Update: I am planning to release a new version of ACD in a few weeks time. It probably won't include the heatmap, at least initially - it's rather complicated to implement and yo need to use an external program (GNUplot) to generate the plot. The level of complication has kept me from releasing it.
The upcoming release should include the ability to generate a line plot and a polar plot. These are used to display the system response for multiple axes while you are designing the crossover.
The upcoming release should include the ability to generate a line plot and a polar plot. These are used to display the system response for multiple axes while you are designing the crossover.
Latest News: I have updated and expanded the tutorial and am about 90% done. I need to have this ready before releasing anything.
I plan to release the new LADSPA based tools in two stages: first, the "on axis" part of the tools, which are similar to the previous ACD version. In the second stage I will release the "off-axis" part of the tools along with another tutorial for setting up the off axis loudspeaker model and the directivity plotting worksheet. There may be a third stage as well, in which I release the code for the GNUplot plotting, since that gets its input data from the directivity worksheet where it is generated by a VBA macro. It takes quite a while to work up an example and then do it again and document the whole thing to create a tutorial.
So, rollout will happen in stages over the next few months. I will launch a new thread for ACD-L (the LADSPA version of ACD) in a few weeks when things begin to be made available.
I plan to release the new LADSPA based tools in two stages: first, the "on axis" part of the tools, which are similar to the previous ACD version. In the second stage I will release the "off-axis" part of the tools along with another tutorial for setting up the off axis loudspeaker model and the directivity plotting worksheet. There may be a third stage as well, in which I release the code for the GNUplot plotting, since that gets its input data from the directivity worksheet where it is generated by a VBA macro. It takes quite a while to work up an example and then do it again and document the whole thing to create a tutorial.
So, rollout will happen in stages over the next few months. I will launch a new thread for ACD-L (the LADSPA version of ACD) in a few weeks when things begin to be made available.
Just keep in mind that the new release no longer supports the miniDSP products. It's meant to be used with ecasound and LADSPA under linux, for instance on a Raspberry Pi.
Ah, OK.. well I'll continue on with the current version I suppose, thanks.
Noooo..... don't say it's so!! Why...?!
I've moved on... now doing DSP in software.
The "old" ACD is still alive and well. It will remain as a separate download. If there is continued interest in the miniDSP platform and people want the new features that I building into ACD-L I can make a revised version of ACD at some point.
Noooo..... don't say it's so!! Why...?!
I thought about it and realized that you CAN use ACD-L to generate the miniDSP advanced biquad input. The transfer function coefficients are still calculated, but the sheet "miniDSP biquads" has been removed. You can find the biquad values, in standard (not miniDSP) format, in the BiquadCalcs sheet at the bottom of each column below where it says "Z-transform Transfer Function with each Coefficient divided by a0, and b0, b1 and b2 normalized to give the specified gain". PLEASE NOTE THAT THESE ARE NOT IN THE MINIDSP FORM. The miniDSP nomenclature reverses the signs of the a1 and a2 terms compared to the typical Direct Form I nomenclature. So, to use the values in a miniDSP product you would need to copy out the set of biquad values for each filter (b0,b1,b2,a1,a2), change the sign of a1 and a2, and finally add in the text around each value and the "biquad1" text before each set, that is required by miniDSP. I'm not really planning on restoring that functionality at this time. The problem is that I have increased the number of filters significantly (to 20 in both system and driver) because, unlike the miniDSP units, there is no hard-coded limit to the number of filters you can use in software DSP. For instance, the 2x4 allows only 6 filters stages and 8 EQ stages for the input, which you must manually load anyway.
I've found that software DSP (for instance on the Raspberry Pi) has some unique advantages and is less expensive for my needs. I'm not shutting the door on miniDSP, just no longer designing ACD specifically for them.
The ACD tools are open and editable, so you can just use the old ACD minDSP_Biquads sheet as an example and make your own version in ACD-L.
I don't remember, but I'm pretty sure the human mind images the world and the objects in it in 3-D.
So whenever you need to represent more than 2-axises on a graph it's better to use a 3-d graph, than a 2-D graph.
For example, it's easier to sculpt an object than to draw an object.
Jimmy, you might want to look back the first few posts in this thread for examples of 3D plots. They didn't seem to be very well received.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.