OzOnE_2k3 said:
The logsweep method on Acourate offers a fair amount of ambient and steady state electronic noise rejection because it computes using your recorded logsweep measurement and an inverse of the original logsweep.
Taking this into account, the signal to noise ratio of the measurement doesn't have to be as high as other methods.
I'll explain what I do and maybe it will help.
For SPL, I aim for around 90dB at the mic position. If you have a cheap SPL meter such as a rat shack you can quickly confirm by playing pink noise, hold the meter near the mic and take a reading.
Noise floor or ambient level in my room is about 20dB but its well damped with treatments so yours may be higher. SNR is therefor 70dB(90-20) or in other words, you can trust data to about -70dB with a measurement normalised to 0dB.
Anything over 50dB SNR is fine for our purposes since were only interested in correcting frequencies to about 30-40dB below peak level, anything greater would be well into the stop bands of drivers and inaudible to the ear.
I am currently building a pcxo system and need to buy an audio interface with 8 channel output.
I was going to buy a Fireface 800, but with the recent price hike of RME products here in the uk i'm looking at something cheaper.
The Echo Audiofire 8 seems to be well regarded and is quite cheap, but I don't know if it has the channel routing capability for what I need.
I'll be using Acourate (or possibly Audiolense) filters in console. For music playback I'll be using J River and for tv/dvd I'll use the analog inputs of the interface (with low latency filters)
Does anyone have experience of the Echo Audiofire interfaces?
thanks,
Martin.
I was going to buy a Fireface 800, but with the recent price hike of RME products here in the uk i'm looking at something cheaper.
The Echo Audiofire 8 seems to be well regarded and is quite cheap, but I don't know if it has the channel routing capability for what I need.
I'll be using Acourate (or possibly Audiolense) filters in console. For music playback I'll be using J River and for tv/dvd I'll use the analog inputs of the interface (with low latency filters)
Does anyone have experience of the Echo Audiofire interfaces?
thanks,
Martin.
mleggett said:
Echo Audiofire 8 may work because the Mackie 400F has driver written by Echo for Mackie. Another user earlier in this thread was seemingly happy with the performance. You'll find his post here:
http://www.diyaudio.com/forums/showthread.php?postid=1664820#post1664820
To be safe I'd probably go with Mackie myself. It has 8 analogue outs like the audiofire 8. Here's the manufacturer page: Mackie Onyx 400F
They not too expensive either and can be had £450 new.
http://www.dolphinmusic.co.uk/shop/flypage/product_id/7071
Or if you don't mind an ex demo model then £379:
http://cgi.ebay.co.uk/B-Stock-Macki...876225&_trksid=p3286.c0.m14&_trkparms=72:1301|66%3A2|65%3A12|39%3A1|240%3A1318
Shinobiwan, still following your thread with great interest but also being occupied with some other time consuming projects.
Yes, the Mackie Onyx 400F gets my recommendation.
Using it for recording, measurement and playback for years now (and abusing it from time to time too), I haven't thought about replacing it – the highest attribute a piece of audio gear can get IMO. There are other places in my chain where I *know* there is a lot room for improvement.
Asked to summarise the ONYX pro's I would highlight three points
- after some burn in, sonically absolutely satisfying "out of the box" (not even made an attempt to mod the electrolytic's what usually is my first "upgrade")
- very, very reliable (though tried with XP only)
- versatile and expandable
As I don't have your experience with such a broad palette of different high quality cards, I really would be interested about your findings if you get a chance to check it out.
Michael
Yes, the Mackie Onyx 400F gets my recommendation.
Using it for recording, measurement and playback for years now (and abusing it from time to time too), I haven't thought about replacing it – the highest attribute a piece of audio gear can get IMO. There are other places in my chain where I *know* there is a lot room for improvement.
