rrTV-PHOTO   New HD TV
HOME   rrTV-PHOTO   GALLERIES   MY GALLERY   HELP-FAQ
myHOME PM pmRR MEMBERS 293 ONLINE 20 EVENTS SEARCH REGISTER  START HERE
 
54 pages [ <<    <     8      9     ( 10 )     11      12     NEXT    >> ]85447 viewsPOST REPLY
GrandRC . CanoMod . Futaba-RC

.
.
Radio - Servo - Gyro - Gov - Batt > TX/RX eCCPM Latency Test Results
 
 
Yug
rrProfessor
Location: UK. Herts

Quote 
I feel the difference directly when doing piroflips etc, actually, first time running the DX7 I had do put in some more expo just to slow it down.
Same here.
When I setup the DX7, I matched everything exactly to my 9C using a scope.
04-28-2007 Over year old.
 
 
hootowl
Elite Veteran
Location: Garnet Valley, Pa.

I can see where you would notice a difference from the 9C. Like I mentioned before, the 9C has some issues besides being horribly slow. It also was a bit choppy. My servos used to move like they were being restricted with my 9C.

All I'm arguing here is once you get into the 50-60ms zone I don't think anyone will feel a difference.

Member Bog Troll Club #1
04-28-2007 Over year old.
HOMEPAGE  
 
 
MrMel
Elite Veteran
Location: Lidingo, Sweden

Quote 
Did you change anything else when you went to the DX7 or just the radio?
When changing radios you are also reprogramming a whole model. The radio itself may translate what your settings were on your old radio differently. Can can confirm that lower latency is THE reason you needed expo?

Of course I changed the receiver on the heli,but if you refer to mechanical setup/throw or servo change, No, not on any of my 5 helis.
So, yep, its the latency that makes it.

Its actually fun now, because before (14MZ with oldstyle PCM, not 5014) the latency made a big difference between trying something in the simulator (very little latency) and in real life.

Now with the DX7, its so simular to the latency in the simulator, its a breeze to go out and do the same stuff you do in the simulator, for as long as your nerves hold togheter that is

I think moves like the piroflips are the most noticable since those milliseconds make a huge difference in the overall completion of the move.
04-28-2007 Over year old.
 
 
hootowl
Elite Veteran
Location: Garnet Valley, Pa.

Mel what radio did you have before?
As far as sims are concerned, what latency do they have? Don't they require other processing that TX/RX systems don't have to do? Graphics cards, CPU and what have you? These comparisons are not very firmly based.

It's funny because I have found flying my heli for real to be more enjoyable than flying the sim. The sim seems less responsive to me. And that's with one of the highest latency setups in the chart. PCM 1024 on a 12Z. Same as you.

Member Bog Troll Club #1
04-28-2007 Over year old.
HOMEPAGE  
 
 
MrMel
Elite Veteran
Location: Lidingo, Sweden

as I wrote 14MZ, but I couldnt use the 5014 receiver due to interference, so I was running old style PCM.

Just do a piro flip in the simulator and try it in real life and see instantly that your timing is off... no exact measurements needed, its just very noticable.
Of course the sims can be tweaked to be slower..
04-28-2007 Over year old.
 
 
hootowl
Elite Veteran
Location: Garnet Valley, Pa.

No kidding timing is different from a sim to real life. You are comparing a $150 software to real life. Corporations spend millions developing "real life" sim software for training pilots.

Member Bog Troll Club #1
04-28-2007 Over year old.
HOMEPAGE  
 
 
MrMel
Elite Veteran
Location: Lidingo, Sweden

What Im saying is that with the DX7 its not different anymore...

well, try one, ask a friend to borrow one if you dont want to buy it, then you will know what Im talking about.

And, Im pretty sure as you say its hard to see 50-60ms, but feel it, its no problem.
04-28-2007 Over year old.
 
 
GimbalFan
Elite Veteran
Location: Copter County, Nv

Quote 
XPS data added. Basically, it adds 25 ms to the PPM stream from the transmitter. - JKos
Thanks again, John. Good to know.

Looks like one can take any tx from the 'PPM and eCCPM' portion of your text table and add 25 ms to its 'Avg' column number to get a good idea where a particular tx/XPS combination will land in this data pool while doing eCCPM.

EDIT -- pic removed. John has updated his data at the start of this thread.

op-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-t
04-28-2007 Over year old.
 
 
Peter Wales
Key Veteran
Location: Orlando Fl

I'm confused!

Where did 25mS come from?

The 9303 AVG is 36mS and the XPS version is 49mS. My Cray computer worked that out to be a difference of 13mS.

Significant but not 25mS.

What am I missing?

Whoops, OK I just realised I should be comparing the figure in the top column with those in the bottom column.

Peter Wales
04-28-2007 Over year old.
HOMEPAGE  
 
 
GimbalFan
Elite Veteran
Location: Copter County, Nv

You've got to look at a tx's PPM performance as a baseline for comparison because it must be in PPM to talk to the XL module.

op-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-t
04-28-2007 Over year old.
 
 
hootowl
Elite Veteran
Location: Garnet Valley, Pa.

