From df8oe at gmx.de Thu Jun 6 06:41:04 2013 From: df8oe at gmx.de (df8oe at gmx.de) Date: Thu, 6 Jun 2013 08:41:04 +0200 Subject: rtl-tcp-server running on avm-fritzbox Message-ID: <201306060841.04785.df8oe@gmx.de> I want to use a rtl-dvb-stick connected to a fritz-box-dsl-router via tcp. Software running on the fritzbox is freetz (www.freetz.org) and cpu-architecture is mips. So I created a makefile and compiled the software without errors. But when I start the software (usb-dongle connected) it throws a usb_claim_interface error -5. rtl_test recognizes the type of stick correctly but throws identical error message. So I started rtl_test with strace - here you can find the output: http://www.df8oe.de/output.txt Any tipps for solving the problem are welcome ;) Best regards df8oe From Robert.Kiendl at automatis.com Thu Jun 6 07:52:09 2013 From: Robert.Kiendl at automatis.com (Robert Kiendl) Date: Thu, 6 Jun 2013 09:52:09 +0200 Subject: AW: rtl-tcp-server running on avm-fritzbox In-Reply-To: <201306060841.04785.df8oe@gmx.de> References: <201306060841.04785.df8oe@gmx.de> Message-ID: Hi, I guess there is a permission problem accessing the device. Under normal conditions udev is doing the permission stuff when plugging in the device. I have had once the similar issue for own compiled binary mipsbe for mikrotik and tplink devices. Because of the lack of udev is simply altered the permission after plugging the dongle (something like /dev/dvb/adapter[something] perms 755) This made it work afterwards. Hope this info helps, Ciao Robert, 73 de oe6rke -----Urspr?ngliche Nachricht----- Von: osmocom-sdr-bounces at lists.osmocom.org [mailto:osmocom-sdr-bounces at lists.osmocom.org] Im Auftrag von df8oe at gmx.de Gesendet: Donnerstag, 06. Juni 2013 08:41 An: osmocom-sdr at lists.osmocom.org Betreff: rtl-tcp-server running on avm-fritzbox I want to use a rtl-dvb-stick connected to a fritz-box-dsl-router via tcp. Software running on the fritzbox is freetz (www.freetz.org) and cpu-architecture is mips. So I created a makefile and compiled the software without errors. But when I start the software (usb-dongle connected) it throws a usb_claim_interface error -5. rtl_test recognizes the type of stick correctly but throws identical error message. So I started rtl_test with strace - here you can find the output: http://www.df8oe.de/output.txt Any tipps for solving the problem are welcome ;) Best regards df8oe From df8oe at gmx.de Thu Jun 6 08:19:38 2013 From: df8oe at gmx.de (df8oe at gmx.de) Date: Thu, 6 Jun 2013 10:19:38 +0200 Subject: AW: rtl-tcp-server running on avm-fritzbox In-Reply-To: References: <201306060841.04785.df8oe@gmx.de> Message-ID: <201306061019.38993.df8oe@gmx.de> Am Donnerstag, 6. Juni 2013, 09:52:09 schrieb Robert Kiendl: > Hi, > > I guess there is a permission problem accessing the device. Under normal > conditions udev is doing the permission stuff when plugging in the device. > I have had once the similar issue for own compiled binary mipsbe for > mikrotik and tplink devices. Because of the lack of udev is simply altered > the permission after plugging the dongle (something like > /dev/dvb/adapter[something] perms 755) This made it work afterwards. > > Hope this info helps, > > Ciao Robert, 73 de oe6rke I have figured out the device is unter /dev/usb/001/004 and I changed the permissions from 744 to 755 but this does not change anything. rtl_test is running with root permissions.... There are only the folders block, bus, char, dsl_vr9, input, net and pts under /dev... Maybe it is a permission problem - but I have no idea where to look. Best regards Andreas, df8oe From jeff at kb9ysj.com Thu Jun 6 10:41:56 2013 From: jeff at kb9ysj.com (Jeff Allen) Date: Thu, 6 Jun 2013 05:41:56 -0500 Subject: No subject Message-ID: <007401ce62a2$7791f040$66b5d0c0$@kb9ysj.com> Please remove me from your email list, thanks -------------- next part -------------- An HTML attachment was scrubbed... URL: From fritz.meier at myway.de Thu Jun 6 13:19:57 2013 From: fritz.meier at myway.de (Fritz Meier) Date: Thu, 06 Jun 2013 15:19:57 +0200 Subject: Need source for rtlsdr.dll !! Message-ID: <51B08C7D.6070100@myway.de> Hello, i would like to work in ARM-Cpu environment with the RTL2832 dongle therefor i need the Arm-WinCE .dll or Arm-Linux .o . Where i can download the C, or C++ or Pascal source ? Thanks for your help Fritz From f.gugliuzza at hacklabproject.org Thu Jun 6 13:53:44 2013 From: f.gugliuzza at hacklabproject.org (Francesco Gugliuzza) Date: Thu, 6 Jun 2013 15:53:44 +0200 Subject: Need source for rtlsdr.dll !! In-Reply-To: <51B08C7D.6070100@myway.de> References: <51B08C7D.6070100@myway.de> Message-ID: Library and tools' sources are here: http://cgit.osmocom.org/rtl-sdr/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From vjt at openssl.it Thu Jun 6 13:55:45 2013 From: vjt at openssl.it (Marcello Barnaba) Date: Thu, 6 Jun 2013 15:55:45 +0200 Subject: Need source for rtlsdr.dll !! In-Reply-To: <51B08C7D.6070100@myway.de> References: <51B08C7D.6070100@myway.de> Message-ID: <73D72B52-E729-4AFA-9552-C53513395A45@openssl.it> On Jun 06, 2013, at 3:19 PM, Fritz Meier wrote: > Hello, > i would like to work in ARM-Cpu environment with the > RTL2832 dongle therefor i need the Arm-WinCE .dll or Arm-Linux .o . > Where i can download the C, or C++ or Pascal source ? Hi Fritz, the source of osmocom projects is on their Git repository: http://git.osmocom.org/ And this is the official project page: http://sdr.osmocom.org/trac/wiki/rtl-sdr Have fun, ~Marcello From 246tnt at gmail.com Thu Jun 6 14:10:38 2013 From: 246tnt at gmail.com (Sylvain Munaut) Date: Thu, 6 Jun 2013 16:10:38 +0200 Subject: Need source for rtlsdr.dll !! In-Reply-To: <51B08C7D.6070100@myway.de> References: <51B08C7D.6070100@myway.de> Message-ID: > i would like to work in ARM-Cpu environment with the > RTL2832 dongle therefor i need the Arm-WinCE .dll or Arm-Linux .o . > Where i can download the C, or C++ or Pascal source ? And you might also want to yell at whoever distributed you the .dll you're using because they're supposed to include an offer for the source code with the package they distribute, as per GPL requirements. And also be sure yourself to include such an offer for your sources (the original ones if unmodified, or your modified source version if you have to do some porting) in any package you distribute. Cheers, Sylvain From alancorey at yahoo.com Thu Jun 6 14:21:25 2013 From: alancorey at yahoo.com (Alan Corey) Date: Thu, 6 Jun 2013 07:21:25 -0700 (PDT) Subject: AW: rtl-tcp-server running on avm-fritzbox In-Reply-To: Message-ID: <1370528485.56354.YahooMailClassic@web161506.mail.bf1.yahoo.com> I don't know anything about your OS, but look at the 3rd line down in your posted listing: open("/usr/lib/freetz/librtlsdr.so.0", O_RDONLY) = -1 ENOENT (No such