Film-Tech Forum ARCHIVE: Sound card for RTA

This is topic Sound card for RTA in forum Film Handlers' Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=1;t=009285

Posted by Marco Giustini (Member # 4544) on 05-15-2008, 09:09 AM:

I need to replace my external sound card that I use as pre-amplifier for my two measurements microphones I use for Screen setup.

I need a very flat sound card, with excellent performances. USB or Firewire.

On the web I can read all kind of reviews, so I'm not able to understand which is good and which is not.

I was even thinking about the USBPre, used on D2 RTA, but read some issues for that too! [Smile]

I saw some good devices from Edirol, EMU, Tascam...

Can someone give me some suggestions?

Thanks and regards

Marco

Posted by Mark Gulbrandsen (Member # 72) on 05-15-2008, 10:35 AM:

What about considering a real 1/3 octave analyzer instead. The new Ivie is not all that expensive and it also does a bunch of other neato stuff. About $1500.00 USD. I didn't care at all for the last Ivie pocket PC gestation of a 1/3 octave analyzer but this time around there are more sensible options available. I guess they learned the hard way. This time they did it right.

Mark

Posted by Robert Minichino (Member # 3495) on 05-15-2008, 12:03 PM:

The M-Audio MobilePre USB is pretty good. I'm not terribly thrilled with its noise performance, but it has very flat frequency response, 48V phantom power, and is fully powered by the USB connection.

Posted by Gordon McLeod (Member # 33) on 05-15-2008, 09:22 PM:

I am very pleased with the USB Pre that was supplied with the D2 and had no issues with it

Posted by Marco Giustini (Member # 4544) on 05-15-2008, 09:41 PM:

Robert

I own an M-Audio USB and it's far from perfection... Perhaps it's my particular model, but I really want to try another brand!

Gordon

Read a review on Internet that says that USBPre has some 1000Hz noise on playback. It's a 500$ piece of equipment and I really like to read only positive review! [Smile]

Mark
Good point. I'll check Ivie web and see what they offer.
However, I'm used to use two microphones at the time to have a real time average of the two, by software. However, my actual system seems to have very poor HF accuracy, despite the microphones has been both calibrated.

Thanks

Marco

Posted by Gordon McLeod (Member # 33) on 05-15-2008, 10:54 PM:

Funny the whine issue is a complaint on all the usb audio devices he reviewed
The level of whine is well below the noise floor that a room analyzer is worried about and in the case of the usb pre it is only in playback which is not used as an rta

you could also check out
http://www.gold-line.com/tef/t-tefpre.htm

Posted by Marin Zorica (Member # 1716) on 05-16-2008, 08:18 AM:

Use E-MU 0404 USB card, it have two mike input, balanced with phantom power!

Posted by Marco Giustini (Member # 4544) on 05-16-2008, 08:00 PM:

EMU 0404 is one of the device I'm watching.

There are lot of devices with two mic input with phantom. My concerns are if these devices are suitable for my needs.

For example, I saw a device from Mackie. Asked to the tech support they told me that it would not be good for me, since it is a device for live recordings and they added a slight raise in high frequencies.

This is why I'm asking.

Thanks

Marco

Posted by Tristan Lane (Member # 1169) on 05-17-2008, 01:39 AM:

Marco,

What program are you using in conjunction with the USB input device?

Posted by Marin Zorica (Member # 1716) on 05-17-2008, 02:40 PM:

quote: Marco Giustini
For example, I saw a device from Mackie. Asked to the tech support they told me that it would not be good for me, since it is a device for live recordings and they added a slight raise in high frequencies.

mhmm....kinda stupid thing, why boost up? I think all cards like 0404 and other are flat!

Posted by Ron Funderburg (Member # 4561) on 05-17-2008, 02:52 PM:

I'm with mark I use an RTA stand alone and it works fine they are not all that expensive. I also still use an Oscilloscope rather than software. It is a bit to lug around but the computer based hardware and software cost are so high and the both these units work so well I see no reason to switch.

Posted by Marco Giustini (Member # 4544) on 05-17-2008, 06:34 PM:

Marin

Seems stupid to me too. But that's what Mackie's support said. And that's why I'm asking everywhere for reviews and comments. I'm asking to manufacturers too of course.

I was wrong about the HF raise, they just told me that it has a "distinct colorization" and that it has not "the linear frequency
range you are looking for".

Tristan
I use many: Spectra, Smaart, TrueRTA...

