X79 and/or rampage IV OC'ing thread for those of us still left....

[Lorec]

So the build is like in my signature.
Had a bit problem with temps tbh, today when I went to work I didnt leave aircon on and my pc reset itself at some point...
I am using Hyper 212 Evo though...
I exchanged one cooler master fan for two push -pull noctua a12x25 and that solved the problem.
Idles at 35C opposed to earlier 45-50.
Crunches at 100%cpu/100%time at around 70C.
While earlier it was 100%cpu/75%time (break every three seconds) at around 77C.
It does make a hell of a lot difference.
I picked up this 3960X for like $50 Im very happy so far.
EDIT:
Core 5 always gets hotter and I wonder if that might be this particular chips gimmick or the fact that hyper 212 doesnt fully cover the IHS...

 

[Kanan]

Except it is a lot slower with chrome. Selenium is an automation tool that controls the browser for you so you can do things like automated tests against web applications.
View attachment 132698

This is a breakdown of every test including the browser tests. The red line is the 3960X. There is almost nowhere other than Zstd compression where the 3960x even gets close and these are all 6c/12t chips.
View attachment 132699

Mind you, I'm speaking as a person with a 3930k.

Doesn't matter, because in games it is a vastly different picture and the 3960X is approximately as fast as a 6700K. If you overclock it to 5GHz it is comparable to a 7700K performance wise. The Ryzen for example is much worse for gaming, only the 3rd gen Ryzens are superior and only if you do not overclock the 3960X. The 5GHz 3960X will be as fast or even faster then the 3600X because of the problems Ryzen has due to the CCX nature of its architecture.

I see a version difference in the Chrome version (but I doubt it would matter too much) but still... I'm surprised in a way but not in others.. The 3xxx series is probably missing out on a lot of instructions sets that the newer CPUs have so it would explain it to be honest

Again, do you expect something 5 to 6 years old to be the same performance as something 'cutting edge' (shall we say)...

I always heard nothing but good things about the 3930k or 3960X, in ways I should have upgraded to them when they came out rather than sticking with the X58 really lol Am I right in thinking some/most of those can do 5Ghz??

Well at least mine could do 5 GHz easily, and even more, but I don't have a water cooler, so I didn't try more than 5.1 GHz. The 3960X/3970X is the best binned 32nm chip ever, and 32nm is the best node for overclocking in general.

 

[Aquinus]

Doesn't matter, because in games it is a vastly different picture and the 3960X is approximately as fast as a 6700K. If you overclock it to 5GHz it is comparable to a 7700K performance wise. The Ryzen for example is much worse for gaming, only the 3rd gen Ryzens are superior and only if you do not overclock the 3960X. The 5GHz 3960X will be as fast or even faster then the 3600X because of the problems Ryzen has due to the CCX nature of its architecture.

Games a lot of the time aren't CPU bound, particularly at higher resolutions, and that means nothing unless all you care about is games. The geometric mean says it's a little under 3 times faster and 5Ghz is not going to net you that kind of improvement. Even if you consider the 3960x's base clock of 3.3Ghz at stock, overclocking to 5.0 on all 6 cores is only about a 50% increase and you'd need a lot more than that to close the gap with the numbers in this review (more like >250%, which isn't realistic.) I'm more apt to believe a review from a reputable source than the claims of someone on a forum on the internet, particularly since the claim is exactly that, a claim. I'm willing to entertain what you're suggesting if you provide numbers to back it up, but I think the reality is that Sandy Bridge is 8 years old and it's starting to look like it as well, at least that's what the numbers say, which I trust a bit more than anecdotes.

That isn't to say I don't enjoy my 3820 or 3930k, they're just dated and I think that review highlights that.

