RunRyder - TX/RX eCCPM Latency Test Results

rrTV-PHOTO

New HD TV

HOME rrTV-PHOTO GALLERIES MY GALLERY HELP-FAQ

myHOME PM pmRR MEMBERS 714 ONLINE 46 EVENTS SEARCH REGISTER START HERE

48 pages [ << < 3 4 ( 5 ) 6 7 NEXT >> ]

74486 views

POST REPLY

Thunder Power RC	.	Real Raptors	.	Mikado Modellhubschrauber
	.		.

Radio - Servo - Gyro - Gov - Batt > TX/RX eCCPM Latency Test Results

JKos
Elite Veteran
Location: Town of California in the state of Maryland

Quote
Is this limitation only on eCCPM?

No, it was there in normal swash mode as well. Same in PCM of PPM mode as well.

- John

Protos -- Logo 14 -- Logo 10

01-06-2006 Over year old.

JKos
Elite Veteran
Location: Town of California in the state of Maryland

Numbers added for the Futaba 9C in PPM mode. A lot slower than I expected. I thought it would be under a frame time, but I checked and double checked my measurements.

- John

Protos -- Logo 14 -- Logo 10

01-07-2006 Over year old.

JKos
Elite Veteran
Location: Town of California in the state of Maryland

I measured the 9C's latency in standard swash mode and observing channel 6, the pitch channel. Then numbers have been added to the chart. So, doing the math for eCCPM is clearly taking the 9C a little longer than it should.

Also, I would like to note why the Stylus tests a bit slower in FM mode than in PCM mode. The two reasons are that the channel pulses are spread out coming out of the receiver and a lower frame rate. Instead of channels 2 and 6 being simultaneous then immediately followed by channel 3 as in the PCM case, for FM, as with all FM on all brands, the channels come out in order 1 through X. The Stylus has a frame time of about 24 ms in FM mode versus a fast 14 ms in PCM mode so missing frame N has the "detrimental" effect of having to wait 24 ms for the next frame N+1.

- John

Protos -- Logo 14 -- Logo 10

01-08-2006 Over year old.

davehour
Key Veteran
Location: Guayama, Puerto Rico

John:

I'm sure your findings have changed the way many of us think about certain brands. You have also changed the way manufacturers should treat and provide us with the appropriate technology for what we are looking for in our hobby. The competition has just begun.

Thanks for your valuable work.

David

01-08-2006 Over year old.

Shortman
Key Veteran
Location: Clovis, California

Definately great information, I think the next great test in radio equipment should be a select group of the higher end digital servos from all major manufacturers. Test to failure and see the results.

01-08-2006 Over year old.

JKos
Elite Veteran
Location: Town of California in the state of Maryland

Measurements for 8103 and 10X added to chart.

BTW, I'm a bit skeptic of the number for the 8103 in PPM mode and normal swash. I'll recheck that tonight.

- John

Protos -- Logo 14 -- Logo 10

01-17-2006 Over year old.

Ron Henry
Senior Heliman
Location: Atlanta area, GA - USA

Thanks for the additional info JKos. You've been awesome about getting this testing done. I would never have thought that the 10x would come out slower than the 8103 but I guess the numbers don't lie

- Ron Henry
http://kaizenrc.com/forum

01-17-2006 Over year old.

JKos
Elite Veteran
Location: Town of California in the state of Maryland

New charts for PCM and CCPM data...

Note: I left out two data sets. One is the 14MZ with an R149DP receiver as that is conjecture and not measured. The other is the 14MZ using the "old" channels of 1,2, and 6. I felt that doing so made the graphs a little clearer.

- John

Protos -- Logo 14 -- Logo 10

01-17-2006 Over year old.

cdyckes
Veteran
Location: Nr. Yeovil, Somerset - UK

Very useful info John.

One question: have you tried the 14MZ in PPM mode with the new channel selections (4,5,6). I'm wondering how it would compare with the 9C and others.

Colin

01-18-2006 Over year old.

JKos
Elite Veteran
Location: Town of California in the state of Maryland

Thanks for the kind words, guys.

NOTE: For anyone interested in what the Spektrum system would do to the latency, simply add 46 ms to the numbers listed for the PPM measurements.

