SDS100 Prerelease Discussion

[jonwienke]

The videos show absolutely nothing about RF performance at all

Right. Including the one where you can see the received signal level shown on-screen (-109dBm). Or the side-by-side comparison of the 436 and SDS100 where the SDS100 is clearly doing a better job receiving and decoding a digital signal than the 436 in a weak signal situation. The scanners may not be connected to service monitors in the videos, but they show enough to get a reasonable idea of the SDS100's performance.

and I've seen no comparison video against a dongle

No, just comparisons with other scanners whose RF performance is fairly well-known and understood.

I bet it wouldn't come close to a $99 SDR Play RSP1A either.

Pure speculation on your part, given that Uniden has only disclosed one component in the receiver.

If it uses a tuner that is only good enough for 8 bit sampling then it doesn't really matter what you put in front of it.

Or perhaps a higher-quality ADC is one of the things that distinguishes the SDS100 from a $10 dongle.

 

[woodpecker]

So I'm pretty certain there's more than just a cheap TV dongle inside of it, same for Uniden.

Uniden stated the tuner it had:-

宏觀微電子

They are designed for use in TV tuners where the whole set top TV decoder costs around $10 - $20, the designs RF performance is limited to be *TV dongle like* by this low quality part.

These tuners are designed for the outputs to be sampled around 10 bits for DTV which isn't great for SDR and doesn't give a great artifact free SFDR.

Rafael have also had very poor part yields as well with the older cousin many only being good enough for 8 bit sampling.

These are facts, not speculation.

 

[Voyager]

This is probably a moot point,from a Uniden lover. On my wish list, a more sensitive closecall, like on a par with spectrumsweeper. This is where Whistler/gre, shined.

SpectrumSweeper operates differently. It performs a search of a 1 MHz segment which is as sensitive as regular searching. The initial hit uses a nearfield receiver to determine the presence of a hit, and it will find many weaker transmissions in that 1 MHz search as well. CloseCall actually tunes to the frequency using a nearfield receiver.

If you want a more sensitive search like SpectrumSweeper, use the SEARCH function. Closecall is as sensitive as it can be given how it works.

 

[mtindor]

Uniden stated the tuner it had:-

å®è§€å¾®é›»å*

They are designed for use in TV tuners where the whole set top TV decoder costs around $10 - $20, the designs RF performance is limited to be *TV dongle like* by this low quality part.

These tuners are designed for the outputs to be sampled around 10 bits for DTV which isn't great for SDR and doesn't give a great artifact free SFDR.

Rafael have also had very poor part yields as well with the older cousin many only being good enough for 8 bit sampling.

These are facts, not speculation.

The R820T2 is used in the Airspy R2. The Airspy R2 stomps the RSP (my opinion) and stomps DVB-T dongles (probably everyones opinion). Clearly there is more to a great receiver setup than the R820T2 all by itself. You can't make a judgement about any product just because it uses an R820T2, or an R836. As Jon is suggesting, you are leaving out large parts of the puzzle (and i suspect that is purposeful).

Time will tell how the SDS performs -- looking good so far.

mike

 

[jonwienke]

These tuners are designed for the outputs to be sampled around 10 bits for DTV which isn't great for SDR and doesn't give a great artifact free SFDR.

Given a decent preamp such that the amplification occurs before the ADC rather than after, even 8 bits of RF is overkill for 5KHz voice. It's only problematic if you're trying to amplify the signal digitally after the ADC.

Rafael have also had very poor part yields as well with the older cousin many only being good enough for 8 bit sampling.

These are facts, not speculation.

You're speculating that:

1. The production issues that affected the R820T2 also affect the R836.

2. Uniden is too stupid or incompetent to test chips before building them into a scanner and handing it to the consumer.

 

[Voyager]

Pure speculation on your part, given that Uniden has only disclosed one component in the receiver.

Uniden has a reputation of putting out industry-leading hardware that does not overload as easily as other manufacturers. I expect this will be true of the SDS100 as well. They know the shortcomings of their competition.

Unihaters don't even need one part known to trash the unit. They classify it as junk before they have even seen one. Speculation? That's another one of their core traits (and negative speculation at that). That is why their credibility is so low.

What you say is true about the videos, Jon. They ignore such things because it does not fit their narrative. They will never stop trashing Uniden and the SDS100, will always expect the worst, and will tell others how bad it will be based on zero evidence. Again, very low credibility.

That said, some of those cheap tuners have very good sensitivity. The front end is just as open as a barn door (by design - they are made to cover 6 MHz wide channels). The SDR mentioned isn't much better. Filtering will make all the difference in the world. In fact, filtering is recommended by the SDR manufacturer mentioned. Again, those facts are ignored because it does not fit their narrative.

 

[Voyager]

You're speculating that:

I would say they are hoping that, but most people know it's very unlikely.

 

[woodpecker]

The R820T2 is used in the Airspy R2. The Airspy R2 stomps the RSP (my opinion) and stomps DVB-T dongles (probably everyones opinion). Clearly there is more to a great receiver setup than the R820T2 all by itself. You can't make a judgement about any product just because it uses an R820T2, or an R836. As Jon is suggesting, you are leaving out large parts of the puzzle (and i suspect that is purposeful).

