|
I've edited my first post to add the CPU features portion. I hadn't realized this was where my TDP limit settings were, so I hadn't adjusted them yet. Ok, that's the piece I was missing. Voltages ARE getting bumped up, but the motherboard is doing it automatically. I was operating under the idea that there was a stock ceiling that I hadn't touched yet. This machine is not going to see much heavy use. Occasional Photoshop and Lightroom use is about the worst it's going to see. No gaming. I paid $100 after rebate for the mobo, CPU and case, so getting a 33% overclock on this without having to gently caress around with it at all is great. I may try to tweak it more in the future when I have some time.
|
# ? Dec 30, 2014 18:32 |
|
|
# ? May 25, 2024 03:57 |
|
Honestly 4.5 is the general ceiling for that combo so 4.3 is quite close. If it does blue screen , just know auto volting may be the culprit. You'll probably see that happen sooner than later though if its going to Encode a video in Handbrake at "Very Slow" in the video settings tab. I've never seen anything thrash my overclock so effectively
|
# ? Dec 30, 2014 18:37 |
|
So I just upgraded to a 4790k with an h80i cooler, using a VII hero board. Just stock, its at 1.220v @ 4.4ghz but my idle temps are 42-45ish I've seen that these chips run hot but figuring my cooling I still think thats a little high. I've got it set up in a P183 case, i got the radiator on the rear exhaust push/pull, top fan on exhaust, 2 front fans on intake. I thought it might be the stock thermal grease that came with it, used some arctic silver 5 and reseated the heatsink, but nothing really changed.
|
# ? Jan 7, 2015 22:09 |
|
AS5 is not meaningfully different than most standard thermal paste. What's load temp like? Idle is much less important, but you could always try bumping the fan speed if it bothers you.
|
# ? Jan 7, 2015 22:24 |
|
Paste at full load can be a matter of like 3 degrees unless there is something seriously wrong with the factory paste. I am in a firm minority where I think those 3 degrees can make a difference (sometimes it wont matter at all though) but thats only when youre pushing it to the max. That being said, I also think 45 degrees is a tad high for idle, but like factory said check your full load temps which is the important information. For example those few degrees just put me under the "comfort" max zone for me (which was like 90 degrees) with IBT and its variants. But it's not a huge deal. I like having good paste on hand more for GPU's having been through so many rather than cpu's. I'd also check that its running at its idle voltages at idle versus 1.20 all the time. I'm not sure its possible without having messed with bios settings but that also might be the case. There's no reason to turn off that power feature as far as I know. Should be a c-state but I'm a little rusty now If at 1.22 vcore it goes way way too high in temperature then there might be something a little off somewhere. I've never had much difference in CPU temp in which way I mounted the radiator (in/out, top or side of case), however that definitely does make a difference for GPU temperatures. But yeah anyways, just try full load. Dont worry about damage for this test. Best case if there is something a little wrong you might not have tightened the cooler enough but I'm weary recommending to really torque stuff down as its so relative from person to person. If you're full load temperatures are just fine I imagine its just the cooler winding down a lot as there is nothing really wrong with those idle temperatures. Maybe over like 30 years or something BTW you'll want to be in the low 80'sat most for that test imo at that voltage for IBT or small fft. I wish I remember more pre-delid numbers for me, but at 1.22 vcore I wouldn't even hit 70's but that's a little irrelevant 1gnoirents fucked around with this message at 22:50 on Jan 7, 2015 |
# ? Jan 7, 2015 22:44 |
|
Reseated the pump again, even switched up the fan config (intake at the top exhaust at the bottom, surprisingly nothing changed temp-wise idle/load that much) Load temps at STOCK go up to loving 99 degrees on Realbench 100% cpu load after just 30 seconds of it running. CPU-Z stays at 1.220V. Obviously something is wrong. Should I just return the chip? I bough it at Microcenter. Or could it be the cooler that isn't working?
