Hola, Jose, I know exactly what graph you speak of but it is not on their web site any longer and I could not find it anywhere else.
When I wrote my initial message I was going to put the claim that SCS had some years ago about how deep into the noise the Pactor modes can go. But I could not remember for sure. My recollection was around -16 db for P2 and P3, it was similar to what you recalled. The graph indicated that there could not be much throughput at low S/N ratios since it was a steep curve at that point. Just like Clover II, you would have a link, hear the signal, but it just would not pass any data, even though it stayed "connected" for a time. Rick, KN6KB has published his interpretation of the throughput of P modes and claimed that by -5 db, the throughput was near zero for all of them. It is my understanding that Pactor is varying numbers of DPSK tones not unlike PSK except faster. At the slowest speed wouldn't it be like PSK100 using two tones or DPSK100? Then they add tones up to 16DPSK if the conditions will allow sufficient throughput. What is it about current sound cards that will not allow this to work on a 2 gig or better computer? The switching of the computer seems adequate on my ICOM rig to tolerate a similar mode. If more time was needed then one could use a longer block, and and a sufficient window listening for the ARQ. If this still was not practical (due to insufficient processing power of the computer) then you could move the computational power requirement to the background and pipeline it as in SCAMP. I wonder how different this would be from the non ARQ (but later correctible) DRM modes they use for SSTV? 73, Rick, KV9U Jose Amador wrote: > Actually, with bad S/N ratios, thruput suffers a lot. > > I have been checking at > http://www.scs-ptc.com/pactor.html before sending this > reply, and there is some info, but I cannot find the > graph > comparing several modes up to P-II, that I think was > published in QST in the late 90s. > > P-I with memory ARQ holds the link down to -5 dB (in a > 3 kHz bandwidth, SNR in pactor´s bandwidth is BETTER > than that). P-II holds the link down to -18 dB, but > that´s all, thruput is almost nil. > > P-III shines at some 20 dB SNR, because with more > complex constellations, the distance between the > constellation points diminishes, and is less noise > tolerant. Interleaving, compression and convolutional > encoding with Viterbi decoding all that helps a lot, > but there is a big distance between "telegraphy" and > "telepathy"...that is, digging for signals down into > the noise. > > After all, the most succesful newer digital modulation > > schemes are using some of the tricks that have made > P-II and P-III successful, but there are physical > limits that cannot be violated. Specially some forms > of FEC, block codes and interleaving are helpful to > avoid signal degradation due to QRM, fading and > multipath, and data recovery without repeats thanks to > block codes. > > But it carries a price tag, the more noise and QRM > resistant a mode is, generally implies more latency, > that is, makes it less "keyboard friendly". > > 73 de Jose, CO2JA > > > > Need a Digital mode QSO? Connect to Telnet://cluster.dynalias.org Other areas of interest: The MixW Reflector : http://groups.yahoo.com/group/themixwgroup/ DigiPol: http://groups.yahoo.com/group/Digipol (band plan policy discussion) Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/digitalradio/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