 

[dalekdukesboy]

That's what I thought about my 3930k until I saw this review that Phoronix did comparing the 3960x to modern 6c CPUs and the results paint a very stark picture.
https://www.phoronix.com/scan.php?page=article&item=sandy-extreme-2019&num=1
View attachment 132697

Mind you, this is basically saying that the 8700k and Ryzen 5 3600X are almost 3 times faster than a 3960x. No overclock is going to fill that gap and is a reason why I'm both running at 4.75Ghz right now and considering an upgrade to a 3900x within the next 6 months.

Games a lot of the time aren't CPU bound, particularly at higher resolutions, and that means nothing unless all you care about is games. The geometric mean says it's a little under 3 times faster and 5Ghz is not going to net you that kind of improvement. Even if you consider the 3960x's base clock of 3.3Ghz at stock, overclocking to 5.0 on all 6 cores is only about a 50% increase and you'd need a lot more than that to close the gap with the numbers in this review (more like >250%, which isn't realistic.) I'm more apt to believe a review from a reputable source than the claims of someone on a forum on the internet, particularly since the claim is exactly that, a claim. I'm willing to entertain what you're suggesting if you provide numbers to back it up, but I think the reality is that Sandy Bridge is 8 years old and it's starting to look like it as well, at least that's what the numbers say, which I trust a bit more than anecdotes.

That isn't to say I don't enjoy my 3820 or 3930k, they're just dated and I think that review highlights that.

I only glanced at review but "Geometric mean of all tests" and a quick glance at tests shows many things not only I don't use but are synthetic benchmarks which are nice to show off (I do) but admittedly I care far more about gaming and real world apps. Yeah, I don't expect X79 to be too close at stock with 6 cores to relatively new tech but...my comment about X79 being competitive still etc is with Ivy Bridge not Sandy and with 8 cores OC'd with fast memory, I'd like to see Phoronix do a review with my system on all these tests.

I'm purposely posting this before I go through Phoronix's article but gut reaction and experience guided by intelligence tells me one article of one set of tests by one guy is not enough to convince me to toss my rig aside for something new/er. Also performance increases with OC's are not strictly linear exercises based on clock speed alone, good way to guide what you expect from OC but depending on architecture and what you overclock CPU/Bus/memory etc the actual results vary to some degree...otherwise, you'd not even need to OC and do an article just crunching #'s on what an OC SHOULD do based on clock speed increase by percentage.

Anyway I like the spirited debate and just my 2 cents and yeah X79 isn't cutting edge and depending on benchmark or application of course it may show it's age, but on others many of which matter most to some users, we can't perceptibly see the difference.

 

[Kanan]

Games a lot of the time aren't CPU bound, particularly at higher resolutions, and that means nothing unless all you care about is games. The geometric mean says it's a little under 3 times faster and 5Ghz is not going to net you that kind of improvement. Even if you consider the 3960x's base clock of 3.3Ghz at stock, overclocking to 5.0 on all 6 cores is only about a 50% increase and you'd need a lot more than that to close the gap with the numbers in this review (more like >250%, which isn't realistic.) I'm more apt to believe a review from a reputable source than the claims of someone on a forum on the internet, particularly since the claim is exactly that, a claim. I'm willing to entertain what you're suggesting if you provide numbers to back it up, but I think the reality is that Sandy Bridge is 8 years old and it's starting to look like it as well, at least that's what the numbers say, which I trust a bit more than anecdotes.

That isn't to say I don't enjoy my 3820 or 3930k, they're just dated and I think that review highlights that.

Unless you're a newbie I dont need to give you links, I hope. It is well known that Haswell has about 10-15% more IPC than Sandy Bridge. Skylake on top of that has another 10% which makes it about 30% total. This means, a 6 core SB processor will match a 4 core Skylake in real life. I do not care much about your website as its benchmarks are meaningless, I've read these things a hundred times, seen a 1000 benchmarks, which are proving what I say.