|
# ? Jan 7, 2015 23:48 |
|
The Quake posted:Reseated the pump again, even switched up the fan config (intake at the top exhaust at the bottom, surprisingly nothing changed temp-wise idle/load that much) Pretty sure that's your cooling system, I'd try with aircooling and see how that works edit: after a few minutes of google I saw that the general opinion was that the h80 is a piece of poo poo, never used pre assembled watercooling kits because they are pointless but I'd expect them to be better than an average aircooler. Try with a good heatsink. Sh4 fucked around with this message at 23:55 on Jan 7, 2015 |
# ? Jan 7, 2015 23:51 |
|
I also bought the h80i at microcenter. I guess i'll try and go return both and see if they'll take them back. Might as well since I just bought them both on Saturday. Hopefully they aren't too uptight about it.
|
# ? Jan 8, 2015 00:02 |
|
The Quake posted:I also bought the h80i at microcenter. I guess i'll try and go return both and see if they'll take them back. Might as well since I just bought them both on Saturday. Hopefully they aren't too uptight about it. Try the stock fan before you return the CPU? Stock fan should work fine, if the H80 is terrible replace it.
|
# ? Jan 8, 2015 00:04 |
|
Sh4 posted:Pretty sure that's your cooling system, I'd try with aircooling and see how that works H80i might not be the most quiet thing ever but that one is simply broken. The Quake posted:Reseated the pump again, even switched up the fan config (intake at the top exhaust at the bottom, surprisingly nothing changed temp-wise idle/load that much) Like others, replace the cooler. Just put the stock one on, it can handle factory settings for the time being. Also it will test the extremely unlikely event your cpu is just bad as well... oh and make sure you plugged your pump in
|
# ? Jan 8, 2015 01:21 |
|
1gnoirents posted:Paste at full load can be a matter of like 3 degrees unless there is something seriously wrong with the factory paste. I am in a firm minority where I think those 3 degrees can make a difference (sometimes it wont matter at all though) but thats only when youre pushing it to the max. That being said, I also think 45 degrees is a tad high for idle, but like factory said check your full load temps which is the important information. I don't disagree that paste can make a 3C difference. It's just that AS5 doesn't live up to its hype and is nothing special any more.
|
# ? Jan 8, 2015 05:19 |
|
Factory Factory posted:I don't disagree that paste can make a 3C difference. It's just that AS5 doesn't live up to its hype and is nothing special any more. Last time I needed some, I had a quick look around and came up with the conclusion Noctua's NT-H1 is the stuff to buy. Also, if it's good enough for them to ship with their coolers, it's good enough for me. HalloKitty fucked around with this message at 14:12 on Jan 9, 2015 |
# ? Jan 8, 2015 08:10 |
|
Seconding NT-H1, it's far better than AS5 as I've tried them both on the same chip and got vastly better results with the latter. The best part is a syringe of it will last for at least a dozen applications.
|
# ? Jan 9, 2015 13:52 |
|
I have a giant tube of MX-4 on hand for GPUs, but I haven't really used it much since I'm still going through the leftover Chill factor III stuff from a review that's been working fine. I usually don't give alot of thought to TIM in general but I prefer to avoid anything like AS5 that requires burn-in or that possibly risks causing a short. Same with weird poo poo like gallium Coollaboratory Liquid PRO or anything that requires more thought than 'squirt and tighten'. Not planning to micromanage loving thermal paste.
|
# ? Jan 9, 2015 19:36 |
|
Looking for some tips! g3258 + msi pc mate; I set the multiplier to 45 and voltage to 1.250. Didn't touch anything else.
