PS3 Frankenstein PHAT PS3: CECHA with 40nm RSX

[vyktormvmpay25]

Yes that is more clear now RIP-Felix should wait until another test is done with another cok001 before any reflow/reball on it.
Today tested few more cd ps2. All going on 576p on board with 40nm and modchip cok002.
On unit from costumer with cok002 with his 90nm rsx most went to 720p.
Can't jump right now to conclude but will do more after another cok002 with 65nm will be done.
It could be better compatibility with 65 because dia002 has same hdmi ic and 65 nm rsx, cok002 use same hdmi ic.
Between 90nm and 40nm is something about upscale image to 720p into hdmi ic. Well 40nm give command to another type of hdmi ic. Just my theory test and probably DeadEnd unit working fine.

 

[Pacorretaco]

Yes that is more clear now RIP-Felix should wait until another test is done with another cok001 before any reflow/reball on it.
Today tested few more cd ps2. All going on 576p on board with 40nm and modchip cok002.
On unit from costumer with cok002 with his 90nm rsx most went to 720p.
Can't jump right now to conclude but will do more after another cok002 with 65nm will be done.
It could be better compatibility with 65 because dia002 has same hdmi ic and 65 nm rsx, cok002 use same hdmi ic.
Between 90nm and 40nm is something about upscale image to 720p into hdmi ic. Well 40nm give command to another type of hdmi ic. Just my theory test and probably DeadEnd unit working fine.

Hmm, so after more tests you are still having problems with COK-002 + 40nm RSX as well?
Ps2 games... Enhanced resolution doesn't work hmm. At least they do work if you disable enhanced resolution.
What about "smoothing"?

Haha, people went crazy and all modchip out of stock, and now we can't even make sure things are working... Hopefully it's just coincidences

 

[vyktormvmpay25]

Smoothing is the only option on PS menu. The ofter unit 90nm will have Screen resolution full screen or normal.
On 40nm only one option is Smoothing. Won't make any difference.
Anyway 90nm unit was given to costumer.

 

[Pacorretaco]

Smoothing is the only option on PS menu. The ofter unit 90nm will have Screen resolution full screen or normal.
On 40nm only one option is Smoothing. Won't make any difference.

Hmm, wait.
And what happens if you go in XMB settings > game settings > ps2 enhance resolution ON

Maybe we are getting confused early

 

[vyktormvmpay25]

It was off on both, I'm noob with those . Is fine now.
So is fine with cok002 and 40nm rsx. We can assume now one more thing about cok001.
We need to test one of this board here.

Now need to figure why so much difference between cpu and rsx and that mosfet exchange.

 

[Pacorretaco]

It was off on both, I'm noob with those . Is fine now.
So is fine with cok002 and 40nm rsx. We can assume now one more thing about cok001.
We need to test one of this board here.

Now need to figure why so much difference between cpu and rsx and that mosfet exchange.

Haha, good good. Now all you need to do is install my fan curves and finally you can close the top plastic cover?

Oh, I forgot about mosfet. Well that difference in temperature is pretty normal. Remember new RSX is working with less than half the Watts as before, but same heatsink. And CPU is still main power hog at over 100w.

Next thing is confirm COK 001 yes. Hopefully somebody who still has modchips can test. Also I'm sure soon enough they will become available again, probably more than before.

 

[vyktormvmpay25]

This unit will always be here as debugging unit. It's good to have always one around for test.

 

[RIP-Felix]

A few more guesses at what CGReset could mean.

• Cell + Graphics Synthesizer reset? On the EE+GS there are /ERST and /GRST pins on the chip. And there is an incoming signal for both. I'm guessing it's to synchronize their startup. Together it's EGRST. On the South Bridge there is a SBReset too. If C refers to Cell and G refers to the Graphics Synthesizer, then perhaps CG reset syncronizes the System CPU with the PS2 HW. CGReset and RSXReset are linked. So this one signal would synchronize all of them at the same time. Perhaps it's needed for PS2 HW synchronization. IDK, this is me looking for patterns that explain a pet hypothesis. It's most likely confirmation bias and CG reset means something totally different.

• Clock Generator Reset? That's just a random combination of C+G that popped into my head.

 

[froggy8]

have just read part of this thread, interesting!

if i still liked the phat version then maybe i would give this a go but i really like my slim now.

