Hey Chris, Thank you for raising questions. We do it to adjust the power. The data is not fake. V2 transmission in more than 500mw will only enhance the ground end of the transmit power.
"V2 transmission in more than 500mw will only enhance the ground end of the transmit power", you mean that it's useless to have 1200 mW ???? And what about the the v1 ? Why so little power, dji has more i guess...
@@jadoofpv3462 I've gone 29km with the v1 on '1200mw' using cheap 9.4db triple feed patches. Range was increased using the 1200 setting compared to 700.
@@tmc-fpv That's the V1 though - which from Chris' data does output more. Have you done any range tests with the V2? What's interesting is on Chris' data, it seems like even on the 1200 mW setting the system only outputs ~350mW, but that's still higher than ~220 mW on the 700 mW setting...
@@Vousie Ive only been using the v1 and 1s vtx. other fixed wing buddies are finding that the range of the v2 is around 15% less than the v1, which they put down to the single antenna. Regardless of the setting or numbers within the menus it appears that there is ample power.
Hi Chris, I think I might be unto to one possible explanation for your unusual V2 power output results. I was looking JB's Avatar V2 tests and noticed that at 700mW he never got above a 14mbps bitrate, but once he went to 1200mW he was able to get 25mbps. Then I remembered that OFDM (the digital modulation scheme used by these systems) consists of multiple sub-carriers each modulated with some form of QAM-n sub-modulation. To achieve higher data bitrates, the sub-carrier need to use higher QAM modulations e.g. switching from QAM-8 to QAM-16 would double the raw channel bitrate of a sub-carrier, however this requires a higher SNR to properly discern the data symbols of the QAM constellation so it's normal to boost the power of the sub-carrier when changing to schemes with a higher bit-per-symbol encoding. That got me thinking, since your tests were all indoor and stationary, the video encoded by the VTX would have been a static image and had the lowest possible bitrate (due to MPEG temporal compression), so the system would not need to increase the bitrate above the 14mbps JB recorded, hence no measurable increase in output power. I would suggest you repeat your tests with the camera pointing at an LCD monitor showing white noise e.g. from an analog VRX noise on an unused channel or a DVR recording of that. This would force the system to produce higher bit rate video and I would expect you'd then see higher output powers when set to 1200mW.
You can also observe the same behaviour through the debug overlay of the goggles by setting the screen size to 10%. When you raise the power above 500mw, only the goggles' transmit power increases but the vtx stays the same like... 500mw => Sky: 26dBm, Ground: 26dBm 700mw => Sky: 26dBm, Ground: 27dBm 1000mw => Sky: 26dBm, Ground: 28dBm 1200mw => Sky: 26dBm, Ground: 29dBm
Very bizzarre behavior indeed! Thanks for showing this! Would really like to see similar testing on the other systems, such as HDZero for example and maybe the O3 Air unit.
I think Walksnail has been lying to us all along. And the VTX isn’t actually capable of the output power they’ve been advertising. I think originally they scaled the output to appear as if each milliwatt setting was appropriate. But with this latest revision they either forgot to apply the scaling, or just couldn’t be bothered to even try anymore. We’ve known from early on that basically anything above 700mw resulted in signal to noise ratio that essentially rendered it useless. So I wouldn’t be surprised if they just lowered output to try to stay out of this noise floor.
WoW, so interesting. I'm guessing that Walksnail will be putting out an update for this, i'm pretty sure they know about this in house and released the V2's anyway, just like when the Walksnail goggles came out, they were still working on some stuff. Awesome find Chris.
I had wondered how they were supposedly getting 1200mW out of the SKY85743 amplifiers in the vtx, which only specify 24dBm (250mW) output power in their datasheets. I'm curious about the 1S vtx too, and whether the maximum output changes when the battery voltage decreases. Power consumption does change as measured by painless360, there may be a tradeoff. Cheers!
