CPU cooler clearance

[red1776]

Well I for one like the fact that there is 200+ MB's available. It is a side effect of competition. When I look for a motherboard I look at as priorities

1) the correct capacities I need
2) Quality VRM/phasing
3) layout and functionality

lots of sub-catagories to that, but you get the idea.

For customer builds, I agree with you most of the time, but some of them are like me and fall into the "enthusiast" class and like the overkill over the top aspect, and are glad that some of this poor price to performance items exist. I will often pay twice as much for that last 10% of performance, just because I like this stuff.
I just rebuilt with a GA-990-FXA-UD7 in anticipation of the Bulldozer. I like getting the front line stuff and know going in that sometimes I am going to pay for it or get burned. It's a risk I have been willing to take, and don't ***** about it when I have gotten burned (see NB problem with the Asus Crosshair IV Formula)
I think this 'thermal armor" thing is a gimmick, but time will tell. and i think the heat-sink market in the air dept anyway is just coming up with ideas to make theirs look exotic rather than perform better. I just read a dreadful review on the new 'supercooler' Genesis...apparently not so good.
The biggest problem here, and places like these forums is that people that take an interest, and want to build for the first time, will spend a lot of money on the advice of the people here who respond. Too many people will defend their bad buying decisions and or insist that something is "the best" or "sucks" because thats the way they wish it was. It doesn't help for those who actually have a lot of experience 'hands on' with much of this product.
I like to think for the regulars though that if you frequent this place for a while , you can ferret out who to listen to. a good example is 'Dividebyzero' you can pretty much take what he says to the bank, and most here know that.

 

[nfiniti9]

Well I for one like the fact that there is 200+ MB's available. It is a side effect of competition. When I look for a motherboard I look at as priorities

1) the correct capacities I need
2) Quality VRM/phasing
3) layout and functionality

lots of sub-catagories to that, but you get the idea.

For customer builds, I agree with you most of the time, but some of them are like me and fall into the "enthusiast" class and like the overkill over the top aspect, and are glad that some of this poor price to performance items exist. I will often pay twice as much for that last 10% of performance, just because I like this stuff.
I think this 'thermal armor" thing is a gimmick, but time will tell. and i think the heat-sink market in the air dept anyway is just coming up with ideas to make theirs look exotic rather than perform better. I just read a dreadful review on the new 'supercooler' Genesis...apparently not so good.
The biggest problem here, and places like these forums is that people that take an interest, and want to build for the first time, will spend a lot of money on the advice of the people here who respond. Too many people will defend their bad buying decisions and or insist that something is "the best" or "sucks" because thats the way they wish it was. It doesn't help for those who actually have a lot of experience 'hands on' with much of this product.
I like to think for the regulars though that if you frequent this place for a while , you can ferret out who to listen to. a good example is 'Dividebyzero' you can pretty much take what he says to the bank, and most here know that.

This is very true. Choice is what makes us human. You two both seem to be well informed but if we just took everything for fact then there would be no point of a discussion forum. There is very little fact in life, especially in computers. Wheres the fun without intelligent discussion

 

[red1776]

This is very true. Choice is what makes us human. You two both seem to be well informed but if we just took everything for fact then there would be no point of a discussion forum. There is very little fact in life, especially in computers. Wheres the fun without intelligent discussion

I agree completely...just don't like it when an OP spends a bunch of money when offered advice that is about as useful as a bumper sticker of Calvin pissing on a Ford emblem! ya know "intel sucks man!" LOL

 

[nfiniti9]

I agree completely...just don't like it when an OP spends a bunch of money when offered advice that is about as useful as a bumper sticker of Calvin pissing on a Ford emblem! ya know "intel sucks man!" LOL

Yeah thats kinda what I was feeling when I first came to this thread. I preach dont waste money but make sure you spend enough to get quality.

 

[red1776]

Yeah thats kinda what I was feeling when I first came to this thread. I preach dont waste money but make sure you spend enough to get quality.

The other part I edited was:

I just rebuilt with a GA-990-FXA-UD7 in anticipation of the Bulldozer. I like getting the front line stuff and know going in that sometimes I am going to pay for it or get burned. It's a risk I have been willing to take, and don't ***** about it when I have gotten burned (see NB problem with the Asus Crosshair IV Formula)

The point being that if BD turns out to be a dud, I'm not going to whine about it.

 

[nfiniti9]

The other part I edited was:


The point being that if BD turns out to be a dud, I'm not going to whine about it.

