Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!
Quick Links
What is the best price/quality motherboard P67 1155?
First I'd like to purchase from canada to avoid brokerage fees at the border.
I've been looking at the MSI (104$) that seems apealing.
http://www.newegg.ca/Product/Product.aspx?Item=N82E16813130576
I don't need anything but some usb ports, one X16 PCIE, optical is interesting, 4 slots of ram and possibility to do SMALL overclocks but not much. I don't care about SLI or CrossFireX or USB3 or eSata.
I have a 2600K intel processor with hyperthreading and 4go of Corsair Dominator ram, plus a 260GTX video card, and an Enermax 500W modular PSU, my case is ATX.
Comments
Do you really want to pair Intel's top of the line Core i7 2600K processor with MSI's bottom of the line P67 motherboard? Really? Seems like something of a mismatch to me.
5 power phases with the power circuitry exposed to the open air without so much as a heatsink is not the sort of motherboard you want to overclock on. But for even a modest overclock, you might want something a little stronger than that, like one of these:
http://www.newegg.ca/Product/Product.aspx?Item=N82E16813128476
http://www.newegg.ca/Product/Product.aspx?Item=N82E16813130583
http://www.newegg.ca/Product/Product.aspx?Item=N82E16813131705
http://www.newegg.ca/Product/Product.aspx?Item=N82E16813128488
Furthermore, if you're only going to do a small overclock, you don't need a Core i7 2600K for that. The non-K version will turbo boost up to 3.8 GHz on its own when it decides you need the extra speed, and as high as 4.2 GHz for short periods of time on a single core. You can turn off turbo boost and manually overclock it to 3.8 GHz, too. The unlocked multiplier only makes sense if you're going to overclock it significantly beyond 3.8 GHz.
Really, though, if you've got the budget for a Core i7 2600K, then there's no need to skimp on the motherboard. You could pick up something a lot more feature laden that will work for whatever you later decide to do with it:
http://www.newegg.ca/Product/Product.aspx?Item=N82E16813130574
http://www.newegg.ca/Product/Product.aspx?Item=N82E16813131703
http://www.newegg.ca/Product/Product.aspx?Item=N82E16813128478
My guess is that you don't actually have a large enough budget for a Core i7 2600K to make sense, but would be better off switching to a Core i5 2500 or 2500K, and spending that extra money elsewhere, such as on a good SSD.
http://www.newegg.ca/Product/Product.aspx?Item=N82E16820148441
I already received my I7 2600K by mail, I do have budget for a good PC I just don't need the fancy features of bigger boards, like USB3 or eSata is of no importance to me.
I have read somewhere that power phase design was not important for small OC's because those processors consume less power and the new VRD will keep the voltage more stable, and also a good 4 phase will outclass a bad 8 phase or 12 phase which would be in smaller chips anyway.
What do you think?
https://i.imgur.com/Dc8dU3M.png
Certainly, not all power phases are identical. But what do you think MSI puts in their bottom of the line motherboard? Do you think they aggressively seek out the highest quality power phases they can find? Or do you think they try to find the cheapest thing that they're pretty sure will work at stock speeds with a hardware-enforced TDP of 95 W?
Imagine that you're MSI. You're trying to produce the cheapest P67 motherboard that you can, to compete on price at the low end. You know perfectly well how to produce a motherboard that can safely deliver 500 W to the processor (well, safe for the motherboard, but not so safe for the processor), while keeping the voltages stable. But that's also rather expensive to build. You also know how to make a motherboard that can properly deliver up to 200 W, or up to 150 W, or up to 95 W, or up to any other number within reason. The lower you set the threshold of how much the motherboard has to deliver, the cheaper it is to build.
Furthermore, you like Intel's Turbo Boost feature, as it lets the processor measure power consumption in real time. If a program makes power consumption spike upward, so that the processor suddenly pulls more than its usual TDP of 95 W, then the processor will clock itself down within a fraction of a second, and without waiting until it has already overheated to do anything about it. If power consumption is lower than 95 W, it can overclock itself a fair bit. This ensures that at stock settings, the processor will never pull more than 95 W for a thermally-significantly length of time.