In the very least, it is ridiculous to believe Skylake is 300% faster than Sandy Bridge. It is only that way maybe in very special situations, but not in important applications or games. It is well known to everyone, the famous Intel Meme, that they barely did anything in the last years when it came to improving their architecture, the IPC gains were minimal and only a few features such as AVX2 were added, which are unimportant however for most users. And now you come here and want to tell me everything people said for multiple years is wrong and Intel did a 300% improvement over SB? Uh no. That's bullshit. Ryzen forced Intel to increase core count, while Skylake is still their newest architecture and was refreshed two times. It is only 30% faster than SB, which makes a lot of sense, given the fact that it is only 2 major architecture changes above it. Ryzen 3000 itself has a slightly higher IPC than Skylake or Coffee Lake, whatever you wanna call it now. There's no magical 300% difference out of nowhere. You're kinda making it into, SB being a FX processor, which it is not. As the video that was posted here, has also proven, it can still up to this day mess with the best CPUs, even with low clocks at 3.6 GHz, and deliver high fps even at 1080p, which isn't optimal for old CPUs as it places the bottleneck on the CPU.

Honestly, it is pretty rare and odd that I read such a nonsense, given the overwhelming evidence that is there since years, and the fact that these CPUs are still used by many people today, even combined with strong GPUs. Also given the fact that it is only 2 major architecture changes apart from the newest stuff Intel has, makes it even more odd. I would say, you have to prove to me, that the new architectures are 200-300% faster, in real life applications and no nonsense like this, or there's no discussion left.

The 3960X at 4.5 GHz is as fast as a 6700K. And at 5 GHz will match or slightly surpass the 7700K, meaning it is still a very capable processor today. The only downside is its power consumption with high overclocks applied.

I asked a few weeks ago if my CPU would be fast enough to support the 5700XT. And you yourself told me it would be no problem. Now you're kinda trying to tell me my processor is outdated garbage. Uh, I'd recommend you to form one opinion and clear your thoughts of meaningless garbage that is posted in the internet. Everything I've said is hard facts btw. Nobody should be derailed by this circumstantial misinformation.

 

[Aquinus]

Unless you're a newbie I dont need to give you links, I hope. It is well known that Haswell has about 10-15% more IPC than Sandy Bridge. Skylake on top of that has another 10% which makes it about 30% total. This means, a 6 core SB processor will match a 4 core Skylake in real life. I do not care much about your website as its benchmarks are meaningless, I've read these things a hundred times, seen a 1000 benchmarks, which are proving what I say.

In the very least, it is ridiculous to believe Skylake is 300% faster than Sandy Bridge. It is only that way maybe in very special situations, but not in important applications or games. It is well known to everyone, the famous Intel Meme, that they barely did anything in the last years when it came to improving their architecture, the IPC gains were minimal and only a few features such as AVX2 were added, which are unimportant however for most users. And now you come here and want to tell me everything people said for multiple years is wrong and Intel did a 300% improvement over SB? Uh no. That's bullshit. Ryzen forced Intel to increase core count, while Skylake is still their newest architecture and was refreshed two times. It is only 30% faster than SB, which makes a lot of sense, given the fact that it is only 2 major architecture changes above it. Ryzen 3000 itself has a slightly higher IPC than Skylake or Coffee Lake, whatever you wanna call it now. There's no magical 300% difference out of nowhere. You're kinda making it into, SB being a FX processor, which it is not. As the video that was posted here, has also proven, it can still up to this day mess with the best CPUs, even with low clocks at 3.6 GHz, and deliver high fps even at 1080p, which isn't optimal for old CPUs as it places the bottleneck on the CPU.

Honestly, it is pretty rare and odd that I read such a nonsense, given the overwhelming evidence that is there since years, and the fact that these CPUs are still used by many people today, even combined with strong GPUs. Also given the fact that it is only 2 major architecture changes apart from the newest stuff Intel has, makes it even more odd. I would say, you have to prove to me, that the new architectures are 200-300% faster, in real life applications and no nonsense like this, or there's no discussion left.

