PS3 Frankenstein PHAT PS3: CECHA with 40nm RSX

Yeah. About syscon swaps. It is actually going to be a bit more complex than just patching eeprom. Syscon firmware must be changed as well. So modchip is still going to be the simplest solution for a new RSX.
Dont forget the syscon patches, the official format allows to apply patches at different offsets (up to 4 as far i remember). I dont know how long can be that patches but lets say 0x40 bytes each
When a retail syscon boots applyes the patch data into RAM (virtually) on top of the syscon base firmware. This is the official way (and the only) to update the syscon firmware in a retail syscon
The goal of how this update procedure is designed is to keep the base firmware intact as a fallback, it allows to return to the original syscon firmware by deleting the patch

Ok, so... if we can patch the base syscon firmware at 4 different offsets in theory we could use them to patch the RSX bittraining data or whatever function from the syscon firmware is related with it
Keep in mind also this official patch format im talking about only is handy to patch the non-rewritable areas. The frankenstein mod requires other changes (the byte that indicates the RSX version, or modifying the thermal config) that doesnt needs to use this official patch format, this ones can be considered "settings" and can be modifyed by the standard UART access
 
Last edited:
I don't know all the deeper details. I was talking about swapping syscon with the F version (flashable variant). You guys speak in a developer language, which my simple mind can't fathom. :D For now, all I have learned is that patching the original eeprom and reflashing it to the new syscon is not enough. We must also flash the correct firmware, but it cannot be done through bus pirate. It is done through UART and a script that currently only works in Linux. @vyktormvmpay25 has tried to use the script in Windows and that didn't work. (thanks to @M4j0r for being patient enough to explain it in simple terms) He will eventually make a working script though, so just be patient.

Honestly, this whole ps3 modding is becoming a bit tedious. It's such a niche considering that only a few retrogamers care enough about it. I've been considering shifting my attention elsewhere...
 
Let's all wait M4j0r to receive and see it. I understand that I may skip something and I will let exxperts to work on programming/software side. I'm trying to understand but I can't skip important steps in learning, like if I didn't start from basics I won't be able to understand well some steps and I'll probably fail so let's all wait and working all together I think is possible to achieve something.
 
hi guys
got my self ps3 fat with rsx swapped by @botakompong itself but seems like the ps2 EE chip is broken ? i just get black screen when playing every ps2 game but ps2 menu works (hold ps button on controller ) it just black screen even though im swap back to ofw

btw if anyone gonna need orbis chip or cheap rsx from botakompong it self pm me i help you guys out
 
i think ps3 ee/gs is same as slim ps2 7500x. but solderballs must added manualy no stensil for that. is bga chip. but ive never see a ps2 that has damage on either ee/gs or ram. also is the 2nd ps3 with breaked ps2 compatibility. Also i dont remember if i have mention it but some ps2 games dont like the upscaling to be turned on try with upscaling off. This has nothing to to with the frankenstein mod. I had read an old treat about it
 
Last edited:
I don't know all the deeper details. I was talking about swapping syscon with the F version (flashable variant). You guys speak in a developer language, which my simple mind can't fathom. :D For now, all I have learned is that patching the original eeprom and reflashing it to the new syscon is not enough. We must also flash the correct firmware, but it cannot be done through bus pirate. It is done through UART and a script that currently only works in Linux. @vyktormvmpay25 has tried to use the script in Windows and that didn't work. (thanks to @M4j0r for being patient enough to explain it in simple terms) He will eventually make a working script though, so just be patient.

Honestly, this whole ps3 modding is becoming a bit tedious. It's such a niche considering that only a few retrogamers care enough about it. I've been considering shifting my attention elsewhere...
Im not sure if you are used to the official syscon patches but as an introduction for everyone else...
Insdide the PUP (PS3 firmware installers) there are always some PKG's intended to update the firmwares inside Bluray controller, the wifi/BT module, and syscon
Lets say... when we install a PS3 firmware we are also installing firmwares for this 3 components
There are only 8 official firmware updates for syscon (and 3 of them was superceeded)... so the modern PUP files only have 5
https://www.psdevwiki.com/ps3/Syscon_Firmware#Syscon_patches

At an intermediate point of the PS3 firmware instalaltion, the installer does a check to syscon to find his "revision" (also named SoftID... is a software identifyer but it can be used as a hardware identifyer too) and his "version" (the syscon firmware version). The syscon patch is installed only if the SoftID matches and the version of the patch is newer

The keys needed to build that official patches are public, so now we can build a "custom syscon firmware patch"... using SoftID 0B8E (for the syscon CXR713120-201GB shipped from factory in the motherboards COK-001) and with a version of 6.66 (it needs to be bigger than the officials)
Then we include this patch renamed to SYS_CON_FIRMWARE_06006006.pkg inside a PUP... install the PUP like if it was a standard custom firmware and our custom patch will be installed inside syscon EEPROM
The result is we can "update" the syscon firmware with custom code
Sure... the amount of code we can update is limited, in size, in the amount of areas we can patch (separated from each others), and how tricky are the functions we are trying to patch