Marco

Posted by Tristan Lane (Member # 1169) on 05-18-2008, 02:04 AM:

SpectraRTA allows you to make a microphone compensation file based on the response curve of the microphone used. I've never used the other two you mentioned. I use SpectraRTA exclusively now in conjunction with a USL mmp-10 4 mic.

My personal preference for RTA's has been a laptop based system. Whether it is an R2, D2, or other program. The input devices I used were always high quality, and I preferred to work with a 4 mic setup rather than a single. The laptop based programs can also average the response on the readout so you get a much better idea of overall room response when cycling through microphones.

Overall I think much of what someone uses depends on what they are comfortable with. I've heard about many techs that still swear by the Abacus / O-scope. If that tool provides them with the best results, then why change I guess. It all comes down to listening to the room when you are done.

I packed an IE-30 as well for quick checks or to use in a pinch.

Mark, remember: All Ivies past the IE-30 are essentially the same as a laptop based system. All of the analyzing is done within the software running on windows. The hardware portion is providing a clean input for their mic. It's gotten even closer to that with the IE-45, which is just a Samsung Q1 UMPC. But, their software is top-notch and provides all the bells and whistles needed for a cinema tech.

Posted by Monte L Fullmer (Member # 2797) on 05-18-2008, 02:59 AM:

True, SpectraRTA is a pretty nifty program to use...very easy to to work with and it does the job needed.

Posted by Marco Giustini (Member # 4544) on 05-18-2008, 06:52 AM:

Hi tristan

Yes, Spectra has compensation. I have my microphones checked and I have a compensation files for them but seems that my sound card has trouble so I have to replace it.
And I want to start with a flat one, to reduce the software compensation at minimum.

Sound check with a known reel at the end is a must, don't worry!

I would buy a D2 for sure, I will do when/if my business activity will justify it. [Smile]

Thanks

Marco

Posted by Catalin Balaceanu (Member # 3712) on 05-18-2008, 09:27 AM:

quote: Tristan Lane
I use SpectraRTA exclusively now in conjunction with a USL mmp-10 4 mic

Maybe it's a bit off-topic, but does SpectraRTA have the X curve reference feature? I know that the latest version of TrueRTA has it.
Catalin

Posted by Marco Giustini (Member # 4544) on 05-18-2008, 10:24 AM:

You can make it by yourself. Spectra has "overlay" curve files, just create some files with all the X curves you want and it will be overlayed on the RTA screen.

I have many kind of overlay: X-500, bigger and smaller screens.

Marco

Posted by Steve Guttag (Member # 268) on 05-18-2008, 12:36 PM:

This is a case where the R2 or D2 really stomps over the other solutions...

The R2 or D2 has calibration files for EACH microphone and they also NORMALIZE the response to mic 1. That is, Mic 1 sets the SPL reference and then the other three are raised/lowered in SPL to have the display not bounce around.

The problem I see with the USL multiplexer on other analyzers is that at best there is one compensation file as the analyzer does not know at any one point in time which mic is active. They also do not normalize the display so the mic closest to the screen is shown louder than the one furthest from the screen. As such, a frequency anomolie nearest the screen will get more weighting than one furthest from the screen. This is not the case with the R2 or D2 due to the normalization feature.

As to those that use microphone compensation files with your analyzers...who is making these files? That is, what is the basis for these files? Are the individual microphones being measured? If so, by whom and how valid is the measurement? How often do you have them rechecked? With the R2 or D2 system, there is a means to have the mics rechecked with a new offset file as often as desired (about once per year for us).

BTW...I too love my Abacus scope attachement...it is the best analyzer I've ever seen in terms of display and ballistics. I still use it for A-chains and for quickie B-Chain checks.

Steve

Posted by Tristan Lane (Member # 1169) on 05-18-2008, 01:49 PM:

Steve, Your remarks are in line with what I have found as well. If there was one problem with the USL plexer, it is that the measuring device is "dumb" as to what mic is being measured and at one time. You can actually see the response fluctuate when the plexer is cycling. In addition to this, there's always an issue with controlling noise gating. I only successfully got it to work once, and found it to be too big of a pain. SpectraRTA has functions for RT-60, but without a precise control of the pink noise from the software, times wouldn't be accurate enough.

My way of dealing with the fluctuation between microphones was to tell the software to average for a given time, and then make adjustments based on overall response. Generally I would repeat this process two more times and take a final read.

To set levels, I stop the plexer at mic 1 and set the channel gain accordingly.

As far as mic compensation files, I've never used them for anything other than x-curve. I liked the way the R2 showed the response as flat on the display when in actuality, the response was in accordance with the curve. I mimicked this within SpectraRTA by adding 1DB of boost consecutively to each 1/3rd octave over 2k. Then when a proper curve was achieved, it would appear flat on the display.