|
# ? Jan 10, 2015 21:14 |
|
rizzo1001 posted:Looking for some tips! If that is stable, that's really good. Call it a day! Edit: wait, is that a water cooler? If so you could go to 1.3 or 1.35 and maybe hit 4.6 or 4.7 with those results but you're in the realm of diminishing returns. Zero VGS fucked around with this message at 21:36 on Jan 10, 2015 |
# ? Jan 10, 2015 21:18 |
|
Zero VGS posted:If that is stable, that's really good. Call it a day! Arighty! To be fair it's a bit cold in my house today (67F outside case temp) so I'd expect those temps are a bit low. edit. air cooler in non-overclocking case... rizzo1001 fucked around with this message at 21:40 on Jan 10, 2015 |
# ? Jan 10, 2015 21:37 |
|
I realized I've had an i5-2500K for two years now and never really played with overclocking it. This thread and the OPs are a great resource. I simply bumped up the "idiot overclocking slider" in my DZ77GA-70K BIOS to 4.0ghz. I have not yet looked at manually tweaking settings, but that's the next step, now that I've done some research. All is stable so far, but I'm seeing temperatures of 83 C during sustained 100% load, (video transcoding). TDP is 79W. Too hot for comfort? I've got the stock Intel cooler.
|
# ? Jan 17, 2015 19:05 |
|
ryangs posted:I realized I've had an i5-2500K for two years now and never really played with overclocking it. This thread and the OPs are a great resource. You wanna get a better cooler for sure. (Haven't paid attention to them for a while so I won't recommend one.) ~80C is OK for a brief stress test but if you'll be hitting it with regularity it's not recommended. Look in CPUz, what's your voltage (vcore) going up to? Some boards aggressively put too much with the default settings and jack up temps. 4.3 or 4.4 is a good goal to shoot for before it starts getting harder to keep the voltage in recommender ranges when ya have a better cooler on it. Still crushes games no prob.
|
# ? Jan 17, 2015 19:20 |
|
That's pretty warm although not unexpected with the stock cooler. The easy-mode overclocking probably bumped up the voltage by alot as well. I'd look into replacing the heatsink with something better, or at least manually set your CPU vcore. The 'safe' upper voltage limit on your chip's going to be about 1.38V but I wouldn't go that high on the stock cooler. Something like a 212+ EVO or that newer Cryorig ~$35 cooler would do the trick.
future ghost fucked around with this message at 19:43 on Jan 17, 2015 |
# ? Jan 17, 2015 19:27 |
|
Got it. The BIOS fan controls were set pretty conservatively, with a target temp of 80 C. I turned that down to 72 C target temp and am having better luck keeping temps in line. The fans were all staying at their lowest speed before; now they're ramping up appropriately. The board's default settings for 4.0ghz are a 40 mV offset, which resulted in about 1.35V during sustained load at 4.0ghz. Not bad? The other power settings seem sort of high: 190W Turbo Boost short max, 180W sustained and 140A current override, but I guess I have no frame of reference. I'll probably leave those there, at the suggested 4.0ghz settings, and manually tweak only the voltage and multipliers. Given my stock cooling, I'm setting 4.4ghz as a goal. Reasonable? Or too extreme with stock? I have a Fractal Design R4 case, and its two 140mm fans are also connected to the motherboard for control, so I feel like my overall cooling situation is solid. Edit: drat, those Coolermaster 212 EVOs are pretty affordable. Might have to get one of those. ryangs fucked around with this message at 21:08 on Jan 17, 2015 |
# ? Jan 17, 2015 20:48 |
|
Get a Cryorig H7 instead - easier to install, smaller, cools better, quieter, and only like $5 more.