Colin,
No, I didn't try the 14MZ in PPM mode at all. When I started this whole campaign, I was so focused on PCM and CCPM that that is the only setup I tested the first radios with. I was a bit too focused and really should have captured all the possible data.

I believe it would be fair to say that the 14MZ in PPM mode will be much faster than the 9C. Really no doubt about that.

- John

Protos -- Logo 14 -- Logo 10

01-18-2006 Over year old.

w.pasman
Elite Veteran
Location: Netherlands

Thanks for the nice plots
So for the really lowest possible latency we have to go for JR8103 in analog mode- That's surprising news

01-18-2006 Over year old.

HOMEPAGE

JKos
Elite Veteran
Location: Town of California in the state of Maryland

Last night I rechecked all the numbers for the 10x and 8103 and did make a slight correction to the 8103 PPM/CCPM numbers.

The 10x in PPM and normal swash mode is amazing. What that number actually means is that the PPM stream starts being sent before the channel 6 pulse width has been determined. I haven't seen that in any other radio yet.

- John

Protos -- Logo 14 -- Logo 10

01-19-2006 Over year old.

rroback
Key Veteran
Location: Irvine (UCI), Ca

wow, those numbers are amazing. Instead of giving up my 8103, and buying a 14mz, I should simply have put a ppm receiver in! Do you think the first two numbers, or the average is more imporant? I'm having a tad trouble digesting the whole chart.

Rhett... There's no power like E Power!

01-19-2006 Over year old.

JKos
Elite Veteran
Location: Town of California in the state of Maryland

Rhett,
> I should simply have put a ppm receiver in!

Well, yes and no. With the 14MZ and the G3 receiver you have the benefit of simulaneous output pulses which makes a difference in eCCPM applications. You also have fail-safe. With respect to PPM and fail-safe, I would love to get my hands on a Multiplex IPD receiver and see what it does with respect to latency.

> Do you think the first two numbers, or the average is more imporant?

Since your inputs are uncorrelated to frames, it will be an average that is important. In the text chart I figured the average of the second and fourth number but in the graph I used the average of all four numbers. I've debated what is the "right" average and now I'm thinking it is probably the average of all four numbers. I may change the text chart to reflect that.

- John

Protos -- Logo 14 -- Logo 10

01-19-2006 Over year old.

Hoverdown3K
Key Veteran
Location: Rochester, New York

Wow Great information, but basicly if I am reading this correctly the Futaba 9C sucks.

too bad that is the radio I have.

Anyway to increase the speed of the Futaba 9C, can it be overclocked like a PC?

Great info and please keep on testing!

-= I know there is Money in RC helicopters. I put it there=-

02-11-2006 Over year old.

HOMEPAGE

Shortman
Key Veteran
Location: Clovis, California

Hey John, I was able to do Tic Tocs with the 7C

02-12-2006 Over year old.

willard
Senior Heliman
Location: Indianapolis, Indiana, USA

nevermind- carry on gentleman...

02-19-2006 Over year old.

DKflyer
Senior Heliman
Location: Huntington, WV

Great Information!

How much difference does this make in the real world? All of the times seem pretty quick in comparison with the time it takes the serovs to move.

DK

02-24-2006 Over year old.

w.pasman
Elite Veteran
Location: Netherlands

Ah I never noted you had als measured the spectrum module
great work

09-15-2006 Over year old.

HOMEPAGE

drdot
Elite Veteran
Location: So. California, Orange County.

fwiw..

Very curious as to whether any data has been collected on a Spektrumized 6102...I have flown a lot of 3D with the standard and modified radio, and feel the 6102 is a bit crisper. It's interesting to note the ccpm on the Spektrum is much better than 9C....

John.

09-18-2006 Over year old.

48 pages [ << < 3 4 ( 5 ) 6 7 NEXT >> ]

74486 views

POST REPLY

GrandRC	.	CanoMod	.	Futaba-RC
	.		.

Radio - Servo - Gyro - Gov - Batt > TX/RX eCCPM Latency Test Results

UPDATE SCREEN PRINT TOPIC

Advertisers

Subscribe to This Topic