50g Random Turn ON

[HPfreak]

Hello all,
I haven't seen any posts on this, so please forgive me if this a repeat.

My HP50g will randomly turn itself on. I have no alarms set. The only reason I know it does it is because when my 50g turns itself off (no key presses) for 3 minutes it gives a set of tones telling me it has shut itself off. No one has been near it, and like I said earlier, no alarms are set.

I was just curious if anyone else has had a similar experience.

I wrote an event log as part of my StartOff routine that I check fairly often and yes, my calculator will do it when no one is around.
Am I missing/overlooking something?

Thank you for any input!

[rprosperi]

(11-28-2016 10:21 PM)HPfreak Wrote:  My HP50g will randomly turn itself on. I have no alarms set. The only reason I know it does it is because when my 50g turns itself off (no key presses) for 3 minutes it gives a set of tones telling me it has shut itself off. No one has been near it, and like I said earlier, no alarms are set.

This is completely normal, all 50g's do this. I've read many suggested explanations, but none of them conclusive. I have 4 50g machines in various places so I hear the small chirps at all hours all around my house. 2 of them seemed to be nearly in-sync, chirping at almost the same time, but that seems to have passed.

[SlideRule]

(11-28-2016 10:21 PM)HPfreak Wrote:  My HP50g will randomly turn itself on. ... I was just curious if anyone else has had a similar experience. ... Thank you for any input!

Sorta ... my CASIO fx-115w S.V.P.A.M would seemingly activate spontaneously until I put in a fresh power cell. After that, it quit and hasn't repeated.

BEST!
SlideRule

[HPfreak]

Thank you for the feedback!
My 50g has been doing this for quite some time independent of the condition of the batteries, although I haven't replaced the CR2032. I could try that I suppose... It doesn't show as bad yet, but I could give that a shot.
Thank you!

[DavidM]

I've never seen this occur with my 50g, nor did I ever see it with the 49g+ that I used before that.

I don't set alarms on my 50g, mainly because I can't count on them working. But I do wonder if the same issue that hampers the alarms from working properly is behind what you are seeing.

IIRC, all RPL-based calcs "wake up" when a particular timer reaches 0, though (in theory) this should only result in a visibly active system if an alarm was the reason that the timer hit 0. A non-alarm timer event of this type is supposed to immediately reset the timer and power down the CPU, without having activated the calculator's normal power-up sequence. Perhaps there's something about these particular systems that fools the code into thinking it should do a full power-up when that timer event occurs.

Since you are keeping logs of this, how often does it occur? Multiple times per week/day/hour? Just curious. In particular, what's the shortest period between unintended power-ups that you've seen?

[rprosperi]

(11-29-2016 02:53 AM)DavidM Wrote:  I've never seen this occur with my 50g, nor did I ever see it with the 49g+ that I used before that.

They probably are doing this, but since you don't have any sounds associated with power-up or power-down, you never noticed when it silently wakes-up and then powers-down.

Add a BEEP command with a few tones to your STARTOFF object and welcome to the world of cuckoo calculators.

Stolen from Joe Horn, I use << { 4111. 2000. 4444. 2222. } .3 BEEP OFF >>

[DavidM]

(11-29-2016 03:26 AM)rprosperi Wrote:  They probably are doing this, but since you don't have any sounds associated with power-up or power-down, you never noticed when it silently wakes-up and then powers-down.

Well the 49g+ currently has NewRPL on it, but for the 50g I'll probably try something with STARTOFF and an assigned OFF-ON app that logs an entry on powerup if the user doesn't press a key within a certain time limit. I don't particularly want it beeping in the middle of the night. I'm curious as to whether the ghost power-ups have any pattern.

[Thomas Radtke]

Out of curiosity: Anyone seen this behaviour on the emulator, i.e. is it a hardware or a software flaw?

[Han]

(11-29-2016 06:05 PM)Thomas Radtke Wrote:  Out of curiosity: Anyone seen this behaviour on the emulator, i.e. is it a hardware or a software flaw?

I believe the behaviors would be slightly different. The emulator's clock, for example, runs off of the computer's clock. Timings would presumably be slightly different. (Technically both the PC emulator and the calc itself are both emulators.)

