Latency is consistent. That's not a problem. If you have latency that changes, there's something wrong with your setup. Drivers, anti-virus software interfering or another problem.
The lower end of the pro interfaces these days is usually driverless (USB audio compliant mode). And they do not require extra software, as there is no DSP, no routing. As long as the interface is 2 channels, this works perfectly.
Correct.
Some consumer cards are bad, but most are suited for measurements. The problem is not in the quality of the chips. These are nearly perfect, but filtering and power. If the interface's power is provided by the USB bus, it depends on the quality of the PSU in the computer.
Yes. It has phantom power, so you can choose from a wide range of mics.
Most computers are suited to the job. Some aren't. And it really depends on that particular laptop.
F.i. My own Mac produces a little noise on the input and output. Typical noise, coming from the screen. It can be heard when you scroll. Most other Macs don't have that noise. Mine is 5 years old an a little battered because of life on the road. It could also be an aging PSU.
PCI is out. That's a market thing, laptops are the thing now, desktops are out. And PCI has changed too much. The consumer doesn't know the difference between the various PCI busses.
you forgot the Mic Part 🙂 It does not have the Mic component. Also, it is expensive in my opinion. If one can afford it and also on top of it an extra component which is a good mic and still justify/afford the cost, then probably you are right.
Most of the PC Sound output is pretty acceptable these days. It is the inputs that suck or are almost non existent in case of Notebooks/Laptops.
With only one UMM-6(and its calibrated file), and your existing onboard output(you do not even need a USB sound card then), you are already into measurement arena, may not be very top notch, but yet acceptable.
Sangram, I believe you are also from India, and hope you understand, the dilemma of the cost factor here. Things needs to be affordable as much as possible. Audio enthusiasts here, at least need something that is measured, affordable and is still at the same time, not compromised. and in my opinion, the combo I mentioned, is unbeatable at what it achieves and specially still being cheap and affordable.
There is no end to what more and more money can buy.
Also, please throw some more light on why and how Latency is one of the very important factors while choosing a sound card (specially in terms of Speaker Measurement practices), and what role would DSP effects (which you say require excellent ASIO) play in a speaker measurement rig.
the Asus U3, windows driver(the one ASUS installs has DSP), but, I always use and try to use Linux, and in Linux, the U3 is completely a plain simple card, and DSP effects are not exposed to ALSA. I have had very great performance of the U3 under linux on Dekstop as well as a Raspberry Pi as a plain simple/Elegant DAC and equally great as a High quality Input Source, both for Stereo Walkman Mics as well as Audio AUX in from CD players etc.
It doesn't have a capsule, true, but neither does the U3. So if you want to do both electronics and speakers, you will anyway need two boxes. Pick your poison.
DSP is extremely useful for applying crossovers, filters and EQ so you can see in real-time, the actual response you need to achieve electrically or acoustically when you see an anomalous response. You don't need 'excellent' ASIO. Pro cards have had DSP for over ten years now. Even the $150 e-Mu 0404 cards have a very good DSP. ASIO itself dates back even further, it's just that consumer cards have started using some form of it recently.
I stick with the fact that USB (referring only to sync USB) does have variable latency. This is not conjecture. There is no driverless async but that's limited to one samplerate, at least on Windows (Mac and Linux users have more leeway).
The reason for latency being important is simple, all measurements require a response to an impulse. If the latency is fixed at say, 20mS, the software will add 20ms to each response. When it is variable, the results will not be correct for each impulse. This is a gross oversimplification of way FFTs work, but it'll do for this post. This is why consistency is important, and not available except under async transfer or with a hardware controller.
As to the question of India, I sourced almost all my gear from studios who were upgrading, leeched off a close friend or two, or built by hand. I'm a cheapskate and proud of it. My last new purchase in the recording business was a e-mu 1212m, which a friend of mine brought in from Singapore. That was 2006. I think resource management is good, but I also think one needs to manage the right areas. It doesn't make sense to cheap out on the important stuff. Nor is it possible to start with a big bang, but given the way things are priced in our country, you have to take things a little slowly. Or you get stuck with stuff that's not worth its street price and you can't upgrade for a long time. Ultimately the adage of 'buy cheap, buy twice' applies no matter where in the world you are.
One would hope that there would be some implicit QC inherent in the calibration process. Certainly the DOA units should be weeded out. The ones with low output or really substandard frequency response should be weeded out as well.
The ECM8000 has no in built source of power. No phantom power input, no output!
I'm currently using one of my ECM8000s with a Presonus Audiobox ($80 at GC) that includes the necessary preamp and phantom power. Seems to be working a treat. This is about the minimum fuss and bother route short of an actual USB measurement mic. The setuup is ECM8000->XLR cable->Audiobox->USB cable->PC USB port.
