp) has a flat frequency response just like the xda-1. But it sounds different though still very good. It also has very high SNR ratings, much better than the xda-1, but it still sounds different and I use the XDA-1 exclusively for my music listening.
|
|
|
|
Post by garbulky on Apr 5, 2013 13:38:27 GMT -5
If you read the article I linked to in the carver thread:
"A transfer function is nothing more than a statement of the relationship between the signal fed into a device and the signal that comes out of it. For example, a frequency-response specification is a description of the transfer function telling us how much an input signal of fixed amplitude and varying frequency will vary in amplitude at the output.
Bob's test hookup would show much more than frequency response differences. In fact, one of his most interesting statements, for those of the "every amplifier is the same except for frequency response" school, was that varying frequency response between the 1.0 and the reference amp made up only about 25% of the significant differences. Relative phase shift, source impedances (damping factors)—in short, every electrical difference between the amplifiers—would produce a signal at that test point between the Plus output terminals. When the amplifier outputs were identical, in all respects, there would be total cancellation—a null—of the difference signal. Bob's goal was a 70dB null, or an 0.03% difference between the two amps."
|
|
KeithL
Administrator
Posts: 10,261
|
Post by KeithL on Apr 5, 2013 14:13:37 GMT -5
Standard amplifier tests are indeed usually performed using a resistive load. However, an amplifier with a relatively low output impedance (meaning a high damping factor) should be able to deliver a similar frequency response into loads that are far from resistive. (Tube amps, because they have relatively high output impedances, and so react more with the load, tend to have varying frequency responses into different loads.) Standard THD measurements are generally taken at a single frequency (either at one point, or one swept frequency). IM (intermodulation) distortion is the measurement of how multiple simultaneous tones interact to produce things that shouldn't be there (distortion). There are all sorts of other things that can be measured or inferred, and which affect the sound. In the old days "SID - slewing induced distortion" got a lot of mention. SID is a type of transient distortion that can occur in badly designed amplifiers, and is very difficult to measure directly. However, designs likely to exhibit it also tend to show rising THD with rising frequency.... so, if an amp doesn't show this, then we can infer that SID won't be present in significant amounts either. Likewise, an amplifier with low THD across the audio band, a wide bandwidth, and a high damping factor is likely to be flat and low in distortion into most loads. I have no personal experience with the Asus Xonar, but I can tell you that many computer sound cards simply don't meet their specs - or fail to do so in a real-world situation (where they are installed in a typical PC with its noisy power supply and noisy environment overall). My guess is that you rarely get anywhere near published specs with a computer sound card... whereas you should get what you expect with a separate piece of audio equipment - especially ours Kraemer: the flat frequency response is usually represented using a fixed load on tests AFAIK. A speaker is not a fixed load and it's resistance changes over the frequency band. Also, music is very rarely in the form of a 20-20khz sweep or fixed hz tones. It usually has multiple peaks over multiple frequencies all at the same time. To me, what a flat frequency response means is that the amp is able to reach that frequency response and is a GOOD but not perfect indication that it should indeed play a neutral sound through the speakers. My asus xonar (agreed it;'s not an amp) has a flat frequency response just like the xda-1. But it sounds different though still very good. It also has very high SNR ratings, much better than the xda-1, but it still sounds different and I use the XDA-1 exclusively for my music listening. |
|
KeithL
Administrator
Posts: 10,261
|
Post by KeithL on Apr 5, 2013 14:23:10 GMT -5
You're forgetting distortion However, on that note: If there is enough distortion that you can hear it, then different types of distortion can (and do) sound different. Obviously, if the amounts involved are inaudible to begin with, then they will sound equally inaudible. However, there is no well established standard for the line below which something is inaudible (and it certainly varies between individuals and different types of distortion.) To take an extreme example. A 50 Hz tone is rather a low hum, and second harmonic distortion on that would be at 100 Hz - it would add a slight "edginess" to the original tone, but would have to be present in significant quantities to be audible. HOWEVER, the TENTH harmonic of 50 Hz is 500 Hz (so a 50 Hz tone with significant tenth harmonic distortion would sound like a low hum plus an annoying whistle - and the annoying whistle would be audible at a quite low level - much lower than the level at which the 100 Hz second harmonic would NOT be audible). So, playing a 50 Hz note, an amplifier with 0.5% THD (but all second harmonic) would sound much more pleasant than one with 0.5% TENTH harmonic distortion. It's in your head. They all sound the same. Some play louder then others. A flat frequency response is just that, if it sounds different, then this response is not flat anymore. |