file or directory) It isn't finding the library.? That same message repeats many times throughout your listing.? It's not finding many libraries from looking at it again.? If you're using cmake you can probably fix the paths to libraries in your CMakeCache.txt in the build directory.? If not you might need to set an LDFLAGS in your environment or something. If the process doesn't have execute permissions on all the directories in the paths it won't find anything either because it can't search there.? Does it work if you run as root? ? Alan ----- Radio Astronomy - the ultimate DX --- On Thu, 6/6/13, Robert Kiendl wrote: From: Robert Kiendl Subject: AW: rtl-tcp-server running on avm-fritzbox To: osmocom-sdr at lists.osmocom.org Date: Thursday, June 6, 2013, 3:52 AM Hi, I guess there is a permission problem accessing the device. Under normal conditions udev is doing the permission stuff when plugging in the device. I have had once the similar issue for own compiled binary mipsbe for mikrotik and tplink devices. Because of the lack of udev is simply altered the permission after plugging the dongle (something like /dev/dvb/adapter[something]? perms 755) This made it work afterwards. Hope this info helps, Ciao Robert, 73 de oe6rke -----Urspr?ngliche Nachricht----- Von: osmocom-sdr-bounces at lists.osmocom.org [mailto:osmocom-sdr-bounces at lists.osmocom.org] Im Auftrag von df8oe at gmx.de Gesendet: Donnerstag, 06. Juni 2013 08:41 An: osmocom-sdr at lists.osmocom.org Betreff: rtl-tcp-server running on avm-fritzbox I want to use a rtl-dvb-stick connected to a fritz-box-dsl-router via tcp. Software running on the fritzbox is freetz (www.freetz.org) and cpu-architecture is mips. So I created a makefile and compiled the software without errors. But when I start the software (usb-dongle connected) it throws a usb_claim_interface error -5. rtl_test recognizes the type of stick correctly but throws identical error message. So I started rtl_test with strace - here you can find the output: http://www.df8oe.de/output.txt Any tipps for solving the problem are welcome ;) Best regards df8oe -------------- next part -------------- An HTML attachment was scrubbed... URL: From df8oe at gmx.de Thu Jun 6 14:51:20 2013 From: df8oe at gmx.de (df8oe at gmx.de) Date: Thu, 6 Jun 2013 16:51:20 +0200 Subject: AW: rtl-tcp-server running on avm-fritzbox In-Reply-To: <1370528485.56354.YahooMailClassic@web161506.mail.bf1.yahoo.com> References: <1370528485.56354.YahooMailClassic@web161506.mail.bf1.yahoo.com> Message-ID: <201306061651.20605.df8oe@gmx.de> Am Donnerstag, 6. Juni 2013, 16:21:25 schrieb Alan Corey: > I don't know anything about your OS, but look at the 3rd line down in your > posted listing: > > > > open("/usr/lib/freetz/librtlsdr.so.0", O_RDONLY) = -1 ENOENT (No such file > or directory) > > > > It isn't finding the library. That same message repeats many times > throughout your listing. It's not finding many libraries from looking at > it again. If you're using cmake you can probably fix the paths to > libraries in your CMakeCache.txt in the build directory. If not you might > need to set an LDFLAGS in your environment or something. > > If the process doesn't have execute permissions on all the directories in > the paths it won't find anything either because it can't search there. > Does it work if you run as root? Hi Alan, I was not using cmake because there is no cmake posible on that embedded machine. I wrote a makefile by myself and gave linkeroption for the paths of the libraries. As you see the binary tries search at different places and fails - until it finds successfully at /usr/lib/... This repeats for all needed libraries. It is nicky - but I don't think it is critical. May be that can be figured out - but that is my first makefile for this machine. Luckily it finds all libraries at the end. The binary is executed as root. For Robert tells me something about missing permissions I took a look at some permissions in /dev/usb and saw some of them are --r--r--r - no w! So I changed this permissions to -rw-rw-rw and the errors disappeared for these files. But you see there are much errors "Inappropriate ioctl for device" and many of them are still present and I can't see why. df8oe, Andreas From peter at stuge.se Thu Jun 6 19:18:25 2013 From: peter at stuge.se (Peter Stuge) Date: Thu, 6 Jun 2013 21:18:25 +0200 Subject: rtl-tcp-server running on avm-fritzbox In-Reply-To: <201306060841.04785.df8oe@gmx.de> References: <201306060841.04785.df8oe@gmx.de> Message-ID: <20130606191825.8748.qmail@stuge.se> df8oe at gmx.de wrote: > when I start the software (usb-dongle connected) it throws a > usb_claim_interface error -5 $ grep -- -5 src/libusb/libusb/libusb.h LIBUSB_ERROR_NOT_FOUND = -5, From df8oe at gmx.de Fri Jun 7 07:07:33 2013 From: df8oe at gmx.de (df8oe at gmx.de) Date: Fri, 7 Jun 2013 09:07:33 +0200 Subject: rtl-tcp-server running on avm-fritzbox In-Reply-To: <20130606191825.8748.qmail@stuge.se> References: <201306060841.04785.df8oe@gmx.de> <20130606191825.8748.qmail@stuge.se> Message-ID: <201306070907.33449.df8oe@gmx.de> Am Donnerstag, 6. Juni 2013, 21:18:25 schrieb Peter Stuge: > df8oe at gmx.de wrote: > > when I start the software (usb-dongle connected) it throws a > > usb_claim_interface error -5 > > $ grep -- -5 src/libusb/libusb/libusb.h > LIBUSB_ERROR_NOT_FOUND = -5, > > From the log: (thanks) > > ioctl(5, USBDEVFS_CLAIMINTERFACE, 0x7fd063c4) = -1 ENOENT (No such file or > directory) > > The kernel is saying that the interface doesn't exist. > > lsusb -v for this device run on the fritzbox would be interesting. If the interface does not exist: Where does the kernel get the following information (have a look at the log): "ezcap USB 2.0 DVB-T/DAB/FM dongl"..., 33ezcap USB 2.0 DVB-T/DAB/FM dongle) I think the interface cannot be opened or the binary thinks it cannot be opened (but it can be opened). When I plug in the stick correspondent devices are created under /dev/usb/ Andreas From df8oe at gmx.de Fri Jun 7 06:40:13 2013 From: df8oe at gmx.de (df8oe at gmx.de) Date: Fri, 7 Jun 2013 08:40:13 +0200 Subject: rtl-tcp-server running on avm-fritzbox In-Reply-To: <20130606191825.8748.qmail@stuge.se> References: <201306060841.04785.df8oe@gmx.de> <20130606191825.8748.qmail@stuge.se> Message-ID: <201306070840.13991.df8oe@gmx.de> Am Donnerstag, 6. Juni 2013, 21:18:25 schrieb Peter Stuge: > df8oe at gmx.de wrote: > > when I start the software (usb-dongle connected) it throws a > > usb_claim_interface error -5 > > $ grep -- -5 src/libusb/libusb/libusb.h > LIBUSB_ERROR_NOT_FOUND = -5, > > From the log: (thanks) > > ioctl(5, USBDEVFS_CLAIMINTERFACE, 0x7fd063c4) = -1 ENOENT (No such file or > directory) > > The kernel is saying that the interface doesn't exist. > > lsusb -v for this device run on the fritzbox would be interesting. > Hi