| Author |
Message |
JovianPyx
Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224
|
 Posted: Sat Mar 19, 2016 4:45 pm Post subject:
|
  |
|
I've added a 16 tap reverb which hopefully will sometime sound great, but at least it's computational load doesn't cause a problem. The synth and the effect runs on the same core in the same thread. It should be possible to have another thread on another core running the same combination - some synth with some effect and then mix the two. This would likely add an additional 1 mSec to the latency, so it would be a total of about 3 mSec for a bitimbral synth with separate effects on each.
I'm also wait for a Raspberry Pi 3 to arrive. I shall write about that after I get it.
_________________
FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
|
|
|
Back to top
|
|
 |
BobTheDog
Joined: Feb 28, 2005
Posts: 4044
Location: England
Audio files: 32
G2 patch files: 15
|
| Posted: Mon May 02, 2016 2:25 am Post subject:
|
|
|
Hi,
I just ordered a Pi 3 and a cirrus logic board.
I'm just wondering if you managed to get your PI 3 and get it working with the CL board?
Cheers
Andy |
|
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224
|
| Posted: Mon May 02, 2016 4:03 am Post subject:
|
|
|
| BobTheDog wrote: | Hi,
I just ordered a Pi 3 and a cirrus logic board.
I'm just wondering if you managed to get your PI 3 and get it working with the CL board?
Cheers
Andy |
Yes, I did. Currently using the pre-compiled 4.4.6 kernel (Arch Linux). In fact, I tested the linux installation on a Rpi2 and it works there too (without modification).
_________________
FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
|
|
|
Back to top
|
|
|
BobTheDog
Joined: Feb 28, 2005
Posts: 4044
Location: England
Audio files: 32
G2 patch files: 15
|
| Posted: Mon May 02, 2016 5:54 am Post subject:
|
|
|
Thanks for the info, that sounds easier than I was expecting!
I was expecting to have to build a kernel. |
|
|
Back to top
|
|
|
BobTheDog
Joined: Feb 28, 2005
Posts: 4044
Location: England
Audio files: 32
G2 patch files: 15
|
| Posted: Mon May 02, 2016 6:00 am Post subject:
|
|
|
| Just had a quick look, and just wanted to check. The version available is only 32bit? |
|
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224
|
| Posted: Mon May 02, 2016 6:13 am Post subject:
|
|
|
| BobTheDog wrote: | | Just had a quick look, and just wanted to check. The version available is only 32bit? |
Yes, at this time there is no 64 bit OS for the pi3. I've not heard much about plans to move on that. However, if ARM technology is moving toward 64 bit universally, I think it's only a matter of time. I'll say this much, the Rpi3 is at least "low end" workstation quality in that it does GUI quite well, is not slow, runs stuff like FireFox and Thunderbird, youtube vids are "watchable", tho not creamy smooth. I plan to use my pi3 this summer as an "AC-on workstation". I live in the southern Californicate desert and my old Pent4 HT machine makes too much heat...
Also, I didn't mention this, but while I did not try it myself, I have read on the forums that Raspbian on pi3 works with the CL board as well - others had posted of their success. So you have a choice of at least 2 distros.
_________________
FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
|
|
|
Back to top
|
|
|
BobTheDog
Joined: Feb 28, 2005
Posts: 4044
Location: England
Audio files: 32
G2 patch files: 15
|
| Posted: Mon May 02, 2016 6:26 am Post subject:
|
|
|
Super, thanks for the info.
When it arrives I will let you know how it goes... |
|
|
Back to top
|
|
|
BobTheDog
Joined: Feb 28, 2005
Posts: 4044
Location: England
Audio files: 32
G2 patch files: 15
|
| Posted: Tue May 03, 2016 11:28 am Post subject:
|
|
|
Hi Again,
Everything arrived and is running Arch 4.4.8-2-ARCH
This does't seem to support the Cirrus Logic, did you get your image from archlinuxarm.org or somewhere else?
