• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

2600+ MPMW ROCKS! 2.7Ghz

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Gogeta_G3D

member
Joined
Apr 17, 2004
Looks like my MPMW theory was right? I wrote down every single bit of info..temps, vcore, 12+ etc..its pretty long ill post that tomorrow. to give a summary. Its a 0310MPMW 2600+ MPMW from excalibur. Due to my low rails, i believe with a better PS ill be able to oc higher, or maybe even with less voltage, i hope.

11x200 @ 1.45v



13x200 2.6g
1.775v
idle: 37
load:40
+12=1.76


1.95v
13.5x200- 2.7g
+12=11.67 - very low
+3.3=3.28
idle=43'C
load=47'C

I test stability with something realistic, UT2K4 for 1 hour play. In my experience with other CPUs ive oc, if they can play UT for at least 1 hour, they are stable. BTW, at 2.8Ghz @ 2v, I can go into windows, but crashes immediatly. il post pics tomowrrow.
 
Your V-rails arent low at all there m8...very normal even. I dont think a better PSU will influence your OC at all
 
Its possible just not the rails doing it...my 2600m hates running past 49c for long periods of time...and it is possible that when you set the core to 2v plus...you are actually running into a temp issue besides a rail issue...

YGPM on a adjustable pot mod for the Tt...
 
Heres testing report: I realy want to know what was going on between 1.825v and 1.93v. Not sure if it was heat, low rails, or voltage or combination of....i hope someone can give insight on this.

Also, there were no error beeps, ever. There were also no system freezes. System either rebooted itself, or crashed to windows. All temps are CPU temps. I had the side open, so system temps were always 20-22. I spent the whole day testing yesterday, UT2k4 stable= at least 1 hout play.





- 15x133, 1.45v, UT2k4 Stable = yes

- 12x200, 1.75v, UT2k4 Stable = yes

- 12.5x200, 1.75v, UT2k4 Stable = yes

- 1.75v, 13x200
Into windows = yes
UT2k4 Stable = NO
Temps: Idle=37'C

- 1.775v, 13x200
Temps: Idle=37, load=40
UT2k4 Stable = YES

- 1.8v, 13.5x200
Post=yes
login screen=yes
into windows=NO

- 1.825, 13.5x200
Post=yes
login screen=yes
into windows=YES, but crashed after 2 seconds.
STABE=NO


- 1.85v, 13.5x200
into windows=YES
Temp idle: 38
idle rails:
+3.3=3.26
+5v=5
+12v=11.75
UT2k4 loaded, about to enter game, crashed to windows with UT2k4 bug report=CPU error
UT2k4 stable=NO

- 1.875v, 13.5x200
inot windows=yes
Temps: idle=39
idle rails:
+3.3v=3.26
+12v=11.72
+5v=5v
UT2k4 loaded, but immediatly crashed to windows
Ut2k4 stable=NO

- 1.9v, 13.5x200
Idle temp= 40
rails idle
+3.3v=3.26
+12v=11.71
+5v=5
UT2k4: Made it into game, crashed after 20 seconds
UT2k4 stable=NO

- 1.93v, 13.5x200
Temp idle: 41
rails idle"
+12v=11.67
+3.3v=3.28
+5v=5
UT2k4: Enter game, crashed after 2 minutes
UT2k4 stable= NO

- 1.95v, 13.5x200
UT2k4 stable= YES
rails idle:
+12v=11.67
+3.3v=3.28
+5v=5.03
Temps: idel=43, load=47


- 2v, 14x200
Made it into windows, but crashed immediatly. I think it was getting too hot, or even rail was low?
 
i think both factors play a part here.

i had mine stable at 2420MHz @ 1.725v with the previous psu, the rail would drop and jump if i went to 1.8v which would crash the system eventually. with better psu 2530MHz is rock stable at 1.8v.

on the other hand, getting past 2.6 takes about 1.95v. i noticed that when i load it and it's hitting 50C, it has a good chance of destabilizing.

i think it's one of the characteristics of mobiles; i had an oportunity to play with a desktop barton, terrible overclocker but could run at high temp no problem.
 
mcoleg said:
i think both factors play a part here.

i had mine stable at 2420MHz @ 1.725v with the previous psu, the rail would drop and jump if i went to 1.8v which would crash the system eventually. with better psu 2530MHz is rock stable at 1.8v.

on the other hand, getting past 2.6 takes about 1.95v. i noticed that when i load it and it's hitting 50C, it has a good chance of destabilizing.

i think it's one of the characteristics of mobiles; i had an oportunity to play with a desktop barton, terrible overclocker but could run at high temp no problem.

i think your right. they seem to be not stable over 50'c with high vcore. but can you explain then why mine wouldnt run [email protected]? not heat issue I think, maybe rails? but if rails, then why would 1.95v work fine?
 
It's a vicious cycle. To sustain stability at higher temperatures, you need higher core voltages, which then in turn require lower temperatures. Usually the 50 C/2v mark is banking on the edge of stability for AXP's, it seems. I needed 35C to remain stable at 2.6GHz/2.1v with my 1700+. Others could do the same in the 50C range so there is plenty of variation.
 
