PS3 CECHA PS3 YLOD Saga

shinichi999 said:
Thanks, I understand the issue a little bit now. I have had this console since 1 year and everything runs flawlessly for now. Of course, I don't pretend to force the console (I played TLOU only this time just to test the stability of the console), since I mainly use it to play PS2 games taking advantage of the backward compatibility, so I really hope to have several years ahead with the console working perfectly.
Click to expand...

For curiosity sakes and for science of course, ill grab a copy of The Last Of Us (unpatched), as many reports for the shutdown suggested the game was 2013 unpatched, ill digg some tests, maybe playing 30/45min and report back, that should conclude this topic once and for all :)

Ill do some research on what the most hotpoints of shutdowns are, some say its in the intro area,others in the downtown area building near the first clicker. Ill take these as measurements and see if the machine will shutdown.
 
normally TLOU crash on downtown area when the clicker attacks or after that climb the wall downtown area is an good way to test.
if some oneuse an save from other user on rebug frimware and go to downtown area when climb up the wall game bugs lol
 
Ok, so after 10 minutes on the intro scene, the PS3 decides to shutdown, its still unconclusive data, as remember, 7 of my 8 NEC/TOKINs are still original, which lead me to believe, that intensive games like The Last Of Us, require more juice through the RSX, and replacing a few NECs with Tantalums will restore the PS3 power cycle,but it wont hold against more intensive games, which needs a full 100% tantalum replacement.

Temps wise, i was getting 71ºC on the RSX, and around 58ºC on the CELL, so the machine was definitely not overheating, just to end the topic, i will try it again but this time at an abysmal 80% fan speed, but i doubt its going to change anything.

Update 2 - Same intro, same results with 85% fan-speed (Sony Airlines), but the results came the same, so overheating its out of the question, CELL: 49ºC - RSX: 51ºC

Conclusion: If you are planning to play heavy intensive 2013 titles, you will have to replace all the 8 NEC/TOKIN capacitors,or else insufficient voltage supply to the CELL and RSX leads to unexpected shutdowns.
 
Last edited:
I made lot of temperature tests in the last of us years ago, actually i played the game entirelly monitoring the temperatures and im very confident there are 2 points of the game where temperature increases brutally (in comparison with the rest of the game)
That 2 points are the only ones of the game where there is water

The first place with water in the last of us is the lake that can be seen in this video:
https://www.psx-place.com/threads/benchmarking-or-stress-test-for-ps3.12690/#post-72784
sandungas said:
I think the test kozarovv suggested is easyer to do because in gta5 you can go to the sea at any point of your playthrought
In TLOU there are only a couple of areas with water, so you need a savegame before that point, or play the entire game just for the test, so is not handy

The area i mentioned is this one btw...
Click to expand...

The second point of the game where there is water is near the ending, when you are entering some kind of hospital (or scientific complex, dont remember well) from the sewers
You know... that sewers are like a river and there is lot of water moving

As a curious detail... that sewers at the ending (with lot of graphic load because there is lot of water moving) had pretty much the same temperature than the other point i mentioned (the lake on the road very calmy, the water doesnt moves in it)

So the conclusion is naughty dog game engine "suffers" a lot with water, and there is no need for the water to move, is just the presence of water what pushes the temperatures to the max
Also... in the point of the video... that lake in the street... you are going to notice how the temperature increases a lot even before you see the water (a couple of streets before reaching the lake), because the game is loading the water before you see it
At the ending happens the same effect... the temperatures increases a lot even before you enter in the sewers
 
Last edited:
umart666 said:
how did you monitor temps while playing?
Click to expand...
Is explained in the link i posted, i did it when the software fan control was not displaying anything in screen
I configured the speeds with a gap of 3ºC in between every step, and assigned very different fan speeds to them
This way i was able to know (by the noise level) in real time if the temperature was moving 3ºC up... or 3ºC down

Also, i played the entire game with low volume keeping attention at the fan noise :D
I completed the game 4 or 5 times btw, i decided to make this test with it because i knew the maps and what it was going to happen very well, this playthrought with low volume was some kind of speedrun for completionism and for science :D