Asked to summarise the ONYX pro's I would highlight three points
- after some burn in, sonically absolutely satisfying "out of the box" (not even made an attempt to mod the electrolytic's what usually is my first "upgrade")
- very, very reliable (though tried with XP only)
- versatile and expandable
As I don't have your experience with such a broad palette of different high quality cards, I really would be interested about your findings if you get a chance to check it out.
Michael
Hi,
I've been messing with the ADSP-21161N dev kit today and I've managed to connect it to my Sabre (Buffalo) DAC. The DSP board has a handy pin header (P3) for the I2S output, but I don't think this is in the documentation? This I2S output is the same signal which is fed to the "aux" DAC (AD1852). I can post a photo of the current mess if anyone's interested?
It seems to produce some nice clean FIR filters. The problem is that at a 48KHz "master" sampling rate, it would take quite a few taps to do the 80Hz LPF for the sub. Could I just skip say every 128 samples to produce an effective 375Hz sampling rate for the sub FIR??
I can only assume that the lowest frequency that can pass "through" an FIR filter is based on the sampling rate and the number of taps? eg...
Fs = 48000Hz, so 48 samples per ms...
80Hz LPF = 12.5ms time period...
12.5ms x 48 samples = 600 taps minimum ??
Or, simply 48000Hz / 80Hz?
I'm sure I've got this all wrong, and I know it depends on many other factors, but is this anywhere near the ballpark, or is it out in space?!
As you can see, I'm not so good with the maths stuff, and don't know the first thing about the algebraic functions etc. But, does anyone know how convert the raw pcm filter files produced by DRC into a "fractional text" format which would import into the DSP program? (I have Matlab if that helps).
I know this DSP isn't cut out for long FIR filters, but it should be able to do around 3000 taps per channel (stereo). So, I just need to "paste in" a cropped version of a DRC filter to see if it would work in priniciple....
The DRC filter files are raw pcm (32-bit floats I think), and the format required by VisualDSP++ are in a fractional text file like this...
0.0001362251,
0.0001199989,
0.0001039734,
0.0000874365,
0.0000696176,
0.0000496927,
0.0000267908,
-0.0000000000,
-0.0000316246,
-0.0000690537,
-0.0001132760,
-0.0001652893,
Any ideas how to convert this?
btw, I'm aware that it should (in theory) be possible to import the raw pcm files into the DSP assembly program directly, but I've only just understood how the basic FIR loop and SIMD stuff works!
OzOnE.
EDIT: I was using Acourate for a while, but the trial expired. I would be willing to purchase it in future, but I can't afford to spend on software until I can confirm that my measurement process is working correctly.
I've been messing with the ADSP-21161N dev kit today and I've managed to connect it to my Sabre (Buffalo) DAC. The DSP board has a handy pin header (P3) for the I2S output, but I don't think this is in the documentation? This I2S output is the same signal which is fed to the "aux" DAC (AD1852). I can post a photo of the current mess if anyone's interested?
It seems to produce some nice clean FIR filters. The problem is that at a 48KHz "master" sampling rate, it would take quite a few taps to do the 80Hz LPF for the sub. Could I just skip say every 128 samples to produce an effective 375Hz sampling rate for the sub FIR??
I can only assume that the lowest frequency that can pass "through" an FIR filter is based on the sampling rate and the number of taps? eg...
Fs = 48000Hz, so 48 samples per ms...
80Hz LPF = 12.5ms time period...
12.5ms x 48 samples = 600 taps minimum ??
Or, simply 48000Hz / 80Hz?
I'm sure I've got this all wrong, and I know it depends on many other factors, but is this anywhere near the ballpark, or is it out in space?!
As you can see, I'm not so good with the maths stuff, and don't know the first thing about the algebraic functions etc. But, does anyone know how convert the raw pcm filter files produced by DRC into a "fractional text" format which would import into the DSP program? (I have Matlab if that helps).
I know this DSP isn't cut out for long FIR filters, but it should be able to do around 3000 taps per channel (stereo). So, I just need to "paste in" a cropped version of a DRC filter to see if it would work in priniciple....