So what do you do? Do you make sure that your lowest end motherboard will be safe at 150 W? Or do you design the cheapest power circuitry that you can that will be safe at 95 W, even if it's unsafe at 120 W? The latter is cheaper, and if someone buys it and fries it, you can blame the customer for being an idiot and overclocking on a bottom of the line motherboard.
If power circuitry didn't matter at all, then why don't motherboards all just go with two power phases? That's the minimum necessary to function, as it needs separate phases for the processor and memory. Motherboard manufacturers know perfectly well how to build a motherboard with two power phases:
http://www.newegg.com/Product/Product.aspx?Item=N82E16813154020
The question answers itself, really. What is safe for a chip with a TDP of 18 W isn't necessarily safe at 95 W. And likewise, what is safe at 95 W isn't necessarily safe at 150 W.
If you overclock, then you lose the hardware-enforced TDP of turbo boost. The processor might well still pull safe levels of power most of the time. But most of the time is not all of the time, and you only have to kill a part once for it to stay dead.
When you pull more power than hardware is designed to safely deliver, a variety of bad things happen. One is that the voltage drops. Pulling more power from various power circuitry naturally causes the voltage to drop anyway, but there are various things that power supply manufacturers can do to correct for it. But they only correct for it within a certain range that the hardware is designed to handle, and outside of that range, voltages can be problematic.
A processor at a given clock speed needs at least a certain minimum voltage to run. Above that voltage, it works just fine, and below it, it crashes. Better power circuitry can prevent the voltage from dropping as far under a given load. If a processor needs at least 1.15 V and is running at 1.2 V, that should theoretically be fine if not for the voltage droop. If in practice, the voltage sometimes drops to 1.19 V, that doesn't matter. If on a low end motherboard, it sometimes drops to 1.14 V, then it crashes. You can correct for that by having it nominally run at 1.25 V, so that even with the voltage drop, it stays high enough. But then you're pulling even more power, and putting out even more heat.
Another problem is ripple. Even if the average voltage that a motherboard delivers is 1.2 V, at a millisecond scale, it's not a steady 1.2 V, but fluctuates. I'm not sure how precisely motherboards can deliver voltages to the processor, so I'll make up numbers for the sake of example. Maybe instead of a steady 1.2 V, the voltage fluctuates chaotically between 1.19 V and 1.21 V. If the ripple is small enough, then it doesn't matter much. If it gets too large, though, then it can cause the processor to crash, or even to be permanently damaged. Is that really what you want to do to your top of the line processor?
Now, the motherboard might work just fine. But overclocking on a bottom of the line motherboard is a dumb risk. If you put one bullet in a revolver, spin it randomly so that you don't know where the bullet is, point it at your head, and pull the trigger, you'll probably survive. But even if you do so and live, that doesn't mean it was a good idea, even if you win a $100 bet by doing so.
Fine, I totaly agree with you on the voltage matter. Then, which one offers the best voltage regulation? Asus's VRD or Gigabyte's true phase design? Or is it MSI? That is very hard to guess from my point of view
All the P67 boards seems very unstable, there are always 50% good reviewsand 50% terrible reviews on sandy bridge motherboards, like they're plagued. I don't remember seeing such a big fail rate on LGA 775 back in the days.
Do you think turbo boost is the cause? People are getting those weird random reboot apparantly? Bad board, or bad cooling etc.. ?
https://i.imgur.com/Dc8dU3M.png
I got the Asus P67 board (have to look the model up later, one of the no frills micro ATX form ones) from the tiger direct store down the way from me with no issues over the pas ~3 months of use. I always buy Asus and out of the ~10 boards only had one dud and they replaced that with a winner fairly quickly.
Cheers!
MMO Vet since AOL Neverwinter Nights circa 1992. My MMO beat up your MMO. =S
Don't worry too much about the reviews. It's easy to just blame the motherboard. Many of the bad reviews are from first time builders that don't know how to troubleshoot thier problems, and then blame the board. Alot more people are building their own rigs these days, so it seems normal.
