Purpose of Grounded Pins on CPU?

YoshiMoshi

Well-Known Member
OP
Newcomer
Joined
Sep 5, 2016
Messages
53
Trophies
0
Age
31
XP
209
Country
United States
I've noticed that on GBATek, that the CPU pins on the original GBA (fat one not SP), pins 109 110 111 and 112 are grounded on the CPU (AGB CPU). Does anyone know if these pins are actually all connected together internally (indicating that they were originally intended to be grounded). This require removing the CPU from the motherboard and testing with a multi meter. I just find it interesting that there are not four pins that are grounded on the CPU GBA B (the GBA SP CPU). I'm wondering if these four pins have some sort of functionality, but they were grounded for commercial release.

Does anyone know the purpose of these pins? I've tried searching and can't find anything on them.
 

FAST6191

Techromancer
Editorial Team
Joined
Nov 21, 2005
Messages
36,798
Trophies
3
XP
28,373
Country
United Kingdom
You could also tell from the decap shots you were on about in the previous thread, indeed if it was going to be debug functionality or some canned feature then that would be more obvious from that.

Functionality nerfed for final release... I doubt it if it was custom from the ground up but dropped features exist. In before TV out or high end debugger discovered.

Most of the time things are grounded to a) provide less resistance to ground (you divide by area after all, and the length aspect can also be a thing here, never mind inductance/impedance considerations for signals), b) possibly serve as different ground in the case of things that don't tolerate noise so well/risk some kind of backfeeding, and c) to avoiding floating voltages. You also get the multiple people maybe working on their own individual element of the die so they kick things over to ground for their little bit to maybe make the job of the integrator that bit easier.

I equally doubt you will get much in the way of hard answers around here -- Nintendo engineers have not really given any talks that I have seen, never mind to that level, so unless it was in the gigaleak or something (and even then you might have to infer from different design stages or what passives surround the CPU where the pins meet the board or any kind of matched impedance traces*) nobody is really going to have heard anything or stopped to consider for terribly long.

*those playing along at home if you pull apart a device and see traces be all wiggly line rather than straight or as near to it as they can do then this is usually going to be matched impedance -- signal delay and strength is good, but if they arrive at different times (incoherent in the more literal definition) you are really troubled where boosting is usually doable enough.

At that point we are waiting on people playing at
, possibly https://trixter.oldskool.org/2015/04/07/8088-mph-we-break-all-your-emulators/
That level or maybe going in for the high end FGPA... I guess it would be simulation at that point. A GBA would probably be among the things I consider that more probable to happen to but I am not aware of such things just yet (the few FPGA things I have seen being more taken from instruction level than recreation level).
 

Site & Scene News

Popular threads in this forum

General chit-chat
Help Users
  • No one is chatting at the moment.
    SylverReZ @ SylverReZ: @K3Nv2, https://www.youtube.com/watch?v=BLdXfFsgvkE