The 3960X at 4.5 GHz is as fast as a 6700K. And at 5 GHz will match or slightly surpass the 7700K, meaning it is still a very capable processor today. The only downside is its power consumption with high overclocks applied.

Yet you won't provide any supporting evidence for your claims and you call what I wrote garbage? At least I have something to substantiate my claims.

I asked a few weeks ago if my CPU would be fast enough to support the 5700XT. And you yourself told me it would be no problem. Now you're kinda trying to tell me my processor is outdated garbage. Uh, I'd recommend you to form one opinion and clear your thoughts of meaningless garbage that is posted in the internet. Everything I've said is hard facts btw. Nobody should be derailed by this circumstantial misinformation.

You must have trouble reading. In case you didn't notice, I have a 3930k that I play games on...

Games a lot of the time aren't CPU bound, particularly at higher resolutions, and that means nothing unless all you care about is games.

Maybe all you care about is games, in which case you're golden, but I do a lot more than just play games on my machine.

I only glanced at review but "Geometric mean of all tests" and a quick glance at tests shows many things not only I don't use but are synthetic benchmarks which are nice to show off (I do) but admittedly I care far more about gaming and real world apps.

I respect that and if you notice, not a single one of those benchmarks in that review are games. Games are testing a whole lot more than just the CPU and most of those benchmarks are focused on just testing the CPU, because that's the intent. My real world situations might be a little different than yours as well, so I might have additional reasons to upgrade. So for context, I'm a software engineer and I write software and run tests on my machine and my language of choice handles multithreading very well. Something like tests I currently run for another piece of software I work on are very CPU bound and literally the faster a single core is, the faster tests runs.

I'll give you a real world example (for me.) That same test suite I mentioned, which I will run on my laptop at work with a mobile i7 8550U, is about as fast as my tower.

 

[Kanan]

Yet you won't provide any supporting evidence for your claims and you call what I wrote garbage? At least I have something to substantiate my claims.

It is not needed to provide evidence for something as well known and 1000 times proven as this. But because I'm not a asshole, here you go, literally a 2 second google: https://www.anandtech.com/show/9483/intel-skylake-review-6700k-6600k-ddr4-ddr3-ipc-6th-generation/9

You have nothing, no. Those benchmarks are not relevant for my usage which is gaming and some editing. As seen on my link, Skylake is nowhere near 300% as fast as Sandy Bridge. Nor did Intel themselves ever claim such a feat. Sorry, you're simply talking nonsense, as with your link, it is bullshit.

You must have trouble reading. In case you didn't notice, I have a 3930k that I play games on...

Strange, so you're basically talking shit about your own CPU.

Even if I would only play games, a lot of them stress the CPU pretty hard and if the CPU would be 33% of the performance a Skylake or Ryzen 3000 has, it would simply not even be fast enough to drive my 980 Ti let alone a 5700 XT I plan to upgrade to. And it would choke on BF1 and other CPU heavy games. So, no, you're wrong.

 

[Aquinus]

It is not needed to provide evidence for something as well known and 1000 times proven as this. But because I'm not a asshole, here you go, literally a 2 second google: https://www.anandtech.com/show/9483/intel-skylake-review-6700k-6600k-ddr4-ddr3-ipc-6th-generation/9

Multi-threaded applications are going to show more pronounced results when you have systems with more cores and you're also talking about quad core CPUs that are 4 years old as opposed to CPUs that are only a year or two old with 6c/12t. Even on the linux benchmarks in that review, the results are pretty pronounced and adding more cores at higher frequency is going to make that difference even more pronounced. Now it's not the same as the review I provided, which is comparing modern 6c/12t chips with a 6c/12t SBe. The simple point is that if you use those cores, you'll notice a difference. Just because it's not your bottleneck, doesn't mean it's just as fast.