I would recommend a Gigabyte board, such as the UD3. I read somewhere that Gigabyte went all out on the quality of the P67 boards. I have only used Gigabyte boards though, so my opinion is a bit biased.
I own the P67 UD3P and it's a good board so far. No problems running my 2600K at 4.5ghz, and it's been a couple months. It looks like newegg is selling the Z68 UD3P for the same price that I paid for the P67 a couple months ago. Not a bad deal for a top quality board.
It's not just the brand name. The major motherboard brands all offer a full lineup of boards designed to fill every market segment. It's kind of like asking whether Nvidia or AMD makes better gaming video cards. The answer is, they both make a full range of cards, going from high end gaming cards to low end cards not meant to run games. If you buy a Radeon HD 6450 or a GeForce GT 520 and then it doesn't run games as well as you'd like, the problem isn't AMD or Nvidia; the problem is buying the lowest end card of the generation.
Gigabyte did go overkill on power delivery for the -UD3P and higher motherboards. They didn't do so for the -UD3 and lower ones. For Asus, the analogous thing is the P8P67 and up, but not the LE or LX versions. MSI really only went overboard on the -GD80, though the -GD55 and up should have plenty of power to not be a problem unless you're trying to fry the processor--and I'd call clocking the processor at 5 GHz trying to fry it, no matter what cooling system and motherboard you're using.
That said, going overkill on power delivery is a huge deal if you want to push the processor to its limits, but not necessary if you just want a small overclock. How small of an overclock is "small" anyway? I'd expect the MSI -G43 and the Gigabyte -UD3 to be adequate if you're adamantly not going to overvolt anything (even if it won't overclock as far as you'd without overvolting!), and not going to try to push the processor above 4.2 GHz or the memory above 1600 MHz. With shipping, the Asus P8P67 isn't that much more expensive than either of those, and would let you clock stuff however you want without the motherboard holding you back, provided that you're using either air or water for cooling.
-----
I'm not sure why there are a lot of negative reviews. There seem to be a lot of negative reviews on New Egg of Asus motherboards in particular for Sandy Bridge motherboards. Whatever problems the random people on New Egg ran into didn't seem to show up in reviews from tech media sites.
There are a few things that I can think of that would cause Sandy Bridge motherboards to have systematically more problems than those of older generations. One is the transition from BIOS to UEFI. Motherboard manufacturers have been writing BIOSes for decades, but this is the first generation of consumer hardware to use UEFI. It's plausible that not having the same sort of experience with it yet means that there tend to be more bugs in the new UEFIs than in the old BIOSes.
Another potential issue is having a single clock generator for everything. If people aren't aware that Sandy Bridge is different, and try to overclock their processor by modifying the base clock, that will make things crash. Sandy Bridge processor overclocking is multiplier-only, apart from the potential for slight tweaks to the base clock. If that's leading to a number of negative reviews, then it's not a problem with the motherboards themselves, but only with idiots who buy them.
A third possibility is that it's something wrong with the platform itself, and not the motherboards of any particular brand. Maybe the SATA 2 bug in the early chipsets isn't the only problem, and something else wrong with the chipsets or sockets or whatever is causing problems. I don't have any reason to believe that this is the case, but if it were, it would explain a lot of negative reviews.
One thing to remember is that people who post reviews are a self-selecting sample. Someone who buys a motherboard, plugs everything in, and has it work, might not think of the motherboard again afterwards. That's certainly a good outcome, but he likely won't post a review of the motherboard. Someone who has a motherboard die and is upset about it is far more likely to post a one egg review. Furthermore, motherboard problems are hard to diagnose, so even if it's a different part that is the problem, he might blame the motherboard.
------
There's also the potential for FUD campaigns. What stops someone working for one brand from writing negative reviews of products of other brands? New Egg does at least mark reviews by whether the person bought a product from New Egg, which clamps down on that somewhat. But likewise, what stops someone working for one company from posting on tech forums, and talking up that company's products, while trashing the competition? It's not hard to pick a product you've never used, register for a forum, and say that you bought that product and it died and the company wouldn't replace it.
