A500: upgrade CPU; need help

I'm trying to replace T6500 with new T9600. I ran into a problem while disassembling laptop. On the last step when I try to remove motherboard, I cannot do it in spite I screwed out all holding screws, disconnected HDD and DVD drives and main fan. Something in the left bottom corner holds it. Question for those who have already done this operation: what could it be and how to work around this problem?





On the 3-d picture there is my second (minor) question. White power connector (on the top): how to unplug it. I tried to pull it back and up - didn't work.

 

Does it have a VGA port? The two holes (visible from the outside) that secure the VGA cable must be unscrewed. I do that from the outside of the case using a small pair of flat nose pliers. (My experience was not with this laptop, nonetheless it may be a fairly universal step.)

 

Uh, one slightly stupid question, why are you removing the motherboard? You don't need to remove the motherboard to upgrade the CPU. Does your CPU use BGA or something?

 

No, I have Core Duo. That's because of Toshiba design: CPU is in the back side of motherboard . There is no direct access to it.

About VGA: good try, but not in this case. I feel there s something below PC Card slot.

 

Are you sure you're not missing a screw at the bottom?

Try to flex the left case a bit (where the usb port is), and see if you can lift the mobo.

 

The power cable looks to be soldered onto the motherboard.

If it's not soldered it should lift up. Dont pull too hard, because I think it is soldered and you will damage the motherboard.

If you want to remove the power cable, you would probably have to flip the motherboard over and un-solder the 4 points.

 

You don't have to take all the motherboard off, just need to turn it around

 

I'm sure that there is no forgotten screw. I even removed rubber corner disk at the bottom expecting to find there hidden screw. No chance...

Yes,I agree that I can avoid complete removing, so I can leave alone power supply connector. Still need to find what piece does not permit me to turn it out.

 

Is there a modem on the other side?

I think you have to take that out.

 

or is it not the side ports that just need to get loosen up from the case ?
have you tried to pull the side case a little to give room so you can pull up the mobo?

edit: if you can tell whats on the side and bottom part of it, it would be helpful

 

OK, I did everything. And it works. But... in spite T9600 (as it marked) it shows E8235. BIOS mistake? I wouldn't mind, but t9600 is 35w, and E8235 is 44W. This is 19W more than original T6500. On the other hand, E8235 declared as iMAC specific, probably it is not supposed to work on other platform?

 

what does written on the cpu?

 

you might have got conned and got a hotter apple CPU... it will work but check CPU-Z again and use the latest version of it..

 

but they are cheaper, yes

 

they're cheaper... but looks like he's been conned... the BIOS should have recognized T9600 since this model came with it....OP , try latest version of CPU-Z and if it comes as E processor, return it and ask for refund..

 

It's funny, but seller changed descrition of product the same day I made my purchace ( see http://cgi.ebay.ca/Intel-Core-2-Duo-Mobile-CPU-T9600-2-80Ghz-6Mb-T9400_W0QQitemZ130366513534QQcmdZViewItemQQptZCPUs?hash=item1e5a73217e). There is also one alarming thread ( see http://forum.notebookreview.com/showthread.php?t=449101). First of all I checked new E8235 using LoadTester-RealTemp couple. Regular temperature was 55C. At 100% CPU usage it started to grew up until reached level 65-68C and stayed on this level forever. Then I requested seller about extra warranty, She agreed on extra 3 months. I decided to stop at this point and wait at the same time observing new CPU behavior.
Thanks everybody for help.

 

Yeah Sean is right, if you purchased it with a price of a T9600, get a refund or partial refund, the E processor should have been cheaper.
And you can undervolt it to get better temps.

 

I payd $230 with free delivery. I think it is a fair price.
How to change voltage? My BIOS doesn't have this settings. BTW There is also no VT settings, it just works

 

You can follow the steps here to undervolt.
http://forum.notebookreview.com/showthread.php?t=235824
credit to flipfire on this forum

edit: that's a good price

 

I would be kind of scared to use 44W CPU in a laptop unless I am 100% sure that its power circuitry can feed it comfortably. It may work fine for a while but may burn out the motherboard in the long run...


--

 

44W isn't the power consumption, it's a heat measurement.

Also, you shouldn't trust the quoted TDP. It's wrong, ask K-TRON.

The chip is exactly the same, same tech, same voltage, same multi, same cache, same everything. The heat and power consumption will be exactly the same.

 

If everything is the same, why is Intel marking one as 35W and the other as 44W? Why aren't they marking every chip in that series as 35W? Are they too dumb to de-grade their products with mis-leading specs?

Have you ever heard of leakage power? Do you know anything about the leakage power variation among different chips?

(I did my Master's work partially on that area.)


--

 

So I performed undervoltage procedure.
Before: 1.250V, 40min test (ORTHOS): 79C/83C
After : 1.1V , 180min test : 69C/74C

Is it acceptable?
It is interesting to know opinion of those who have experience.

 

stock T series have 1.175V as stock processor voltage... so that's why ur E seires are rated with higher TDP... i would personally try pushing max multiplier voltage to 1.0V...

 

Yes, I wanted to do it but in the combo box for x10 values start from 1.1V. Any idea how too set less value?

 

That's alright, I'd be aiming for something lower though.

 

Maybe the Imac specific CPUs are the lower binned T -series and they just bump up the voltage to make them stable. But they are the same chips made from the same silicon wafer.
Don't give me that leakage power BS, the chips are the same, the leakage power will not vary that much. Maybe 1-2W but not 9W.

Anyway, don't ask me about it, talk to K-TRON.
He has a very good argument for why the T and P series are actually the same TDP.
That argument can apply in this case also.

 

You have a max multiplier of 10.5... so if u have RM clock to undervolt , u need to mod RM clock... so it sees ur 10.5 multiplier as 11x or something else...

Here's the undervolting guide and do as it says and then mod RM clock like in the thread link which i am giving u... its very useful so subscribe to it and keep it...

http://forum.notebookreview.com/showthread.php?t=411168

 

Bingo!I just spoofed RMClock. The problem was that this guy did not recognize my cpu and used wrong limit values. Actually it took current values as a low limits. I installed trial version CPUGenie64 (there is a good wizard). It detected CPU correctly and I passed all the process of undervolting. Then I ran RMClock once again, and this time it read current values that CPUGenie64 had set. I saved this values in profile, uninstall CPUGenie64 and restarted.I have now 70-71C.