Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[mattkhan]

I want to measure my convolution setup to verify that what is coming out of each channel is "correct". My device (focusrite saffire pro 24) presents 16 asio channels of which the last 2 are its loopback channels. I currently configure as follows

REW output: DAW7
saffire mixcontrol: DAW7 -> Loopback1 (aka asio channel 14)
jriver asio line-in: channel offset 14, 1 channel

This gives me a single channel playing through JRiver. Am I right in thinking that routing that channel through R/C/SL/SR requires using PEQ to move the signal from L to the channel I want to measure & have PEQ above convolution in the DSP plugin list? (i.e. like the attached pic) or is there another way to do this?

Thanks
Matt

[mwillems]

I want to measure my convolution setup to verify that what is coming out of each channel is "correct". My device (focusrite saffire pro 24) presents 16 asio channels of which the last 2 are its loopback channels. I currently configure as follows

REW output: DAW7
saffire mixcontrol: DAW7 -> Loopback1 (aka asio channel 14)
jriver asio line-in: channel offset 14, 1 channel

This gives me a single channel playing through JRiver. Am I right in thinking that routing that channel through R/C/SL/SR requires using PEQ to move the signal from L to the channel I want to measure & have PEQ above convolution in the DSP plugin list? (i.e. like the attached pic) or is there another way to do this?

Thanks
Matt

Forgive me if you've already discussed this in a prior post, but did you try using REW's ASIO output send a signal directly to JRiver's ASIO driver instead of passing through your soundcard loopback?

[mattkhan]

Forgive me if you've already discussed this in a prior post, but did you try using REW's ASIO output send a signal directly to JRiver's ASIO driver instead of passing through your soundcard loopback?

no I haven't tried that, I thought using >1 ASIO device at once was not possible. Has this changed?

[mwillems]

no I haven't tried that, I thought using >1 ASIO device at once was not possible. Has this changed?

I'm not sure I understand the limitation you're describing (or how it applies to this case). JRiver has it's own ASIO driver that can be chosen as an output in programs that use ASIO.  The sound sent to the JRiver ASIO driver gets picked up as an input by JRiver (like ASIO line in), and then gets passed through JRiver's audio chain to whatever the output is.  It's very similar to how ASIO line in works, but it's initiated by an external playback source. 

Basically, if ASIO line in works, I don't know why this wouldn't work?

[mattkhan]

I'm not sure I understand the limitation you're describing (or how it applies to this case). JRiver has it's own ASIO driver that can be chosen as an output in programs that use ASIO.  The sound sent to the JRiver ASIO driver gets picked up as an input by JRiver (like ASIO line in), and then gets passed through JRiver's audio chain to whatever the output is.  It's very similar to how ASIO line in works, but it's initiated by an external playback source.  

Basically, if ASIO line in works, I don't know why this wouldn't work?

the problem is that REW doesn't let you pick >1 ASIO device & jriver asio only offers an output path so I could send a signal out but then I can't capture the recorded signal. Hence if you pick the jriver driver then REW greys out the input side.

[mwillems]

the problem is that REW doesn't let you pick >1 ASIO device & jriver asio only offers an output path so I could send a signal out but then I can't capture the recorded signal. Hence if you pick the jriver driver then REW greys out the input side.

Oh I see, I thought you were describing an issue with JRiver's ASIO implementation that's what confused me; I don't use REW for measurement very often (because it lacks certain features I need) so I missed the issue.  Now that I see it, I'm not sure how to work around it in REW, so you may be stuck copying the output in PEQ.

FWIW, if you're willing to try a different measurement suite, Holm Impulse will let you measure with different ASIO devices on input and output if you just need to do a sanity check measurement, although it unfortunately only supports 4 channels out directly (I think), so you may be stuck with PEQ routing for some channels even so.

[mattkhan]

Oh I see, I thought you were describing an issue with JRiver's ASIO implementation that's what confused me; I don't use REW for measurement very often (because it lacks certain features I need) so I missed the issue.  Now that I see it, I'm not sure how to work around it in REW, so you may be stuck copying the output in PEQ.

