Loading ...
Sorry, an error occurred while loading the content.

Re: One more CW try

Expand Messages
  • sv1eia
    Doug, you can find past posts with measurements and you will see that PSDR has the smallest latency by far of all (counting of course with the same
    Message 1 of 19 , Feb 1, 2010
    • 0 Attachment
      Doug,
      you can find past posts with measurements and you will see that PSDR has the smallest latency by far of all (counting of course with the same parameters).
      Also PSDR has WAY more functionality than the other bare-bones applications.
      IMHO you cannot compare apples and oranges.



      --- In softrock40@yahoogroups.com, Mr Doug - <dsc3507@...> wrote:
      >
      > You are right, PSDR does have much more latency than other programs. In no other program do I need to run special drivers or set the latency so high as in PSDR to make it function. Why, would best be answered by those intimate with programming in PSDR. Somewhere internally there is either more processing time, more buffering, or just plain inefficiency.
      >
      >
      >
      >
      > ________________________________
      > From: kb3kjs <galicic@...>
      > To: softrock40@yahoogroups.com
      > Sent: Mon, February 1, 2010 8:34:05 AM
      > Subject: [softrock40] Re: One more CW try
      >
      >
      > I too feel Tom's pain. Rocky works fine with my paddles and keyer. How is it that there is no delay in Rocky but h delay in PSDR. Same AVR code and USB connection.
      >
      > I'm not a programmer but my theory is that it is related to the way PSDR interacts with the sound card or AVR code to make CW monitor sounds. Too much latency somewhere?
      >
      > I love PSDR-IQ and would love to use it with CW paddles/keyers.
      >
      > --- In softrock40@yahoogro ups.com, Bruce Tanner <bet110@> wrote:
      > >
      > > Glad that worked for you to a great degree. I have also experienced the
      > > 'popping' with Rocky that you describe. If you can locate it I would
      > > sure like to know what it is as well. I mostly experience with my USB
      > > audio interface (Match Box II). In my case something almost goes in to
      > > oscillation after popping and crackling for a short while.... (Seconds).
      > > At that point I get a very strong, consistent sound almost like a
      > > 'chopper' working. My screen goes nuts on the Rocky window and the only
      > > way I can restore that is also to turn it off and then back on. That
      > > always restores the normal functioning.
      > >
      > > I have thought there may be some sort of IRQ conflict causing it but I
      > > have not been able to investigate that yet. It is annoying when it
      > > occurs and almost always happens when I am making changes to settings
      > > requiring a new screen paint so it seems there is audio and video
      > > involved somehow.. Fortunately, it does not happen too often.
      > >
      > > Carry on...
      > >
      > > Bruce
      > > ------------ ------
      > >
      > > tom wrote:
      > > > Bruce,
      > > >
      > > > I tried Rocky again today as per your suggestion, this time with a Delta 44
      > > > sound card, with the keyer out of the loop and the paddle connected directly
      > > > to the USB-I2C card, and there did not appear to be any latency between the
      > > > paddle and the transmitted signal, as heard on my 746 or through the monitor
      > > > speaker of the PC which I have plugged into the native intel sound card
      > > > speaker-out jack. I was impressed after struggling for the past week with
      > > > PSDR.
      > > >
      > > > Although most of the time it seemed to work fine, on occasion the
      > > > transmitted signal would start popping, as otherd have heard in previous
      > > > threads. I will have to see under what circumstances this popping occurs. It
      > > > does not seem to be related to other apps running on the PC, but I have not
      > > > had a chance to study it. When the poppping started it did not clear up
      > > > till I shut down Rocky and restarted it. I can hear it on both PSK and CW.
      > > > It is not heard on the pc monitor speaker so it seems to be only on the
      > > > actual transmitted signal. I assume it is audible on the received end but
      > > > have not worked anyone yet to check that.
      > > >
      > > > I have no problem, manually switching from transmit to receive. I normally
      > > > don't care to work full break-in anyway.
      > > >
      > > > W7qiq pointed me to a thread (#12262) with a possible solution also, but
      > > > that uses a separate receiver and transmitter. I will study that as well.
      > > >
      > > >
      > > > Tom
      > > >
      > > > ----- Original Message -----
      > > > From: "Bruce Tanner" <bet110@>
      > > > To: <softrock40@yahoogro ups.com>
      > > > Sent: Saturday, January 30, 2010 5:20 PM
      > > > Subject: Re: [softrock40] Re: One more CW try
      > > >
      > > >
      > > >
      > > >> Tom, it is not as full featured but if you don't mind that... Rocky, is
      > > >> very good with CW. The keyer in that software sounds and feels like real
      > > >> time processing. I understand your issue very well as I also use both my
      > > >> kinesthetic awareness to sync with what I am hearing to use a paddle
      > > >> well. It makes no difference to the person on the other end if they hear
      > > >> a delay of a few micro-seconds but it can sure mess up your sending
      > > >> accuracy when you are transmitting.
      > > >>
      > > >> I find I have no problem with Rocky in this regard, but it is much less
      > > >> CPU thirsty so there is hardly a perceptible latency.
      > > >>
      > > >> I have used an outboard keyer with both programs and find Rocky works
      > > >> best with that as well... at least for now. With Rocky you can turn off
      > > >> the monitor embedded in the software and then you only hear your keyer's
      > > >> monitor which will be pretty much in real time. The receiving station
      > > >> may hear a delayed signal but they won't know the difference as long as
      > > >> you are able to send accurately. In either Rocky or PSDR all you need do
      > > >> is set the 'Iambic' feature to "OFF" (Unchecked) and then it will act as
      > > >> though you are using a straight key (on/off carrier). When you feed the
      > > >> output of your keyer to either software program with its "Iambic", OFF,
      > > >> it will act as though it is receiving the on/off of a straight key but
      > > >> it will do it in sync with what your keyer is sending as 'dits' and