Since you can attenuate the signal and increase latency it would be cool to see your latency test compared to the advertised numbers at various signal strengths.
There's a post on the official Walksnail FB channel of this video. I can’t post the link, comment gets deleted here. In a comment on the post, Caddx states that Chris' data is correct - the V2 VTX is limited to 500mw output, all higher settings only increase output power of the TX on the goggles/VRX.
My understanding is that analog is measured as an average and digital is measured as a peak... I got that information from the ImmersionRC power meter. You seem to be using a different method ? Did I miss a video you go deep in RF power measurement ? Good luck on your battery testing quest... I hope it won't be too time consuming as I love your videos and don't want less of them :) CNHL has been my 'go to' brand for 3 years now because it's half the price of premium packs like r-line and about 90% the same performance. Impatient to see your results !
Very interesting findings! Nice of Caddx to answer swiftly also. I have to ask though, what is the thought behind "enhance the ground end of transmit power" (the Goggles, i presume)? Range is a VTX thing, isn't it?? Double the power on the goggles does not increase the range, or does it? 🤔 I use the Avatar HD Pro and Goggles X so this would be nice to learn.
I think this is done to not give the impression that the system has lost power (it effectively gained power) while trying to match the actual power to what the setting is called. So I would guess they did the "equivalent to analog" thing in the beginning, but realized it was a mistake later on and wanted to fix it without giving the impression that the system lost power (it would probably sell worse and it wouldn't be true). Still would have been bettr to openly communicate that they did this.
Hi Chris! I think this is a feature/simplification. On the lower power levels, they are bringing the numbers in line with what you see, as well as having users notice a big upgrade on penetration/range. As most casual users will not use the highest output settings, they may have capped the power. Hear me out with potential reasons. 1) newer hardware, loathe to have many RMA due to bench burnouts. 2) from electronics packed tight, low airflow, refer above. 3) if things seem to stay stable, and only a few fires here & there, they can unleash more with fw updates for LR records. 4) they could just cap it for stability and reliability. They might want this iteration of the vtx to have a longer service arc and want to build rep for those reasons. Cruising vs racing won't blow the engine. 5) they want to stay under the radar, so to speak, even if the stated power is above legal limits in the US, they can claim it was capped the whole time, if investigated? Last reason is reaching, and I don't know enough about the nuances of liability here in the US.
Really nicely thought out comment, thank you! There is some tolerance allowed on power output (DJI have +/- 2.5dBm!) so the V2 settings would be within that accuracy from a power output standpoint. V1 seems more like "equivalent to analog" or a bit on the low side if you take the combined output power of both antennas.
RushFPV max Solo is also not every piece can deliver 2.5 W, as the chip inside has nominal output power 26 dBm (0.4 W), and in overdrive modes sometimes delivers 1.5 W, sometimes 2.1 W. Tested on a few VTXs
Did you look at the amount of power drawn by the system ? Is the power source able to deliver enough ? The walksnail system being bidirectional, I wonder if obstructing the signal to measure its power level does not modify the whole system behavior. (measuring the power drawn would also show the efficiency of the walksnail solution and similar measures could be done on DJI & HDZERO)
In practice I've gotten really good range on a long range wing with the v2 vtx. Penetration has been ok, better than analog. But miles out still great link.
@Caddx WTH!?? Please explain this? I wondered why the range was horrible on the V2 system, I just thought it had to do with the single antenna. But if DJI can do a single antenna with the Airunit lite/vista with no issues then Walksnail should be able as well🤷🏻
i am sure you made a video allready about that but. I have a 6s freestyle/cinematic quad. but i bought 2080kv motors. i have put a 80% cap in betaflight. Is there some disadvantage? performance issues?
Is it possible that the walksnail system is limiting the power output based on the distance from the goggles? Maybe if you get some distance between the Rx and Tx that it might output more power?
There have been several firmware updates since this video was published. 33.39.10 - > 37.42.5 Would it make sense to repeat this test and see if anything has changed?
