2500 barton temp

[theboogieman]

Help i've got it upto 2758mhz but it's running at 61c without load any idea's how to get the temp down, it's doing 197fsb x 14 and is stable on my msi kt6 delta (msi's own overclocking board)

[Jay23]

woah dude... Watercooling maybe? Maybe some hefty air cooling, Delta case fans! and an slk-800? unless theres a better HSF combo out there im unware of(most likely), someone else wiill help ya... im just going on arnt I?

[whatever2003]

is stable on my msi kt6 delta
Since you have 61C idle I am assuming you did not even try Prime95. Run that program's torture test for three hours. Then you can claim it is stable.

How many times people claim the setup is stable but it cannot run 3 minutes of Prime95? Stability is very tricky thing, it is a perception that means very little, the less you know the more stable it looks.

[Gautam]

2758? Wow. Even if its not the slightest bit stable, its beyond commendable that it even manages to boot at such a high speed. Looks like its got a lot of potential. What core voltage and cooling are you using at the moment?

[theboogieman]

I'm using vcore of 1.62 and got a cooler master fan and it sounds like a leaf blower runnina at 4900rpm lol

[whatever2003]

its beyond commendable that it even manages to boot at such a high speed.

It certainly is. The good question is how many times he will be able to boot it. In AMD systems the temperature gets reported in two different ways. Abit boards like NF7-S underreprt it for a whopping 10-15C. Some MSI boards report it correct some like K7N2G underreport it for more than 10C.
In other words, 61C might be really very tentative. ;)

[Gautam]

Sounds like you've got a very very nice processor on your hands. I highly doubt that it would be stable at that speed, though. If it is, don't worry about temps; they are meaningless. Use Prime95 to check for stability. And if you really can boot at that speed, I'd second the watercooling suggestion; it will unlock the full potenital of your processor.

Edit: MSI board's use on-diode thermistors, hence are quite accurate. For such a low amount of voltage, 61 is about what one would expect. I take a liberal stance to what is "tentative." My processor has ran has high as 110°C, and still works perfectly. AMD rates 95°C as the maximum safe temperature; anything below this will not damage the processor in the slightest. The only problem is maintaining stability at such high temperatures, which is nearly impossible. But damage will not occur.

[whatever2003]

Can you post a shot of CPU-Z of that setup?
My processor has ran has high as 110°C, and still works perfectly. AMD rates 95°C as the maximum safe temperature; anything below this will not damage the processor in the slightest.
Yeah, right. :D :D
As if I didn't burn any CPU before. Be careful for what you wish.That is a silicone not wolfram or titanium.

Beside 85C (not 95C) is maximum die temperature for Barton 2500, as said by AMD. That is ... the "max die temperature" not the "safest die temperature".

[Gautam]

I understand your finding it hard to believe. Granted, it was for under 1 minute, but my point is that even temps in the high 60's to low 70's will not cause any physical damage. Of course, no one really has had a chance to test this, as its not even normally possible to run stabily at such high temperatures. Although, on stock cooling, mid 60's are common at stock, and remain stable.

[whatever2003]

Every time you shave you cause thousands of micro injuries to your skin that you do not notice. Your skin can heal afterward. Silicone cannot.
Meaning every time you USE you processor you are causing damage to it. Using it at 50C these damages will not be serious and your CPU will last very long. Every time you increase temperature for another 10C the life of that CPU will be cut in half. And this is about "safe temperatures". Once you reach the burnout zone, the situation changes, the CPU will burn out.
It might take 10 sec, it might take 2 mins, or even longer. Sure there are some that can resist more than others, but we are not talking about wolfram here, again.

[Gautam]

Agreed. The electromigration process is ever-happening, and only accelerated by higher temperatures. But, processors are rated for such a high life, that high temperatures will cause their life to be cut in such a way that few would notice. No one actually uses their processors for more than a year or two at a time. Thus, its not a heavy concern. The main concern is stability. If your system is stable, the chances of you being at a dangerous temperature are next to zero.

[theboogieman]

Well it was running well but the video card seems to object to the extra heat lol so i need to get a fan for that aswell now but it is running at 2484 (216fsb x 11.5) and it is reporting a temp of 49 at the moment and will stay that way for now with any luck

[whatever2003]

300 MHz drop for 12C?
That is still remarkable too.
How about posting Prime95 running and CPU-Z on the same screenshot?
You apparently have a remarkable processor, why not to post a proof for all to see.

[Deadly Dude]

running at 61C idle isnt too bad, my 1800+ has been running at about 67C load for 1.5 years now and theres no damage to the cpu.

[whatever2003]

Is that Palomino or Thoroughbred?

[Lithan]

"Abit boards like NF7-S underreprt it for a whopping 10-15C."

*Cough, Cough*BS*Cough*

"Every time you shave you cause thousands of micro injuries to your skin that you do not notice. Your skin can heal afterward. Silicone cannot."

