Where am I Going Wrong?

[reddiesel]

Dont enter keys with auto updating flash it stops the auto updating working.

[Tesla]

The eeproms, exteep, do not have an infinate number of write cycles the manufacturer will normally guarantee them for say a minimum of 100,000 writes. If the code processed the keyroll emm everytime bearing in mind the emm comes down every 10 minutes or so this will repeatedly write the keys every time eating into the eeprom life. To prevent this when the a new emm comes along the card will store part of the hash code effectively a signiture for that emm so it will only process and write the keys one time. If you then manually change the keys when the emm comes along the card thinks I have already processed this one so already have these keys no need to write them then, it dosent know that you have manually changed them, and no update. You would have to wait untill a new keychange occurs which I know at present is a couple of times a day so not too long to wait!.

[reddiesel]

Thats what i said , nice one tesla, i wish i could see the noobs face when they read that.

[fitz]

Tesla, Red diesel, although I appreciate what you're talking about with the emm hash, if you enter the keys (incorrect or otherwise) into a freshly programmed funcard, the flash would still run the emm the first time the card sees it. It's only seen the keys before..not the emm..therefore doesn't have a hash for it.

And Reddiesel, put the wrong keys in..it'll still AU, otherwise how would keyrolls (keys, not method) work?

If you're gonna take the **** outta 'noobs', get it right.

Regards, fitz.

Edit: Tesla I see you posted that at 3:49am, for this reason you are excused. reddiesel...you posted at 1:33pm, and for this reason, you remain a scoundrel.

[Tesla]

Yea I was on emergency call out this weekend so had a bit of time on me hands. From what I can remember the funs store the emm hash in the intrnal eeprom so its the inteep file that needs to be virgin to the emm if you manualy update the keys if that makes sence.

;The following checks 3 previously stored bytes of the last successful keyroll EMM's hash.
;If they all match we dont bother processing it.(Note , not foolproof)
LDI R29,0x00
LDI R28,0x98
LDI R27,0x00
LDI R26,0x4D
RCALL ReadEEP
LD R17,Y+
CP R16,R17
BRNE _940
RCALL ReadEEP
LD R17,Y+
CP R16,R17
BRNE _940
RCALL ReadEEP
LD R17,Y+
CP R16,R17
BRNE _940
RJMP EndEMM

This is copied from the fun card asm the coders who put the fun simulation code together must have realised that it was a bit hit and miss so if your lucky it might update if your not it wont!!! If you have the time you could test this and let us know if a virgin inteep files always allows a key update.

[reddiesel]

We have had this from key changes before with people putting correct keys in with au flash , the card goes off premature, not always at the next keyroll , others can confirm this.

[fitz]

Soz for delay. Hats off to you Tesla you're right about the intEEP.

[test] = [result]:

exteep with wrong keys, blank intEEP = Updated fine

extEEP with wrong keys, int with hash of last executed pre hit EMM = Updated fine

exteep with wrong keys, int EEP with hash for current EMM = No update.

I can only comment about the Infinity USB Phoenix software, but i had to purposely engineer it for the previous intEEP to remain intact...reading it out first then writing it back as the last operation.

Even writing just the extEEP then just the flash resulted in the int getting 0'ed.

I understand the chaos source code a bit better now, the hash can be seen by opening the intEEP in bin format with a hex editor. It is, as the code suggests, at 0x4D 0x4E and 0x4F.

So kids, when you've finished, always wipe your int.

RedDiesel, you still remain a scoundrel .

Cheers, fitz

[reddiesel]

I wouldnt go anywhere without wiping my int.

[Tesla]

I was conidering changing the fun card flash to write the inteep 4D,4E and 4F all to FF when you insert the card. This would effectively make the inteep 'virgin' everytime you take the card out and put it back in. The problem is with the proliferation of people doing the changes the flash files will become very different in content. It is feasible that a coder has thought whats all this hash stuff about it prevents the keys updating and removing that segment of code from the flash thereby significantly increasing the number of writes to your card and you would never know! Whereas before most people were using the same flash that I know came from a very trusted source.