Help again! Problem about the Pinmod overclocking! I've got the 9LPRS325's datasheet.

Hi,my friends!
A few months ago, I posted a thread here, with some questions about overclock my Toshiba Satellite M100, T2350, from 133MHz to 166MHz:
http://forum.notebookreview.com/hardware-components-aftermarket-upgrades/551113-help-could-someone-tell-me-how-find-9lprs325-datasheet.html

I did not succeed at that time, but I am so grateful to the people for their guidance. moral hazard, Dufus, nX3NTY...
I don't really know how to express my gratitude.

And now, I shield CPU's B23-pin, and put a tiny wire into the socket's B22-B23 hole, so my CPU can be run at 166x6, with Intel Chipset 6x multipler lock.
I didn't shield the B23 before, so B22's high potential was pulled down to low potential by B23, FSLC/B/A=0 0 0, Bus Speed would be 266MHz!
It's the reason of my failure at that time.

OK,Let's go on!
The frequency table:


The FSLB pin:


The circuit diagram about FSLB:


So I cut off pin 45(FSLB) from the pad, then connect it to a reliable high potential pin(such as: pin 41 FSLA, pin 7 +3.3v VDDA, VCCP etc.), but it can not start, I still get a black screen.

Could someone please tell me Why?

I've updated my RAM's SPD to a slower timing, 266MHz works with 5 6 6 18 24 now, which is 400MHz's original:


I also disabled a core:


For testing, I tried these:
1.)When I get rid of R593, and place the 0k resistor to R587, then FSLB and MCH get a 1, so It would be starting at 166 Bus Speed, with the multipler 6x lock of course.
Test result:The CPU could run at 166x6.

2.)When I get rid of R593, and place the 0k resistor to R586, then FSLB and MCH get a 0, so It would be starting at 133 Bus Speed, as it was before.
Test result:The CPU could run at 133x14.

3.)When I get rid of R593, and leave R586/R587 OPEN.
Test result:The CPU could run at 133x14.So my pin 45 has a internal pull-down resistor?

Why I disconnect FSLB from MCH, then give FSLB a 1, and give MCH a 0, it's not working?
Is it unacceptable that FSLB and MCH have different electric potential?
With this question, I tried this:
4.)When I get rid of R593, and place the 0k resistor to R587, then connect FSLB to GND, What would happen?
Test result:The CPU could run at 133x6, Yes, the multipler was locked!

So, I can not OC to 166x14, because of my CPU is too weak?
I do not really want to believe it.

moral hazard said, Maybe a TME pinmod is needed.
But I can't find anything about TME in 9LPRS325's datasheet.

Help me please!
Any advice is appreciative!
If any PIC or PDF are needed, I'll post it at once.

 

I found the setPLL tool in NBR. It's so cool! Great Thanks to the author, tester...

But I dont know how to get my [BCLK_B1]/[BCLK_B2]/[BCLK_B3]...
When I set any FSB in abo's SetFSB, my notebook freezed or black screen.
Is there other way to get the [BCLK_B1]/[BCLK_B2]/[BCLK_B3]...?

 

Can you post a screenshot of setfsb with the diagnosis tab (after pressing 'getfsb')?

 

Sure.


Use this PLL, It can read my real frequency, but cant be set.

 

I'm not sure about it, can't think of anything right now.

 

Thank you! moral hazard.
I'll borrow a similar NB tomorrow, I hope it's helpful.

BTW, Could you please tell me how to get the strings [BCLK_B1], [BCLK_B2] .. [BCLK_B5] in the setPLL? or some threads I need to read? Thanks!

 

I think you would only need [BCLK_B1] and [BCLK_B2]. They would be the M/N divider programming bytes, you might find useful info about those bytes here:
http://forum.notebookreview.com/6188065-post115.html

 

Thank you for the useful thread!
I'm reading it now, But I cant modify the Default Byte Count Register.
It is the register 0Eh(14) in my PLL:


So I try to change it bigger, but failed.
In Diagnosis[Read 48byte] mode:

Ater I changed it to 16, Update/Apply, My NB crashed and black screen.