|
# ? Jan 18, 2015 00:50 |
|
Forgive me if this would be better suited to the GPU thread, but I've been taking babby steps into OC'ing with my GTX 760 and I'm running into some stuff that I'm not sure is normal. I put my system together a couple weeks ago and got all my drivers and such updated, Windows Update, etc. And I installed the newest Nvidia drivers for my card (347.09). However I've been getting crashes during games like Metal Gear Solid: Ground Zeroes, Watch_Dogs, and Alien: Isolation anywhere from five to thirty minutes in. It seems to happen when the card's temp gets up around 70c. The crashes vary a bit, and I've gotten at least one BSOD, but usually the video will freeze and cut to black screen, and the sound will either stutter/freeze or just keep playing until I open Task Manager and end the game's process. So I figured maybe the card just needed a little extra push. I started messing with the GPU using EVGA Precision X16 the other night, and gradually worked my way up to the highest setting Anandtech was able to safely reach. http://www.anandtech.com/show/7103/nvidia-geforce-gtx-760-review/17 The only benchmark I was using was Unigine Heaven, and even without any overclocking I would get the same crash to black within the first ten seconds of the benchmark, though unlike every game I've tried it with, Heaven would pop back on and continue benchmarking instead of crashing outright. Now I'm wondering if it's just a drivers issue. Would rolling back to the last driver (344.xx) be the logical next step? Specs: CPU: Intel Core i7-4770K (not OC'd) Cooler: Corsair H100i Mobo: Gigabyte Z87X-UD7-TH RAM: 8GB Corsair Vengeance DDR3 PSU: Rosewill Capstone 750w GPU: EVGA GTX 760 DammitJanet fucked around with this message at 22:27 on Jan 20, 2015 |
# ? Jan 20, 2015 22:23 |
|
Even if a little extra voltage fixed the card, crashing at stock settings means the card is faulty. Try rolling back the drivers, and if it's still as misbehaved, then get an RMA. -- I used my overclocking knowledge for something strange today: I undervolted my Surface Pro. Took 50 mV off the top, resulting in a 10 C temperature drop under LINPACK and a small but noticeable improvement in battery life. Crazy. Supposedly the real payoff will be when I game on the thing, as the lower heat will allow higher turbo clocks.
|
# ? Jan 21, 2015 02:49 |
|
Factory Factory posted:Get a Cryorig H7 instead - easier to install, smaller, cools better, quieter, and only like $5 more. Ooo, this looks great. I hate noise. Any experience with the be quiet Pure Rock? These guys love it just slightly more than the Cryorig H7, and it's the same price. I guess it's Taiwan vs. Germany. ryangs fucked around with this message at 06:20 on Jan 21, 2015 |
# ? Jan 21, 2015 06:08 |
|
ryangs posted:Ooo, this looks great. I hate noise. Any experience with the be quiet Pure Rock? These guys love it just slightly more than the Cryorig H7, and it's the same price. I guess it's Taiwan vs. Germany. I have never seen any cryorig cooler in my life but I've used a dark rock and it was the best heatsink I've had, it's so big it has fitment problems on some motherboards though so I'm using an Evo 212 with be quiet shadow wings fans on it now and I can tell you it's absolutely silent, you can't tell the computer is running
|
# ? Jan 21, 2015 16:28 |
|
Factory Factory posted:get an RMA. On it. Many thanks.
|
# ? Jan 21, 2015 19:52 |
|
ryangs posted:Got it. The BIOS fan controls were set pretty conservatively, with a target temp of 80 C. I turned that down to 72 C target temp and am having better luck keeping temps in line. The fans were all staying at their lowest speed before; now they're ramping up appropriately. 1.35 is fairly high for 4.0 ghz. You can probably get a lot more speed out of it without touching the voltage at all. There are many who will say don't go over 1.35 at all, although I believe for these cpus its technically 1.375
|
# ? Jan 23, 2015 22:51 |
I figure there's no reason to upgrade my CPU from my 3570K since I'm just gaming/emulating with it, so I figured I'd try and overclock it to 4.5 today. I used 1.5 V, and the CPU ran at about 65 C with a Evo 212 in there, but then I got a blue screen after about 5 minutes using Prime95. Is that a voltage issue, and if so do I just wanna back up on the voltage until I get stable?
|
|
# ? Jan 24, 2015 01:16 |
|
Do research. 1.5V is a crazy high voltage that a 212 EVO isn't going to be able to handle. start from 1.2V and work your way up with .01 increments.
|
# ? Jan 24, 2015 01:46 |
Hace posted:Do research. 1.5V is a crazy high voltage that a 212 EVO isn't going to be able to handle. start from 1.2V and work your way up with .01 increments. Well it would help if I had put the 3 before the 5 in my original post, but I still figured that was probably too high, I'll lower it down to 1.2.
