Jump to content
Corsair Community

Big power, what to do.


MaFi0s0

Recommended Posts

Hi, I have researched how to OC already and managed to cook my valueselect by setting the volts just .01 more then what it should of been, I had my 3.6@4.3 but with spontanious reboots at random, no overheating, so im suspecting the ram was the culprit.

 

My motherboard keeps my CPU 10degrees lower then normal, it boots up between 37-41degrees(celc) thanks to the U-plus DPS.

 

My stock FSB etc is

CPU:200x18 MEM:666MHz(altough my ram handles 1GHz).

 

Does this mean my computer should be able to handle:

FSB: 234X18 MEM: 779MHz

?

 

I want to keep ALL voltages normal if i was to over voltage anything it would be the cpu and by .02 volts, I can overclock on the fly within windows using gigabytes easy tune 5.

 

I have artic silver 5 heatsink paste(not on yet though) and my tower consists of 4 fans(2intake 2 exhuast) + PSfan GPUfan Southbridge CPUHeatsink.(8 Fans total)

The highest ive ever gotton my CPU was 55degrees and aparently prescott can handle 74.

 

So i was wondering if its worth trying something like

 

300x18 MEM:999MHz

 

this would overclock a P4 prescott from 3.6 to 5.4 :bigeyes:

 

all my computer specs are in profile, thanks

Link to comment
Share on other sites

Looking at CPU-Z:

My ram runs@ Dual interleaved

Max bandwidth: PC6400(400MHz)

Part number: CM2x512-8000UL

 

SPD timing table

 

Freq 270 400

 

CAS# latency 4.0 5.0

RAS# to CAS# 4 5

RAS# precharge 4 5

TRas# 13 18

 

 

Memory tab:

 

Freq 332.7 MHz

FSB:DRAM 3:5

Cycle time (Tras) 15 clocks

 

CPU

P4 660

Prescott

LGA775

tech: 0.09µ

Voltage 1.344-1.38

 

family F

model 4

stepping 3

ext family 0

ext model 0

revision N0(n and zero)

 

clocks:

bus speed 798.5MHz

mulitplier (jumps back and forth from 14 - 18)

 

Cache

L1 data 16 kbytes (assoc 8way) (line size 64 bytes) (context adaptive)

L1 trace 12 kµops (assoc 8way)

 

L2:

location onchip

size 2048 kbytes

assoc 8way

line size 64 bytes

ratio full

bus width 256bits

prefetch logic yes

Link to comment
Share on other sites

Ok atm i have my cpu@ 4.166GHz

232x18 mem:772MHz

 

any higher and it fails the memory test.

 

now im testing the cpu, i have volted it to 1.4 wich is determined safe by intel.

 

i have had instability even before overvolting it dont know why though!

it reboots when i run 2 instances of CPU burn in is this normal?

i reached 59-60degrees and my motherboard started beeping every 20 seconds or so. then it would reboot.

 

atm i have 1 instance of CPU burn in running with error checking and my cpu temp is 56-57

throttle watch says it has about 52% load, the only way for me to get it to 100% load is by runing 2 CPU burn ins, this is when it crashes.

Link to comment
Share on other sites

Your motherboard starts beeping at you when the CPU overheating protection is enabled. It's shutting down to keep it from burning up. The temperatures you're reporting are from a bit of software inside Windows, no? They're probably lagging just a little bit.

 

Regardless, a 660 runs at 3.6GHz and trying to OC that isn't going to be terribly easy. In truth, the slower CPU's overclock better due to a lower multiplier.

 

Also, don't expect much over 4GHz if you're just cooling with air, these CPU's need better cooling than that for big OC's.

 

Since you're still getting error issues, even with PC6400 memory, I'd say it's not the problem.

 

Looking at your specs, I see one glaring issue... 430 watts is NOT enough for your rig. A single 7800 requires a minimum of 22 Amps on the +12v rail or it can (and will) cause voltage drooping when the system is stressed. To successfully OC, you need a very sturdy power supply.

 

PLUR

CK

Link to comment
Share on other sites

...why?

