|
Not trying to claim overclocking OG status or anything (I was there when you could overclock so long as you had a pencil  )... And I understand that it's basically the huge push for much more affordable, non-confusing overclocking that's leading everyone to hop on board (which really makes the clock rate a peculiar commodity item and makes those who buy anything that could have a K on the end, but doesn't, seem really off in my eyes). But thinking back to when overclocking was WAY more of a pain in the rear end, it's a little bit adorable to see people talking about how the UEFI we get today is too confusing. I mean it even has a mouse, and you change one thing for the simple overclocks, and only a few, comparatively speaking, for the complex ones... It's beautiful.Hell of a long way from all the pain in the rear end involved in screwing with the front side bus and making sure your processor and RAM frequency lined up appropriately for the type of setup you were running. Everything's so easy these days. I love it. I don't fault anyone for using software, apparently it's got better at its job so go hog wild - but the UEFI is just so elegant, and there's so much you just do not have to think about or worry about at all (regardless of whether you do BIOS or operating-environment overclocking) that it's pretty mind boggling. Like, the decision to get stupid high clocks pretty literally comes down to "how much do I want to spend on my motherboard and my heat sink?" and that's cool.
|
#
?
Jun 8, 2012 04:24
|
|
|
|
| # ? Apr 29, 2024 20:29 |
|
|
Agreed posted:it even has a mouse Once I stopped trying to use the mouse and treated it more like a classic bios it got much simpler. Changing settings with a mouse makes me stop thinking or something. Modern memory technology is really fantastic though, I'll be glad to never mess with the balance of dividers, FSB black holes, and timings that change on every motherboard even when you keep the same sticks.
|
|
#
?
Jun 8, 2012 04:38
|
|
|
I just upgraded from a P55/i5-750 to a Z77/i5-3570K and I still can't believe it's overclocking itself to 3.8GHz at 1.0V  I've never seen a Vcore that low. Problem is since I'm a few generations behind, I have to relearn what all these new BIOS options mean, but with a 212+ (100% load on all cores running prime95 topping out at ~45C) I feel like I have so much headroom to go hog wild 
|
|
#
?
Jun 11, 2012 22:20
|
|
|
Watch the temperatures fly above 1.2V.  On the other hand, I've seen that some chips can make it well over 4 GHz at stock Vcore.
|
|
#
?
Jun 11, 2012 22:35
|
|
|
Factory Factory posted:Watch the temperatures fly above 1.2V. I have no idea what stock Vcore is, really (which was why I said "gently caress it, let's put it on 1.0V and see what happens). First time boot, I didn't mess with any BIOS settings, so the voltage, Bclk, and multipliers kept changing depending on how much load was on. I'd like to set the Vcore manually to whatever "stock" is and see how high I could push it, though. And setting it manually kills any of the power saving features, like on Sandy Bridge, right? I'd have to figure out where the voltage offset options are, and go that way, wouldn't I?
|
|
#
?
Jun 11, 2012 22:41
|
|
|
It's mixed. The processor has a fine-tuned "default" full-load VID that was set during binning, but it's the motherboard that controls voltage. The default VID can vary, and differences in manufacturing on the board can vary, so the "default" voltage can change on the same chip with different motherboards (even of the same model), and on the same board with different chips (even of the same model). Most IVB CPUs come with a default Vcore somewhere a bit north of 1.05V. They undervolt swimmingly - AnandTech's sample for their overclock/undervolt review managed 3.9 GHz undervolted as far as it would go, 0.9V. I'd recommend reading that linked article, by the way. But the "default" voltage is really pretty worthless except as a curiosity once you're overclocking - stability and safety is what matters. As for power saving, a Sandy Bridge-like "manual" voltage would kill off EIST, yes. But there's a wrinkle involved with Ivy Bridge motherboards: a number of them (mostly MSI) have abstracted away Offset voltage. Instead, they just read the "default" voltage off the CPU and bake in the offset, so you set an explicit VID target and the board handles the proper offset for that itself. If your board still offers Offset, then just do Offset +0.000, and that's the default. Simple. Make sure LLC is off, though, until you need it.
|
|
#
?