Peter, here the output of lsusb -v: http://www.df8oe.de/lsusb.txt df8oe, Andreas From peter at stuge.se Fri Jun 7 10:37:34 2013 From: peter at stuge.se (Peter Stuge) Date: Fri, 7 Jun 2013 12:37:34 +0200 Subject: rtl-tcp-server running on avm-fritzbox In-Reply-To: <201306070840.13991.df8oe@gmx.de> <201306070907.33449.df8oe@gmx.de> References: <201306060841.04785.df8oe@gmx.de> <20130606191825.8748.qmail@stuge.se> <201306070840.13991.df8oe@gmx.de> <201306060841.04785.df8oe@gmx.de> <20130606191825.8748.qmail@stuge.se> <201306070907.33449.df8oe@gmx.de> Message-ID: <20130607103734.29622.qmail@stuge.se> df8oe at gmx.de wrote: > > > usb_claim_interface error -5 > > > > $ grep -- -5 src/libusb/libusb/libusb.h > > LIBUSB_ERROR_NOT_FOUND = -5, .. > > The kernel is saying that the interface doesn't exist. > > If the interface does not exist: > > Where does the kernel get the following information Interface has a very specific meaning in USB. Please look into chapters 5 8 and 9 of the USB 2.0 spec available at usb.org to learn more. > I think the interface cannot be opened The interface can not be claimed because the kernel doesn't think that it exists. The error is actually really clear. > When I plug in the stick correspondent devices are created under /dev/usb/ USB is a packet bus with highly structured communication. Device nodes in /dev are not a good reference. df8oe at gmx.de wrote: > here the output of lsusb -v: http://www.df8oe.de/lsusb.txt Unfortunately the fritzbox has a useless mini lsusb program as opposed to the actual lsusb utility from the usbutils package. It would be great if you could build the real one. Somewhat less effort and still interesting would be to mount debugfs and then cat /sys/kernel/debug/usb/devices Other than that, it would be helpful if you could build a libusb binary with debug logging enabled. See http://libusb.org/wiki/debug for copypaste instructions that would apply to a host build. You must modify them as appropriate for your cross build of course. //Peter From df8oe at gmx.de Fri Jun 7 12:37:16 2013 From: df8oe at gmx.de (df8oe at gmx.de) Date: Fri, 7 Jun 2013 14:37:16 +0200 Subject: rtl-tcp-server running on avm-fritzbox In-Reply-To: <20130607103734.29622.qmail@stuge.se> References: <201306060841.04785.df8oe@gmx.de> <201306070907.33449.df8oe@gmx.de> <20130607103734.29622.qmail@stuge.se> Message-ID: <201306071437.16549.df8oe@gmx.de> Am Freitag, 7. Juni 2013, 12:37:34 schrieb Peter Stuge: > > I think the interface cannot be opened > > The interface can not be claimed because the kernel doesn't think > that it exists. The error is actually really clear. > > > When I plug in the stick correspondent devices are created under > > /dev/usb/ > > USB is a packet bus with highly structured communication. Device > nodes in /dev are not a good reference. > > df8oe at gmx.de wrote: > > here the output of lsusb -v: http://www.df8oe.de/lsusb.txt > > Unfortunately the fritzbox has a useless mini lsusb program as > opposed to the actual lsusb utility from the usbutils package. > It would be great if you could build the real one. > > Somewhat less effort and still interesting would be to mount debugfs > and then cat /sys/kernel/debug/usb/devices > > Other than that, it would be helpful if you could build a libusb > binary with debug logging enabled. See http://libusb.org/wiki/debug > for copypaste instructions that would apply to a host build. You must > modify them as appropriate for your cross build of course. > For debugging is not enabled by default in this kernel and building a new one and/or complete lsusb, I looked for another way to get the needed information. I noticed that the debugging package "listusb" is available for the box and I build it. Here the output of listusb -v: http://www.df8oe.de/listusb.txt Can you see there what is going wrong? df8oe, Andreas From peter at stuge.se Fri Jun 7 14:44:25 2013 From: peter at stuge.se (Peter Stuge) Date: Fri, 7 Jun 2013 16:44:25 +0200 Subject: rtl-tcp-server running on avm-fritzbox In-Reply-To: <201306071437.16549.df8oe@gmx.de> References: <201306060841.04785.df8oe@gmx.de> <201306070907.33449.df8oe@gmx.de> <20130607103734.29622.qmail@stuge.se> <201306071437.16549.df8oe@gmx.de> Message-ID: <20130607144425.19612.qmail@stuge.se> df8oe at gmx.de wrote: > For debugging is not enabled by default in this kernel What is your kernel version and what is your libusb-1.0 version? > http://www.df8oe.de/listusb.txt One thing that sticks out is: wTotalLength 8704 Another is: wMaxPacketSize 0x0002 1x 2 bytes These clearly show an endianness problem which would more likely be in your version of libusb-1.0 than in the kernel. //Peter From df8oe at gmx.de Fri Jun 7 16:20:15 2013 From: df8oe at gmx.de (df8oe at gmx.de) Date: Fri, 7 Jun 2013 18:20:15 +0200 Subject: rtl-tcp-server running on avm-fritzbox In-Reply-To: <20130607144425.19612.qmail@stuge.se> References: <201306060841.04785.df8oe@gmx.de> <201306071437.16549.df8oe@gmx.de> <20130607144425.19612.qmail@stuge.se> Message-ID: <201306071820.15893.df8oe@gmx.de> Am Freitag, 7. Juni 2013, 16:44:25 schrieb Peter Stuge: > df8oe at gmx.de wrote: > > For debugging is not enabled by default in this kernel > > What is your kernel version and what is your libusb-1.0 version? > > > http://www.df8oe.de/listusb.txt > > One thing that sticks out is: > > wTotalLength 8704 > > Another is: > > wMaxPacketSize 0x0002 1x 2 bytes > > These clearly show an endianness problem which would more likely be > in your version of libusb-1.0 than in the kernel. > > > //Peter Linux version 2.6.32.41 (gcc version 4.6.3 (Buildroot 2011.11) ) #1 Fri Feb 15 15:09:13 CET 2013 mips GNU/Linux libusb-1.0.9 http://svn.freetz.org/trunk/make/libs/libusb1/ The problem is I cannot take a "normal-standard" distribution because this is a dsl/voip/Wlan/DECT-Router working with mips-processor and I must use the installed system adding some useful components. So I choose "freetz", with is an extension for the installed software for this embedded system. Because important modules are not open source I am not free in choosing a quite different kernel eg. But I dont give up hoping the problem can be fixed... I have contact with the freetz-maintainers and if there is a problem in the libusb1 and I can tell them where it is - they will update to a version without the bug (??) df8oe, Andreas From peter at stuge.se Fri Jun 7 23:12:11 2013 From: peter at stuge.se (Peter Stuge) Date: Sat, 8 Jun 2013 01:12:11 +0200 Subject: rtl-tcp-server running on avm-fritzbox In-Reply-To: <201306071820.15893.df8oe@gmx.de> References: <201306060841.04785.df8oe@gmx.de> <201306071437.16549.df8oe@gmx.de> <20130607144425.19612.qmail@stuge.se> <201306071820.15893.df8oe@gmx.de> Message-ID: <20130607231212.31162.qmail@stuge.se> df8oe at gmx.de wrote: > > What is your kernel version