      > > >> 'dahs'.
      > > >>
      > > >> I have recently found that if I do this with PSDR that it sends good CW
      > > >> as long as I can't hear the delayed sound... it is not the software's
      > > >> issue... it is that my reflexes in sending become very out of rhythm and
      > > >> I end up sending crazy stuff because to the delay between what I hear
      > > >> and the timing of my wrist action! The way I over come this is a little
      > > >> unhandy right now because I have not yet found the features in PSDR
      > > >> which indicate "Disable Monitor" to be working. I get the sound no
      > > >> matter how I make those settings to disable it... I am assuming that
      > > >> those features have not yet been enabled in this new version of PSDR-IQ.
      > > >> I have used the CWX feature and find that it send perfect code but
      > > >> again, I am not using any wrist action at a paddle when that is being
      > > >> sent.
      > > >>
      > > >> My work around for it is to use the keyboard, shortcuts to 'Mute' the
      > > >> receiver audio which effectively turns off the errant, delayed audio,
      > > >> and allows me to hear only the audio from my keyer. I then can send
      > > >> decent CW. I have actually monitored this on an alternate receiver and
      > > >> find that the code being sent by PSDR and the Softrock are actually
      > > >> pretty good but if you listen to the audio coming from the PSDR while
      > > >> you are sending it sounds like 'crap'!
      > > >>
      > > >> The un-handiness of this arrangement is that you have to keep taking
      > > >> time to manually turn on and off your audio since you will need it back
      > > >> on in order to hear the other station's signals.
      > > >>
      > > >> Hope this is helpful...
      > > >>
      > > >> Bruce, K2BET
      > > >>
      > > >>
      > > >> tom wrote:
      > > >>
      > > >>> Thanks Christos, that explains part of it...
      > > >>>
      > > >>> ----- Original Message -----
      > > >>> *From:* sv1eia <mailto:sv1eia@ ...>
      > > >>> *To:* softrock40@yahoogro ups.com <mailto:softrock40@yahoogro ups.com>
      > > >>> *Sent:* Saturday, January 30, 2010 12:20 PM
      > > >>> *Subject:* [softrock40] Re: One more CW try
      > > >>>
      > > >>> Tom,
      > > >>>
      > > >>> The fastest that USBtoI2C ATtiny45/85 AVR chip can send back info
      > > >>> is 12ms (~83 times/sec).
      > > >>> And it is not USB2, its plain old 1.1 (or even slower..)
      > > >>>
      > > >>> --- In softrock40@yahoogro ups..com
      > > >>> <mailto:softrock40@yahoogro ups.com>, "tom" <loeblt@> wrote:
      > > >>> >
      > > >>> > Christos,
      > > >>> >
      > > >>> > What I find odd about the latency issue is that the computer
      > > >>> processes inputs in fractions of a microsecond and USB 2 is very
      > > >>> fast, so why is the time delay between touching a keyer paddle and
      > > >>> hearing the result in the speaker take milliseconds instead of
      > > >>> microseconds. ...
      > > >>> >
      > > >>> > Tom
      > > >>> >
      > > >>> >
      > > >>> > ----- Original Message -----
      > > >>> > From: sv1eia
      > > >>> > To: softrock40@yahoogro ups.com
      > > >>> <mailto:softrock40% 40yahoogroups. com>
      > > >>> > Sent: Saturday, January 30, 2010 11:29 AM
      > > >>> > Subject: [softrock40] Re: One more CW try
      > > >>> >
      > > >>> >
      > > >>> >
      > > >>> > Tom,
      > > >>> >
      > > >>> > > I wonder whether the FLEX series radios have the same issues
      > > >>> with CW.
      > > >>> >
      > > >>> > Going into apples and oranges here in every aspect..
      > > >>> > :-)
      > > >>> >
      > > >>> > --- In softrock40@yahoogro ups.com
      > > >>> <mailto:softrock40% 40yahoogroups. com>, "wa1vta01452" <loeblt@> wrote:
      > > >>> > >
      > > >>> > > Hi,
      > > >>> > >
      > > >>> > > I know I have asked about this several times in the past and I
      > > >>> know there are numerous past threads on this but I am not quite
      > > >>> ready to give up trying to use CW with the RXTX and PSDR.
      > > >>> > >
      > > >>> > > Has anyone had success in connecting an electronic keyer to
      > > >>> the RXTX or used the paddle directly with a sidetone in a way that
      > > >>> sent clean CW using PSDR with no latency? Is Rocky a better choice
      > > >>> for this?
      > > >>> > >
      > > >>> > > I have not tried to connect the keyer directly to the RXTX
      > > >>> dot/dash and serial ports. I wonder if this would work vs the
      > > >>> USB-I2C connection that I used.
      > > >>> > > Christos recommended using CWX, which works fine with PSDR but
      > > >>> I prefer to use a keyer paddle rather than a keyboard for sending CW.
      > > >>> > >
      > > >>> > > I wonder whether the FLEX series radios have the same issues
      > > >>> with CW.
      > > >>> > >
      > > >>> > > I use a 2.4 GHZ dual core PC.
      > > >>> > >
      > > >>> > > Tom
      > > >>> > >
      > > >>> >
      > > >>>
      > > >>>
      > > >>>
      > > >>>
      > > >>> __________ Information from ESET Smart Security, version of virus
      > > >>> signature database 4821 (20100130) __________
      > > >>>
      > > >>> The message was checked by ESET Smart Security.
      > > >>>
      > > >>> http://www.eset com
      > > >>>
      > > >> ------------ --------- --------- ------
      > > >>
      > > >> Yahoo! Groups Links
      > > >>
      > > >>
      > > >>
      > > >>
      > > >
      > > >
      > > >
      > > > ------------ --------- --------- ------
      > > >
      > > > Yahoo! Groups Links
      > > >
      > > >
      > > >
      > > >
      > > > __________ Information from ESET Smart Security, version of virus signature database 4821 (20100130) __________
      > > >
      > > > The message was checked by ESET Smart Security.
      > > >
      > > > http://www.eset com
      > > >
      > > >
      > > >
      > > >
      > > >
      > >
      >
    • kb3kjs
      So you are using a keyer with the sidetone muted in PSDR-IQ. Have you been successful with paddles? Since I cant get the sidetone muted as you described my
      Message 2 of 19 , Feb 1, 2010
      • 0 Attachment
        So you are using a keyer with the sidetone muted in PSDR-IQ. Have you been successful with paddles?
        Since I cant get the sidetone muted as you described my workaround is turning down the RX volume on my external sound cards monitor loop. It works but its a real pain to have to ride the volume control every time I TX. Thanks for sharing your findings. -Joe

