How do I measure the quality of a pre-amp?

I

ironpony

Carbonite Member
So when it comes to audio I am guessing analogically, that a pre-amp is kind of like a sensor, for a field recorder, if compared to a camera. You can measure the quality of a sensor by it's size. The bigger the sensor, usually the bigger the quality. But is there any way of measuring pre-amp quality?
 
OK, here goes... :-)

Not owning or even wanting to own the lab equipment to test pre-amps, here is how I test pre-amps in low cost cost recorders and cameras.

THE GEAR

1- I use a sound devices mixer to output a 1 kHz test-tone

This is a high quality piece of audio gear that has a minimum dynamic range of 110 dB, so it's very unlikely to generate much audio noise when the test-tone is switched off.

2- I use an ART SPLITMix4 4-Channel Passive Splitter/Mixer

This is a low cost passive mixer with a dynamic range of 90+ dB. I really like passive audio gear because it's cheap and often transmits very good audio signal.


THE TEST

1- Connect the Sound Devices mixer to the recorder / camera MIC or LINE input.

2- Switch on the Sound Devices test-tone, and adjust the recorder / camera pre-amp gain to record the test-tone signal at -12 dB.

3- Record 10 seconds of the test-tone, then switch the test-tone off and continue to record for another 10 seconds.

4- Connect the Sound Devices Mixer to the ART passive mixer and then to your recorder / camera MIC or LINE input.

5- Raise the GAIN on the pre-amp of your RECORDER / CAMERA, then use the ART passive mixer to bring the mixer test-tone back to a recording level of -12 dB.

6- Record 10 seconds of the test-tone, then switch the test-tone off and continue to record for another 10 seconds.

7- Repeat steps 5 and 6 until you have exhausted the GAIN range of your RECORDER / CAMERA. ( the recorder or camera pre-amp GAIN is at the maximum level )


CHECKING THE RESULTS

1- Load the recordings you made into your video or audio editor.

2- Play the recording and see where the RMS audio LEVEL indicator sits when the test-tone is on, it should be close to -12 dB if you did everything correctly.

3- Then see where the RMS audio LEVEL indicator sits when the test-tone is switched off, this is the RMS noise-floor of your RECORDER / CAMERA at the GAIN level you set when recording.

...In a normal listening environment, most people will find that the audible noise-floor will disappear around -60 dB. So when the noise-floor value is less than -60 dB ( say something like -51 dB ) you will hear some "hiss", and when it's greater than -60 dB ( say something like -68 dB ) the audible noise floor is gone.

Using this system of testing, my Panasonic GH2 and HMC-150 cameras show a noise-floor of -63 dB when fed a clean LINE level signal from the Sound Devices mixer.

The best low-cost ( under $400 ) digital audio recorder I've tested to date is the Tascam DR-100 MK2, which shows a noise-floor of about -83 dB when using any LINE level setting and any MIC level setting with the pre-amp GAIN set to LOW or MEDIUM. This recorder gets fairly noisy when the HIGH GAIN setting is used, so I would avoid using the HIGH setting if possible.
 
Last edited:
Analogically, the microphone is equivalent to the sensor of the camera, not the mic preamp.

But there are standard scientific tests that are used to measure the performance of any particular circuit. The top two would be the SNR (signal-to-noise ratio), and the distortion figures.

The higher the SNR, the quieter the mic preamp. The preamp noise is the biggest problem of lower-quality preamps, including all DSLRS, cheaper camcorders, and cheaper audio recorders, etc.

The lower the distortion, the better the preamp. Distortion figures significantly less than 0.05% would be expected for high-quality preamps.
 
Richard Crowley said:
Analogically, the microphone is equivalent to the sensor of the camera, not the mic preamp.

But there are standard scientific tests that are used to measure the performance of any particular circuit. The top two would be the SNR (signal-to-noise ratio), and the distortion figures.

The higher the SNR, the quieter the mic preamp. The preamp noise is the biggest problem of lower-quality preamps, including all DSLRS, cheaper camcorders, and cheaper audio recorders, etc.

The lower the distortion, the better the preamp. Distortion figures significantly less than 0.05% would be expected for high-quality preamps.
Click to expand...

This. The gain or equivalent control for low light adjustment is the equivalent of a pre-amp.

The rule of thumb for measuring the quality of a pre-amp is the price. I've had some cheaper preamps used for music recording that have had reasonable SNR but add a lot of 'colour' (uneven frequency response). For music this is fine, take a flaw and make it a feature. After all the classic distortion sound is created typically by faulty gear (despite what many musicians think, correctly maintained tube preamps are cleaner than transitors). SNR is the most important for you, the other thing that makes quality preamps expensive is even frequency response. Do a search for 'pink noise' to find out how to test frequency response. Thing is the size and the shape of the room you are in will effect frequency response, let alone your speakers, A/D and D/A converters etc. So unless you have a professional sound studio set up you aren't going to be able to do much there.

