Re: [Discuss-gnuradio] ANCI-C vs Gnuradio/C++ speeeed
I run the following simple test, compiled with
g++ test.cc -o test
and I got the following results.
I see a 4-fold speed reduction using STL.
What am I doing wrong?
Achilleas
---
$ time ./test 1 1
real0m0.121s
user0m0.120s
sys 0m0.001s
while
$ time ./test 1 2
real0m0.462s
user0m0.459s
sys 0m0.003s
and
$ time ./test 10 1
real0m1.185s
user0m1.184s
sys 0m0.001s
while
$ time ./test 10 2
real0m4.597s
user0m4.595s
sys 0m0.002s
--
test.cc=
#includevector
int main(int ac,char **av) {
const int M = 1000;
std::vectorint y(M);
int *x;
x=(int*)malloc(M*sizeof(int));
int N=atoi(av[1]);
if(atoi(av[2])==1) {
for (int i=0;iN;i++)
for(int j=0;jM;j++)
x[j]=i/(j+1);
} else if(atoi(av[2])==2) {
for (int i=0;iN;i++)
for(int j=0;jM;j++)
y[j]=i/(j+1);
}
}
==
___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
Re: [Discuss-gnuradio] ANCI-C vs Gnuradio/C++ speeeed
Yes, with -O the STL code is only 1.7 times
slower.
Even better, with -O2 the two cases execute
in exactly the same time.
So, is Gnuradio compiling with -O2
option?
Achilleas
Philip Balister wrote:
Try g++ -O test.cc -o test
That solved the problem for me.
Philip
Achilleas Anastasopoulos wrote:
I run the following simple test, compiled with
g++ test.cc -o test
and I got the following results.
I see a 4-fold speed reduction using STL.
What am I doing wrong?
Achilleas
---
$ time ./test 1 1
real0m0.121s
user0m0.120s
sys 0m0.001s
while
$ time ./test 1 2
real0m0.462s
user0m0.459s
sys 0m0.003s
and
$ time ./test 10 1
real0m1.185s
user0m1.184s
sys 0m0.001s
while
$ time ./test 10 2
real0m4.597s
user0m4.595s
sys 0m0.002s
--
test.cc=
#includevector
int main(int ac,char **av) {
const int M = 1000;
std::vectorint y(M);
int *x;
x=(int*)malloc(M*sizeof(int));
int N=atoi(av[1]);
if(atoi(av[2])==1) {
for (int i=0;iN;i++)
for(int j=0;jM;j++)
x[j]=i/(j+1);
} else if(atoi(av[2])==2) {
for (int i=0;iN;i++)
for(int j=0;jM;j++)
y[j]=i/(j+1);
}
}
==
___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
Re: [Discuss-gnuradio] 48KHz audio board
On Sat, May 13, 2006 at 11:49:03PM +0200, Vincenzo Pellegrini wrote: sorry, i know that the question might be silly but: my audio board requires 48Ksps the fm receiver in gnu radio examples delivers to it 32Ksps instead. setting a overall decimation factor of 1320 i can get as near as 48484 samples/sec. but even if much better, this still gives problems. and there's no integer factor that can take 64Msps to exactly 48Ksps. can anyone help a newbie? Thanks Vincenzo Pellegrini Italy If you're using ALSA, try plughw:0,0 or use gr-audio-oss instead. Eric ___ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
Re: [Discuss-gnuradio] 48KHz audio board
On Sun, May 14, 2006 at 08:37:11AM +0930, Daniel O'Connor wrote: On Sunday 14 May 2006 07:19, Vincenzo Pellegrini wrote: my audio board requires 48Ksps the fm receiver in gnu radio examples delivers to it 32Ksps instead. setting a overall decimation factor of 1320 i can get as near as 48484 samples/sec. but even if much better, this still gives problems. and there's no integer factor that can take 64Msps to exactly 48Ksps. can anyone help a newbie? I think you need a resampling function (or a smarter sound card ;) eg sox can resample from one frequency to another, I don't know if there is such a block in GNURadio though. gr.rational_resampler. See gnuradio-examples/audio/test_resampler.py Eric ___ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
Re: [Discuss-gnuradio] ANCI-C vs Gnuradio/C++ speeeed