You know AMD is going to a new socket shortly after BD which will allow BD to really shine. Its going out on AM3+ now so they can get something out. I don't think they can handle the cpu and chipset all at once.

Personally I'm done with AMD for now unless I want a budget computer (they rock for $100 or less). I don't see how BD will really be much faster than Thuban at release. I was an AMD guy until Core 2 wiped the floor with them.

 

[red1776]

You know AMD is going to a new socket shortly after BD which will allow BD to really shine. Its going out on AM3+ now so they can get something out. I don't think they can handle the cpu and chipset all at once.

Yeah it sounds like 6-9 months after BD with the 'Trinity ' release.

I'm a little more optimistic about AMD and the new architecture, but I will know in about 60 days.

I don't see how BD will really be much faster than the Thuban at release

well a new architecture, a shrink , and going from SOI to HKMG holds a lot of potential. The details of the architecture have been few, so I am just going to buy the first FX 8130p or 8150P and let'er rip.

 

[nfiniti9]

I hope so. We need competition for Intel.

 

[red1776]

I hope so. We need competition for Intel.

One guy posted in another thread " I hope Bulldozer rocks!....so I can afford a i72600k!"
:haha:

 

[dividebyzero]

See the OC ability is what I question. I've been overclocking since the C300 days and I have rarely seen a motherboard magically give you and extra 100-200 mhz without going into unstable voltages etc.

Kind of depends on what era/chipset you're talking about. For the most part, with so much of the conventional chipset now part of the CPU architecture it now comes down to the component tolerances. The biggest killer for OC stability is Vdroop in general terms- some boards are afflicted, some aren't. From my own experience (I build OC, watercooled and TEC systems more often than not, or overclock and mod for people who don't have the expertise, resources or time for repetetive stability testing) overclocking for 24/7 operation isn't about wringing out every last CPU cycle, it's about getting the best core freq at the lowest stable Vcore. Most people with a 2600K are more than happy with 4.5-4.7G on a daily basis, which for SNB is eminently do-able at a safe voltage (~1.35 or less...maybe a tad more on some CPU's if you need hyper threading)...maybe a couple of customers that wanted the epeen of 5G or more

Anytime I've dealt with really high overclocks its not long before something goes wrong and it wont post and resets to safe mode or is no longer stable at said voltage/speed. It just doesnt seem worth it to squeeze that last 100mhz outta the processor.

High overclocks and instability go together hand-in-glove. Having said that, a thorough stability regime will lessen the chances of it happening quickly. So will preventative maintenance on the cooling system -including the systematic replacement of old TIM- and periodic stress testing. Anyone who thinks a substantial overclocked system is a "set-and-forget" proposition probably shouldn't be running one.

98% of the time you will see your cpu fail before you motherboard will, unless you get to cherry pick from many

Of the systems I've built, I'd say that around a quarter (into three figures) were NF4, 680/780i and 790U + Conroe/Kentsfield CPU. Virtually all would hold a stable OC better than what ever else was around at the time ( Asus Commando/Blitz/Rampage, Giga DS5/DQ6, various Abit, DFI flavours), and with the possible exception of a couple of X6800's I would say the board's capacitors, or tension in the northbridge heatsinks gave way before the CPU started to throw errors. Luck of the draw maybe, but I get called to RMA/replace boards much more frequently than CPU's - and most of the CPU's that die tend to be because of external problems (static, incorrect handling).

I've even done that but no matter what it seems like something always crops up over time with major oc's, I now prefer ultimate stability over a few hundred mhz. Its just not worth the time/energy.

Precisely the market the Sabertooth is aimed at

I cant count the nights I've been up all night tweaking/adjusting etc to get more out of the processor. In the end it just seems like luck if you get one of the mb/cpu combos that runs great at really high oc's

Such as it always was, probably as it always will be. There are no end of threads devoted to the overclock potential of certain Sandy Bridge FPO batchs. For every tired ex-overclocker there is probably one budding tweaker in the making- albeit with a better CPU arch to work with, but less manual overclocking voltage parameters to work with. Having said that, I don't think this thread is about finding the golden chip that can hit the 57 multi

One guy posted in another thread " I hope Bulldozer rocks!....so I can afford a i72600k!"
:haha:

That's actually a sentiment I've run into a number of times when discussing upcoming builds with prospective customers. They value the performance of Bulldozer as a means to make Intel lower it's prices, and not for what BD might bring to the table. A seemingly odd view, but on questioning these people it comes down to seeing AMD as the red-headed stepchild of the manufacturers family. I point out that an AMD chipset usually offers more connectivity than a compareably priced Intel board, but it seems that AMD's past woes ( read southbridge I/O performance), BD release passing into Duke Nukem territory, and at least locally -pricing of FX chipset boards, doesn't make an attractive proposition.
Add in that Intel supplies AMD with a superior GbLAN over the generic Realtek (and others) for their boards, and I think the general perception is that AMD are doing a P1ss poor job of playing follow the leader. A very hard attitude to shake.

