DSP - ANF Performance Specification

"I can wait for a fix." -- Dan KG0AQ

1)  How long have we all been waiting?

2)  How long should we wait?

3)  What is a reasonable expectation for timelines for fixes to longstanding, very basic software defects that never seem to get addressed, while lots of other ****-**** features are being added to up-charged releases?

When I was trying to make up my mind whether to buy a Flex or not, I really struggled with the knowledge that I would be painting myself into the corner of being 100% dependent on one company for hardware and software support. But being ever the believer in American quality and customer support I pulled the trigger and bought into it. I've been really disappointed in Flex's commitment to software support ever since. I'll never voluntarily make that mistake again.

Bill:

No offense intended, but when you say:
"We need to be carful in comparing the Flex DSP to other radios. Flex is doing the DSP in a much different way then other radios. Most modern radios do not do DSP 100% in software, they combine hardware filters with software."

I say HOGWASH!  The only thing that matters at the end of the day for ANY company, is the resulting performance of the product, NOT how the engineers
choose to implement it.  No company, not even Flex, gets a free pass because
they were trying to use some novel different method.    Again, I will repeat, it
does not matter how elegant or complicated or new or unique the engineering
approach is, IF it does not provide a useful working solution for the customer.

No car maker, in a crowded market, can make any sales if their car model
lacks tires, a reverse gear for backing up, or has a non-working turn-signal
system and a non-working air-conditioner and terribly loud wind noise in the
passenger cabin.   These are base standard features.  They simply must work
on par with other cars, and are considered base or commodity features. 
If base standard features are not up to consumer expected standards for a
specific competitor's product, sales will suffer.  Consumers do not tolerate
missing or poorly peforming base features.  These are a foundation block
that must be solid, before any new manufacturer starts to layer their unique
value proposition and innovation on top.  
NR, NB, and ANF are not unique innovations introduced by Flex, they are
base standard features that every recent radio on the market has today.
As numerous other posters in this thread point out;   Place an Anan or Icom or
Yaesu or Elecraft on the same table with the Flex, and compare the base features.
They may differ in the amount of noise they can reduce, but today, after all these
years, NONE of them should distort the audio badly.

So, once again, please explain to me WHY we have to be careful in comparing
Flex DSP to other radios?   Either the base features work as well as competition,
or they don't.  Like the little kid's used to say on those funny AT&T Television
commercials,  "It's not complicated"  :-)       It either works or it doesn't.
No amount of political hand waving will change that.

Now, we know for a fact that good DSP is a limited art and quite difficult and
complex (It's not a circuit that can be copied, it is proprietary code, amazingly
difficult to invent and refine).  Which means that there are very few qualified
engineers around the world who know how to implement high quality DSP.   
If FLex has not been able to improve their DSP notch and noise features in a
noticeable way in the last 5 years, then either they lack the proper engineer on
staff, or they lack the funding required to do so.   There is no other logical
explanation for base features that perform so poorly compared to the competition.

OK, yes, I know, I am done and will go crawl back under my rock now :-)

Cheers,

Neal

Does Spectrum lab run OK on Windows 10?   Is this the site;
https://www.qsl.net/dl4yhf/spectra1.html

Is there a tutorial or getting started document somewhere for interfacing the
Spectrum Lab program to Flex DAX audio stream?

Cheers,

Neal

Maybe they need a group of new Consultants to To Write new DSP algorithms for them Since they are so limited in their resources Since spreading themselves to thin these Days with other priorities that consume their Entire staff with day to day issues and fires

Neil,there really is nothing I could possibly say, To you it's very simple, black and white. With out considering the technology being used and the complexity within coding for SSDR.

I said we should be carful while comparing the Flex DSP with other radios only in the sense they work differently, and they have a much different set of problems. I understand it matters not to you what is involved, just the simple bottom line.

Flex has not really done much with the DSP since Version 1.5. And many of us have asked  Flex to get back to working on it, I would like that as well.

But if they don't? it is not a deal breaker. I could go with a radio with a better DSP performance, but that radio would have some things that work poorly as well. Choises, choises.

Maybe FRS should hire Simon and point him at the steaming pile of DSP
code :-)

Flex ANF is 90% unusable.  ANF on every other radio I have works great, I have many.

If my memory serves me well flex Power SDR had a decent ANF...….

I'll upvote that post!

Strange. My WNB only causes my entire spectrum to jump up and down (like AGC pumping) and yes it happens at all different signal levels and gain/threshold settings. Completely useless on my 6700 and ANF is not selectable on AM where I need it most. Odd that some see benefit with these functions while many see only audio distortion (I see that with NB)... The other radios I have actually work reasonably well with noise reduction modes so not sure how this could be too location dependent.

