i want to download gc2 firmware

[Ronsc1985]

I think I was pretty clear on where I got my numbers by linking to the document. That is an application note from a person that works for a company that makes accelerometers. So he is not assuming or guessing. He knows.
The key point to understand about all of this is that the accelerometer resolution and relative accuracy is not the problem. Both are way much better than this application needs.
I have no idea of the GC2 uses a temperature compensated accelerometer. If you know that for a fact then that would be interesting info.
I have never seen an accelerometer that has built in automatic temperature calibration. The process to calibrate it to work across temperature with high accuracy is way much more complicated than placing it on a leveled surface and storing the offset.

The point is there were no numbers in the article. You made up the "fact" you should have the unit perfectly level when hitting balls. This is false. The tilt calibration routine supplied by Foresight, takes care of this if you level the unit when running the calibration.

As to temperature compensation each and every semiconductor accelerometer made today has it designed in , especially for the narrow range encountered in hitting golf balls. How do I know this, 35 years of designing IC sensors of all types, a PHD in EE. and a lot of experience using them in race car instrumentation.

I have no idea which sensor Foresight uses but only and idiot engineer would pick one that compromises the unit performance over the tilt and temperature range these units operate in. Even inexpensive Harry Homeowner electronic protractors have .05 degree or better resolution and built in push the button zero calibration,.

[davray666]

I don't think we need to debate weather there where any numbers or a relevant point in the article I refereed to. Anybody that can read can check it out for them self and draw their own conclusions.
It's clear that any further discussions with you on this topic will result in a heated debate and probably insults. I don't have any will or time to engage in that.
So I am out. Anybody else that wants to discuss this in a more civilized manner feel free to engage.

[SFR]

Latest firmware version is 3.10. I have no idea what is different from 3.07 since I haven't run it yet.

Bump to get this back on topic of firmware.

3.10 is new and if anyone has tried it with Protee/TGC can you report back. Unless it changes how it reads the ball I don't plan on updating. I like to know what my flash count is and that was lost on the 3.7 version. Possibly that is back?

Protee tends to test these out for us just in case they don't work.

[FaultyClubs]

I read the article. They state that Automatic Leveling varies from real to fantasy. They don't provide numbers for our particular application and we don't know how the GC2 fits so it is hard to make a valid conclusion based on the article.

The "Automatic Leveling" mentioned above covers accuracy over wide temperature ranges AND multiple arbitrary orientations and accelerations none of which apply in our case. The problem is much easier in static conditions over limited temperature ranges and where one is ONLY measuring the gravity vector. Then calibrating the device offsets and (also important) sensitivity errors can yield good results. The fantasy part comes when the device is moving all over the place over wide temperature ranges and one still wants better than 0.5 degree accuracy. Varying offset and sensitivities in all the axis including cross axis coupling issues make it hard. On the scale of easy to hard the GC2 application is towards the easy end IMHO.

Also, if accelerometers are being used in the calculations their errors will apply regardless if the unit is level or not so I don't see (much) benefit to putting it on a leveled surface. I say "much" because the sensitivity errors are non-linear and hard to calibrate out over large ranges but I don't think a big deal over the small tilt angles and temperatures we are concerned with.

Having said that I have no idea how sophisticated the GC2 or Skytrak is with their accelerometers nor how sophisticated their calibration tools are...ideally one wants multiple calibration points per axis (at least two each for total of six) for good results.

I also have a hard time thinking GC2 engineers would allow their system accuracy to be compromised by their accelerometers. Especially when simple alternatives exist. A couple precision bubble vials (the kind found in machinist levels) with leveling feet easily exceed the leveling requirements for the system. Such a system is cheap, easy to implement and simple to use.

Bottomline: If the article had stated the limits of their accelerometers in a GC2 like application then it would have been useful. I notice they have a mechanism to question the authors so perhaps anyone with enough interest can ask about limits in our specific application. That would be interesting and move the discussion from philosophy to engineering. As it stands, I didn't think the article tells us anything useful. But that is just my opinion.

[davray666]

Good point. The article obviously considers a case where the full range in each axis is used. It might well be that the very narrow tilt use case is much more accurate. I don't have my GC2 yet, when i get it I will check what accelerometer is used and if the tilt calibration tool reports values it's easy to verify how accurate it is.
Let's hope my suspicions are wrong and your assumptions are correct. We probably should open up a new thread for this topic in case anybody has anything more to add.

[Bruce]

Don't know what changed, but 3.10 runs fine with TGC. Still no flash counter :-((((((((

[mthunt]

Why is everyone so worried about the flash counter? Trust me. You'll know when the flash is done.

[SFR]

Why is everyone so worried about the flash counter? Trust me. You'll know when the flash is done.

Well... for me I would like to know when my flash does go, if you got the 400,000 to 500,000 flashes out of it or did it only last half as long as it should have. Either way it is a good idea to have one on hand in case. I personally think that this reading should not have been removed from the info screen to begin with.

[Js1010]

Anyone else try firmware 3.10 yet or get any info on whats changed?

[Mxwoodman]

I had 3.10 and sent my GC2 in for repairs and it now has 3.9.2 not sure what the changes are, but before I sent it in had installed a older firmware version and it started auto hitting in the FS software and they said it was because of a mismatch of the firmware and the FS software.

[Ronsc1985]

Anyone else try firmware 3.10 yet or get any info on whats changed?

I've had it installed for a couple of weeks and as far as I can tell nothing of note has changed. I'm only using the system with FS1 course, range and fit software and have used the course and range sim but not the fit portion of the system software. I am not streaming anything and only using the mini usb connector from the GC2 to the sim computer.

[Js1010]

Thanks - Will give it a try soon. Seems only Bruce has tried this with TGC so far. Hopefully there will always be the option to downgrade by reinstalling the older firmware if need be.