DFI P35 Value Motherboard, Infinity INF P35-T2RL BLOOD-IRON

[Sailindawg]

@Motf

You have coils on that board in 2 locations. One is right above the 24 pin power connector, the second is below the IDE header.

Try to isolate the squeal. Usually, it's the coils vibrating. The black boxes around the cpu, with R35 are the PWM for controlling 12V voltage to the cpu. About the only thing you could try those is to put a RAM sink on each one for cooling.

[Lome]

those "black boxes" are coils

[Sailindawg]

The copper wire wrapped doughnut shaped coils are usually the culprits on DFI's boards that start to squeal when raising the volts and overclocking. These coils oscillate as the voltage increases.

[TSFroggy]

I have been watching diligently for information involving this motherboard and not much has come up. It has gotten great reviews but not much other information is coming out of the woodwork about it. Would there be much of a difference running this over a Lanparty UT P35-T2L?

[Sailindawg]

Performance wise for the average user, that remains to be seen. I would rather see consistently good overclocks that are stable, rather than suicide shot overclocks. The BIOS, based upon screen shots, for this board seems to be up to par with the LP P-35.

[TSFroggy]

Performance wise for the average user, that remains to be seen. I would rather see consistently good overclocks that are stable, rather than suicide shot overclocks. The BIOS, based upon screen shots, for this board seems to be up to par with the LP P-35.

Agreed. I was even surprised they allowed GTL reference voltage manipulation on the Infinity version...

[MotF Bane]

Agreed. I was even surprised they allowed GTL reference voltage manipulation on the Infinity version...

I haven't found GTL voltage adjustment on my board.

[wtz54321]

I haven't found GTL voltage adjustment on my board.

Find the latest beta bios , it should be there ~

[Bbq]

Thinking of buying this board to replace my now-dead P5B Deluxe..

Think it's a good choice? I'm looking to hit 9x400 (3.6ghz) on an E4300 (no problems on the p5b deluxe).

[CyberDruid]

Should have one of these in hand in a few days and I will post up some results. I think it's going to be a sleeper. I like the BIOS--thanks for all the pics Andre. But man those blue slots are driving me nuts...I might have to get some black covers or something...

Yellow and Orange or Green together seem about right for DFI...not Red and Blue...and White....and Black...

I like the Red PCB...but whatever...as long as it cranks out the WUs...

Does anyone know if the board is UV reactive? It's common knowledge UV increases ability to OverClock (especially in the dark)...

[GAR]

very clean and simple board, love it.

[RLM]

I've just got this board on loan until my P35-T2R replacement comes in. Have to say I'm pretty impressed

Just a few points relative to my experiences with the P35-T2R. Obviously feel free to correct me on these, if I'm way off

Negative points for me.

1. Feels a bit flimsy compared to the P35-T2R. Especially when inserting ram and the 24pin power connector. The right side of the board overhangs the mounting standoffs on my case by about an inch or so.

2. Bios is a bit limited.

a. No save profiles option, so when the board does on occasion decide to reset to default, it means entering all your settings again.

b. There's no vdroop control, and from running some early tests I'm getting about -0.06 droop.

c. Vcore selection takes a bit of getting used to. It requires setting additional voltages, rather than entering a specific value.

Positives.

1. Vdimm droop is about -0.02, and what you set is what you get (pretty much)

2. No squealing (Yet)

3. Seems to run pretty cool

4. Memory so far is running stable at cas 4. For instance my current run is 400x9 memory at 1000mhz 4-4-4-12-36 with EA and EDT set to FAST. Performance Level 5. No problems at all. A 1m PI run comes in at about 14.016secs. All good

5. Decent clearance between the graphics card slot and memory

I've only been running this board for a short time, but it's encouraging. With a vdroop control option or a mod, it would be very good indeed.

Right then. Here's a small FTT run. Only a quicky of about 18 minutes, as I'm an impatient sod.

You'll have to take my word for it, but the vcore actually sat at 1.34v for most of the run, fluctuating occasionally to 1.36v as per piccy.

Prime Small FTT run Q6600 GO 400x9 3.6 Ghz Memory 1000mhz 4-4-4-12 EA Fast EDT Fast


RLM

[GripS]

Like you said. Only 14 minutes. But even at that. Only 1.36v for 3.6ghz!! Very cool.

[RLM]

Like you said. Only 14 minutes. But even at that. Only 1.36v for 3.6ghz!! Very cool.