You have nothing, no. Those benchmarks are not relevant for my usage which is gaming and some editing. As seen on my link, Skylake is nowhere near 300% as fast as Sandy Bridge. Nor did Intel themselves ever claim such a feat. Sorry, you're simply talking nonsense, as with your link, it is bullshit.

What you use your CPU for doesn't change the numbers, it just says what you care about which isn't what I'm talking about. You have yet to disprove anything I've said and you're getting awfully defensive when I say that Sandy Bridge is getting old and you're basically saying that Michael over at phoronix is full of crap... because it doesn't matter for your use case? Think about what you're saying, a review is full of crap and that an 8 year old CPU is going to keep up with a modern CPU as if it were brand new. I understand that you might not be fully taxing your CPU for your uses, but for those of us that do, it makes a difference, whether you want to admit that or not. As I said earlier:

I'll give you a real world example (for me.) That same test suite I mentioned, which I will run on my laptop at work with a mobile i7 8550U, is about as fast as my tower.

My laptop is thermally constrained with lower clocks, yet it does the same job just as well with a fraction of the power.

If you want to validate his numbers, Phoronix Test Suite, the tool he made and uses for his benchmarks which is widely used in the Linux community, can be run on your own machine to run the same tests if you really don't believe it. So instead of talking shit, I challenge you to prove him wrong.

Strange, so you're basically talking shit about your own CPU.

I'm saying that it's dated and that 6c/12t CPUs nowadays are significantly faster than our SBe chips while consuming a lot less power. It's ignorant to pretend like SBe isn't 8 years old.

Even if I would only play games, a lot of them stress the CPU pretty hard and if the CPU would be 33% of the performance a Skylake or Ryzen 3000 has, it would simply not even be fast enough to drive my 980 Ti let alone a 5700 XT I plan to upgrade to. And it would choke on BF1 and other CPU heavy games. So, no, you're wrong.

Games are a poor benchmark of CPU performance since it depends on a lot of other things and if your CPU isn't your bottleneck you're not going to notice a difference. Your CPU could be 10x faster, but that won't matter if it's not your bottleneck. It's like buying more ram, if you're not running out, you're not going to notice a difference.

 

[dalekdukesboy]

Just be nice guys, that's all I ask, or mods will be back in here bashing heads. Also like all your posts generally Aquinus and Kanan as well so no favoritism here other than my own biased feeling owning the platform myself.

Aquinus, I'm not a linux guy and thanks for providing that test suite, because I will run the tests because I'm curious on what my results will be vs. his with my system, I got Gimp downloaded and I can say doing the various photo tests whatever size picture I used etc it finished in only a few seconds...so maybe he's using massive photo's/files but I will check out his suite so I can do apples to apples hopefully it won't depress me but is what it is lol.

 

[Aquinus]

Just be nice guys, that's all I ask, or mods will be back in here bashing heads. Also like all your posts generally Aquinus and Kanan as well so no favoritism here other than my own biased feeling owning the platform myself.

Aquinus, I'm not a linux guy and thanks for providing that test suite, because I will run the tests because I'm curious on what my results will be vs. his with my system, I got Gimp downloaded and I can say doing the various photo tests whatever size picture I used etc it finished in only a few seconds...so maybe he's using massive photo's/files but I will check out his suite so I can do apples to apples hopefully it won't depress me but is what it is lol.

I wouldn't be here if SBe hadn't treated me well for almost the last 8 years, but it has been just that, almost 8 years.

As for PTS, it has settings for running all of the tests and stuff, so using the tool will replicate what Michael does.

