Highlighting Major Software Issues with the RX888 SDR

The RX888 is a $200 software defined radio that has a 16-bit ADC and tuning range from 1 kHz up to 1.8 GHz, with a bandwidth of up to 64 MHz between 1 KHz to 64 MHz and 10 MHz between 64MHz - 1700MHz. The design is based on the RX-666 which is turn was based on Oscar Steila's (IK1XPV) BBRF103 original open source hobby design. The product is designed and manufactured in China and is sold on Aliexpress and eBay without any official company backing it. 

While on paper, the RX888 has great specs and a great price, it appears that the software driver support from the manufacturer has been extremely poor, and no one has really been able to get this SDR working in practice without it constantly throwing errors and locking up.

@Aang245 & @Ryzerth are the developers of the popular open source SDR programs 'SatDump' and 'SDR++'. They often get queries to support the RX888, but have been unable to get much working due to broken drivers and no support from the manufacturer.

They have written up a scathing several page document about all the problems they have with the RX888, in an attempt to bring awareness to the problems, and to hopefully initiate a push on the manufacturer to properly support the device. 

In the document Aang245, author of SatDump, discusses the technical problems he's been having with the library and drivers, noting that almost all of the library drive code is broken, leaving him unable to support the device in his software.

When actually attempting to use the library on any of my machines, pretty much nothing would work reliably, and even when streaming samples would work, gain control, frequency or simply loading the firmware would fail horribly.

Later, I tried instead using the ExtIO code, maintained by the original project maintainers. By then, libsddc mentioned above had been merged into that main repo. Seemed great… Until I tried to use it. To put it simply : It was bad before, but now it relied on an entire ExtIO, segfaulted seconds after trying to do anything, and of course what worked before didn’t even anymore.

A good reason for that is that when the library was “merged”, it instead was made to rely on ExtIO internals, with barely half of the functions even implemented or working...

...In summary, it just feels like the BBRF103 hobby project commercialized without any thoughts about consequences or the ecosystem, and not even usability. Same hardware, usually sold as a premium, but really just a bunch of parts hacked together.

Ryzerth, author of SDR++ then adds the following reinforcing viewpoint:

The code quality in the library was absolutely horrendous. Functions were unimplemented, stuff was hardcoded everywhere and it was just generally hacked together. Same goes for the firmware, it seems to be a barely modified “streaming” example from Cypress (the FX3 chip manufacturer)...

...I have tried multiple times to reach back to the manufacturer on twitter but they have been radio silent since April 2022...

...Something else that bothers me is that SDR seems to be popular in the SWL community. A bunch of people recommend it when the performance can only be described as mediocre. Making a wideband HF frontend is an art, and you’re not gonna get any good result from something built down to a price like it. It’s a cool ham radio project, but not something that can be marketed as a commercial SDR. I’ve seen people claim that it has superior performance to Airspy and SDRplay SDRs, which is complete bullsh*t...

...This SDR has been unlike any other SDR I’ve had to support. Other manufacturers have clean APIs, proper drivers and libraries. It usually takes me at most a day or two to support the hardware properly. Being an “aliexpress special”, I guess I shouldn’t be surprised, you get what you pay for. All the money went into the BOM and none into the R&D and software.

This entire saga highlights the fact that software defined radios are not only about the hardware specs. The support and state of the drivers from the manufacturer is key to allowing third party developers to integrate the device into their software.

The RX-888 Software Defined Radio


  1. Frank

    13 years ago someone reversed engineered RTL DVB USB sticks with zero knowledge or hardware engineering support, look where we are today I’m sure someone can start from zero with some hardware and do a with some other cheap but high spec SDR hardware like RX888, saying that SDR Play proves good support will make it a marked leader

  2. Phillip

    I have been eyeing the large number of “alternative SDR Radios” Available on Amazon.ca Alibaba

    1 If the specs of any SDR look too good to be true ” then it probably crap” I don’t buy it.

    2.when i see an SDR that has no signs of being supported with website/email/phone/driver I don’t buy it.

    3 when i spend 3 day online looking for the company name, I get frustrated, I don’t buy it.

    A very good example of this is the knockoff of SDRPlay sdr. A black silk screened board called the sun-ray (with 8 separate inputs from 0 – 2ghz using MSI based chipset that clams software compatibility with SDR plays software, for only 40$ I have to ask, is that possible?
    I recall the SDR play Rsp 1 devices achieve it stunning performance from using Texas Instruments DSP to dig out faint signal at the -180 db
    can your 40 sun ray SDR do that? , if it can I should buying it, But I dont buy without knowing what I’m getting.

    4. I saw ads for the RBX888 and the first thing i noticed was it uses direct sampling that should be a good hint the SDR radio is crap
    I spent 2 years modding EZcap 686 TV tuner dongles for Direct sampling, to know It’s crap.

    5. It’s better to support the talented designers of original SDR devices like SDR play / Hack RF/Kiwi SDr / Sdr-blogs V3 and just buy the well designed.stuff. and say no to junk.

    6. If your just curios about how the cheap stuff like the sun ray SDR can perform, you can do that too. its your choice, but i suspect at at some point the cost of going cheap will cost you more then buying a quality device in the first place. and you wont have tech support when you have problems.

    • Simon Brown

      Re: 4 – Direct sampling with an ADC is the better option. The RX888 MkII is a decent piece of hardware, we are only complaining about the low-level software support. Direct sampling is used by SDRs such as WinRadio, ELAD S2 / S3, Perseus etc.

    • Frank

      Cheap SDR Play RSP1 knockoffs are basic OEM chipset with not filtering or control firmware, need to use dip switches to select frequency range, is the OEM of hardware allowing vanilla use of the product or are the knock offs illegal ? Like the OEM sells the chips right

  3. Guy Atkins

    Bravo, Simon, for your hard work to add the RX888 to your SDR-Console. My RX888 Mkii has worked flawlessly in SDRC (as well a V1 RX888); and I flog it pretty hard for MW and HF DXing. The authors of the “scathing document” may not believe me but the MKii is many times neck-and-neck with my Perseus SDR for digging out weak signals adjacent to “flame throwers” in my local RF jungle with an active Wellbrook loop. I’ve likewise used it with great results pulling out overseas MW signals 1-3 kHz away from strong signals on coastal DXpeditions.

    The RX888 MKii is stable without problems up to its 30 MHz bandwidth, though I rarely use any sampling <2 MHz. I've used numerous SDRs since my first Flex SDR-1000 in 2005, and I don't find the RX888 MKii to have "mediocre performance". The few times I've used the receiver with HDSDR software, it's run very well also. Perseus is still my first choice (because it works with the medium wave-specific Jaguar Pro software), but I think the RX888 is a great value. I don't doubt that it's a coding nightmare to support, but as an end user I find it to be worth the cost, easily.

    Check out this interesting article for some testing done on the original receiver (not MKii): http://dc4ku.darc.de/RX888.pdf

  4. JTC

    If there’s one thing I absolutely loathe, it’s software devs that peddle a totally broken product and then offer no update or patch or support whatsoever. Thanks for posting this, I might have wasted my money on one of these pieces of garbage.

  5. Simon Brown

    Support and interface documentation from the manufacturer is essentially non-existant. To get this device working you have to do all the work yourself, I started by reading Oscar Steila’s code to understand more, then started coding from scratch. This took me at least three weeks, by contrast supported hardware often takes me just one day, two at the most to implement.

    So, if you want to support the RX888 familly be prepared for a lot of learning and coding.

Post a comment

You may use the following HTML:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>