|
Hey so I have a really weird overclocking issue that I've never seen or experienced before. I have a 6600k setup and a 1070. Both overclock VERY well, and work without a care in the world... nothing to ever suggest that they are being pushed too hard whether it's benchmarking or playing an intensive game for hours. EXCEPT when the drat computer goes into hibernation. When my computer hibernates there is like a 30% chance that it will fail to boot back. Then, after restarting, the BIOS resets all of my overclocking (XMP/6600K). It's infuriating and impossible to understand.
|
#
?
Aug 11, 2016 03:43
|
|
|
|
| # ? May 16, 2024 04:48 |
|
|
Phlegmish posted:What does XMP do? Is it not enough to set your RAM speed manually? Technically the DDR3 spec only extends up to 1600 and DDR4 only goes up to 2400. Anything past that is technically overclocking and the motherboard won't do it automatically just to be safe. XMP is a way for the memory to automatically set its speed and timings past the "safe" limit. Sh4 posted:Didn't know about the leaking defect but yet again half the people I know with this kind of AIO cooling had failures so I thought it was common knowledge they were poo poo (hint: price) Most AIOs are fine. Honestly if there's anything cheap about them, it's the pumps. I've seen quite a few people complain their pump died and that's game over for the AIO unit. I wouldn't buy a used AIO of any kind, or run them beyond a 5 year lifespan. I also would not install them if they're older than 3 years just to give a reasonable lifespan before they should be pulled. Whether or not you need liquid cooling depends on what you're doing. A stock CPU in a normal case? No reason to do anything more than a Hyper 212 Evo. But not everyone's running stock clocks and not everyone has room in their case for a 212 Evo. A thin 120mm AIO is going to beat the poo poo out of most low-profile coolers and the higher the TDP gets the more effective liquid cooling becomes. At a certain point air coolers stop being viable, i.e. OC'd HEDT chips or OC'd/high-end AMD chips. You can't push 250W+ through a 212 Evo. Paul MaudDib fucked around with this message at 07:21 on Aug 11, 2016 |
|
#
?
Aug 11, 2016 05:40
|
|
|
Paul MaudDib posted:Technically the DDR3 spec only extends up to 1600 and DDR4 only goes up to 2400. Anything past that is technically overclocking and the motherboard won't do it automatically just to be safe. XMP is a way for the memory to automatically set its speed and timings past the "safe" limit. Kingston hyperX will run automatically at faster than 1600 without xmp. My 1866 kingston hyperx runs at.. 1866. Dont have to touch anything on the board.
|
|
#
?
Aug 11, 2016 14:52
|
|
|
Anyone else here tried overclocking their Gigabyte 1070 G1? I was able to get a core clock of +81 and a memory clock of +600 at stock voltage simply by raising the power limit. But I ran into a weird issue trying to go beyond that: If I move the core voltage at all my previously stable overclock becomes unstable in Heaven. I know the voltage is capped by the factory BIOS, so it only nets me 0.03V but I thought it was a bit odd to see that instability show up. I was hoping to push my core a little further since some reviews suggest it should be possible to get closer to 115 on the core with a voltage boost.
|
|
#
?
Aug 17, 2016 18:01
|
|
|
edit: whoops wrong thread.
Welmu fucked around with this message at 14:08 on Aug 18, 2016 |
|
#
?
Aug 18, 2016 13:56
|
|
|
I've got 2 GPUs in a full ATX tower mining ethereum, and am adding a 3rd. Right now the case has one single exhaust fan on the back at the top, and I've got one 140mm fan to add to try and keep things sane. Do I want to put it on the top of the case as an exhaust, or on the side panel at the vent directly over the GPUs as an intake or exhaust?
|
|
#
?