 

[k_9virus]

awesome discussion guys! for me ill be in this platform for couple more years lol btw 3960x @4.5ghz and vega64 runs all my games on ultra in decent frame rate @2560x1080 60hz monitor

 

[Aquinus]

awesome discussion guys! for me ill be in this platform for couple more years lol btw 3960x @4.5ghz and vega64 runs all my games on ultra in decent frame rate @2560x1080 60hz monitor View attachment 132781

I've been fairly satisfied with my 3930k and Vega 64 at 4k. CIties Skylines is the only game that I've been playing lately that hits the CPU pretty hard, particularly with bigger cities. `htop` was showing ~4 full threads consumed by it last time I was monitoring load while playing. Overclocking more seems to help, so I've probably hit a CPU bottleneck. Other than that, I haven't really had any issues. I didn't start with the 3930k either, I started with a 3820 which was more than enough, but I wanted a cheap upgrade about a year and a half ago. I still have the 3820 in another machine with another P9X79 Deluxe. It doesn't get a lot of use these days though.

 

[k_9virus]

i started with 3820 too then 2640 v2 which not OC friendly i guess 3960x is the sweet spot.

 

[phill]

I've managed to get a little lucky and grab a few X79 boards of late... Here's some quick pictures

I think I have one more up in the loft, a Rampage 4 Extreme (4 memory slots on the board??) been after a black edition for a while but not had any luck with a friends board, so I bought another

Love the board, hear it's a little flakey with random RAM and that Corsair is the safe bet to go for.. Not a major fan of Corsair but I've a few 4Gb sticks so that'll do just nicely

I hope to get a few hours with the boards to test them a bit more, see what they can do

 

[Lorec]

I've managed to get a little lucky and grab a few X79 boards of late... Here's some quick pictures

View attachment 132933

View attachment 132934

I think I have one more up in the loft, a Rampage 4 Extreme (4 memory slots on the board??) been after a black edition for a while but not had any luck with a friends board, so I bought another

Love the board, hear it's a little flakey with random RAM and that Corsair is the safe bet to go for.. Not a major fan of Corsair but I've a few 4Gb sticks so that'll do just nicely

I hope to get a few hours with the boards to test them a bit more, see what they can do

Whats with X79 that keeps even the mainstream users still so interested?
X58 and X99 is everywhere, while X79 is so rare.
Searching for my Asus Sabretooth X79 took me around a year

 

[FreedomEclipse]

4960X for over $1000 anyone?


::EDIT::

Fixed typo and link

 

[Lorec]

4969X for over $1000 anyone?

It has to be a higly binned golden sample no less

 

[dalekdukesboy]

I wouldn't be here if SBe hadn't treated me well for almost the last 8 years, but it has been just that, almost 8 years.

As for PTS, it has settings for running all of the tests and stuff, so using the tool will replicate what Michael does.

I admit I'm still figuring out how to get it installed and use the benchmarking stuff, I downloaded it and unzipped it...then from there it's not so obvious to me what to do with it.

 

[phill]

Whats with X79 that keeps even the mainstream users still so interested?
X58 and X99 is everywhere, while X79 is so rare.
Searching for my Asus Sabretooth X79 took me around a year

Well if I'm honest, I have X58 and X99 already... X79 was missing from my collection so.... But in all seriousness, I just enjoy hardware. I never really went the 115x route, I seemed to go the X58 etc.

 

[Aquinus]

I admit I'm still figuring out how to get it installed and use the benchmarking stuff, I downloaded it and unzipped it...then from there it's not so obvious to me what to do with it.

It's a command line application. Its options are probably in the man page for it, but this might help:
https://www.phoronix-test-suite.com/documentation/phoronix-test-suite.pdf

 

[dalekdukesboy]

It's a command line application. Its options are probably in the man page for it, but this might help:
https://www.phoronix-test-suite.com/documentation/phoronix-test-suite.pdf

Thanks, dickheads who wrote it though are typical arrogant pointy headed geeks never laid who insult you by calling everything (very basic) and (simple) like this in parenthesis to insult you as you struggle to understand what they are saying but is just oh so easy and basic it's demeaning for them to even have to write it to us pleebs who read it. Just sayin... .