The Asus U3 makes no sense to me as almost all modern computer sound card ouptuts are more than good enough for acoustic/speaker measurements.
I don't know if you are aware of this but for decades a ton of professional speaker designers developed speakers you probably know and love with 12 bit audio interfaces. This is acoustical measurements, not power amp testing.
The alternative I recommended uses a common fairly inexpensive pro audio product that is both a preamp, audio interface and phantom power supply and plugs straight into a USB port that also powers it.
The inclusion of USB may be gratuitous. An onboard or add-on internal audio interface communicates directly to the system bus. The input side of an onboard audio interface may be a little less wonderful than its output side, but it is generally still good enough for speaker and room measurements.
Your criteria (minimize components) is questoinable, and the way you are counting them may be questionable. But cutting to the chase the reason why people go the non-USB mic route is that the selection of USB mics is far narrower than the conventional phantom powered measurement mic, and to a lesser degree the mics with built-in USB may involve performance compromises. As a measurement mic the compromises may be moot because measuring speakers and rooms doesn't require high performance audio, but people have provided me with objective evidence that at least some of the extant USB measurement mics have higher than usual internal noise when you want to use them for recording music. In contrast the phantom-powered measurement mics are generally more usable for general purposes like recording.
From a noise standpoint USB has some built in internal and external noise issues of its own, some of which that may be accentuated by trying to pack the whole enchilada into a small package for a price. Again, this is not so much of an issue for measurements but it is if you want to double dip on your investment and use the mic for some recording sessions.
In side by side comparisons phantom/mic preamp systems have been found to be as quiet or more quiet than USB mic systems in actual use.
If you stop right there, you are as right as rain.
Now the discussion veers off the rails. One thing about that Phantom powered mic/ USB audio interface with phantom power system is that the USB audio interface is often a really very nice audio interface with line and mic inputs and line outputs, all by itself. Typically there is balanced I/O in and out, for example.
It is true that a lot of laptops are kinda challenged on the audio input side often by simply offering only a cheap electret mic input, but its been over a decade since I saw a desktop whose line inputs were not good enough for speaker and room measurements.
Agree with that!
To summarize, for speaker and room measurements one of the popular USB measurement mics is a wonderful alternative with several very usable alternatives. You don't have to put down the alternatives because it is more than good enough, and so much simpler and lower in cost than the alternatives.
For speaker and room measurement purposes my usages of USB mics and that of associates who I trust has been just fine. If you have no measurement mic, don't pass Go, don't try to collect the $200, just run right out and buy one. The ones from Dayton audio and MiniDSP are proven products.
That is IMO a false claim. The best ADCs and DACs are professional audio tools that sit inside the PC. I base that on experience with literally 100s of audio interfaces over almost 2 decades. They are often simply more solid in use than their USB alternatives and since everybody buys the DAC chips out of the same pail, there is no necessary superiority to either alternative.
If you want to split hairs, USB has noise exposures in the PHY and protocol areas that can relate to both internal and external noise. I don't want to make mountains out of these molehills but these are objectively observable facts that are often manifest in real world usage. Fortunately for speaker and room measurements, these are exceedingly moot issues.
But you raised the issue.
Last edited:
All of the modern Impulse response based software I am familiar with (example: Holme Impulse) does not even actually use an impulse for testing for reasons of unmanageable sensitivity to room background noise. Holme Impulse and competitive software usually uses a Fourier Transformed sine sweep. The impulse response is key to the process, but it can be obtained from other signals by means of relatively easy and well-known math.
Furthermore, in a speaker/room measurement context, the latency in the audio interface vanishes during the management of the inherent acoustical latency of the process.
Modern Impulse response based software automatically finds the primary peak in the impulse response and in the process zeroes out audio interface latency. If you are struggling with audio interface latency during your speaker/room measurements, please update your software!
Last edited:
These seem to be getting discontinued left and right. Who's still actually supplying them? I *think* that the Lynx is still around, but M-Audio and ESI seem to have abandoned that market.
I don't have any issues because I do not, and do not intend to use USB. At least not sync USB.
The comments you are quoting are being misunderstood. The assertion is that sync USB has variable latency - defined by a range - and it is not suitable for audio work because audio is real time and the bus does not work in real time.
Does that mean everyone using sync USB will face a problem? No.
Does it mean that some using it may? Yes.
The key is to find out if your soundcard does or does not, one red flag is a 16/48 limitation.
I have all the internal interfaces and external mic pres I need for measurements. I thought I lost my 1212, but it came back from the dead. I guess the cap replacements I did in the first year were uncalled for, and are now causing issues.
So then you have no personal problem with this allegdly severe problem. Curiously I have a good stash of USB and bus-attached audio interfeaces and I have no problems. What might be learned by comparing these two sets of experiences?