Aug 19, 2016 15:23
|
|
|
Twerk from Home posted:I've got 2 GPUs in a full ATX tower mining ethereum, and am adding a 3rd. Right now the case has one single exhaust fan on the back at the top, and I've got one 140mm fan to add to try and keep things sane. Do I want to put it on the top of the case as an exhaust, or on the side panel at the vent directly over the GPUs as an intake or exhaust? My personal philosophy here is to try and maintain positive pressure to keep dust out of the case, so I would put it on the side as an intake with a dust filter. The air will also force itself out a little better through seams/louvers/the top fan. Configuring it as a top or side exhaust may be more effective at getting hot air out however.
|
|
#
?
Aug 19, 2016 16:37
|
|
|
Temperature and overclocking question for you guys: So I have this case + mobo (not my setup in pic):  I'm running an i7 6700 with a low profile Noctua NH-L9x65mm cooler. The case can take up to an 85mm cooler. I've been having some issues with temps getting up into the high 80s and low 90s when stress testing. At idle, temps are at 22-25C in "balanced" mode with windows and 37-40C when in "high performance" windows mode where the processor just runs at 4.2Ghz all the time. I've reapplied thermal paste multiple times and the temps are always in the same range. The low profile cooler seems to be the issue since reviews say it is just not good enough to really cool a 6700k especially if it is overclocked. If I OC to 4.5Ghz with 4.8Ghz turboboost, I easily get 100C temp spikes and throttling. Temps in games are in the 60s-70s at stock speeds so that's good. Any thoughts on what I can do to improve CPU cooling at this point that WOULD NOT involve going the water cooling route? I suppose I could mount a massive air cooler on the thing and just keep the side panel of the case off but that would look pretty awful. Also, any thoughts on just buying a desk/room fan, popping off the side panel, and just blowing a metric fuckton of air into the case when I'm gaming and overclocking? I usually keep the side panel unscrewed anyways and take it off when gaming since I can get temps to go down by about 5C with the glass side panel off. It's my fault for buying this case, but it is sturdy as gently caress, and I just really like it aesthetically.
|
|
#
?
Aug 28, 2016 02:45
|
|
|
What are you using to generate those temps during stress testing? Something like prime95 will run your temps much higher than anything you are going to do naturally. It all depends on if you are building the machine for stress tests or actual use. Your temps seem fine to me if they don't get that high outside of stress testing.
|
|
#
?
Aug 28, 2016 03:14
|
|
|
The 3dmark Timespy benchmark will cause temps to spike to the 100C range when the thing is OC'd. Temps spike occasionally in BF4 to 92-94C during OC but on average temps are in the mid to high 70s. The temp spikes literally last for 1-2 seconds and then go back to the high 70s when reading via HWiNFO64. Should I really be concerned about intermittent temp spikes? Or be more focused on average cpu temp?
|
|
#
?
Aug 28, 2016 03:24
|
|
|
Those low profile CPU coolers arent really that great for overclocking, if it were me I'd either leave it at stock speeds or invest in a better cooler.
|
|
#
?
Aug 29, 2016 15:18
|
|
|
On my noctua dh14, temps at 4.4ghz / 1.45v (my boards auto) will hit high 80s. The same clocks at 1.28v (the actual value it needs, set manually, and manually setting vccio and use agent voltages sanely) result in 65deg + better, crash proof stability. I stress test using Asus real bench (I don't think it's that tough a test, cinebench and games). Auto can contribute to temps being /way/ higher then they need to be. Even with great cooling you'd not want to run auto voltages as it's just reducing the chip lifespan and dumping extra heat into the system. GRINDCORE MEGGIDO fucked around with this message at 00:50 on Aug 30, 2016 |
|
#
?
Aug 30, 2016 00:41
|
|
|
How many volts are you pumping through that thing to hit 4.8ghz? (also, turn off turbo boost) GPU OC question - The Sapphire Trixx 6.0 software doesn't let me use voltage control on my Sapphire Fury. MSI Afterburner works just fine though - is there a trick to Trixx to unlock that setting?
|
|
#
?
