Home Theater Forum and Systems banner

1 - 2 of 2 Posts

·
Registered
Joined
·
9 Posts
Discussion Starter #1
In the controls of the impules graph there are 3 t=0 offset adjustments

t=0 offset (samples)
t=0 offset (ms)
t=0 offset (distance)

1. The "t=0 offset (distance)" does not accept negative distances(with the up/down buttons)
2. The "t=0 offset (distance)" changes "-1 ft" into "-11 in"
3. The "t=0 offset (distance)" does not like partial inches - it rounds 0.5 up to 1 and -0.5 up to 0
4. The "t=0 offset (distance)" would be nice in metric units (0.0 mm would be great - 0.1mm is 2 degrees phase at 20 KHz)
5. The "t=0 offset (distance)" messes up the impulse - changes its shape (move left and right a few inches to see this)
6. The "t=0 offset (distance)" moves the "red dashed line timing reference" relative to the impulse (move left and right a few inches to see this)
7. The "t=0 offset (distance)" assumes speed of sound of 343 m/s but this is not mentioned anywhere in the docs
It might seem like nothing but this 343 versus 344 is 0.3mm per meter which is 13 degrees @ 20 KHz @ 2 metres
It would be nice to be able to change this - ARTA uses 343.85 changeable and HOMLImpulse uses 344 fixed.

The last thing is the spike impulse when using a sweep with a from frequency of less than 20 Hz. Tried with HOLMImpulse and result is fine so problem is not my hardware. Note this test was performed with a digital loopback as reference and signal
Stange impulse.png

View attachment Strange Impulse.mdat
 

·
Premium Member
Joined
·
6,711 Posts
1. The "t=0 offset (distance)" does not accept negative distances(with the up/down buttons)
2. The "t=0 offset (distance)" changes "-1 ft" into "-11 in"
The distance spinner didn't handle negative distances when in feet and inches mode, I've fixed that for the next beta.

3. The "t=0 offset (distance)" does not like partial inches - it rounds 0.5 up to 1 and -0.5 up to 0
The units are inches

4. The "t=0 offset (distance)" would be nice in metric units (0.0 mm would be great - 0.1mm is 2 degrees phase at 20 KHz)
The distance units can be changed between metres and feet in the room sim or the EQ window, I've added another control in the View preferences for the next beta. The resolution of those distance spinners is 1 inch or 1 cm though.

5. The "t=0 offset (distance)" messes up the impulse - changes its shape (move left and right a few inches to see this)
Fractional sample shifts in sampled data cause resampling, that is just how sampling works. If you select "Show points when zoomed in" the individual samples will be displayed when the time spans are short enough for them to be distinguished.

6. The "t=0 offset (distance)" moves the "red dashed line timing reference" relative to the impulse (move left and right a few inches to see this)
Fractional sample shifts weren't handled correctly, I've fixed that for the next beta.

7. The "t=0 offset (distance)" assumes speed of sound of 343 m/s but this is not mentioned anywhere in the docs
It might seem like nothing but this 343 versus 344 is 0.3mm per meter which is 13 degrees @ 20 KHz @ 2 metres
It would be nice to be able to change this - ARTA uses 343.85 changeable and HOMLImpulse uses 344 fixed.
I've added a note about the value used to the Impulse graph help.

The last thing is the spike impulse when using a sweep with a from frequency of less than 20 Hz. Tried with HOLMImpulse and result is fine so problem is not my hardware. Note this test was performed with a digital loopback as reference and signal
REW's sweeps start an octave below the selected start and end an octave above the selected end (or at Nyquist), so that artefacts from band limited sweeps do not affect the measurement range. If the selected start frequency is below 20 Hz the signal begins with a linear sweep from DC to 10 Hz, followed by a logarithmic sweep from there to the end frequency. That avoids spending excessive time at very low frequencies. For a measurement to produce a 'perfect' impulse the stimulus and response both need to span the full range from DC to Nyquist. A digital loopback ensures the response matches the stimulus, a start frequency lower than 20 Hz ensures the stimulus covers the full frequency range. Start frequencies at or above 20 Hz will show ripples that correctly reflect the reduced sweep bandwidth.
 
1 - 2 of 2 Posts
Top