|
Hi all
WARNING: Alpha! Not even BETA! I just would love some feedback, from a small number of people, I am not ready to release it officially...?
Are there any?super-brave souls who could try out this ALPHA firmware, 1_00_015 candidate which I labelled version 00000000 (I shall call alphas 8 numbers, for identification). The link is:
Note that when you install this, it will automatically perform a factory reset, because I did a lot of re-arrangement to the EEPROM to fix some mapping issues and tidy up.?
I am mostly interested in whether or not there's an improvement to the CW keying glitches, and to the CW sidetone latency. I admit I am nowhere?near as good at CW as a lot of you are, I trundle along at 16wpm. So I cannot easily tell if the sidetone latency is still an issue and my mistakes are more frequent than glitches too.?
This is the full list of changes SO FAR (not final), which is mostly a bunch of bug-fixes. A nice one is that it now remembers the frequency by band. Not returns you to the center frequency (by default FT8 frequency) when you change band.? - Bug fix: Another attempt to improve CW timing glitches.
- Removed unused S-meter step parameter.
- Remembers frequency by band.
- Re-arrangement of EEPROM config - automatically performs factory reset on upgrade to 1_00_015.
- Bug fix: if you tuned below 3.2MHz it locked up; now added minimum 3.2MHz frequency to avoid this out of range issue.
- Bug fix: PTT output port wasn't enabled; now PTT output is enabled
- Bug fix: When in Japanese band limits mode, message sending did not work.
- Bug fix: It allowed you to send CW messages when you're in Digi mode!
- Bug fix: Some of the Message presets got truncated, due to an EEPROM page boundary issue.
- Bug fix: when message editing, pressing the left button on a space didn't move right (if the space was really a null).
- Bug fix: substantial reduction in keying to sidetone latency, and sped up sidetone amplitude shaping 10x.
- Bug fix: when message editing, on exit of a long message it didn't scroll back to the beginning by itself.
- Moved CW-R parameter to the System config menu, alongside Sideband selection
Any feedback would be very welcome please.?
73 Hans G0UPL
|
Hans, just a check re: item 5:
Last I checked, for every band there is an upper and lower
frequency limit. It seemed that, if I hit that limit, everything
froze (IIRC). Couldn't even turn the unit off!
Thanks for your persistence in getting all this right!
73, Paul -- AI7JR
On 2/1/24 11:15, Hans Summers wrote:
Hi all
WARNING: Alpha! Not even BETA! I
just would love some feedback, from a small number of
people, I am not ready to release it officially...?
Are there any?super-brave souls who could try out this
ALPHA firmware, 1_00_015 candidate which I labelled version
00000000 (I shall call alphas 8 numbers, for identification).
The link is:
Note that when you install this, it will automatically
perform a factory reset, because I did a lot of re-arrangement
to the EEPROM to fix some mapping issues and tidy up.?
I am mostly interested in whether or not there's an
improvement to the CW keying glitches, and to the CW sidetone
latency. I admit I am nowhere?near as good at CW as a lot of
you are, I trundle along at 16wpm. So I cannot easily tell if
the sidetone latency is still an issue and my mistakes are
more frequent than glitches too.?
This is the full list of changes SO FAR (not final), which
is mostly a bunch of bug-fixes. A nice one is that it now
remembers the frequency by band. Not returns you to the center
frequency (by default FT8 frequency) when you change band.?
- Bug fix: Another attempt to improve CW timing glitches.
- Removed unused S-meter step parameter.
- Remembers frequency by band.
- Re-arrangement of EEPROM config - automatically performs
factory reset on upgrade to 1_00_015.
- Bug fix: if you tuned below 3.2MHz it locked up; now
added minimum 3.2MHz frequency to avoid this out of range
issue.
- Bug fix: PTT output port wasn't enabled; now PTT output
is enabled
- Bug fix: When in Japanese band limits mode, message
sending did not work.
- Bug fix: It allowed you to send CW messages when you're
in Digi mode!
- Bug fix: Some of the Message presets got truncated, due
to an EEPROM page boundary issue.
- Bug fix: when message editing, pressing the left button
on a space didn't move right (if the space was really a
null).
- Bug fix: substantial reduction in keying to sidetone
latency, and sped up sidetone amplitude shaping 10x.
- Bug fix: when message editing, on exit of a long message
it didn't scroll back to the beginning by itself.