and what is your libusb-1.0 version? > > Linux version 2.6.32.41 (gcc version 4.6.3 (Buildroot 2011.11) ) #1 > Fri Feb 15 15:09:13 CET 2013 mips GNU/Linux This kernel version is over two years old. I don't want to think about how many THOUSAND commits there have been in the kernel since then. > libusb-1.0.9 > http://svn.freetz.org/trunk/make/libs/libusb1/ This should be OK, but I'm not sure that it was tested very much on BE machines. Please build a version of the library with debugging. > The problem is I cannot take a "normal-standard" distribution > because this is a dsl/voip/Wlan/DECT-Router working with > mips-processor Of course you have to cross-compile libusb-1.0.9 (better yet libusb.git code) for the target. //Peter From df8oe at gmx.de Mon Jun 10 09:14:19 2013 From: df8oe at gmx.de (df8oe at gmx.de) Date: Mon, 10 Jun 2013 11:14:19 +0200 Subject: rtl-tcp-server running on avm-fritzbox In-Reply-To: <20130607231212.31162.qmail@stuge.se> References: <201306060841.04785.df8oe@gmx.de> <201306071820.15893.df8oe@gmx.de> <20130607231212.31162.qmail@stuge.se> Message-ID: <201306101114.19541.df8oe@gmx.de> Am Samstag, 8. Juni 2013, 01:12:11 schrieb Peter Stuge: > df8oe at gmx.de wrote: > > > What is your kernel version and what is your libusb-1.0 version? > > > > Linux version 2.6.32.41 (gcc version 4.6.3 (Buildroot 2011.11) ) #1 > > Fri Feb 15 15:09:13 CET 2013 mips GNU/Linux > > This kernel version is over two years old. I don't want to think > about how many THOUSAND commits there have been in the kernel since > then. > Because the kernel is modified by the supplier AVM I choose an very easy possibility: I installed Debian on a Raspberry Pi and compiled rtlsdr there. It was working at the first try :) Many thanks for your replies - but I see no easy way to get the driver running on the embedded system without the risk of destroying other functions... The power consumption of the RP is not so high - so I have 2 "servers" running in my network ;) Andreas, df8oe From zilvinas at augma.lt Wed Jun 12 08:13:33 2013 From: zilvinas at augma.lt (Zilvinas, LY2SS) Date: Wed, 12 Jun 2013 11:13:33 +0300 Subject: e4000 r820t signal output Message-ID: <51B82DAD.8010402@augma.lt> Hi all ! Please excuse me if this is a bit off topic. Is there any sense to tap the IQ signals from e4k or r820t to the internal or external soundcards. Is it possible at all (with r820t particularly) ? (It could seem like reinventing funcube dongle, bet the performance/price would be still right) Best regards, Zilvinas From leif at sm5bsz.com Wed Jun 12 15:20:09 2013 From: leif at sm5bsz.com (Leif Asbrink) Date: Wed, 12 Jun 2013 17:20:09 +0200 Subject: e4000 r820t signal output In-Reply-To: <51B82DAD.8010402@augma.lt> References: <51B82DAD.8010402@augma.lt> Message-ID: <20130612172009.a5cb7ebb.leif@sm5bsz.com> Hello Zilvinas, This should work very well:-) The IQ signals are balanced and the proper procedure would be to use a differential amplifier to convert the voltage difference between I+ and I- to a voltage difference between the ground reference and the input of the soundcard. You can try just putting a wire from I+ to the soundcard input. A capacitor in series might be needed. The simple solution might add hum and noise near the center frequency. Such problems are cured with the differential amplifier. Then, do the same for Q:-) Regards Leif / SM5BSZ > Hi all ! > > Please excuse me if this is a bit off topic. > > Is there any sense to tap the IQ signals from e4k or r820t to the > internal or external soundcards. > Is it possible at all (with r820t particularly) ? > > (It could seem like reinventing funcube dongle, bet the > performance/price would be still right) > > Best regards, > > Zilvinas > From f.gugliuzza at hacklabproject.org Wed Jun 12 15:58:42 2013 From: f.gugliuzza at hacklabproject.org (Francesco Gugliuzza) Date: Wed, 12 Jun 2013 17:58:42 +0200 Subject: e4000 r820t signal output In-Reply-To: <20130612172009.a5cb7ebb.leif@sm5bsz.com> References: <51B82DAD.8010402@augma.lt> <20130612172009.a5cb7ebb.leif@sm5bsz.com> Message-ID: If the tuner output impedance is known, you could use a balun transformer of the right type to convert the signal, and get total galvanic isolation as an extra. Also, you need a good lowpass filter because the audio cards sample at 44100 or 48000 Hz and the lowest tuner output bandwidth is surely > 1 MHz. 2013/6/12 Leif Asbrink > Hello Zilvinas, > > This should work very well:-) > > The IQ signals are balanced and the proper > procedure would be to use a differential > amplifier to convert the voltage difference > between I+ and I- to a voltage difference between > the ground reference and the input of the soundcard. > > You can try just putting a wire from I+ to > the soundcard input. A capacitor in series > might be needed. The simple solution might > add hum and noise near the center frequency. > Such problems are cured with the differential > amplifier. > > Then, do the same for Q:-) > > Regards > > Leif / SM5BSZ > > > > Hi all ! > > > > Please excuse me if this is a bit off topic. > > > > Is there any sense to tap the IQ signals from e4k or r820t to the > > internal or external soundcards. > > Is it possible at all (with r820t particularly) ? > > > > (It could seem like reinventing funcube dongle, bet the > > performance/price would be still right) > > > > Best regards, > > > > Zilvinas > > > > -- Francesco Gugliuzza B.Sc. in Computer Engineering HackLabProject.org Administrator Linux user #374630 Tel (VoIP geographic number): +39 0921440446 Tel (Libera il VoIP number): 5125320 E-mail: f.gugliuzza at hacklabproject.org -------------- next part -------------- An HTML attachment was scrubbed... URL: From jsalsburg at bellsouth.net Wed Jun 12 16:22:27 2013 From: jsalsburg at bellsouth.net (Jay Salsburg) Date: Wed, 12 Jun 2013 11:22:27 -0500 Subject: e4000 r820t signal output In-Reply-To: <51B82DAD.8010402@augma.lt> References: <51B82DAD.8010402@augma.lt> Message-ID: Yes! Tap. The only problem is a sound card can only accommodate 48K - 96K bandwidth, some maybe as wide as 192K. The noise at the center of tuning may wipe out much of this tuning. What is needed is a tunable "Zero IF" double-balanced converter like a SA602(612) or NC1496 on the output of the Tuner chip driven by a very clean VC oscillator, these chips have differential I/Q inputs so the tuner can be used as it was designed and not constrained by that funky 8-bit A to D chip. A 16 bit I/Q sound card conversion of the double-balanced IF tuner output would reveal much the slow 8 bit converter does not, like very good noise cancelation, ultra narrow-band RADAR signals, very high speed frequency hopping, and spread spectrum modulation. I would create such a thing if I were not involved in more profitable ventures. Jay Salsburg -----Original