cuban8 wrote:Sorry, but I just don't understand the hysteria surrounding the "Spektrum Problem". Yes I fully understand and accept the findings regarding the system possibly selecting channels that are relatively close, but just because it's been decreed that this might prove fatal if other random outside influences occur at the same time, doesn't mean the system's no good.
You are absolutely correct.
As I've said in my articles on the matter
The sky is not falling.
However, it's not an insignificant issue either.
That I was able to completely disrupt the DSM2 link using something as simple as a video-sender and that this was replicated by a colleague using just a 50mW video sender is somewhat worrying. These video senders are sold in large numbers and some people are using them on RC models for FPV purposes which means the scenario where a DSM2 system selects 2 closely-spaced channels and then someone turns on a video sender that also occupies the same part of the band is a small but definite possibility.
What is of concern to me (and clearly other people) is not that this vulnerability exists in the DSM2 system, but that Horizon/Spektrum won't acknowledge it and fix it. All it would take is a tiny change to the code in the DSM2 transmitter -- something that could be done and tested in a day.
The only reason I can think of is that they don't want huge numbers of existing DSM2 users returning their equipment for an update to the new code -- so they'd rather not acknowledge the problem. Perhaps the new stuff is already fixed but they're not going to say anything for fear of that wave of "please update my old gear" requests, and the costs that would involve.
Yes, Spektrum's S/W designers have potentially left their product open to possible failure if a very unlikely set of circumstances occur. Too many ifs, buts, maybes, and random effects to really cause owners to lose confidence in their equipment IMHO.
The thing is that if Spektrum was a cheap Chinese system that cost just a few bucks I'd agree. However, Spektrum is pitched as a "premium" product and a leader in the field of 2.4GHz RC, with a matching price.
Why should people have to pay top-dollar for a bottom-shelf product?
It is the goal of RCModelReviews to help people make *informed* decisions when purchasing new stuff. They need to know that DSM2 is past its best-by date (for a number of reasons) and that it makes no sense to buy an RF system which is outdated and suffers performance penalties when, for the same or even less money, they can buy much more advanced and resilient systems.
One only has to look at a DSM2 installation in a large model to see how far behind the 8-ball Spektrum is now. Spektrum/JR advise you should use at least 3 satellite receivers and a data-logger to ensure reliable operation in such models and to keep an eye out for potential issues. By comparison, there are a huge number of FASST and other FHSS installations that work perfectly -- with just a single receiver and two antennas. Technology has advanced significantly since Spektrum designed their system but they are steadfastly sticking with old designs and (it would appear) buggy software.
What became very apparent in my tests is that there are now only two brands of 2.4GHz RC systems that select and stick with only two DSSS frequencies. These are DSM2 and the Chinese Assan (one of the first Chinese 2.4 products).
Every other manufacturer has now either gone to constantly-agile (FHSS) technology or opted to use at least three frequencies for redundancy. JR's launch of DMSS is another clear indicator that they don't wish to have their brand associated with the now aging and (in some cases) inadequate DSM2 technology.
How many wide band 2.4 gig transmissions (video transmitters?) would it take to swamp the whole band and render a FH system useless as well?
Well a video sender tends to consume about 1/4-1/5 of the entire band so in theory, as few as 4 video senders operating on different parts of the band would cover it. However, there are peaks and troughs in the video-sender's signal so even then, *some* of the FHSS system's signal would get through. In my testing, it's enough to provide a measure of control that would allow you to get your plane down in one piece. FHSS systems tend to degrade more gradually as you fill up the band with noise. DSSS systems just stop dead.
I don't know whether it's 4 or 6 or 8 or however many. Presumeably it could happen, so why don't we start worrying about that?
It'd only take 4-5 to cover the band -- but a good FHSS signal will still get through adequately to provide a modicum of control. And what's more, the chances of this happening are clearly a lot less than the chances of a single video transmitter being turned on while someone's flying their DSM2 system.
In summary: DSM2 still works for most people but it is now very much a "second tier" system -- so, why would anyone buy a second-tier system for a top-tier price when there are a raft of better options available that provide more insurance against being shot down by interference or suffering a receiver brownout?
The sky isn't falling -- but it's getting awfully cloudy in the Spektrum camp.
Posted: Mon Sep 06, 2010 1:46 am