So you are saying eCCPM adds about 25ms on the average?

In that case new topic needs to be started on why mCCPM is better than eCCPM

Member Bog Troll Club #1
04-28-2007 Over year old.
HOMEPAGE  
 
 
MrMel
Elite Veteran
Location: Lidingo, Sweden

No, they are saying the XPS module add 25ms to the PPM stream regardless of radio.
04-28-2007 Over year old.
 
 
GimbalFan
Elite Veteran
Location: Copter County, Nv

Quote 
So you are saying eCCPM adds about 25ms on the average?

Looking at the numbers on John's table, it appears that most tx's computing of servo mixing in eCCPM only adds around 5 ms to their respective latencies, with the biggest difference seen on a 9C using an FM rx (19 ms). The data for a 9C in mCCPM (normal) using a PCM rx isn't listed.

And using a 9C with a PCM rather than a PPM rx adds another 19 ms to its latency.

op-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-thwõp-t
04-28-2007 Over year old.
 
 
hootowl
Elite Veteran
Location: Garnet Valley, Pa.

OK now I'm confused.

Futaba 14MZ |R5014DPS|14 | 14| 31.5| 31.5| 22.75 *
Futaba 14MZ |PCM1024 |29 | 93| 42 |106 | 67.5 ***

What I don't see yet is how will the XPS system compare running PPM through the XPS system? Or other radios for that matter?

I guess we don't have the module yet to get the results? And the only example we have is the 9303.

OK I understand now... you are looking at the 9303 before XPS and after.

PCM, DSM, DSM2, FASST, XPS in eCCPM Mode
Transmitter |Receiver|MLF|MLL|MaxLF|MaxLL|Avg
JR 9303 |649S |15 | 47| 36 | 47 | 36.3
JR 9303 |770S |15 | 47| 36 | 47 | 36.3
JR 9303 w/XPS |XPS 8-ch|35 | 39| 57 | 65 | 49


PPM and eCCPM
Transmitter |Receiver|MLF|MLL|MaxLF|MaxLL|Avg
--------------+--------+---+---+-----+-----+-----
JR 9303 |FM |10 | 14| 32 | 40 | 24




.

Member Bog Troll Club #1
04-28-2007 Over year old.
HOMEPAGE  
 
 
MrMel
Elite Veteran
Location: Lidingo, Sweden

I think JKos didnt expect people to use PPM again so he didnt test those on all radios, unfortunally.


Quote of JKos at another forum
Quote 
I started collecting the PPM data well after I had measured a bunch of radios in PCM mode. My main thrust at the beginning was PCM. It was only later that I decided PPM might be useful. If I had only known what is going on now back then, I could have had PPM numbers for all the radios listed (where appropriate) like 9Z and 14MZ.
04-28-2007 Over year old.
 
 
Wyn
Senior Heliman
Location: Oregon, USA

John (or someone else who's tested), do the XPS RX outputs appear roughly simultaneous (in banks?), or sequenced regularly 1 to X from beginning to end of the frame?

Wyn
VoltMagic
04-28-2007 Over year old.
 
 
JKos
Elite Veteran
Location: City of California in the state of Maryland

Wyn,
Output pulses on the XPS receiver are sequential 1 to N with the current firmware. Mr. Drew stated that if you use the channel mapping feature to duplicate a channel, it would be simultaneous. I have not tested that feature.

- John

MSH Protos
04-28-2007 Over year old.
 
 
flyboy0413
Senior Heliman
Location: CT, USA

Hmm... I remember XPS as being advertised as only adding 2ms to latency, on top of the PPM latency. This got me really excited, because this would have put a 9303 radio w/ XPS ahead of the DX7. But according to Jkos testing, the number is more like 25ms.

So where did the 2ms that XPS was advertising come from? What am I missing?
04-29-2007 Over year old.
 
 
hootowl
Elite Veteran
Location: Garnet Valley, Pa.

The XPS site says ...
Latency (measured from stick input to receiver output) 2ms to 1 frame

Member Bog Troll Club #1
04-29-2007 Over year old.
HOMEPAGE  
 
 
flyboy0413
Senior Heliman
Location: CT, USA

Yep,

I saw that on the website too. But according to John's testing, it's nowhere near 2ms to 1 frame (about 20ms).
04-29-2007 Over year old.
 
 
54 pages [ <<    <     8      9     ( 10 )     11      12     NEXT    >> ]85447 viewsPOST REPLY
A Main Hobbies . Boca Bearings . Modefo's RC Helicopters

.
.
Radio - Servo - Gyro - Gov - Batt > TX/RX eCCPM Latency Test Results
 PRINT TOPIC Advertisers 

Subscribe to This Topic

Saturday, November 22 - 11:28 am - Copyright © 2000 - 2008 runryder.com | email | link to rr | runryder needs cookie