Edit:
My point is there are parts of the game that have a temperature around 5ºC over other parts of the game
This is what i use to call the "temperature peaks" because happens very fast... there is people that doesnt belives it or doesnt gets it though... they thinks the temperature inside a game is mantained stable along the whole game, and thats not realistically at all
So... when someone tells... "i have X temperature inside Y game" for me doesnt means anything because i know the meassurement was not made in a accurate way, is just a random meassure taken at a random point of that game
 
Last edited:
umart666 said:
so there's no way to monitor temperatures while in game?
Click to expand...
Nowadays you can do it with the web interface of webman, either in a PC or android smartphone, this way you have the screen of the PC (or the screen of the phone) displaying the temperature at all times
And... i think it can be made also with CCAPI or similar

*Is needed to mention this software fancontrols are not displaying temperature on real time though.... they reads the temperature vaule (by sending a syscall) every X seconds
And as far i know X = 3 seconds
So... if there is a very fast temperature peak that goes up and down in less than 3 seconds... then is technically bypassing the custom code
 
sandungas said:
Nowadays you can do it with the web interface of webman, either in a PC or android smartphone, this way you have the screen of the PC (or the screen of the phone) displaying the temperature at all times
And... i think it can be made also with CCAPI or similar

*Is needed to mention this software fancontrols are not displaying temperature on real time though.... they reads the temperature vaule (by sending a syscall) every X seconds
And as far i know X = 3 seconds
So... if there is a very fast temperature peak that goes up and down in less than 3 seconds... then is technically bypassing the custom code
Click to expand...
One thing that crossed my mind its to try on a 65nm CECH-2000 Slim,with the original NECs, to see what kind of a behaviour will display.
 
Naked_Snake1995 said:
One thing that crossed my mind its to try on a 65nm CECH-2000 Slim,with the original NECs, to see what kind of a behaviour will display.
Click to expand...
Btw, the CECH-20xx have 4 tokins (instead of 8 like all the previous PS3 models) because the CECH-20xx have a:
CELL = 45nm
RSX = 65nm <--- with 4 tokins

In the next PS3 model CECH-21xx (without tokins):
CELL = 45nm
RSX = 40nm <--- they reduced size a lot

https://www.psdevwiki.com/ps3/Talk:SKU_Models#PS3_Slim
https://www.psdevwiki.com/ps3/Motherboard_Revisions
 
sandungas said:
Btw, the CECH-20xx have 4 tokins (instead of 8 like all the previous PS3 models) because the CECH-20xx have a:
CELL = 45nm
RSX = 65nm <--- with 4 tokins

In the next PS3 model CECH-21xx (without tokins):
CELL = 45nm
RSX = 40nm <--- they reduced size a lot

https://www.psdevwiki.com/ps3/Talk:SKU_Models#PS3_Slim
https://www.psdevwiki.com/ps3/Motherboard_Revisions
Click to expand...
Mate, thank you for the information but i already knew that [emoji23][emoji23]

But that's not my main goal, what i am trying to see its if by having the NECs on the RSX the PS3 would shutdown at some point, remember 65nm RSX Fatties are prone to this issue as well, so i wanted to see if it's on a RSX level or CELL, but so far no luck! [emoji848]

From what info i managed to gather, all the symphoms are affecting the Phat Models mainly the CECHA/B/C/E etc... didnt saw a single 2000 Slim on this list.

The only way to make sure its the NECs and not the CELL or the RSX, its testing with an unpackaged unit, never touched,but thats out of the question! [emoji23]

Sent from my G8341 using Tapatalk
 
Last edited:
Naked_Snake1995 said:
Mate, thank you for the information but i already knew that [emoji23][emoji23]
Click to expand...
I realized about it yesterday, previouslly i was not sure which motherboards had the infamous tokins :D

Naked_Snake1995 said:
But that's not my main goal, what i am trying to see its if by having the NECs on the RSX the PS3 would shutdown at some point, remember 65nm RSX Fatties are prone to this issue as well, so i wanted to see if it's on a RSX level or CELL, but so far no luck! [emoji848]

From what info i managed to gather, all the symphoms are affecting the Phat Models mainly the CECHA/B/C/E etc... didnt saw a single 2000 Slim on this list.