 

[DeadEnd]

This post is a review of the software modifications made by squeept in the second attempt of the frankenstein, is also a continuation of the recap written by @RIP-Felix here

After thinking a bit in it... i realized the best way to discuss all this details is if some of you gets used to the software procedures, we need at least 4 or 5 persons used to it, also some of the steps are optional, is like a cook recipe which ingredients needs to be discussed so please dedicate some time to this explanation and practise a bit with the file samples im going to upload in this post
First download this samples ---> https://workupload.com/file/H9RJAPHxpy6

And take a read at the README.TXT ... is like a tutorial for people not used to hexeditors, i suggest to download HxD and follow the steps "like a robot" to generate some files and to see by yourself what squeept did

Also, is included this image that is a resume in itself, this is what we need to do
As you can see there are 3 steps that are optional, squeept did 2 of the 3 optional steps (missed one of them... but we dont know if that one was important)

In resume... the only things that can be considered controversial in the "2nd_try_modified_by_squeept.bin" was the invalid thermal config region, and 1 of the options missing

Please @RIP-Felix @dead-end @vyktormvmpay25 @DoublesAdvocate try to complete the exercise at the bottom of the README.TXT

I managed to squeeze in a moment to return to the experiments.

I think, because originally I was attempting them a while ago. I am still going to need a few clarifications on the matter.

The examples which you have provided only involve modifying syscon EEPROM. But If I remember correctly, there was talk about needing to dump Cok002 syscon firmware completely? And to do that, we needed to apply patches to syscon EEPROM in order to get it to dump syscon FW over UART? This is what I have been trying to do.

 

[DeadEnd]

Well, the syscon swap was done correctly this time (I believe), a cok-002 eeprom was patched and flashed to a new syscon with M4j0r's help . The system has a 65nm on board with a diagonal 10k resistor. But I get errors 4002 and 3034, so typical to a bad solder joint. Perhaps more hardware mods are needed? Gonna try to add 40k resistor.

 

[RIP-Felix]

3034/4xxx was normal if the modchip is installed incorrectly, so it probably has to do with that resistor mod. Unless you get the software side of things correct, the console should error with a 3034.

Remember Ohm's law. V = IR. The higher that resistance the higher the voltage. I'm concerned you may fry something if you pull voltage too high! So I wouldn't recommend adding a 40K resister. We know 10K works, which was why I was comfortable doing that, but unless someone pulls the resistor off of an official SONY refirb and confirms it's value is higher, I wouldn't chance it.

I suspect there's something in the software that we haven't found yet.

 

[DeadEnd]

3034/4xxx was normal if the modchip is installed incorrectly, so it probably has to do with that resistor mod. Unless you get the software side of things correct, the console should error with a 3034.

Remember Ohm's law. V = IR. The higher that resistance the higher the voltage. I'm concerned you may fry something if you pull voltage too high! So I wouldn't recommend adding a 40K resister. We know 10K works, which was why I was comfortable doing that, but unless someone pulls the resistor off of an official SONY refirb and confirms it's value is higher, I wouldn't chance it.

I suspect there's something in the software that we haven't found yet.

Oh apologies, the value was actually 47k. The additional 47k resistor is needed for the modchip (on cok-002). Just to be clear, I'm not talking about replacing 10k with a 47k. Though you do have a point, as someone here with a 65 refurb posted the pics and they didnt' have a 47k placed.

 

[DeadEnd]

My only other suspicion is the capacitors on top of RSX, which I had knocked off earlier. I've tried my best to solder them back on, but possibly some of them ,uhm, switched places and one is missing. I don't remember now because it was still a practice board and my RSX lifting technique was far from perfect.

Either ways, this test is not conclusive. The board has been through a lot before, although it was untouched when I got it, I feel like I may have abused it a bit during practicing.

 

[Pacorretaco]

Well, the syscon swap was done correctly this time (I believe), a cok-002 eeprom was patched and flashed to a new syscon with M4j0r's help . The system has a 65nm on board with a diagonal 10k resistor. But I get errors 4002 and 3034, so typical to a bad solder joint. Perhaps more hardware mods are needed? Gonna try to add 40k resistor.

I think this is the same brick wall that squeept was hitting.

If my hypothesis is, let's say there is something wrong with this method and 4002 means some kind of "unsupported RSX" error.
What if you try to solder a modchip there right now, see what happens?
If it works it would confirm this and save many headaches.

But I get errors 4002 and 3034, so typical to a bad solder joint.

