Posted: 6/21/2016 5:55:22 AM EDT
|
If you haven't backed this you should. This is gearing up to be one heck of an SDR system. Want your own micro cell? Download it from the App Store, want to broadcast digital TV? Download it from the App Store. Etc etc. plus at less than $300 the hardware alone is a great deal when compared to a hackrf, bladerf, or airspy.
Only a few hours left https://www.crowdsupply.com/lime-micro/limesdr |
|
That PCI-E version is the future. This is the direction the openHPSDR community is going. The idea is to minimize the cost of the FPGA hardware and the complexity of the FPGA code as much as possible. These have been real impediments to the open source developer community. Big FPGAs are expensive. The FPGA development environments cost money and require serious machines to do simulations and builds. Simulations and builds can take a long time making iteration painful. Achieving timing closure is its own special torture. And, most important of all, there are not a lot of skilled FPGA developers in the open source community.
For an HF direct conversion SDR, like an Apache Labs or Flex radio, this approach can be taken to the limit. With sample rates under 150MHz, raw ADC and DAC samples can be streamed over the PCI-E bus to a GPU for processing. The FPGA on the SDR card need only the most rudimentary functionality associated with the PCI-E interface and general housekeeping functions. Ideally, it would be completely eliminated by using an appropriate GPP. It can be small, cheap and will probably never need a firmware update. Meanwhile, all of the processing is occurring in the GPU. A perfectly capable NVIDIA CUDA GPU costs on the order of a mere $100. And, most important of all, it can be programmed by mere mortals in some variant of C. This makes the development cycle SO much easier, quicker, and open to a massive population of potential developers. The only downside to the simple SDR + GPU approach is SWAP (size, weight and power). But for hobbyist or ham radio base station applications, I dare say most people would happily trade-off SWAP for low cost, easy to develop architectures. |
|
Quoted:
You can order it with either. Quoted:
Quoted:
A board that big should not have a male USB plug on it. Not mechanically robust at all. It should have a female USB receptacle instead and use a M-M cable as is convention with all USB devices. You can order it with either. Yeah but they shouldn't be showing the default image with the male connector! I'm sorry, but that just doesn't make a whole lot of sense to me... |
|
Quoted:
Yeah but they shouldn't be showing the default image with the male connector! I'm sorry, but that just doesn't make a whole lot of sense to me... Quoted:
Quoted:
Quoted:
A board that big should not have a male USB plug on it. Not mechanically robust at all. It should have a female USB receptacle instead and use a M-M cable as is convention with all USB devices. You can order it with either. Yeah but they shouldn't be showing the default image with the male connector! I'm sorry, but that just doesn't make a whole lot of sense to me... I kept it on mine. I won't plug it into the host but use an extension cable instead. I have plenty of SDRs that all use different connectors. Micros break easy, B plugs are hard to come by in a pinch, and like full size b mini b is getting faded out. Full size B like the SDRplay is the best connector but they are getting hard to find and my tool bag is turning into a cable bomb. I will just use an extension cable or hub like I do with all my RTLs and am happy with. To compare this to a flex or Anan is like comparing planes and helos because they both fly. This isn't a talk to your buddies ham SDR, this is a research tool that makes setting up test scnerios really easy. Just the other day they released a VNA demo for the unit. That is the cool versatility of the design. It may not be the best at any one thing but for $300 it sure does a lot of things |