Jun 11, 2012 23:00
|
|
|
That's definitely an interesting read. I'm going to need to spend more time getting acquainted with the P8Z77-V LX BIOS (I'm still floored at mouse support in a BIOS) so I can figure out which settings change what. The Kingston RAM I bought is 1600 but at 1.65V. It defaults to 1066 at 1.5V in the BIOS though and I have to manually change it to 1600@1.65V and the OP says anything over 1.58V is dangerous? Should I downclock it to 1066@1.5V and just let it climb up as I overclock (or is that not how it works anymore)?
|
|
#
?
Jun 11, 2012 23:15
|
|
|
Fhqwhgads posted:That's definitely an interesting read. I'm going to need to spend more time getting acquainted with the P8Z77-V LX BIOS (I'm still floored at mouse support in a BIOS) so I can figure out which settings change what. Memory voltage is fixed; you only need that higher voltage because Kingston says you do for stability at that speed. I don't think memory overclocking is worth it, to be honest. If you have 4 sticks, I'd just bump up to 1.52V or so to improve stability and leave memory at stock speeds.
|
|
#
?
Jun 11, 2012 23:21
|
|
|
movax posted:Memory voltage is fixed; you only need that higher voltage because Kingston says you do for stability at that speed. I don't think memory overclocking is worth it, to be honest. If you have 4 sticks, I'd just bump up to 1.52V or so to improve stability and leave memory at stock speeds. Thousand times this. I do stuff that involves memory and noticed no qualitative performance increase with the pain in the rear end and higher voltages (both to the dimms and VCCIO) involved in getting it to 1T from the stock 2T. Might be worth 2% difference on a gigantic decompression? Maybe?
|
|
#
?
Jun 11, 2012 23:23
|
|
|
It won't climb, as you won't be adjusting BCLK (the reference clock for all other buses now that the FSB is gone). I would return the RAM, actually, but if you're determined to keep it, I'd just do 1.58V and whatever speed you can get out of it. It may do 1333 (use memtest to check it) just fine. Technically, the voltage limit on RAM is based on maintaining a less-than-0.5V difference between the DRAM itself and the memory controller. But I wouldn't start fiddling with VCCIO or VCCSA voltages just for that RAM.
|
|
#
?
Jun 11, 2012 23:23
|
|
|
Factory Factory posted:It won't climb, as you won't be adjusting BCLK (the reference clock for all other buses now that the FSB is gone). I would return the RAM, actually, but if you're determined to keep it, I'd just do 1.58V and whatever speed you can get out of it. It may do 1333 (use memtest to check it) just fine. I wasn't planning on overclocking the RAM, I was just used to it climbing since I'm used to adjusting Bclk. What's funny is that Corsair Vengeance 8GB kit that the parts thread is telling people not to buy (because it's loving expensive and has the huge fins, etc) is actually on sale right now for cheaper than the Kingston set I bought. At least that RAM is listed as 1600@1.5V. ~$50 on Amazon isn't bad.
|
|
#
?
Jun 11, 2012 23:30
|
|
|
Fhqwhgads posted:I wasn't planning on overclocking the RAM, I was just used to it climbing since I'm used to adjusting Bclk. Yeah, the Vengeance occasionally crashes in price below the more "normal" RAM. At that point the only real disadvantage are the obnoxious fins, as you said.
|
|
#
?
Jun 11, 2012 23:36
|
|
|
Word of warning, Amazon sometimes doesn't know what the gently caress when it comes to RAM. They've sent me the wrong kind of RAM twice within a week of placing the orders. Granted, the only difference is printed on a rather small sticker, not in big typeface 9-9-9-24 1.5V as opposed to 9-9-9-24 1.65V... But that's a mistake that could, maybe, possibly kill an Ivy Bridge processor? 1.65V is beyond the excursion range of the system agent and memory controller alike innit?
|
|
#
?
