# ABP frame counters in memory

**URL:** <https://www.thethingsnetwork.org/forum/t/abp-frame-counters-in-memory/5522>\
**Category:** Hardware\
**Created:** [February 20, 2017, 6:59am UTC](https://www.thethingsnetwork.org/forum/t/abp-frame-counters-in-memory/5522 "2017-02-20T06:59:32Z")\
**Posts on this page:** 2\
**Page:** 2

<div class="post-metadata">

**Author:** ![Mirmit](https://www.thethingsnetwork.org/forum/user_avatar/www.thethingsnetwork.org/mirmit/32/11119_2.png) [@Mirmit](https://www.thethingsnetwork.org/forum/u/Mirmit)\
**Post date:** [March 16, 2017, 5:59pm UTC](https://www.thethingsnetwork.org/forum/t/abp-frame-counters-in-memory/5522/22 "2017-03-16T17:59:49Z")

</div>

Not sure on AVR, but on some other EEPROM/EPROM/FLASH modules, the data can be overwritten when the bits goes in one direction. AFAIR, it was going 1 to 0, this way you can writes the 8 bits before moving to the next one. This has to be checked chip by chip.

This is explained here:  
[http://electronics.stackexchange.com/questions/21232/100k-eeprom-writes-per-bit-or-as-a-whole](http://electronics.stackexchange.com/questions/21232/100k-eeprom-writes-per-bit-or-as-a-whole)  
and also have some suggestions to write the counter forward in the memory as it increase, distributing it on the whole chip area rather than in one location.

---

<div class="post-metadata">

**Author:** ![BoRRoZ](https://www.thethingsnetwork.org/forum/user_avatar/www.thethingsnetwork.org/borroz/32/44234_2.png) [@BoRRoZ](https://www.thethingsnetwork.org/forum/u/BoRRoZ)\
**Post date:** [March 16, 2017, 6:02pm UTC](https://www.thethingsnetwork.org/forum/t/abp-frame-counters-in-memory/5522/23 "2017-03-16T18:02:06Z")

</div>

possible… I was referencing to this [fram chip](https://www.adafruit.com/product/1895), which has another advantage besides [very long](https://www.adafruit.com/product/1895) lasting, no memory pages 🕶

[Previous page](https://www.thethingsnetwork.org/forum/t/abp-frame-counters-in-memory/5522.md?page=1)