Time will tell how the SDS performs -- looking good so far.

mike

I have all these SDRs and have compared the lot, on my antenna array the Airspy is more bullet proof but is poor on sensitivity and hence not as good for weak signal DX, they are both very good but in different areas.

Airspy had to ditch their last whole reel of R820T2 chips because that was the performance limiter, Airspy and SDR Play are both budget SDRs, only a small step up from TV dongles.

 

[woodpecker]

Given a decent preamp such that the amplification occurs before the ADC rather than after, even 8 bits of RF is overkill for 5KHz voice. It's only problematic if you're trying to amplify the signal digitally after the ADC.

Even if you had an ENOB of 8 bits the SFDR would be awful, you don't seem to understand how sampling works.

 

[Voyager]

The videos show absolutely nothing about RF performance at all and I've seen no comparison video against a dongle

That's because the advantage is against traditional scanner designs. Dongles are, by their nature, IQ devices that will perform better than those traditional designs on LSM. Their performance is also highly dictated by the PC used to run them, as the software defines the receiver in SDR.

Many SDRs exist, and some of the best use the same R820T2 chip that the cheap dongles use. It's all about what is surrounding that R820T2 chip, and also about the software (or firmware in this case).

BTW, ENOB becomes less important the higher in frequency you go, and the SDS100 is not an HF receiver.

 

[woodpecker]

That's because the advantage is against traditional scanner designs. Dongles are, by their nature, IQ devices that will perform better than those traditional designs on LSM. Their performance is also highly dictated by the PC used to run them, as the software defines the receiver in SDR.

Many SDRs exist, and some of the best use the same R820T2 chip that the cheap dongles use. It's all about what is surrounding that R820T2 chip, and also about the software (or firmware in this case).

BTW, ENOB becomes less important the higher in frequency you go, and the SDS100 is not an HF receiver.

The best SDRs do not use TV tuner chips, the cheap $10 - $200 ones do.

ENOB is important at all frequencies, it determines the SFDR.

I haven't seen a single RF spec for the SDS yet.

 

[jonwienke]

Even if you had an ENOB of 8 bits the SFDR would be awful, you don't seem to understand how sampling works.

No, I've only been doing professional-level digital photography and audio recording for 20+ years, and playing with 8-bit cheapo dongles for about 4 years. If an 8-bit RF signal can be decoded into HDTV video with HD surround sound, then it surely can be parsed into a decent quality mono audio with a maximum frequency under 5KHz.

 

[jonwienke]

I haven't seen a single RF spec for the SDS yet.

So you admit that pretty much everything you've posted is unfounded speculation.

 

[mtindor]

The best SDRs do not use TV tuner chips, the cheap $10 - $200 ones do.

ENOB is important at all frequencies, it determines the SFDR.

I haven't seen a single RF spec for the SDS yet.

This is not the thread for me to use to ask you this question, but I'm begging to hear the answer so I'll ask. Please list a couple of the "best SDRs" that do not use tuner chips, that cover ~50 to 1000 mhz. I might be interested in one, at least interested enough to go read up on them. Forget about HF, as HF is not part of the SDS' coverage. I'm only interested in VHF/UHF.

And when you come up with that list of SDRs, we can then compare the features / price with that of the SDS-100. I am betting that'll end up making the SDS-100 pricing look like an awesome bargain.

Mike

 

[Voyager]

The best SDRs do not use TV tuner chips, the cheap $10 - $200 ones do.

ENOB is important at all frequencies, it determines the SFDR.

I haven't seen a single RF spec for the SDS yet.

I said SOME OF the best. And that is true.

No RF specs? As UPMan has said, the specs are constantly improving. That is also true with SDRs, too. The software/firmware is a primary component and since it's constantly being updated the specs are constantly being improved.

I would not expect specs before it's released.

I also didn't say ENOB is not important. I said it's LESS important at higher frequencies. Please stop parsing my statements and quote me accurately.

 

[woodpecker]

No, I've only been doing professional-level digital photography and audio recording for 20+ years, and playing with 8-bit cheapo dongles for about 4 years. If an 8-bit RF signal can be decoded into HDTV video with HD surround sound, then it surely can be parsed into a decent quality mono audio with a maximum frequency under 5KHz.

Well my designs sit in the Inmarsat spacecraft I/Q baseband processors and in some of the deep space transponders that have ventured a long way off this planet.

Your post just shows you don't understand digital signal processing, demodulating HDTV 64 QAM and such like with moderate strength HDTV is nothing like the same.

 

[Voyager]

I'm waiting for the criticism Uniden didn't use the same tuner chip that was used in a $XXM spacecraft.

 

[jasonhouk]

Us finding out the insides of the SDS100 are no better than a $10 TV dongle, maybe...

This screen is better than any $10 dongles! [emoji3]

Ole hell who am I kidding, better than any other handheld scanner ever!

Houk

Sent from my Moto G (4) using Tapatalk

 

[W8RMH]

Not so fast, looks like Uniden (UPMan) is requesting for NXDN Fine Tuning Help. https://forums.radioreference.com/u...xdn-fine-tuning-help-request.html#post2917359

I have no problem receiving NXDN with my TRX-1. No fine tuning here.

This scanner is still in testing.

 

[jasonhouk]

This scanner is still in testing.

I'd say it's tweaking, as we all know more data=better & more efficient decoding.

Houk

Sent from my Moto G (4) using Tapatalk