Aug 30, 2016 22:55
|
|
|
Is it even possible to have a 6600k that cannot pass Very High IBT at 4.2GHz, 1.35v (auto voltage), with a liquid cooler (stayed well under 60°C during the test)? And if i try 4.4GHz with the same settings it consistently blue-screens at boot? I know it's called the "silicon lottery", but I've never heard of a chip this bad. Is there anything else I could be screwing up, I hope?
|
|
#
?
Aug 31, 2016 21:32
|
|
|
NihilCredo posted:Is it even possible to have a 6600k that cannot pass Very High IBT at 4.2GHz, 1.35v (auto voltage), with a liquid cooler (stayed well under 60°C during the test)? And if i try 4.4GHz with the same settings it consistently blue-screens at boot? Of course it is "silicon lottery"
|
|
#
?
Aug 31, 2016 22:04
|
|
|
NihilCredo posted:Is it even possible to have a 6600k that cannot pass Very High IBT at 4.2GHz, 1.35v (auto voltage), with a liquid cooler (stayed well under 60°C during the test)? And if i try 4.4GHz with the same settings it consistently blue-screens at boot? It seems pretty drat unlikely. I'd try setting voltage manually and maybe messing with RAM profiles before accepting that kind of bad luck
|
|
#
?
Aug 31, 2016 22:41
|
|
|
NihilCredo posted:Is it even possible to have a 6600k that cannot pass Very High IBT at 4.2GHz, 1.35v (auto voltage), with a liquid cooler (stayed well under 60°C during the test)? And if i try 4.4GHz with the same settings it consistently blue-screens at boot? You could be a really unscrupulous consumer and return it?
|
|
#
?
Aug 31, 2016 22:44
|
|
|
Ninkobei posted:It seems pretty drat unlikely. I'd try setting voltage manually and maybe messing with RAM profiles before accepting that kind of bad luck RAM profiles? I have only 2400-clocked RAM, so I just left them as they were. Could it make a difference to bring them down to 2133? And if it does, would that mean it's the RAM sticks that are crap, and I should replace them? Twerk from Home posted:You could be a really unscrupulous consumer and return it? Thinking about it, but it's not really worth it. Replacing the CPU itself would be more work than usual (it's an ITX build so I'd need to remove a lot of other stuff to get to it), and I bought it from a small retailer that has the best prices but is very slow with deliveries and processing - I may well be desktop-less for multiple weeks. At that point, if it's an option, I think I'd rather just crank up the voltage to "dangerously high" levels and have to replace the chip a few months sooner than planned.
|
|
#
?
Aug 31, 2016 23:26
|
|
|
Go to whatever voltage level you need to and buy OC insurance from Intel itself for when it dies, if you really want to feel safe.
|
|
#
?
Aug 31, 2016 23:45
|
|
|
NihilCredo posted:RAM profiles? I have only 2400-clocked RAM, so I just left them as they were. Could it make a difference to bring them down to 2133? And if it does, would that mean it's the RAM sticks that are crap, and I should replace them? It could be that the RAM isn't getting enough voltage, but that would cause stock clocks to be unstable too. It's probably fine, I probably shouldn't have said anything.
|
|
#
?
Sep 1, 2016 00:05
|
|
|
Hey folks. I just finished putting my system build together and had a few quick questions and didn't read the op. I have 4 ddr-3000 dimms, which are recognized by XMP, but won't actually boot at 3000. The bios guard thing pops up after a while and resets them back to 2133. If I set it manually to 2900 everything seems to work fine. Should I bother doing anything about this or just shrug and move on? CPU side I'm idling at ~43c (6700k) on air that seems about ballpark right? I haven't tried bumping it at all yet. And finally for the GPU it's literally just drag the power target in afterburner to the right these days?
|
|
#
?
Sep 1, 2016 18:27
|
|
|
Fuzzy Mammal posted:Hey folks. I just finished putting my system build together and had a few quick questions and didn't read the op. Try raising the memory voltage a little.
|
|
#
?