[DavidM]

(11-29-2016 06:05 PM)Thomas Radtke Wrote:  Out of curiosity: Anyone seen this behaviour on the emulator, i.e. is it a hardware or a software flaw?

I've never seen this behavior using Emu48, but I also know that clock-related functions aren't the same due to the same point that Han made -- namely, that Emu48 uses your computer's system clock as a reference.

I don't know that it's related, but I do know that after extensive testing I've never seen Emu48 report a subsequent time which steps backward, which is something that a real 49g+/50g will do quite easily when checked frequently enough to witness it (eg. running Saturn code).

I'm still curious as to how often people experience these ghost power-ups. I'd like to have some ballpark for how long I need to inconvenience myself while checking for this. Pressing an extra key at power-up isn't too bad, but I'm sure it will get old quickly.

[toml_12953]

(11-28-2016 10:21 PM)HPfreak Wrote:  Hello all,
I haven't seen any posts on this, so please forgive me if this a repeat.

My HP50g will randomly turn itself on. I have no alarms set. The only reason I know it does it is because when my 50g turns itself off (no key presses) for 3 minutes it gives a set of tones telling me it has shut itself off. No one has been near it, and like I said earlier, no alarms are set.

My 50g resists all efforts to keep it off. Like yours it turns itself on. I call it HAL and would never rely on it to open the pod bay doors. Hmm. Maybe if it had removable memory elements....

Tom L

[Claudio L.]

(11-29-2016 06:05 PM)Thomas Radtke Wrote:  Out of curiosity: Anyone seen this behaviour on the emulator, i.e. is it a hardware or a software flaw?

I can say that at some point during newRPL development, my machine didn't have auto power off. Therefore if it had turned on by itself it would've remained on and drained all the batteries. This did not happen during those couple of weeks, so I'm inclined to think it's a software issue, but needs some more confirmation as newRPL doesn't enable all sources for powerOn, just the On key (at the time of testing, now we have alarms but I haven't tested if if powers on by itself or not).

[TravisE]

I'm confused, too. I have two HP 50g's (one since 2009 and one since 2012), and I've never known either of them to turn themselves on nor make any beeps in the process. I have, though, always experienced the buggy alarm/clock/keyboard/interrupt-handling behavior DavidM mentions, which has always been extremely irritating, but I've never heard of a real fix for it.

Could the calculators which do this have any libraries installed that might possibly be modifying the behavior of a “stock” HP 50g? Could they have a particular ROM version or range of versions in common?

[rprosperi]

(12-04-2016 02:36 AM)TravisE Wrote:  I'm confused, too. I have two HP 50g's (one since 2009 and one since 2012), and I've never known either of them to turn themselves on nor make any beeps in the process. I have, though, always experienced the buggy alarm/clock/keyboard/interrupt-handling behavior DavidM mentions, which has always been extremely irritating, but I've never heard of a real fix for it.

Could the calculators which do this have any libraries installed that might possibly be modifying the behavior of a “stock” HP 50g? Could they have a particular ROM version or range of versions in common?

The "beeping" discussed above only occurs because my STARTOFF object (which is run when the 50g times-out and powers-off) contains some BEEP statements. So the 50g turns-itself on (I suspect due to the buggy clock issues you referenced) at various times, and then when it times-out and powers-off, STARTOFF executes, plays the BEEP sounds and shuts down.

Your 50g almost certainly turns-on at random times like every other 50g, but you never notice it as it quietly turns-off a minute later.

To verify this, create a STARTOFF object in HOME, containing BEEP commands (I suggest several, as one is easy to miss) and then just set it aside and within a few days you will notice it too beeps.

The most popular 50g STARTOFF tune is listed above - try that one (It runs on about 8-10 50g's I'm aware of - darned unlikely there are more 50g's playing any other tune).

[TravisE]

Oh, I see. This is interesting. Since I don't consider the beeper loud enough in my 50g's, I'd probably just set up a logging program. But it turns out that I essentially already have one: I've always had both calculators set up to automatically back up the HOME directory to SD card on every auto-off (with timestamps encoded into the filenames, along with an “auto-off vs. manually-invoked” flag), giving me a way to go back and track such events.

