hackrf icon indicating copy to clipboard operation
hackrf copied to clipboard

Feature Request: use all the 10 bits of the DAC

Open Paco93 opened this issue 7 years ago • 5 comments

Sorry if this is the wrong place for asking it. For certain applications 8 bits DAC is quite limiting. As the DACs in MAX5864 are10 bits wide it would be great to have an USB connection mode transferring 16 bits wide samples (accepting a decrease in the maximum usable sample rate) in addition to the defaul mode transerring 8 bits samples .  Thanks

┆Issue is synchronized with this Basecamp todo by Unito

Paco93 avatar Jan 19 '17 17:01 Paco93

This is theoretically possible, but it would require a great deal of work. In my opinion, it would provide very little benefit. It is already possible to simulate higher resolution at lower sample rates by using a high sample rate with a low bandwidth reconstruction filter (e.g. hackrf_transfer -s 20000000 -b 1750000).

Can you describe a specific example of an application that would benefit greatly from 10 bit TX resolution?

mossmann avatar Jan 27 '17 06:01 mossmann

My application is in a traffic emulator for random spread spectrum signals In this application i can generate a large number of cofrequency spread spectrum signals with the number of signals which can change with time to emulate random variabiliti of traffic

Paco93 avatar Jan 27 '17 07:01 Paco93

Sorry i closed inadvertently the message. To conclude I agree that my case is a fringe case. Feel free to close the issue if you think it is not Worth the effort and thanks for the good work you are done.

Paco93 avatar Jan 27 '17 07:01 Paco93

by using a high sample rate with a low bandwidth reconstruction filter (e.g. hackrf_transfer -s 20000000 -b 1750000 Is this trick possible in gnuradio? I also think it's great idea to have 10 bits onboard for more precise accuracy. For some modulations is not necessary to have 8 bit.

Radiomix2000 avatar Jan 27 '17 17:01 Radiomix2000

I'm happy to leave this issue open in case anyone wants to work on it, but I don't plan to spend time on it myself.

mossmann avatar Jan 27 '17 18:01 mossmann

I'm closing this issue because I no longer believe it to be theoretically possible without a complete reimplementation of our CPLD and SGPIO solutions, removing some other functionality that we rely on. A custom solution for a particular use case may be possible, but it would be external to this repository.

mossmann avatar Sep 09 '22 16:09 mossmann