Cheers
Andy |
|
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224
|
| Posted: Tue May 03, 2016 11:43 am Post subject:
|
|
|
Right, the basic kernel doesn't support the card. For reasons unknown to me, the driver is compiled into the kernel instead of being a separate loadable driver file.
This thread: http://www.element14.com/community/thread/43711?start=0&tstart=0 is where I found the precompiled kernel. Note that at the time I installed mine, kernel 4.4.7 was available, but the precompiled version is 4.4.6. I've had no problems with the back-version kernel. The guy who maintains this is "hias".
Basically the process is to install Arch first, then use the instructions on hias's site to download and deliver the modified kernel.
_________________
FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
|
|
|
Back to top
|
|
|
BobTheDog
Joined: Feb 28, 2005
Posts: 4044
Location: England
Audio files: 32
G2 patch files: 15
|
| Posted: Tue May 03, 2016 12:00 pm Post subject:
|
|
|
Nice one, thanks for the info.
The stock kernel seems to have lots of overlay drivers for other DACs just not the Cirrus 
I'll have a look at your links.....
Once again thanks
Andy |
|
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224
|
| Posted: Tue May 03, 2016 1:31 pm Post subject:
|
|
|
There's a few things to know and deal with regarding this, so I'll try to barf out what I know.
One is that there are two LEDs on the audio card. The RED LED will remain on as long as the driver is not present/running. As far as I know, the GREEN LED is just power. Once the driver is running, the RED LED goes out. This doesn't mean that the card will work, it just means that at least part of the process is working. If the RED LED is out and you still can't "see" the card, then there is more configuration stuff that needs done or was misspelled or missing.
I use to see if ALSA finds the card. If that works - you should be able to play an mp3 file with something like mplayer.
Make sure you have downloaded the scripts too. The scripts allow you to set the audio channel for (as example) line out. I've done this a few times now, my end-of-line test is mplayer playing an mp3 file. If that works - you're golden. I've also discovered that youtube isn't always very friendly with audio. The card/driver/pi supports most youtube videos, but I encountered one that had static for sound. No idea what that's all about and I don't care much because my purpose for this audio card is music. I've been able to write synth code that achieves a less than 1.5 millisecond latency between MIDI note on receipt and sound start. (this is using the ARM hardware UART as my MIDI device, not a USB device - I've not tested that at all).
In the instructions hias gives, there are a few edits that need to be made to various system files. /boot/config.txt contains a few of them. Of note, I2C may be turned off by default, so it needs an edit to turn it on. I believe all of this is described by hias on his site - at least that's where I found it. There is also a way to disable the HDMI sound device (it's only 10 bits anyway) so that the Cirrus card is the only one that shows up and is therefore also the default sound device (which makes it easy for a lot of applications).
It's wise to go through that whole thread because even if it's for older versions, there is good troubleshoot info in there.
_________________
FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
|
|
|
Back to top
|
|
|
BobTheDog
Joined: Feb 28, 2005
Posts: 4044
Location: England
Audio files: 32
G2 patch files: 15
|
| Posted: Tue May 03, 2016 10:46 pm Post subject:
|
|
|
Hi,
Thanks for all the info, only just seen it. Notifications never work for me on this forum.
I failed last night with arch and got to the "Bloody Linux" stage so gave up!
This morning I just stuck on Raspian Lite 4.1.19 and followed the instructions here: http://www.horus.com/~hias/cirrus-driver.html
The red light is now out and the driver seems to be there:
| Code: |
pi@raspberrypi:~/bin $ aplay -l
**** List of PLAYBACK Hardware Devices ****
card 0: sndrpiwsp [snd_rpi_wsp], device 0: WM5102 AiFi wm5102-aif1-0 []
Subdevices: 1/1
Subdevice #0: subdevice #0
pi@raspberrypi:~/bin $ arecord -l
**** List of CAPTURE Hardware Devices ****
card 0: sndrpiwsp [snd_rpi_wsp], device 0: WM5102 AiFi wm5102-aif1-0 []
Subdevices: 1/1
Subdevice #0: subdevice #0
|
Fingers crossed I can get some audio out now..... |
|
|
Back to top
|
|
|
BobTheDog
Joined: Feb 28, 2005
Posts: 4044
Location: England
Audio files: 32
G2 patch files: 15
|
| Posted: Tue May 03, 2016 11:13 pm Post subject:
|
|
|
Success, audio output working 
Now to get a cross compiler going....