The only way to make sure its the NECs and not the CELL or the RSX, its testing with an unpackaged unit, never touched,but thats out of the question! [emoji23]

Sent from my G8341 using Tapatalk
Click to expand...
Is hard to know if is only related with CELL or RSX, and could be both actually
Also, another confusing thing is the failure is related with power lines, but the temperature is an indirect indicator of the power lines consumption, this is why i mentioned that there are 2 very well defined points in the last of us where temperature increases a lot

In that points... additionally to the temperature increase i mentioned... i bet the power lines consumption should be high too (higher than other points of the game)
Some of you have been mentioning that the game crashes at beginning (in the intro or in the first clicker... something like that)... and i bet is going to crash too in the 2 points i mentioned (where the game engine generates the water)

Dunno... initially i guess probably is more related with RSX... but the crash reports with the teken game goes a bit against this theory, because tekken doesnt have a huge graphic load

------------
In your PS3 (where you replaced only 1 tokin) it looks you improved it but is not completly fixed
My guess is you need to replace the other tokins next to the one you replaced (4 tokins in total for the RSX, and you only replaced 1). The other 3 are having some problem

As i mentioned somewhere the tokins works in groups, and there are 2 groups for CELL or RSX, and every group is composed by 4 tokins
So the good fix is to replace 4 tokins (the ones for RSX goes first).... or 8 tokins (for CELL too)

You are used to it so is not much dangerous because you know how to do it... is just is needed to solder a lot of tantalum capacitors so is a lot of solder job, but is worthy because this will solve the problem definitivelly and it prevents failures in the other tokins in long term
 
sandungas said:
I realized about it yesterday, previouslly i was not sure which motherboards had the infamous tokins :D


Is hard to know if is only related with CELL or RSX, and could be both actually
Also, another confusing thing is the failure is related with power lines, but the temperature is an indirect indicator of the power lines consumption, this is why i mentioned that there are 2 very well defined points in the last of us where temperature increases a lot

In that points... additionally to the temperature increase i mentioned... i bet the power lines consumption should be high too (higher than other points of the game)
Some of you have been mentioning that the game crashes at beginning (in the intro or in the first clicker... something like that)... and i bet is going to crash too in the 2 points i mentioned (where the game engine generates the water)

Dunno... initially i guess probably is more related with RSX... but the crash reports with the teken game goes a bit against this theory, because tekken doesnt have a huge graphic load

------------
In your PS3 (where you replaced only 1 tokin) it looks you improved it but is not completly fixed
My guess is you need to replace the other tokins next to the one you replaced (4 tokins in total for the RSX, and you only replaced 1). The other 3 are having some problem

As i mentioned somewhere the tokins works in groups, and there are 2 groups for CELL or RSX, and every group is composed by 4 tokins
So the good fix is to replace 4 tokins (the ones for RSX goes first).... or 8 tokins (for CELL too)

You are used to it so is not much dangerous because you know how to do it... is just is needed to solder a lot of tantalum capacitors so is a lot of solder job, but is worthy because this will solve the problem definitivelly and it prevents failures in the other tokins in long term
Click to expand...
Yes thats, true, right now its like having it with the original NECs, as far as i am aware THE LAST OF US and Gran Turismo 6 will only hold when all the NECs are replaced, but i am not too worried, as i barely use the C04 besides playing PlayStation2 Titles, so it doesnt bother me one bit, but for any other lads that only own one unit, full NEC replacement its advised.
 
Wow! I cannot believe this thread is still going :D

Been a while since I used my PS3, but been thinking of getting it out of its box again to play some classics, but the mere existence of the YLOD issue with Tekken really puts me off :(

Worst yet, is knowing nobody around my local area has any idea about tokin capacitor replacement or even wants to service PS3's anymore.
 
tldr of the thread is to use it and only suspect caps when having instability and don't put the ps3 in the oven. :)
 
Yes, this thread is back at the top again!

Hello everyone.

I have managed to locate a friend who is able to re-solder some new capacitors on my unit in the next few weeks. I'm very interested to see how this will go and if it will finally solve the problem of my unit and TTT2.

In the mean time, i have noticed that when i play some older games (Tekken 5 DR) there is some weird slow down effect during intro/outro poses of certain characters. This was not present as far as i can remember on this unit, not has it ever had such a problem on my launch Slim model 2000b.

I am wondering what would be the cause of the weird sudden slow down on certain games that was not there before? Temps on both chips are normal as they always were on my 60gb (65 - 68 cell and 60-69 rsx) but yet this problem has started occuring?