That's not to say that everyone who is far too enamored of one particular brand is actually working for this or that company and trying to spread FUD. There are a lot of fanboys in the world who love this brand and hate that one. A thread about Apple can easily turn into a flame war, but that hardly means that everyone who posts in such a thread works for either Apple or Microsoft. It really only means that Apple manages to inspire strong reactions, both for and against their products.
Good points! I won't overclock much but either way the Gigabyte Z68 does seem to put emphasis on voltage stability, I'll read more on that one tonight. Thanks!
+ I'll keep you updated with how well the 2600K goes!
https://i.imgur.com/Dc8dU3M.png
In the end the Asus P8P67 (rev3) seems to have all the features I like, and the special price of 149.99$ and free shipping is just perfect for me (newegg.ca).
I'm sending in my order at the moment, hope it won't be a bad board like some had haha :P
Will my 500W Enermax PSU be enough for my I7 2600K, 4GO of DDR3 Corsair Dominator, 1 HDD, and a 260GTX without any overclocks?
https://i.imgur.com/Dc8dU3M.png
The Asus P8P67 is basically their equivalent of the Gigabyte GA-P67A-UD3P. It's got plenty of power circuitry so that you could clock the processor however you want, and the motherboard won't hold you back.
Which Enermax power supply do you have? Any of their modern ones should have enough power for that system at stock speeds. The Naxn 80+ should be fine, though you'd want something stronger than that if you're going to overclock. The Tomahawk doesn't have enough power connectors to plug everything in. If you've got some older power supply before basically everything moved to the +12 V rail, that could be a problem.
I think it's an Enermax Liberty 500W Modular PSU
http://bigbruin.com/reviews05/review.php?item=enermaxliberty&file=1
https://i.imgur.com/Dc8dU3M.png
That was a nice enough power supply in its day. The problem is that its day came way back in 2005. Today, it's a relic of another era, and a 500 W power supply targeted at the same market segment would offer an awful lot more than 32 A combined on the +12 V rails.
Furthermore, some components degrade with the passage of time, so even if, when the power supply was brand new, it would have handled your system without any problems, it might not still be able to do so today. My guess is that it would probably work just fine, but I wouldn't want to risk it, when a more modern power supply isn't that expensive.
Good enough quality:
http://www.newegg.com/Product/Product.aspx?Item=N82E16817371044
Higher quality, lower wattage:
http://www.newegg.com/Product/Product.aspx?Item=N82E16817371047
Very good quality, and a nice deal if you regard rebates as "free":
http://www.newegg.com/Product/Product.aspx?Item=N82E16817139020
What will happen if I try it and my power supply doesn't deliver enough amps? Can I damage components or will my system just crash randomly whenever the demand is too high in energy?
https://i.imgur.com/Dc8dU3M.png
It depends. Some power supplies will recognize that you're pulling way too much power and shut down, ideally before damaging hardware rather than after. That probably wouldn't happen to you, as the overcurrent protection seems to be there mainly to protect against a malfunction that causes an enormous, unexpected power draw.
You'd certainly get very low energy efficiency if you try to pull too much power from a power supply. Voltage regulation and ripple tend to get much worse as you pull too much power. That can cause crashing, or damage hardware. That can be an abrupt case of parts dying outright, or a slow case of intermittent problems that are an awful pain to reproduce, isolate, and diagnose.
Note also that how much power is "too much" for a power supply decreases as time passes, due to the aging of various components in the power supply. Even if it works great today, it might not still work so great in a couple of years.
The power supply will deliver the amperage that components try to pull from it. The question is how well it will deliver it. The rated wattages on a power supply basically mean that the company said, it can safely deliver up to this much. Beyond that, it will deliver more wattage/amperage, but not necessarily safely. Enermax is honest with its power supply wattage ratings, unlike some companies.
To expand on Quiz's answer a bit:
A) The power supply will run hot. Probably very hot.
C) In extreme cases, the power supply will shut down. Overcurrent protection isn't really there to protect you exceeding the rating, it's there in case you accidently short circuit a rail to prevent the power supply from catching on fire, acting more like a fuse or circuit breaker (see A).