 

[nfiniti9]

Of the systems I've built, I'd say that around a quarter (into three figures) were NF4, 680/780i and 790U + Conroe/Kentsfield CPU.

Ohh not dont get me started on the 680/780 boards. If I ever see a damn Evga/XFX board again I might just snap lol. So many failed. So many were completely different of the same model in potential. Honestly they are why I don't OC like I used to.

One note, my watercooled setups minus the maint on the wc part were pretty much set-and-forget. Watercooling is so nice for stability. Well if you have a decent setup. But you cannot build a wc system for someone, they have to be able to maintain it, so air is pretty much necessary for amateurs/novices.

 

[dividebyzero]

Ohh not dont get me started on the 680/780 boards. If I ever see a damn Evga/XFX board again I might just snap lol. So many failed.

That comes down to the chipset voltage more often than not. Striker Extreme's and it's gimped P5N32-E brethren are actually a worse board for failure rates. I never actually had much problem with the reference boards (EVGA/XFX/Biostar) since I tended to lift the heatpipe/heatsink array and replace the sh*tty thermal pads and screw down the heatpipe (as opposed to those stupid push-pins) -shades of the Crosshair Formula/Extreme.
Having any warpage in the PCB usually meant an early demise for these boards precisely because of the bad cooling ability of the stock setup.

One note, my watercooled setups minus the maint on the wc part were pretty much set-and-forget. Watercooling is so nice for stability. Well if you have a decent setup. But you cannot build a wc system for someone, they have to be able to maintain it, so air is pretty much necessary for amateurs/novices.

My experiences are somewhat different then. When I build a system (water/TEC) I usually set up a maintenance timetable, so every six months/year I (for a modest sum) will change thermal compound, flush the loop(s), blast dust out of rad fins, clean and check fans for wear, change/clean jet plates/block fins if necessary, check fittings/o-rings, refill etc. But then, I tend to get a lot of referrals so it makes sense to advertise a whole service...some murky-looking fishtank scene with tubing that looks like it was scavenged from a '65 Nova doesn't do much to attract new customers you understand.

 

[nfiniti9]

That comes down to the chipset voltage more often than not. Striker Extreme's and it's gimped P5N32-E brethren are actually a worse board for failure rates. I never actually had much problem with the reference boards (EVGA/XFX/Biostar) since I tended to lift the heatpipe/heatsink array and replace the sh*tty thermal pads and screw down the heatpipe (as opposed to those stupid push-pins) -shades of the Crosshair Formula/Extreme.
Having any warpage in the PCB usually meant an early demise for these boards precisely because of the bad cooling ability of the stock setup.

My experiences are somewhat different then. When I build a system (water/TEC) I usually set up a maintenance timetable, so every six months/year I (for a modest sum) will change thermal compound, flush the loop(s), blast dust out of rad fins, clean and check fans for wear, change/clean jet plates/block fins if necessary, check fittings/o-rings, refill etc. But then, I tend to get a lot of referrals so it makes sense to advertise a whole service...some murky-looking fishtank scene with tubing that looks like it was scavenged from a '65 Nova doesn't do much to attract new customers you understand.

"minus the maint on the wc part" i abbreviate too much. anyone who uses wc has to do maintenance on the setup regularly. i just meant the mb/settings etc.

 

[dividebyzero]

I just read a dreadful review on the new 'supercooler' Genesis...apparently not so good

That would be the bit-tech review I assume. Not the most flattering. Hardly surprising with the fan choice. Sharkoon Silent Eagle 1000 x 2...the website lists 64.8m³ flow rate - a quick conversion to cfm gives us 37.7. Not the most vigorous of airflows I would suspect...always pays to check the fine print.

I just rebuilt with a GA-990-FXA-UD7 in anticipation of the Bulldozer.