|
|
|
Post by garbulky on Apr 5, 2013 16:56:21 GMT -5
I feel like I just got schooled Thanks for the info Keith! I was wondering what IMD was... The difference in audibility in harmonics is also very interesting. |
|
|
|
Post by arthurz on Apr 5, 2013 22:13:04 GMT -5
Class A amps make different sounding distortion than Class A/B amps (the distortion spectra are different). This means that, to the people who are sensitive enough to hear those tiny amounts of distortion produced by any of our amps, they sound different. Some people may hear the difference, while others don't; likewise, some may hear the difference, but not find it important to them. Pretty much everybody who CAN hear the difference prefers the way Class A amps sound. [But you can't "read" this difference on a THD graph.] OK, this makes sense. Unfortunately EE is not my field, so I probably can't state my case very well – intuitively I find it hard to believe that there's no way to put a number on this . I totally appreciate that numbers are just abstractions, but that doesn't mean they're not useful. For example, could you not test with a set of tones of unrelated frequencies, get a frequency domain chart, throw out anything 100 dB below the peaks, and group the remaining signals into the following categories: (1) original signal, (2) early and other musically-related harmonics, and (3) other distortion (including late harmonics that sound out-of-tune)? Then plot everything! OK, now I'm getting excited  . Why not embarrass your competition while you're at it (and all those stereo fetish rags), and include them as well? Please take into account I mean for the above to be just an example – you as an expert can surely come up with far superior ways of showing off and comparing your wares. We prefer NOT to encourage people to compare the number with a microscope because, well, the numbers don't tell the whole story anyway. (We'd also rather spend the time developing audio equipment instead of writing fancy JAVA apps for overlaying graphs.) I have to respectfully disagree. This just means you don't have enough numbers. I don't think you need a Java app to show a meaningful comparison. You could start with a combined chart of the sort I put together, but hopefully showing more than just THD+N. It would be great to see a detailed product comparison chart based on consistent criteria. For example, it seems to be the standard industry practice to show max power at THD == 0.1% (except for tube amps...), but on the XPA-100 page all I see is "THD < 1%". That's odd! The comparison chart of the sort I'm imagining would allow you to click/hover over a category, such as "Gain", "SNR", "Input Sensitivity", etc., and get an idea of what it means, what range is acceptable, etc. Ideally, you could add videos explaining things in more detail. Even if none of this makes technical sense, I'm nearly certain you'd see results in terms of increased sales, so maybe treat it as a marketing exercise. I think this type of approach is consistent with your openness. I would never expect anything like this from the likes of Marantz. The MAIN reason I didn't dismiss you guys when I first came across your website last November was that I saw numbers and tests results EVERYWHERE. At the time, I was dead set on buying MC452 or 2x601 and their C50 preamp for my system. Your prices were so much lower, I was in disbelief, and was certain you guys were just some random fly-by-night operation. But then the numbers started talking! |
|
|
|
Post by creimes on Apr 6, 2013 1:27:00 GMT -5
Man I am glad I'm not smart enough to know what all these #'s mean....lol I just enjoy listening to the music...  sometimes people get wayyy too technical haha |
|
|
|
Post by pedrocols on Apr 6, 2013 1:48:11 GMT -5
The least thing I am thinking when listening to music is numbers.....maybe one day I learn to play the harmonic... ;D
|
|
|
|
Post by AudioHTIT on Apr 7, 2013 17:37:06 GMT -5
The least thing I am thinking when listening to music is numbers.....maybe one day I learn to play the harmonic... ;D Maybe while you're surfing a standing wave. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 8, 2013 0:17:56 GMT -5
The least thing I am thinking when listening to music is numbers.....maybe one day I learn to play the harmonic... ;D Obviously you've never listened to Revolution 9 on the Beatles' White Album. |
|
|
|
Post by Tungx2 on Oct 13, 2015 12:00:44 GMT -5
We'd also rather spend the time developing audio equipment instead of writing fancy JAVA apps Well but as the XPA-1L FW bug shows, or also the UMC-1 and UMC-200 quirks, there might be something to be learned from "fancy JAVA apps" which even helps in "developing audio equipment". It is getting more and more software today... Btw. thanks for the graph! There was a XPA-1L firmware bug before? |
|
. Why not embarrass your competition while you're at it (and all those stereo fetish rags), and include them as well?
sometimes people get wayyy too technical haha
