New G2 Firmware 20200116 - Problems - FIXED w/ 20200121

[Jet1945]

cyrille - Tim

Edited to add more info:

DId test with four firmware versions:
20181016
20191120 beta
20200116
20200120 beta

Ran ON-SYMB x 5 (ea firmware) - with 2018 and 2019 firmware - only number displayed: 3

NOTE: with Beta firmware - entered diagnostics and ran Touch Screen test (x3) - Passed

Additional info displayed with s/n: 9CJ9440073 HW:8

The HW: 8 is the "new" info.

Ran ON-SYMB x5 with 20200116 firmware and numbers returned was 3 3

Updated to 20200120 Beta

No Touch.

Ran ON-SYMB x5 - numbers returned 3 3

[Wes Loewer]

(01-20-2020 02:39 PM)rombva Wrote:  With FW 20181016 my Prime shows number 3 in the upper left corner.
With FW 20200116 my Prime shows number 3 and after space number 2 rundomly changes to number 1.

I'm getting something similar with my unit in which touch intermittently comes and goes (Serial 9CJ9230xxx). With FW 20200116 and now with 20200120 both show mostly "3 1" but it briefly flickers "3 3" or "3 2". (I've got video if needed.)

When I check the Touch IC, it shows GT5688-516 when touch is working, and ilitek-00 when touch is not working (with the same firmware).

Earlier I reported that a simple ON-C reset fixed the touch. I now suspect the touch worked or not based on whether the hardware was reporting correctly at the moment that the device turned on.

[JDW]

If HP will swap my pristine G1 for a G2, I'd be happy to join in the Touchscreen testing! :-)

[Mark Hardman]

(01-20-2020 07:08 AM)cyrille de brébisson Wrote:  On the top left, you should see 2 numbers a 3 and some other number. Can you please report what this number is?

Thanks,
Cyrille

Following your instructions I get:

3 3


ETA: Upgrading to 20200120 did not fix the touchscreen.

[Wes Loewer]

(01-20-2020 06:49 PM)Wes Loewer Wrote:  I'm getting something similar with my unit in which touch intermittently comes and goes (Serial 9CJ9230xxx). With FW 20200116 and now with 20200120 both show mostly "3 1" but it briefly flickers "3 3" or "3 2". (I've got video if needed.)

When I check the Touch IC, it shows GT5688-516 when touch is working, and ilitek-00 when touch is not working (with the same firmware).

Earlier I reported that a simple ON-C reset fixed the touch. I now suspect the touch worked or not based on whether the hardware was reporting correctly at the moment that the device turned on.

Here's something that may be of interest.

The flickering of my calculator HW number reminded me of a digital voltmeter that jumps back and forth between two values when the voltage is right at the threshold. This made me wonder if the flickering of the HW number might be voltage related. So I thought I'd see if running the battery down would make any difference.

At 50% battery, the flickering reduced substantially such that it mostly stays on "3 1" with an occasion flicker to "3 2". As soon as I plug in the USB to a power source, the number immediately starts flickering again between "3 1", "3 2", and "3 3".

What does this mean? I have no idea. :-(

[cyrille de brébisson]

Hello,

Wes:

>The flickering of my calculator HW number reminded me of a digital voltmeter that jumps back and forth between two values when the voltage is right at the threshold.
>At 50% battery, the flickering reduced substantially such that it mostly stays on "3 1" with an occasion flicker to "3 2". As soon as I plug in the USB to a power source, the number immediately starts flickering again between "3 1", "3 2", and "3 3".

This was exactly my thinking, the 3 IO pins that I use to do the detection are "energized", "read" and "de-energized" when needed to reduce battery use when not needed...
I assumed that I did not have enough delays between the operations and that the analog parts did not have time to "stabilize". This is the change I made on the newest version, just add delays between the energize and reap portions. In my experience, this is usually enough to deal with such issues...
Unfortunately not this time :-(

Thanks to you guy for finding the issue before we went public!!!!

In order to avoid filling the forum with the problem and continuing to publish "bad" versions, would it be possible for 3 or 4 of those of you having an issue to send me their email in PM so that I can directly send you some test versions until I get it right since I unfortunately do not have any units showing the problem at my disposal.

Thanks again to a GREAT HP calc community!
Cyrille

[CyberAngel]

(01-21-2020 06:16 AM)cyrille de brébisson Wrote:  Hello,

Wes:

>The flickering of my calculator HW number reminded me of a digital voltmeter that jumps back and forth between two values when the voltage is right at the threshold.
>At 50% battery, the flickering reduced substantially such that it mostly stays on "3 1" with an occasion flicker to "3 2". As soon as I plug in the USB to a power source, the number immediately starts flickering again between "3 1", "3 2", and "3 3".

This was exactly my thinking, the 3 IO pins that I use to do the detection are "energized", "read" and "de-energized" when needed to reduce battery use when not needed...
I assumed that I did not have enough delays between the operations and that the analog parts did not have time to "stabilize". This is the change I made on the newest version, just add delays between the energize and reap portions. In my experience, this is usually enough to deal with such issues...
Unfortunately not this time :-(

Thanks to you guy for finding the issue before we went public!!!!

In order to avoid filling the forum with the problem and continuing to publish "bad" versions, would it be possible for 3 or 4 of those of you having an issue to send me their email in PM so that I can directly send you some test versions until I get it right since I unfortunately do not have any units showing the problem at my disposal.

Thanks again to a GREAT HP calc community!
Cyrille

Cyrille
Do you call this a hardware related driver problem?
VPN

[cyrille de brébisson]

Hello,

Personally, I call this a crappy documentation problem, because, if I had good doc, the problem would not be visible...
However, it is so hard these days to get good documentation on anything...
When you have doc, it tells you how it ought to work... never what might fail, and even less why it would do so and how to fix it...