Message----- From: osmocom-sdr-bounces at lists.osmocom.org [mailto:osmocom-sdr-bounces at lists.osmocom.org] On Behalf Of Zilvinas, LY2SS Sent: Wednesday, June 12, 2013 3:14 AM To: osmocom-sdr at lists.osmocom.org Subject: e4000 r820t signal output Hi all ! Please excuse me if this is a bit off topic. Is there any sense to tap the IQ signals from e4k or r820t to the internal or external soundcards. Is it possible at all (with r820t particularly) ? (It could seem like reinventing funcube dongle, bet the performance/price would be still right) Best regards, Zilvinas ----- No virus found in this message. Checked by AVG - www.avg.com Version: 2013.0.2904 / Virus Database: 3199/6404 - Release Date: 06/12/13 From leif at sm5bsz.com Wed Jun 12 17:45:14 2013 From: leif at sm5bsz.com (Leif Asbrink) Date: Wed, 12 Jun 2013 19:45:14 +0200 Subject: e4000 r820t signal output In-Reply-To: References: <51B82DAD.8010402@augma.lt> <20130612172009.a5cb7ebb.leif@sm5bsz.com> Message-ID: <20130612194514.5ab78bbd.leif@sm5bsz.com> Hi Francesco and all, If you do it the simple way, that is use the signal as is like is done in the Funcube you can not use a balun transformer. Going from DC to 48 or 96 kHz is not possible. Audio transformers can not provide that bandwidth. It is possible to go from perhaps 5 to 96 kHz in a transformer, but highly impractical because of the heavy magnetic shielding that would be required. A differential amplifier is cheap and efficient and can repalce an ideal transformer. Something like the output amplifier in fig 1. here: http://www.sm5bsz.com/linuxdsp/rxiq/antiali.htm R124 and R125 would go to the two outputs of one audio channel. A typical soundcard samples at HF. Maybe 5 MHz or higher. The soundcard uses digital filters for anti-aliasing and there is no need to add a good filter. Aliasing may be caused by signals near the Nyquist frequency and multiples of it, but very simple filters are needed to suppress those signals since the Nyquist frequency is typically several MHz. Regards Leif > If the tuner output impedance is known, you could use a balun transformer > of the right type to convert the signal, and get total galvanic isolation > as an extra. > Also, you need a good lowpass filter because the audio cards sample at > 44100 or 48000 Hz and the lowest tuner output bandwidth is surely > 1 MHz. > > > 2013/6/12 Leif Asbrink > > > Hello Zilvinas, > > > > This should work very well:-) > > > > The IQ signals are balanced and the proper > > procedure would be to use a differential > > amplifier to convert the voltage difference > > between I+ and I- to a voltage difference between > > the ground reference and the input of the soundcard. > > > > You can try just putting a wire from I+ to > > the soundcard input. A capacitor in series > > might be needed. The simple solution might > > add hum and noise near the center frequency. > > Such problems are cured with the differential > > amplifier. > > > > Then, do the same for Q:-) > > > > Regards > > > > Leif / SM5BSZ > > > > > > > Hi all ! > > > > > > Please excuse me if this is a bit off topic. > > > > > > Is there any sense to tap the IQ signals from e4k or r820t to the > > > internal or external soundcards. > > > Is it possible at all (with r820t particularly) ? > > > > > > (It could seem like reinventing funcube dongle, bet the > > > performance/price would be still right) > > > > > > Best regards, > > > > > > Zilvinas > > > > > > > > > > -- > Francesco Gugliuzza > B.Sc. in Computer Engineering > HackLabProject.org Administrator > Linux user #374630 > Tel (VoIP geographic number): +39 0921440446 > Tel (Libera il VoIP number): 5125320 > E-mail: f.gugliuzza at hacklabproject.org From jsalsburg at bellsouth.net Thu Jun 13 19:07:46 2013 From: jsalsburg at bellsouth.net (Jay Salsburg) Date: Thu, 13 Jun 2013 14:07:46 -0500 Subject: e4000 r820t signal output Message-ID: Yes! Tap. The only problem is a sound card can only accommodate 48K - 96K bandwidth, some maybe as wide as 192K. The noise at the center of tuning may wipe out much of this tuning. What is needed is a tunable "Zero IF" double-balanced converter like a SA602(612) or NC1496 on the output of the Tuner chip driven by a very clean VC oscillator, these chips have differential I/Q inputs so the tuner can be used as it was designed and not constrained by that funky 8-bit A to D chip. A 16 bit I/Q sound card conversion of the double-balanced IF tuner output would reveal much the slow 8 bit converter does not, like very good noise cancelation, ultra narrow-band RADAR signals, very high speed frequency hopping, and spread spectrum modulation. I would create such a thing if I were not involved in more profitable ventures. Jay Salsburg -----Original Message----- From: osmocom-sdr-bounces at lists.osmocom.org [mailto:osmocom-sdr-bounces at lists.osmocom.org] On Behalf Of Zilvinas, LY2SS Sent: Wednesday, June 12, 2013 3:14 AM To: osmocom-sdr at lists.osmocom.org Subject: e4000 r820t signal output Hi all ! Please excuse me if this is a bit off topic. Is there any sense to tap the IQ signals from e4k or r820t to the internal or external soundcards. Is it possible at all (with r820t particularly) ? (It could seem like reinventing funcube dongle, bet the performance/price would be still right) Best regards, Zilvinas ----- No virus found in this message. Checked by AVG - www.avg.com Version: 2013.0.2904 / Virus Database: 3199/6404 - Release Date: 06/12/13 From mrzfrn at gmail.com Thu Jun 13 19:26:29 2013 From: mrzfrn at gmail.com (mrzfrn) Date: Thu, 13 Jun 2013 21:26:29 +0200 Subject: MiricsFlexiTVMSi3101-LinuxSupport Message-ID: <51BA1CE5.2050807@gmail.com> I got a quite good (AM/SW)/FM/DAB/DVB-T USB dongle : Mirics FlexiTV MSi3101. http://www.mirics.com http://www.mirics.com/node/31 WFM , DAB/DAB+ DVB-T signals on air well decoded by Mirics software ( FMDABplayer even on slow netBook, while DVB-T needs fast PC) but Windows binary . I noticed basic Linux software at address http://cgit.osmocom.org/libmirisdr/ What the status of this software just now ( june 2013 )? Thanks. Francesco From zilvinas at augma.lt Fri Jun 14 07:30:58 2013 From: zilvinas at augma.lt (Zilvinas, LY2SS) Date: Fri, 14 Jun 2013 10:30:58 +0300 Subject: e4000 r820t signal output In-Reply-To: <20130612194514.5ab78bbd.leif@sm5bsz.com> References: <51B82DAD.8010402@augma.lt> <20130612172009.a5cb7ebb.leif@sm5bsz.com> <20130612194514.5ab78bbd.leif@sm5bsz.com> Message-ID: <51BAC6B2.6040402@augma.lt> Hi and thanks all for the remarks guys! On 2013.06.12 20:45, Leif Asbrink wrote: > A differential amplifier is cheap and efficient and can > repalce an ideal transformer. Something like the output > amplifier in fig 1. here: > http://www.sm5bsz.com/linuxdsp/rxiq/antiali.htm > R124 and R125 would go to the two outputs of > one audio channel. AD797 - (ouch) this