Last night I spent about 2 hours on my 6600M.. 20,40 and 80 plus several Marine radio frequencies and SWL stations..I figure I could test out the ANF if a hetrodine was to appear..,, I ran into a situation 3 times where I felt it was a good test to turn on the ANF, and see what happens.. My expectations was very low after reading all these threads.. I dont use the ANF that often, I am more focused on the NR function.. Well, in all 3 cases the ANF completely eliminated the hetrodine, I mean 100% gone with zero impact to the quality of the audio.. So for me my testing left me 100% satisfied! I did not switch to SmartSDR to see if running remote would change anything? I would expect the results to be the same.. Still on version 2.xxxxxx.. I have had very good results from the NR, WBN and NB functions, just depends on what the noise source is.. With a ICOM IC-R8600 sitting next to the Flex, but better yet, a JRC NRD-545 that also has a Auto Tracking notch filter, i will try to make some comparison.. I know the NRD-545 notch has been very impressive in the past.. Not sure why so many folks are not having decent results from the ANF?? 

Robert

Just an FYI for those of us hoping that the upcoming v2.50 maintenance release would contain improvements to the ANF, the short answer is no. According to Tim Ellison on the FRS Facebook page, when I posed the question to him, if the ANF issues would receive some needed attention, his response was: "James Whiteway If it wasn't in 3.0.19 and the upcoming maintenance release for 3.0.19, then the answer is no. If we decide to do a 2.6, it might, but that is just speculation on my part. There is no talk at this time about doing a 2.6 " I find his answer disappointing, but at least it is a clear, and honest, answer. It gives me an idea as to what, if any, options, I might look for to address this problem for myself. I can either upgrade to v3.0.10 and hope someday, this issue, along with others, finally receive the attention they need. Or, I can look towards other manufacturers of SDR radios that already have a basic features working to their full potential. ( along with some other, useful to me features) Time will tell. James WD5GWY Oh, Bill. It appears.you were spot on, with your statement regarding the .v2.50 update. If I came across as a bit harsh in my response to you, please accept my sincere apologies

Benchmarking against the best performance in the category is a very simple and objective method. Those with Flex and ANAN in their shacks of course can A-B compare them easily. This is where the trouble starts: Those few of us who are fortunate to have both systems see a large performance deficit between the effective and excellent NR2 and whatever is in the Flex. The trouble then gets bigger letting years pass after this was clearly recognized, and doing nothing about it.

Mike,

I tend to keep the AGC-T set pretty low. I have always thought the noise reduction in SmartSDR was great. And for the most part it is. But, Warren Pratt's NR2 completely removes any background hiss. Had I not experienced the difference, I woul, and really am, satisfied with the noise reduction in SmartSDR.

But, the ANF is not good. A tuner upper tuning on top of a qso can still be heard popping thru a conversation no matter the settings used.

On the other hand, the ANF in OpenHPSDR and SDR Console knock it completely out without distorting the audio in any way. All done in software with no special filters in the radio itself.

As for an All Mode Squelch, I know several people find such a thing useful. I know FRstack has one. ( implementation suggested by me a frw years ago) And it works well. But, a Squelch should be a basic part of a modern HF radio. I have 30 year old radios with all mode Squelch and working Notch Filters.

Someone said earlier, ( which caused me to respond to this thread) that maybe in V4 of SmartSDR we would gat an all mode Squelch and a working ANF. But, I won't be purchasing a V4 to get basic WORKING features that should have been in SmartSDR long ago.

I love the 6000 series hardware. But, in my opinion, the software desperately needs attention. And I have pretty well come to the conclusion that basic features will never get the attention they need.

Lucky for me I found a good deal some time ago on a Timewave 599zx outboard filter. The ANF is absolutely great.

James

WD5GWY

For me, the noise reduction on the Flex is hit or miss. Under some conditions it is absolutely great but it does take futzing to get it right. On other occasions, it seems to do nothing no matter what settings I try. And yes, I have watched Tim's video and read the guidance multiple times.

https://helpdesk.flexradio.com/hc/en-us/articles/360029494371-How-does-the-Automatic-Gain-Control-AGC-work-in-SmartSDR

https://www.youtube.com/watch?v=-WUWuAS24uM&t=4s

I recently acquired a TX 500 and have commented elsewhere that the noise reduction on it works with just a push of the button regardless of the conditions. It is a little disconcerting.

73,

Bob, WK2Y