Talking of reviews, Kyle and Co. just threw this board under a bus.....Wouldn't be the Asus and MSI sponsorship of their upcoming "Texas GamExperience" jizzfest coming into play...would it?
(Personally I rate [H] only slightly above Tom's on the paid shill-o-meter...but that's me) :suspiciou

 

[red1776]

Talking of reviews, Kyle and Co. just threw this board under a bus.....Wouldn't be the Asus and MSI sponsorship of their upcoming "Texas GamExperience" jizzfest coming into play...would it?
(Personally I rate [H] only slightly above Tom's on the paid shill-o-meter...but that's me)

Wow ROFL,
that is the only review I have seen that was not a stellar review. funny how it runs perfectly with Corsair 1600+OC for me...and everyone else.
4.26Ghz on my 1100T, guess I hit the binning lottery...and MB as well.
That was not review, it was an all out frontal assault.
well, I will go with the 30 other reviews I have read. Geez, thats about the most obvious case of shill I have seen in quite some time.

 

[dividebyzero]

double post

 

[nfiniti9]

It's been pretty apparent for some time that Kyle and co are now up for hire.
Case in point; How many [H] graphics reviews this year don't feature overclocking numbers?
Answer: 1 card in 2 seperate reviews
So what happens when you review a card that specifically has been built for overclocking ?
MSI GTX 580 Lightning XE 3GB....sorry no OC included
MSI GTX 580 Lightning XE 3GB SLI....nope, not for you either

Considering bit-tech pushed their card to 980/5000, Hardware Canucks 1009/4380, G3D 961/4600, Tweaktown 980/5140, and TPU 898/4840 (stock voltage), with some predictably very good results (fps wise), all rather surprising that a site that fancies itself as hard can't be bothered OC'ing the King Kong of single GPU's.

Also note the kicking they gave the Gigabyte GTX 560 this week- citing no voltage control on the Gigabyte software, and going no further after finding out Riva Tuner wasn't updated for 275.33 drivers. Considering they aren't adverse to downloading and using Afterburner, it seems strange that they didn't try using the software with the Giga card...much the same as HT4U and TweakPC did with their review samples...maybe something else for the Too [H]ard Basket for Kyle.

I seem to remember that Kyle got the jump on the rest of the sites with his exclusive preview of the HD 6990. Five will getcha ten that there's a Bulldozer exclusive preview with Kyle's name on it at his upcoming GamExperience - book it.

Personally I will not use Gigabyte anymore. I got to see how many of their P35/P45 boards failed personally. Almost all the guys I work with had them during the C2D era and they are ALL dead but one.

 

[red1776]

Personally I will not use Gigabyte anymore. I got to see how many of their P35/P45 boards failed personally. Almost all the guys I work with had them during the C2D era and they are ALL dead but one.

So you have written off/sworn off the entire Gigabyte brand because of some problems with a chipset during the C2D era? ....interesting.
I have a different philosophy about tech/brands. The time from the C2D era to now in technology is about the equivalent to the Ford Pinto to the current lineup of vehicles. Brands like Asrock, Biostar, and now even ECS, once producing boards of questionable to terrible quality products, are producing very good, or vastly improving quality boards. PSU's are a good example of this as well. what is great one month, may not be so good the next as they change the primary build of their products. I think you are missing out on a lot of potential options by swearing off a brand for a bad experience years ago. But thats just me.

 

[nfiniti9]

So you have written off/sworn off the entire Gigabyte brand because of some problems with a chipset during the C2D era? ....interesting.
I have a different philosophy about tech/brands. The time from the C2D era to now in technology is about the equivalent to the Ford Pinto to the current lineup of vehicles. Brands like Asrock, Biostar, and now even ECS, once producing boards of questionable to terrible quality products, are producing very good, or vastly improving quality boards. PSU's are a good example of this as well. what is great one month, may not be so good the next as they change the primary build of their products. I think you are missing out on a lot of potential options by swearing off a brand for a bad experience years ago. But thats just me.

Could be... But the only power I have as a consumer against a big corporation is through my wallet.

Ok to be fair I shouldn't make a statement like that without all the facts.

First a lot of boards failed within 1.5 years - bad
second their customer support was terrible at that time
third the boards that were rma'd took 3-4 weeks to get fixed and sent back.
fourth they sent a board back that wouldnt run any ram in dual channel (we tried 5 different brands and used slower ram just to test, all failed any memory test or prime etc) without any testing it appears. it posted so they sent it back.
fifth they release multiple bios updates that cause more problems than they fixed.
sixth they made no attempt to explain or admit those boards even had a problem.
seventh they refused rma on a board that had changed the northbridge heatsink (its their policy and they can enforce it but asus doesnt as long as there is no physical damage)

so Its not just that I had a bad board or two. its much more than that.