one is quite expensive over here: ~26EUR for both I&Q channels. I mean not that expensive one couldn't buy but maybe one could find cheaper solution without degrading overall unit characteristics? I am not specialist in operational amplifiers (just an amateur) and don't know what are criteria for choosing right device in this place. What about kind of instrumentational opamp f.e. INA333 ? (This would cost half of that . Besides they are single power supply) (or even dual ne5532? less than 1EUR but not sure how to avoid dual power supply) Best regards, Zilvinas > > A typical soundcard samples at HF. Maybe 5 MHz or higher. > The soundcard uses digital filters for anti-aliasing > and there is no need to add a good filter. Aliasing > may be caused by signals near the Nyquist frequency > and multiples of it, but very simple filters are needed > to suppress those signals since the Nyquist frequency > is typically several MHz. > > Regards > > Leif > > > >> If the tuner output impedance is known, you could use a balun transformer >> of the right type to convert the signal, and get total galvanic isolation >> as an extra. >> Also, you need a good lowpass filter because the audio cards sample at >> 44100 or 48000 Hz and the lowest tuner output bandwidth is surely > 1 MHz. >> >> >> 2013/6/12 Leif Asbrink >> >>> Hello Zilvinas, >>> >>> This should work very well:-) >>> >>> The IQ signals are balanced and the proper >>> procedure would be to use a differential >>> amplifier to convert the voltage difference >>> between I+ and I- to a voltage difference between >>> the ground reference and the input of the soundcard. >>> >>> You can try just putting a wire from I+ to >>> the soundcard input. A capacitor in series >>> might be needed. The simple solution might >>> add hum and noise near the center frequency. >>> Such problems are cured with the differential >>> amplifier. >>> >>> Then, do the same for Q:-) >>> >>> Regards >>> >>> Leif / SM5BSZ >>> >>> >>>> Hi all ! >>>> >>>> Please excuse me if this is a bit off topic. >>>> >>>> Is there any sense to tap the IQ signals from e4k or r820t to the >>>> internal or external soundcards. >>>> Is it possible at all (with r820t particularly) ? >>>> >>>> (It could seem like reinventing funcube dongle, bet the >>>> performance/price would be still right) >>>> >>>> Best regards, >>>> >>>> Zilvinas >>>> >>> >> >> -- >> Francesco Gugliuzza >> B.Sc. in Computer Engineering >> HackLabProject.org Administrator >> Linux user #374630 >> Tel (VoIP geographic number): +39 0921440446 >> Tel (Libera il VoIP number): 5125320 >> E-mail: f.gugliuzza at hacklabproject.org From leif at sm5bsz.com Fri Jun 14 11:50:24 2013 From: leif at sm5bsz.com (Leif Asbrink) Date: Fri, 14 Jun 2013 13:50:24 +0200 Subject: e4000 r820t signal output In-Reply-To: <51BAC6B2.6040402@augma.lt> References: <51B82DAD.8010402@augma.lt> <20130612172009.a5cb7ebb.leif@sm5bsz.com> <20130612194514.5ab78bbd.leif@sm5bsz.com> <51BAC6B2.6040402@augma.lt> Message-ID: <20130614135024.ffde57ab.leif@sm5bsz.com> Hi Zilvinas, There are many OP-amps that you can choose from. The dynamic range of the E4000 is good, but not super-high so you do not need something expensive. (The WSE system is ultra-high performance, but today old.) Have a look at what they use in Softrock and other soundcard based hardware. You should be able to find something very cheap for a single 5V supply. I am not up-dated on what is available, but first I suggest you have a look in your junk-box:-) A poor OP-amp could cause an increased noise floor. That can easily be checked by connecting the two resistors to the same output pin of the E4000. If the noise does not go down significantly, OP-amp noise is too high. A poor OP-amp could also cause distortion. Typically a signal twice as far from the center as a strong signal. There could also be intermodulation. Dynamic range performance will be much better than for the Funcube because there is not a LNA in front of the E4000. With appropriate isolation from USB and computer noise you should get the same NF as possible with the Funcube. Regards Leif / SM5BSZ > Hi and thanks all for the remarks guys! > > On 2013.06.12 20:45, Leif Asbrink wrote: > > A differential amplifier is cheap and efficient and can > > repalce an ideal transformer. Something like the output > > amplifier in fig 1. here: > > http://www.sm5bsz.com/linuxdsp/rxiq/antiali.htm > > R124 and R125 would go to the two outputs of > > one audio channel. > > AD797 - (ouch) this one is quite expensive over here: ~26EUR for both > I&Q channels. I mean not that expensive one couldn't buy but maybe one > could find cheaper solution without degrading overall unit characteristics? > I am not specialist in operational amplifiers (just an amateur) and > don't know what are criteria for choosing right device in this place. > > What about kind of instrumentational opamp f.e. INA333 ? > (This would cost half of that . Besides they are single power supply) > > (or even dual ne5532? less than 1EUR but not sure how to avoid dual > power supply) > > Best regards, > > Zilvinas > > > > > A typical soundcard samples at HF. Maybe 5 MHz or higher. > > The soundcard uses digital filters for anti-aliasing > > and there is no need to add a good filter. Aliasing > > may be caused by signals near the Nyquist frequency > > and multiples of it, but very simple filters are needed > > to suppress those signals since the Nyquist frequency > > is typically several MHz. > > > > Regards > > > > Leif > > > > > > > >> If the tuner output impedance is known, you could use a balun transformer > >> of the right type to convert the signal, and get total galvanic isolation > >> as an extra. > >> Also, you need a good lowpass filter because the audio cards sample at > >> 44100 or 48000 Hz and the lowest tuner output bandwidth is surely > 1 MHz. > >> > >> > >> 2013/6/12 Leif Asbrink > >> > >>> Hello Zilvinas, > >>> > >>> This should work very well:-) > >>> > >>> The IQ signals are balanced and the proper > >>> procedure would be to use a differential > >>> amplifier to convert the voltage difference > >>> between I+ and I- to a voltage difference between > >>> the ground reference and the input of the soundcard. > >>> > >>> You can try just putting a wire from I+ to > >>> the soundcard input. A capacitor in series > >>> might be needed. The simple solution might > >>> add hum and noise near the center frequency. > >>> Such problems are cured with the differential > >>> amplifier. > >>> > >>> Then, do the same for Q:-) > >>> > >>> Regards > >>> > >>> Leif / SM5BSZ > >>> > >>> > >>>> Hi all ! > >>>> > >>>> Please excuse me if this is a bit off topic. > >>>> > >>>> Is there any sense to tap the IQ signals from e4k or r820t to the > >>>> internal or external soundcards. > >>>> Is it possible at all (with