On 7 Feb 2020, at 17:20, dave.roberts via Groups.Io <dave.roberts@...> wrote:

Any luck with it Steve?

Should interface using Soapy - all set up using Quisk Radios screen.

Works fine with Lime board. TxRx also working but flaky. See Jim's notes. Lime also works with GNU

73? Bob? g3udi




On Feb 5 2020, Jonathan Guthrie wrote:

I know it's a little late after sending a couple of emails to the group, but hello the group!

I'm interested in using Quisk with this HackRF One that I have. I'd like to eventually use it as an HF radio.? I realize that Quisk may need some modifications for this to work, but I'm a reasonably capable programmer, so I'm expecting to have to help with that.

Hello Mario,

Does this problem still exist? If so I will work on a fix.

Jim
N2ADR

Just FYI, I've run Linux as my primary desktop OS since, um, 1998 maybe??? 2002 certainly.?? It's been a long time, anyway.? I tend to just not use stuff that's inconvenient for whatever reason.?? Sometimes, I have the temerity to complain that some free software package or other would suit my needs better if it were different, and I tend to get grief about that opinion.?? My attitude is that you can play the piano in the closet, if you want to, but if you think it would be cool for people to use the software you write (whether you charge them for it or not) then it behooves you to listen to their gripes.?? You may not choose to change what you do based upon those gripes, but telling opinionated users to, um, just leave if they don't like something about what you've written tends to be received less well.

I see that Jim N2ADR has already weighed in on Quisk vs GRC.? My opinion is slightly different than his is, so I'm going to express it.?? The problem with trying to implement an actual application in a tool like GRC is that it is harder than implementing it without the toolkit.?? While I might try to use the gnuradio libraries if I was doing something like Quisk myself, that wouldn't really relieve me of the responsibility to understand how it works so I'd still have to know the algorithms and then on top of it I'd have to do the UI design and then I'd have to understand the details of how GRC works to make an application that works in that environment.?? Understanding how GRC works is a nontrivial exercise.?? If you're doing an application from scratch, you need to know all of that except you don't need to understand the details about GRC.

There comes a point where a platform stops being a help and starts being a hindrance, and it's up to the judgment of the implementor to decide when that point is reached.

I like GRC.?? I think it's a really nice tool for messing around with software radios, but it's not really the tool I would use for implementing an application.

On 2/4/2020 12:13 PM, ag5gt@... wrote:

Hi Terry, Jonathan,

I guess we have all suffered frustrations for all of the reasons you point out. GRC has certainly not been stable through its early development years. While learning, I had trouble finding existing examples that still worked in the gnuradio/GRC release I had just installed. My recent impression is that has tended toward stabilization as it has gained exposure. My further impression is one might insulate oneself from some of that by maintaining code "outside of tree" which is to say, maintaining it such that dependencies on the rapidly evolving, experimental parts of the "tree" are narrowly circumscribed. A further thought was that the user interface would be pure python, built outside of GRC. But, that's said with little experience on my part.

What drove me back to Linux recently was microsoft's termination of win7 support, with no sure path to install win10 on my perfectly good PC hardware. That, of course, was preceded by many rounds of marketing-driven changes in behavior, combined with frustration that my PC was often too busy doing Microsoft's bidding to be available for actual use by me. Relative to those considerations, Linux Mint has been a breath of fresh air. That's where I have been running Quisk lately. In theory at least, I have control of my Linux PC, so I need not ever again change anything, once it all works to my satisfaction. So, I am done with the Windows ecosystem for reasons that go beyond bug handling.

With respect to Linux library changes and Quisk, I recently had to navigate around the effects of python3 relative to python2 as the two interacted with my Linux window manager. What works best for 3 was not entirely functional in 2, given the light weight window manager, apparently. I moved to python3. I take that as the price of progress. I'm sure GRC has some issues with the python transition, too, but I have not yet explored that.

Except for the longer term stability concerns in gnuradio or linux generally, how reasonable do you think it would be to try to implement Quisk in GRC?

bruce, ag5gt

Just FYI, I've run Linux as my primary desktop OS since, um, 1998 maybe? 2002 certainly. It's been a long time, anyway. I tend to just not use stuff that's inconvenient for whatever reason. Sometimes, I have the temerity to complain that some free software package or other would suit my needs better if it were different, and I tend to get grief about that opinion. My attitude is that you can play the piano in the closet, if you want to, but if you think it would be cool for people to use the software you write (whether you charge them for it or not) then it behooves you to listen to their gripes. You may not choose to change what you do based upon those gripes, but telling opinionated users to, um, just leave if they don't like something about what you've written tends to be received less well.