I have another question if it's ok, how have you gone about getting midi going? |
|
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224
|
| Posted: Wed May 04, 2016 7:47 am Post subject:
|
|
|
MIDI... yes...
One thing I didn't want was MIDI over USB. Mainly because the pi has only one USB controller, so all USB traffic is shared on that controller. This same USB controller also operates the ethernet port, so more random traffic... The alternative is the hardware ARM UART. In the pi2, this UART is programmed to be the system console which I don't use. The pi3 is a little different. They added bluetooth support for keyboard/mouse and they used the hardware UART to do that. This was done because the hardware UART has a hardware clock that doesn't change it's rate when the CPU clock changes. Then they put a software "miniUART" in for the system console.
On the pi3 it is still possible to use the hardware UART for MIDI, but it means disabling bluetooth, so if you depend on a bluetooth keyboard/mouse, then this won't make you happy.
Assuming you can live without bluetooth, here is what I did:
You need to make a current loop to logic circuit (standard thing) using an optoisolator like 6N138. The 6N138 logic output is connected to the little 3 pin UART connector on the sound board. The problem with that is the pins for the UART are rx, tx and gnd and no power pin. I had to steal 3.3v power from the larger connector.
Bluetooth must be disabled and the UART needs to be configured to connect to the original GPIO pins. This is done with a device tree overlay file called pi3-disable-bt.dtbo in /boot/overlays. I can't remember if I had to download that file or if it already existed. This file must be enabled by adding
| Code: | | dtoverlay=pi3-disable-bt |
to config.txt
Once this is done, the UART pins on the audio board are connected to the hardware ARM UART.
Bluetooth uses a very high baud rate, so the default UART clock frequency is 48 MHz. This needs to be changed to 3 MHz.
| Code: | | init_uart_clock=3000000 |
If this is all done correctly, you will see a device /dev/ttyAMA0 in the /dev directory. That is the UART device.
Console functions need to be disabled as well. Here is my cmdline.txt which does this:
| Code: | | rootfstype=ext4 root=/dev/sda1 rw rootwait console=tty1 selinux=0 plymouth.enable=0 smsc95xx.turbo_mode=N dwc_otg.lpm_enable=0 elevator=noop |
Please note that the above command also changes the root file system to a hard drive, so you can't just copy/paste it. You will see in your own cmdline.txt mention of console= in 2 places (I think), those must be removed. I believe that console=tty1 is required, but tty1 is just a soft device.
I spent a good deal of time getting this working, there were all kinds of posts about getting the UART to work properly at 31250 baud, but none of that worked due to how the pi has changed over the years. The old methods used to work, but no longer do. The only method I found that worked correctly and could be incorporated into my code was to uses mmap() functions to manipulate the hardware UART device registers directly. Part of the problem is that 31250 has never been considered a "standard" baud rate, even after all of these years. So the only way to get there is to set the UART and it's clock circuitry for a nonstandard baud rate. Essentially, this means dividing the 3 MHz clock by 6. 500000 is exactly 3000000/6 and it's also 16 x 31250, so it's exactly the correct MIDI baud rate.
I can't properly describe how to make that happen in English, it's a bunch of number stuffing into device registers. I will include some code in an attachment that does it for a program called ttymidi. ttymidi is a converter that receives MIDI data from the UART and translates it to an ALSA MIDI port that can be used to connect to MIDI software. I tested with ZynAddSubFX, a MIDI synth that can be downloaded and installed on the pi.