Sep 1, 2016 18:43
|
|
|
NihilCredo posted:Is it even possible to have a 6600k that cannot pass Very High IBT at 4.2GHz, 1.35v (auto voltage), with a liquid cooler (stayed well under 60°C during the test)? And if i try 4.4GHz with the same settings it consistently blue-screens at boot? Yeah mine is worse, liquid cooled, 58c max load, would not go above 4.1 no matter what. Sold the AIO liquid cooler and I'm now running it at stock speed. Silicon lottery.
|
|
#
?
Sep 1, 2016 18:43
|
|
|
So I have been tinkering with my new skylake build. It's a MSI Z170A SLI and i7-6700k with the cryorig H7 cooler. Been running at 4.6ghz so far with my vcore set to adaptive with offset. Set my voltage to 1.3 with a offset of +0.04. Everything seems stable so far with my voltage bouncing between 1.336 and 1.4 with temps hovering between 60 and 70 C. I have only been using the intel XTU to stress test at the moment and gaming has been stable. Does this sound reasonable or is there anything else I should be trying?
|
|
#
?
Sep 10, 2016 19:14
|
|
|
I've been stress testing my i5-6600k, and I've noticed that at idle all the cores are the same temperature, but when at full load there's one core that gets five degrees hotter than the rest which are all around the same temp. Is this common or did I mess up with the thermal paste or something?
|
|
#
?
Sep 10, 2016 19:20
|
|
|
Every multi-core processor I've ever had would vary a bit from core to core even on a full-bore 100%-on-all-cores load. Five degrees seems to be well within normal range.
|
|
#
?
Sep 10, 2016 19:40
|
|
|
CharlieFoxtrot posted:I've been stress testing my i5-6600k, and I've noticed that at idle all the cores are the same temperature, but when at full load there's one core that gets five degrees hotter than the rest which are all around the same temp. Is this common or did I mess up with the thermal paste or something? Mine's that way too, Core #1 is consistently about 5 degrees hotter under load.
|
|
#
?
Sep 10, 2016 19:59
|
|
|
Up to 10 degrees is normal.
|
|
#
?
Sep 10, 2016 20:55
|
|
|
I've got my 6600K at a 1.35 vcore and 4.4 GHZ OC, how much can i push the vcore if i want to try going for 4.5 or 4.6 GHZ?
|
|
#
?
Sep 10, 2016 21:32
|
|
|
I'm a super noob and stuck on the auto overclock mode on my motherboard, and all seems to be going well other than my ram speed according to speccy is no longer the 1600mhz it should be, is this cause for concern? (This is running with witcher 3 going)
|
|
#
?
Sep 12, 2016 02:17
|
|
|
Speccy reports the clock speed rather than the transfer speed so it should be running at 1333. The performance impact is probably very small but you can try going into your motherboard's overclocking settings in BIOS and setting it back to 1600.
|
|
#
?
Sep 12, 2016 07:12
|
|
|
I recently got an msi 1070 (armor), and tried to overclock it. I've never done it before so I just followed a guide, and everything seemed to work fine. I ran the Nvidia benchmark (Valley) for like an hour and there were no issues, and I've played Witcher 3 for like 4 hours straight with no hitches, but I'm still getting occasional driver crashes in a few games like Overwatch. From what I can tell I was pretty unlucky with my card, the core clock is at +190, but the memory clock only goes up to about +280 before everything starts crashing. I thought I had gotten it stable there, but apparently not! Is there a more reliable way of testing this stuff? Edit: I tried out Furmark, and although it ran fine, I did get slight artifacting until I turned down the memory to about 230. I guess that's what it takes, but it seems very low when I can read that others are getting it up to like 700! Hakkesshu fucked around with this message at 13:40 on Sep 13, 2016 |
|
#
?
Sep 13, 2016 13:13
|
|
|
Hakkesshu posted:I recently got an msi 1070 (armor), and tried to overclock it. I've never done it before so I just followed a guide, and everything seemed to work fine. I ran the Nvidia benchmark (Valley) for like an hour and there were no issues, and I've played Witcher 3 for like 4 hours straight with no hitches, but I'm still getting occasional driver crashes in a few games like Overwatch. Silicon lottery etc etc
|
|
#
?
