New NXT-CV7 won't finish booting

fogled@mizzou wrote: »

The unit is stuck at the square brushed aluminum AMX logo bootup screen. I'm not sure it's getting a DHCP address, and I know it's on a different subnet than the controller at the moment. I figured it was grinding away trying to auto-config, but... 2 hours going and still no timeout?

How do I get to the setup to manually configure this thing?

Thanks,

Sounds like it needs reflashed. If it's new, it's under warranty. I'd call AMX or see if you can flash it yourself.

TurnipTruck wrote: »

Sometimes you have to power cycle and they will come to life on a second try. I've seen A LOT of this with intercom panels.

Yes, I've seen that too, and was about to suggest the same. But if it still won't boot after a power cycle (cold power cycle, pull the plug, not just "reboot" it), then it's almost certainly a CF failure. When they fail, it's often as they power up, and it's not usually just a need to be re-flashed; the card itself has failed. The last panel I had do that to me, I stuck the card in my reader and ran some tests on it, and there was a whole section that couldn't be written to.

a_riot42 wrote: »

Apparently these solid state drives have a much higher failure rate than do hard disks. I hope they can improve that or else the technology is going to be toast.

Have you seen how microchips are made? They make this big sheet that could have dozens of the same chip on it all in one process, then they cut it up into individual chips. They test a very small percentage of those chips to make sure it works; if the test sample passes, the rest go out the door. If there is any inconsistency across the sheet, it isn't necessarily going to show up in the test, and there are frequently sections that the temperature wasn't high enough, or was too high, and weakens that segment. Hard drives are made one-by-one, and each one is tested in some shops. It's not that the manufacturing is necessarily any more reliable, just the testing process. I think the philosophy is that chips are inexpensive to make, and too much testing would drive the cost up too much ... they would rather just make more, or even replace bad chips than to ramp up the costs for more extensive testing.

As an aside, you know how they rate processor chip speeds? They don't make them in different speeds, they are all made in the same process. When they do the aforementioned testing, they ramp up the buss speed on the test chip until it fails, then rate the entire sheet at the speed level just below that.