Glitch times is chip dependent, you cannot change unless you use a different eMMC. You can try a different firmware but I doubt anything would change.
improve that, that´s the cause of your long glitchings time
you need to do it this way in order to have instant boot. A good ground points also helps. Clean as much as possible
Best regards
Attachments
Last edited by one-armed,
Same issue here on switch lite, used a hwfly rp2040 based, shipped already flashed.
When it exceeds 20seconds of glitching times, even if it finally succeed, switch won't boot.
I have this behaviour 1 time every 25 times.
Post automatically merged:
Points are connected well in his case. Contact is made, multimeter says it, period.
Quality of solder can impact longevity of the soldering but not glitching time
This sentence you wrote can´t be more unexact for the purpose we have here.
I highly recomend you to read and also understand how the glich attack is achieved on tegra x1 before advise each other here.
https://media.ccc.de/v/c4.openchaos.2018.06.glitching-the-switch
I really like this video
Best regards
Hi! I have recently installed one of these RP2040 HWFly Core boards on a Mariko V2 with Samsung EMMC.
https://a.aliexpress.com/_EuMhJjZ
It came without preloaded FW, installed 2.73 everything went fine, good solder work and am getting insta boot all times.
Hekate just loads fine but got the EMMC on slow mode while doing the NAND bak which completed successfully two times, but took longer.
It boots with EMMC @400Mhz but after some reads goes into slow mode (backups / benchmark)
Hekate EMMC screens tells me init error counter as 2. No R/W erros and the EMMC goes into slow mode.
Was able to boot HOS just fine but games with the updates installed on the EMMC are crashing almost always.
Do you have any ideas? Defective chip? Badly installed EMMC module on the Pico board?
Am thinking into buying an SX based HWFly which are still available on AliExpress.
Thanks!
https://a.aliexpress.com/_EuMhJjZ
It came without preloaded FW, installed 2.73 everything went fine, good solder work and am getting insta boot all times.
Hekate just loads fine but got the EMMC on slow mode while doing the NAND bak which completed successfully two times, but took longer.
It boots with EMMC @400Mhz but after some reads goes into slow mode (backups / benchmark)
Hekate EMMC screens tells me init error counter as 2. No R/W erros and the EMMC goes into slow mode.
Was able to boot HOS just fine but games with the updates installed on the EMMC are crashing almost always.
Do you have any ideas? Defective chip? Badly installed EMMC module on the Pico board?
Am thinking into buying an SX based HWFly which are still available on AliExpress.
Thanks!
Since these board are clones of the original work that started here in GBA, anything is possible. They just want to be the first in the market and don't give a shit when they customers (like you get into trouble) and thats why we always suggest that for this mod you go with the original supported boards (listed in the guide)
That being said, it seems that your issue is related to the resistors used in this board.
You should try to solder another 47ohm resistor to the dat0 line and it should be fine.
Thanks, do you know which size are they used by eye seems like 0402 resistors.
I guess I'll replace the HWFly board one with a 100Ohm one and test.
Should I change the DAT0 one only or the CMD aswell?
Thank you very much for your support.
I have no idea since I have no clue about the board you are using but you can just eyeball it and as long as it sits in place without touching anything around it should be fine. I would just do dat0, but if you are ok with soldering those little fckrs you could do cmd as well. You are welcome.
Ok so in order to help some guys that have problems with those clones. Today arrived some of the new ones called rp2040 PLUS 
.
The values of the reistors are these :
oled old
point a cmd 47 ohms
point d clk 47 ohms
point c dat0 47 ohms
point b no resistor
oled new
point a cmd 47ohms
point d clk 47 ohms
point c dat0 240 ohms
point b no resistor
v1-v2
point a 50 ohms
point d 50 ohms
point c 330 ohms
point b 50 ohms
lite
point a 50 ohms
point d 50 ohms
point c 240 ohms
point b 50 ohms
instinct-v6
point a 240 ohms
point d 240 ohms
point c 240 ohms
point b 33 kiloohms
From the measurements above we can see:
1) The oled chips dont have resistors at rst point b which i think is good. But we can see that lite and v1-v2 models use 50 ohm resistors at point b which is not recommended.
2) On newer PLUS chips the main difference is the addition of 240 ohm resistors, instead of 47-50 ohms, at c dat0 points.
I dont know if they are influenced from instinct-nx chips which use 240 ohms at all a,d and c points. This also doesnt make a lot of sense since the recommendation is to add 100 ohm resistors at point c dat0.
3)for some reason the v1-v2 clone uses 330 ohms at point c. Even more than 240 ohms which are used at oled and lite rp 2040 clone boards.
First of all check if you have slow emmc errors after the installation. If yes then proceed with the following:
On old oled boards add 2 more 0201 47 ohm resistor at point c and point a as you can see in my picture (or simply replace them with 100 ohm resistors) . First cut the traces and then add the 47 resistors if you dont replace with 100 ohm.
On new oled models replace the 0201 resistors of point a and c with 100 ohm resistors.
On v1-v2 board replace point a and c with 100 ohm 0402 resistors and if it still doesnt work remove the 50ohm point b resistor and bridge the points.
