New version 2018-06-17:
Windows portable (.zip)
Other (.jar)
- Some graphs on speaker where missing after refresh
- Prevent adding/moving low/high pass filters or L-Pad below speaker, because it doesn't work as might be expected
- Copy/paste now works between multiple instances
Windows portable (.zip)
Other (.jar)
New version 2018-06-18:
Windows portable (.zip)
Other (.jar)
- Fixed: missing refresh after item has been deleted
- Removed unnecessary refresh on item move
Windows portable (.zip)
Other (.jar)
New version 2018-07-04:
Windows portable (.zip)
Other (.jar)
My future plan is more or less to find and fix bugs, cleanup code and publish code under GPLv3.
Feedback would be very helpful (but it looks like nobody is using it
).
- IEC 268-5 is now configurable in settings and is disabled by default (better be pessimistic)
Windows portable (.zip)
Other (.jar)
My future plan is more or less to find and fix bugs, cleanup code and publish code under GPLv3.
Feedback would be very helpful (but it looks like nobody is using it
).
I wanted to thank you for adding the aperiodic mode to the simulation.
I didn't find much in the way of bugs, it seems pretty stable.
The tree-based electrical model makes it seem like the software is fundamentally limited from the outset, which is a problem when you want to modify the crossover in ways that aren't already programmed in. It's hard to understand how where the component is placed in the tree corresponds to it's placement in a schematic.
I do like the fact that at every level of the model you can view the charts for various measurements. This makes a lot of sense to me. For instance with Boxsim, you get only the measurements Boxsim considers important and taken from only certain places. SpeakerSim puts the user in charge of managing complexity rather than the software. The user can just ignore the charts he doesn't want to deal with at the moment. Some people might be intimidated by the amount of data but I think this is a good way for SpeakerSim to differentiate itself from other tools.
In the end SpeakerSim is easier for me to pick up and immediately start doing what I want. BoxSim has dialogues that disable each other or refer to other dialogues in different menus which always has me worried that some setting in some forgotten dialogue is messing up my simulation.
I didn't find much in the way of bugs, it seems pretty stable.
The tree-based electrical model makes it seem like the software is fundamentally limited from the outset, which is a problem when you want to modify the crossover in ways that aren't already programmed in. It's hard to understand how where the component is placed in the tree corresponds to it's placement in a schematic.
I do like the fact that at every level of the model you can view the charts for various measurements. This makes a lot of sense to me. For instance with Boxsim, you get only the measurements Boxsim considers important and taken from only certain places. SpeakerSim puts the user in charge of managing complexity rather than the software. The user can just ignore the charts he doesn't want to deal with at the moment. Some people might be intimidated by the amount of data but I think this is a good way for SpeakerSim to differentiate itself from other tools.
In the end SpeakerSim is easier for me to pick up and immediately start doing what I want. BoxSim has dialogues that disable each other or refer to other dialogues in different menus which always has me worried that some setting in some forgotten dialogue is messing up my simulation.
Nice to hear it is usable.
About aperiodic I would just like to point out it is tuned (to volume). You have to use correct vent dimensions otherwise results are slightly different.
I'm thinking how to improve that (make it configurable), because traditional (cut-and-try) aperiodic enclosures are usually tuned higher.
Tree-based model is a compromise between free form editor and fully predefined. Free form editor is too complex for me (maybe some day).
I was thinking about adding circuit sketch to make it easier to understand, but I kinda gave up after I failed to implement room sketch. I might try again, but I'm really not into graphics stuff.
About aperiodic I would just like to point out it is tuned (to volume). You have to use correct vent dimensions otherwise results are slightly different.
I'm thinking how to improve that (make it configurable), because traditional (cut-and-try) aperiodic enclosures are usually tuned higher.
Tree-based model is a compromise between free form editor and fully predefined. Free form editor is too complex for me (maybe some day).
I was thinking about adding circuit sketch to make it easier to understand, but I kinda gave up after I failed to implement room sketch. I might try again, but I'm really not into graphics stuff.
