Hacking Hardware Picofly on "old" Samsung EMMC glitching fails sometimes

[superfiege]

Hey together.
I modded an old Switch OLED with Samsung EMMC. Glitching works well and fast, but sometimes (about 1 out of 15) glitching fails and gives **= (No eMMC block 1 read (should not happen))
Has anyone of you had a similar or samne problem?

This setup workes fine for SkHynix and newer Samsung but on old ones sometimes fails.

Setup:
Picofly

0.1mm wire (RST, DAT0, CLK, CMD)
0.2mm wire (VCC)
v6 APU Flex-cable (CPU, GND)

 

[superfiege]

Am I the only one with this problem? I was wondering because it happened on two similar devices…

 

[Gobracket]

Maybe bad solder connection on point d?

 

[superfiege]

Maybe bad solder connection on point d?

I dont think so, because when i swap the picofly with an instinct-nx it works fine
But I dont like instinct-chips so I would prefer to stay with the picofly....

 

[rehius]

slight adjustments to the boot waiting code

 

[Hayato213]

I dont think so, because when i swap the picofly with an instinct-nx it works fine
But I dont like instinct-chips so I would prefer to stay with the picofly....

Probably have to do with your soldering. Why you don't like Instinct NX?

 

[superfiege]

I dont know why you think its because of soldering...all other switch-models work like a charm. I suspect the specific switch-model.

Post automatically merged: Apr 10, 2024

slight adjustments to the boot waiting code

hey thank you. I will desolder the instinct-chip and resolder the picofly and give feedback.

Post automatically merged: Apr 10, 2024

slight adjustments to the boot waiting code

ok, I tried, but now with 2.77 its another behaviour:
In case of failure the led indicator now shows long static blue, then long static white and then boots to ofw.

I think failure-rate decreased, but cant prove.

 

[rehius]

In case of failure the led indicator now shows long static blue, then long static white and then boots to ofw.

in case the firmware fails to boot, it rolls back to the previous firmware, long white looks like it.

weird. how about muuuch longer timeout for every glitch attempt?

(hard to tell the real reason without logical analyzer. it is probably easier to deal with this rare issue)

 

[superfiege]

in case the firmware fails to boot, it rolls back to the previous firmware, long white looks like it.

weird. how about muuuch longer timeout for every glitch attempt?

(hard to tell the real reason without logical analyzer. it is probably easier to deal with this rare issue)

hey thank you for your support. same as before but its not a big problem for me, but i was wondering, because its the second switch where its happening. and both are ones with an old samsung-emmc. so i thought it could have a correlation.