On lite board
replace point a and c with 100 ohm 0201 resistors and if it still doesnt work remove the 50ohm point b resistor and bridge the points.
.The values of the reistors are these :
oled old
point a cmd 47 ohms
point d clk 47 ohms
point c dat0 47 ohms
point b no resistor
oled new
point a cmd 47ohms
point d clk 47 ohms
point c dat0 240 ohms
point b no resistor
v1-v2
point a 50 ohms
point d 50 ohms
point c 330 ohms
point b 50 ohms
lite
point a 50 ohms
point d 50 ohms
point c 240 ohms
point b 50 ohms
instinct-v6
point a 240 ohms
point d 240 ohms
point c 240 ohms
point b 33 kiloohms
From the measurements above we can see:
1) The oled chips dont have resistors at rst point b which i think is good. But we can see that lite and v1-v2 models use 50 ohm resistors at point b which is not recommended.
2) On newer PLUS chips the main difference is the addition of 240 ohm resistors, instead of 47-50 ohms, at c dat0 points.
I dont know if they are influenced from instinct-nx chips which use 240 ohms at all a,d and c points. This also doesnt make a lot of sense since the recommendation is to add 100 ohm resistors at point c dat0.
3)for some reason the v1-v2 clone uses 330 ohms at point c. Even more than 240 ohms which are used at oled and lite rp 2040 clone boards.
First of all check if you have slow emmc errors after the installation. If yes then proceed with the following:
On old oled boards add 2 more 0201 47 ohm resistor at point c and point a as you can see in my picture (or simply replace them with 100 ohm resistors) . First cut the traces and then add the 47 resistors if you dont replace with 100 ohm.
On new oled models replace the 0201 resistors of point a and c with 100 ohm resistors.
On v1-v2 board replace point a and c with 100 ohm 0402 resistors and if it still doesnt work remove the 50ohm point b resistor and bridge the points.
On lite board
replace point a and c with 100 ohm 0201 resistors and if it still doesnt work remove the 50ohm point b resistor and bridge the points.
Attachments
After reading your post, I think we should adopt/test with these resistence values based in the switch version, rather than changing them on the ribbons. I suspect that the "official"(100ohm) method tried to use universal resistence values which might not be optimal for every switch version.Ok so in order to help some guys that have problems with those clones. Today arrived some of the new ones called rp2040 PLUS
The values of the reistors are these :
oled old
point a cmd 47 ohms
point d clk 47 ohms
point c dat0 47 ohms
point b no resistor
oled new
point a cmd 47ohms
point d clk 47 ohms
point c dat0 240 ohms
point b no resistor
v1-v2
point a 50 ohms
point d 50 ohms
point c 330 ohms
point b 50 ohms
lite
point a 50 ohms
point d 50 ohms
point c 240 ohms
point b 50 ohms
instinct-v6
point a 240 ohms
point d 240 ohms
point c 240 ohms
point b 33 kiloohms
From the measurements above we can see:
1) The oled chips dont have resistors at rst point b which i think is good. But we can see that lite and v1-v2 models use 50 ohm resistors at point b which is not recommended.
2) On newer PLUS chips the main difference is the addition of 240 ohm resistors, instead of 47-50 ohms, at c dat0 points.
I dont know if they are influenced from instinct-nx chips which use 240 ohms at all a,d and c points. This also doesnt make a lot of sense since the recommendation is to add 100 ohm resistors at point c dat0.
3)for some reason the v1-v2 clone uses 330 ohms at point c. Even more than 240 ohms which are used at oled and lite rp 2040 clone boards.
First of all check if you have slow emmc errors after the installation. If yes then proceed with the following:
On old oled boards add 2 more 0201 47 ohm resistor at point c and point a as you can see in my picture (or simply replace them with 100 ohm resistors) . First cut the traces and then add the 47 resistors if you dont replace with 100 ohm.
On new oled models replace the 0201 resistors of point a and c with 100 ohm resistors.
On v1-v2 board replace point a and c with 100 ohm 0402 resistors and if it still doesnt work remove the 50ohm point b resistor and bridge the points.
On lite board
replace point a and c with 100 ohm 0201 resistors and if it still doesnt work remove the 50ohm point b resistor and bridge the points.
@abal1000x what do you think?
On second thought, if it works there might be no need to change the values. I was just hoping to eliminate some of the more common and seemingly random issues encountered until now. This is way over my head anyway
Last edited by twins333,
I think the new hwfly is already good.After reading your post, I think we should adopt/test with these resistence values based in the switch version, rather than changing them on the ribbons. I suspect that the "official"(100ohm) method tried to use universal resistence values which might not be optimal for every switch version.
@abal1000x what do you think?
On second thought, if it works there might be no need to change the values. I was just hoping to eliminate some of the more common and seemingly random issues encountered until now. This is way over my head anyway
Adding the dat0 resistor is a good one.
Its not always 100 ohms. Its just that 100ohms is the most people used, to fix the 'slow emmc' issue.
And if its tested by most, then its the best.
I myself have made an experiment until 1kOhms, and it seems works.
Adding the resistor to the nvidia reset line, seems 'not wise' decision.
i don't see anyone do it, so it might triggered an unknown issue.
'If it aint broken then dont fix'
Similar threads