At the end of the day you have to use your ears. It's good to train them anyway. You think you know how to listen but it takes training.
 
Okay thanks. I thought analogically, that the lens would be the mic, and that the pre-amp would be the sensor, haha... Is there anything on the box that indicates pre-amp quality when you buy it? That would help.
 
When it comes to pre-amps in most of the low-cost recorders/cameras I find that it's the measured noise-floor that makes the difference between an ok recorder/camera and an absolutely great recorder/camera, and most of the time you can't hear the difference with your ears when it comes to very low noise-floors.

And mic performance ( and placement ) is often more important than the pre-amp performance. Most low cost recorders/cameras have terrible sounding mics.
 
As Egg Born Son says, the price is a pretty good indicator of quality, but not 100% reliable. There are some notable low-cost mic preamps that have good reputations, while there are also high-price gadgets of mediocre quality.

I can't say that I agree with Mr. Son about tubes vs. solid-state, however. Tubes require transformers for impedance matching which carry their own quality issues (except for those that cost around $100 per channel just for the transformer. Jensen, et.al.) Furthermore, tubes themselves can't touch modern solid-state circuits when it comes to either SNR or distortion. But many people like the quality of the distortion, so they continue to enjoy some small segment of popularity.

However, TheDingo makes the most important point here. Even when dealing with the terrible mic preamps in cheap camcorders and all DSLRs, mic selection AND PLACEMENT will make a MUCH MUCH bigger improvement to your recorded sound track than the difference between the cheapest mic preamp and the most expensive.
 
Richard Crowley said:
I can't say that I agree with Mr. Son about tubes vs. solid-state, however. Tubes require transformers for impedance matching which carry their own quality issues (except for those that cost around $100 per channel just for the transformer. Jensen, et.al.) Furthermore, tubes themselves can't touch modern solid-state circuits when it comes to either SNR or distortion. But many people like the quality of the distortion, so they continue to enjoy some small segment of popularity.
Click to expand...

I did say properly maintained (and as you point out) expensive tube set ups, properly tuned by a technician although admittedly my own direct experience is limited to guitar amps and in that case a dirty signal is the desired end goal. Many guitarists get upset after spending a great deal of money and getting quality tubes that the signal is clean and the distortion when overdriven is sweet, not crunchy. The cheap tubes, dry solder joints and other failures is generally where the crunch comes from (as artists in the 60s well understood, trashing their gear, cutting their speakers cones etc). Not really an issue here though, unless ironpony has a tube based DSLR (how cool would that be!). Don't know why I even brought it up really, how does my mind work?

+1 for effective use of the gear you have. The difference between low end and mid range audio gear is massive. The difference between mid range and high end is much smaller and the gap closes every year. The high end and ultra high end may be costing $10,000 more for less than 5% improvement. Yet good mic placement and great mic placement can make an 80% difference*

*arbitrary value, no science involved! but the point is made.
 
I would suggest using your ears mate. It may sound like I'm taking the Michael, but honestly the best bit of gear is the bit that sounds best, simple as that. You can try to quantify preamp quality with signal to noise ratios and frequency response graphs but the only way to know is to listen.
 
Will Turner said:
You can try to quantify preamp quality with signal to noise ratios and frequency response graphs but the only way to know is to listen.
Click to expand...

...With low end audio gear ( recorders under $400 and cameras under $3,000 ) the biggest drawback is the noise-floor, as many of them can barely produce a noise free recording, and when you have to boost the volume of a recording there is a definite "hiss" that should never be there.

From experience I put the RMS noise-floor threshold at -60 dB under normal listening conditions ( most people can't hear any of the self-noise of the recorder ), so a recorder has be able to produce a -60 dB noise-floor to produce a good recording without any modifications to the audio.

When a recording level is accidentally set too low ( which happens occasionally for lots of reasons ), you may need to boost the audio volume by as much as 20 dB to achieve good sound in the finished piece, and this will also boost the noise-floor of the recorder. So if your $200 recorder has a noise-floor of -63 dB ( which sounds great when everything is recorded perfectly ), boosting the audio recording by 20 dB to compensate for the accidental low recording level, means that the noise-floor is now -43 dB, and may sound like scrambled eggs frying in the background of your recording.

If you had used a recorder with a noise floor of -83 dB ( like the Tascam DR-100 MK2 when set correctly ), you can boost the volume of your recording by 20 dB and the final noise-floor is -63 dB, which most people will not be able to hear, so your recording is still noise free.

