RTM875t-606( Info gathering and OCing)

Alright thanks. i understand. did u say u just connected to the card reader housing?

and the reason i dont just use setFSB is because i have multi OSes on my machine ;-)

 

SoftWare OCing works in Linux too.

Ok, here is a better PIC of what I'm saying.


PIN 4 PIN 10

See the Odd way PIN 10 Runs?

If you look next(Right) to the Resister for PIN 4, that is what the PAD will look like with no resister.

 

I dont have linux. Leopard. But still i would like to just make a global system change that is constant.

 

ok thanks this is what i am looking for thanks for the details in describing how you took off the resister can you post a pic up of were u grounded it to?

 

ok.. so i did the solder and everything is working... except my system wont boot when the wireless is installed....

correction... Bad Solder... Works now, but system locks up in OS

 

Snap a few shot's of your Mod, we'll see how your pin 10 trace is routed.

What chipset are you using, ICH?

Did you try with the resister?

Also, PM me the out put of kextstat | grep SM

 

Dear Darth_nVader, Can I put Setfsb in startup when windows start? May I use command to set 00h bit2 to 1 when start?

 

Ok, guys. Here is utility posted in another thread that allows to overclock RTM875T-606 without hardmodding. FSB-166 and FSB-250. It perfectly works for my Timeline 3810 with su7300 onboard (@1300 in stock and @1630 OC'ed).

The only problem is that we have binaries and no source code or description. Readme.txt says that utility is based on original setfsb. Do we have developers to disassemble and understand how does it work?

Original post, OC utility


Full thread

 

Put a batch file in the startup folder:

Code:

@echo off
set incr=5
set FSB=240 
set PLL=ICS9LPR501SGLF


c:\setfsb\setfsb -w0 -s%FSB% -i%incr% -cg[%PLL%]

Similar to that^.

I don't think you can change that bit at startup though, either ask abo (creator of setfsb) for help or do this:
http://forum.notebookreview.com/showthread.php?t=430230

 

Hi people... I have an hp dv9700t with an RTM875T-606 PLL. I've tried the first steps listed above, but I'm running into issues. Specifically, nothing shows up in the registers area on the diagnosis page with this laptop! I know it's not my copy of SetFSB and it's not Win7 x64 because I have a desktop with an EX58-Extreme mobo and it displays PLL register info just fine.

I get this: and this where you can see neither have any info in the control registers after clicking Get FSB and in the first capture, you can see where I selected PLL diagnosis as the clock generator, the FSBs just say Unknown / Unknown ???

This is a picture I took of the PLL on the motherboard, so I know for a fact that it's an RTM875T-606 PLL and I'll be happy to hardware mod it if that's what it takes, I just want to know that it'll work before putting all that effort in.

Oh, and I've also tried that custom SetFSB for this PLL from http://forum.notebookreview.com/showthread.php?p=5833099, but that just freezes the system up hard... which could be for any number of reasons, but just so no one asks - yes, I tried it.

Suggestions? Thanks for your help!

Oh, and I attached page 2 (clock generator) from the Quanta block diagram for the Hasee HP500 (the very same system that the Chinese guy who modded the custom SetFSB for the RTM875-606) has, and note, it lists pin 4 as TME and the PLL as a ICS9LPRS365AGLFT. Does this perhaps mean that the RTM875-606 and LPRS365AGLFT are the exact same pinout? If so, that's good right? Please, someone do something with this!

 

My Timeline has su7300 (it is an underclocked chip from 1066 to 800, for lower power consumption), GM45 chipset (it can run @1066), DDR3@1066.

So software overclock is great for me, and it works тшсудн without hardmodding (Timeline has TME enabled!).

All we need is to understand how does it work.

 

Could you post your setfsb diagnosis window before and after overclocking. That way can see what bits it's setting, like shown and explained here. Will also mean other Timeline owners can compare PLL states to see if truly your system is TME locked.

 

Nobody has any ideas for me on my last post?

 

I've overclocked my timeline using SetFSB. Just set register 1f to 56 and than 0e to 93. It seems 0e is the register which control fsb speed

Which system do you have? x64 or x86?

 

Nando,

i forgot to run cpuid before OC. So cpuid screen only for OC.

Acer Timeline 733g25i, it runs @200 in stock. Seems advan20092 is right!

no OC:

http://img522.imageshack.us/i/nooc.png/

OC:

http://img525.imageshack.us/i/22263931.png/
http://img96.imageshack.us/i/cpuid.png/

 

morpax,
You can also try set 0e to 80 without changing 1f. This may cause restarting but FSB will be set to 266

So rtm875t-606 doesn't need hardware mod to be overclocked in acer timeline models

 

Windows 7 Ultimate x64, hp Pavilion dv9700t.

