Apache Labs ANAN-200D a competitor?

Walt - KZ1F

While I do agree Flex has some very smart people, they don't corner the market on very smart people. Their size presents challenges. That is one of the reasons I expect to see the 7000 sooner rather than later.

np2g

Hope the 7000 will be fully functional. And like all computers a quantum leap from one to another. And yes. There is a very large case of the smarts out there just waiting to be taped

Bill -VA3WTB

I have not any information about a Flex 7000, were can I find it.

Bill -VA3WTB

Walt, are you working on some software,,do I have that right?

W9OY

What a bizarre comment np2g.

How can I be 90% incorrect?  THIS IS MY EXPERIENCE DADDYO  I have both radios sitting on my desk, I follow all the tech audio feeds and the yahoo groups  I never turn on my ANAN anymore because it's just not a compelling radio to me.  All I have to do is press the ON button and I can't even be bothered to do that.  One more thing the ANAN 100D's fan is insanely noisy

Bill -VA3WTB

Lee, as I read your post I was thinking, what are we really saying here. To me this is not really about Flex verses Anan. There is room for both. Each comes from different directions and both are very good radios. The last thing I would want to see is any Anan bashing here, I think it is clear why many of us buy Flex and with very good reasons. But the Anan is fun too.It has a lot of parts that can be replaced with experimental boards, that's what they like. The flex stands on it's own merits very well, lets enjoy both.

Walt - KZ1F

It's OK Barry. I've done that too, most embarrassing (it was for me). I don't really follow the HPSDR subject. Personally, from a user perspective I think PSDR made more sense. However, from a product life cycle standpoint, the radio is in the box makes FAR more sense which is likely why FRS did it. This is why there are marketing departments, to sell the package.

Bill -VA3WTB

Walt did you see my question to you?

Walt - KZ1F

Oh Bill, have you not been following me for the last year? I am hurt. This is why I don't have a facebook page, what would I do if nobody showed up? Quick aside, the "don't like IBM?" comment is, in software, Blue == IBM. Well, for those that go back that far.
Yes, Bill, I am. I've been in development and architecture for 43 years. Now my sights are full time on this project. After I figure out how to do a YouTube video, I will introduce it. I was hoping to do that next month in Dayton but it won't be ready in time. Actually, I shouldn't say full time, I did, after all, buy the radio to actually transmit on.

np2g

Ok . Try to be professional on the analysis. A. very poor comparison. As the 6300 is missing most of the architecture that makes the 6500 or 6700 better Both of these will make a comparison. And indeed Grow to a fantastic radio Please remember. It is not a 1500. Or the 3000. Or the 5000 and not the 6300 that has the gusto. Clarification : I tried to say that the hpsdr was more mature. Translated it means everything works. What is there now and what you have to look forward too is the defining issue with what,s better .

Bill -VA3WTB

I am just trying to understand this is all, sounds interesting, I would like to know more some time

Walt - KZ1F

I gotta ask though Bill, why now? What prompted you to ask?

Bill -VA3WTB

I need to talk off here Walt, it will make sense to you

Walt - KZ1F

kz1f
at
arrl
dot
net

W9OY

Bill, I have both.  In fact I have 2 ANAN's the 100D and a 10.  I like them as clearly I have $1500 more invested in ANAN than I do Flex.  I've been running SDR since the 3 board SDR-1000 days.  It should be clear that ANAN/HPSDR is actually an offshoot of Flex, the SDR-1000 version.  I was there for that as well, and bought and built all the boards from TAPR and got that radio going as well.  From that you should discern I have considerable experience in both camps.  If you read the second line it states in bold both are strong radios, and they are.  My point is ANAN sells their radio as if it's a system and it is NOT a system.  It is a piece of hardware unsupported by them from a software point of view, and after all this is a software defined radio so perhaps the software aspect is important.  This is one of the "merits" of the radio which is hardly ever covered.  Another merit is the fact that if a developer decides, he can simply head down a new path leaving your hardware high and dry.  This is precisely what is going on with their "new architecture"   Phil decided he doesn't like coding FPGA and very few interested hams are competent FPGA coders so he wanted to get back to a platform that uses some variant of C.  Using the format of a computer which does all the processing connected to a "radio" modem over a ethernet cable, which is the "new architecture" virtually takes us back to the SDR-1000 topology from more than ten years ago.  This was all well covered in a coupe of team speak sessions

His calculation was he could just squeeze adequate bandwidth to run the radio on a GB ethernet line and he proceeded to perform a proof of concept which did pass the concept.  Sounds rosy doesn't it except what happens when the ADC strength goes up or some feature needs to be added?  Do you now need 1.5 gb of ethernet?  How are you going to get that on a single board computer?  How are you going to get that out of a ANAN radio?  It may all work out but I can easily see a total abandonment of the present platform just based on moores law.  Who's going to write the software for this brave new world of SDR?  I'm a tester for Flex, and the complexity of the software, interactions and bugs soars dramatically as the number of processes goes up.  When you start claiming 14 receiver capability (7 per ADC) and you haven't even written software for more than 2 you are basically lying to people or at best pretending when you claim all this horsepower.  So this is the true state of the HPSDR universe not the imagined unicorn universe of someday np2g likes to push.  When you buy something from ANAN you own a piece of hardware.  If you can write the code great, otherwise if it doesn't exist, pound sand.