The DRC filter files are raw pcm (32-bit floats I think), and the format required by VisualDSP++ are in a fractional text file like this...
0.0001362251,
0.0001199989,
0.0001039734,
0.0000874365,
0.0000696176,
0.0000496927,
0.0000267908,
-0.0000000000,
-0.0000316246,
-0.0000690537,
-0.0001132760,
-0.0001652893,
Any ideas how to convert this?
btw, I'm aware that it should (in theory) be possible to import the raw pcm files into the DSP assembly program directly, but I've only just understood how the basic FIR loop and SIMD stuff works!
OzOnE.
EDIT: I was using Acourate for a while, but the trial expired. I would be willing to purchase it in future, but I can't afford to spend on software until I can confirm that my measurement process is working correctly.
mige0 said:
Thanks Michael.
I was so close to buying the Onyx 400f before eventually settling on the Fireface 800. It looks like a great interface for the money.
Your mention of mods to these sorts of devices had me peering inside the Fireface 800. Sure enough there's some unbranded electrolytic in the analogue sections. I might consider swapping these to see if anything improves. Blackgate would be a likely candidate. I think my second hand unit is already out of warranty so no harm in trying.
The Thread slows down, what happend ?
No more interested users ?
Meanwhile, i played with a Soundblaster Audigy and Console.
But it is frustrating because of the drivers.
My goal was, to get a routing from analog input over
convolver plugin to analog output. Only in stereo, so i can
plug a cdplayer in and test the basics.
The problems start with the basic installation (i skip a few hours
searching). The old drivers can´t find hardware on irqs higher
than 16, but with modern boards with APIC, the card gets
higher irq. With manual install, i am not shure i install all needed.
So console only shows me a "primary sounddevice", no word
about audigy and of course no ASIO.
Now i try the kx project, but with the many manual install and
deinstall of the audigy drivers, xp won´t start at least,
so i have to install xp first.
No more interested users ?
Meanwhile, i played with a Soundblaster Audigy and Console.
But it is frustrating because of the drivers.
My goal was, to get a routing from analog input over
convolver plugin to analog output. Only in stereo, so i can
plug a cdplayer in and test the basics.
The problems start with the basic installation (i skip a few hours
searching). The old drivers can´t find hardware on irqs higher
than 16, but with modern boards with APIC, the card gets
higher irq. With manual install, i am not shure i install all needed.
So console only shows me a "primary sounddevice", no word
about audigy and of course no ASIO.
Now i try the kx project, but with the many manual install and
deinstall of the audigy drivers, xp won´t start at least,
so i have to install xp first.
sonopanic- 1010LT routing setup here
http://www.mp3car.com/vbulletin/hig...tting-up-your-carpc-using-m-audio-1010lt.html
Seems like a very capable card. Wish someone would chime in if they daisy chained more than one 1010LT for more than 8 channels. Manual clearly states it's possible.
frode-
In directwire, make wires from WDM OUT(for DVD software players), input (if using line in), MME OUT (winamp/ foobar) to ASIO IN all in parallel for each channel. Then mute WDM and MME out by clicking out the OUT button. This will "disconnect" the internal routing of the audio stream from the actual hardware outputs so you can use a VST host.
I have used the prodigy hifi with no issues (Windows XP Athlon LE1640, 1G RAM) except if using allocator full. The latency settings do not go high enough to use the arbitrator portion (slight clicking). I do not use DRC and do not plan to. The Prodigy might not have big enough latency settings to use FIR filters. I have no trouble using other basic filtering plugins using standard analog models. There are bigger gains to be had elsewhere (acoustics) outside of electronic manipulation especially DRC/bass mode room correction.
http://www.mp3car.com/vbulletin/hig...tting-up-your-carpc-using-m-audio-1010lt.html
Seems like a very capable card. Wish someone would chime in if they daisy chained more than one 1010LT for more than 8 channels. Manual clearly states it's possible.
frode-
In directwire, make wires from WDM OUT(for DVD software players), input (if using line in), MME OUT (winamp/ foobar) to ASIO IN all in parallel for each channel. Then mute WDM and MME out by clicking out the OUT button. This will "disconnect" the internal routing of the audio stream from the actual hardware outputs so you can use a VST host.