        --- In softrock40@yahoogroups.com, "Graeme" <gvj@...> wrote:
        >
        > Hi Bob,
        >
        > The sidetone can be muted in PSDR-IQ. On the main screen in the CW subsection check "Disable Monitor" and then on main screen under the "START" button ensure the "MON" button is deselected. Initially I guess due to a bug I needed to do this more than once before the settings were sticky. Christos has said several times that v1.19.3.15 is in beta (maybe even pre beta) and bugs are to be expected.
        >
        > Regarding latency as previously commented on. There is a finite time to process the slices of information, apply the mathematical formulae and assemble them. The time is dictated by how fast the computer is and how sharp you want the filters etc. and is near enough to constant for all software (not quite true) and if one is faster or has less latency than another it is probably not filtering as sharply or applying less of some process. In practice with everything set to sharpest filters, noise blanker on and all that stuff you can expect about 200 mSec with a 2 Gig dual core machine and a CPU load of 30-40%.
        >
        > On transmit the latency can be reduced on PSDR by adjusting the buffer size in Settings/Options/Buffer Size. On Tx with "CW Rx=2048 and Tx=1024" my CPU load is around 15%
        >
        > CW is working 100% with PowerSDR v1.19.3.14 and 15 for me. It took a bit of messing around and a lot of time reading the Flex 1000 manual and posts on this forum and the PowerSDR-IQ forum. It is worth doing this as when you understand most of how it all goes together and runs you can apply meaningful changes with some expectation of a certain result and if you don't get it then something is probably wrong.
        >
        > I do get some dropouts and have identified one of them. I have my mailer check the Internet every 10 mins and each time it does it puts a spike on the CPU and causes a brief dropout on my RX (or TX). I have some other dropouts which will be caused by similar devices stealing CPU time and of course they will be difficult to find. Christos pointed the PSDR forum to a neat little utility called DPC-latency http://www.thesycon.de/eng/home.shtml which is just the thing for this job.
        >
        > Now that I have PSDR going on CW there is nothing I would change except the tone from my keyer which is a 800 Hz square wave. I really must modify this to be a sinewave.
        >
        > My setup
        > Dell Inspiron E1505 Laptop (only USB and firewire ports)
        > Dual Core 2.2 g processor and 1.5 gB ram
        > Windows XP SP3
        > PowerSDR-IQ v1.19.3.15
        > Edirol FA-66 sound card
        > RX/TX6.3
        > Mobo4.3
        >
        > 73 Graeme ZL2APV
        > <SNIP>
        > > I have recently found that if I do this with PSDR that it sends good CW
        > > as long as I can't hear the delayed sound... it is not the software's
        > > issue... it is that my reflexes in sending become very out of rhythm and
        > > I end up sending crazy stuff because to the delay between what I hear
        > > and the timing of my wrist action! The way I over come this is a little
        > > unhandy right now because I have not yet found the features in PSDR
        > > which indicate "Disable Monitor" to be working. I get the sound no
        > > matter how I make those settings to disable it... I am assuming that
        > > those features have not yet been enabled in this new version of PSDR-IQ.
        > > I have used the CWX feature and find that it send perfect code but
        > > again, I am not using any wrist action at a paddle when that is being sent.
        > >
        > > My work around for it is to use the keyboard, shortcuts to 'Mute' the
        > > receiver audio which effectively turns off the errant, delayed audio,
        > > and allows me to hear only the audio from my keyer. I then can send
        > > decent CW. I have actually monitored this on an alternate receiver and
        > > find that the code being sent by PSDR and the Softrock are actually
        > > pretty good but if you listen to the audio coming from the PSDR while
        > > you are sending it sounds like 'crap'!
        > <SNIP>
        >
      • Mr Doug -
        OK, you should know such things. I guess this means that the other programs have built-in buffering you do not see or cannot change. The reason I came to that
        Message 3 of 19 , Feb 1, 2010
        • 0 Attachment
          OK, you should know such things. I guess this means that the other programs have built-in buffering you do not see or cannot change. The reason I came to that conclusion was that PSDR was so much more finicky in audio settings requiring very high manual latency settings for MME or DirectSound drivers while the other programs, for the most part, did not require that. I never measured actual latency, so as you say the others might have that built-in to make them work out of the box without the ASIO driver.