That being said PowerSDR_mrx is a very mature piece of software.  It's a much modified version of Flex's PowerSDR.  It is claimed virtually none of the code in PowerSDR_mrx is the same as the PowerSDR from which it sprang.  This may be true but the radio basically behaves exactly like PowerDR and I can't really tell much difference in practical performance compared to my F5K.  Maybe some differences around the edges.  I can buy a F5K for like 1500 on swap QTH.  

I had network trouble setting up my ANAN 10.  The issue was with the subnet wired in the radio.   I needed help so I went to the Yahoo group and was led down 15 dead ends by the "experts" and then summarily dismissed from the group because my "helpers" got **** their suggestions were not working.  The real problem was their suggestions were ineffective solutions and they failed to understand what was going on, but they felt if they could just yell louder somehow the ineffectiveness would resolve.  One engineer took pity on me and between us we solved the problem in 10 minutes after 2 weeks of being yelled at and then being banned on the "yahoo group".  So much for support.  I also had audio issues with the 100D which I was basically left to solve on my own, which I did.  There is no Dudley WA5QPZ in ANANland. The yahoo group is nothing but a dispenser of free, often wrong advice often worth exactly what you pay for it.    It leads you down the wrong path where you spend hours and hours doing things that are not even close to a solution.  In addition any addition to PowerSDR_mrx I inquired about with the authors of PSDR_mrx, like restoration of a previous macro that worked in PowerSDR was blown off 

So yes by all means let us discuss the merits, ALL THE MERITS.  If we need a list:

1 unstable platform support into the future
2 non existent professional support
3 often destructive amateur support
4 pure signal that requires a lot of screwing around to keep it pure
5 No interest in implementing user requests for software changes

I listen to the ANAN freq and those guys are always disappearing to go tune up the pure signal.  When it works it is astounding in what it brings to SSB legibility but at present it's still a kludge IMHO

You will note the title of this thread.  Is the 200 a competitor?  Unknown by me.  I have worked a ton of DX on both the 100D and 6300 so they are definitely competitive and IMHO either is a better choice than legacy radios so from that understand if you are moving from something like a K3 or a Yaesu I think you will be pleased but if I'm going to buy another SDR I'm buying a 6700 or what ever follows  

Bill -VA3WTB

Thank you Lee, well stated....

Walt - KZ1F

Lee, tell us what you really think. As for your last paragraph, I believe if you look at shereng.com the K3 is neck and neck with the 6700. What difference does it make what the generation is when, in the lab, they appear virtually identical? I am not advocating purchasing a K3. It's just you and others appear more interesting in the technology used than the performance of it. I do not own a K3, but if I did the reason I would like it is because I plug in the KAT500, I plug in the KPA-500, and I plug in the P-3. Everything works. As I change frequencies on the K3, the linear, tuner, and panadapter know exactly what I just did. That is user friendly. You can't say that about the 6000 series...at all. So if the K-3 works just as well and is truly plug n play, I would be careful about bashing it as "legacy technology".

Not everything sold is consumer grade. I've owned a brand new C-182S and a brand new C-172ME. If I didn't know how to fly, they would have been REALLY expensive bookends. As for your comment about Kenwood, Yaesu, Icom, and Elecraft...they haven't gone out of business yet. My guess is, they won't.  The USB-2 is 48Mbs, compared to 1Gbs. There is over a 20x speed differential between pushing data across a NIC and pushing it across a USB as the 1500, 3000, and 5000 did. I think you are trying too hard to push a position. Me thinks people need to be less emotionally wrapped around their radio decision.

Bill -VA3WTB

It's ok I see what Lee's point really was, it had more to do about moving from a legacy radio to SDR, even as many legacy radio's still perform well.

np2g

Which is better . It really doesn't matter. I just got tired of waiting. So in the next 2 years or so ill just suffer

David Decoons, wo2x

But the receiver passband doesn't line up with my timezone.

np2g

But it will display gray line

Barry N1EU

I know this thread has just about run its course.  But I just received a rather full explanation of the new OpenHPSDR/ANAN DFC architecture from VK6PH that I wanted to share here. 