In Auto[PLL parameter] mode:

After I set to 16, Update/Apply, My NB looks no change.

What should I do?

 

Did you try using setpll to change that byte?

 

Thank you!
First, I run "setpll -read" in prompt window, then I modify the mypll.lut like this:

Code:

[INIT]
set bytecount=0x0F
set Local0=0xFFFFFE0030003981
set Local1=0x070EB075007F[BCLK_B1]00
RW Exit
[BCLK]
16,16

run "setpll mypll 16", My NB crashed.
Is that right?

 

Because my READBACK_bytes is not 0xC(12) but 0xE(14), so I modified some place with red font in setPLL.bat:

Code:

:read_pll
  set pll_bytes=%2
  :: Print a PLL dump based on specified bytes
  echo :: Start setPLL.rw                   > setPLL.rw

  echo :: set local30=PLL's readback byte [COLOR="Red"]14[/COLOR] (0x[COLOR="red"]E[/COLOR]) >> setPLL.rw
  echo ^>smbus read block 0xd2 0 0x[COLOR="red"]E[/COLOR]         >> setPLL.rw
  echo ^>local30=and(local1, 0x000000[COLOR="red"]00[/COLOR]00[COLOR="red"]FF[/COLOR]0000) >> setPLL.rw
  echo ^>local30=shr(local30,32)             >> setPLL.rw

  if "%pll_bytes%"=="" goto skip_readbackwrite
  echo :: Write PLL read_back bytes          >> setPLL.rw
  echo ^>local1=and(local1, 0xFFFFFF[COLOR="red"]FF[/COLOR]FF[COLOR="red"]00[/COLOR]FFFF) >> setPLL.rw
  echo ^>local31=shl(%pll_bytes%, 32)        >> setPLL.rw
  echo ^>local1=or(local1, local31)          >> setPLL.rw

  echo :: set READBACK_BYTES. Write only 0x[COLOR="red"]E[/COLOR] bytes. >> setPLL.rw
  echo ^>smbus write block 0xd2 0 0x[COLOR="red"]E[/COLOR]        >> setPLL.rw
  echo ^>local30=%pll_bytes%                 >> setPLL.rw

and run "setpll -read 16" to re-generate mypll.lut & setPll.rw. but It makes a error:

What's wrong?
I cant understand "shr local30,32"&"shl pll_bytes,32", Does the "32" need to be mod?
Thanks for your explanation!

 

That is beyond me

 

Ah....

 

Did you talk to nando4 about it?
If anyone would know, I think that it would be him.

 

Below is the substitute code fragment to use byte 14 instead of byte 12 to set PLL readback byte count.

Code:

:read_pll
  set pll_bytes=%2
  :: Print a PLL dump based on specified bytes
  echo :: Start setPLL.rw                   > setPLL.rw

  echo :: set local30=PLL's readback byte 14 (0xE) >> setPLL.rw
  echo ^>smbus read block 0xd2 0 0xE         >> setPLL.rw
  echo ^>local30=and(local1, 0x00FF000000000000) >> setPLL.rw
  echo ^>local30=shr(local30,48)             >> setPLL.rw

  if "%pll_bytes%"=="" goto skip_readbackwrite
  echo :: Write PLL read_back bytes          >> setPLL.rw
  echo ^>local1=and(local1, 0xFF00FFFFFFFFFFFF) >> setPLL.rw
  echo ^>local31=shl(%pll_bytes%, 48)        >> setPLL.rw
  echo ^>local1=or(local1, local31)          >> setPLL.rw

  echo :: set READBACK_BYTES. Write only 0xE bytes. >> setPLL.rw
  echo ^>smbus write block 0xd2 0 0xE        >> setPLL.rw
  echo ^>local30=%pll_bytes%                 >> setPLL.rw

 

Thank you for your recommendation!

 

Thank you very much! The code works perfectly!


OK, I'm going to change the frequency.