I only have about 25 years with computer audio and at least 12 years professional experience with live recording so what could I possibly know that is relevant?
Hey some people are going to have problems with everything.
Why would one need more than 16/44 for room tuning or loudspeaker work?
OK, if you want to play with hypersonic tweeters you need more than 16/44, but that is a tiny minority. Most people are most interested in the normal audible range last time I looked. Not that hypersonic recording and measurements stop me. I have these B&K 4006s and many fine audio interfaces that work up to 24/192, but back in the real world I've done a lot of work of all kinds with 16/44 audio interfaces.
Yup a lot of those questionable so-called upgrades can come back to bite you. A high proportion of the broken caps I've replaced lately said Audiophile on them. Advice from an old hand - if it ain't broke don't fix it. ;-)
That is my experience as well.
If I have to use legacy cards in computers with legacy OSs then so be it. I just picked up a AP 24192 on eBay just in case...
Lynx and RME seem to be going strong. Even if he market shrinks there can be good business for years for the survivors who have less competition.
However today we are talking about speaker measurements and there are many contemporary USB devices with current support can suffice.
Good news! We are working with speakers and room, not amps DACs and preamps. ;-)
Last edited:
Thanks Guys, Sangram and arnyk
These discussions are so much healthy, wealth of information exposed easily, by a very slight sensitive trigger(low latency in Humans you see).
Please keep spewing all that knowledge for the benefit of all.
These discussions are so much healthy, wealth of information exposed easily, by a very slight sensitive trigger(low latency in Humans you see).
Please keep spewing all that knowledge for the benefit of all.
You don't. A 16/48 card is a dead giveaway that it is using sync USB. 16/48 is fine for speaker measurements. Sync USB cards are not.
But hey, since you don't have a problem, nobody else should, right?
Given that you seem to have admitted that you don't have the problem either, one might wonder what personal experience you have with the issue, or whether we are talking basic fear of the unknown. ;-)
I'm not saying that USB cards are perfect, but being able to use a laptop and not cart around a desktop PC to do some audio system tuning is the kind of thing that is attractive to many people.
I take it that the world of touring SR or outside consulting is not part of your life? There is work there, friend!
Even with desktops, the ability to plug in a simple USB device and mic, and not go inside the box, has quite a bit of romance for those of us who do work outside of our personal shops.
This is speaking as a guy who goes inside PCs and laptops all the time. It is still work and it takes time I'd prefer to do other things with, like listening to music.
BTW I found a fairly technical review of the Presonus Audiobox that I mentioned earlier here:
Review: PreSonus AudioBox iTwo by Paul Dormer | UCan Play.
He says latency is about 5 msec, which seems low enough even for recording.
Last edited:
I think the inconsistency referred to in some USB soundcards is due to different sample clocks being used for record and play. They can never stay is lockstep unless the clocks are locked (or the same), otherwise they continually drift apart (and there can be a few samples mismatch when averaging repeated responses over time). If you use no averaging, and don't require the 'reference distance' to stay stable, then it does matter. but averaging can be a powerful tool when you need more dynamic range in a non-perfect environment.
Big offenders are any of the TI PCMxxxx-based USB soundcards (such as the Behrenger USB Audio dongles). those chips use the the USB clock for one function and an internal crystal for the other. Presumably so the sample rates can easily be different on normal Windows Skype-type use, but really, godknowswhy.
Big offenders are any of the TI PCMxxxx-based USB soundcards (such as the Behrenger USB Audio dongles). those chips use the the USB clock for one function and an internal crystal for the other. Presumably so the sample rates can easily be different on normal Windows Skype-type use, but really, godknowswhy.
Focusrite Scarlet 2i2 from BAJAAO.COM Music equipment superstore. 30000 products. Free Shipping. | Trusted by professional musicians and sound engineers since 2005
I just bought one. It works very well. Better than the EMU 0204 which I bought sometime ago.
The Focusrite was on sale a short while ago. It was priced just under the equivalent US price ! That's a great deal and no shipping charges or customs hassles.
Its Rs 10,800 now. If you wait it might be on sale again , later on ( Rs 9760/- ! ). Amazon price US$ 150 /- plus shipping plus duty !
http://www.bajaao.com/products/focusrite-scarlett-2i2-usb-2-0-award-winning-audio-interface
I just bought one. It works very well. Better than the EMU 0204 which I bought sometime ago.
The Focusrite was on sale a short while ago. It was priced just under the equivalent US price ! That's a great deal and no shipping charges or customs hassles.
Its Rs 10,800 now. If you wait it might be on sale again , later on ( Rs 9760/- ! ). Amazon price US$ 150 /- plus shipping plus duty !
http://www.bajaao.com/products/focusrite-scarlett-2i2-usb-2-0-award-winning-audio-interface
Last edited:
- Status
- Not open for further replies.