This is in response to a comment misplaced in another thread on the old Forum website about this situation.
Original posters question: "what can i do about the recurring short dropouts, where the waterfall freezes for about three to four seconds and then when sending the ptt also fails. after this short time everything runs normally for a while, and then the same game starts again."
My response: This dropout is actually a lockup due to a HRD system file running in the background called KeepAlive. HRD support was "unaware" of this until one of the technicians logged onto my computer and witnessed it first hand. It happens every 5 minutes (to the second from when DM780 is started) and when it does, all things DM780 cease for 2 to 5 seconds or so, depending on the speed of your CPU and RAM, until this process complete whatever its doing.
I was told that this process should NOT be causing this problem, but they witnessed it first hand several times during the connection. To display this properly for them, I had WSJTx also running and I was sending my call using the CW portion of DM780 since it is the most noticeable way of showing the lockup happening. The WSJTx MSK144 mode waterfall showed exactly what was happening to my call sign when the HRD waterfall just stopped and then caught up after the lockup, thus indicating NO stoppage while ultimately displaying the data received during the lockup, which is in error. The problem is during TRANSMIT, when all tone output stops during this KeepAlive operation. There was a noticeable gap in the transmitted tone detection in the WSJTx waterfall that was not evident in the HRD waterfall thus sending a wrong/invalid CW character, or invalid data in other modes. But it does pickup where it left off, just sending wrong data in the process.
You can see the results in the attached image of the WSJTx MSK144 waterfall with DM780 sending my call (WB5JJJ). If you look carefully and decode the dots and dashes in your mind, you will see it happening. This is totally repeatable and has been unresolved for the past few years through numerous updates. In the lower graph you will see the following:
......? B 5 J J J...... W B ?.....<gap>......? J J J....W B 5 J