Sandy Bridge i5-2500k with 4-pin CPU Power Connector

fullofzen

Distinguished
Jan 25, 2011
280
0
18,810
In a few months, I'm planning to pick up a Asus P8P67 Pro and a i5-2500k. My existing power supply (PCP&C Silencer 470) has only a 4-pin CPU power connector. I have done several searches and many have asked whether this will do. What I haven't been able to find is someone saying "Yes, I'm using the 4-pin cpu connector with the 2500k and it's working just fine."

For its part, Asus' P8P67 manual suggests that either an 8-pin or 4-pin is acceptable, but it's just a picture showing how to do it in either scenario. There is no documentation.

To cut unwanted responses off at the path, understand that 470w is not a lot. However, since the i5-2500k is overkill for almost any game I'd want to play in the next few years I'm not looking to overclock immediately, so the Silencer 470 will last me for a while.

So the question: Is anyone currently powering a 2500k or 2600k (or any of the other Sandy Bridge i5s or i7s) with only the 4-pin CPU power connector?
 
Solution
Okay, NO I don't have a 2500 or 2600......

BUT, let me try to explain the electrical engineering principles behind the 4 wire vs 8 wire debate.....

If the power supply can produce the power (Amps...)

Then bottleneck is NOT the for 4 wires running from the power supply to the motherboard. The true bottle neck is the trace on the PCB itself.

Resistance is the culprit here and the heavy gauge wires from the power supply are VERY low in resistance. On the other hand, the flat, thin, metal traces on the circuit board are relatively very HIGH in resistance.

(this is true since the equation for resistance is heavily dependent upon the surface area of the conductor....)

So, the $5 four pin to eight pin adapter in NOT a patch or a...
I'd get a new PSU, you should use a 'real 8-pin' from your PSU especially if OC is the 'goal' by the assumption of your 'K' CPU.

However, there is an adapter 4-pin -> 8-pin. This will stress your PSU longterm.
http://www.newegg.com/Product/Product.aspx?Item=N82E16812201005
12-201-005-02.jpg

 

fullofzen

Distinguished
Jan 25, 2011
280
0
18,810


OC is the goal long-term. I will eventually purchase a new power supply, but I hate to toss a good power supply that will only be 7 months old by the time I go to Sandy Bridge.

Given this, how do you know the 4-pin will stress my PSU long term? There's a single 12V rail on almost all consumer power supplies. To suggest that the power supply is going to be worn-out by the current having to go down 2 wires (with the 4-pin) rather than 4 wires (with the 8-pin) seems, frankly, tenuous and superstitious. Reminds me of Comcast's suggestion that DSL is slower because coax is thicker than UTP...

I don't want to disregard your advice and want to emphasize appreciation for your quick response, but can you provide any documentation?
 
Unless half of the 8-pin is 'capped' by the Mfg chances are it won't run or run properly {depending on the MOBO's Phase circuitry design} with the 4-pin. Using an 'adapter' to split a 4-pin with {X} AMPS into an 8-pin with {X+} AMPS can work. I have NOT looked-up your PSU to 'see' how many AMPS the 4-pin is rated to draw nor the MOBO/CPU AMPS needed particularly with OC nor how many RAILS your PSU has -- I prefer (1)/single rail PSUs. If your PSU is a single rail, I assume it is not, then using an adapter shouldn't be an issue.

Next, if your GPU is equivalent to a GTX 460 {vanilla/average} then you're under the recommend 550~650W with OC.

I am anything if trying to scare you or in particular if you are somehow 'thinking' I am clueless about electrolytic capacitor aging. You need to determine Wattage requirement X 1.2~1.3. In other words 20~30% above minimum requirements otherwise your PSU will indeed - guaranteed - have a short lifespan and worst degrade performance if under sized. ZERO DOUBTS!
 

fullofzen

Distinguished
Jan 25, 2011
280
0
18,810
Right now this power supply is powering a 84w prescott 2m. The 2500k pulls in 135w at full tilt -- and full tilt on a 2500k probably won't even be realized for a few years. We're talking another 50 watts, so an extra 4 amps on a rail rated at 26amps.

Anyway, the question is -- is anyone currently running a 2500k with a 4-pin cpu connector?
 
I've been running an i5 760 (TDP 95W) on a single 4-pin with no issues at all. The i5 2500k also has a TDP of 95W, so I don't see how it will be an issue. According to my Gigabyte mobo manual, you only need the 2nd 4pin connector for the 135W TDP i7 processors. If you are going to seriously overclock, you will also likely need the extra juice.
 

fullofzen

Distinguished
Jan 25, 2011
280
0
18,810


nitrium -- bingo, thanks. Exactly what I was looking for. Don't know why I thought 2500k was 135watts......