If you look at the ttymidi source code, you should be able to find my hack. And please note - it is a hack, the code no longer accepts baud rates on the command line. There's probably a bunch of crap in it that could be removed, but I didn't care enough to be honest. I used ttymidi only to test as I don't use ALSA MIDI ports in my own synth software. For my synths, the driver for the UART is in the synth code (in the MIDI controller) itself. This is non-standard, so you're free to do it however you like, standard or otherwise.
One of the rubs came during testing of my own synth code. Mysteriously, some MIDI messages never made it into the synth. The reason for this was that the linux UART driver for the console was running and responding to interrupts. This would cause the driver to occasionally grab a byte and the synth never saw it. There was also an issue of the UART's FIFO. The UART would only indicate it had new data when there were already several bytes in the FIFO, and then it would belch them all out at once destroying the MIDI byte delivery timing. This caused funky timing. The ultimate solution was to use mmap() functions to disable the UART's hardware interrupts which then essentially "killed" the driver - if it never receives interrupts - it never steals bytes. But then one can't simply use the read() function to get data from the UART. The way around that is to implement a "polling driver" in the MIDI controller program. The code literally bit spins on the data-ready bit in the UART hardware. Once the bit goes high, the code grabs the new data byte from the device's data register ( using mmap() ) and then is processed by the MIDI controller code. The MIDI controller code runs well fast enough that there's no latency or other timing issues from it. And in fact, I use isolated CPUs to dedicate a core to just the MIDI controller.
Unfortunately, the ttymidi program uses the kernel's UART driver, so it will not have the best timing, however, it will work and can be used as a basic test.
I will also attach the MIDI controller for my karplus-strong synth - but be warned - the code is a bit awful... However, it should demonstrate what I did. Due to the nature of that synth, the bit spin loop that waits for a new MIDI byte does some other stuff. I've tried to comment it, so maybe you can decipher it.
ttymidi can be compiled with:
| Code: | | gcc ttymidi.c -o /usr/local/bin/ttymidi -lasound -lpthread |
The MIDI controller won't compile because it's only one of many files in the project.
ALSO --- The above is all based on an Arch Linux installation. Arch is now on kernel 4.4.x where Raspbian is 4.1.x, so there may likely be differences in the linux structure. The names of the overlay files were changed from .dtb to .dtbo and the word "overlay" was removed from all overlay file names. I have not messed with Raspbian, so I really can't give much advice on specifics for Raspbian.
| Description: |
|
Download (listen) |
| Filename: |
MIDIcont.c |
| Filesize: |
18.29 KB |
| Downloaded: |
387 Time(s) |
| Description: |
|
Download (listen) |
| Filename: |
ttymidi.c |
| Filesize: |
24.52 KB |
| Downloaded: |
345 Time(s) |
_________________
FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
|
|
|
Back to top
|
|
|
BobTheDog
Joined: Feb 28, 2005
Posts: 4044
Location: England
Audio files: 32
G2 patch files: 15
|
| Posted: Wed May 04, 2016 11:13 am Post subject:
|
|
|
Thanks very much for the detailed reply and code, it's very good of you.
I can live without bluetooth so I will have a proper read through and give it a go.
Many thanks
Andy |
|
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224
|
| Posted: Sun May 15, 2016 12:50 pm Post subject:
|
|
|
Project update...
I have a working 24 voice plucked karplus-strong synth with reverb.
I increased the sample rate from 44.1 ksps to 88.2 ksps to raise the synth an octave and get better quality high notes.
I am using 3 cores for the synth and 1 core for Linux.
Core 0 is for Linux
Core 1 is for reverb
Core 2 is for MIDI controller
Core 3 is for synth and DAC DMA RAM feed
At this point the reverb is a 57000 element delay line with 16 taps at prime proportions of the delay line length. With some fooling around in it's config file, I was able to make a fair room sound.
The code runs on both my Rpi2 and Rpi3.