r820t particularly) ? > >>>> > >>>> (It could seem like reinventing funcube dongle, bet the > >>>> performance/price would be still right) > >>>> > >>>> Best regards, > >>>> > >>>> Zilvinas > >>>> > >>> > >> > >> -- > >> Francesco Gugliuzza > >> B.Sc. in Computer Engineering > >> HackLabProject.org Administrator > >> Linux user #374630 > >> Tel (VoIP geographic number): +39 0921440446 > >> Tel (Libera il VoIP number): 5125320 > >> E-mail: f.gugliuzza at hacklabproject.org > > From smile_k at libero.it Sun Jun 16 19:13:26 2013 From: smile_k at libero.it (Favati) Date: Sun, 16 Jun 2013 21:13:26 +0200 Subject: rtl_fm question Message-ID: Hi! I'm modifying rtl_fm to change frequency and demod based on the date/time. For the frequency i've no problem, but changing the modulation from wfm to any other gives me no outout at all. Which is the best approach to change the demod type in realtime? (whitout shutting down rtl_fm of course!) Many thanks! From jdowjunkmail at earthlink.net Mon Jun 24 08:50:42 2013 From: jdowjunkmail at earthlink.net (jdowjunkmail at earthlink.net) Date: Mon, 24 Jun 2013 01:50:42 -0700 Subject: Enhancement Request Message-ID: <51C80862.8000004@earthlink.net> Background: Suppose you have four dongles. (That's easy for me. I do.) Suppose they all identify as one of two types. (Three of one and one of the other.) Suppose you want to load a specific dongle when you start the program. And you don't want it to matter if you move the dongles around. How do I do that? Welllll, I hacked. Unfortunately rtlsdr.dll does not allow me to do this even if I alter the serial numbers, at least not the way I want. There is a fly in the rtlsdr ointment when fishing for the USB strings for Manufacturer ID, product ID, and serial number. You must formally open the device to read these values. The dongle can only be opened by one opener at a time. Unfortunately SDRSharp is laid out inside such that it searches for dongles whenever you open Config. And any time you open Config you also have a dongle open. So far this is not too bad a problem. If there's a dongle I cannot poll for strings, that's the one I have open. Er, not so fast, Joanne. What if you have two or more copies open? There will be more than one dongle you cannot poll. I had to figure out how I could identify the dongle I have open even if I move dongles. So I dug in and found some things libusb.dll can do. The chief one is get the USB path for the dongle. And it can do that without actually opening the dongle. Now when filling out the dialog if I cannot read strings for a dongle I get the dongle's path from the device list used and by cheating on the device handle rtlsdr.dll hands out for opened dongles I extracted information I needed to get the path that way. If they compare it's obviously my dongle. If they do not compare it's another dongle that is already opened so I don't show it in the list. Request: Please implement two new entry points which expose the libusb_get_port_path functionality, one by index (same as for strings) and one for rtlsdr handle which I finessed to prove the technique above. The libusb_get_device function will take the device handle and spit out the device list entry that matches the handle. Thanks for considering this. {^_^} Joanne/W6MKU From Francesco.Marzano at roma1.infn.it Wed Jun 12 18:58:22 2013 From: Francesco.Marzano at roma1.infn.it (marzano) Date: Wed, 12 Jun 2013 20:58:22 +0200 Subject: MiricsFlexiTV-MSI3101 Message-ID: <51B8C4CE.2080003@roma1.infn.it> A quite good (AM/SW)/FM/DAB/DVB-T USB dongle. Mirics FlexiTV MSi3101 (two chips MSi001+MSi2500) http://www.mirics.com http://www.mirics.com/node/31 Data sheets available, register yourself at Mirics portal. RF Tuner: MSi001/MSi002 ( used by FUNcubrePro+ ?), seems better than R820T ADC and HiSpeed USB interface: MSi2500 . ADC 10 bit , better than 8 bit RTL2832U aDC. Windows binary software available: FMDABplayer, DVB_T decoder, Windows lib.dll and API doc. IO-DATA GV-TV100 stick sold on Japan . Stick, chips and dev board available from Mirics. Linux software at address http://cgit.osmocom.org/libmirisdr/ It seems project on stand-by . Can author Steve Markgraf comment ? Francesco From llaniscudani at gmail.com Tue Jun 11 16:10:13 2013 From: llaniscudani at gmail.com (daniel .) Date: Tue, 11 Jun 2013 18:10:13 +0200 Subject: rtl-sdr line to support Asus U3100 MINI_PLUS_V2 Message-ID: Hello The Asus U3100MINI_PLUS_V2 rtp_eeprom output is: $ rtl_eeprom Found 1 device(s): 0: Afatech Using device 0: Afatech Found Fitipower FC0013 tuner Current configuration: ______________________________ ____________ Vendor ID: 0x1b80 Product ID: 0xd3a8 Manufacturer: Realtek Product: Rtl2832UDVB Serial number: Serial number enabled: no IR endpoint enabled: no Remote wakeup enabled: yes the line to add support to this is in librtlsdr.c - known_devices { 0x1b80, 0xd3a8, "Afatech" }, thanks. -------------- next part -------------- An HTML attachment was scrubbed... URL: From coinchon at yahoo.com Wed Jun 26 07:21:31 2013 From: coinchon at yahoo.com (Mathias Coinchon) Date: Wed, 26 Jun 2013 00:21:31 -0700 (PDT) Subject: realtek rumor Message-ID: <1372231291.8167.YahooMailNeo@web162106.mail.bf1.yahoo.com> Hello, I heard a rumor from someone from industry that Realtek RTL2832u chip would soon be phased out and replaced by another chip (with lower sensitivity according to my contact). Does anybody has more information about this ? Regards Mathias -------------- next part -------------- An HTML attachment was scrubbed... URL: From david.jacobowitz at gmail.com Thu Jun 27 08:32:09 2013 From: david.jacobowitz at gmail.com (David Jacobowitz) Date: Thu, 27 Jun 2013 01:32:09 -0700 Subject: minimum sample rates? Message-ID: I'm trying to write some code that uses a Raspberry Pi not just to capture the samples, but also to do the DSP. The signal I'm working (aviation VOR) is narrow and the DSP is not too hard-core, but it'll be easier if I can sample at a lower rate. Though rtlsdr_set_sample_rate returns no errors and rtlsdr_get_sample_rate returns the same value I gave it (or quite close), it does not appear to be working. I had been working at 250ksps but wanted to try 125ksps or even 62500sps, but none of those seem to work. Is this just a limitation of the hardware? I have a NooElec RTL2832 / R820T device if that matters. Regards, Dave J -------------- next part -------------- An HTML attachment was scrubbed... URL: From 246tnt at gmail.com Thu Jun 27 08:49:01 2013 From: 246tnt at gmail.com (Sylvain Munaut) Date: Thu, 27 Jun 2013 10:49:01 +0200 Subject: minimum sample rates? In-Reply-To: References: Message-ID: > Is this just a limitation of the hardware? Yes. Cheers, Sylvain From david.jacobowitz at gmail.com Thu Jun 27 09:11:28 2013 From: david.jacobowitz at gmail.com (David Jacobowitz) Date: