[Fwd: Re: [linuxham] S/N numbers for Olivia]
- Hi Dave,Just joined the group, this post will probably bounce until I'm accepted as a member. Post it on my behalf if you want please.I'm using Pavel's code stack, when I last looked at fldigi's Olivia code it was the same as Pavel's code. Any different would be due to soundcard calibration I think - there shouldn't be any other reason I can think of. FWIW I changed the code (but not the logic) so that it would compile with the aged Windows Visual Studio 6.0 C++ compiler.I hope to attack DM780 next year, there's a lot that could be done, apart from anything else I want to get Thor added.Simon Brown, HB9DRV
- --- In email@example.com, w1hkj <w1hkj@...> wrote:
>Also when comparing Linux to Windows there will be different soundcard
> I hope to attack DM780 next year, there's a lot that could be done,
> apart from anything else I want to get Thor added.
drivers and native sample rates.
FWIW I do know Linux programming, just write Windows 'cos I like it :-)
- Welcome to the group, Simon. There are many Windows users in this
group. Everyone here is civil most of the time. ;)
--- In firstname.lastname@example.org, "Simon Brown" <simon.brown@...> wrote:
> --- In email@example.com, w1hkj <w1hkj@> wrote:
> > I hope to attack DM780 next year, there's a lot that could be done,
> > apart from anything else I want to get Thor added.
> Also when comparing Linux to Windows there will be different soundcard
> drivers and native sample rates.
> FWIW I do know Linux programming, just write Windows 'cos I like it :-)
> Simon HB9DRV
- Perhaps some would like to know how I compared DM780's and fldigi's
performance when copying olivia.
First, my setup:
Flex 5000 A transceiver
Generic PC with AMD64 duo-core processor running WinXP home, PowerSDR
1.14.0, DM780, and fldigi 3.30
line out from the F5K to the line-in on my MacBook Pro
I started out by running fldigi on both the MacBook pro and on the PC
on the same signal. Fldigi was connected to PowerSDR via VAC on the
PC. Fldigi on the Mac was getting the analog signal. Over time I
monitored both copies of fldigi receiving the same signals in varying
conditions. From what I could tell by eyeball looking at the S/N
display and at the actual copy, the performance of fldigi on both
platforms was virtually identical. S/N numbers never varied by more
than a couple tenths of a dB (as fast as I could look from one to the
other). The copy was also virtually identical. From this I deduced
that the two computing platforms were performing virtually identically.
After that I stopped fldigi on the PC platform and started DM780.
Since the S/N report from DM780 does not appear to be the same as for
fldigi I could not compare. All I could do is compare the copy from
DM780 and from fldigi. This comparison was more difficult but one
thing was clear, fldigi would continue to copy correctly at lower
signal levels than DM780. Copy would become garbled from DM780 before
it would from fldigi. Trying to estimate signal level change on
signals experience slow QSB and comparing that to copy quality let me
to believe that fldigi is giving equivalent copy to DM780 when S/N is
decreased by about 3-4dB.
Now this was a quick eyeball comparison and hardly qualifies as a
rigorous scientific test but it has satisfied me that there is an
improvement over DM780 when running fldigi.
73 de Brian, WB6RQN
Brian Lloyd - brian HYPHEN wb6rqn AT lloyd DOT com