Thread: About 37xx converted to 34xx satellites reception accuracy

Yes these files are about switch on prediction (HotFix) not about GPS Accuracy.
Reason: Data in files gets over-written after a while with the true data from the GPS.
If you convert 1300 to Jap Hotfix is not as Good as EU/US and takes longer for FIX, But it's still usable.
I think this is the case with 37>34 Real 34 is a newer device after all.
Also I think Real 34 is a bit over-optimistic by declaring 3M Accuracy all the time even when weak satellite signals are present.

@dasilvarsa
Agree with all that, and particularly your comment regarding the 'over-optimistic' 3 metre accuracy all the time by original 34xx. I've commented before about that and i flat out don't believe it. Same goes for 3590, rock-sold 3m constantly. I reckon it's rigged. Maybe real 3790 is a bit optimistic too at times, but it's somewhat more believable at 3 to 6 metres, as i found in the test of it side-by-side with 3490 convert [Only registered and activated users can see links. ].
I'm hoping Loly gets over the blues and mods 37xx gps fw into 34xx before I open a vein or strangle an innocent [Only registered and activated users can see links. ].

On balance I think you are right Angelo. Right at least to the extent that the SBAS info cannot be properly processed by 3xxx series imo. But obviously my converted 3790>3490 can 'tune' sbas sats. I've seen PRN 153 and 193 with decent signal strength, but with conflicting results. With 153 showing, the accuracy seemed worse, with 193 it was better, (apparently). But this was maybe just random and the presence of the sbas signal was simply because the unit was unable to get enough signals from true gps sats so it locked on the sbas sat in desperation, who knows? Much more testing would be required to test that theory and prove one way or another if it can actually use the signal info. The fact that it can receive the signal is however indisputable. My layman's guess is that sbas is a 'red herring' and the real answer for all converts lies in modding the fw because the real problem is fw/hw partial-incompatibility. In my convert checking Predicted EPH and Cold Start then following Loly procedure resulted in faster lock and apparently improved accuracy. As dasilvarsa points out, this is about 'hotfix', and is predictive to give the unit a 'heads up' on where the sats should be. So to a certain extent, it's illusionary 'mirrors and cigar smoke' and is not really about accuracy. But in some units, accuracy 'apparently' improves. I say apparently because maybe that's illusionary too in that i don't really believe that original 3490 and 3590 really can maintain constant 3m accuracy, and my restored 3590 attained constant 3 m with fast lock after the procedure was done on it and it didn't demonstrate that before.

Summing up, this is what i think:
1. SBAS/WAAS is of no consequence and merely incidental. It happens because the 34xx convert is sometimes not able to lock to sufficient true GPS sats, so it picks up a SBAS sat signal but really can't use the broadcast info.
2. Setting Predicted EPH and Cold Start helps some units to show fast lock and improved accuracy, but it may not be really helping accuracy in those units in fact, and doesn't seem to help accuracy in others at all.
3. The real problem (especially for the unfortunates with 15-20m) is because of partial-incompatibility and the only way to fully overcome that is by a fw mod.

Just putting it out there as usual. Simply playing 'blind man's bluff' with no scientific backing.... but my money's on modding fw. This needs more brain-storming, where's Loly?

Originally Posted by giuseppe66

In my test i've see that is not important how many satellities you have but how many signal you have on i single satellites. another test that i've do is for receive more satellites: i've delete all files in GPS directory then turn ON and (on open space), go to secret menu and select EPHER....and COLD.. and wait for 10-15 minutes for best reception without go out from that menu. I've see the reception of all satellites but accuracy always 8 - 12 about. So i've understund that accuracy is depending from quality and not from number

Hello,

Exactly, I agree with you.

Guys, look my sattelite page:

[Only registered and activated users can see links. ]

All the time my signal looks worse just running slowly. This was my better accuracy today.
The image above I was stopped with sky clear and no clouds.

PS: I was using your GPS files @lolypop000. So looks me worse than the other GPS files that @catymag share with us.
PS2: I will flash again the original 37XX firmware just to compare the satellite page.

[]'s

Originally Posted by rfb

Hello,

Exactly, I agree with you.

Guys, look my sattelite page:

[Only registered and activated users can see links. ]

All the time my signal looks worse just running slowly. This was my better accuracy today.
The image above I was stopped with sky clear and no clouds.

PS: I was using your GPS files @lolypop000. So looks me worse than the other GPS files that @catymag share with us.
PS2: I will flash again the original 37XX firmware just to compare the satellite page.

[]'s

Hello,

Look the image below:

[Only registered and activated users can see links. ]


Just flashed back the original 37XX firmware.
When I did that screenshot, I was moving during a big thunderstorm and a lot of trees between me and the sky.
So comparing the accuracy, the diference is so big! Same day, same places, just diferent firmwares.
Now the signal have a lot of power....


[]'s

I think that those that have "unacceptable GPS Performance" should revert to 37xx if it solves the problem.