geneltd
New Head-Fier
- Joined
- Aug 31, 2010
- Posts
- 13
- Likes
- 0
Introduction:
After I have learned to use RMAA software as a non-bias tools to measure audio gear performance, I have tested my ipod touch 4 and got incredible result. What if I want to further upgrade the portable source? As the ipt 4 is so good (technically), what I can think of is the latest portable cd player. I know it is not apple to apple. But why not give it a try? A cd player is so cheap today
. Then I begin my research on pcdp. It then come up with this:
http://www.testfreaks.com/portable-cd-players/sony-d-ej011/
This is what I can buy the latest model in Hong Kong. It cost me $40 USD (8 March 2011). Things is pretty expensive in HK.
The audio/AV gears are always cheaper in US than in HK.
Testing environment:
AMD Phenom x4 940 with Asus xonar STX via Cirrus Logic CS5381, a 24-bit A/D converter chip. (Not a professional gear, but good enough for this)
1M Cable (made in japan)
Windows 7 32bit
Volume max on both devices when measure
G protection off in EJ011
Result:
sony EJ011 G off
Testing chain: External loopback (line-out - line-in)
After I have learned to use RMAA software as a non-bias tools to measure audio gear performance, I have tested my ipod touch 4 and got incredible result. What if I want to further upgrade the portable source? As the ipt 4 is so good (technically), what I can think of is the latest portable cd player. I know it is not apple to apple. But why not give it a try? A cd player is so cheap today
http://www.testfreaks.com/portable-cd-players/sony-d-ej011/
This is what I can buy the latest model in Hong Kong. It cost me $40 USD (8 March 2011). Things is pretty expensive in HK.
Testing environment:
AMD Phenom x4 940 with Asus xonar STX via Cirrus Logic CS5381, a 24-bit A/D converter chip. (Not a professional gear, but good enough for this)
1M Cable (made in japan)
Windows 7 32bit
Volume max on both devices when measure
G protection off in EJ011
Result:
sony EJ011 G off
ipt4
RightMark Audio Analyzer test
Testing chain: External loopback (line-out - line-in)
Sampling mode: 16-bit, 44 kHz
Summary (no load)
Test | sony EJ011 G off | ipt4 |
Frequency response (from 40 Hz to 15 kHz), dB: | +0.08, -0.65 | +0.01, -0.05 |
Noise level, dB (A): | -90.7 | -92.0 |
Dynamic range, dB (A): | 91.3 | 93.2 |
THD, %: | 0.257 | 0.0020 |
IMD + Noise, %: | 0.153 | 0.0065 |
Stereo crosstalk, dB: | -83.8 | -90.3 |
Summary (loaded with earphone)
Test | ipt4 loaded | sony ej011 loaded |
Frequency response (from 40 Hz to 15 kHz), dB: | +0.10, -0.09 | +0.31, -0.92 |
Noise level, dB (A): | -91.2 | -89.0 |
Dynamic range, dB (A): | 92.5 | 90.3 |
THD, %: | 0.0044 | 2.313 |
IMD + Noise, %: | 0.082 | 1.027 |
Stereo crosstalk, dB: | -65.2 | -54.0 |
Frequency response(no load)
Loaded with earphone
Noise level
Dynamic range
THD + Noise (at -3 dB FS)
Intermodulation distortion
Stereo crosstalk
[size=xx-small]This report was generated by RightMark Audio Analyzer 6.0[/size]
Interpretation:
I think you guys are better than me. But if you really don't know what it is, than look for the green line i.e.ipod touch 4 result. It almost come close to perfect.
Discussion:
Please feel free to discuss on this thread. Technically, Apple wins all in any technical aspect. It even come close to a semi professional DAC like Benchmark DAC1. But the lack of driving power hurt the ipt4 a lot. The remaining question is : should I throw the poor sony in rubbish bin? Absolutely No. It has its own advantage. I will explain later.
-------------------------------------------------------------------------------------------
faichiu
500+ Head-Fier
- Joined
- Mar 10, 2007
- Posts
- 660
- Likes
- 17
RMAA test means nothing to me. For sound quality, there's a lot of things that "Graph" cannot show.
Etc, soundstage, output power, details...........
Etc, soundstage, output power, details...........
elfary
Headphoneus Supremus
- Joined
- Sep 29, 2005
- Posts
- 1,521
- Likes
- 123
Not to undermine your effort. But measuring the Touch under no load (Not plugged earphones) is like measuring a car acceleration without starting the engine.
geneltd
New Head-Fier
- Joined
- Aug 31, 2010
- Posts
- 13
- Likes
- 0
thx, elfary. Can you teach me how to give it a load to measure?
elfary
Headphoneus Supremus
- Joined
- Sep 29, 2005
- Posts
- 1,521
- Likes
- 123
I'm not into RMAA procedures but a few people around here know the proceedings. I think that you need and y-splitter.
geneltd
New Head-Fier
- Joined
- Aug 31, 2010
- Posts
- 13
- Likes
- 0
Thx, I will get one and do it again.