...Beyond noise-floor, the second distinguishing quality of low cost recorders for me is the quality of the built-in mics. Very few low cost recorders have decent mics, and these I would definitely evaluate with your own ears by recording a live event ( like an acoustic music performance with vocals ), and seeing what the recorded sound is like.
 
ironpony said:
So when it comes to audio I am guessing analogically, that a pre-amp is kind of like a sensor, for a field recorder, if compared to a camera. You can measure the quality of a sensor by it's size. The bigger the sensor, usually the bigger the quality. But is there any way of measuring pre-amp quality?
Click to expand...

The light sensor in the camera is equivalent to the mic for audio. Both signals are given amplification, but the 'pre-amp' in the camera is buried, and between the sensor and the Analog-to-Digital (A/D) converter, further the digital data is heavily processed by the compression processors which yields the digital data stream that is recorded.

Because of the lower data rate for audio signals, these days, one can have 'no compression' option for the digital stream. DSLRs may never have a 'raw' option for internally recorded image data streams... they most likely will require an external capture device if the 'raw' sensor values are available on a connector. This is why some number of people look to the HDMI port as a method to access the image date before the compression processor.
 
Richard Crowley said:
...Furthermore, tubes themselves can't touch modern solid-state circuits when it comes to either SNR or distortion. But many people like the quality of the distortion, so they continue to enjoy some small segment of popularity.
Click to expand...

In the visual side... Some 'digital' imaging is judged as inferior to Film film because of the lack of 'grain'... 'grain' in Film film is in fact 'noise'... and for much of the history of film film development 'less grain' was the goal... now... it has been turned into an aesthetic...
 
Will Turner said:
I would suggest using your ears mate. It may sound like I'm taking the Michael, but honestly the best bit of gear is the bit that sounds best, simple as that. You can try to quantify preamp quality with signal to noise ratios and frequency response graphs but the only way to know is to listen.
Click to expand...

The point of determining these parameters is to set bounds on what can be expected. It may be that one can not afford an expensive 'quality' option, and very simply... sometimes expensive does not == quality... so by knowing the parameters and how they affect the eventual 'listening' one may be able to better identify and buy 'cheaper' solutions.

I personally don't like the 'sales pitch' that runs along the lines of 'just listen'... It has been my experience that has led to being offered a product that has a higher price tag than if I compared specs, and requested products in a given 'window' of performance vs. price.
 
Can your pre-amp handle a decent gain range without introducing noise, distortion and frequency restriction? If it can, I'd be happy using it. The idea of using valves (tubes) on anything that gets thrown in the back of the van worries me, and I've never been one to appreciate the warmth they produce (aka, 'nice' distortion) on location. In the studio, on certain voices or instruments, the tube compression and distortion does sound nice - but it's not a faithful reproduction of the input - it's been affected, but hopefully nicely.

So much audio advice is snake oil based!
 
Will Turner said:
I would suggest using your ears mate. It may sound like I'm taking the Michael, but honestly the best bit of gear is the bit that sounds best, simple as that. You can try to quantify preamp quality with signal to noise ratios and frequency response graphs but the only way to know is to listen.
Click to expand...

FINALLY someone says this. I find it funny, especially that I'm on the manufacturing side of things, that people get all wrapped up in graphs and charts and numbers and all that. Just use your ears. Is it noisy? Then it's no good. Of course, you need to know how to set up a proper gain structure, but that's audio 101 stuff right there. Also, I do find that "you get what you pay for" does generally hold true, with a few minor exceptions, for all audio equipment.

Here are my guidlines that I personally follow:

Going camera audio only for something serious? Better go line level in. No line level in? Then don't do it.
Low cost equipment means noise..from the pre's to the a/d converters to the headphone amps. Corners have to be cut all over the place to deliver you a $400 hand held recorder.
Want to use a low cost recorder anyway, put at least a sound devices mix pre in front of it and go line level in, but even then that isn't optimal.
Good/Fast/Cheap...pick two.
 
davefisk said:
Just use your ears. Is it noisy? Then it's no good.
Click to expand...

...But what if what you are trying to find out is beyond the limits of your own hearing?

Such is the case where we are trying to find out how much better than the human threshold of hearing is the noise-floor of a portable recorder ?

This inaudible noise-floor becomes an issue when you are trying to fix a recording that was accidentally made with a very low recording level. So now you want to boost the entire recording by 20+ dB, and this inaudible noise-floor suddenly becomes a problem when the recording volume is boosted.

Recorders with better pre-amps that have a noise-floor that is 20 dB lower than your hearing threshold won't have a problem when the recording is boosted by 20 dB.

Really good recorders like the Sound Devices models, may have a noise floor that is 50 dB below your hearing threshold, so you can manipulate the heck out of the recorded audio and not worry about noise from the recorder itself.
 
You must log in or register to reply here.
Back
Top