G2 and the future of remote operations

[RobB]

In the sales blurb for the G2, Apache labs claim the G2 will be remote operation capable but this feature will rely on further software development. Can anyone say if this development work is in progress and what form it is likely to take? True thin client like the Flex or simply via a VNC?
We can only hope this prospect is not like the adaptive pre distortion promises the Flex have made for ages and are only now acknowledging the 6000 series hardware is not powerful enough but we are happy to sell you a new model that can (subject to future development)

[w-u-2-o]

In the sales blurb for the G2, Apache labs claim the G2 will be remote operation capable but this feature will rely on further software development.

While that's a true statement, it was very disappointing to see it posted as part of the website advertising copy when the G2 was first announced. Again, true statement, but it's awfully forward looking and might be taken by some people as something that would ultimately become true.

Can anyone say if this development work is in progress and what form it is likely to take?

AFAIK, there is nobody working on this at all. It has been discussed many times. More than one developer has said that the coding of Thetis is so muddled that it would literally be easier to start over with a new thin client/server architecture than to try to separate Thetis into client and server elements. It's likely this would be much easier for piHPSDR but, again, it does not appear that there is any developer interest in doing so.

True thin client like the Flex or simply via a VNC?

We can already use screen sharing software to remote operate Thetis-based stations (VNC is too slow to be suitable, but there are other choices). Hence what we are really seeking is a true thin client/server approach.

It's also worth noting that Apache claiming this is a capability of the G2 and not prior hardware is misleading. Any openHPSDR compatible radio from the original Hermes to the latest and greatest could easily be run by provisioning a server alongside the radio. It need not be a built-in Pi or, for those who have them, one of the 7000's that has a built-in PC. Inside or outside the Apache box makes no difference at all.

We can only hope this prospect is not like the adaptive pre distortion promises the Flex have made for ages...

It depends on whether you read the Apache ad copy as a promise or as a potential capability.

Given that we are currently in one of the periodic development lulls that have been regular occurrences in the openHPSDR world it's safe to say thin client/server app's are not likely to appear anytime soon, if ever. I am very happy to be wrong about that, though!

[Trucker]

We can only hope this prospect is not like the adaptive pre distortion promises the Flex have made for ages and are only now acknowledging the 6000 series hardware is not powerful enough but we are happy to sell you a new model that can (subject to future development)

The Flex 6000 radios already have the necessary connections made internally to make Adaptive Pre-Destortion work. And well over a year ago one of Flex Radio's developers told me they had it running on a 6000 on their test bench. And that it was working fine. But, he said that he did not know if it would be activated anytime soon because they ( Flex Radio) have not had that much demand for APD.
The new 8000 radios are a result of unavailability of components that were used in the 6000 radios. Getting the new components that are faster and getting rid of the SD card does give them more opportunities to add features to the 8000 radios.
Either way, time will tell if they actually add features people want.
James
WD5GWY

[laurencebarker]

I can't see it happening with Thetis, for the reasons Scott gives. But that's not the only option.

piHPSDR already has a part implemented client-server mode that could be developed. that would need the remote end to be executing an instance of piHPSDR.

And at Friedrichshafen last year there was a presentation "RemoteSDR - Browser Based SDR Transceiver Control" by André Buhart, F1ATB. He uses his own DSP code, but to could be adapted to use WDSP. Already a bidirectional TX/RX architecture. This looks to be like a viable starting point.

The problem as always is one of time and insufficient developers. If others want to join in, there is lots of opportunity.

[oe3ide]

I am also using Thetis (Reid's version for HL2, mi0bot) in combination with my HL2. I am also working remote with it.
It works flawless up to 192khz bandwidth. In my case I use a simple port-forwarding in both directions, so I can grab also the TCI information to control the amp, ant-switch, autom. ant-tuner, etc...
WAN bandwith:
"server"-side: 9 MBit/s up, 92 MBit/s down, 20ms
"client"-side: 19 MBit/s up, 185 MBit/s down, 22ms

Not sure if this would work also with the Anan firmware versions...

73 de Ernst

[w-u-2-o]

Not sure if this would work also with the Anan firmware versions...

It will. Other than perhaps a few bits here and there there is no difference between the HL2 and Apache firmware interface.

However, to operate in a truly remote fashion, Thetis on local computer, hardware remote, the network connection must be fast, nearly flawless, and not exhibit significant changes in latency. Few people enjoy that sort of situation where the internet is involved.