FWIW, if you're willing to try a different measurement suite, Holm Impulse will let you measure with different ASIO devices on input and output if you just need to do a sanity check measurement, although it unfortunately only supports 4 channels out directly (I think), so you may be stuck with PEQ routing for some channels even so.

OK thanks, I'll take a look at holmimpulse.

[mattkhan]

FWIW, if you're willing to try a different measurement suite, Holm Impulse will let you measure with different ASIO devices on input and output if you just need to do a sanity check measurement, although it unfortunately only supports 4 channels out directly (I think), so you may be stuck with PEQ routing for some channels even so.

I did try HolmImpulse quickly last night and I think I confirmed impulse alignment (of L and SW) using the time lock feature, v handy. One Q for you, do you happen to know what the markers in the attached pic are? I can't see it mentioned in the user guide.

[mwillems]

I did try HolmImpulse quickly last night and I think I confirmed impulse alignment (of L and SW) using the time lock feature, v handy. One Q for you, do you happen to know what the markers in the attached pic are? I can't see it mentioned in the user guide.

I'll be honest, I've never seen those markers, but I don't ordinarily ever look at impulse responses with a scale in the 100's of meters.  I usually only look in the centimeter/low meter range, so it's possible I just missed them. The first one appears to be over a hundred "meters" backwards in time, or close to half a second before the impulse (unless I'm misreading your graph).  The "2" might be the point at which Holm started playing the sweep, and the half-second interval between that and the impulse may be the software and hardware latency and time of flight delay.  But that doesn't explain the higher numbers, unless you took 11 or more measurements?

Those are just guesses, I'll play around when I get home tonight and see if I can figure out what they are.  

[mattkhan]

I'll be honest, I've never seen those markers, but I don't ordinarily ever look at impulse responses with a scale in the 100's of meters.  I usually only look in the centimeter/low meter range, so it's possible I just missed them. The first one appears to be over a hundred "meters" backwards in time, or close to half a second before the impulse (unless I'm misreading your graph).  The "2" might be the point at which Holm started playing the sweep, and the half-second interval between that and the impulse may be the software and hardware latency and time of flight delay.  But that doesn't explain the higher numbers, unless you took 11 or more measurements?

Those are just guesses, I'll play around when I get home tonight and see if I can figure out what they are.  

no I'd just taken 5-6 measurements, they appear on the 1st measurement only anyway. I think it's just a curiosity but thought I'd ask.

I didn't get anywhere using the JRiver ASIO driver though, clicking measure bought up the usual warning (protect your ears) screen and then did nothing. I did however use the time lock feature to show that the alignment (of main channels) appears to have no worked at all so back to the drawing board on that one. Thanks for suggesting holmimpulse, that is a v useful feature.

[mwillems]

no I'd just taken 5-6 measurements, they appear on the 1st measurement only anyway. I think it's just a curiosity but thought I'd ask.

I didn't get anywhere using the JRiver ASIO driver though, clicking measure bought up the usual warning (protect your ears) screen and then did nothing.

Interesting; I've had good luck with it so I'm not sure what's up.  You might have to open ASIO line before starting Holm, as Holm is sometimes a little "grabby" with sound interfaces when it's open.  If you're inclined to give it another go, I'd be curious to know what you see happening in JRiver when the sweep is played (i.e. does JRiver think it's playing something, or does playback not even initialize?).  

I did however use the time lock feature to show that the alignment (of main channels) appears to have no worked at all so back to the drawing board on that one. Thanks for suggesting holmimpulse, that is a v useful feature.

Glad you like it    Holm's pretty great, and the time lock works a treat in most cases (sometimes it's a little flaky with USB microphones due to clock drift).

Another great feature that Holm has, which REW does not, is frequency dependent windowing, which is similar to the advanced/psychoacoustic filtering that Acourate and Audiolense do.  It's not identical (the pay suite filtering is pretty sophisticated), but taking the same raw measurement in the Audiolense demo and Holm, the results of FRD filtering in Holm look pretty similar to the TTD filtering in audiolense, etc.

It's a shame that Holm has more or less stopped development of the software, but thankfully it's more or less complete as is (and free!).