73, Barry N1EU

DFC (Direct Fourier Conversion)  
DFC is intended to extent the **** pipe architecture to the extreme  (‘like drinking from a Water Main!’) in that the entire SDR hardware to PC interface  bandwidth is used. 
 
What is holding back more open source SDR development is the specialised skills necessary to program an FPGA.  If they can eliminate the DSP processing in the FPGA by moving all the DSP  to a Single Board Computer (SBC) or PC then the number of potential programmers who can contribute increases significantly.
 
DFC  effectively directly connects the ADC and DAC to the SBC/PC and all processing is done on the SBC/PC.  The ADC data is immediately applied to a Fast Fourier Transform (FFT) that processes the entire HF spectrum in real time.   Individual receives are generated by selecting the appropriate range of FFT bins.  The number of receivers implemented, and their features, is then solely limited by the SBC/PC processing power. 
 
This architecture is not new – just an alternative approach to the DDC/DUC that virtually every ADC based SDR is currently using, This architecture has been possible since the 1980’s – if you had a Cray supercomputer!  Well today we have the equivalent in GPUs on a low cost Graphics card.
 
With the DFC approach a SBC will have to be 100% dedicated to the SDR task – you will not be running any other code on the board.   However, the required processing power is available via a dedicated, low cost, SBC and in the current development they are using a  sub $200 Nvidia Jetson board.
 
As cheaper/faster SBCs become available these can be upgraded  relatively inexpensively.  The Jetson board, running Linux,  initially simply emulates the openHPSDR protocol so that PowerSDR etc software (running on an Windows PC connected via Ethernet to the Jetson board) can be used directly for development and testing.  This will be extended in the future to become a full server so that a ‘thin pipe’ interface to a tablet/cell phone/PC can be used.

WA2SQQ

Barry this is all very interesting, but in the end the progress of the product is only as good as the team behind it. I enjoy my 6500 as it serves the purpose I got it for - to talk with fellow hams. Just as I drive an auto for transportation, whether it's a Toyota or a Corvette, in the end my **** will get there. We are fortunate enough to be able to buy anything we want.i agree that this debate has run its course so let's get on the air and enjoy our hobby!

np2g

In plain language. **** or thin it doesn't matter. The cud aboard is a pretty neat approach There are a load of smart guys out there not on flex pay. And we are the beneficiaries of all the developments.

Steve-N5AC

N2PG: As the 6300 is missing most of the architecture that makes the 6500 or 6700 better 

As tempting as it is, I'm going to stay mostly out of the larger discussion, but this statement is simply incorrect.  The architecture of the 6300 is identical to the 6500 and 6700.

Steve-N5AC

N2PG: As the 6300 is missing most of the architecture that makes the 6500 or 6700 better 

As tempting as it is, I'm going to stay mostly out of the larger discussion, but this statement is simply incorrect.  The architecture of the 6300 is identical to the 6500 and 6700.

np2g

I might be confused i guess diversity receive In the 6700 is also available in the 6300 and 6500 .



I know one of your stumbling blocks on pre distortion was the manual attenuation function .
This objection has been overcome by the open source guys .

The adaptive pre distortion now has auto attenuation function .
Hands free. Yep.

Joe, KQ1Q

It appears a major redesign of the Anan was needed to support thin pipe clients.  VK6PH himself calls it a "new SDR architecture", which has currently only reached the proof-of-concept stage.

As part of that redesign, using the Tegra K1 SBC was a good idea since it's very fast and power-efficient. The nVidia CPU it's based on  (code-named Denver) has good CPU and GPU performance at a low cost. Potentially the SBC could be upgraded.

In essence the **** pipe is moved to the radio server and functionality is moved off the FPGA to the Tegra K1 SBC which has an easier software development environment. From a high-level viewpoint, they are moving toward the overall solution FRS decided on years ago, just using a different implementation.

In yet future versions beyond the current proof of concept, this theoretically could allow a thin pipe client if the SBC only forwarded a lower-bandwidth more GUI-oriented data stream. That would also require development of the software protocols for thin pipe communication, unless that's already been done and I missed it.

Whether the actual production Tegra K1 SBC would exist inside a future Anan radio or be housed outside in a separate chassis (possibly for retrofitting current radios) isn't clearly stated.

EA4GLI

I find the information provided in this thread very interesting with the exception of some odd entries and some out of place comments. I really don't find information about these SDR platforms on regular ham radio magazines as pertinent and insightful as some of the things I have read here. I hope the thread remains open a bit longer.

Bill -VA3WTB

It goes to show us Flex guys sure love learning, not only about Flex but other things going on with other radios as well. Most of us here are interested in what Anan is doing for interest sake and I think most wish them good luck in bringing that radio up to the 3rd generation SDR. As Howard said. Long Live SDR