- Moved CW-R parameter to the System config menu,
alongside Sideband selection
Any feedback would be very welcome please.?
73 Hans G0UPL
--
Paul -- AI7JR
|
Hmm, the archive seems to be 1k in size and contains a 0 byte file.
Or am I doing something wrong?
YO3GFH
op. Adrian
|
No the file is 000000000.qmx and 0 bytes in size.? Think you need to re-zip it Hans :)
|
> No the file is 000000000.qmx and 0 bytes in size.? Think you need to re-zip it Hans :)
Ooops sorry about that. No idea how that could happen! I have re-zipped it and re-uploaded it.?
On Thu, Feb 1, 2024 at 10:33?PM Danny Bower < m0sdb@...> wrote: No the file is 000000000.qmx and 0 bytes in size.? Think you need to re-zip it Hans :)
|
That looks better - installing now....
|
Hi Paul
Hans, just a check re: item 5:
Last I checked, for every band there is an upper and lower
frequency limit. It seemed that, if I hit that limit, everything
froze (IIRC). Couldn't even turn the unit off!
Thanks for your persistence in getting all this right!
I just tried that here, it seemed to work fine for me, I didn't see any issue. I set the lower limit of 7MHz to 6,990,000? and tuned down below it, and there wasn't any problem. No hanging, and no issues... Please could you try that again???
|
I will-- This was on the 0014 FW, not the newest! I was sort of
responding to how specific your note was for item 5.
It'll be a day or two for me to get to it. If you've checked for
it, personally, I'm satisfied. ;-)
Paul -- AI7JR
On 2/1/24 11:41, Hans Summers wrote:
Hi Paul
Hans, just a check re: item 5:
Last I checked, for every band there is an upper and
lower frequency limit. It seemed that, if I hit that
limit, everything froze (IIRC). Couldn't even turn the
unit off!
Thanks for your persistence in getting all this right!
I just tried that here, it seemed to work fine for me, I
didn't see any issue. I set the lower limit of 7MHz to
6,990,000? and tuned down below it, and there wasn't any
problem. No hanging, and no issues... Please could you try
that again???
--
Paul -- AI7JR
|
So - very quick test and yes, outright latency seems noticeably improved?
However, the clunking is still there and I am not convinced the keyer speed is correct.? 25wpm sounds too fast - nearer 27.? And there are still short dahs and long dits however they are less distracting from a pure keying sense as the latency is reduced but it still is just wrong.
For me the clunking is worst on 20m
|
Thanks Danny
So - very quick test and yes, outright latency seems noticeably improved?
However, the clunking is still there and I am not convinced the keyer speed is correct.? 25wpm sounds too fast - nearer 27.? And there are still short dahs and long dits however they are less distracting from a pure keying sense as the latency is reduced but it still is just wrong.
For me the clunking is worst on 20m
I didn't tackle clunking, only the things in my list.?
Very interesting about the keyer speed seeming incorrect... I don't see how that could happen. The dit duration is defined as 1200 / WPM milliseconds. So for 25wpm?you get 48 milliseconds. Isn't that correct??
So I just rattled out a long string of dits and measured them on the 'scope. See attached. 20 dits measures 1.924 seconds, which divided by 40 (because there is one space for every of the 20 dits) you get 0.0481s. So 48.1 milliseconds - as close to 48ms as measurement error would probably allow.?
Am I missing something?
73 Hans G0UPL
|
Running it into Morse Expert (Android app) that is measuring it at 26wpm however I often find slight discrepancies between key speed measuring tools.? It may just be that the clunking is making it sound faster as my brain may be registering those as extra dits.? Feel free to go with your calculations on this one :).
There are still mis-constructed dits and dahs though however they do seem to be less frequent.
Noted re: the clunking.
|
Done some fiddling here too, some observations: - used a single lever paddle, practice mode and live into a dummy load, 25wpm, iambic A and B, changed weight to 600; - still found shortened dahs and lenghted dits, but less frequently, same as Danny found; - while in practice mode, latency is very good, however, if I go live into a dummy load, seems that things are worse (latency almost as before update, and significantly more dit/dash malforming, worse in mode A). If I switch back to practice, things are not improving until a power cycle; - compared speed with my IC7300@25wpm and it seems to match; - clunking more pronounced in 20 and 80m, for me the worst is in 80m (I noted that you didn't work here); - just fyi: immediately after update and factory reset, it starts in 80m with very low volume. Increased the vol to a normal level, then switched band... boy was it loud.. turned the vol down and powered off, then powered on.. the audio level was so potent it blew my knickers clean off. However, I can't seem to reproduce it again; Hope it helps! YO3GFH op. Adrian ?
|
Some update regarding iambic A/B: seems that even in practice mode, if I switch to A and use it for a while, latency increases.. bit by bit, progressively? If I go back to B, things are ok again.