I see from the datasheet that the AD5902 has a 2K input impedance which is much higher than a typical omni antenna (< 50 ohms). Could it be that the V2 VTX is able to detect the high WSVR such an impedance mismatch would generate and power limits the RF amp e.g. to protect against a damaged antenna? We all know reflected energy can damage the transmitter, and more so with the V2 VTX since it has a single antenna outputting all the power. FWIW Analog Devices already incorporates this functionality into some of its amps. Search for "Log amps and directional couplers enable VSWR detection".
Bardwell crowd funded some battery test equipment a few years ago. His channel doesn't focus on testing so he didn't use it much which is unfortunate. I doubt he would want to give it to you but, it did come to mind it would be nice if he gave you the equipment he doesn't use any more. Either way your channels focus is more about testing and I look forward to watching.
really curious about the numbers. Does anyone know if there is any real world improvements on avatar V2 at the higher power levels? Clearly there isn't more power, but I wonder if they are doing something else on these different modes? I don't know the science enough to know what that could be.
A simple explanation could be, that your DIY power meter goes into saturation somewhere around 450mW - either in the sensor or in the Arduino processing? All the V1 measurements were below this level. I fully realize you may have tested your measurement setup with a higher powered source, but it is not clear what readings you obtain with say a known good 1W source. I look very much forward to seeing your battery tests - great initiative! Edit: I am now a Patron.
It doesn't saturate. I "calibrated" the setup using a 1.6W analog VTX and even that is nowhere near saturation. Saturation would be around 33dBm (2000mW)
@@ChrisRossertry to check your system on the dji module. If it behave dufferently on the high power so we have a problem with the walksnail vtx, otherwise with the power meter...
Hi chris, have you seen the system to calibrate the v2 vtx ? The one that JB published in one of his video lately using zoom 10% to check values and putting some files on. It would be extremely interesting if you could make a test looking if that calibration process change something or not . About power , it will be interesting to compare dji air unit vs caddx vista too
Speaking to Mads about it we can be confident that the calibration will change the output power, but what I can't say for certain is if it will increase range as it will also amplify noise.
@@ChrisRosser ok Mads is the master there :) if any of you two could certify that it would be interesting to run some penetration test with various values. I was thinking to check power output with immersion Rc power meter but I guess it's not reliable
Can't use an analog power meter like the ImmersionRC for digital VTXs due to the high crest factors for LTE signals and the frequency hopping. That's why I built this RMS power meter. It can measure EVERYTHING, analog, digital, HDZero, 5.8GHz, 2.4Ghz, 900MHz it doesn't care. Power is power. I believe it works by measuring the heating up of a small element.
@@ChrisRossergoogles make more power! You can test.. but make very more noise - snr decreasing 200% at 1200mw - im want make penetration tests with different calibrations values fof goggles. Maybe 500mw for goggles and vtx make good results.. and please say - can you test what frequency's WS using for downlink - same as uplink or have some offsets?
Yo Chris sent you a msg through AOS website weeks ago but no reply . Anyways i got the EVO 7 " and im looking for 3d prints mainly for DJI 03 au but with 2 antennas and GPS mount does EVO 5" prints fit the 7" ?
@@ChrisRosser yeah I saw the 1s on the AOS page but the antenna mount is for the dji 03 au single antenna I was looking for a double sma antenna print.. does any of the other AOS frame mounts fit the EVO as I found a lot available for AOS 5" / 5.5" / 7" / UL5 V2
A little more actually, but the V1 has 2 antennas so in total it outputs more power. 2 antennas at 400mW is not equal to 1 antenna at 800mW though because the power doesn't add linearly.