Thu, 27 Jun 2013 02:11:28 -0700 Subject: minimum sample rates? In-Reply-To: References: Message-ID: Thanks, Sylvain. That makes sense. I probably wouldn't have spent so much time on it if the lib gave some indication that it wasn't setting those rates. On Thu, Jun 27, 2013 at 1:49 AM, Sylvain Munaut <246tnt at gmail.com> wrote: > > Is this just a limitation of the hardware? > > Yes. > > Cheers, > > Sylvain > -------------- next part -------------- An HTML attachment was scrubbed... URL: From 246tnt at gmail.com Thu Jun 27 09:16:57 2013 From: 246tnt at gmail.com (Sylvain Munaut) Date: Thu, 27 Jun 2013 11:16:57 +0200 Subject: minimum sample rates? In-Reply-To: References: Message-ID: > Thanks, Sylvain. That makes sense. I probably wouldn't have spent so much > time on it if the lib gave some indication that it wasn't setting those > rates. Well, we don't really know exactly what rates are supported by the hardware or not since there isn't exactly good documentation on it ... and the lib will just pass that to the hardware as long as the range fits into the "registers". This way users can experiment ... But if you want reliable operation you should stick to "known" working samples rates like 250k 1M 1.024M 2M 2.048M 2.4M Cheers Sylvain From david.jacobowitz at gmail.com Thu Jun 27 10:52:37 2013 From: david.jacobowitz at gmail.com (David Jacobowitz) Date: Thu, 27 Jun 2013 03:52:37 -0700 Subject: minimum sample rates? In-Reply-To: References: Message-ID: One could probably write a program that can tune to a known signal and then test through a range of sample rates to determine which ones generate a data stream that when treated at the expected sample rate gives the expected signal .... a project for another day. On Thu, Jun 27, 2013 at 2:16 AM, Sylvain Munaut <246tnt at gmail.com> wrote: > > Thanks, Sylvain. That makes sense. I probably wouldn't have spent so much > > time on it if the lib gave some indication that it wasn't setting those > > rates. > > Well, we don't really know exactly what rates are supported by the > hardware or not since there isn't exactly good documentation on it ... > and the lib will just pass that to the hardware as long as the range > fits into the "registers". This way users can experiment ... > > But if you want reliable operation you should stick to "known" working > samples rates like 250k 1M 1.024M 2M 2.048M 2.4M > > Cheers > > Sylvain > -------------- next part -------------- An HTML attachment was scrubbed... URL: From f.gugliuzza at hacklabproject.org Thu Jun 27 11:03:41 2013 From: f.gugliuzza at hacklabproject.org (Francesco Gugliuzza) Date: Thu, 27 Jun 2013 13:03:41 +0200 Subject: minimum sample rates? In-Reply-To: References: Message-ID: You shouldn't need to do any tuning. http://cgit.osmocom.org/rtl-sdr/tree/include/rtl-sdr.h search "test mode" in the file. Once you have enabled the counter, just check the incoming samples to see if you drop any data. I *think* rtl_test does it when running in async mode, but I'm not so sure. I'd like an official word from the devs on this. 2013/6/27 David Jacobowitz > One could probably write a program that can tune to a known signal and > then test through a range of sample rates to determine which ones generate > a data stream that when treated at the expected sample rate gives the > expected signal > > .... a project for another day. > > > On Thu, Jun 27, 2013 at 2:16 AM, Sylvain Munaut <246tnt at gmail.com> wrote: > >> > Thanks, Sylvain. That makes sense. I probably wouldn't have spent so >> much >> > time on it if the lib gave some indication that it wasn't setting those >> > rates. >> >> Well, we don't really know exactly what rates are supported by the >> hardware or not since there isn't exactly good documentation on it ... >> and the lib will just pass that to the hardware as long as the range >> fits into the "registers". This way users can experiment ... >> >> But if you want reliable operation you should stick to "known" working >> samples rates like 250k 1M 1.024M 2M 2.048M 2.4M >> >> Cheers >> >> Sylvain >> > > -- Francesco Gugliuzza B.Sc. in Computer Engineering HackLabProject.org Administrator Linux user #374630 Tel (VoIP geographic number): +39 0921440446 Tel (Libera il VoIP number): 5125320 E-mail: f.gugliuzza at hacklabproject.org -------------- next part -------------- An HTML attachment was scrubbed... URL: From alexey.kruglov at gmail.com Sun Jun 30 19:05:01 2013 From: alexey.kruglov at gmail.com (Alexey Kruglov) Date: Sun, 30 Jun 2013 23:05:01 +0400 Subject: Delay after rtlsdr_cancel_async Message-ID: <51D0815D.7080702@gmail.com> Hi. I noticed there is a 1 second delay in librtlsdr between calling rtlsdr_cancel_async and releasing from rtlsdr_read_async. I'm writing a radio scanner and this delay limits scanning speed considerably. I tracked the delay down to this piece of code in librtlsdr.c: while (RTLSDR_INACTIVE != dev->async_status) { r = libusb_handle_events_timeout(dev->ctx, &tv); [...] if (RTLSDR_CANCELING == dev->async_status) { next_status = RTLSDR_INACTIVE; [...] if (dev->dev_lost || RTLSDR_INACTIVE == next_status) { >>>> libusb_handle_events_timeout(dev->ctx, &tv); break; } The marked line is almost always(?) invoked when all the events are already processed, hence it blocks for tv = 1 second. In fact I don't understand what this `if' is supposed to do at all. I tried to modify this line by changing the delay to 0 making it a non-blocking call (leaving the timeout of 1 sec for the former libusb_handle_events_timeout call). This worked with my project and the 1 second delay disappeared. So, I suppose, it's better to either a) remove the whole `if' or b) make the latter libusb_handle_events_timeout non-blocking as I did. I checked that b) works for me, and didn't test a) yet. The question is: what that `if (dev->dev_lost ...' is supposed to do? And the two possible fixes for 1 second delay bug depending on the answer are (diffs): === 1630a1631 > struct timeval tv_nb = { 0, 0 }; 1696c1697 < libusb_handle_events_timeout(dev->ctx, &tv); --- > libusb_handle_events_timeout(dev->ctx, &tv_nb); === === 1694,1698d1693 < < if (dev->dev_lost || RTLSDR_INACTIVE == next_status) { < libusb_handle_events_timeout(dev->ctx, &tv); < break; < } === -- GPG fp: 6DFA 1186 7576 DE60 6CCB EB39 AD81 1733 EEBB 970A From jaysalsburg at gmail.com Wed Jun 26 18:15:02 2013 From: jaysalsburg at gmail.com (Jay Salsburg) Date: Wed, 26 Jun 2013 13:15:02 -0500 Subject: a new and more mature SDR Dongle Message-ID: Have any of you seen this? It appears as though we have a new and more mature SDR Dongle available. http://www.mirics.com/sites/default/files/MSi3101_new.pdf http://www.mirics.com/node/6 Thanks goes out to Francesco for supplying the links to this site -------------- next part -------------- An HTML attachment was scrubbed... URL: