Which USB Sound Card to buy?

Status
Not open for further replies.
I would see if you can find a used product of higher quality, such as e-mu - some options may be two-part cards, but may have better A/D spec.

One thing to watch out for if you use Windows is that Win7 normally has slightly higher DPC latency than older versions of the OS. There is a long thread on the SOS forums (where you also might find some pointers to search for used cards) that held you optimise the system to get better latency numbers.

Stay away from the Asus cards for any serious audio work. They do have decent D/A, but the latency of their WDM drivers is terrible and the ASIO drivers do not allow control over levels as they do not have hardware mixers.

I believe we are quite far off topic now.
 
It will be quite interesting to know what kind of speakers are to be measured with these super tech and super highend Sound Cards with super low latency. Also, what kind of Rooms they will be tested in. and what difference would it really put towards measuring FR and THD of these speakers by using cards ranging from .0000008 vs .01 THD.

In my opinion, hardly any DIY stuff (specially speakers that are not manufactured using High Precision machines), and that too fabricated at home using home tools, would benefit from a super high latency or super smooth DAC/ADC card. I would rather spend the extra money to get better speaker drivers, and spend more time in the cabinet planning and design.

I am doubtful if there is any sense on attempting to measure a 0.01db of difference in a speaker's response across a couple of HZ.

but that is all my thought, I may be wrong.
 
IF latency means anything for loudspeaker measurements, you should at all cost avoid using a smartphone (unless it's an iphone). All smartphones have a lot more latency than that Behringer. And by a lot more I mean > 100 msec. The Behringer has around 25 msec.

I can't even imagine what latency has to do with audio measurements. So can you explain?

The only thing the Behringer UCA202 lacks, is a microphone input with phantom power. Hence the low price. Next up is the Focusrite. Almost everything in between is not worth the money, imho.

You could buy a couple of Chinese C-media USB sticks @ 3$ a piece and use the one with the best linearity. Usually these aren't too bad for measurements. You need to check if they aren't missing input and output capacitors tho. They're so cheap, some manufacturers "forget" those.

And besides the ECM8000 you can use most small diaphragm condenser mics too. Only, they aren't any cheaper than the ECM8000. And for the ECM8000 you can get a calibration file on Behringer's site.

If you don't have phantom power on your mic preamp, go for a battery powered electret. These are often fairly linear too.

PS stay away from USB mics. These work allright usually, but some people are losing hair over these because of whinig sounds making them completely useless for measurement (and recording too). There are a couple of professional USB (measurement) mics, but these are really expensive.
 
You could buy a couple of Chinese C-media USB sticks @ 3$ a piece and use the one with the best linearity. Usually these aren't too bad for measurements. You need to check if they aren't missing input and output capacitors tho. They're so cheap, some manufacturers "forget" those.

Manufacturers Forget that ? They Don't Forget, they Intentionally do it. In any case, at least the output of these cards have horrible harmonics. If you have a FFT, try running the output of these cards on the FFT and you will see why they are pure utter crap. Being Flat or not, who cares, when they produce a 5Khz Harmonics(70% of amplitude of primary wave and of multiple frequencies) of a pure 1Khz wave ? I don't know why would do that though. Bad implementation of the DSP(D/A) or the poor analog output stages. C-media chips are not bad. These people just do the worst with them that is possible in the world.

PS stay away from USB mics. These work allright usually, but some people are losing hair over these because of whinig sounds making them completely useless for measurement (and recording too). There are a couple of professional USB (measurement) mics, but these are really expensive.

I use a Dayton UMM-6 (USB), no "whinning" and no issues of any kind so far. I have used it as a measurement Mic and also Recording mic. Used it on the PC, Laptop as well as an Anrdoid Tablet(no iGenPhone) (tested on Nexus 7 2nd Gen and Xiomi Redmi 1S also) as well. So not sure where you get those ideas from.
 
Manufacturers Forget that ? They Don't Forget, they Intentionally do it. In any case, at least the output of these cards have horrible harmonics. If you have a FFT, try running the output of these cards on the FFT and you will see why they are pure utter crap. Being Flat or not, who cares, when they produce a 5Khz Harmonics(70% of amplitude of primary wave and of multiple frequencies) of a pure 1Khz wave ? I don't know why would do that though. Bad implementation of the DSP(D/A) or the poor analog output stages. C-media chips are not bad. These people just do the worst with them that is possible in the world.

Forget was between quotes...

"These cards": I suppose you've tested ALL these cards. Some are crap, some are quite good. Almost as good as the Behringer. It's just an attempt at making the OP's budget work. That's why I said: buy a couple and test.

I use a Dayton UMM-6 (USB), no "whinning" and no issues of any kind so far. I have used it as a measurement Mic and also Recording mic. Used it on the PC, Laptop as well as an Anrdoid Tablet(no iGenPhone) (tested on Nexus 7 2nd Gen and Xiomi Redmi 1S also) as well. So not sure where you get those ideas from.