Jun 11, 2012 23:40
|
|
|
Yeah, those voltages are typically 1.1V, which 1.65V outstrips.
|
|
#
?
Jun 11, 2012 23:56
|
|
|
Turns out I can manually set the RAM to 1600@1.5V without issue, so I'll keep it for now. You know, I'm not new to OC'ing, but obviously new to this generation's iteration. I've read the OP, the other guides, and pretty much everything I can find on IVB overclocking, and last night's results are confusing me. To start, I can set the voltage to manual (not offset) of 1.0V, and leave it at the default turbo of 3.8GHz with no problems whatsoever, and keeping it cool at load. It's when I try to do anything else that my computer keeps freezing at login. So let's turn everything to Auto and crank the turbo setting to 4.0GHz. Now (according to HWiNFO64) is idling at almost 40C because it loves cranking the voltage to 1.20V idle, and 1.25V at load. Trying 4.4GHz scared the hell out of me because I ended up at 85C almost instantly because the Vcore jumped to something absurd like 1.45V. Offset does the same thing. Auto offset (and any other offset I define) just makes the Vcore sit at 1.20V-1.25V even when idling. I'll need to play with it more but holy hell I thought this was supposed to be simpler 
|
|
#
?
Jun 12, 2012 14:00
|
|
|
Well, you picked a motherboard with the chintziest of VRMs. It's a bit of a given that you will need higher voltages for the same clocks vs. a beefier board. Reset everything to default, update your BIOS, and make sure load-line calibration is set to the lowest setting (Regular?). Then see if you can adjust just the voltage offset and see changes.
|
|
#
?
Jun 12, 2012 14:15
|
|
|
Factory Factory posted:Well, you picked a motherboard with the chintziest of VRMs. It's a bit of a given that you will need higher voltages for the same clocks vs. a beefier board. I was going to say it makes me sad that a $140 mobo is considered chintzy, but then I looked at all of the Z77 boards and when did they all start costing $2-300?
|
|
#
?
Jun 12, 2012 14:24
|
|
|
When Asus realized they had better engineers than the other guys and decided to price like Sony.
|
|
#
?
Jun 12, 2012 14:27
|
|
|
So this is getting into more academic questions, but I'm curious now. I'm looking at the ASUS site and what makes the LX the chintzy one vs the LE, the straight up V, etc? Looking at them side by side, they only seem to vary in peripherals (extra connectors, different audio/lan chipset, etc). Did I just unluckily pick the 'worst' one of the group?
|
|
#
?
Jun 12, 2012 14:37
|
|
|
Fhqwhgads posted:So this is getting into more academic questions, but I'm curious now. I'm looking at the ASUS site and what makes the LX the chintzy one vs the LE, the straight up V, etc? Looking at them side by side, they only seem to vary in peripherals (extra connectors, different audio/lan chipset, etc). Did I just unluckily pick the 'worst' one of the group? They differ in VRM phases, probably
|
|
#
?
Jun 12, 2012 15:11
|
|
|
Power phases. The LX and LK both have 4 phases for the CPU, and other boards have more. LE has 6, vanilla has 8, and higher SKUs have more. The short version is that more phases equals faster and more efficient conversion from low-amp +12VDC from the power supply to high-amp Vcore for the processor, as each phase has to do less work. Around 8 phases you hit diminishing returns, even for the cruddiest of VRM technologies. The Digi+ stuff used on Asus boards is far from the cruddiest, but the LX still has a small number of phases, and so can't put out huge amounts of current without overheating.
|
|
#
?
Jun 12, 2012 15:12
|
|
|
I believe this problem is related to overclocking. I have a 2500k with a hyper 212. I had it set to 4.2GHz and using it seemed perfectly fine, the temps were fine at load and everything. Here comes the problem, about an average of two to three times a week, I would come to my computer, move my mouse to get my displays back on, and nothing would happen. The computer is still on, but the display won't show up. I don't have my computer set to sleep or hibernate, just the displays turn off after 5 minutes. My screens never shut off while I am playing games/doing things on the computer, only when I come back to the computer later in the day or something. I set my BIOS/UEFI back to defaults and it seems to not happen anymore. Any suggestions? Am I overclocking wrong? I am just setting the multiplier to 42 and saving and exiting. i5 2500k w/hyper 212 ASRock Z77 Pro4-M MSI 6950 Any help is appreciated, I'm probably doing something wrong.
|
|
#
?