          Doug


          From: sv1eia <sv1eia@...>
          To: softrock40@yahoogroups.com
          Sent: Mon, February 1, 2010 11:34:05 AM
          Subject: [softrock40] Re: One more CW try

           

          Doug,
          you can find past posts with measurements and you will see that PSDR has the smallest latency by far of all (counting of course with the same parameters).
          Also PSDR has WAY more functionality than the other bare-bones applications.
          IMHO you cannot compare apples and oranges.
           



        • Giulio Scaroni
          unsubscribe
          Message 4 of 19 , Feb 2, 2010
          • 0 Attachment

            unsubscribe

          • Giulio Scaroni
            unscribe
            Message 5 of 19 , Feb 2, 2010
            • 0 Attachment
              unscribe

            • g8voip
              Hi Giulio, Your messages will not do anything! To leave the group you must do it yourself. Go to the top of the page, open Edit Membership and scroll down to
              Message 6 of 19 , Feb 2, 2010
              • 0 Attachment
                Hi Giulio,

                Your messages will not do anything!

                To leave the group you must do it yourself.

                Go to the top of the page, open 'Edit Membership' and scroll down to the bottom and click on the 'Leave Group' button.

                Press the button and 'Goodbye'!

                73, Bob G8VOI

                --- In softrock40@yahoogroups.com, Giulio Scaroni <ik2ded@...> wrote:
                >
                >
                > unsubscribe
                >
              • k5nwa
                ... Cecil k5nwa Never take life seriously. Nobody gets out alive
                Message 7 of 19 , Feb 2, 2010
                • 0 Attachment
                  At 04:56 PM 2/2/2010, you wrote:


                  >unscribe
                  >
                  >Every email includling this one has a link at the bottom to unsubscribe.

                  Cecil
                  k5nwa
                  < www.softrockradio.org > < www.qrpradio.com >
                  < http://parts.softrockradio.org/ >

                  Never take life seriously. Nobody gets out alive anyway.
                Your message has been successfully submitted and would be delivered to recipients shortly.