If you supply netlist files, SPICE programs will let you run AC analysis on a given circuit. Run the analysis to determine Zin, Zout and gain, then the user can create any kind of bizzarre crossover they want.
I envision the process like this:
SpeakerSim provides the user with a template LTSpice schematic. This contains a generic crossover of whatever type was selected. The input and output nodes have fixed names which the user must not change.
The user modifies the schematic.
SpeakerSim runs the simulator in console mode on the schematic to generate a netlist and simulation output, say a .raw file which it can interpret for the necessary data.
LTspice seems like a good candidate:
Command Line Switches
I envision the process like this:
SpeakerSim provides the user with a template LTSpice schematic. This contains a generic crossover of whatever type was selected. The input and output nodes have fixed names which the user must not change.
The user modifies the schematic.
SpeakerSim runs the simulator in console mode on the schematic to generate a netlist and simulation output, say a .raw file which it can interpret for the necessary data.
LTspice seems like a good candidate:
Command Line Switches
I'm never used SPICE. I tried LTspice now, but that's really complex (I was unable to run simulation for simple high pass filter).
I like the idea, because advanced users could use their favorite editor and I don't have to make any GUI.
If I would add something like "Custom circuit" item in current design it would be limitless for one speaker, but it would not allow serial crossovers.
However, I would need SPICE library in Java, but I have only found JSpice which does not support AC sweep (I think that is what I need).
Running LTspice from command line isn't an option as it would not work on all platforms.
It seems it won't be possible.
I like the idea, because advanced users could use their favorite editor and I don't have to make any GUI.
If I would add something like "Custom circuit" item in current design it would be limitless for one speaker, but it would not allow serial crossovers.
However, I would need SPICE library in Java, but I have only found JSpice which does not support AC sweep (I think that is what I need).
Running LTspice from command line isn't an option as it would not work on all platforms.
It seems it won't be possible.
New version 2018-07-17:
Windows portable (.zip)
Other (.jar)
- Bugfix: baffle, driver position, enclosure values did not refresh when selecting different speakers (introduced with 2018-06-16)
Windows portable (.zip)
Other (.jar)
But it is not same binary, it would be tricky handle all possible configuration. It would be much better to have self-contained solution.
Thanks, I will send you PM, if I get there.
This feature is definitely on my wish to implement list (I'm not very optimistic about these features), like room sketch, circuit sketch, quarter wave simulator. I don't expect it to be soon.
There are some other possibilities I can try, like converting more complete C# library to Java.
Can't reproduce this (even on Windows 8.1). Is it possible that it was still processing?
You have selected "Speaker (Visaton BG20)" and wanted to change "Enclosure" type or see "Enclosure response" graph?
New version 2018-07-25:
Windows portable (.zip)
Other (.jar)
- Add item above currently selected item (very convenient )
- L-Pad can now calculate resistors from decibels
- Bugfixes
Windows portable (.zip)
Other (.jar)
New version 2018-08-13:
Windows portable (.zip)
Other (.jar)
- Excursion graph is now shown only when it makes sense (only one speaker below selection)
- Zobel filter added
Windows portable (.zip)
Other (.jar)
Important new version 2018-08-18:
Windows portable (.zip)
Other (.jar)
- Critical impedance simulation bug fixed
- Causing invalid filter simulation
- Present from first released version
- Many other less critical bugs fixed
Windows portable (.zip)
Other (.jar)
Last edited:
New version 2018-08-19:
Windows portable (.zip)
Other (.jar)
- New active filter: Delay
- AES2-1984 power filter (when loading older projects, power filter settings are reset)
- Per driver power filter
- Volume occupied by port is shown in Port calculator
- Bugfixes
Windows portable (.zip)
Other (.jar)
New version 2018-08-30:
Windows portable (.zip)
Other (.jar)
- Rectangular drivers
- Bugfixes
Windows portable (.zip)
Other (.jar)