c vw [Tue, 6 Nov 2018 08:38:34 +0000 (09:38 +0100)]
- updated the CAT CW stuff, now maintaining a ring buffer.
Several corrections such that hitting a "local" CW key aborts
CAT CW correctly and continues with manual CW.
- moved choice of active level for CW buttons to GPIO menu
- added support for "CW letter spacing" in CW menu
- at many places corrected cast from a generic pointer to int. The correct
way of passing an int when receiving a pointer is:
void extern(void *p) {
int state=(uintptr_t) p;
call intern(state);
}
What is certainly wrong is "call intern((uintptr_t) p)" since this
passes on many systems a 64-bit value where a 32-bit value is expected
unless strong prototyping is used.
- cleaned up configure.c (at least: "gtk_widget_destroy(dialog)" in configure_gpio
had to be deleted)
- use cw_breakin instead of vox_hang in CW-keyer
- improved GPIO side tone creation, no longer use softTone but do this within the keyer thread
(but only if GPIO side tone is requested).
- Hide button names such as CWL_BUTTON from iambic.c
- Major overhaul of iambic keyer: see extensive comment in iambic.c
Jim Ancona [Sun, 23 Sep 2018 15:15:44 +0000 (11:15 -0400)]
Add support for Pi SDR. Note that in order to get it to run, I had to disable PURESIGNAL, STEMLAB_DISCOVERY and STEMLAB_FIX, which are enabled by default in Makefile.mac.
c vw [Tue, 24 Jul 2018 08:00:12 +0000 (10:00 +0200)]
Changes that affect the user interface
======================================
A. applies only if compiled with the STEMLAB_DISCOVERY_OPTION:
unified non-AVAHI and AVAHI versions of stemlab_discovery.
STEMLAB_DISCOVERY_MAC in Makefile.mac no longer needed.
Made the user interface more specific in the NO_AVAHI version.
Currently, the no-avahi version is only activated in Makefile.mac
but could be useful for *nix systems where you do not have avahi.
B. applies only if compiled with SPLIT_RXTX defined:
if using two receivers, the TX panel only "hides" the first one.
it is possible to listen to RX2 while TXing on the RX1 frequency,
very useful for measurements where you feed-back the attenuated
TX signal to the RX2 antenna jack. This implies that if RX2
is active while TXing, you see the S-meter of RX2 and not the
TX levels (to be able to measure the signal strength).
RXing with RX2 and TXing with TX on the RX1 frequency implies
that you go "split".
This is so useful that it is the default option for me now.
C. applies only if compiled with DIGI_MODES defined:
when switching to DIGU or DIGL, you automatically get wide
RX filters (0 ... 3000 Hz) and all noise reduction OFF
without changing the current settings. So when returning to
another mode (say, CWL or USB), the filter/NR settings
effective before switching to digi are restored.
This is so useful that it is the default option for me now.
D. if *not* compiled with the PURESIGNAL_OPTION, you still get
a two-tone generator for testing purposed. The new button
TT is located exactly where the PS button is placed if
compiled with PURESIGNAL.
E. Small change to make B. meaningful:
- Upon TX, mute only that RX that is bound to the TX frequency
- Upon TX, mute the audio of the active_receiver only in CW mode
(to give room for the side tone).
Error corrections (only the most important ones)
================================================
- avoid core dump if using PURESIGNAL with METIS devices:
create receiver[PS_T(R)X_FEEDBACK] in any case
- colouring the correct TX frequency in red in the VFO panel
- use correct attenuation scale for filter_board == CHARLY25
- initialize tx->attenuation
- explicitly specifying levels and frequencies for two-tone
(the defaults were silently used).
- changed an #ifdef FREEDV to #ifdef PURESIGNAL in vfo_menu.c (obvious error)
projects / pihpsdr.git / log