_________________
FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
|
|
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224
|
|
|
Back to top
|
|
|
AlanP
Joined: Mar 11, 2014
Posts: 746
Location: New Zealand
Audio files: 41
|
| Posted: Fri May 20, 2016 5:12 pm Post subject:
|
|
|
| Jupiter has some rockin' muso's |
|
|
Back to top
|
|
|
BobTheDog
Joined: Feb 28, 2005
Posts: 4044
Location: England
Audio files: 32
G2 patch files: 15
|
| Posted: Thu May 26, 2016 1:23 am Post subject:
|
|
|
| The ks24 sounds really good, nice job. |
|
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224
|
| Posted: Thu May 26, 2016 3:31 pm Post subject:
|
|
|
| Blue Hell wrote: | Cool, you have liftoff!
The timing seems a bit odd here and there though?
It could be that the MIDI file was played in by a human tho ... in fact .. odd is not the right word, it's played nicely ... but hard to hear possible timing issues in the synth this way .. or maybe my brains are off .. that happens  |
To revisit this, I've revamped the 32 voice sine organ synth. In the process of doing that, I saw that the MIDI controller was an earlier version in which the UART was controlled by the kernel using interrupts and it had the UART's FIFO enabled. The kernel's driver was written with the idea that the UART would be connected to a terminal, so to use a FIFO and interrupts made sense and didn't cause problems for the user by storing a few bytes before burping them out as a glob of bytes. If this is done with MIDI, variable amounts of latency are introduce and likely caused the problem you refer to.
In the revamp, I've replaced the MIDI controller with updated code that disables UART interrupts and the FIFO. The MIDI controller waits for new MIDI bytes by polling the UART data-available bit and reading the new byte directly from the UART hardware register (using mmap). This code causes bytes to be processed by the MIDI controller as soon as they are received.
I'm going to produce a new demo once I've added reverb to the organ synth.
Other improvements made include variable second partial flat to sharp detune (using CC) and conversion of integer based fixed point arithmetic to float (this causes the compiler to use the NEON floating point processor instructions).
_________________
FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
|
|
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224
|
| Posted: Fri May 27, 2016 11:48 am Post subject:
|
|
|
I got reverb (such as it is) working on the sine organ today, so I recorded a demo. It's the same piece as when I first demoed it (Handel's Air and Variations in E Major), but I've added a second part an octave below the first to fill it out.
Next is to do a better reverb and maybe some other special effects.
EDIT ADD: I've also added a Bach fugue.
| Description: |
| Handel's Air and Variations in E Major |
|
Download (listen) |
| Filename: |
handel-airnvari.mp3 |
| Filesize: |
3.72 MB |
| Downloaded: |
795 Time(s) |
| Description: |
| Fugue in G Major - J. S. Bach |
|
Download (listen) |
| Filename: |
fugue_in_G_major_JSBach.mp3 |
| Filesize: |
3.85 MB |
| Downloaded: |
773 Time(s) |
_________________
FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
|
|
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224
|
| Posted: Wed Aug 03, 2016 10:05 am Post subject:
|
|
|
I've taken a break from synth development. For one thing, my 15 year old "main" workstation finally gave up the ghost and since I had no money at the time, I decided to try something brave and daring. I'd accumulated 4 raspberry boards, three are pi3B and one is pi2B. I've built a server out of one pi3B and a workstation out of another pi3B.
So far, I'm actually satisfied with the performance of the server. It supports Apache2 web server which I use for accessing locally stored documentation for software and electronics. It also supports Postfix email server, Fetchmail email fetch daemon, and Samba file server for any Windows machines on my network.
I plan to buy a fast Intel based machine to run Windows and Linux, but for the time being, I'm using a pi3B for that. As a workstation, the pi3 is more than adequate do to C development. My synth C programs compile in 1 to 3 seconds depending on complexity, so using this environment I have acceptably fast edit-compile-test cycles. I am running Arch Linux with Xorg (LXDE desktop). Using Midori as a light weight web browser, I get acceptable performance when accessing sites with little or no advertising such as linux, programming, mathematics, etc. sites for documentation. However, using it to read world news sites is very sluggish, so I'm just patient... One really good thing is that it's summer here in Los Angeles as I write this and not having the Intel workstation on reduces the heat in my lab which allows the AC to cool the room about 5 degrees F lower (considering I also got rid of a CRT TV set). I'm working to finish all of the server conversion so I can shut my intel server down too for further heat reduction. At about 35 watts or so, the raspberry pi makes very little heat compared to Intel machines (over 400 watts), so this will be my summer mode of operation. When I get my Intel workstation, I can use it for a heater in winter. If I find performance problems with the server, I can always copy project files from the server to the workstation and update the server as things change since the server can do backups. Both the server and the workstation use USB 1 TB hard drives.