Because we know better than this. Those errors just mean failed CPURSX bittraining. Yes they will appear if there's bad soldering under RSX (Or CPU). But the bittraing can fail for many more reasons, even with perfect soldering.
Sounds like old story hehe.

Btw I notice you edited your comment, before you were talking about standby triple beep or something.
Probably bad data inside SYSCON or something like that? Human error while writing?
At least it looks like you solved it. If you are getting to bittraining, I guess is because the Syscon "remarry" was successful.

Cheers

 

[Pacorretaco]

Oh apologies, the value was actually 47k. The additional 47k resistor is needed for the modchip (on cok-002). Just to be clear, I'm not talking about replacing 10k with a 47k. Though you do have a point, as someone here with a 65 refurb posted the pics and they didnt' have a 47k placed.

View attachment 34387

Good, now we have the bringup that @squeept probably had but didn't show!

Interestingly it mentions something like:
Error RSX FlexIO ID...

Which sounds a lot like what I was theorizing. Could "RSX FlexIO_ID" be the proper name of the thing that the modchip is intercepting?
The "identification" of the RSX, that is not recognized and maybe this is the "only" problem?

Basically the thing that we want to overcome in this whole frankenstein endeavor hehehe

 

[DeadEnd]

I'll be honest, this board is very difficult to rely on. Before error 3034 I had temperature sensor errors 2030, 2031, 2033. Ironically, I got them back again. It could be because we were experimenting with changing fan tables. The last one I flashed is from the refurbished console. But this board additionally has a bulged spot from my earlier mistakes with the top heater. So yeah, it could be causing some unknown issues as well.

I don't have a modchip yet to come to any proper conclusion, but it should arrive next week. So I'll check it out then.

Yeah, we know better than this about the bittraining errors . I just wanted to quickly post an update.

Edit. Got the 4002 and 3034 errors now again after flashing dia002 fantables. Bringup works now too. With other fantables I can't use bringup command because syscon "locks up" and starts blinking red around 2 seconds after plugging it into the wall. Power on button also doesnt respond. That is when errors 2030, 2031 and 2033 show up.

 

[M4j0r]

If my hypothesis is, let's say there is something wrong with this method and 4002 means some kind of "unsupported RSX" error.

Well, we know that the new Syscon supports 65nm RSX, it's from the same board.
If you change the RSX value in the Syscon EEPROM to something it doesn't support, it won't even start the bit training, it'll return error 3000 instead. The value has been set to 0x21, which means 65nm and there's no error 3000, it actually tries to do something.
I mean the actual RSX software part is straightfoward, it's just one byte.

Interestingly it mentions something like:
Error RSX FlexIO ID...

Which sounds a lot like what I was theorizing. Could "RSX FlexIO_ID" be the proper name of the thing that the modchip is intercepting?
The "identification" of the RSX, that is not recognized and maybe this is the "only" problem?

The ID command is the first FlexIO command, so if it can't communicate at all it'll show that error (https://pastebin.com/QZAqHT1u), especially if it shows up right after the SB link speed setting.

I'll be honest, this board is very difficult to rely on. Before error 3034 I had temperature sensor errors 2030, 2031, 2033. Ironically, I got them back again. It could be because we were experimenting with changing fan tables. The last one I flashed is from the refurbished console. But this board additionally has a bulged spot from my earlier mistakes with the top heater.
Edit. Got the 4002 and 3034 errors now again after flashing dia002 fantables. Bringup works now too. With other fantables I can't use bringup command because syscon "locks up" and starts blinking red around 2 seconds after plugging it into the wall. Power on button also doesnt respond. That is when errors 2030, 2031 and 2033 show up.

It's really weird the neither the refurbished COK-001 thermal config nor the one for 65nm refurbished COK-002 by @sandungas works, but at least the original DIA-002 config works. Maybe that's why they had to introduce the 301GB.

 

[RIP-Felix]

@botakompong I finally got around to installing your mod chip on a COK-001 with a 40nm CDX5301. It worked, but PS2 games don't load. It just sits there on a black screen, like a GLOD, except I can press the PS button to bring up the menu and return to the XMB! If I choose the reset game option in the menu it retries. If I keep doing that it will eventually YLOD with error 80 1701 (BE: Attention):

>$
[SERV DEVPM] CONTROL_PCI_BUS_POWER_STATE CMD
[SERV NVS] READ CMD
[SERV NVS] READ CMD
[SERV DEVPM] CONTROL_PCI_BUS_POWER_STATE CMD
[SSM] Attention BE : Detected !
[SSM] BE Attention signal is detected !!
[SSM] state: 0400 -> 0700
[POWSEQ] AV Backend Letup
[SSM] ssmCb_AfterBeOn() called.
[SSM] Shutdown mode : syspm_stat=00000000/00000000
Wait WmMcCom_DeadEvent timeout
[ERROR]: 0xa0801701
[POWSEQ] PowerSeq_Letup called.
[SSM] state: 0700 -> 0600
(PowerOff State) (Fatal)

[mullion]$
>$

Everything else works. PS3/PS1, no problem. It's like PS2 is being held in reset!? Have you ever seen that on a COK-001?

