[hpsdr] New Hermes board

[Marc OLANIE]

Hi Helmut

Many thanks for your precisions and agreed with all of them. 

Frontend is more than lpf/bpf is exactly what I meant. Particularly if you intend to address both ADC independently (impossible without a fork of Alex SPI "protocol". At least two assignation bits are missing (amongst other things). A good frontend could quickly turns into an incredible route-switching nightmare. 
I'm convinced that modular architecture is a necessity, I was just insisting on the fact that this modularity should also concern the first power amp. It was not very clear for me at first glance, it is now with your precisions. 
Concerning the "market positioning" of the Hermes V2.0, I've been lead into error by the targeted price of the main board. 300 bucks is approximately the price of the FPGA used on Angelia alone. That doesn't leave much for 2 16 bits ADC, 1DAC, memory, Ethernet output and some other guts. But you reassured me.

I should precise that I'm definitely convinced of the importance of diversity, even if the exploitation conditions are relatively hard to please. It's just a matter of weighting.  So far, the second ADC has 3 main ways to be used : second RX without any kind of filter , predistortion when TX, and Diversity. But when I'm looking the way most of Hermes (or alike) users around me deal with their equipment, they all use the second ADC on a specific band/antenna, with or without external filtering.
73' Marc, F6ITU





-----Message d'origine-----
De : Helmut Oeller [mailto:oeller at freenet.de] 
Envoyé : dimanche 4 juin 2017 12:27
À : Marc OLANIE <marc.olanie at decision.fr>; 'ad0es' <ad0es at ad0es.net>; hpsdr at lists.openhpsdr.org
Objet : AW: [hpsdr] New Hermes board

Hi Marc,

I really appreciate your fresh ideas concerning this thread. Although in a hurry I'll try to respond on some points:
Absolute no dissent in necessity of 'open source SW/HW'.
Frontend is more than a LPF/BPF.
Modular architecture, in particular the separation of the RF and FPGA-CPU part is most essential for more flexibility and finally for the economic success.  Apache Labs cancelled the Hermes board due to the competition of other designs and Gerd, DJ8AY a second supplier will do perhaps the same. I think it's appropriate to discuss competitive options keeping openHPSDR powerful.
In no aspect the successor of Hermes should be a degradation, so it has to implement the audio-codec and the given EER/ET control interface as well.
Diversity capability will become more and more important in the man-made-noise of urban areas.

Many thanks for Steve's, KF7O input. That information seems very helpful to define a promising design approach.

73, Helmut, DC6NY