This procedure could be used for any other patches, im not telling the problem of the RSX bittraining is something easy to achieve, the last time i was talking with m4j0r about this he said there is no room in the official patch formwat to "update" all the code related to RSX bittraining... but you know... as usually with hacking it could happen that eventually he or someone gets the illumination from the sky and finds how to do it in a unconventional way :D

Anyway... this trick to apply custom syscon patches using the official format is something handy for many other hacks, if there are some features that can be unlocked just by patching a few bytes it should work

I don't even know more of that EE side, we could swap that? Or is kind tied in wierd way?
Is not logically married, you can replace it without worryes
I would say before reballing or replacing it is important to take a look at the other components dependant of it, it have a couple of RAM chips (16mb each), doesnt seems to be much stressed by temperatures or workloads but the manufacturing technology looks a bit old, maybe are prone to failure
There is also another misterious component that worths to be checked https://www.psdevwiki.com/ps3/CXD9208GP
This one is related with video output, so i guess is going to be tricky to check because it have a lot of lines with video signals
 
i dont think there is a way to patch syscon even if it meens 2 different firmwares one for 65nm rsx and one for 40nm rsx. why sony bother to swap syscon with newer if there is a way to swap rsx and keep the same syscon? although i would like to be wrong
 
i dont think there is a way to patch syscon even if it meens 2 different firmwares one for 65nm rsx and one for 40nm rsx. why sony bother to swap syscon with newer if there is a way to swap rsx and keep the same syscon? although i would like to be wrong

You misunderstood something. The latest test was about patching eeprom from original syscon, then copyin it to the flashable syscon and also flashing correct firmware to it. CXR713F120A https://www.psdevwiki.com/ps3/CXR713F120A .You can't patch the original syscon to be compatible with 40nm. You need another chip (like the one I just linked). Although I admit I didn't read sandugas's text.
 
Last edited:
oops youre right. then modchip is the way to go. more bga work much expensive right?

I don't believe there will be any easier way than the modchip. Indeed it requires less steps and no need for bus pirate, reflashing or bga soldering. The only issue with it is ordering from Indonesia isn't very convenient. The person on Fiverr hasn't been responsive for our last orders, so if you have nobody to forward it through then it's tricky to buy from there. Plus customs declaration adds an extra step.
 
@feel2death I could try fix this issue for you. Well if you keep it until my next order and send it all together, return will going to be bit harder but not impossible. You don't pay for this fix, return only fees, this is experimental test. Just if you look to fix it keep it for my next order. If someone else wants to try just do what you consider.
I also have one incoming with this problem, ps3 working and ps2 failing but that is on original 90nm rsx so is pretty sure is board issues. I'll try 40nm rsx on that but don't think will help. So much more work with those rare cok001 boards :)
 
Last edited:
Every month I get asked about a cok001 board, often from USA. I also don't want to work on them, mostly because I'm not yet confident in my equipment (especially after Trustmaster's board issues). Additionally, a lot of people don't realize that heatgunned boards are not worth repairing. Even the untouched boards could exhibit all kinds of extra issues just like described above; ps2 games sometimes failing to work, difficulties soldering rsx because of balls merging or extra errors that aren't common in other boards. It's a very sensitive model and yet everybody wants it. Plus think how tricky it is to work on them for anybody in EU due to the lack of spares.

Just a reminder, CECHC/CECHE may not be perfect for emulating every single ps2 game, but it sure does upscale them noticeably better than CECHA.

 
Last edited:
so if ps2 issues hapen also on 90nm then something is really wrong. but before calling it a faulty do this
1 reset video settings i think is holding power button untill 2 beeps or manualy set them again in settings
2 try with upscaling off. monitor must support low resolutions 480p ntsc games 575p pal games
 
2 try with upscaling off. monitor must support low resolutions 480p ntsc games 575p pal games

I think you should be able to run the supported games regardless of the resolution or upscaling option. At least that's how it goes on the working models.
 
@feel2death I could try fix this issue for you. Well if you keep it until my next order and send it all together, return will going to be bit harder but not impossible. You don't pay for this fix, return only fees, this is experimental test. Just if you look to fix it keep it for my next order. If someone else wants to try just do what you consider.
I also have one incoming with this problem, ps3 working and ps2 failing but that is on original 90nm rsx so is pretty sure is board issues. I'll try 40nm rsx on that but don't think will help. So much more work with those rare cok001 boards :)

yeah maybe next time actually im gonna try snatch ps3 metal gear solid 4 edition if my board doesnt work kinda cool and rare to have

[QUOTE="so if ps2 issues hapen also on 90nm then something is really wrong. but before calling it a faulty do this
1 reset video settings i think is holding power button untill 2 beeps or manualy set them again in settings
2 try with upscaling off. monitor must support low resolutions 480p ntsc games 575p pal games[/QUOTE]

my tv seems support lower resolution since when im open emulator menu the resolution already lower to 480p but the game doesnt run only black screen

WhatsApp%20Image%202022-02-08%20at%2012.35.33.jpeg


still reminder if anyone need rsx or orbis (fresh from botakompong itself :v ) from indonesia just ask me in pm
 
Back
Top