Jump to content

M17x R4 CPU VCore Limit?


Recommended Posts

As I recently shared, I've upgraded my m17x R4 with a 3920xm ES CPU I bought on eBay - and I've been tweaking it a lot in the past days.

I have my multipliers set at 44x1, 43x2, 42x3, and 41x4 cores. This appears to be as high as I can go without significant instability. The current settings ARE stable as long as the notebook is on a good hard surface (I also added bigger feet to the back to aid in air intake) but I'm noticing WHEA error 19's in my event log.

These are processor parity check errors, basically saying it went back and checked its work and found that it got a different result the second time, forcing it to redo the calculation. This signifies that I'm on the edge of instability, and a google search indicated that the usual solution was to bump the CPU voltage up a *tiny* bit - However it appears that I cannot exceed my current 1.331v max (via HwInfo64 monitoring.) I can set the value higher in Intel Extreme Tuning platform and it appears to take the value (it even shows the increase in BIOS upon reboot) but going higher than +105mv (corresponds to 27 "additional turbo voltage" in BIOS) stops yielding higher max voltages in HwInfo64 - and appears to have no effect on CPU stability.

When I look at Intel XTU / ThrottleStop screenshots of this same CPU being used in m18x's, I see vcore voltages of ~1.5v being achieved - enabling notably higher multipliers, but after lots of tinkering I'm unable to exceed 1.331v on my system.

Is this a hardware limit of the M17x R4's mobo? A firmware/BIOS limit? Any suggestions for bypassing it?

Other posts indicate that if I could even ramp up my voltage by something like +.05v, that could be enough to stop the parity check warnings without requiring me to reduce my speed - and I'll admit I'd sure love to see the 46x4 multi's that are being achieved at 1.5v... someday... with an overclocker's warranty from Intel.

Link to comment
Share on other sites

Update: (and possibly the solution to my own question)

I went back an further analyzed the screenshots I'd seen for the 3920xm and took notice of the 120w turbo power limits being used. Now My BIOS implies a long-duration power limit of 85w when you select Level 3 overclocking - and it doesn't allow the user to manually enter anything greater than 80w - so I kind of assumed that was as high as I'd ever be able to go on that setting but I was able to raise both the short and long turbo power limits in Intel XTU to 120w, matching what I see in the screenshot I'm working from. I also matched his additional turbo voltage setting of 195mv (which was previously ignored) and I'm seeing slightly higher voltages in HwInfo64. My multipliers are still the same as before but I'm seeing a max. of 1.351v up from 1.331v on the 85w turbo power limit.

I also copied his setting of 1300 for primary plane current limit in BIOS. Changing that setting in XTU results in a very low number being shown in BIOS upon restart - something silly like 165 - and if I remember it may have crashed when I originally tried to apply it that way, so I believe that limit has to be modified in BIOS to be done correctly on this machine. He also limited his secondary plane (iGPU) limit to 128 in BIOS but I don't want mine that low (fullscreen flash videos run slow) so I'm starting with 256 which equals 32 amps for the iGPU. I'm not sure if that limitation will help me get more out of my IA cores (or whether I was ever really maxing out the iGPU at its default 50a limit) but we'll see how it goes from here...

[Giving credit where it's due, the screenshots I'm working off of are done by Mr. Fox over at Laptop Forums and Notebook Computer Discussion and are available near the bottom of this page: M18x R1/R2 Sandy Bridge and Ivy Bridge CPU Overclocking Thread - Page 159

Link to comment
Share on other sites

Update: I'm passing my stress tests with the new settings, and so far no WHEA CPU parity errors either! Due to the extra voltage, my CPU is throttling down after ~30 seconds at full load test, compared to ~40 seconds prior, but if it's more stable I'll take it!

My next project is going to be a cooling upgrade for this beast. I'm looking into swapping my CPU fan for something with a higher air output than 2.3CFM. I've noticed something interesting - the M14x R2's fan is in a shroud of identical size and shape, with identical wiring appearance, but the blade design is different and it pushes a lot more air! Even limited by the M14x's slimmer exhaust port I can feel there's clearly more air coming out when I manually set both to maximum fan speed. I stumbled across a page somewhere yesterday that claimed the M14x R2's fan is rated at 10.4CFM!! OMG right? The only trouble is I really don't want to cannibalize my secondary laptop for this project and spares of its fan look difficult to come by. Everyone is selling the R1's fan which is only a pitiful 2CFM unit with a blade design like what I'm trying to replace in the M17x.

I'll post more on that later but any ideas/suggestions are appreciated in the meantime! Anyone out there already upgrade their M17x's CPU cooling system? I know the voltage/current requirements are the same for both fans in question (M17x CPU fan and M14x R2 fan) and the connectors look the same, as do the fan shrouds save one mounting screw position, but I wonder how the M17x's mobo will react to this different fan being attached. Like how universal is the speed control/sensor mechanism? I know the 14's fan has a higher top speed for instance.

Link to comment
Share on other sites

Sounds like a fun project. You could call Dell and buy a fan from them, that should work.

2cfm for the M14x sounds weird, my M15x has a 7.8cfm fan, at least for the GPU.

Link to comment
Share on other sites

My biggest concern is will the system recognize the fan, that is a really great idea if it works. The last thing you want is to run a manual switch off the fan just to operate it, keep us posted.

It should be fine. Fans are controlled by a standardized 25KHz PWM signal, so if one fan works then any should work. The only problem could be if you're putting in a 3.3V fan and the motherboard is supplying 5V, but this mismatch is unlikely. Usually fans are rated for 5V.

Link to comment
Share on other sites

The fan is rated at 5v, for the CFM, I can not seem to find specific specs on it, so 10.3 was possibly measured using a personnel air sensor tool, but if the exhaust port is actually smaller in size then that would explain the perception.

Link to comment
Share on other sites

  • 2 weeks later...

Temisan - installing the new XM cpu was very easy in this system. Everything you need to access is easily accessible in the M17x. The only aftermarket/extra thing I used was high-end heatsink paste.

Regarding the fan swap, the M17x mobo CAN control the M14x's fan. That much has been verified. The issue with the first attempt was the difference in fan shrouds between the M17x and M14x. Simply put, the 14's fan shroud will fit into the 17 but it is shorter from exhaust-end to opposite end and none of the screw holes line up. My second attempt will be to adapt for these issues using the 14's original shroud in the 17 because the first attempt tried to free the 14's fan from its shroud and "transplant" it into the 17's shroud. That was an epic fail. The fan lives in its shroud and cannot be removed from one and put back into another without damage - so I have a replacement M14x fan on the way for my next try. Also the top and bottom circular air-intake holes in the 14's shroud are slightly larger than the 17's, which means it will be better to use that shroud anyway. I'll just have to fabricate an adapter to hold it in place. I will post pics n stuff when I get it working!

Link to comment
Share on other sites

The fan is rated at 5v, for the CFM, I can not seem to find specific specs on it, so 10.3 was possibly measured using a personnel air sensor tool, but if the exhaust port is actually smaller in size then that would explain the perception.

The 10.3CFM rating is actually on a yellow factory sticker on the fan shroud itself. That's what I'm going by. There's a similar 2CFM sticker on my M17x CPU fan. I'll admit that great a difference seems hard to believe but those are the ratings on the actual hardware.

I can attest that the M14x fan does truly push more air than the M17x's because I ran both outside the respective laptops for comparison purpose and while outside the laptops the exhaust port size is the same on both fans, but it was still clear by my own observation that the airflow was greater with the 14's fan.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use. We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.