Has an RTM875T-606 PLL which I know for a fact (please see my previous post for all specific details). Thanks.

 

advan,

I`ll try, thanx. No laptop here )-:

I follow Timeline thread@ixbt (-: You've done a great job (bios and OC), so thank you a lot one more time(-:

BTW, can you post there that su7300 can be overclocked too in such maner? I'm not registered.

Did you getFSB after rebooting@266? May be some changes in registers you don't notice.

Not only timeline can be OC! Remember, original developer of oc_rtm875t.exe owns HP500, few chinese guys successfully OC other laptops, guy from this forum OC Vostro V13, guy @ixbt succesfully OC Acer 5930G with P7350 and so on.

 

Correct. I just said what I exactly know.

Yes, 1c register has changed from 23 to 13. Sometimes it allows to OC without restarting, sometimes not. I'm not sure what is a problem with it.

It looks like unsigned driver issue. Try to boot in 'disable signed driver enforcement' mode. Just press F8 while system boots and select this mode.

 

Unfortunately, the answer isn't that simple. My WinRing0x64.sys is fully signed. Just for good measure, I booted into disabled signature enforcement mode, but there's no difference. Nothing shows up in the PLL or SMBus boxes, but I can pull frequencies if I use the RTM875T-587... just I can't get bit-level diagnostic data, which means that I can't set bits either... which means I don't know whether setting pin 4 to ground will work for me or not, etc etc. I was just about to open it and try, because if you look at the picture of my PLL, you'll see that pin 4 (TME) connects directly to a resistor and there's a empty solder point for another resistor directly after it... if I were to bet, I would say that (the empty point) it's probably ground... I guess my multimeter will tell me either way.

 

Okay. This is what I found out:



See that - pin 4's first connection is a 33.3Ω resistor, voltage is approx 1.67V.

Directly down the trace from there, it's still 1.67V because that's inline. The next pin down, across there where another resistor could be is ground!

So those of you who know about this kind of thing -- do you think the resistor can just be moved to the solder point further down to ground the TME, but still allow PCI2 to work (if it's being used)?

I've looked at a lot of similar Quanta board schematics, and it seems like there's often points just like this, where if the resistor is at location 1, TME is enabled and if location 2, then it's disabled...

I'm wondering if I move the resistor, if I might be able to adjust using sub 1MHz increments with SetFSB on the RTM875T-587 setting.

Thoughts? Assistance?

 

pyfgcrl,
why can't you try pll registry adjustment first as described above? May be it is old SetFSB version or some bugs in your system so software problems cause SetFSB can't display registers. I think you should check for solution before make hardware mod.

 

Can't try adjusting PLL registers because I can't access the PLL registers. I'm running the latest SetFSB downloaded directly from Abo's website.

Anyway, I executed the hardware mod -- system still runs fine, and now I can use the little app to underclock my FSB to 167MHz, but the one that overclocks it to 250MHz still crashes it. It ran once, and my 2.5GHz Penryn processor was running for a few seconds at 3.375GHz!! (IDA 13.5 multi x 250 MHz) -- that is, until it froze, 10 seconds later.

So anyway, my TME is now hardware disabled. My holy grail will be to have the ability to use SetFSB to adjust in sub-1MHz increments, but until someone tells me how I get the registers to magically appear, it looks like I'm screwed.

UPDATE: Okay! I figured it out... sort of. It appears that my SMBus is disabled by my laptop's BIOS at startup. I see where there's a thread here that discusses using GRUB 2 to access the register and enable it, but how do I do that from Windows? Is there a way?

 

I'm testing HWDirect now. It is small app which allow access to hardware registers, memory, smbus etc. I'm not sure if it'll help in your case.

 

I have RW-Everything 1.4.7 and according to this, my laptop's RCBA is 0xFED1C001 and I'm supposed to add 3418h to it and then read that from somewhere... 'course the instructions are for using GRUB2, and I'd rather not have to make a GRUB2 USB stick (don't have a spare one lying around, nor an Ubuntu Karmic LiveCD) but I have no idea what type of addresss that is that I'm supposed to read from - a memory address? PCI register?