On Sun, May 14, 2006 at 09:40:16AM -0400, Achilleas Anastasopoulos wrote: I run the following simple test, compiled with g++ test.cc -o test and I got the following results. I see a 4-fold speed reduction using STL. What am I doing wrong? Achilleas You're not using the standard level of optimization -O2 E.g., [EMAIL PROTECTED] tmp]$ g++ -O2 test.cc -o test [EMAIL PROTECTED] tmp]$ time ./test 10 1 real0m1.797s user0m1.792s sys 0m0.004s [EMAIL PROTECTED] tmp]$ time ./test 10 2 real0m1.797s user0m1.796s sys 0m0.000s ___ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
Re: [Discuss-gnuradio] ANCI-C vs Gnuradio/C++ speeeed
On Sun, May 14, 2006 at 11:13:47AM -0400, Achilleas Anastasopoulos wrote: Yes, with -O the STL code is only 1.7 times slower. Even better, with -O2 the two cases execute in exactly the same time. So, is Gnuradio compiling with -O2 option? Yes. Eric ___ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
[Discuss-gnuradio] Re: Graphical programming with Ptolemy [rather: Deep dependencies]
John Clark wrote: It seems to becoming an 'annual' event that I setup a GNURadio environment, and this year's candidate machine seems to have almost nothing of the antecedent packeges... This is not exactly the content you would expect from the subject. I know i'm a little late, but shouldn't the subject be changed? Moreover the references are kept with replying, so diligent clients thread the messages misleading... Patrick -- Engineers motto: cheap, good, fast: choose any two Patrick Strasser patrick dot strasser at tugraz dot at Student of Telematik, Techn. University Graz, Austria ___ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
[Discuss-gnuradio] 'module' object has no attribute
I am attempting to build modified versions of gr_file_source, and
gr_file_sink outside of gnuradio-core. I have successfully built, and
used my version of file_source (called rad_wvfm_source). However, I
cannot get my version of file_sink (called rad_wvfm_sink) to build into
a module so that it is callable from python. My python QA script fails
with an AttributeError: 'module' object has no attribute 'wvfm_sink'.
I've attached my swig files. Any help is appreciated.
Thanks,
- Lee
P.S.
You can browse all the code at http://svn.pattoncentral.org, just click
gr_radar. If it is easier, you can check out the code with
$] svn co http://svn.pattoncentral.org/gr_radar --username=anon
and empty password.
/* -*- c++ -*- */
/*
* Copyright 2004 Free Software Foundation, Inc.
*
* This file is part of GNU Radio
*
* GNU Radio is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2, or (at your option)
* any later version.
*
* GNU Radio is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with GNU Radio; see the file COPYING. If not, write to
* the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
* Boston, MA 02111-1307, USA.
*/
%feature(autodoc, 1); // generate python docstrings
%include exception.i
%import gnuradio.i// the common stuff
%{
#include gnuradio_swig_bug_workaround.h // mandatory bug fix
#include stdexcept
%}
%{
#include rad_wvfm_sink.h
#include rad_wvfm_source.h
%}
%include rad_wvfm_sink.i
%include rad_wvfm_source.i
GR_SWIG_BLOCK_MAGIC(rad,wvfm_sink)
rad_wvfm_sink_sptr
gr_make_wvfm_sink (size_t itemsize, const char *filename);
class rad_wvfm_sink : public gr_sync_block
{
protected:
rad_wvfm_sink (size_t itemsize, const char *filename);
public:
~rad_wvfm_sink ();
/*!
* \brief open filename and begin output to it.
*/
bool open(const char *filename);
/*!
* \brief close current output file.
