[WP-34S] Cable and flashing options for Mac users

[John Galt]

(01-23-2015 11:24 PM)walter b Wrote:  Don't be afraid. Watch this: http://youtu.be/diJA2cCQenI . It isn't Mac-specific.

LOL!

Done! Came out beautiful thanks to Eric's video. But it was still nerve-wracking.

[Dave Frederickson]

(01-25-2015 02:15 AM)John Galt Wrote:  Done! Came out beautiful thanks to Eric's video. But it was still nerve-wracking.

Perhaps an assistant could help.

[MarkHaysHarris777]

(01-25-2015 05:41 AM)Dave Frederickson Wrote:  Perhaps an assistant could help.

I agree totally! When I post my process pics I will be sure to include my assistant pic just for you... I actually have two assistants... the one is called Joules (I'll let you figure out why) and the other Ava. Joules is ADHA, while Ava is the intellectual (and she keeps my feet warm while I solder....

[walter b]

Seems here's a nest of cacas.*

d:-)

* calculator cats

[Massimo Gnerucci]

(01-25-2015 09:35 AM)walter b Wrote:  Seems here's a nest of cacas.*

d:-)

* calculator cats

Even if it doesn't sound so nice in italian, I admit there's a couple of them here too.

[rprosperi]

(01-25-2015 10:08 AM)Massimo Gnerucci Wrote:  Even if it doesn't sound so nice in italian, I admit there's a couple of them here too.

In the rare times that calculators get boring, stimulate your assistant with a laser pointer around the room. Terrifically entertaining. For me too.

[Massimo Gnerucci]

(01-25-2015 01:36 PM)rprosperi Wrote:  

(01-25-2015 10:08 AM)Massimo Gnerucci Wrote:  Even if it doesn't sound so nice in italian, I admit there's a couple of them here too.

In the rare times that calculators get boring, stimulate your assistant with a laser pointer around the room. Terrifically entertaining. For me too.

Right

[Dave Frederickson]

(01-25-2015 02:08 PM)Massimo Gnerucci Wrote:  

(01-25-2015 01:36 PM)rprosperi Wrote:  In the rare times that calculators get boring, stimulate your assistant with a laser pointer around the room. Terrifically entertaining. For me too.

Right

Because cats can never catch a laser pointer I don't play with one, he's neurotic enough.

(02-23-2014 07:08 PM)Massimo Gnerucci Wrote:  As much as I carefully check and clean beforehand I always find a cat's hair in a corner whenever I do something like applying a screen protector to my phone...

Yes, that is an issue. My GF likes to say that cat hair is both an accessory and a condiment. On Kai's behalf, he's a Bengal and doesn't shed very much.

Dave

[Dave Frederickson]

(01-25-2015 09:35 AM)walter b Wrote:  Seems here's a nest of cacas.*

d:-)

* calculator cats

If anything can be learned from watching The Big Bang Theory, a group of cats is a clowder or glaring.

[MarkHaysHarris777]

(01-25-2015 04:36 PM)Dave Frederickson Wrote:  If anything can be learned from watching The Big Bang Theory, a group of cats is a clowder or glaring.

Yeah, and that's why I only have two... because herding a clowder of cats is a glaring impassability (which can cause brain injury and death).

[Dave Frederickson]

(01-25-2015 10:43 PM)MarkHaysHarris777 Wrote:  Yeah, and that's why I only have two... because herding a clowder of cats is a glaring impassability (which can cause brain injury and death).

Very true. For those who haven't seen the video:

[John Galt]

Some interesting timing measurements:

(first picture) Flashing the calculator sends many records of length 22 mS (upper trace). The required replies from the calculator (acknowledgements, lower trace, at the trigger point "T") occur approximately 9 mS after each record is sent - I think each and every record gets a reply, but I am not certain.

The entire file, most recent version takes 19 or 20 seconds to send from my Mac so the approximate total number of records can probably be derived from that.

(second picture) The reply (lower trace) consists of only a few bits but they're always the same, and of 78 uS in duration. It corresponds to ASCII character Y (upper case). Or, it's just a nice symmetrical character to choose - zero one zeeeero ooooone zero one stop :-)

I suppose N would represent a negative acknowledgement.

(third picture) The initial transmission from MySamBa. It corresponds very closely to 115200 bps and completely confirms Mr. von Cube's explanation of the protocol. Subsequent transmissions occur at exactly that same speed on my calculator, but it's unlikely that everyone else's will do that and MySamBa won't need to adjust transmission speed... however it does that.

If MySamBa receives no reply at all, it sends the exact same thing and at the exact same speed, a total of three transmissions before it gives up. This puzzles me, and I still wonder how it's supposed to cope with a calculator lacking a crystal timebase that might not recognize the bitrate.

(fourth picture) During normal communication, subsequent transmissions occur a minimum of 5 mS after the previous one's reply. Sometimes it's more than 5 mS.

With respect to Mr. Meheut's wp34sflash program, the reason it doesn't work is obvious. What it sends is completely different and bears no resemblance to what MySamBa sends. What it does, and what I have not been able to observe MySamba doing, is to vary the bitrate from the first transmission to the fourth (it sends four before giving up). The hard work in porting MySamBa's essential functions to the Mac is surely done though, and if he would be so kind to point me in the right direction for its source, perhaps I could make that program work.

If you don't have a scope, Mr. Jebem's Tx and Rx LEDs are VERY helpful diagnostic tools. They clearly show activity when it's supposed to occur. If the calculator is OFF (a bonehead mistake I've made more than once) the Rx LED will glow faintly but steadily, indicating something is wrong.

Maybe this information will help others if they have trouble, I suppose most people won't care.

On a related subject, it seems pretty difficult to kill the calculator. If I haven't fried something by now I probably never will :-p

I realize flash memory can only withstand a finite number of flash cycles (10,000 according to Mr. Bonin's manual, footnote 49). I'm two or three dozen into that, and will start counting! For this reason I am limiting testing to a sacrificial machine.

[pascal_meheut]

(01-26-2015 05:41 PM)John Galt Wrote:  With respect to Mr. Meheut's wp34sflash program, the reason it doesn't work is obvious. What it sends is completely different and bears no resemblance to what MySamBa sends. What it does, and what I have not been able to observe MySamba doing, is to vary the bitrate from the first transmission to the fourth (it sends four before giving up). The hard work in porting MySamBa's essential functions to the Mac is surely done though, and if he would be so kind to point me in the right direction for its source, perhaps I could make that program work.

The source code is in SourceForge (https://sourceforge.net/p/wp34s/code/HEA...nk/QtFlash) but you should know that:
- it has been written by looking at MySamba source code and converting it
- I've written some tools to check that I was sending the exact same thing as MySamba
- it works more reliably on Windows & Linux than on OSX

Of course, I may have missed something obvious and you are more than welcome to have a look at it: some people helped me to test it but as far as I know, nobody else tried to debug it.

[John Galt]

Thank you Pascal! I will do that.

I really appreciate you starting that project, that's the hardest part. There's no way I would have had even that amount of motivation.

[John Galt]

some more finished product pictures...

I decided to forego Jebem's Reset LED, for no other reason than I ran out of the panel mount LEDs I had in stock. That left a lot of empty board space.