Yes and we all know that electricity behaves exactly like a razor. Let's all stick our faces in a bathtub and toss a toaster in the water to shave from now on.

"Meaning every time you USE you processor you are causing damage to it."

Everytime you use ANYTHING that wears out this holds true. What's your point?

"Using it at 50C these damages will not be serious and your CPU will last very long. Every time you increase temperature for another 10C the life of that CPU will be cut in half."

What are you baseing this on?

"And this is about "safe temperatures". Once you reach the burnout zone, the situation changes, the CPU will burn out.
It might take 10 sec, it might take 2 mins, or even longer. Sure there are some that can resist more than others, but we are not talking about wolfram here, again."

The system will usually crash or shut down before this ever becomes an issue. And I've not seen an amd crash due to temps at anything under 84*C (reported by a sensor that actually was inaccurate, a shuttle kt266).


Seems to me that you can't overclock your bartons worth a damn, so you go around, dead set on convincing everyone else that their overclocks aren't real. Because god knows, your experiences are the science of the universe. People are hitting 2.7-2.9+ghz on bartons with extreme cooling (sustainable). People are hitting 2.4ghz at stock voltage. People are hitting 2.2ghz at stock voltage with stock cooling. Some people even hit 2.3 on stock voltage and cooling. Just because you don't want to believe it doesn't change anything. Of course these aren't guarenteed overclocks. There is no such thing. Not in bartons or in your 2.4c's.

[pacino]

Originally posted by Lithan Seems to me that you can't overclock your bartons worth a damn, so you go around, dead set on convincing everyone else that their overclocks aren't real. Because god knows, your experiences are the science of the universe. People are hitting 2.7-2.9+ghz on bartons with extreme cooling (sustainable). People are hitting 2.4ghz at stock voltage. People are hitting 2.2ghz at stock voltage with stock cooling. Some people even hit 2.3 on stock voltage and cooling. Just because you don't want to believe it doesn't change anything. Of course these aren't guarenteed overclocks. There is no such thing. Not in bartons or in your 2.4c's.

And you find this chip/OC is real? 2.7 on almost default vcore? give me a break!

[gazzrawly]

right erm...i wud say anything above 60*C or even 60*C would definitly be my limit, if not less......I call 50*C my limit and atm i run 2200mhz (10X220) and its fine @ 41*C idle and 45/46*C Load..I feel this is quite an achievemtn @ 1.725Vc 1700XP @ 3200XP on a chip that cost £38 :) w/o screenshots i still dont believe a barton 2500XP will run @ 3600XP+ w/o a voltage increase (or 1.625V whatever u said) i feel thats near impossible or u have the best chip ever produced by AMD :)....SCREENSHOT AWAY!

[PCluva]

Originally posted by pacino


And you find this chip/OC is real? 2.7 on almost default vcore? give me a break!

I agree, need a screeny!!

I cant believe any chip can load an O/S with an overclock like that without a higher voltage and some very good cooling, so if you did get your chip that high and only saw the POST and BIOS you are not stable, I wouldn't say a system is stable until it does some long hard work without a crash

[userA]

Barton 2500+ @2758mhz with only 1.62v??? Regardless of how the CPU is cooled, I don't believe it is possible with Vcore 1.62v!

[Cytomax]

Just to add some info i know of a friends computer that ran at about 55C for about 1 yr and still no prob so i wouldnt recomend it but if its stable at 60 C then more power to you
Eddie

[RedDawg41]

why such disbelief? when boogieman gets time, asks about his proc stepping AQXEA ? AQUCA? which week? 0330? 0320? 0333?
depending on his setup, assumptions about his particular board and cpu not getting 2.7GHz may be wrong. Could be a type-o error... I wouldn't step on someone's posting on what he got his cpu up... let see if he's able to post up some screen shots and applaud his effort if true. Find out what other little Leprechaun's are in his rig or hanging around the cpu. :)

[userA]

You're right. I shouldn't make any assumption though it seems impossible. I too would like to see a screen short :)

[theboogieman]

Sorry i didnt make it clear the vcore @2700 was 2.1v and 1.62@2300

[Steven4563]

well for saying things about temps i ran a athlon 850 @ 900mhz at temps off 68Cidle before i even started to come to overclockers and it ran like that for about 8 months and i killed my msi bios chip :) lol and to this day my mate has got it running at 900mhz with no problems and thats about 4 years old :)

[RedDawg41]

cool beans, tks for the info.
now, getting back to your 2.1v for 2.7GHz, that would explain why you're getting temps of 61c, which is common in that part of the statusphere.
check this site out...
http://www.benchtest.com/calc.html#stress
this will give you some insite on your cpu, which by using 2.1vcore
your cpu is putting out 163.2 watts of microwave fun!
Compare this to the normal 68watts of dissipation of energy, you've increased it almost by 10times the normal amount. I'm surprised that you're coolmaster alone transferred heat that effective for you to boot into windows. Please download CPU-Z v1.19a from cpuid.com. there's an option to quickly annotate all you rigs info on html format. this will record your rig's settings and such. Then you can comeback an posts up your info back here.

