While I somewhat agree that there is a perception of ARQ modes being slower, this has not been based upon my actual testing of FAE400. The ability of FAE400 to work at least as deep into the noise as PSK31, and probably a bit deeper with its memory ARQ capabilities, not available in any other sound card mode at this time, and its much greater throughput, often makes it difficult to keep up with the flow if your keyboarding speed is much below 40 wpm.
If you want to send a file or an image, you need to have ARQ, but it really does appear that most hams just want relatively quick boiler plate contacts. On VHF, you have more tolerance, but I have found PSK31 to be less than desirable since any multipath (aircraft, hills, etc.) can make it scramble data. The WRAP program makes a lot of sense when you do not want a connected mode, such as a one to many bulletin. Using MT-63 on phone circuits will work because you typically have very good signals. If you did not have this, you could not use phone as well as MT-63. But for weaker signal applications, MT-63 can not compete well with other modes. I do not agree that the learning curve is too great for FAE400 compared with NBEMS or any other system. If you are a user of Multipsk, then only a few things need to be learned. The clutter of the program does mean that very few hams are moving in that direction anymore. Even myself, who at once time mostly used Multipsk, have moved to fldigi and for almost total rig control as well as non-ARQ digital modes find Ham Radio Deluxe to be the best for a completely integrated program. Nothing else can come even slightly close for now. But if you want to run packet radio at 300 or 1200 baud, or FAE400 and the faster FAE (2000) and similar modes that have a basis from the old ALE protocols, the only freely available program is Multipsk. One the other hand fldigi has the unique capability of non only being the only cross platform multimode digital program, but acts as the core program for NBEMS and PSKmail. What is complicated that to get basic rig control and centralized logging, quite a few programs need to be running and that is quite complicated. Bottom line: If you don't use a given program and mode on a regular basis (daily or at least weekly) you are not going to be using it for public service either. And if we are to ever develop ARQ BBS systems that can work with sound card modes and provide a superior solution to 300 baud packet on HF, we have to have these technologies. 73, Rick, KV9U wrote: > Rick, > > ARQ is perfect for being sure emcomm and other messages are delivered > error-free, but for chatting, most people will not want to slow things > down waiting for an acknowledgment. Rather, they just ask for a repeat > when it is needed. In addition, we can correct errors (a single > apparently misspelled word, for example) with what we think is the right > word, or fill in a missing word with our brains (since we can visualize > things in context). Overall, this is usually faster than using ARQ and > good enough for casual conversation. > > However, for sending pictures, ARQ is sometimes absolutely necessary, > especially with a compression technique in which a single byte ruins the > whole picture. > > The Western Pennsylvania emcomm group has fully implemented NBEMS over > both repeaters and simplex, but mostly over VHF, and, because VHF tends > to be more constant and tends to be much more error-free than HF, did > not want to spend the extra time (on any mode or speed) to slow down for > ARQ, so we developed the Wrap program, which sends a checksum at the end > of the message, and error-free reception can be verified that way. > > On our MARS emcomm net, MT63 on HF usually produces error-free copy on > the statewide net, and Wrap is useful with MT63 also just for verifying > that there were no errors, or indicating that a resend is necessary. > > However, far enough away, there may always be some stations, under poor > conditions, that either need a repeat of the whole message, or need to > have ARQ used to repeat bad blocks if there are many. The advantage of > Wrap is that a one-on-one ARQ link is not needed except when that is the > only way to get the message through. Bulletins can be transmitted in > MT63 and received error-free by most stations, with others needing a > resend, or perhaps a relay. > > On VHF SSB weak signal phone, it is common practice to use "vocal FEC" > (to coin a term!) and just repeat callsigns twice or "over" twice to > accomplish the contact during poor conditions. The standard call on CW > is a 3x3 call, which is a type of "manual" FEC to try to get at least > one of each callsign through. > > Most files these days are very large, compared to those in DOS days, and > with the bandwidth limitations on HF, it just takes too long to send a > very large file, even using a fast mode and ARQ, so I think there is > little interest in file transfer on the bands either. Still, I have > always though it would be very convenient to be able to send a schematic > to explain something, but these days, that can be done with most > stations by using the Internet. > > FAE400 is a great development, but the learning curve is too steep for > emcomm operators thrust into a position without much training. That is > why we elected to use commonly used digital modes and provide ARQ with > flarq when necessary, and the learning curve is not as steep that way. > > ARQ definitely has its place, but is usually needed for messaging or > when poor conditions require it (for example, if QSB is strong). I think > that is why only a handful of hams have any interest in ARQ modes for > chatting. > > That is how I see it. Other's opinions may vary, of course. > > 73, Skip KH6TY > NBEMS Development Team > >