|
|
# ? Jan 24, 2015 01:49 |
|
1gnoirents posted:1.35 is fairly high for 4.0 ghz. You can probably get a lot more speed out of it without touching the voltage at all. There are many who will say don't go over 1.35 at all, although I believe for these cpus its technically 1.375 Indeed. I left the power settings at the BIOS's overclocking "default" for 4.0ghz and increased the multiplier to 44, and 4.4ghz is rock solid. It's a dynamic voltage with a 40mV "turbo boost," so it's not fixed at 1.35. It only peaks at that under 100% load (stress test or video encoding). It idles at around 1.15V. I might try bumping the multiplier above 44 once my new cooler arrives, but for now I'm totally happy and not feeling any need to go much higher.
|
# ? Jan 24, 2015 20:24 |
|
... Why not? posted:Idle temps. The lesson to take away here is, clamp the heat spreader by the square, not by the flanges on the side. Overclocking results for this poor mangled thing coming soon, if it survives. Sidesaddle Cavalry fucked around with this message at 10:47 on Jan 27, 2015 |
# ? Jan 27, 2015 10:43 |
|
Oh dear God. Well, because it works, that was definitely a filtering cap. If your PSU and motherboard have great ripple control, the CPU will be fine. But ripple increases wear and tear, and your CPU now has less protection against it. How do you feel about getting a replacement cap and an SMD soldering iron?
|
# ? Jan 27, 2015 14:32 |
|
1.52v suicide run for a month or two?
|
# ? Jan 27, 2015 19:53 |
|
Factory Factory posted:Oh dear God. Well, because it works, that was definitely a filtering cap. If your PSU and motherboard have great ripple control, the CPU will be fine. But ripple increases wear and tear, and your CPU now has less protection against it. The difficult part about putting a new cap back on is that even if I bit the bullet to attempt it, I wouldn't be confident enough to know if I did a good enough job for the new cap to actually be working. I certainly can't tell exactly how well the CPU is taking ripple right now, much less how it would after such an operation. 1gnoirents posted:
I still haven't touched the multiplier or anything else yet, but I do look forward to getting more adventurous for the remainder of this life!
|
# ? Jan 27, 2015 22:07 |
|
Sidesaddle Cavalry posted:While this chip may have caught electricity cancer, suicide is still not the moral thing to do. Euthanasia sounds much better.
|
# ? Jan 27, 2015 22:09 |
|
So far so good with my old overclock at 4.4 GHz and 1.25 V at load. Before the delid, this was stepped down from 4.7 at a substantially higher voltage due to temperature and resulting case noise preferences. I maxed out temperature at 63° C this time around, so there's room to come back.
|
# ? Jan 28, 2015 05:07 |
|
SMD soldering is really hard, even with a lot of experience. Taking a hammer to the chip is practically normal and easy in comparison.
|
# ? Jan 28, 2015 08:05 |
|
|
# ? May 25, 2024 03:57 |
|
I just finished this last night as part of my "computer-to-take-to-work-on-the-weekends-and-play-video-games" PC, it's a Pentium G3258 on an ASUS Z97 itx board crammed into a Corsair 250D (not the smallest box in the world, but small enough and it's lightweight). I bought an H100i to go with it, but I had to take the front 140mm fan out just so the radiator would fit. I plan on upgrading to an i5 or i7 in the future, which is why the board is a little bit overboard for the processor. Anyways, both fans attached to the radiator are the only ones actually blowing, and both of them are intakes. I'm gathering this is blowing hot are directly onto the video card? It's a little Geforce 750ti, but I'm wondering if I am creating a situation in the future where I am not allowing enough air to come in during the summer and/or when I upgrade. Should 1 fan on the radiator be an exhaust and the other an intake? Right now on the Pentium it's running about 28c under load, and I let it run over night and woke up to 30c.
|
# ? Jan 30, 2015 23:39 |