 

The memory should be able to run up to DDR800 / PC6400 at 2.1 volts @ 5-5-5-12 latencies.

 

As I said, I don't think it's your memory giving you the errors.

 

Also, I'd suggest any PSU with >22 Amps on the +12v rail. Personally, the PC Power & Cooling 510 SLi is what I'd buy in your shoes.

 

PLUR

CK

Link to comment
Share on other sites

I looked through my bios, it doesnt specificly say the voltage it just has either

auto(+0) +.1 +.2 +.3

 

it has this for DIMM and FSB voltage control. i remember it was DIMM i used to cook my valueselect.

 

if you want i can post screenshots of anything including the bios screens

 

atm im asuming the voltage for my mem is 2.1.

where it says DDRV fail, is that defently the ram voltage or something else?

Link to comment
Share on other sites

sandra info

 

Voltage Sensor(s)

CPU Voltage : 1.34V

Aux Voltage : 2.19V

+3.3V Voltage : 3.26V

+5V Voltage : 6.85V

+12V Voltage : 11.92V

-12V Voltage : -1.42V

-5V Voltage : -6.31V

Standby Voltage : 2.85V

Battery Voltage : 4.08V

 

Memory Module 1 and 2

Manufacturer : Corsair

Model : CM2X512-8000UL

Type : 512MB DIMM DDR2-SDRAM

Technology : 8x(64Mx8)

Speed : PC2-6400U 5.0-5-5-18

Version : 1.2

Date of Manufacture : Saturday, 25 December 1999

Set Timing @ 400MHz : 5.0-5-5-18

Set Timing @ 270MHz : 4.0-3-3-12

 

System Memory Controller

Location : Mainboard

Error Correction Capability : None

Number of Memory Slots : 4

Maximum Installable Memory : 4GB

Bank0/1 - A0 : None None None None DIMM 512MB/64 @ 533Mt/s

Bank2/3 - A1 : Empty

Bank4/5 - A2 : None None None None DIMM 512MB/64 @ 533Mt/s

Bank6/7 - A3 : Empty

 

Chipset 1

Model : Giga-Byte Technology 82955x Processor to I/O Controller

Bus(es) : PCI PCIe IMB USB FireWire/1394 i2c/SMBus

Front Side Bus Speed : 4x 232MHz (928MHz data rate)

Maximum FSB Speed / Max Memory Speed : 4x 333MHz / 4x 200MHz

Width : 64-bit

HTT - Hyper-Threading Technology : Yes

IO Queue Depth : 12 request(s)

 

Logical/Chipset 1 Memory Banks

Bank 0 : 512MB DDR2-SDRAM 5.0-5-5-15 1CMD

Bank 4 : 512MB DDR2-SDRAM 5.0-5-5-15 1CMD

Channels : 2

Speed : 4x 193MHz (772MHz data rate)

Multiplier : 10/12x

Width : 64-bit

Refresh Rate : 7.80µs

Performance Acceleration Technology : Yes

Power Save Mode : No

Fixed Hole Present : No

Link to comment
Share on other sites

Also, I'd suggest any PSU with >22 Amps on the +12v rail. Personally, the PC Power & Cooling 510 SLi is what I'd buy in your shoes.

 

 

 

not sure about the amps but the brandnames i can get for dirt cheap are

 

SHAW, ATRIX, AcBel, Thermaltake, Coolermaster, Super flower, Top power, Antec,

ranging upto 600-680watt

 

im gonna email the store maybe they might know

Link to comment
Share on other sites

  • IMHO, the PSU is the most commonly overlooked piece of hardware in a system, just sitting there in the top of our case. I would argue that the PSU is one of, if not the most important piece of hardware in your system. If the PSU dies, your computer dies and what many people don't realize is that if the PSU dies in a way that creates a voltage spike or fluctuations, etc, then it can seriously damage the rest of your hardware. This sort of damage isn't usually covered by warranty.
     