Anyhow, luckily, I will be able to fix the problem relatively fast.

Cyrille

[toml_12953]

(01-21-2020 06:16 AM)cyrille de brébisson Wrote:  In order to avoid filling the forum with the problem and continuing to publish "bad" versions, would it be possible for 3 or 4 of those of you having an issue to send me their email in PM so that I can directly send you some test versions until I get it right since I unfortunately do not have any units showing the problem at my disposal.

Thanks again to a GREAT HP calc community!
Cyrille

Don't forget to also test your new versions on the unit that doesn't show the problem now. We wouldn't want the fix to break Primes that already work!

[ijabbott]

Do the IO pins have any configurable internal pull-ups or pull-downs and are they configured? (Or are they using external pull-ups/pull-downs?) If they are just floating, then waiting for them to settle to get a reading after "energizing" them probably isn't that effective!

[Gene222]

(01-20-2020 07:42 PM)CyberAngel Wrote:  My HW A works with the 2020 ROM

I have hardware version A, and successfully updated the software, but the text size on the Connectivity Kit seems a lot smaller. Is there anyway I can make the text size larger on the CK? It is really hard to read the text.

[Eddie W. Shore]

Doing the On + symb just flashes the screen for me - no numbers. Not sure if I'm doing it right. But I now notice a short "load" time of the screen when I turn on the Prime.

The touch screen on my G2 is still working properly after updating it a few days ago.

[Mark Hardman]

(01-22-2020 02:55 AM)Eddie W. Shore Wrote:  Doing the On + symb just flashes the screen for me - no numbers. Not sure if I'm doing it right. But I now notice a short "load" time of the screen when I turn on the Prime.

The touch screen on my G2 is still working properly after updating it a few days ago.

Eddie, you need to press and hold On + Symb and then release the On key while keeping Symb pressed until the calculator switches to Update Mode (3-4 seconds).

[cyrille de brébisson]

Hello,

Note: The IO pins do not need PU/Down as they are directly tied to ground of VCC...

The calc goes to update screen if it detects the SYMB key down at boot time (SYMB which I continue to call the 'C' key, a free cookie to the first one who figures out why!!!)

On G2, the boot sequence is quite long as the CPU will first load the whole application in RAM from flash before booting it. So you are looking at a couple of seconds just for that.
Then the application will kick in, booting the various drivers, including the file system which itself can take a while to boot as it needs to scann the flash to find its bearings.
Then the "main" function will start and check the Symb key. If down, it will stay in bootloader mode. Else it will start the calculator application.

Cyrille

[Joe Horn]

(01-22-2020 06:08 AM)cyrille de brébisson Wrote:  The calc goes to update screen if it detects the SYMB key down at boot time (SYMB which I continue to call the 'C' key, a free cookie to the first one who figures out why!!!)

Do you call SYMB the 'C' key because ON+C is the warmstart key sequence on many older HP calculators, including the RPL models (HP 48, 49, 50), the 40gs and its ilk, the 17bII, and probably many others too?

[CyberAngel]

(01-22-2020 06:41 AM)Joe Horn Wrote:  

(01-22-2020 06:08 AM)cyrille de brébisson Wrote:  The calc goes to update screen if it detects the SYMB key down at boot time (SYMB which I continue to call the 'C' key, a free cookie to the first one who figures out why!!!)

Do you call SYMB the 'C' key because ON+C is the warmstart key sequence on many older HP calculators, including the RPL models (HP 48, 49, 50), the 40gs and its ilk, the 17bII, and probably many others too?

Divine help is forbidden by the intrinsic competition rules!

[rprosperi]

(01-22-2020 07:20 AM)CyberAngel Wrote:  Divine help is forbidden by the intrinsic competition rules!

Aw, c'mon... he just wants a cookie...

I never knew the Prime used cookies, heck it doesn't even have a browser does it?

[Gene222]

(01-22-2020 12:35 AM)Gene222 Wrote:  I have hardware version A, and successfully updated the software, but the text size on the Connectivity Kit seems a lot smaller. Is there anyway I can make the text size larger on the CK? It is really hard to read the text.

I am using windows 10, 64 bit. The virtual calculator displays prime sans mono font, but the Connectivity Kit is displaying some other font. The CK font has proportional spacing, appears to be about an 8 pt font size, and does not display number superscripts. So, clearly the CK font being displayed is not the prime sans mono font. When I copied the CK program and pasted it into notepad, notepad displayed the program in prime sans mono font at 10 point font size.

I think my problem started when I loaded prime sans font into my windows 10 font folder. Normally, when you install the CK and virtual calculator, the prime sans mono font is not loaded into the Windows 10 font folder, but I manually installed the prime sans mono font onto Windows 10 font folder. The CK and virtual calculator font folders both contained the prime sans mono font.

After backing up the virtual calculator and moving the backup file onto my desktop, I uninstalled both the virtual calculator and CK, and reinstalled both programs. I then manually copied my App and program files onto the virtual calculator. Unfortunately, the font problem with the CK still persisted.

Any help with this problem would be appreciated.

[Jet1945]

I just installed the 20200121 Firmware with a Mac. Available on hpcalc.org (Thanks Cyrille, Tim, Eric, and everyone.)

Touch works.

[Voldemar]

(01-22-2020 09:01 PM)Jet1945 Wrote:  I just installed the 20200121 Firmware

I also. Touch Works. Looks like everything is OK.
Thank You very much, Cyrille, Tim, Eric…

Now numbers in Update Mode: 3 2
EA656 Version Info
Data Version : V14433
OS Version : 2.1
PCB Version : D
Touch IC FW Ver : GT5688-516