Sep 14, 2016 12:17
|
|
|
Taima posted:Hey so I have a really weird overclocking issue that I've never seen or experienced before. What version of Windows? Quick online searching shows 7 and 10 had some weird hibernation issues that are similar to yours but they each have different fixes/workarounds. The BIOS resetting seems weird, have you checked with your board vendor for any BIOS updates?
|
|
#
?
Sep 14, 2016 15:57
|
|
|
So, I upgraded to a 6600k over the weekend and was wondering what a good maximum safe voltage level would be for an overclock. I've read in a couple of places that it's 1.45 and I'd rather not push it that far but would around 1.35 be okay? I got it to 4.5ghz with a 1.285 vcore and tested the stability with prime95 for three hours with no errors but I wasn't able to hold that once I set my ram to it's 3200mhz xmp and had to drop it to 4.4ghz to get it stable at that voltage. Also, I realize that my motherboard may just not be able to do both beyond that point but I would at least like to try. e: also temps are maxing out at 65c for the 6600k and my exhaust fan died right before I overclocked but it was 5 years old and I'm planning on replacing it in the next week or two. e2: It's been stable for an hour of prime95 at 4.5ghz with 1.308 vcore and a 70c max temperature. I'll probably just leave it here until I can buy a new exhaust fan but I would like to try for 4.6ghz after I get it because this seems to be a pretty stable chip. Regrettable fucked around with this message at 20:23 on Sep 16, 2016 |
|
#
?
Sep 16, 2016 18:22
|
|
|
I think i either lost the silicon lottery or im not setting my vcore correctly, because 4.4 is proving unstable for my 6600K
|
|
#
?
Sep 16, 2016 21:35
|
|
|
Regrettable posted:So, I upgraded to a 6600k over the weekend and was wondering what a good maximum safe voltage level would be for an overclock. I've read in a couple of places that it's 1.45 and I'd rather not push it that far but would around 1.35 be okay? I got it to 4.5ghz with a 1.285 vcore and tested the stability with prime95 for three hours with no errors but I wasn't able to hold that once I set my ram to it's 3200mhz xmp and had to drop it to 4.4ghz to get it stable at that voltage. Also, I realize that my motherboard may just not be able to do both beyond that point but I would at least like to try. Yeah 1.35 should be fine, I just got a 6600k as well and I settled at 4.6GHz at 1.37V with 3000MHz XMP. I want to try to push for 4.7 but my cooler hasn't been working as well as expected, I might have to reseat it.
|
|
#
?
Sep 17, 2016 22:04
|
|
|
MaxxBot posted:Yeah 1.35 should be fine, I just got a 6600k as well and I settled at 4.6GHz at 1.37V with 3000MHz XMP. I want to try to push for 4.7 but my cooler hasn't been working as well as expected, I might have to reseat it. Yeah, after further research I bumped it up and got it stable 4.6Ghz with 1.37v. e: I solved the other issue I was asking about. Regrettable fucked around with this message at 05:11 on Sep 18, 2016 |
|
#
?
Sep 18, 2016 03:52
|
|
|
I want to give a shoutout to my champion I5-750 2.6 ghz ancient CPU that has been holding 3.8gz since early 2010. Upgraded to a 6600k and some 3200 DDR4 today, hoping to win the lottery again. Is 4.2 an unreasonable goal with air cooling? (hyper 212 evo)
|
|
#
?
Sep 18, 2016 04:27
|
|
|
|
| # ? May 16, 2024 04:48 |
|
|
fargom posted:I want to give a shoutout to my champion I5-750 2.6 ghz ancient CPU that has been holding 3.8gz since early 2010. Upgraded to a 6600k and some 3200 DDR4 today, hoping to win the lottery again. A lot of it depends on what voltage you have to run.
|
|
#
?
Sep 18, 2016 04:30
|
|