Jun 12, 2012 16:39
|
|
|
That's really more of a Haus of Tech Support question, as that's not generally a problem you see from properly functioning components, overclocked or not. The only thing it might be is the Internal PLL Overvoltage bug - go into the UEFI setup and switch that to "off" and see if that helps. I'm not sure it will, though, as that's a bug related to standby and hibernate, which you're not doing.
|
|
#
?
Jun 12, 2012 16:56
|
|
|
So I threw together a system with a 3570k, Sabertooth Z77, TRUE 120, and 16gb of Crucial 1.5v RAM (e: 2 8gb sticks). Anyone familiar with that board/combo have any advice before I dig into it this weekend?
Lowclock fucked around with this message at 05:11 on Jun 14, 2012 |
|
#
?
Jun 14, 2012 05:08
|
|
|
I think I might remove the IHS on my 3570k too. I practiced some on my old q6600 and with rocking the razor back and forth to get under the corners it seems really easy. Are people gluing them back on after replacing the TIM, or just using the socket itself to hold the IHS tight on the core? Seems like glue would just be redundant. Hard to find much good info beyond Japanese/German youtubes and such.
|
|
#
?
Jun 15, 2012 23:53
|
|
|
I don't know if this has been mentioned already, but there's a mistake in the OP. It advocated for having positive case pressure, which is correct, but then says you achieve this by having more exhaust fans than intake fans. In any case, I came here for this question. Cooler Master Hyper 212 Plus is $20 AR at Newegg. My current CPU temps are 35C idle with 30C ambient and max out at 86C under Prime95, but realistically the most CPU-intensive stuff I do is play Diablo 3 and Starcraft 2 and Civ 5 (assuming H.264 video is being decoded by my 560 Ti). Is it worth the $20? Before you suggest I spend money on case fans, all 6 spots are occupied, albeit by lower dB/CFM 120mm fans.
|
|
#
?
Jun 16, 2012 06:51
|
|
|
Josh Lyman posted:I don't know if this has been mentioned already, but there's a mistake in the OP. It advocated for having positive case pressure, which is correct, but then says you achieve this by having more exhaust fans than intake fans.
|
|
#
?
Jun 16, 2012 07:44
|
|
|
grumperfish posted:What CPU and clocks? 80C+ is actually really high for Prime95 for an Intel chip, and way too high for an AMD CPU.
|
|
#
?
Jun 17, 2012 02:47
|
|
|
For the love of God, don't do that. The stock Intel cooler is designed to be just enough to get 99.99% of chips past the warranty without dying on stock speeds. The temp you're at is on par with suicide e-peen run; you'll be lucky if it lasts you the year.
KillHour fucked around with this message at 04:42 on Jun 17, 2012 |
|
#
?
Jun 17, 2012 04:37
|
|
|
Well switching the TIM seems to be a pretty good success. It's dropped about 3-5c off my idle temp, and like 12-15c off full load Prime95 tests. I have yet to see if this will get me a higher clock but whatever, it was free and easy. As a note, the paste that came on mine was a darker grey that doesn't look like any of the pictures I've seen. It was applied relatively well, but it still feels like cheap silicone stuff.
|
|
#
?
Jun 17, 2012 04:41
|
|
|
I'm just waiting for someone crazy enough to try soldering the heat spreader on to see if it's possible.
|
|
#
?
Jun 17, 2012 05:02
|
|
|
Seems reasonable enough so far.
|
|
#
?
Jun 17, 2012 05:58
|
|
|
KillHour posted:For the love of God, don't do that. The stock Intel cooler is designed to be just enough to get 99.99% of chips past the warranty without dying on stock speeds. The temp you're at is on par with suicide e-peen run; you'll be lucky if it lasts you the year. That said, are there any better options for $20 than the Hyper 212 Plus? Rebate expires today so I have to make a decision soon. 
|
|
#
?