What's your PSU wattage? You got a dual or single 12v rail?
 


My PSU is a 5 year old(!!!) Coolermaster 450W, dual 12V rails (rated at 12A and 10A continuous, or 18A and 16A burst) . Easily runs the i5 760, an OCed GTX260, 2 HDDs, and DVD. System draw at the wall at load (stressing both GPU and CPU) is only 275W, idle is about 120W.
I kept the PSU after upgrading the system from an AMD Athlon X2 4200+ (TDP 89W - barely lower than the quad core i5 760!).
 


My Gigabyte mobo manual specifically states that a single 4-pin connector is sufficient for the 95W CPUs. Further, how does insufficient power blow up the CPU??? If the system power draw is too great, it's the PSU that will overheat and die, the PC components should be fine. Anyway, like I said, I wouldn't overclock my CPU on a single 4-pin, but no way is it being overloaded stock - it says right on the PSU itself that it will do 120W continuous (12V at 10A) on a 12V rail, that is clearly ample for a CPU drawing 95W peak load!.
 
I appreciate GA, it utilizes it phases differently than ASUS. Don't confuse TDP with actual CPU use wattage; see TDP http://en.wikipedia.org/wiki/Thermal_design_power the OP CPU will draw 110~120 Watts non-OC.

If your GA had 4-pins covered then you can get by with using 4-pin unless your CPU requires more. The potential problem is if the 8-pin is open some the MOBO's phase power controller 'might' not function properly. Per my initial post the ADAPTER solves this problem ($5).

I have no clue 'why' the OP blew their CPU ; I saw it several hours ago on http://hardforum.com I do recall it wasn't running 'properly' e.g. cores a replaced 8-pin PSU AND new CPU solved the problem.

I've posted here too...ooo long not to see 4 vs 8 pin issues before and the subsequent problems.

My Motto doesn't say Do it half a$$ ... is says "Do it Right..."

I don't care one way or the other what anyone else does with their rig, nor do I give Jerry-Rigged or Poor Advice.
 

jb6684

Distinguished
Okay, NO I don't have a 2500 or 2600......

BUT, let me try to explain the electrical engineering principles behind the 4 wire vs 8 wire debate.....

If the power supply can produce the power (Amps...)

Then bottleneck is NOT the for 4 wires running from the power supply to the motherboard. The true bottle neck is the trace on the PCB itself.

Resistance is the culprit here and the heavy gauge wires from the power supply are VERY low in resistance. On the other hand, the flat, thin, metal traces on the circuit board are relatively very HIGH in resistance.

(this is true since the equation for resistance is heavily dependent upon the surface area of the conductor....)

So, the $5 four pin to eight pin adapter in NOT a patch or a band-aid but will work EXACTLY like a "true 8 pin connector" wired straight from the power supply (anyone who tells you different has only the tiny fractions in the 4th or 5th decimal place to back them up....)


 
Solution
^Don't disagree as long as the 4-pin from the PSU is sufficient for the Amps drawn by the CPU. Essentially what I said, my 'caveat' was/is the PSU rated EATX12 V 4-pin.

What are your thoughts on using ONLY 4-pin and NOT 8-pin? {edit: the OP is sporting a 470W PSU}

If: [PSU] 4-pin (split into 8-pin) rated Amps ≥ [CPU] power requirements
Then: Good.
 

jb6684

Distinguished
I believe the 4 pin vs 8 pin is really all about getting that last tiny bit in your overclocking.

My favorite motherboards are ASUS & Gigabyte. If you look at an particular motherboard, Gigabyte usually runs 8 pin connectors, and, for essentially the Same board ASUS almost always runs 4 pin.

Both boards meet all specs, have great warranties, overclock like crazy, and last a long time.......


SOOOOO, 8 pins is nice, but, I think it's more hype than substance.... (hey, even boards with 8 pin connectors most often "cover" the other 4 pins with a black plastic blank, so, it's Certainly OPTIONAL......

 


For sure, if I had an 8-pin I would use it. If my PSU had 2x4-pins, I would have used them both. Nothing wrong with providing extra headroom! The issue for me was that my (admittedly antiquated) 450W PSU only has a single 4-pin connector. But after looking at the TDP of my previous Athlon 4200+ which was successfully powered (and OCed from 2.2GHz to 2.6GHz) for 3-4 years (on a mobo that only had a single 4-pin header anyway), I had zero doubt that the same single 4-pin would also power my new system. This was verified in the mobo manual, which clearly states a 4-pin is sufficient for the 95W i5/i7's. In so far as it is practical I like to keep using my computer components for as long as they are fit for purpose. A single 4-pin is a 100% sure fit to power 95W TDP processors. It will not, however, give you much scope for overclocking!
 