I see that Jim N2ADR has already weighed in on Quisk vs GRC. My opinion is slightly different than his is, so I'm going to express it. The problem with trying to implement an actual application in a tool like GRC is that it is harder than implementing it without the toolkit. While I might try to use the gnuradio libraries if I was doing something like Quisk myself, that wouldn't really relieve me of the responsibility to understand how it works so I'd still have to know the algorithms and then on top of it I'd have to do the UI design and then I'd have to understand the details of how GRC works to make an application that works in that environment. Understanding how GRC works is a nontrivial exercise. If you're doing an application from scratch, you need to know all of that except you don't need to understand the details about GRC.

There comes a point where a platform stops being a help and starts being a hindrance, and it's up to the judgment of the implementor to decide when that point is reached.

I like GRC. I think it's a really nice tool for messing around with software radios, but it's not really the tool I would use for implementing an application.

On 2/4/2020 12:13 PM, ag5gt@... wrote:

Hi Terry, Jonathan,

I guess we have all suffered frustrations for all of the reasons you point out. GRC has certainly not been stable through its early development years. While learning, I had trouble finding existing examples that still worked in the gnuradio/GRC release I had just installed. My recent impression is that has tended toward stabilization as it has gained exposure. My further impression is one might insulate oneself from some of that by maintaining code "outside of tree" which is to say, maintaining it such that dependencies on the rapidly evolving, experimental parts of the "tree" are narrowly circumscribed. A further thought was that the user interface would be pure python, built outside of GRC. But, that's said with little experience on my part.

What drove me back to Linux recently was microsoft's termination of win7 support, with no sure path to install win10 on my perfectly good PC hardware. That, of course, was preceded by many rounds of marketing-driven changes in behavior, combined with frustration that my PC was often too busy doing Microsoft's bidding to be available for actual use by me. Relative to those considerations, Linux Mint has been a breath of fresh air. That's where I have been running Quisk lately. In theory at least, I have control of my Linux PC, so I need not ever again change anything, once it all works to my satisfaction. So, I am done with the Windows ecosystem for reasons that go beyond bug handling.

With respect to Linux library changes and Quisk, I recently had to navigate around the effects of python3 relative to python2 as the two interacted with my Linux window manager. What works best for 3 was not entirely functional in 2, given the light weight window manager, apparently. I moved to python3. I take that as the price of progress. I'm sure GRC has some issues with the python transition, too, but I have not yet explored that.

Except for the longer term stability concerns in gnuradio or linux generally, how reasonable do you think it would be to try to implement Quisk in GRC?

bruce, ag5gt

I know it's a little late after sending a couple of emails to the group, but hello the group!

I'm interested in using Quisk with this HackRF One that I have. I'd like to eventually use it as an HF radio. I realize that Quisk may need some modifications for this to work, but I'm a reasonably capable programmer, so I'm expecting to have to help with that.

Hi Terry, Jonathan,

I guess we have all suffered frustrations for all of the reasons you point out. GRC has certainly not been stable through its early development years. While learning, I had trouble finding existing examples that still worked in the gnuradio/GRC release I had just installed. My recent impression is that has tended toward stabilization as it has gained exposure. My further impression is one might insulate oneself from some of that by maintaining code "outside of tree" which is to say, maintaining it such that dependencies on the rapidly evolving, experimental parts of the "tree" are narrowly circumscribed. A further thought was that the user interface would be pure python, built outside of GRC. But, that's said with little experience on my part.

What drove me back to Linux recently was microsoft's termination of win7 support, with no sure path to install win10 on my perfectly good PC hardware. That, of course, was preceded by many rounds of marketing-driven changes in behavior, combined with frustration that my PC was often too busy doing Microsoft's bidding to be available for actual use by me. Relative to those considerations, Linux Mint has been a breath of fresh air. That's where I have been running Quisk lately. In theory at least, I have control of my Linux PC, so I need not ever again change anything, once it all works to my satisfaction. So, I am done with the Windows ecosystem for reasons that go beyond bug handling.

With respect to Linux library changes and Quisk, I recently had to navigate around the effects of python3 relative to python2 as the two interacted with my Linux window manager. What works best for 3 was not entirely functional in 2, given the light weight window manager, apparently. I moved to python3. I take that as the price of progress. I'm sure GRC has some issues with the python transition, too, but I have not yet explored that.

Except for the longer term stability concerns in gnuradio or linux generally, how reasonable do you think it would be to try to implement Quisk in GRC?

bruce, ag5gt