I have used the prodigy hifi with no issues (Windows XP Athlon LE1640, 1G RAM) except if using allocator full. The latency settings do not go high enough to use the arbitrator portion (slight clicking). I do not use DRC and do not plan to. The Prodigy might not have big enough latency settings to use FIR filters. I have no trouble using other basic filtering plugins using standard analog models. There are bigger gains to be had elsewhere (acoustics) outside of electronic manipulation especially DRC/bass mode room correction.
durwood said:
Here's how I've done it:
An externally hosted image should be here but it was not working when we last tested it.
WMP, chose U46 Ch123456:
An externally hosted image should be here but it was not working when we last tested it.
This result in very distorted sound. I used Bidule because my Console trail period is over, but the problem was the same when i tested with Console.
Frode
Frode
Re: Lynx
Thanks for the reminder but I'm not sure the AES16 supports this. The Lynx2 does but is limited to 2 channels.
Here's the updated list.
RME:
HDSP 9632
HDSP 9652
HDSP AES 32
HDSPe AIO
HDSPe RayDAT
HDSPe AES
Digiface
Multiface 2
Fireface 400
Fireface 800
MOTU:
828 mk2 (loopback of upto 8 channels)
828 mk3 (confirmed loopback of 2 channels but possibly 8)
896 mk3 (confirmed loopback of 2 channels but possibly 8)
UltraLite mk3 (confirmed loopback of 2 channels but possibly 8)
TC Electronic
Studio Konnect 48 (confirmed loopback of 2 channels possibly more due to extensive routing capabilities)
EMU:
EMU 1820m (discontinued but an excellent interface that pops up occasionally on ebay for not much money)
EMU 1820 (discontinued but same deal as 1820m)
1616m PCI
1616 PCI
1212m PCI0404 PCI
Audiotrak:
Prodigy 7.1 Hifi (good value and probably one of the most popular to test the waters)
Mackie:
Mackie Onyx 400F
Mackie Onyx 1200F
Lynx:
Lynx TWO C (2 channel loopback only)
mikela said:
Thanks for the reminder but I'm not sure the AES16 supports this. The Lynx2 does but is limited to 2 channels.
Here's the updated list.
RME:
HDSP 9632
HDSP 9652
HDSP AES 32
HDSPe AIO
HDSPe RayDAT
HDSPe AES
Digiface
Multiface 2
Fireface 400
Fireface 800
MOTU:
828 mk2 (loopback of upto 8 channels)
828 mk3 (confirmed loopback of 2 channels but possibly 8)
896 mk3 (confirmed loopback of 2 channels but possibly 8)
UltraLite mk3 (confirmed loopback of 2 channels but possibly 8)
TC Electronic
Studio Konnect 48 (confirmed loopback of 2 channels possibly more due to extensive routing capabilities)
EMU:
EMU 1820m (discontinued but an excellent interface that pops up occasionally on ebay for not much money)
EMU 1820 (discontinued but same deal as 1820m)
1616m PCI
1616 PCI
1212m PCI0404 PCI
Audiotrak:
Prodigy 7.1 Hifi (good value and probably one of the most popular to test the waters)
Mackie:
Mackie Onyx 400F
Mackie Onyx 1200F
Lynx:
Lynx TWO C (2 channel loopback only)
King Nothing said:well I just ordered an fireface 800
Good choice.
With the move back to an RME, I'm reminded why I'm so found of them. The drivers really are rock solid. Everything works and with a of minimum effort. No hunting around tracking down problems, you simply install and it works.