As I've been TRYING to say, it all depends on how your MOBO is designed 'wired' to handle the incoming power and phases. The capping of 4 of 8 pins is a tell tale 4 or 8 is fine; obviously depending on the power requirements.

Clearly, my first post -- the "adapter 4-pin -> 8-pin" for $5 is the safest option to your expensive CPU. On my i7 it simply will NOT post with the use of the 4 of 8 pins. In addition, I've seen plenty of 4 vs 8 pin failures in Toms. It may work fine, it may 'seem' fine, it may damage the CPU, or it may fail all together. I take the safest option.

Honestly, I DON'T CARE what people do, I've stated my peace.

Pin 1 -> Phase
Pin 2 -> Phase
Pin 3 -> Phase
Pin 4 -> Phase
Pin 5 -> Phase
Pin 6 -> Phase
Pin 7 -> Phase
Pin 8 -> Phase

Pin 1 -> |
Pin 2 -> |
Pin 3 -> |
Pin 4 -> |
-> Phase bank
Pin 5 -> |
Pin 6 -> |
Pin 7 -> |
Pin 8 -> |
-> Phase bank

Pin 1 -> |
Pin 2 -> |
Pin 3 -> |
Pin 4 -> |
Pin 5 -> |
Pin 6 -> |
Pin 7 -> |
Pin 8 -> |
-> Phase bank
 

fullofzen

Distinguished
Jan 25, 2011
280
0
18,810
I know I'm reopening an old thread here, but wanted to post a conclusive answer of my experience on this. Since Tuesday, I have been running a P8P67 (plain-vanilla version; no pro) with a i5 2500k with only the 4-pin power connector provided by my PCP&C Silencer 470 PSU. I did some modest overclocking (40 multiplier; auto-voltage) and the system ran perfectly stable with Prime95 -- if a bit too hot with stock cooler (hyper 212+ on the way as we speak) and the excessive Asus-engineered auto-voltage -- with only the 4-pin power connector.

As a note, the 4-pin CPU power cord is too short to reach comfortably to the 4-pin connector when mounted at the bottom of this case (it was built for a case with both the power connector and the power supply at the top of the case) and hangs awkwardly over the video card. As a result, I have indeed ordered the 4-pin to 8-pin converter as a means to extend the length and avoid infringing upon the GPU's personal space.

I am also sympathetic to jaquith's suspicion that the PCB's "power paths" between the power connector and the voltage/phase regulators may be set-up with a 8-pin connector in mind (less power over more lanes). However, it does not seem strictly necessary for the motherboard to run to have 8-pins connected -- I just don't know how to test this hypothesis without a 5 year study.
 

CtJackal

Distinguished
Jun 1, 2011
3
0
18,510
fullofzen, I hate to tell you this but I blew my first P8P67Pro board up because of a 4 pin connector in the 8 pin socket.
The lanes on the motherboard burned out.
Very smelly.

To the OP.
Get a 4*12v rail modular supply and feed one of them exclusively to the CPU via all 8 pins, it's the only way to be sure.
 

fullofzen

Distinguished
Jan 25, 2011
280
0
18,810
CtJackal -- interesting. I certainly appreciate you providing the input. (I am in the OP, btw). A few questions, actually:

1) What chip were you running? Were you OC'd?

2) What PSU were you using? Did it end up damaged or is it fine?

3) What was the computer up to when it failed? Were you running a full-load with prime95 or something?

3) How can you tell that the 4-pin set-up was what caused the issue? You mentioned that the "lanes on the motherboard burned out." Did you happen to take pictures? I'm really curious of what they would have looked like.
 

CtJackal

Distinguished
Jun 1, 2011
3
0
18,510
Hi fullofzen, it's a 2500k and yes it was overclocked to 4.6ghz at the time.
The PSU is a cheap generic 500W supply which amazingly still works although I have no confidence in it after the hammering it took.
I was showing off my overclock to a friend and span the cores up with Prime95, I had done this previously for a couple of hours with no ill effects, this time there was an almighty POP after about 10 minutes.
The reason I assume it was the 4 pin connector is that I initially tracked the location of the failure by smell, the burnt component/ozone pong was concentrated around the ATX 12V connector and when I removed the 4 pin plug there were scorch marks actually inside the socket.
I have to admit I didn't see any blackened traces on the motherboard but needless to say it never worked again.
The yellow power LED would light up but when I pressed the power button it would "tic" the fans and then nothing.
Exactly the same behaviour that a motherboard exhibits when it has no ATX 12v connected :/