Unless you sent your mic off to TXH, or another similar lab for reference testing, I doubt you would gain anything other than a general figure of what the response of the mic was. And like you said: It only applies to one of the 4 mics. I still believe that the overall measurements taken by the USL system are going to be better than many of the single mic setups being used.

Posted by Steve Guttag (Member # 268) on 05-18-2008, 05:10 PM:

The only way I could see the USL system being notably better than a single calibrated mic is if you took the time to balance out the inputs each time you set up the system. That is, play pink noise and note the level in mic 1...then set the other three mics to match (EACH TIME YOU set up the system in each room).

For the offset file, you could then send all four mics out for measurement...then create an average file that represents the average off all four mics since hopefully the analyzer under use is showing an average of the four mics (time sample is longer than it takes to sweep through all four mics).

By in large, this is not what I see. I see most people just plunking the four mics down and tuning up. Well, without offset files, you are tuning the room to the mic's response, not for "flat"...and furthermore you are tuning primarily to the "loudest" mic. The USL mics are not necessarily the flattest things and neither are the R2 or D2...the key to them was having accurate offset files.

Prior to the R2, the AKG C451 with an omni module seemed to be the defacto standard for measuring theatres. I know that AKG supplied a individual response curve for mine...including when I sent it back for repairs and it was pretty darn flat. That mic was many hundreds of dollars though.

Steve

Posted by Marco Giustini (Member # 4544) on 05-18-2008, 07:20 PM:

Hi Steve

R2 and D2 are the best RTA analyzer on the market, this is a fact.

Said so, I use two channel soundcard (L + R) and have SPECTRA averages both channel in amplitude in real time. Both channels have its compensation file, made by a lab in Germany. I sent my mics to this lab, they provided a file with amplitude and phase compensation in TXT format, plus PDF graphs.

For my needs it's enough. I agree with you that once (or IF) I'll need more, the next and unique step will be the D2.

Thanks for the amplitude hint. I really never thought about it.
Spectra, however, shows "algebraic average" of the amplitudes of the channels. The response curve should not change if one mic is higher in level than the other.
But using the USL yes, you're right.

Thanks
Marco

Posted by Tristan Lane (Member # 1169) on 05-18-2008, 11:38 PM:

Steve, You may be right in that your method would provide the most accurate readings for each time you adjust, but the USL has proven to be fairly reliable. USL claims that each mic is calibrated at the factory, and that no compensation is needed to achieve desired results. Yes, the mics are not inherently flat, but overall gain on all mics can be perfect since the plexer has adjustments for each mic's gain.

I used to be as skeptical of the USL kit, as my prior experience was strictly with an R2 or Ivie. Since I had doubts as to the accuracy of the new product, I decided to tune a room with it and then save the readings for future reference. I then set up an R2 in the same room, taking care to place the mics right next to the USL mics. I compared the readings of both spectra RTA and of the R2 software and they were very close (+/- 2db on a few frequencies). I even connected the USL plexer to an input on the R2 box and took a reading within R2 software and got similar readings.

I'm certainly not denying that the R2 or D2 are a much more robust and accurate measuring tool, but I'm weary of single mic tuning, and prefer the USL over it.

Posted by Marco Giustini (Member # 4544) on 05-19-2008, 06:18 AM:

Microphones cannot be "hardware calibrated". Every microphone has its response curve and it can be corrected only by software. USL device will not calibrate it since the software will see only one input.

About the ability to adjust each input of the USL, is what Steve has said, but you should adjust each microphone's level for each screen.

Regards

Marco

Edit: best reply was from EMU support. They can't say if the EMU 0404 suits my needs, the best thing is to test it from a dealer by myself and decide if it is good for me [Smile]

Posted by Mark Gulbrandsen (Member # 72) on 05-19-2008, 07:47 AM:

All of this has been discussed man many times before in these threads and I'm surprised the F-T police haven't slammed this thread in with some other one from 2002. Fact always was and still is. R-2 and D-2 are the most accurate standard analyzer for cinema. If you don't own one then you can't work to any sort of accurate standard period. The fact that the mics are normalized is but touching on what either of them are capable of. Personally on the offset files I have often wondereed just how accurate the R-2 and D-2 offset files are done... Since it is also done as a side by side comparison with a mic "standard" it may depend more on how the person who is doing them feels on any given day than anything else and its this part of the R-2/D-2measurement system that I always thought should be automated at the manufacturing end. Aside from this there is no other accurate enough analyzer in my book.

Mark

Mark