Photos of the MSi.SDR Dongle: A New SDRplay RSP1 Clone

Thank you to reader Mikael for writing in and letting us know that over on Aliexpress a new clone of the SDRplay RSP1 has been starting to show up. The sellers are terming it as the "MSi.SDR" in reference to that fact that it uses the Mirics MSi chips, which are the same chips used in all SDRplay SDRs. Currently the MSi.SDR is selling for about US$65 on Aliexpress.

In the past we've seen other SDRplay clones but this is the first version in a dongle format, and the first time that it is actually priced cheaper than the original. SDRplay and Mirics are owned by the same people, and Mirics note that they have not distributed any chips in over 5 years, so it a mystery as to why there are suddenly a lot of unauthorized Mirics chips available on the Chinese market. It is unclear if SDRplay will be able to enforce their IP rights in China, but it seems unlikely.

Mikael has provided some good photos of the PCB from the version he received which we show below. It appears to be similar to the RSP1, so it is missing the new features and filtering improvements of the newer RSP1A. When connected to a PC it shows up as an RSP1, so all compatible SDRplay software including SDRuno and HDSDR with the official SDRplay EXTIO should function with it.

Mikael writes:

I have seen this sdr on various China sites for quite some time and lately on ebay as well, I was kind of sure it was a rsp1 in a new pcb layout as it claimed it to be 10khz to 2Ghz with a 10Mhz bw and 12 bit resolution and containing an msi chipset with a 0,5ppm tcxo.

I ordered one two weeks ago from aliexpress for about 67$ incl postage and got it in the mail today, it was delivered with a micro usb cable and a magnetic telescope antenna with an about 2m coax cord.

pluged it in and watched the device manager, as I have a rsp1a and a rsp2 I already had the drivers loaded for those, and sure, it was identified as an rsp1, so I did a quick test and and started hdsdr and and selected the extio.dll for the rsp1, it started up as it should and it behaves just as a rsp1, did a quick tunearound in the 3,5 and 7Mhz band and it was working well, have not done any in depth analys yest, will examine the pcb to see if there is any kind of filtering in there.

In an upcoming post we plan to compare the various SDRplay and Airspy clones available on the market with the originals.

Notify of

Inline Feedbacks
View all comments

Potwiedzam odbiór na HDSDR po skopiowaniu SDRPlay-EXTIO Ciekawi mnie jak się ma do odbioru HF 1.8 MHz- 29.999 KHz


The usual china bashing and knockoff lament. This device has a complete different circuit board layout, has a metal case (the claimed ‘original’ does not have) , just using the same MSI chip series, therefore compatible with the SDRPlay. In the beginning every non-IBM PC was called a knockoff, because it used an Intel CPU. Did you hear that recently?

Jay Bree

The only thing the Chinese won’t steal and copy is first world business ethics.


I guessed as much – thanks for reply

Wong Lee

Sometimes the hard work and self-innovation in an exiting domain acts like acid on the skin. 🙂


@ Peter1066 – those are pure cosmetics…

Look here:


Is the magnetic ring sold with some MSI.SDR units worth getting?


I bought one of these off Ebay after owning an RTL SDR for about 4 months. It was $55 shipped. I figured if it was junk as some here suggested I could just send it back thanks to eBay’s generous return policy.

Turns out it’s not junk, and is functionally identical to an Sdrplay RSP1 with the metal case upgrade. I can hear WWVB at 60 kilohertz just fine, I get the full 10 MHz of bandwidth, and it’s relatively free of spurious signals. Signals that are difficult to hear on the RTL dongle come in better on the rsp1 clone (due to it’s 12 bit resolution). For the price I couldn’t be happier with it and it feels like a solid upgrade from the RTL dongle.

Craig Cavallaro

Update I am super happy with mine. I made a loop block or ground loop eliminator for my PC. Let me tell you for under $2 in parts it dropped the DB floor so far down and got rid of a lot of hum on all my SDR. It is just two diodes in a 1K resistor underground wire of your power cord. I started building a few of them with a box in the middle of the cord and the two diodes in the resistor all done up but heat shrink it came out very good I’m thinking about selling some of them. Now I have been able to find in medium wave but there’s not much of this way so I get one station down that low. But as proof of concept it goes all the way from 1 kilohertz up and my testing. A way better than using an up converter a lot less noise.


Did you use a circuit like this, and using a resistor instead of the shown C ?


Yes two diodes I made a bunch of cords with a box about 3/4 of the way down and did it to the ground wire. Works very well for my piecing SDR power supply and my USB power station. 100% eliminated my ground loop. and don’t get me wrong I have some major ground thing on my house I just don’t know why I had such a background loop. I have really bad power coming into the house and now I have zero AC hum.

Ty Bower

Prices on these MSI.SDR are falling all the time. Search AliExpress… there are some sellers asking less than twenty USD for a unit. I may take a risk and order one just to find out if they are real.


Î am using one, go ahead and try it. I am happy with mine.

Craig Cavallaro

Been testing mine I have been able to pickup Canada time @ 3.330 mhz with no trouble at all. I use HDSDR for of my stuff. The unit works great I like how small it is to plug right into an usb hub. Very happy with it.


