Worklog N64-AIO

Gman

RTFDS
Staff member
.
.
.
.
.
Joined
Feb 25, 2016
Messages
1,434
Likes
2,871
N64 AIO (All-IN-ONE)

1602089838646.png


20201003_133415.jpg


The 3X3 core is confirmed working. Now it's time to kick it up a notch.

n64_AIO.png


This here is N64 core + PMS + digital audio amp. FFC connectors for the cartslot and digital video (for ultraVGA/HDMI or N64advanced)
 

Shank

Moderator
Staff member
.
.
Joined
Jan 31, 2016
Messages
1,286
Likes
2,723
Portables
6

CrashBash

No emulation, real hardware
Staff member
.
.
.
Joined
Feb 5, 2016
Messages
129
Likes
301
Location
MGC
I'm going to be looking forward to see everything that comes as a result of this. Seeing everything kind of come full circle from ye old N64 days back when I lurked is really interesting.
 

StonedEdge

a.k.a. ClonedEdge
.
.
Joined
Nov 16, 2018
Messages
362
Likes
1,482
Location
Japan, Tokyo
Portables
2
2xsl8b.jpg
 

MRKane

.
2021 3rd Place Winner
Joined
Nov 5, 2017
Messages
405
Likes
488
I've been pottering on this ever since talking with some of the Goldeneye old-hands about it years ago (ShootersForever) so am frankly thrilled to see that you're taking this on as a project as it's always been beyond my financial means. I'm keen to see your development here and can help with brotherly hugs and the odd cup of coffee even!
 

MRKane

.
2021 3rd Place Winner
Joined
Nov 5, 2017
Messages
405
Likes
488
I've got a "nuts and bolts" question:

Ram line length. How did you get this to work as it appears shorter than the original lengths and this was something that scared me away from it in the past. I do know that xtal overclocks (on the RCP) lead to image glitching and the going theory had to do with ram read times. This lead us down a path of swapping out the RD ram for stuff with lower latency and that's where I really sunk my teeth into it. I'm very interested to hear how the length difference affects things, and if this would open up room for a system optimisation somehow.
 

Gman

RTFDS
Staff member
.
.
.
.
.
Joined
Feb 25, 2016
Messages
1,434
Likes
2,871
I've got a "nuts and bolts" question:

Ram line length. How did you get this to work as it appears shorter than the original lengths and this was something that scared me away from it in the past. I do know that xtal overclocks (on the RCP) lead to image glitching and the going theory had to do with ram read times. This lead us down a path of swapping out the RD ram for stuff with lower latency and that's where I really sunk my teeth into it. I'm very interested to hear how the length difference affects things, and if this would open up room for a system optimisation somehow.
The goal of the rambus channel layout is to meet a 50 ohm impedance. According to some simple calculations mine is a bit more than that actually. It's determined by trace thickness, height, and distance between parallel traces but their could be many more sophisticated factors. But to answer your question: pray.
 

Gman

RTFDS
Staff member
.
.
.
.
.
Joined
Feb 25, 2016
Messages
1,434
Likes
2,871
What a day it's been. I assembled the AIO and not even five minutes after building it I accidentally fried the RCP (didn't see tiny solder ball shorting the cartridge ffc and then accidentally put the cartridge in the wrong way, guess it didn't like that). It's not the easiest to assemble with large QFPs, 0402 components, and double-sided component assembly. I was able to swap a new one on though I ripped a few pads (hence the jumpers). So grateful it's working after all that.

1603166976305.png

Second mistake I made was schematic level. I misread the symbol and swapped SDA/SCL from D1 and D2 of the dual mosfet for the I2C logic level converter for controlling the audio amp via the PMS MCU. I was able to lift the pins and rewire it correctly fortunately. I'll fix this in the next PCB.

The new cartridge daughter board allows the AIO to sit directly ontop of the cartridge, this will be the ideal placement for a portable.

20201019_230540.jpg

20201020_000352.jpg
 

CrazyGadget

Crazy Helpful, CrazyGadget
.
Joined
Jun 6, 2020
Messages
600
Likes
832
Portables
???
Holy guacamole, this is really coming together, and quite fast at that! I'm very much looking forward to seeing what kind of portable this gets used in when the time comes! Phenomenal work as always, Gman; you never cease to amaze me!
 

MRKane

.
2021 3rd Place Winner
Joined
Nov 5, 2017
Messages
405
Likes
488
Oh the money you could have saved Nintendo if you'd one this 25 years ago!
 
Joined
Apr 12, 2020
Messages
128
Likes
242
From seeing that bare custom mobo to being fully populated... what a sight to behold. There's no stopping the gman!
 
Joined
Jan 18, 2019
Messages
26
Likes
49
I can see I've logged back into Bit Built for my annual winter N64 tinkering at the perfect time!!

How many monies would you like for one of these once they're ready? (I don't actually care how much, I just need to know how much to tuck away! Haha)
 

E64

.
Joined
Nov 15, 2016
Messages
125
Likes
99
Portables
2
What a day it's been. I assembled the AIO and not even five minutes after building it I accidentally fried the RCP (didn't see tiny solder ball shorting the cartridge ffc and then accidentally put the cartridge in the wrong way, guess it didn't like that). It's not the easiest to assemble with large QFPs, 0402 components, and double-sided component assembly. I was able to swap a new one on though I ripped a few pads (hence the jumpers). So grateful it's working after all that.

View attachment 14672
Second mistake I made was schematic level. I misread the symbol and swapped SDA/SCL from D1 and D2 of the dual mosfet for the I2C logic level converter for controlling the audio amp via the PMS MCU. I was able to lift the pins and rewire it correctly fortunately. I'll fix this in the next PCB.

The new cartridge daughter board allows the AIO to sit directly ontop of the cartridge, this will be the ideal placement for a portable.

View attachment 14669
View attachment 14670
4C49D933-9F24-439F-B1EE-C39F20E48BC0.gif
 
Top