 

[Aquinus]

Thanks, dickheads who wrote it though are typical arrogant pointy headed geeks never laid who insult you by calling everything (very basic) and (simple) like this in parenthesis to insult you as you struggle to understand what they are saying but is just oh so easy and basic it's demeaning for them to even have to write it to us pleebs who read it. Just sayin... .

It's probably one of the more fully featured benchmark applications IMHO. There are a lot of things that go into doing benchmarks across different applications and having it done in a consistent way. It's complicated because you can customize most of it. Mind you, I live in Linux all day every day for the last several years, so when I look at something like PTS, it's really not too bad, but it's definitely not a tool where it's just like "run the good benchmark," and you call it a day.

I wouldn't look into it too much though. The whole "hey, look how easy it is," is probably more marketing than anything else. I wouldn't take it personally.

 

[dalekdukesboy]

It's probably one of the more fully featured benchmark applications IMHO. There are a lot of things that go into doing benchmarks across different applications and having it done in a consistent way. It's complicated because you can customize most of it. Mind you, I live in Linux all day every day for the last several years, so when I look at something like PTS, it's really not too bad, but it's definitely not a tool where it's just like "run the good benchmark," and you call it a day.

I wouldn't look into it too much though. The whole "hey, look how easy it is," is probably more marketing than anything else. I wouldn't take it personally.

No worries, I did some basic coding on a 383 in 1993 and hated it, done a little for a few occasions since (all for high school/or college) and though all of it was fairly simple programs compared to game/full application codes I recoiled like Dracula shown the cross most of the time.

It was kind of late and of course I wanted to do something with it quickly...which as you said even for you it's not just "run a good benchmark" and you're done. It requires what many men hate doing, reading instructions and actually following them, and with pc code/commands you can't use any duct tape/string/and/or tie wraps to make it work, which I have frequently done with my physical pc hardware because often I'm throwing parts in a box in such a way that they weren't thought of to be used like my cooling rads which fit but there are no holes predrilled in case etc to actually hold it there.

Anyway, in all seriousness I don't take much personally as you can see on this thread few other people have lost their cookies and I think even the mods got a bit cranky cracking down on us but I just for the most part laughed it off. I bitch about people who write programs or things that are PITA's to me and who I will never meet so I can complain and feel better, and their feelings won't be hurt because not like I'll ever meet them or even know who they are and certainly vise versa .

I did pull up the instructions you gave and I was going to throw in a quip that it looked like instructions to the NASA space shuttle but yeah it isn't like some benchmark tools with presets and you just pick the presets in Heaven/Aida etc and you call it a day.

I'm actually headed to Canada for about a month starting tomorrow so I've been trying to get my stuff together it's a work/cation with some fun (hopefully) with my Dad but I admit I just kinda wish I was staying in my own house at home and wasn't a captive audience to his schedule for a month...he does need help with a lot of work up there so I'll make some money but it'll be like having your boss and your father with you at all times...not sure how long it'll be before I'm wishing for a fast forward button on time or that I'm back at my own house, so I know this is way off topic OT but I am the thread starter and explaining why I'll have no internet access for a while and won't be on.

So with that said I hope you guys have plenty to fill in till early November with and I'll be back and hopefully play around with that Phoronix suite and do a little hack work on my build and finally get the side on etc but with rad hoses etc just very hard to do and I hate flexing hoses too much and risking leaks etc just to fit things in.

 

[Fastie]

Great thread, I recently got hold of an Asus P9X79 WS which I have an e5-2680v2 and 32gb 1866 ECC ram in.
I did some very short testing with some BCLK overclocking but had no luck. Havent found many people trying to OC this cpu, and not much info on voltages etc to make it run.
Would like to get some higher clocks on it without messing up the ramspeed too much.

I run servers on it as a hobby and some gaming, thus the need for many threads.

 

[Grog6]

I realized today that the OC problem I'm having is due to two bent pins.

I'm going to try to repair them; I wonder what it costs to get a socket replaced in the States?



EDIT: ~$100, I'll go that route if I must.