I am looking over a lot of audio card and my question is about the SNR; does it really make the big different in price between 116dB and 124dB? Or does it really make a different in sound quality?
SNR is generally not an important performance measure for audio systems these days because at a reasonable signal level, the noise level of the gear is going to be below that of ambient noise anyway, so you wouldn't hear the difference. Other factors and performance specs play a larger part. Also, the electronics these days play an even lesser role than seems intuitively right for many people, because that part can be done so well for cheap.
That said, manufacturers often lie about specs. Or at least, they are misleading; if they say they are using a chip with an SNR of 120 dB, other parts of the circuit may contribute noise and make the SNR of the entire system more like 100 dB. If a sound card is picking up EMI from the rest of the computer (as opposed to their evaluation board in a lab in pristine conditions), that can be degraded further in practice. Furthermore, they are often inconsistent about which signal level they are using for the SNR quote. It's a ratio. Sometimes if you crank up the signal level and the noise does not scale completely with the signal level, you can record a higher SNR.
The simple answer is it does. For every 3db of SNR, it makes a whole lot of difference, so something that has 124 SNR is going to be a whole lot better than the 116 SNR.
However.....It also depends on your headphones. If you are getting really cheap and poor quality headphones the SNR isn't going to matter much because you are going to have a harder time hearing the difference. As soon as you start looking at headphones like audio technica, AKG, Sennheiser, grado, ultrasone, and similar companies with headphones that are about 150 and up, then you start to notice the difference between the two cards (or at least I can). If you are looking at headphones that are 300 plus, then you shouldn't even bother with a 116 SNR sound card.
Lastly, once you hit the 500 dollar headphones price point, it is time to get yourself a dedicated USB DAC and a tube amplifier.
So in a sense it is going to come down to which part will bottle neck the other.
SNR is generally not an important performance measure for audio systems these days because at a reasonable signal level, the noise level of the gear is going to be below that of ambient noise anyway, so you wouldn't hear the difference. Other factors and performance specs play a larger part. Also, the electronics these days play an even lesser role than seems intuitively right for many people, because that part can be done so well for cheap.
That said, manufacturers often lie about specs. Or at least, they are misleading; if they say they are using a chip with an SNR of 120 dB, other parts of the circuit may contribute noise and make the SNR of the entire system more like 100 dB. If a sound card is picking up EMI from the rest of the computer (as opposed to their evaluation board in a lab in pristine conditions), that can be degraded further in practice. Furthermore, they are often inconsistent about which signal level they are using for the SNR quote. It's a ratio. Sometimes if you crank up the signal level and the noise does not scale completely with the signal level, you can record a higher SNR.
That first part is wrong. the higher the SNR, the clearer the sound. You just need to make sure you have a good enough pair of headphones that can take advantage of the higher SNR.
The second part can be true. This is why a dac amp combo with a good pair of headphones is actually your best bet so long as you are using high quality parts. The DAC takes a very clean digital sound input and converts it into analog. Then the amp will amplify the sound without adding any noise to it. Depending on the op amps you use, you can also get a much better experience than with any sound card on the market.
The simple answer is it does. For every 3db of SNR, it makes a whole lot of difference, so something that has 124 SNR is going to be a whole lot better than the 116 SNR.
Maybe in comparison to each other, but in a real life scenario, where a 16-bit record has the quantization noise of -93dBFS, both of these systems would be audibly transparent to a CD-quality source. If you run your DAC without digital volume control, and deliver the full scale signal, you should get the reported dynamic range/noise floor. Enormous noise floors only matter if you run a digitally attenuated signal, which you then amplify with a pre-amp, which is a silly practice anyway.
So no, the SNR in modern gear doesn't really matter in conventional listening scenarios. Furthermore, manufacturers lie. For instance, the measurement dBA means frequency-weighted sound pressure, which means they're omitting something from the measurement. What is there to hide? Second of all, as stated earlier in lab conditions, EMI/RFI free environments with lab-quality power supplies, gear can work a lot better (implying a bad PSRR in the developed circuit, if it only works with super high quality PSUs). Furthermore, sometimes manufacturers only state the performance of the CHIP (op-amp, or DA-chip, not the whole circuitry), which can be completely misleading in relation to the whole circuitry and it's design choices.
Don't go by manufacturer specifications, and don't think these humongous signal-noise ratios matter at all. Spend your money elsewhere.
I have asus Phoebus sound card and I just ordered Sennheiser G4me Zero headphones which is 150ohm. Do you think I should replace my Phoebus with Creative SoundBlaster ZxR? Phoebus is 116dB and ZxR is 124dB. By the way I managed to get rid of the Phoebus' EMI noise issue completely. (I'm a CS:GO competitive player) Your thought will be effecting my decision literally.
Signal to noise issue isn't so much hearing the "noise" as it is the noise signal added to the sound being reproduced by the amplifier and the speakers.