     
  • Not all PSU's are made the same. I wouldn't ever buy a case with a PSU in it because they're usually second rate as far as quality is concerned. Re-read the last point and think about that. Many lower quality PSU's don't live up to the specifications they claim, have weak rails (lower than expected/required voltages), can be a major source of system instability and dramatically shorten the lifespan of your hardware. Just last month, I pulled a proprietary (came with case) PSU from a system and the whole PCB (Printed Circuit Board) inside of it was singed, not from burnt hardware, but because of how cheaply the heat-sink was made. That was a cheap PSU and it actually took two mainboards dying before my buddy decided to buy a new one. The system is now perfectly fine, no dead boards.

 

Bottom Line: You WILL get what you pay for with a PSU. Buy "dirt cheap" and you'll get a piece of crap liable to fail and damage your shiney new computer.

 

'Better' brands would include PC Power & Cooling, Tagan, Fortron Source/FSP, and Sparkle. Of them all, the first two are probably the best, with PC P&P being THE best.

 

PLUR

CK

Link to comment
Share on other sites

When i say i get them dirt cheap, i dont mean there cheap brands, i mean i get them just above wholesale cost, i got 2x512mb Valueselect ram for $135AU ($95US). im sure some of the brands listed are good brands seeing as though thermaltake is arguably the leader in cases, and coolermaster seems to have some sexy looking aftermarket parts.

if i was to go with any other brand id probly be paying almost twice as much.

Link to comment
Share on other sites

Actually, some Thermaltake PSU's are worthless and some are good, so you've got to be careful which you go for. I have my TWV due to a great price, but I still felt like I was taking a chance when I got it.

 

I would not buy a Coolermaster PSU and the Antec PSU's have been going down in quality for quite awhile now.

 

PLUR

CK

Link to comment
Share on other sites

i was playing CSS yesterday and it rebooted wich is just more evidence its the power.

 

the shop didnt know about amps, the thermaltake site doesnt mention amps, i spoke to my cousin who is an eletrician and he said a computer should only need 2 or 3 amps, and that 22 amps would be equivilant to something like 20 wall mounted power points.

 

http://www.thermaltake.com/purepower/w0049/w0049.htm

 

im thinking about that PSU because i will be getting a 2nd dvd burner and 3rd HDD and a tv tuner fairly soon, plus in a year or so will be upgrading to dual core, raid0, and bigger ram, and i forgot to mention i have 5 usb slots in use,(i know it sounds dumb not upgrading psu sooner but i had no problems till now).

 

with all the features that PSU claims to have ontop of all the awards, i cant imagine it not doing the job sufficient.

Link to comment
Share on other sites

It might not be the PSU causing the reboots, im beta testing kaspersky anti virus 2006 and one of the bugs at the moment seems to be spontanious reboots, come to think of it i dont think the reboots are related to the overclocking at all, because i get "Crash caused by faulty device driver", but could the PSU still be the reason for the fail of mem test @ 233FSB or higher?
Link to comment
Share on other sites

Yes, the PSU could cause failures under duress... you stress the system and increase it's power consumption when overclocked.

 

As for Amps... heheh. I figured you'd realize that every PSU lists the Amperage per rail (I.E. 12 volt rail) on the sticker on the side.

 

Your computer will only use 2-3 Amps at 120 volts, but we're only talking 3, 5, and 12 volts. FYI, 12v @ 20A = 120v @ 2A if memory serves me.

 

PLUR

CK

Link to comment
Share on other sites

  • 1 month later...

Ok, after searching high and low and a long wait wich ended in disapointment i have hooked up a 2nd PSU to my comp, i have hooked it upto the GFX card and DVD burner. i know the GFX card requires a PSU of atleast 400w plus it has 2 plugs for power. so this should take alot of strain of the 1st PSU, unfortunetly i am still getting DDRV FAIL could i have bad ram? i am going to contact my motherboard manufacturer and ask them what exacly DDRV fail means.

 

also, im thinking maybe i haveto overvolt my ram to get it past 700mhz?

is this safe, my ram cost me $415AU so i dont want to cook it like i did with my valueselect...

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...