Have you checked whether the measurement device you use can even measure higher output powers? It could be saturated and therefore not show any increase
the chip walksnail uses is rated for 250mW. its not uncommon to get more power out of rf amps, especially with cooling, but you wont be getting close to 700mW with that chip.
its not just your PC... sadly his videos are quite often out of sync.. i cant watch and concentrate i have to look away to listen and process without my brain fixating on the out of sync
The V2 is outputting more power on it's antenna than the V1 does on one of its antennas so should get more range all else equal if there's no multipath going on.
It will be interesting to see how the power numbers for all the digital systems compare to the nominal mW values given. In general I've always assumed 1000mW meant "equivalent to an analog VTX at 1000mW" rather than actually 1000mW of output.
@@siltres05s112064 you got it wrong pal. Had a bad day? Firstly you are answering not to this comment, but to another I deleted because it was already answered down there. I was just throwing ideas, and the saturation was a possibility. Chris already answered that the equipment can measure in excess of 1.6W. I love this video and my mind explodes with ideas of different tests that coud be made. At first, it's outraging that walksnail would be doing this, so I wish this could be some other issue.
In my case v1 vtx works better than v2 and its for range and penetration. I'm flying on 700 or 1000mW and can say that 700mW is better because noise ratio is smaller, 1200mW is almost unflyable on both v1 and v2.
I mean they output too much at the lower settings, but still more than the v1. so not a complete fail, just strange that they don't scale the power correctly to the settings.
Doesn't Walksnail use actually dynamic power levels and only boosts up if necessary up to the set limit? To reduce closeup interference and overheating? I mean.... Someone flew 50km on WS. I highly doubt that this would be even remotely possible on just 350mW on the V1 VTX.
No the sensor isn't saturating, it'll measure much higher power than that, tested with a 1600mW analog VTX and still nowhere near saturation. Theoretically the system saturates somewhere between 33 and 39dBm (2 and 8W).
@@ChrisRosser But wouldn't digital's crest factor play a role in signal level causing saturation? Although realistically, that would not be a sharp cutoff after 500mW. Another way to confirm the power is the same is measuring VTX current consumption.
I encourage anyone whinging about the "output" chris has measured here to do some real research on the subject. Accurately measuring the "real" output power requires instruments worth more than a luxury car. In the FCC documentation, even DJI doesnt represent their output power accurately. When it comes down to it, walksnail is capable of 50+ km of range with the right setup and conditions. I say this and someone that owns both systems (G2 + O3 and Fatshark goggles + v1/v2 VTX), WS just isnt as polished, yet.
accurately measuring the power is absolutely possible with this cheap setup, down to around 10% accuracy (which is good enough for fpv purposes). it just needs to be calibrated against a reference which can be difficult to get access to. uncalibrated, as shown in the video, its perfectly usable for relative measurements.
the test setup is not calibrated, so the mW figure you see means nothing. its only good for relative measurements which is exactly what this videon does
Before I buy an Avatar system by Walksnail they need to sort out 1. Their bitrate modes 2. Their power modes 3. How well it can hold a link. So many inconsistencies between the quality/power of their modes. It's really disappointing that they're working on new products without ironing out these fundamental issues.
Chris, congratulations on the technical level of your work. Your analysis is appalling. I feel absolutely cheated. If China were a free nation founded on democratic legal foundations, it would be up to local prosecutors to prosecute for false advertising, fraud and consumer crime. But we're talking about China... Could you run tests with the calibration files (+/- 1, 2, 3...) and analyze the effectiveness of these adjustments in increasing the output RF power? I use +1 on my VTX V1 and I've had an improvement in penetrability, but I know there's a more than proportionate rise in noise. Congratulations on the job.
... adding: I did several tests in a site with vegetation obstructing the direct line of sight, changing the calibration between -3 and +3. There is real improvement at +1, but evident degradation of signal Penetrability above +1 and below 0.
So you are getting the first in the world low delay 1080p fpv system for the lowest price point which does its best to please its users and you are still complaining? Since you are a "free nation founded on democratic legal foundations", go sue yourself for being a nag!!!
