Different Switch Models and there number on the mainboard

Hi,I would like to know which switch Version (unpatched V1, patched V2) has which indications in the Mainboard.
Like HAC and HAD.
And why Do they sometimes have CPU-10 and CPU - 20 interchangeable?

I got both HAC and HAD CPU 10.

Which of These are a an old unpatched CPU and which are newer ones?

I also find it wierd some CPU 10 Boards missing the dioad Arrays, and other CPU 20 Having them in place?

These board markings don’t really mean much to us I’m afraid. Nintendo switched to the patched type mid way through the CPU 10 board rev, the best it can tell you is, there is a higher chance of it being unpatched if it’s a CPU - 10 board but there is a good amount of patched CPU 10 boards which are also patched… the inverse is true also, there are a few unpatched CPU 20 boards but the probability is the vast majority are patched.

The only thing (from memory) that these marking definatively tell you is which region the switch was sold in, the chinese and austrailian boards have a different marking (if I’m remembering right)

This was a cost cutting excercise mid way through, same as earlier, the vast majority of CPU-10 boards have the diodes and a small fraction don’t, CPU-20 boards - a small fraction has them, the vast majority don’t

2 Likes

Shouldnt the CPU 10 or CPU 20 indicate if it’s a erista or mariko SOC?

Also, Do you have any resistance messurements picture or table for the V2 Boards?
I could only finde one for V1 by @SheriffBuck

nope, the SoC’s are interchangeable between boards so it has no bearing.

Mariko is easy to identify as they all use the alt CPU/GPU Max IC reg etc - and of course Switch lites are Mariko

if your talking about CPU-20 marked boards, as mentioned above they’re pretty much identical, if your talking about Mariko revs, they’re largely the same as original rev, the CPU 0 rail is approx half what you’d get on a regular rev switch… few other subtle differences here and there (most related to SoC rails)

1 Like

@Severence

Ah Thanks.
So is around 15-20 Ohm’s on 1,15 normal for a Mariko.
My switch lites which are working All have this Reading, but on the non lite Version I am not sure if this is still a good Reading or not.

On Erista this line is normally 40-60 Ohm’s.

So you mean there are patched Switches which still use the first board Layout and arent mariko?

I thought all mariko Boards are patched and also uses the improved SOC.

That’s fine. You can use your lite boards as a reference as they’re virtually identical

Right, so the first SoC revision had the vulnerability which was subsequently “patched” in the second revision SoC, third revision SoC is termed as “Mariko” and does not have the vulnerability at all - this SoC is also used on Lites.

Right, it’s more efficient (hence why all SoC related rails will measure lower etc) which in turn improves battery life etc

1 Like

Ah nice, thought there are only 2 cpus with both patched and improved Layout.

All Switches in the Red Packaging are Mariko since they changed the package with the updated CPU or?

Yeah I think Mariko always comes a red box, though your probably better off verifying this by either looking at the board yourself (for the alt max IC’s etc) or putting the serial number in ismyswitchpatched which if I remember right will tell you if it’s Mariko

1 Like

Yeah there are some serial numbers which are marked as unknown/potentially patched so you cant sure by it only.

How Do you tell if the CPU on a mariko is bad?
Also any good way to check MAX77620H or MAX77812 for functionality?

Thank you very much :blush:

The same way you would on a standard rev, checking the related power rails, though this is not truly conlusive it rules out or confirms most issues.

measuring resistance to ground on all produced rail outputs, checking the voltage on said outputs, checking enables (if applicable/known - I don’t remember off the top of my head) … though some caveats, depending on the fault condition, some rails may not be produced for a completely unrelated board fault which I’ve seen before, for example, on Mariko it’s quite common for 1V35 to not be produced which is not down to any fault relating to the PMIC (usually) but rather some unknown fault elsewhere. My theory is this particular rail is enabled later on in the chain compared to a regular rev Switch in which the rail ordinarily comes up following prompt, so on Mariko during what I presume is a power on self test of sorts - something is failing

1 Like