Gogeta_G3D, when pushing higher voltages with a psu that is stragling, the voltages would jump downwards. let's say you should be ok with 1.9v, right? the psu would provide it most of the time but there would be some micro fluxuations that would bring the voltage down to (hipotetically) 1.85v for a millisecond or so. since 1.85v is not enough to have the cpu stable at the given speed, it will destabilize. when setting the voltage at 1.95v, the rail would drop to 1.9v or so, which is enough to keep the cpu stable but only at a cost of running it at higher voltage and therefore higher temps. most of the time.

at least that's what i think.

btw, have you done L12 mode? if not, go for it. simple and reversable. won't give you better speed as mhz are concerned but sure as heck would help you to utilize those samsung chips of yours better :p
 
Gautam said:
It's a vicious cycle. To sustain stability at higher temperatures, you need higher core voltages, which then in turn require lower temperatures. Usually the 50 C/2v mark is banking on the edge of stability for AXP's, it seems. I needed 35C to remain stable at 2.6GHz/2.1v with my 1700+. Others could do the same in the 50C range so there is plenty of variation.

You are quite correct. Once I get over 80°C there is no hope for stability.

Anyways, I was no too surprised when I saw the stepping. A 0310, these were all pretty good. My 1700 is a 0310XPMW. It does not get 2.7Ghz (yet.....I have very crappy cooling...anymore than 2.4Ghz and 1.85Vcore yeilds 80+ °C temps.

Funny part was...back when I bought the chip I heard that the MPMW's did better on stock voltage and the XPMW's got an overall higher overclock. Kinda funny, yours is a MPMW and got a really nice high end. Oh well, yours is a 2600, probably different.

Also, damn nice overclock!

Cheers!
-0cer
:cool:
 
@md0Cer said:


You are quite correct. Once I get over 80°C there is no hope for stability.

Anyways, I was no too surprised when I saw the stepping. A 0310, these were all pretty good. My 1700 is a 0310XPMW. It does not get 2.7Ghz (yet.....I have very crappy cooling...anymore than 2.4Ghz and 1.85Vcore yeilds 80+ °C temps.

Funny part was...back when I bought the chip I heard that the MPMW's did better on stock voltage and the XPMW's got an overall higher overclock. Kinda funny, yours is a MPMW and got a really nice high end. Oh well, yours is a 2600, probably different.

Also, damn nice overclock!

Cheers!
-0cer



i can also get 2.3Ghz @ 1.45v and 2.4Ghz @ 1.6v....just in case your wondering.
 
Last edited:
mcoleg said:
Gogeta_G3D, when pushing higher voltages with a psu that is stragling, the voltages would jump downwards. let's say you should be ok with 1.9v, right? the psu would provide it most of the time but there would be some micro fluxuations that would bring the voltage down to (hipotetically) 1.85v for a millisecond or so. since 1.85v is not enough to have the cpu stable at the given speed, it will destabilize. when setting the voltage at 1.95v, the rail would drop to 1.9v or so, which is enough to keep the cpu stable but only at a cost of running it at higher voltage and therefore higher temps. most of the time.

at least that's what i think.

btw, have you done L12 mode? if not, go for it. simple and reversable. won't give you better speed as mhz are concerned but sure as heck would help you to utilize those samsung chips of yours better :p

hey, thanks alot. thats exacly what I needed know. Im trying to sell my current 480w TT. Im gonna go for a TC 510. If that doesnt keep my rails steady, im gonna shoot myself.
 
You test for stability using games?

Try using prime95 for 24hrs and get back to us to let us know if it is stable then. As i truly do not deem anything stable till it runs prime95 prority 10 for 24hrs. I do not think the priority comes into to play as much as just running prime95 itself. It is just safe just in case other apps are stealing cpu cycles.

keep clockin
 
jkb2198 said:
You test for stability using games?

Try using prime95 for 24hrs and get back to us to let us know if it is stable then. As i truly do not deem anything stable till it runs prime95 prority 10 for 24hrs. I do not think the priority comes into to play as much as just running prime95 itself. It is just safe just in case other apps are stealing cpu cycles.

keep clockin


haha, people are SO obsessed with prime95 its scary. I only test with real world apps, UT 2k4 is especially CPU intensive and sensitive, and it happens to be a game. I, truly deem anything stable if it can run UT2k4 for at least 1 hour. Doesnt make sense to use apps that I never run in the "real world", to test my system stability. I dont care to know if my CPU can or cant process a silly non-real world app like prime95. In my book, if games run fine, DVD burning runs fine, word and ie run fine, movies play fine, photoshop runs fine, and my games run, then my system is stable.
 
jkb2198 said:
You test for stability using games?

Try using prime95 for 24hrs and get back to us to let us know if it is stable then. As i truly do not deem anything stable till it runs prime95 prority 10 for 24hrs. I do not think the priority comes into to play as much as just running prime95 itself. It is just safe just in case other apps are stealing cpu cycles.

keep clockin

Heres the thing though..you can be prime stable and still crash on 3d apps...I have had systems that ran prime95, folding, cpu burn in...all of them 100% stable...but as soon as 3d marks loads or a heavy 3d game...system would crash...

So you cant really say..."if you are prime stable for xxx amount of time that would be 100% stable"...you really are not 100% stable until you can do both...heavy calucations and 3d apps...

One of the best ways to test for stability is...run folding in the background...or prime95...and play a heavy 3d game or run 3d marks in a loop of 10+...if your system can handle both...then it is 100% stable...

Key thing to remember...Prime95 mainly uses memory...3d apps is CPU/CPU intense...
 
Back