The amp and speaker try to reproduce the sound signal, which is extremely complex in the first place. with the added complexity of the noise added in.
How much this affects overall sound quality depends greatly on now accurately your system can play the signal presented to it. A system with poor sound quality isn't going to be effected much by noise since it can't play the subtle changes introduced by the noise anyway. While a system able to reproduce high sound quality will be greatly effected as it will reproduce a sound with the signal and the noise added together.
That's ridiculous. The transducers have no problems reproducing noise, signal, or signal plus noise. It's not like lower-end systems aren't affected by noise in the same way.
There are noiselike sounds in the signal itself, like with some percussion, that get played back just fine (with the limitations of recorded sound, mixing, imperfect transducers, etc.). Also, the noise floor of recordings from ambient sound in the studio, the mics, and any of the electronics along the way frequently dwarf the noise floor of the electronics you use for playback. Part of the "signal" the transducers will have to reproduce is noise anyway, even with a theoretical perfectly noiseless playback system.
That's ridiculous. The transducers have no problems reproducing noise, signal, or signal plus noise. It's not like lower-end systems aren't affected by noise in the same way.
There are noiselike sounds in the signal itself, like with some percussion, that get played back just fine (with the limitations of recorded sound, mixing, imperfect transducers, etc.). Also, the noise floor of recordings from ambient sound in the studio, the mics, and any of the electronics along the way frequently dwarf the noise floor of the electronics you use for playback. Part of the "signal" the transducers will have to reproduce is noise anyway, even with a theoretical perfectly noiseless playback system.
You can have whatever opinions you want, but transducers can't even reproduce clean music signals accurately. If they could all music playback would sound like the live sound that was recorded. Once noise levels get below -50db or so the "sound" ceases to be an issue. -100db is like trying to hear someone whispering while standing behind a jet airplane readying for takeoff! Noise is an electronic reproduction problem. becoming more evident as the quality of the speakers increases. If speakers won't play the difference you won't hear the difference.
That's ridiculous. The transducers have no problems reproducing noise, signal, or signal plus noise. It's not like lower-end systems aren't affected by noise in the same way.
There are noiselike sounds in the signal itself, like with some percussion, that get played back just fine (with the limitations of recorded sound, mixing, imperfect transducers, etc.). Also, the noise floor of recordings from ambient sound in the studio, the mics, and any of the electronics along the way frequently dwarf the noise floor of the electronics you use for playback. Part of the "signal" the transducers will have to reproduce is noise anyway, even with a theoretical perfectly noiseless playback system.
You can have whatever opinions you want, but transducers can't even reproduce clean music signals accurately. If they could all music playback would sound like the live sound that was recorded. Once noise levels get below -50db or so the "sound" ceases to be an issue. -100db is like trying to hear someone whispering while standing behind a jet airplane readying for takeoff! Noise is an electronic reproduction problem. becoming more evident as the quality of the speakers increases. If speakers won't play the difference you won't hear the difference.
I didn't imply a perfect or even similar level of accuracy across different speakers, just that noise is itself a signal. From an engineering and physics perspective, the distinction you draw doesn't make sense, unless I've misunderstood your position. It's not like lower-end speakers can't reproduce noise. They can do whatever desired signal you like, whatever noise, etc. and all combinations of those, just with certain and different kinds of distortion and issues. No, they aren't completely linear, and the output given signal+noise won't be quite exactly the same as the output given signal + the output given noise, but as an approximation on listenable speakers it should be close enough. The implication is that having some kind of noise or error signal riding on top of the desired signal should not make the system misbehave in strange ways.
Another main point was that other sources of noise are already in the playback system because they're in the source, and the transducer doesn't care how much noise was added by the electronics and how much encoded into the source audio data itself. And though I didn't mention it, acoustic noise in the listening environment even in a quiet room can also dominate the other contributors of noise.
Also, claiming that accurate reproduction of clean music signals would make music playback sound like live sound is disingenuous, ignorant, or maybe just forgetful. The positioning of a finite number of microphones for recording and a finite number of loudspeakers on playback, never mind whatever best illusion is created during the mixing and mastering make that an impossibility. The experiences would be different even with loudspeakers of perfect fidelity.
The typical audio amplifiers we use will produce 33-34 db gain so you aren't going to "hear" noise unless it's at least greater than - 35 db or so.
I'm just trying to point out that hearing the noise isn't the issue, it's just the added complexity of the amplified signal sent to the speaker(s) that is the issue.
I think we're both on the same wavelength!
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Related Threads
?
?
?
?
?
Ask a question
Ask a question
Overclock.net
27.8M posts
541.2K members
Since 2004
A forum community dedicated to overclocking enthusiasts and testing the limits of computing. Come join the discussion about computing, builds, collections, displays, models, styles, scales, specifications, reviews, accessories, classifieds, and more!