I get those ideas because I support a lot of people doing audio recordings on all platforms. These are new users (to recording), using Audacity, with minimal technical knowledge and experience. They get to buy their own gear, so a lot of them bought USB mics, unfortunately. Internal laptop mics are often better, despite their bad reputation.

No matter what brand, some combinations of computer/USB mic are impossible to use. And there's no clear reason why you get a whine sometimes. If it's the PSU, it can be resolved by operating the laptop from the battery. But sometimes it is the power converter in the mic that starts whining and there's nothing you can do. Besides, there's nothing the manufacturer will do, except exchange the mic for another one. They know the problem very well.

We've tried USB repeaters and USB isolators. Every now and then they resolve the problem. Usually not. Unfortunately we can't use recorders, because of software needed.

That's a problem that doesn't exist with any decent USB audio interface. Even the cheap Chinese ones don't whine very often. And when they do, it's usually on a laptop with a PSU that's on it's last legs.

There are thousands of users out there for whom it works, obviously, but there's no way to tell if it will work for the next user. And there are far too many cases that it doesn't work.

I've seen the inside of Behringer, MXL, Blue, AKG, Shure and other USB mics. They usually are very alike. And not very good for the money. I also have the gear to compare and test with other good mics.

In some cases, when there is a "normal" and USB version of the same mic type number, the USB model isn't the same quality either. Sometimes it's inferior, even vastly inferior.
 
Some of the Cheap Chineese ones work because, they still use the same C-Media Chip that most of the cheap ones use.

The C-Media chip is not bad at all, These guys care a damn about power filtering, Do not employ any kind of RF protection, employ bad PCB design, and what not. So you are really lucky if some of them actually did take care while manufacturing. Telling someone to actually buy a few pieces to test with may not be a good advice(how many you think would make it a chance of getting a decent one ?) .

I really can't say about your experiences with USB mics, but for me, I have had great success at the Dayton UMM-6 Mic. I am pretty happy with the measurement setup with using just another USB sound card which is the Asus Xonar U3, the U3 uses a slightly modified version of C-Media itself internally(probably built to OEM), however, gives a very clean sound without Harmonics in either inputs or outputs. I will any day avoid any Mic with an analog pre-amp component/Phantom Power in the route to the sound card input, specially, in case of cheaper mics.

From all my combing of information and experiences shared by others, and to achieve the cheapest route towards having some sane measurement abilities (with a calibration as reference), by far the best Combination is the UMM-6 and the Asus Xonar U3.

I have no idea about the more expensive ones out there, and I think, a budget conscious person couldn't bother either.
 
Last edited:
The C-media chip in itself is about as good as any other out there. It's the rest that counts. Audio and power filtering.

For ME, personally, all these USB mics work. Unfortunately, because it makes trouble shooting impossible. And you can't count on the rule "same laptop, same USB mic, so it'll work". It's crazy.

All my mics are phantom powered (except for a few T-12 powered Sennheiser MKH long shotguns that I love and dynamic mics, of course). It's no disadvantage but you need a decent interface to go with them. And "decent" usually means "NOT USB powered". There's only 2,5 W to be had from a typical USB port and a mic preamp needs about one watt typical. Add 48V 10 mA phantom power to that and it leaves you with very little to power the ADDA. That's why some of my (older) mics don't work very well with USB powered interfaces. Either they need a minute or so to stabilize (AKG C451/CK8) or they don't work at all (Behringer C4).

But that's not very interesting to the OP, I'm afraid...

Is that Dayton calibrated?

Now that I think about it, there's another USB measurement mic out there that is not too expensive: The miniDSP UMIK-1 USB Measurement Calibrated Microphone. But I've never used it.

And, ouch, that Behringer ECM8000 is expensive in the US of A. It's about 30 € incl. VAT here, I see it's 49 $ without sales tax. Better buy the identical Audix in the States!
 
The Dayton comes with Calibration File downloadable and is unique to the serial number of each mic. Also, Someone is selling them further with his own calibration files as well, using very expensive equipment. Some claim the personally resold calibrated ones are better than the company provided calibration files.

Does the ECM8000 have a Pre Amp built in other than the usually Unusual(at least for starters) 48V supply ? If not, what would a decent Pre-Amp cost along with the phantom power Supply ?

You are not getting the point that I tried to convey. Once again here it goes, the most fool proof cheapest way to measurement heaven is the route of the USB UMM-6 Mic and the Asus U3. That also does not restrict you to the confines of a small room with AC power supply for the Phantom Power and also for the Pre-Amp. It is one sure shot simplest and fastest and reliable combination to go measurement.

If someone wants to take a more complex route, arrange for a lovely 48V supply, Spend money on a Super Pre-Amp, then go for it, no one is stopping. You still need a decent Sound Card with Guaranteed Quality Input.