My daily-use 50g doesn't show anything suspicious yet for this month and last month. However, my older 50g (not used much anymore lately) does show about a half-dozen auto-off backups from last month as well as one for this December (the last three days), during which I'm pretty sure I wasn't using it.

So it seems like it really is happening. Given the amount of use I've given them, I would have expected to have noticed it myself long by now.

[Garth Wilson]

Quote:I don't set alarms on my 50g, mainly because I can't count on them working.

Has anyone had any problem with alarms not working on the 50g? My 41cx has been my alarm clock for 30 years and never failed. I've never tried it on my 50g though.

[Gerald H]

Edit 2016-12-05: Corrected TOFF to STARTOFF.

My 50g has been in use every day for over 1 year & I have never noticed spontaneous start up, nor on any of the 50g models I used previously.

On 29th Nov I installed a STARTOFF programme that leaves a message on the stack if actuated: So far zero spontaneous start ups.

[TravisE]

(12-04-2016 04:52 AM)Garth Wilson Wrote:  Has anyone had any problem with alarms not working on the 50g? My 41cx has been my alarm clock for 30 years and never failed. I've never tried it on my 50g though.

My experience is that when the 50g is off when alarms come due, they intermittently fail to work properly. It may turn on and activate the alarm on time, or it will do so extremely late, or it won't do it at all, with no apparent pattern.

One way to hit this seems to be to run a program to set up a long sequence of alarms every X minutes, where X is longer than the idle turn-off period set in TOFF, then let the calculator set. Sooner or later, it will stop turning itself on to signal the alarms, even when there are due alarms remaining, and only upon manual turn-on will the tardy alarms immediately activate after the fact.

[DavidM]

(12-04-2016 04:52 AM)Garth Wilson Wrote:  Has anyone had any problem with alarms not working on the 50g? My 41cx has been my alarm clock for 30 years and never failed. I've never tried it on my 50g though.

See the bug report here. The alarm issues are specific to the ARM-based systems, and at least one knowledgeable source has suggested that they may be a result of a problem with Kinpo's code. I believe there has been a subsequent posting in this forum from another knowledgeable source that questions if JYA's post is still relevant, but I haven't been able to find it.

While I definitely remember the alarm issues being discussed, this is the first thread I can recall that has mentioned the spontaneous power-up problem. That doesn't mean I doubt its veracity, though. What I am most curious about is how frequently the power-ups occur, and if there is some correlation between the power-ups and the "jitter" of the system clock in these ARM-based systems (which seems to vary with each individual calculator).

I've been running a logger on my daily-use 50g for several days now with no instances of unattended power-ups. Two other calcs (a newer 50g and a 49g+) are also equipped with the same software, but I'm leaving them on the shelf for a while before checking as I want to allow them plenty of idle time before disturbing them.

[TravisE]

(12-04-2016 04:05 PM)DavidM Wrote:  See the bug report here. The alarm issues are specific to the ARM-based systems, and at least one knowledgeable source has suggested that they may be a result of a problem with Kinpo's code. I believe there has been a subsequent posting in this forum from another knowledgeable source that questions if JYA's post is still relevant, but I haven't been able to find it.

I've wondered myself if it's a problem with the Kinpo code or if it could be even an unfixable fault in the hardware design (though hopefully not!).

One mysterious issue I've always had myself which I feel might also be related, but never heard anyone else mention, is occasional, complete spontaneous lockups while the calculator is on but the CPU is in idle state. This happens randomly to me, roughly every several weeks or months or so. I've never been able to find any pattern at all. The calculator becomes completely unresponsive to keys (including ON+C), and I'm forced to remove and reinsert a AAA battery to force a warm start. I've never lost Port 0 or HOME when this occurs, but losing the unsaved work in an ongoing editing session or stack calculation still isn't too funny.

There's also the “busy bug”, where pressing a key just as the CPU is going into idle causes the key to be ignored, only to then be processed as soon as another key is pressed. It is impossible for me to go a single day of normal 50g usage without this happening more than once.

It's frustrating for HP never to have done anything about such intrusive issues; it really blemishes an otherwise excellent and reliable product. But now I'm bordering on rant mode, so I'd better stop.