[RedDawg41]

CPU-Z Report

--------------------------------------------------------------------------------

CPU-Z version 1.19a.


CPU(s)
Number of CPUs 1

Code Name Thoroughbred
Specification AMD Athlon(tm)
Family / Model / Stepping 6 8 1
Extended Family / Model 7 8
Technology 0.13 µ
Supported Instructions Sets MMX, Extended MMX, 3DNow!, Extended 3DNow!, SSE
CPU Clock Speed 2481.1 MHz
Clock multiplier x 11.0
Front Side Bus Frequency 225.6 MHz
Bus Speed 451.1 MHz
L1 Data Cache 64 KBytes, 2-way set associative, 64 Bytes line size
L1 Instruction Cache 64 KBytes, 2-way set associative, 64 Bytes line size
L2 Cache 256 KBytes, 16-way set associative, 64 Bytes line size
L2 Speed 2481.1 MHz (Full)
L2 Location On Chip
L2 Data Prefetch Logic yes
L2 Bus Width 64 bits



Mainboard and chipset
Motherboard manufacturer http://www.abit.com.tw/
Motherboard model NF7-S/NF7-M/NF7 (nVidia-nForce2), 1.X
BIOS vendor Phoenix Technologies, LTD
BIOS revision 6.00 PG
BIOS release date 03/17/2003
Chipset nVidia nForce2 rev. C1
Southbridge nVidia MCP-T rev. A3
Sensor chip Winbond W83627HF




Memory
DRAM Type DDR-SDRAM
DRAM Size 512 MBytes
CAS# Latency 2.0 clocks
RAS# to CAS# 3 clocks
RAS# Precharge 3 clocks
Cycle Time (TRAS) 8 clocks
# of memory modules 2
Module 0 Corsair DDR-SDRAM PC3200 - 256 MBytes
Module 1 Corsair DDR-SDRAM PC3200 - 256 MBytes



Software
Windows version Microsoft Windows XP Workstation Service Pack 1 (Build 2600)

[RedDawg41]

cool huh?

[RedDawg41]

CPU-Z Report
running @1.95vcore
--------------------------------------------------------------------------------

CPU-Z version 1.19a.


CPU(s)
Number of CPUs 1

Code Name Thoroughbred
Specification AMD Athlon(tm)
Family / Model / Stepping 6 8 1
Extended Family / Model 7 8
Technology 0.13 µ
Supported Instructions Sets MMX, Extended MMX, 3DNow!, Extended 3DNow!, SSE
CPU Clock Speed 2539.4 MHz
Clock multiplier x 11.0
Front Side Bus Frequency 230.9 MHz
Bus Speed 461.7 MHz
L1 Data Cache 64 KBytes, 2-way set associative, 64 Bytes line size
L1 Instruction Cache 64 KBytes, 2-way set associative, 64 Bytes line size
L2 Cache 256 KBytes, 16-way set associative, 64 Bytes line size
L2 Speed 2539.4 MHz (Full)
L2 Location On Chip
L2 Data Prefetch Logic yes
L2 Bus Width 64 bits



Mainboard and chipset
Motherboard manufacturer http://www.abit.com.tw/
Motherboard model NF7-S/NF7-M/NF7 (nVidia-nForce2), 1.X
BIOS vendor Phoenix Technologies, LTD
BIOS revision 6.00 PG
BIOS release date 03/17/2003
Chipset nVidia nForce2 rev. C1
Southbridge nVidia MCP-T rev. A3
Sensor chip Winbond W83627HF




Memory
DRAM Type DDR-SDRAM
DRAM Size 512 MBytes
CAS# Latency 2.0 clocks
RAS# to CAS# 2 clocks
RAS# Precharge 4 clocks
Cycle Time (TRAS) 8 clocks
# of memory modules 2
Module 0 Corsair DDR-SDRAM PC3200 - 256 MBytes
Module 1 Corsair DDR-SDRAM PC3200 - 256 MBytes



Software
Windows version Microsoft Windows XP Workstation Service Pack 1 (Build 2600)

[ogboot]

that tRas is an awful lot of clocks, would it be stable at about 6?

[RedDawg41]

It might be awful, however, I'm running PC3200 ddr ram, which is already out of specs... this is my sweet spot where I'm not incurring any instability at this time, inwhich i'm running above 200FSB.

[okayfine]

who cares if you are damaging the cpu. If you shorten their life by 1-2 years, no big deal, coz they are so cheap nowadays. Just updrage to a new cpu in 2years.