#linuxcnc-devel | Logs for 2019-04-21
Back
[01:00:32] -!- c-log has quit [Ping timeout: 246 seconds]
[01:06:57] -!- c-log has joined #linuxcnc-devel
[02:53:43] -!- ve7it has quit [Remote host closed the connection]
[03:14:02] -!- selroc has joined #linuxcnc-devel
[03:19:10] -!- selroc has quit [Quit: Leaving]
[04:00:14] -!- c-log has quit [Ping timeout: 250 seconds]
[04:03:20] -!- c-log has joined #linuxcnc-devel
[11:47:09] -!- JT-Shop has quit [Remote host closed the connection]
[11:49:15] -!- JT-Shop has joined #linuxcnc-devel
[13:11:18] -!- c-log has quit [Ping timeout: 245 seconds]
[13:13:45] -!- c-log has joined #linuxcnc-devel
[13:48:46] -!- jthornton has quit [Read error: Connection reset by peer]
[13:48:47] -!- JT-Shop has quit [Read error: Connection reset by peer]
[13:49:21] -!- jthornton has joined #linuxcnc-devel
[13:49:23] -!- JT-Shop has joined #linuxcnc-devel
[13:49:29] -!- JT-Shop2- has quit [Read error: Connection reset by peer]
[13:49:56] -!- JT-Shop2- has joined #linuxcnc-devel
[15:28:52] <rene_dev_> Pcw I recall there is a mesa solution for laser rastering
[15:59:13] <pcw_home> Yes, the DPainter module but no driver/sensible scheme for feeding data (embedded in Gcode? separate stream? etc)
[16:00:56] <pcw_home> Basically a stepgen clone that clocks out data (PWM or bitstream) from a FIFO
[16:02:04] <pcw_home> intended to be position locked to an axis command,encoder,or calculated surface length for arbitrary shape painting
[16:11:43] <rene_dev_> is that included with the hostmot source?
[16:12:09] <pcw_home> yes
[16:12:16] <rene_dev_> I could write a driver for it, if no one else is up to it.
[16:12:33] <rene_dev_> any documentation? :D
[16:12:38] <pcw_home> what do you do about the data stream?
[16:13:37] <rene_dev_> that would come straight from a image, as a 2d grayscale.
[16:13:48] <rene_dev_> and I guess that needs to be fed into the fifo?
[16:14:06] <pcw_home> yeah no hal FIFOs on LinuxCNC
[16:14:28] <pcw_home> I guess Machinekit has ring buffers
[16:15:11] <rene_dev_> do you have a register map?
[16:15:14] <rene_dev_> could not find it here
[16:15:15] <rene_dev_> https://github.com
[16:15:40] <pcw_home> probably only in recent source from Mesa
[16:22:36] <pcw_home> http://freeby.mesanet.com
[16:28:46] <rene_dev_> dpainter count is sampled by host read
[16:29:09] <rene_dev_> how do I clock it?
[16:31:13] <pcw_home> the clock is the same as a stepgen
[16:32:05] <pcw_home> feedback is the same also (16 bits of clock and 16 bit of fractional clock)
[16:32:42] -!- jthornton has quit [Read error: Connection reset by peer]
[16:32:53] -!- JT-Shop has quit [Read error: Connection reset by peer]
[16:32:53] -!- JT-Shop2- has quit [Read error: Connection reset by peer]
[16:33:01] <rene_dev_> ok, so it just shifts at a fixed data rate?
[16:33:31] -!- jthornton has joined #linuxcnc-devel
[16:33:34] -!- JT-Shop2- has joined #linuxcnc-devel
[16:33:41] -!- JT-Shop has joined #linuxcnc-devel
[16:33:49] <pcw_home> like a stepgen (changeble every servo thread)
[16:34:22] <pcw_home> 32 bit rate register
[16:38:01] <pcw_home> imagine a step/dir driven system with N pixels/step, if you fed the stepgen and dpainter from the same position stream you would have 1 pixel clocked out per step
[16:39:29] <pcw_home> but of course you can scale the position stream to have any desired ratio (or just run a a fixed rate if that's what you want)
[16:48:43] <pcw_home> bbl supper
[16:55:49] <rene_dev_> pcw_home https://github.com
[17:59:50] -!- ve7it has joined #linuxcnc-devel
[19:15:00] -!- skunkworks has quit [Ping timeout: 246 seconds]