Don't want to speak too soon, but it's looking good so far. Cheers

Here's a 32 minute blend test. Maybe pointless in some people's eyes.

Spoke a bit soon about the memory. Had to up the vdimm just a tad, but glad to see it responded. Vdimm is set to 2.25 in bios. Reads 2.2v in SMG at idle and 2.19v at load. Gskill 6400 GBHZs

Prime Blend Q6600 GO 400x9 3.6 Ghz Memory 1000 mhz 4-4-4-12-36


RLM

[Inteleron]

You know its all about the ECS purple pcb

[h3x0r_1984]

Has anyone overclocked a quad on this board for a decent amount of time?Maybe someone bothered to measure the vcore ripple so we can see if there's any problem with the 4phase vcore circuit.

[RLM]

After a decent start with this board, it's gone down hill. The droop has got worse now drooping by as much as -.15.

I think in answer to your question, it isn't really up for the job of running a quad.

I took out my cpu and ram, and tried it on a friend's DFI P35-T2R today. It performed pretty well, happily priming away for half an hour or so at 3.825 ghz with a vcore of 1.47v.

On the blood I'm struggling now to achieve 3.6 stable, with 1.47v set in bios. Just too much droop.

RLM

[Praz]

I think in answer to your question, it isn't really up for the job of running a quad.

The specs on the PWM circuit didn't seem like it was going to be up to the task of pushing a quad. I've had similar results with the 965-S.

[RLM]

Let me just correct my previous post.

At 3.4 guz, I've just watched my vcore drop from 1.47v down to 1.27 gradually over the period of a minute, before prime failed. Given I want to use this kit for 3d rendering, it's basically unusable

In comparison I ran my Quad GT with a QX 6700 overclocked to 3.55 ghz with 1.55 volts, and had no troubles at all.

Basically not too impressed.

RLM

[saint-francis]

I have this board 9 hours prime stable at 395 x 9 fsb with 1.42 volts. I'm using a q6600. getting there took me about an hour or two. Getting to 400 fsb is a little more difficult but very doable. I like this board. I have encoded video with it for 19 hours straight with out any problem at all.

[RLM]

Likewise my initial impressions were very good. Thought I was on to a bit of winner with this board. In fact it was a piece of cake to get to 400x9. You can see by my results a few posts ago, that it was running okay. I know not 9 hour tests

Did a few hours of 3D rendering on it a few days ago 5000x5000 pixel images, which it eat through like a champ.

However now, Kaput. Last prime was at 2.8 ghz. 1.35v, which dropped to 1.26v. It gave one last squeak and then died on me. No longer boots.

I did actually test those 3d renderings again at 800x600 prior to it dying on me. Less of a clock, and the machine crashed.

As I said previously, I've tested the cpu and ram on a DFI P35-T2R and it's all good. In fact that's what I'm using right now.

Anyway good luck to you saint-francis, hope you have better luck than me

RLM

[celemine1Gig]

...
At 3.4 guz, I've just watched my vcore drop from 1.47v down to 1.27 gradually over the period of a minute, before prime failed. ...

There's either something major wrong with your board, or your multimeter is giving in. A Vdroop of 200mV isn't really within the normal range.

How was the PWM temperature when that happened to your board? Was it extremely hot or just normal temp?

I'm really shocked for now, but as your see, there are others (post just above me) who fortunately don't seem to experience such a huge Vdroop.


Edit:
Just had a quick look at the VRD11 design guide by Intel and guess what, their loadline window spec says that at the max specified 125A or current draw, a VDroop of upto 188mV is within spec.
So I guess you just passed that point, although it seems quite unlikely. That would mean over 180W of power dissipation.

Edit2: Just saw your last posting. Had missed it. That proves that the board was bad. Sorry to hear nonetheless. Hope you'll get a replacement or your money back as fast as possible.

[RLM]

There's either something major wrong with your board, or your multimeter is giving in. A Vdroop of 200mV isn't really within the normal range.

Definitely something majorly wrong with the board. It's not working anymore. (I shouldn't laugh)

It's the way the vdroop increased over time that was a little odd. I don't understand the ins and outs, you've seem to have a better idea than me, but obviously something had packed in.

Temps were fine. Cpu under water. Fans over the Northbridge and pwm area, aircon on, nothing out of the ordinary in SMG.

That's all by the by now. All packed up now, and ready to return.

RLM

[boblemagnifique]

I recieved my parcel today

[Okda]

damn you :P