Received mine today. Works, sensivity similar to the FCDPro+. Will check in detail next days.
@Scooby Do: the internals look differen than the ones shown at your link


Finished my comparison with my FuncubeDonglePro+ and I could not find any difference. Maybe under strong signal conditions one will see a difference, but I had no such conditions. To call it a ‘clone’ is a little bit far-fetched. The chips used in all those designs, FcDPro+, RSP1 and now in the ‘MSI.SDR’ are the MSI1001 and MSI2500. All of them using their individual unique circuit design around the MSI chips. Option was, either to struggle with an additional interface and ExtIO for the MSI.-SDR – or to jump on the existing protocoll of the RSP1 interfacing. Good for us they used this kind of solution. The shielded housing of the MSI.SDR is definitely better than any of the others with their plastic around the boards. The only problem I encoutered was with SDR# 1361 and using the RSP1 ExtIO. Scanning and frequency settings from memories did not work. HDSDR, SDR-Console 2 and 3, SDRuno ExtIO and SDRuno 1.3, all of them worked with the MSI.SDR flawless. All in all, the MSI.SDR is a inexpensive, well built nice toy, with a surprising functionality.


The metal enclosure sure seems to be upgrade compared to the crappiest possible plastic SDRplay use, that’s for sure, plus it’s more compact.


It might be even better if the metal enclosure was in any way electrically connected to the ground plane of the board (it’s not)

Scooby Do

ahh so the plot thickens like pea soup…word from the grapevine is when this was disco’d roughly ~7500 units were packed away and then sent out for disposal (bad idea, they should have burnt them at the local mortuary!! themselves!) so now those are in the wild and this is what were left with. I don’t think we’ll be seeing many of these after this initial burst of sales….and also no support obv.


I take it a cloned Airspy doesn’t come with the developers lousy attitude that ships with the original version?


I see I am not the only one who has had poor experiences with the company. Their stuff is interesting but there is no way I would buy another product from Airspy.


How did airspy get in the this , I thought the comments were about the sdrplay clone ???????????????????


I see that people have questions about the clone dongle . The question that I have is it any better than the SDRplay RSP1 ?????????? Normal clones never live to the original .


As far as I remember the original RSP1 (which is no longer sold) had a +/-10 ppm crystal (which never appear in any specs for the device). A modern 0.5 ppm TCXO, which have dropped dramatically in price since the first RSP1 receivers were sold, is going to win. So it would be a biased comparison comparing old hardware to new.

But on the flip side when buying really cheap knockoffs you are basically guaranteed that any minerals used are going to be from the cheapest vendors, which typically means conflict resources (natural resources extracted and sold to perpetuate the killing over the resources in that area) will be used. e.g. tantalum capacitors made from Coltan bought from the conflict region in the Democratic Republic of Congo, also Cassiterite for tin used in solder. Reputable supply chains have auditing and reporting requirements for the resources that they use.

Bendail Vam

ya um…that would be “first world” countries giving a schitt….china is a 3rd world country except for the 5% of the commie party that enjoy the spoils…they don’t care about where something came from conflict or not which btw, is about what 88% of the rest of the globe does also. ChiCOM doesn’t innovate, they steal/replicate and saturate. Just because something has a first world price tag doesn’t mean china has to follow any pricing rules…they have no rules except if you can make money off of it – do it.


From a article at

If you are interested you can look at the MSi2500 and MSi001 datasheets. In addition to the RSP’s, the Mirics chipset is used in the Funcube+ dongle and recently in the CommRadio CR-1a receiver.
So how exclusive is it?


Since we are all speculating, wonder if perhaps a second-party had some excess stock that they sold off (either qc rejects or just bought too much).


This could be chips that have been scrapped because of quality issues. This happens here and there in China, that chips are taken out of the garbage bin, tested again elsewhere (at room temperature and nominal voltage) and used in cheap copies of brand products.

You may get lucky, but you also may get a device that fails very quickly, or has problems at high or low temperatures.


Or QC issues created to scrap the chips.
There is a third possible explanation; it is not rare that manufacturers allow privately and deny pubicly an agreement to make copies, see Baofeng… after all this is an old design that competes with RTL+Upconverter more than with the RSP1A.


So instead signing a NDA and buying 10k of really cheap chips from Mirics (where SDRplay buys it chips), they would use garbage bin chips – that makes total sense to me.

Bendail Vam

you have to remember 30% of the chips mass produced today are identified low QC….of those 15% are still “usable” to certain spec. that’s why very seldom, if never, when an order is agreed upon for say a 20k unit run (and that’s a low quantity order) will you see less then 50k physical units produced. This is why big name chips players have in-house auditors on site at the fab to make sure the overruns get sent or destroyed per order.

Jay Bree

I can’t see how anyone can make anything with 30% fallout. That’s beyond terrible since about 1981.


I do not see how SDRplay can enforce their IP rights if the manufacturer used an original chip.


Yep, the system is floored. China is notorious for ignoring the intellectual rights of those outside of its borders…but who can stop them? Pretty soon the tail will start wagging the dog. Essentially, this the activity of thieves.