I've had a number of cards and interfaces over the years and quickly realised that this is perhaps the most important attribute.
I stumbled across a very nice utility called WVR for remote controlling the volume of RME products.
It works by watching the master volume control within windows mixer then issues MIDI commands whenever a change is noted. These MIDI commands are then interfaced with RME's Totalmix application to allow level changes.
So you'd have something like an iMON remote installed that controls windows volume and then this software links to the hardware mixer.
RME is particularly suited to this because of its 42bit mixer resolution. Even very low levels of -50dB will have absolutely negligible truncation. I'd argue that active line level attenuation is better than passive for a number of reasons too. Overall its a solution that lays to rest the volume problem of old.
You can download WVR from here: http://home.comcast.net/~lapstand/WVR.zip
Its compatible with Vista and XP as well as all RME products.
Here's the setup instructions:
EDIT: I forgot to add that you need a loopback cable that connects the MIDI IN to the MIDI OUT on your RME for this to work. Here's what I'm talking about:
http://www.ramelectronics.net/music-sound/keyboards-and-midi/midi-cables/midi/prodMIDI.html
It works by watching the master volume control within windows mixer then issues MIDI commands whenever a change is noted. These MIDI commands are then interfaced with RME's Totalmix application to allow level changes.
So you'd have something like an iMON remote installed that controls windows volume and then this software links to the hardware mixer.
RME is particularly suited to this because of its 42bit mixer resolution. Even very low levels of -50dB will have absolutely negligible truncation. I'd argue that active line level attenuation is better than passive for a number of reasons too. Overall its a solution that lays to rest the volume problem of old.
You can download WVR from here: http://home.comcast.net/~lapstand/WVR.zip
Its compatible with Vista and XP as well as all RME products.
Here's the setup instructions:
An externally hosted image should be here but it was not working when we last tested it.
EDIT: I forgot to add that you need a loopback cable that connects the MIDI IN to the MIDI OUT on your RME for this to work. Here's what I'm talking about:
http://www.ramelectronics.net/music-sound/keyboards-and-midi/midi-cables/midi/prodMIDI.html
Hi Shin and the others, the card arrived.
But now i am very confused.
I managed accourate to build filter for six channels with center
as gain reference. So i have six .dbl files.
I have installed console and have downloaded the convolver vst Plugin from sourceforge.
I think, i have to write a configfile for the convolver or is there
another convolver Plugin which is better for 5.1 ?
Which settings i have to do with the PatchMixer ?
I think in this piece i route the channels, right ?
But now i am very confused.
I managed accourate to build filter for six channels with center
as gain reference. So i have six .dbl files.
I have installed console and have downloaded the convolver vst Plugin from sourceforge.
I think, i have to write a configfile for the convolver or is there
another convolver Plugin which is better for 5.1 ?
Which settings i have to do with the PatchMixer ?
I think in this piece i route the channels, right ?
Patchmix, like most of the other full featured mixers out there, can initial appear complicated. Its not however.
You need to create 3 x 2 channel input strips using windows wave as the source, these will 1/2, 3/4, 5/6 and enable multichannel playback over windows sources.
Next you create however many ASIO outputs as you require. This could 4 for a total of 8 channels for example. For each of these you need to assign a physical out using the 'SEND' functionality.
I've attached a 44Khz patchmix session that does what I'm describing. Take a look at what options and strips have been added then try to understand the signal routing. Pretty soon everything will become clear.
You need to create 3 x 2 channel input strips using windows wave as the source, these will 1/2, 3/4, 5/6 and enable multichannel playback over windows sources.
Next you create however many ASIO outputs as you require. This could 4 for a total of 8 channels for example. For each of these you need to assign a physical out using the 'SEND' functionality.
I've attached a 44Khz patchmix session that does what I'm describing. Take a look at what options and strips have been added then try to understand the signal routing. Pretty soon everything will become clear.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.