 

According to this thread:
http://forum.notebookreview.com/6188065-post115.html
I cant find the byte of SATA_SEL/SRC_Main_SEL/CPU_DIV_EN/PCI_DIV_EN/SRC_DIV_EN in my PLL's datasheet, So I just only disabled the CPU outputs spread(B2b0), disable the 27/96/100M Spread Spectrum(B9b1), enable the CPU M/N(B9b3).
Then I try to modify the N Divider(B16b7:0&B15b7:6), I found this:

Code:

Bus Speed=16xN/M
Default: N=110010=50, M=110=6, Bus Speed=133

Up to 143.6MHz Bus Speed with 2 cores, highest multipler, 1.26v:


Up to 149.0MHz Bus Speed with 1 core, lowest multipler, 1.26v:


Why?
Is the DDR/PCI-E/PCI frequency too high or the CPU has a weak constitution?
How could I do next?
Thank you!

 

What happens when you try a higher fsb?

 

The NB became black screen or grey screen.

 

What graphics card do you have?

 

Code:

	Intel(R) Graphics Media Accelerator Driver for Mobile Report


Report Date:			04/17/2011
Report Time[hr:mm:ss]:		14:25:07
Driver Version:			8.14.10.1930
Operating System:		Windows 7 Professional* , Service Pack 1 (6.1.7601)
Default Language:		Chinese (Simplified)
DirectX* Version:		11.0
Physical Memory:		3062 MB
Minimum Graphics Memory:	8 MB
Maximum Graphics Memory:	256 MB
Graphics Memory in Use:		110 MB
Processor:			x86 family 6 Model 14 Stepping 12
Processor Speed:		1862 MHZ
Vendor ID:			8086
Device ID:			27A2
Device Revision:		03


*   Accelerator Information   *

Accelerator in Use:		Mobile Intel(R) 945 Express Chipset Family
Video BIOS:			1585.0
Current Graphics Mode:		1280 X 800 True Color (60 Hz)



*   Devices Connected to the Graphics Accelerator   *


Active Notebook Displays:	1


*   Notebook   *

Monitor Name:			Generic PnP Monitor
Display Type:			Digital
Gamma Value:			2.20
DDC2 Protocol:			Supported
Maximum Image Size:		Horizontal: 11.8  inches
				Vertical:   7.4  inches
Monitor Supported Modes:
1280 X 800 (60 Hz)
Display Power Management Support:
	Standby Mode:		Supported
	Suspend Mode:		Supported
	Active Off Mode:	Not Supported

* Other names and brands are the property of their respective owners.

 

Can you underclock that IGP with setpll?

Something like Stage4 here:
http://forum.notebookreview.com/hp-...ebook-2530p-owners-lounge-10.html#post7121833

 

Good idea! Thank you!
I underclock the GMA950 to 166MHz(original: 250MHz). But the Bus Speed cant Breakthrough 143MHz yet.
So I try to OC 166MHz Bus Speed directly, but it still became black screen.
After I power it off, Tragedy occurred... My NB cant be started. Always black screen.
I dont knonw what's the matter with it. Did the CPU or North Bridge damaged?

 

Remove battery, AC, and try to restart system with stock settings. If it works, then use Throttlestop (preferred) or Rmclock to set all CPU multipliers to highest voltage, then try to OC again. Then it's a matter of RAM being able to accomodate the higher FSB and perhaps compensating with lower bootup timings (SPDTool/ThaiPhoon Burner).

 

Thank you!
I've tried this and short circuit the battery in MOBO to clear CMOS, but it still doesnt work.

Yes, I've set all CPU multipliers to highest voltage(1.26v) with Rmclock in the earlier test.
I'm going to borrow a T4200 for testing tonight.

 

Oh My Lady GAGA! My NB didnt dead! Thanks GOD! Thanks nando4! Thanks moral hazard!
I borrowed a T4200 yesterday, but the pins are not same as mine.
When I assembled my CPU back, The NB posted incredibly!
It seems that I must remove the CPU, otherwise the CMOS clear maybe not efficaciously.
What should I do next?