UPDATE: Okay, I made a GRUB2 1.97~beta4 boot CD, and my register data at 0xFED1F418 is 0x31C0001 - so (according to that link listed above at least) bit 3 is already 0, and that would supposedly indicate that SMBus isn't really disabled... so it must be some other issue.

 

What memory do you have?

I think that it can`t run at 1000 MHz, so oc_rtm875 crashes system.

And i`m not sure that any laptop can have SMBus really disabled.

 

Crucial 4GB (2 x 2GB) 200-Pin DDR2 SO-DIMM DDR2 667 (PC2 5300)
Dual Channel Kit Laptop Memory Model CT2KIT25664AC667
CAS Latency in JEDEC EEPROM: 5-5-5-15.

It may also simply be the processor, 'cause that's a lot to ask of it on stock voltages.

Actually, SMBus read/write can be disabled by a certain bit set in BIOS, from everything I've read in this forum. Take a look at some of the content here, like nando4's post #4 for example.

 

1. Definitely need to slow your RAM timings. You have 800Mhz CL5 FSB timings. At 1066Mhz you'd probably need CL6.

2. Yes, smbus can be disabled by setting: RCBA+3418, bit 3=1 as shown here. Setting that bit means the smbus won't appear in a PCI bus scan at all.

3. Enthusiastically I tried both hwdirect AND rw-everthing's SMBUS read feature to read a 22-byte smbus block of data at 0x69 (the PLL). This is on a 2510P's ics9lprs355 where setfsb reads the PLL data without any problems

Neither worked successfully. hwdirect gives a "error! smbus busy!" when do a block read. rw-everything gives "smbus device access error". Setting hwdirect to do byte read gives nonsense "0D" string result.

Would be curious if anyone can successfully read their PLL using either tool. For the time being then only have the following to read the PLLs:

* Linux: grub2 bootloader, setfsb binary or i2cdetect/i2cdump
* Windows: setfsb for defined/compatible PLLs

EDIT: r-w everything has a Access -> Clock Generator that does correctly read the first 13 bytes of my ics9lprs355 PLL. CLicking write however freezes my box. setfsb's author told me it was a 22byte block that needed to be written all at once. rw.ini has a lot of options.. so likely needs a 22-byte setting somewhere.

 

Technically, the RAM is running at 665 MHz (332.5 x 2) ... but yes, I definitely agree that CL6 is probably the way to go. What do you suggest - Thaiphoon Burner?

I didn't try 22-byte blocks, but that's because I didn't know any better. I'll give it a shot and get back to you with the results on my 875T-606.

So, I understand that setfsb for Windows works for 'defined/compatible PLLs' but it seems that Darth nVader got setfsb to read his SMbus and PLL even though the RTM875T-606 isn't defined or compatible. Any idea why my registers come up blank when I select PLL diagnostic or RTM875T-587?

 

Using RW-Everything I got access to 10 first PLL registers (Clock Gen button). I guess it should be just editing rw.ini to get access to other ones.

 

Is the clock generator button really PLL information? I thought PLL information would come out of the SMBus button.

 

Yes it is really PLL registers. I've compared with SetFSB

 

Oh, excellent... so I have an RTM875T-606 and my output is

Code:

41 88 FF FF FF 00 30 1E 10 65 7D 00 0D

do you know if there's something there I can change to alter FSB?

 

There should be 30h bytes instead of 13d. SetFSB changed a lot of them to increase speed. I've changed 0e and 1f bytes to do so. Try edit rw.ini to allow all bytes to be displayed

 

Awesome. My hp Pavilion dv9700t with T9300 (2.5GHz) Penryn has been successfully overclocked to 3.13GHz!




Steps:

1. Executed pin mod to TME (pin 4) on RTM875T-606 as illustrated by me here, thanks to moral_hazard, nando4, darnok44 and Darth_nVader as described here. This step actually might not have been strictly necessary, but I'm glad I did it - I'd really like to see if I can get some sort of micro-adjusting FSB tool to work... maybe then I could go even higher than 250MHz FSB!

2. Modified RAM SPD settings with SPDTool 0.63. I used Thaiphoon Burner 6.7.0.0 for advice on changing the latency parameters, and then SPDTool to fix the checksum and burn them into the module. This was all done on an old 1GB Samsung module to test (so if I screwed up, I wouldn't care so much) and now that it's working, I'm going to mod my Crucial (Micron) 2GBx2 sticks -- UPDATE: That didn't work due to the modules' SPD EEPROM being hardware write-protected, so I used two G.Skill 4GB SODIMMS instead, which did work. Details are here.