So far as I know I'm the first person here to install the mod chip on a COK-001, so I have no idea what to expect. I do not know if the PS2 functionality was working before the install, as the console was previously having a YLOD (3034 only). I figured it needed a rebal and decided to attempt the RSX swap since I needed to reball anyway.

I have not attempted to verify all the PS2 Voltages yet. I was hoping someone else would attempt the mod to see if the same thing happens to them or if it's just me. If it is just me, then I'll troubleshoot the PS2 hardware, but if it's not then it's a compatibility issue with the mod.

Please note, I did the voltage mod as well, replacing Q6200 with one from a KTE-001, and bridging just as you showed. Here are some pics...

Spoiler: top

Spoiler: Btm

 

[botakompong]

@botakompong I finally got around to installing your mod chip on a COK-001 with a 40nm CDX5301. It worked, but PS2 games don't load. It just sits there on a black screen, like a GLOD, except I can press the PS button to bring up the menu and return to the XMB! If I choose the reset game option in the menu it retries. If I keep doing that it will eventually YLOD with error 80 1701 (BE: Attention):

>$
[SERV DEVPM] CONTROL_PCI_BUS_POWER_STATE CMD
[SERV NVS] READ CMD
[SERV NVS] READ CMD
[SERV DEVPM] CONTROL_PCI_BUS_POWER_STATE CMD
[SSM] Attention BE : Detected !
[SSM] BE Attention signal is detected !!
[SSM] state: 0400 -> 0700
[POWSEQ] AV Backend Letup
[SSM] ssmCb_AfterBeOn() called.
[SSM] Shutdown mode : syspm_stat=00000000/00000000
Wait WmMcCom_DeadEvent timeout
[ERROR]: 0xa0801701
[POWSEQ] PowerSeq_Letup called.
[SSM] state: 0700 -> 0600
(PowerOff State) (Fatal)

[mullion]$
>$

Everything else works. PS3/PS1, no problem. It's like PS2 is being held in reset!? Have you ever seen that on a COK-001?

So far as I know I'm the first person here to install the mod chip on a COK-001, so I have no idea what to expect. I do not know if the PS2 functionality was working before the install, as the console was previously having a YLOD (3034 only). I figured it needed a rebal and decided to attempt the RSX swap since I needed to reball anyway.

I have not attempted to verify all the PS2 Voltages yet. I was hoping someone else would attempt the mod to see if the same thing happens to them or if it's just me. If it is just me, then I'll troubleshoot the PS2 hardware, but if it's not then it's a compatibility issue with the mod.

Please note, I did the voltage mod as well, replacing Q6200 with one from a KTE-001, and bridging just as you showed. Here are some pics...

Spoiler: top

Spoiler: Btm

Hello friends, well I've heard from DeathEnd, I just found out there's a problem with the PS2, unfortunately I haven't been able to try it (because of isolation), when the shop opens (August 9th) I'll look for the problem. So far, no one (Indonesian) has talked about the PS2, so I just found out.
Little explanation:
One of the advantages of the cok-001 and cok-002 motherboards is that there is additional PS2 hardware embedded in the motherboard, with such conditions you can say that PS2 & PS3 each work independently, what distinguishes cok-001 from cok-002 is the AV output line, cok -002 there is no problem on the ps2 because the AV output line is directly from the ps2 hardware itself (as far as I remember), while for cok-001 the ps2 hardware AV output line is on the RSX AV line, so when we replace the RSX, there may be something wrong with it. The AV output line, could also be due to interference from the AV output of the RSX itself, as you have experienced.
In the past, there was also a problem with the installation of the cok-001, where after replacing the RSX, the PS3 (GLOD) output image did not come out, after exploring, we finally found the solution by adding 47K
see picture.


I'm sorry for not paying attention to PS2 (cok-001), I'll find the cause later, tq