Anyway, that's where I am right now. I'm quite pleased mostly with these boards, the only downside being slow access to world news sites so I'm both patient and careful about the selection of news stories to read.
_________________
FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
|
|
|
Back to top
|
|
|
Dan Lavin
Joined: Nov 09, 2006
Posts: 649
Location: Spring Lake, Mi, USA
Audio files: 21
|
| Posted: Thu Aug 04, 2016 6:16 am Post subject:
|
|
|
Scott, Good job on the home Pi set-up! I've thought about a set-up like yours with the Pi's, but I have too much invested in 4 Windows machines at this point. Have you considered a cheapo Android tablet for news? I find that with my tablet, I rarely turn on the Win machines at home anymore. I just use Win for my DAW and audio editing. My wife uses one Win machine and the other 2 now are back-ups or in convenience places around the house, but rarely are used.
_________________
Synth DIY since 1977! |
|
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224
|
| Posted: Thu Aug 04, 2016 8:37 am Post subject:
|
|
|
| Dan Lavin wrote: | | Scott, Good job on the home Pi set-up! I've thought about a set-up like yours with the Pi's, but I have too much invested in 4 Windows machines at this point. Have you considered a cheapo Android tablet for news? I find that with my tablet, I rarely turn on the Win machines at home anymore. I just use Win for my DAW and audio editing. My wife uses one Win machine and the other 2 now are back-ups or in convenience places around the house, but rarely are used. |
No, I'm not a fan of mobile devices because of cost, the screen size (vision problem) and other idiosyncracies that I find annoying including wifi (all of my lab devices including the pi boards are wired). There's also "another thing to waste time learning that I'd use only rarely". I usually read news only in the morning for an hour or so. This pi workstation solution is only for the hot summer months here, I would far prefer to use a Windows machine with a big monitor except for the heat. As it is, for news, while the pi3 is a bit slow for news sites, it's fast enough for doing the research I need on software, electronics and mathematics, so the slow response to "fat" websites is a mild inconvenience.
As for the server, I'd prefer to keep that on a pi if it stays reliable which it has been. I don't need a screamer and I need the heat from it even less. So even after I get an 8 core machine (I do development on FPGA too which needs a LOT of CPU horsepower), I'll still use the pi server so that backups are automatic - it can even reach out and backup a workstation if it 's online. I'm very lazy as well, I leave my workstation running so that when I sit down, I can just use it instead of waiting for it to boot. I do this because I'm always doing something with a computer (whether windows or linux). Yes, of course, with a big nasty Intel machine, I'll have to shut that down when I'm not using it due to the power consumption, however, the pi will still be available (I have a 4 port HDMI switch) to use while the Intel boots up. Some of this chaos is due to the cost of electricity going up here because California refuses to replace the nuclear reactors that have gone out of service.
_________________
FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
|
|
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224
|
| Posted: Wed Oct 05, 2016 7:35 am Post subject:
|
|
|
Latest development is a tonal noise monosynth with portamento and a tonal noise polysynth. Both are based on a simple patch of white noise put through 2 series state variable bandpass filters. The polysynth was tested with 16 voices configured without problems.
Both of these synths were built on the existing synthesizer framework code that I developed for the sine organ and karplus-strong synths.
Once all of the bumps are smoothed out, I'll post some sound samples.
_________________
FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
|
|
|
Back to top
|
|
|
|
Forum index » DIY Hardware and Software » Developers' Corner
FAQ
Calendar
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
Chat Room