There is almost a certain chance the computer will crash, and a very good chance that something will break: either the voltage droop frying something electrical (almost always RAM, motherboard, or video card power circuitry), or the power supply burning out a capacitor or solder trace because it is running too hot.
Bad/inadequate power supplies are one of the hardest problems to troubleshoot. The voltage droop can cause any part of the computer to cause a BSOD: One time you get a video card error, another time a network card driver reset, another time it's invalid memory exception. And that's if your lucky and it doesn't just outright fry something, in which case you think "A-HA, I found the problem", replace the fried component, and fry it again in a week or two because the real problem was the power supply in the first place.
This is why people who know what they are doing with computers ~always~ harp on getting a good power supply.
Well I now have a very awesome 900W power supply built by Huntkey, which is one of their well-rated PSU unit (and very stable too).
I'll probably never ever need a new PSU for years, and it can run my system fully overclocked without a hitch!
I'm running my 2600K at 4.4 GHZ right now and my 260GTX overclocked by 20-30% as well.
No crashes, no hard drive or processor stalls, no blue screens!
I absolutely love it!
The motherboard is awesome so far!
https://i.imgur.com/Dc8dU3M.png
Huntkey doesn't make any "very awesome" power supplies. Or even kind of awesome. At best, they make kind of "meh". And then it goes downhill from there. Way downhill, to the point that most of their products do this:
http://www.hardwaresecrets.com/article/Huntkey-Green-Star-550-W-LW-6550SG-Power-Supply-Review/668/8
The only 900 W power supply that Huntkey lists on their site is the Huntkey X7 900 W, which should probably be all right. Not super awesome, but all right. Not having any +12 V rail rated above 18 A is not what you want for overclocking, though.
I don't know what you paid for the Huntkey power supply, but you'd probably have been better off with the Corsair TX650 V2 that I linked above. That's better quality, and $93 including before a $10 rebate is probably less than you paid, too.
Read their review of the 700W Rocketfish. They say it's a completely different design, so is the 900W.
Obviously Huntkey low-end PSU's are NOT trustable and some cheap crap, but these guys know what they're doing with the 900W and 700W units I think, at least that's what I got off these reviews. Nice components, nice filters, etc...
The other PSU I was looking at is a Silent Pro 700W, think it's better?
https://i.imgur.com/Dc8dU3M.png
Why is your choice only between a couple of power supplies that are kind of all right, when there are a lot of power supplies out there that are actually quite good? The Cooler Master Silent Pro is all right, but it's not that good. If you had it, I'd see no dire need to run out and replace it, but it's not something I'd buy in the first place unless it's dirt cheap and you're on an extremely tight budget.
If this is what you mean by the Rocketfish 700 W review, then that's not exactly a glowing review:
http://www.hardwaresecrets.com/article/Rocketfish-700-W-Power-Supply-Review/556/10
Hardware Secrets gives a "golden award" to a large fraction of the products that they test (e.g., maybe half), and a lesser award to quite a few others. That product got no award. That review found mediocre build quality, mediocre ripple suppression, terrible energy efficiency, and an absurd price tag. When they highlight is that it stayed in spec and didn't explode, that's not a glowing review. And that's by 2008 standards; the competition is much better today.
Manufacturer reputations matter. Companies that have made good power supplies in the past are more likely to make good ones in the future. If a power supply is made by Seasonic, for example, then unless you get a defective unit by rotten lock, at worst, it might be kind of all right. And it's likely to be pretty good, or even really great. If you pick up a Diablotek unit, then you should expect something to die, and hopefully only the power supply.
If a power supply company wants to say, yeah, we've made some units that weren't so good in the past, but this one is different, then they have to get it reviewed to prove it. Just saying, this is our best model, trust us, it's good, means it probably isn't very good. If it were a really awesome power supply, then Huntkey would have done what Super Flower did with their gold platform recently: send it to every decent review site they can find for as many reviews as they can get, sending out a variety of wattages, and under a variety of brand names. Get back a bunch of glowing reviews from a bunch of different reputable tech sites and you've demonstrated that you've got an excellent product.