Now Asking cyrano and more knowledgeable people in this forum, a very simple but fundamental question, what happens when you connect a ECM8000/Phantom Supply/Cool Pre amp for recording something through a PC ? Can you explain the Audio Route from source to destination?

From what I understand, here is the route that it would take.

1) Voice Vibrations in Air
2) Electrical Impulse Generation at Mic Diaphragm
3) input to Fet/Mosfet Powered by Phantom Power
4) Capacitance/Electrical Variations Amplified by the FET within the Mic Capsule
5) Impulse Feed Into the External Pre Amp(the cool one of your choice)
6) Analog Amplifier Amplifies Signal
7) Amplified Signal Fed into Line In (PC Sound Card)
9) Sound Card Performs its own analog stuff.(with its own amplification/attenuation analog circuitry).
8) Sound Card Converts into Digital Sample Data
9) Transmits Digital Data over USB Bus/System BUS(whether internal or USB depending on type of sound card)
10) Consumption of the Digital data at the main Software

Looking at the above, I do not see the ECM8000 do anything differently than what the UMM-6 would do, except the ECM8000 requiring a bunch of more electrical components that are exposed to the more noisy analog world and if not done very right, then it would do more harm than good.

The UMM-6 covers all of the points in one neat little package(shielding you from additional equipment, additional RF noise in the outside world and cables) and at the same time giving you a free reliable sound card at least for the signal input part. eliminating the need completely for purchasing a better sound card with better inputs. On Top of it, it works well on Tablets as well, so carry it wherever you need and measure and record where ever you need easily. It comes with its calibration files, that can be used across most industry standard measurement software with little or no effort at all.

So where do you see the trouble in a USB mic ? I would be keen to know about it more. I think a lot of other people will also understand the rational hidden in the bad reputation of being USB. Those people probably overlook, that the best DACs out there all work so swell on USB. So what is different here ?

Sometimes if you drive a Left Handed Car, you think all Right Handed cars anywhere else in the world are bad. Tough to accept something new and possibly better and easier.

I am requesting more experienced folks here to throw more light on this. I was myself keen to spend the least possible to have a decent measurement system and was under the impression that, what I have was sufficient for most measurement needs. If not, then I would like to know and it will help other members as well.
 
USB has variable latency. High latency itself is not an issue, as most programs can adjust for it. It just needs to be consistent.

Read this great explanation on latency (warning, long!): https://www.presonus.com/community/Learn/The-Truth-About-Digital-Audio-Latency

Most pro audio interfaces use the USB connection only as a transport (electrical) but the control is by using their own driver and not Windows'. Most consumer cards including the little Asus talk to and through Windows. The ones that do have some ASIO support (such the Asus and it's C-Media chip) are only capable of routing, which means the input 'pin' gets connected directly to the hardware input, and similar for output. No other controls are available at this time.

Note that most of the pro interfaces are actually controlling a hardware chip inside the interface and using system memory directly. Thus all functions including mixing, levels and DSP (if present) are available when ASIO is used.

There are other ways to get around this problem (of USB not being a real-time bus and not being suitable for audio playback). TI has some documentation on the development of the PCM2702 which highlights some of the peculiar problems of audio playback over USB.

However there are a number of USB measurement mics, some of which are highly regarded.
I believe that was a specific response to a post about consumer cards, and in the original context of the thread. Please correct me if I am wrong.

If someone wants to take a more complex route, arrange for a lovely 48V supply, Spend money on a Super Pre-Amp, then go for it, no one is stopping. You still need a decent Sound Card with Guaranteed Quality Input.

I believe the Focusrite mentioned in a few posts here is a one-box solution that has all of these functions, is as portable as the combination you mention, offer hardware mixing and a very, very good A/D. It costs more for sure, but is all a hobbyist will ever need to buy for any kind of measurement including sources, preamps, power amp and loudspeakers.

Every laptop I have in house has a mike preamp. Not a great one, mind you, but good enough for speaker measurement.

I do hope they're better than the Acer Aspire 1810:

fr.png


I was looking around but couldn't find more results for laptops in loopback mode and with the additional gain enabled. Macs will surely be better, as they are designed to work with recording apps right out of the box.

You can read the rest of the review where I pinched the image from here, Asus' Xonar U3 USB audio device - The Tech Report - Page 2

Notice how a cheap PCI card is always better, but also how horrendous the audio quality of the laptop is. The bulk of the issue is in the recording side of things. I believe the D/A is not that bad.
 
Last edited:
Hi

By design the phantom power is sent to all inputs, this is common to pretty much all these low-cost interfaces and mixers.

The voltage is present on both the + and - nodes of the input and does not damage dynamic mics as the voltage across the capsule is effectively zero.
 
Status
Not open for further replies.