Also would like to mention that i have been playing a lot of Sonic Unleashed on this unit lately and there is certain parts that push my fan speed up to 40% (temps dont exceed 70 on either chip) so kudos to webman software, a real life saver. But this game also has a ton of slow down at random places, but i've heard that the game runs poorly anyway.
 
Last edited:
SeanRanklin said:
Yes, this thread is back at the top again!

Hello everyone.

I have managed to locate a friend who is able to re-solder some new capacitors on my unit in the next few weeks. I'm very interested to see how this will go and if it will finally solve the problem of my unit and TTT2.

In the mean time, i have noticed that when i play some older games (Tekken 5 DR) there is some weird slow down effect during intro/outro poses of certain characters. This was not present as far as i can remember on this unit, not has it ever had such a problem on my launch Slim model 2000b.

I am wondering what would be the cause of the weird sudden slow down on certain games that was not there before? Temps on both chips are normal as they always were on my 60gb (65 - 68 cell and 60-69 rsx) but yet this problem has started occuring?

Also would like to mention that i have been playing a lot of Sonic Unleashed on this unit lately and there is certain parts that push my fan speed up to 40% (temps dont exceed 70 on either chip) so kudos to webman software, a real life saver. But this game also has a ton of slow down at random places, but i've heard that the game runs poorly anyway.
Click to expand...

I experienced such slowdowns on games and even on the XMB sometime ago on my early Slim (2003A), turned out it was caused by a defective hard drive which I eventually changed and everything went back to normal. Yeah that's it, you should check your hard drive.. there's a 20% chance that I might be wrong anyways.

Sidenote; I've been meaning to create an account here, finally I had to register to reply your question, though I've been a follower of this forum for quite some time now. I've learnt a lot from the great men here, their consistency and knowledge blows me away:biggrin:. I'm glad to be a part of this awesome community and I hope to be very useful to others as this forum have been to me.
 
Naked_Snake1995 said:
Ok, so after 10 minutes on the intro scene, the PS3 decides to shutdown, its still unconclusive data, as remember, 7 of my 8 NEC/TOKINs are still original, which lead me to believe, that intensive games like The Last Of Us, require more juice through the RSX, and replacing a few NECs with Tantalums will restore the PS3 power cycle,but it wont hold against more intensive games, which needs a full 100% tantalum replacement.

Temps wise, i was getting 71ºC on the RSX, and around 58ºC on the CELL, so the machine was definitely not overheating, just to end the topic, i will try it again but this time at an abysmal 80% fan speed, but i doubt its going to change anything.

Update 2 - Same intro, same results with 85% fan-speed (Sony Airlines), but the results came the same, so overheating its out of the question, CELL: 49ºC - RSX: 51ºC

Conclusion: If you are planning to play heavy intensive 2013 titles, you will have to replace all the 8 NEC/TOKIN capacitors,or else insufficient voltage supply to the CELL and RSX leads to unexpected shutdowns.
Click to expand...

hey mate i know its been a few months but did u get around to replacing more of these tonkin caps on ur own console to see if it was an issue?
 
wrx884 said:
hey mate i know its been a few months but did u get around to replacing more of these tonkin caps on ur own console to see if it was an issue?
Click to expand...
I did on a Backwards Compatible CECHC04 60GB,once i did on the RSX, after that on the CELL, to stabilise the CELL shutdown, as upon a normal shutdown it would beep, which was a bit annoying.

So far the results have been pleasing, been using the console on a daily basis, didn't encounter a crash or a shutdown, running PS3 content holds up well, but to run more demanding games such as TLoU or GT6 a full replacement is needed, which i recall someone who did do just that, and thoes games don't crash anymore.

So far it has been 3 months since the replacement, already fast approaching to 4,i am setting a few marks for testing, 5 months for the first test, which is the time frame a Reflowed 60Gb lasted me a few years ago, and 1 year, for a Reballed machine i had, which lasted me just that.

Of course, there will always be conflicting opinions about this method, and this is kind of expected, people already made the replacement in multiple units, and they've reported great results, of course the only downside is to have the proper tools to diagnose the poor NEC Caps,so for now we will have to do it blindly, one by one, but as this method matures, new ways of diagnostics will be implemented, this isn't solely my doing, anyone who wants to contribute, they are more than welcome to do so,and there are many users already contributing.

Sent from my G8141 using Tapatalk
 
You must log in or register to reply here.

Similar threads

Back
Top