*/
void close();
};
%constant int SEEK_SET = 0; /* Seek from beginning of file. */
%constant int SEEK_CUR = 1; /* Seek from current position. */
%constant int SEEK_END = 2; /* Seek from end of file. */
GR_SWIG_BLOCK_MAGIC(rad,wvfm_source)
rad_wvfm_source_sptr
rad_make_wvfm_source (size_t itemsize, const char *filename, bool repeat=false);
class rad_wvfm_source : public gr_sync_block
{
protected:
rad_wvfm_source (size_t itemsize, const char *filename, bool repeat);
public:
~rad_wvfm_source ();
bool seek (long seek_point, int whence);
};
___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
Re: [Discuss-gnuradio] 'module' object has no attribute
Lee Patton wrote: I am attempting to build modified versions of gr_file_source, and gr_file_sink outside of gnuradio-core. I have successfully built, and used my version of file_source (called rad_wvfm_source). However, I cannot get my version of file_sink (called rad_wvfm_sink) to build into a module so that it is callable from python. My python QA script fails with an AttributeError: 'module' object has no attribute 'wvfm_sink'. If you use: GR_SWIG_BLOCK_MAGIC(rad,wvfm_sink) Then should use: rad_make_wvfm_sink in stead of: gr_make_wvfm_sink I hope this helps, Martin I've attached my swig files. Any help is appreciated. Thanks, - Lee P.S. You can browse all the code at http://svn.pattoncentral.org, just click gr_radar. If it is easier, you can check out the code with $] svn co http://svn.pattoncentral.org/gr_radar --username=anon and empty password. ___ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio ___ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
Re: [Discuss-gnuradio] depad for atsc 2
Charles Swiger wrote: I just added atsc_depad to cvs gr-atsc. The 2.x code worked thru derandomizer, but the output was atsc_mpeg_packet padded out to 256, which xine/mplayer won't play. Kludging atsc_mpeg_packet pad to 0 in atsc_types.h worked with a warning from gnuradio about output not being a power-of-2. atsc_depad reads padded atsc_mpeg_packets and writes 188 byte char packets. That function might be better built into derandomizer anyway. Also put the 0.9 boundary checking in the atsc_viterbi_decoder. That was causing problems getting past bumps in the input signal like usrp overrun dropped packets. Also added the missing gr_agc.h for atsc_fpll. gr_agc.h is still in gnuradio-2.x but it was renamed to gri_agc.h 4 months ago. The i was added to make it consistent with the rest of the gnuradio framework, to distinguish flowgraph blocks from helper function objects (gri_agc is a helper object). At the same time a compex agc was added as gri_agc_cc.h These are now also used in the gr_agc_cc and gr_agc_ff blocks. Greetings, Martin --Chuck ___ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio ___ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
[Discuss-gnuradio] running
Hello!I've just started playing with GnuRadio and the USRP, and was hoping to run it on Windows XP. I found Martin's installer "setup-gnuradio-complete-2.6cvs.exe" and it worked great. I then triedto run the example to listen to some radio (I had this running with a fedora core 5 linux installation with the same usrp). I received the following error messages:C:\caruthd\usrp\gnuradio-examples-0.7\python\usrppython usrp_wfm_rcv.py -f 101.1 found 5 bussesUsing RX d'board A: TV Rx gr_fir_ccf: using SSE gr_fir_fff: using SSETraceback (most recent call last): File "usrp_wfm_rcv.py", line 268, in ? app = stdgui.stdapp (wfm_rx_graph, "USRP WFM RX") File "C:\Python24\Lib\site-packages\gnuradio\wxgui\stdgui.py", line 36, in __init__ wx.App.__init__ (self, redirect=False) File "C:\Python24\Lib\site-packages\wx-2.6-msw-ansi\wx\_core.py", line 7700,in __init__ self._BootstrapApp() File "C:\Python24\Lib\site-packages\wx-2.6-msw-ansi\wx\_core.py", line 7352,in _BootstrapApp return _core_.PyApp__BootstrapApp(*args, **kwargs) File "C:\Python24\Lib\site-packages\gnuradio\wxgui\stdgui.py", line 39, in Oninit frame = stdframe (self.flow_graph_maker, self.title, self._nstatus) File "C:\Python24\Lib\site-packages\gnuradio\wxgui\stdgui.py", line 60, in __init__ self.panel = stdpanel (self, self, flow_graph_maker) File "C:\Python24\Lib\site-packages\gnuradio\wxgui\stdgui.py", line 81, in __init__ self.fg = flow_graph_maker (frame, self, vbox, sys.argv) File "usrp_wfm_rcv.py", line 93, in __init__ False) # ok_to_block File "C:\Python24\Lib\site-packages\gnuradio\audio_windows.py", line 213, insink return _audio_windows.sink(*args)NotImplementedError: No matching function for overloaded 'sink' I had seen a previous post where Eric replied that a hex string needed to be replaced by a gr function with the hex string as an argument, but from looking through the .py files called here, i did not see anything obvious. has anyone else had this problem? thanks in advance for any help! oh, and by the way, "usrp_fft.py" ran without a problem.Thanks a lot! Dave Get amazing travel prices for air and hotel in one click on Yahoo! FareChase ___ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
Re: [Discuss-gnuradio] 'module' object has no attribute