Jun 17, 2012 08:30
|
|
|
Nope, buy that, cool them temps down and raise them clocks up. Them. Them. themm
|
|
#
?
Jun 17, 2012 08:44
|
|
|
Josh Lyman posted:In any case, I came here for this question. Cooler Master Hyper 212 Plus is $20 AR at Newegg. My current CPU temps are 35C idle with 30C ambient and max out at 86C under Prime95, but realistically the most CPU-intensive stuff I do is play Diablo 3 and Starcraft 2 and Civ 5 (assuming H.264 video is being decoded by my 560 Ti). Is it worth the $20? Before you suggest I spend money on case fans, all 6 spots are occupied, albeit by lower dB/CFM 120mm fans. If you ARE wanting to overclock, then obviously you should upgrade the cooler, preferably to something even more powerful than a Hyper 212, unless you just can't stomach the extra dollars. A Thermalright HR-02 Macho isn't really that much more expensive, and it's an entirely better class of performance and quiet.
|
|
#
?
Jun 17, 2012 16:26
|
|
|
Alereon posted:If you're not planning to overclock or otherwise run in a thermally stressful environment there is no reason to upgrade beyond the stock cooler. If you're using the stock TIM, replacing it with thermal paste (no need to buy Arctic Silver or other stuff) will improve temperatures noticeably and for free. You should also probably remove some of your case fans, six fans is pretty ridiculous and you're generating a lot of extra noise for not much benefit. I run 6 case fans, although that's with 7 HDD's, high-end fans, and a weird full-tower case (although many mid-towers are approaching its size at this point). For most usage cases, 6 basic case fans are far too many.
|
|
#
?
Jun 17, 2012 16:55
|
|
|
Under IntelBurnTest after 2-3 cycles my cores peak at 70/76/72/72. Is it okay for that one core to potentially get that high for brief moments? I only use this machine for gaming so I doubt I'll ever be capping CPU load for more than a few seconds anyway but just want to be safe. Edit: I have an H60 with one fan in a pull exhaust setup, so I'm actually thinking about swapping it out for two better fans in push/pull anyway, should help CPU temps and let me get a bit more of an OC from it. Noise is not really an issue, the machine feels silent compared to my previous Crossfire 6870 machine. V V Thanks  .
Tunga fucked around with this message at 23:05 on Jun 17, 2012 |
|
#
?
Jun 17, 2012 22:42
|
|
|
Yeah, that's fine. It's Prime95 (or Metro 2033) that's a better representative of a real-world sustained load. IBT is a synthetic load that I'm pretty sure is more intensive than even h.264 video encoding, which I believe is about as intensive as real-world tasks get.
|
|
#
?
Jun 17, 2012 22:53
|
|
|
Alereon posted:If you're not planning to overclock or otherwise run in a thermally stressful environment there is no reason to upgrade beyond the stock cooler. If you're using the stock TIM, replacing it with thermal paste (no need to buy Arctic Silver or other stuff) will improve temperatures noticeably and for free. You should also probably remove some of your case fans, six fans is pretty ridiculous and you're generating a lot of extra noise for not much benefit.
|
|
#
?
Jun 17, 2012 23:41
|
|
|
|
| # ? Apr 29, 2024 20:29 |
|
|
If I undervolt my CPU too far, what are the effects? Just instability? Can it damage it? I'm getting better results out of dropping VCore when increasing the multiplier because it keeps my temps down. Increasing VCore is a waste of time because it seems stable anyway and gets too hot. Also any idea of a rough guide for too far? It currently idles at 0.960v and peaks at 1.152v, that's a -0.050 offset running at 4.2GHz. If I set it to -0.1v the number turns red which scared me, heh. Tunga fucked around with this message at 23:55 on Jun 17, 2012 |
|
#
?
Jun 17, 2012 23:49
|
|