geneltd
New Head-Fier
- Joined
- Aug 31, 2010
- Posts
- 13
- Likes
- 0
surprisingly, the performance of Sony drop like hell after loaded with earphone. But the ipod seems unaffected.
yuriv
100+ Head-Fier
- Joined
- Apr 30, 2008
- Posts
- 156
- Likes
- 270
Thank you for taking the time to investigate. You've confirmed something we have all suspected with the fourth generation iPod Touch. I measured the iPhone 4 and the original iPad, and got similar numbers even while driving a load. I have a few questions:
It's also good to know an amplifier's output impedance so that we can more easily predict how the amp will behave with different headphones. We can estimate the output impedance with additional measurements: How much does each player's output drop when driving the load? You can use the numbers in the RMAA level adjustment window. Just loop the calibration signal or use a different test tone. What's the difference in level between load and no load? Try this with a few headphones and report your findings. With this test, set the players' volume controls at the levels that you would use to listen to music normally. That may be different for each pair of headphones.
- With no load, how much stronger is the Discman's output? At maximum volume, what are the numbers in the RMAA level adjustment window?
- Which earphones did you use as the load? The one that came with the Discman or the iPod? Another model?
- Were the players still at the maximum volume when you retested with the load?
It's also good to know an amplifier's output impedance so that we can more easily predict how the amp will behave with different headphones. We can estimate the output impedance with additional measurements: How much does each player's output drop when driving the load? You can use the numbers in the RMAA level adjustment window. Just loop the calibration signal or use a different test tone. What's the difference in level between load and no load? Try this with a few headphones and report your findings. With this test, set the players' volume controls at the levels that you would use to listen to music normally. That may be different for each pair of headphones.
geneltd
New Head-Fier
- Joined
- Aug 31, 2010
- Posts
- 13
- Likes
- 0
Q1. at max volume, both devices are between -10 to -5 dB
Q2. original ipod eyebud
Q3. Yes, otherwise the test will fail due to low signals.
Yours suggestion is interesting. But i think the calibration signal generated by rmaa is too weak to measure/calibrate. It sounds like white noise. I cannot have any reading while play the calibation wave file. But the test itself is okay. Is it normal?
I have a voltmeter at home. Do you know how to use it to measure?
Q2. original ipod eyebud
Q3. Yes, otherwise the test will fail due to low signals.
Yours suggestion is interesting. But i think the calibration signal generated by rmaa is too weak to measure/calibrate. It sounds like white noise. I cannot have any reading while play the calibation wave file. But the test itself is okay. Is it normal?
I have a voltmeter at home. Do you know how to use it to measure?
yuriv
100+ Head-Fier
- Joined
- Apr 30, 2008
- Posts
- 156
- Likes
- 270
- They're within 5 dB of each other, so the maximum output of the Discman isn't that much hotter. The iPad and iPhone 4 can manage around 950 mVrms. I will guess that your iPod Touch is similar.
- The original iPod earbud: Use your meter to measure the DC resistance of the earphones. The three conductors on the headphone plug are termed tip, ring, and sleeve. Measure the resistance from tip to sleeve or ring to sleeve. Earbuds usually don't have crazy impedance-vs-frequency graphs, so the the impedance at 1 kHz shouldn't be too far off from the DC resistance. I'm guessing 32 ohms as a ballpark figure.
- Some sound cards (like the E-mu 1212m) will let you boost the input signal. That will let you take a good measurement in RMAA even with low signal levels from the players. I don't know if your STX will let you do it. Or you could play the test signal while being recorded by Audacity (freeware) or another software recorder, normalize the recording, then let RMAA analyze the resulting file.
But we don't even need to do all of that. We just want to know how much weaker the signal gets after you plug in the earbuds. You can use Audacity to generate a 1000 Hz sine wave. 30 seconds should be long enough for you to take measurements. What does the RMAA level adjustment window say when you're playing that tone? How much does the output drop when you plug in the earbuds in the splitter? We don't need to leave the level adjustment window for this test.
Alternatively, you can use your meter. Set the volume level to the normal listening level for the earbuds. Play the tone, record the voltage. Now plug in the earphones into the splitter. What does the meter read now? Most inexpensive meters aren't as accurate at 1000 Hz as they are at 60 Hz, but don't worry. We're interested only in the relative levels. If you're interested, you could conduct the test also at 60 Hz, just to get another data point.
These measurements will give us a rough idea of how close the player's headphone output is to an ideal voltage amplifier. The headphone amp is like the water supply to your bathroom. How much weaker is the stream from the faucet when you flush the toilet? How much is the output weakened by the additional load?
An ideal voltage amp doesn't change the pressure at all. The Sandisk Sansa Clip+ is a stiff supply--its voltage doesn't get much weaker when you plug in the supplied 16-ohm earbuds. Does the newest iPod Touch exhibit the same grace under pressure? So far, your measurements say that it's getting close.
BTW, if you have a passive inline volume control and a headphone extension cable, there's a quick and dirty way to get a more precise test load than a pair of earbuds.
geneltd
New Head-Fier
- Joined
- Aug 31, 2010
- Posts
- 13
- Likes
- 0
Thx for the useful info. especially the Audacity!! The DC resistance is 16 ohms.