3. Downloaded custom RTM875T-606 overclocking tool based on SetFSB, found here.

4. Ran the executable OC_RTM875T_250.

5. Voila! T9300 running at 3.13GHz (max temp 70°C) stable enough to run BOINC at 100% CPU for 25 minutes now while I wrote this all up. By the way, IDA is 3.38GHz

 

wow dude thats awesome

 

So is it safe to assume that this method will not work with the RTM875N-606 square realtek pll chip?

 

If your system FAILS..............STOP, you need to try and setup the PIN17/18 Pin Config (I'll add a table for this later if we have anyone that needs it, post a reply to this topic.)

could you add the config

 

Hi all.

I tried this mod on my aspire 8920g and it worked a treat.
Just grounded pin 4 of rtm875t-606 to ground.
Then i ran set fsbset and slowly ramped the cpu to 2.6 gig and is still very stable.
My notebook has the t8300 in it.

hope this helps.

 

RTM875T-606 / RTM875N-632 datasheet

Did anyone ever obtain a RTM875T-606 datasheet? I am specifically looking for info on the RTM875N-632 variant. This is the PLL visually identified to be in a i5-430M Compaq CQ41. I contacted setfsb's abo requesting access to it. Abo offered no level of support without a 'donation'.

So atm I'm not sure if the RTM875N-632 can even be overclocked.

The good part:

* the TME bit, register 09, bit 6, is disabled
* register 7 is 0x1E indicating it is a Realtek PLL

The not-so-good part:

* quite a few of the other registers are 0xFF which may mean it is not overclockable. Certainly differs to the RTM875T-606 screenshots posted earlier in this thread.
* the RTM875T-606 specific setfsb binary doesn't do any overclocking on the RTM875N-632.
* selecting any of the other Realtek PLLs in the freeware setfsb altered the CPU clockspeed or made any difference to system performance.


Image: RTM875N-632 PLL registers

With no datasheet at hand, can any RTM875* guru have a look and make any suggestions?

 

hi tanjo-15,

you wrote: Hi all.

I tried this mod on my aspire 8920g and it worked a treat.
Just grounded pin 4 of rtm875t-606 to ground.
Then i ran set fsbset and slowly ramped the cpu to 2.6 gig and is still very stable.
My notebook has the t8300 in it.

hope this helps.

where about is the RTM875t-606 postion on the mainboard and what exactly did you do, you connected pin 4 to GND(ground) and did you disconnect anything for pin 4 ? resistor or something else? i want to do this mod also,i have an x9000 (Q174 cpu) and then let s have a look what i can do..

so info or picture wanted Tanjo-15 of you mod.

thanks in advance

 

charlyd why do you need to overclock the PLL when you can just increase your multi:
http://forum.notebookreview.com/har...unlock-core-2-extreme-multiplier-windows.html

 

hi yes you are right i saw the article earlier but the unlock was not working but yes of course i had an other cpu then... ;-) yes yes it is working.... thanks very very much, and what about the guy in this topic who grounded pin 4??? is that not working or... i am lucky with this CPU x9000 Q174

 

is this topic still live?
need help, my fails.


cpus T5250 1,5Ghz and backup T7500 would try oc too.

 

I have an Eisystem laptop and the clock gen is the RTM875t-606. I took the laptop apart and checked it.
I went to Diagnosis->PLL Diagnosis (Clock generator drop down menu)->Click FSB. The result is shown below.

Do i have to do the PLL mod? Does the mod work with the intel GL40? and does it add support for FSB1066?

Edit:
When I select RTM875T-606 in SetSFB 2.3.166.131 (trial version), I get a "PLL ID Error"!

 

For what its worth, I have finally had some succes with that RTM875T-606 on my Extensa 5630G with currently a T7250 cpu (2.0GHz, 2.2GHz IDA).

RTM875T 250MHz overclocking tool alters the FSB to 250MHz. No version of setfsb i have tried have worked before other than freezes.

(Nvm all the other I wrote as it is not relevant for this thread)

 

Picture in post 0 is gone

 

i all and thanks i have an hp hdx x16 with rtm875n-606 (square version)
i have grounded pin 11 and i have this now View attachment 77594
nando4 validate i have the tme enable
but when i do this my system freeze
i think i have to do another pinmod on pin 17 18??but i dont know how.
thanks for your help

 

I also have rtm875n-606. My board is for an HP G56 intel based.
Is there another msg board for this?

 

I have rtm875n-606.
Can you please refer me to the post that shows how exactly you ground pin 11.