Hope it's not just me being tired and hallucinating, will try again tomorrow.
YO3GFH
op. Adrian
|
For those who use Facebook. Latency? What latency?
-Steve K1RF
------ Original Message ------
Date 2/1/2024 3:16:26 PM
Subject Re: [QRPLabs] #qnx #firmware ALPHA 1_00_015 called 00000000.zip
Thanks Danny
So - very quick test and yes, outright latency seems noticeably improved?
However, the clunking is still there and I am not convinced the keyer speed is correct.? 25wpm sounds too fast - nearer 27.? And there are still short dahs and long dits however they are less distracting from a pure keying sense as the latency is reduced but it still is just wrong.
For me the clunking is worst on 20m
I didn't tackle clunking, only the things in my list.?
Very interesting about the keyer speed seeming incorrect... I don't see how that could happen. The dit duration is defined as 1200 / WPM milliseconds. So for 25wpm?you get 48 milliseconds. Isn't that correct??
So I just rattled out a long string of dits and measured them on the 'scope. See attached. 20 dits measures 1.924 seconds, which divided by 40 (because there is one space for every of the 20 dits) you get 0.0481s. So 48.1 milliseconds - as close to 48ms as measurement error would probably allow.?
Am I missing something?
73 Hans G0UPL
|
Hello Hans
Just installed the test firmware.
After testing the WSPR mode, turned off my QMX now its dead .
Cannot get into boot loader even tried the jumper on the board.
?the QMX is totally dead.
Philip G7JUR
|
Hi Hans, 00000000 loaded, initially receive was quite low (not volume) but rx sensitivity.? ? ? ? ?keyer speed set to 27wpm it sounded quite a bit faster! switched of and on again , rx sensitivity came right up to good and keyer speed confirmed with RBN was spot on. still a little clunky but manageable and no sign of eratic dits or dahs... cheers Scotty gi0bey
On Thursday, 1 February 2024 at 22:25:33 GMT, Philip G7JUR <philip.g7jur@...> wrote:
Hello Hans
Just installed the test firmware.
After testing the WSPR mode, turned off my QMX now its dead .
Cannot get into boot loader even tried the jumper on the board.
?the QMX is totally dead.
Philip G7JUR
|
Hi Hans, 00000000 loaded, initially receive was quite low (not volume) but rx sensitivity.? ? ? ? ?keyer speed set to 27wpm it sounded quite a bit faster! switched of and on again , rx sensitivity came right up to good and keyer speed confirmed with RBN was spot on. still a little clunky but manageable and no sign of eratic dits or dahs...
How do you mean, "still a little clunky" - in what way??
|
Just installed the test firmware.
After testing the WSPR mode, turned off my QMX now its dead .
Cannot get into boot loader even tried the jumper on the board.
?the QMX is totally dead.
I've been playing about with WSPR mode. It doesn't work properly, all sorts of problems, but we knew that already - and I had not (in the Alpha) made any attempts to fix that (yet). But certainly I haven't seen any actual DAMAGE to a QMX, so your situation is perplexing, very.?
|
hi, audio wise probably sidetone volume as i run it through an external audio amp, maybe clunky is a poor description. sidetone is just not as well rounded as my qcx.
On Friday, 2 February 2024 at 13:32:59 GMT, Hans Summers <hans.summers@...> wrote:
Hi Hans, 00000000 loaded, initially receive was quite low (not volume) but rx sensitivity.? ? ? ? ?keyer speed set to 27wpm it sounded quite a bit faster! switched of and on again , rx sensitivity came right up to good and keyer speed confirmed with RBN was spot on. still a little clunky but manageable and no sign of eratic dits or dahs...
How do you mean, "still a little clunky" - in what way??
|
I found the fault, one of the BS170's had gone S/C.
Maybe the drive to the HF amp timing was incorrect.
73 Philip G7JUR
|
Paul - AI7JR