Thanks Martin! I was staring at that code for a half hour, and didn't see it. I even grepped 'gr_' to see if it was something like that, and I still didn't see it Thanks for your time. On Mon, 2006-05-15 at 00:55 +0200, Martin Dvh wrote: Lee Patton wrote: I am attempting to build modified versions of gr_file_source, and gr_file_sink outside of gnuradio-core. I have successfully built, and used my version of file_source (called rad_wvfm_source). However, I cannot get my version of file_sink (called rad_wvfm_sink) to build into a module so that it is callable from python. My python QA script fails with an AttributeError: 'module' object has no attribute 'wvfm_sink'. If you use: GR_SWIG_BLOCK_MAGIC(rad,wvfm_sink) Then should use: rad_make_wvfm_sink in stead of: gr_make_wvfm_sink I hope this helps, Martin I've attached my swig files. Any help is appreciated. Thanks, - Lee P.S. You can browse all the code at http://svn.pattoncentral.org, just click gr_radar. If it is easier, you can check out the code with $] svn co http://svn.pattoncentral.org/gr_radar --username=anon and empty password. ___ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio ___ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
Re: [Discuss-gnuradio] Modem and/or ADSL Daughter card
On Wed, May 10, 2006 at 07:24:07PM -0700, Eric Blossom wrote: On Thu, May 04, 2006 at 02:01:31PM -0400, Michael Milner wrote: Hello, I was just wondering if anyone has used GNURadio for any non-radio applications, specifically phone line modulation/demodulation. It shouldn't be too difficult to design a daughter board for the USRP to sample phone line voltage (with an appropriate line interface circuit of course). After that it should be rather easy in GNURadio to generate DTMF tones for dialing, and modem tones for data communication. Any thoughts? Mike A simple DAA (Data Access Arrangement) plus the LF daughterboard would probably work. There are two chip DAA's available. Been a while since I looked, but I think that Crystal Semiconductor used to make them. Google solid state daa There's also the classic Midcom transformer plus a couple of other discretes. These things are tricky to get right. You definitely want to find somebody's reference design and use that. Or use the pre-approved chipset. The USRP is pretty much thermonuclear overkill for this application. You could just use a sound card with a DAA. That's how all the soft-modems work. IIRC some of the the AC97 codecs support a second channel for telecom (modem) apps. Here is an example that uses a sound card, http://www.araneus.fi/audsl/. idle-musing It would be neat if there were the PC audio-out equivalents of the iGo iTips power adapter plugs, with plugs that adapt audio-out to media such as AC powerline, phone wire, Cat5, and television cable. A suitable software modem would adapt to whatever medium was on hand, adding a new dimension to ad hoc networking. /idle-musing Dave -- David Young OJC Technologies [EMAIL PROTECTED] Urbana, IL * (217) 278-3933 ___ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
Re: [Discuss-gnuradio] Problem with making sound card as source
Thanks Breett:) I've solved this problem by turn on capture item in the mixer. And I guess that's why there are many people say that there linux cannot recording sound. Hanwen2006/5/10, Brett L Trotter [EMAIL PROTECTED]: I had a similar problem- turned out not to be the hardware.Run alsamixer and make sure to turn on capture on your line input (withspacebar) and turn up the volume all the way, then turn on capture oncapture and turn the volume up all the way again. You should then be getting audio and adjust the volume down to reduce clipping.2 1 wrote: Hi, Matt :) I do set the line-in as the record source in the mixer. But I found that the audio recorder app of my FC4 can't record either. I'm doubt that the sound drive of FC4 cannot support the card well. Besides, I tried on 3 different machines and all of them failed. They all have a intel integrated sound card. I'm using gr-audio-alsa.0.3.tar.gz. Have you ever met this problem before? And do you know how to make the sound card well supported under linux? Thanks a lot. Hanwen 2006/5/10, Matt Ettus [EMAIL PROTECTED]: Can you record sound in other applications?Have you set the audio mixer properly? Matt On Tue, 2006-05-09 at 20:34 +0800, 2 1 wrote: Hi, all My USRP is on delievering, so I try to do someting with the sound card. Unfortunately, the integrated AC97 sound card of my pc seem to not work properly as source. I connect a mp3 player with the line-in port of sound card with an audio wire. In a script, the sound card is set to source and connected to a file sink, so the sampled raw audio is written to a file. In another script, the raw audio file is set to the source and replayed by the sound card as sink. But, I hear nothing from my sound card. The dialtone.py can work making the sound properly, so the sound card as sink is ok. Hanwen ___ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio -- Matt Ettus [EMAIL PROTECTED] ___ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio ___Discuss-gnuradio mailing listDiscuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio ___ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
