[Lowfer] Re: Wolf Development

[David Willmore]

> Most of my WOLF decoding has been done using a P133 with 16MB of RAM and 
> Win95. The processing time required is roughly equal to the duration of the 
> audio recording that is being processed. More recently, I have had a PII 
> 300 with 128MB of RAM and Win98- this does the job about twice as fast, but 
> apart from that makes no difference to the result that I can detect. Most 
> of the time, both machines are quite OK to use with the current off-line 
> WOLF processing.

The current program will still exist and any features that can be squeezed
back into it will be.  So, you're not going to lose any functionality.

> Apart from the very significant cost issue, I am wary of rushing off to buy 
> new PCs at the drop of a hat - the maker's specs say nothing about the RF 
> noise level a PC will produce, or peculiar software & hardware bugs, so 
> there is always a risk of spending a lot of money on something that 
> actually makes reception more difficult. One also has to consider the 
> benefits to be achieved - if having the latest hot PC allows you to achieve 
> 0.03dB improvement in SNR, I would prefer to put up with waiting a while, 
> whilst if there were major improvements to be obtained, I might think about it.

If we can change the code to make an improvement and that change comes at the
cost of more CPU/Memory, we'll evaluate the usefullness of it.  But, the first
thing to do will be to optimize the existing version.  I don't think that
it is unreasonable to expect the processing rate to increase by a decent
factor.  I'll quote no numbers until I have some actual data.  Given that,
existing machines will get faster.  If we then trade some of that for 
better receive performance, we might get back to the speed is at now.  Is that
bad? ;)

There may very well be improvements that knock slower machines below the
'real time' threshold.  That doesn't stop you from doing ofline decoding
nor night-to-night QSOs--unless you're dropped about 2x below the threshold.
Even then a little creativity could allow that to creap up to 3x.

All said, we have a lot of room to play with.  But the fact will always be
that noone will be *forced* to get a faster machine.  At best, I'd call it
'encouraged'.

> I am no computer expert; one day I might play with different operating 
> systems, but at the moment all the radio software I use is either DOS or 
> windows based, so introducing other operating systems would probably just 
> render me totally confused!

The current app is command line based--a dos box under windows.  Not true
DOS as that would limit the application too much, but just text-in-a-Window(s).

Since I use Linux--in command line mode--you can be assured that the current
text mode functionality will remain.  If we do anything, the GUI version
will only be an *addition* to that functionality.

> Good luck with the WOLF development - will be interested to see the results.

I'm a bit interested, too. ;)

Cheers,
David