Phenom II 955 BE - almost stable at 4.2 GHz? - Page 3 - Overclock.net

Forum Jump: 
Reply
 
Thread Tools
post #21 of 54 Old 03-26-2011, 05:55 PM - Thread Starter
333mhz
 
pablo.santiago's Avatar
 
Join Date: Mar 2011
Location: Buenos Aires, Argentina
Posts: 67
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 1
Quote:
Originally Posted by sleepergsr View Post
where is the cpu-nb at clock at? 4.0ghz with cpu-nb 2800 beats and overclock of 4.2ghz with cpu-nb at 2.0ghz by a far margin.
CPU/NB Frequency is at 2.7 GHz. You sayin' I should decrease it?

pablo.santiago is offline  
Sponsored Links
Advertisement
 
post #22 of 54 Old 03-26-2011, 05:59 PM - Thread Starter
333mhz
 
pablo.santiago's Avatar
 
Join Date: Mar 2011
Location: Buenos Aires, Argentina
Posts: 67
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 1
Quote:
Originally Posted by falconkaji View Post
Well I've read that there is no point going over 2000MHz, but I haven't noticed any improvement whatsoever in taking it above the stock speed, so I figure it isn't worth messing with. In regards to the max safe temp, you're definitely fine at your current temperature, but I've noticed that I get more stability around 50c than when I let it start heading towards 60c.

If you can't get it stable with that much voltage, it might be time to settle for the best you can get. I run my 955 at 4GHz, 1.44v - it runs relatively cool, and seems happy. Of course I understand wanting to push higher, which is why I had mine at 4.25 for a while. Just feels nice to say you did it, haha.
Haha, totally. In fact, I could keep my current settings and use it as is. No crashing is what really matters (don't do games). But then again, would like to give it a try before going back to 4.1 GHz. It's all about the fun

pablo.santiago is offline  
post #23 of 54 Old 03-26-2011, 06:00 PM
Overclocker
 
sleepergsr's Avatar
 
Join Date: Jan 2010
Location: Orange County, California
Posts: 2,003
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 99
Send a message via AIM to sleepergsr
Quote:
Originally Posted by falconkaji View Post
I think he said in the OP that it is at 2700.

Someone made an interesting thread about the effect of the NB on actual performance, I'll try to find the link.
thanks. i see it now. ht frequency should be close to 2000ghz. its enough bandwidth arlready. i found that anything higher usually cause instability at high overclocks.

sleepergsr is offline  
Sponsored Links
Advertisement
 
post #24 of 54 Old 03-26-2011, 06:02 PM - Thread Starter
333mhz
 
pablo.santiago's Avatar
 
Join Date: Mar 2011
Location: Buenos Aires, Argentina
Posts: 67
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 1
Quote:
Originally Posted by sleepergsr View Post
thanks. i see it now. ht frequency should be close to 2000ghz. its enough bandwidth arlready. i found that anything higher usually cause instability at high overclocks.
HT freq? No, I didn't change that. I left it at stock value (guess 2 Ghz?).
As for CPU/NB freq, I set it to 2.7 Ghz.

pablo.santiago is offline  
post #25 of 54 Old 03-26-2011, 06:05 PM
3.5ghz
 
falconkaji's Avatar
 
Join Date: Feb 2011
Location: Maine
Posts: 629
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
Is the NB frequency the same now as it was when you tested your 4GHz overclock as stable? It seems weird that it would cause trouble now if it didn't before, but then again you never know.

Also, what software are you using to check your core temperatures?
falconkaji is offline  
post #26 of 54 Old 03-26-2011, 06:07 PM
Graphics Card Aficionado
 
trulsrohk's Avatar
 
Join Date: Feb 2011
Posts: 2,329
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 130
Ht will go up with your fsb so you probably need to downclock it a bit to get it back to 2000. As far as the impression that phenoms like multiplier more then fsb clocking....depends on the chip. Mine has a healthy dose of fsb right now and seems to be happier that way.

sneaky.gifAMD 4GHz Clubsneaky.gif
trulsrohk is offline  
post #27 of 54 Old 03-26-2011, 06:09 PM - Thread Starter
333mhz
 
pablo.santiago's Avatar
 
Join Date: Mar 2011
Location: Buenos Aires, Argentina
Posts: 67
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 1
Quote:
Originally Posted by falconkaji View Post
Is the NB frequency the same now as it was when you tested your 4GHz overclock as stable? It seems weird that it would cause trouble now if it didn't before, but then again you never know.

Also, what software are you using to check your core temperatures?
No, at 4 GHz I was using stock speed for CPU/NB. Actually, I tried using stock speed and, as it failed the stability test, that's when I decided to up it.

I'm on Linux, therefore I'm using lm-sensors.

pablo.santiago is offline  
post #28 of 54 Old 03-26-2011, 06:12 PM
3.5ghz
 
falconkaji's Avatar
 
Join Date: Feb 2011
Location: Maine
Posts: 629
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
Are all the cores around the same temperature? Is it a particular one that fails each time or a random one?

Seems unlikely that this would be the problem, but hey, worth asking.

Also, as someone else mentioned above, when you increase the reference clock everything else goes up as well, so if you don't go in and lower your RAM, HT, and NB speeds, they would be slightly overclocked with your 225 reference clock. Again, not sure that this would actually cause a problem, but it is worth double checking.
falconkaji is offline  
post #29 of 54 Old 03-26-2011, 06:57 PM - Thread Starter
333mhz
 
pablo.santiago's Avatar
 
Join Date: Mar 2011
Location: Buenos Aires, Argentina
Posts: 67
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 1
Quote:
Originally Posted by falconkaji View Post
Are all the cores around the same temperature? Is it a particular one that fails each time or a random one?

Seems unlikely that this would be the problem, but hey, worth asking.

Also, as someone else mentioned above, when you increase the reference clock everything else goes up as well, so if you don't go in and lower your RAM, HT, and NB speeds, they would be slightly overclocked with your 225 reference clock. Again, not sure that this would actually cause a problem, but it is worth double checking.
Can't check temp by core, I only get "CPU Temperature" (I know, Linux hasn't the best OC tools, but is very stable).
Ok, so answering your question and taking everyone's else suggestions I tried the following settings: https://spreadsheets.google.com/pub?...UE&output=html

I couldn't even get the temperature for the first 4 tests since it failed on the first Prime95 loop.
Last test is still running but I'm pretty sure it'll eventually fail. Should I keep raising VCore?

pablo.santiago is offline  
post #30 of 54 Old 03-26-2011, 07:08 PM - Thread Starter
333mhz
 
pablo.santiago's Avatar
 
Join Date: Mar 2011
Location: Buenos Aires, Argentina
Posts: 67
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 1
While the test runs there's something really odd I'd like to discuss with you... the voltage I set and the one both BIOS and lm-sensors display is not the same. For example, when I set VCore to 1.525v, BIOS' hardware monitor said it's at 1.476v and lm-sensors 1.43v.
Any idea what could be causing different readings?

pablo.santiago is offline  
Reply

Quick Reply

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off