 

[dividebyzero]

It's generally accepted fact ( sales numbers, feature set, customer reviews, sketchy RMA return figures) that Gigabyte went from being a distant second to Asus, to parity at the very least, on the strength of their P45/X48 boards. With the noteable exception of the RoG boards, Asus' P5Q and P5E series boards are basket cases*. The quality control was abysmal, and the attention to detail slipped noticeably in the 2008-10 timeframe.
I might be somewhat biased to a degree in that both of my rigs are built on Gigabyte boards (both rock solid), as well as well over half of my customer builds. That is starting to swing back to Asus and more increasingly ASRock since they got out from under Pegatron's sphere of influence. Pricing for the given feature set being the prime motivator here.

Here's Hardware France updated article (Apl 2011) on return rates to France's version of Newegg, which would tend to support the fact that failures probably come down to either individual models or SKU's.
As for customer service, I guess my country is blessed with a Gigabyte distributer that cross-ships RMA replacements- and has done since at least 2006. That said, Asus and MSI replacements are usually expedited delivery also (~3 days from RMA notification) while the enthusiast offerings I've had to RMA from Abit and DFI took closer to 4-6 weeks (back when they were still viable entities).
As the saying goes: Your mileage may vary

*RAM compatibility issues, DIMM slot failures, PCI-E slot failures, BIOS/board that wont hold VTT, PLL, MCH and ICH voltage, misaligned heatsinks, missing pushpins, warped boards and the ever present loose capacitors rolling around in the packaging.

 

[red1776]

RMA stands for what now?

It's generally accepted fact ( sales numbers, feature set, customer reviews, sketchy RMA return figures)

about a year ago I got into a great conversation with a rep at Newegg that said to me that 93% of the components RMA'd were components that were functioning perfectly well.
-buying incompatible parts
-not looking at QVL's
-taking the PSU out of the box, plugging it into the wall, and returning it because it didn't start up (thats my favorite):haha:
-various other not knowing what they are doing , or reading instructions
- buyers remorse

...etc etc

add a few more percent for damaged in shipment and you start to get the idea where a lot of these "Brand X Sucks man!" comments come from.

 

[nfiniti9]

It's generally accepted fact ( sales numbers, feature set, customer reviews, sketchy RMA return figures) that Gigabyte went from being a distant second to Asus, to parity at the very least, on the strength of their P45/X48 boards. With the noteable exception of the RoG boards, Asus' P5Q and P5E series boards are basket cases*. The quality control was abysmal, and the attention to detail slipped noticeably in the 2008-10 timeframe.
I might be somewhat biased to a degree in that both of my rigs are built on Gigabyte boards (both rock solid), as well as well over half of my customer builds. That is starting to swing back to Asus and more increasingly ASRock since they got out from under Pegatron's sphere of influence. Pricing for the given feature set being the prime motivator here.

Here's Hardware France updated article (Apl 2011) on return rates to France's version of Newegg, which would tend to support the fact that failures probably come down to either individual models or SKU's.
As for customer service, I guess my country is blessed with a Gigabyte distributer that cross-ships RMA replacements- and has done since at least 2006. That said, Asus and MSI replacements are usually expedited delivery also (~3 days from RMA notification) while the enthusiast offerings I've had to RMA from Abit and DFI took closer to 4-6 weeks (back when they were still viable entities).
As the saying goes: Your mileage may vary

*RAM compatibility issues, DIMM slot failures, PCI-E slot failures, BIOS/board that wont hold VTT, PLL, MCH and ICH voltage, misaligned heatsinks, missing pushpins, warped boards and the ever present loose capacitors rolling around in the packaging.

Well we switched to Gigabyte primarily (for business and EVGA/XFX for gaming) when C2D came out so I didn't get to see many of those boards except for my friends. I did hear of a few problems but nothing like you mentioned. Mainly the bios, voltages and the heatsinks (paste didnt even touch chip). I can't say I've ever heard of a capacitor on an Asus board loose in the box. I have seen a Biostar like that.

Oh and Abit and DFI are the other two I would never buy again. But they had very serious issues and some of their boards just never worked properly. We see how they ended up. I still use Gigabyte everyday (do a lot for universities and government) and honestly I do not see many failures anymore. The H55 and P55 I think I've only seen 1 or 2. Just their customer service got me so frustrated, they just acted like they didn't care. When its my personal board I'm not that worried, when its a customers and its been 3.5 weeks and you cant even get a response from anyone it will drive you nuts. I'm no Giga-hater I just think at some point you have to speak up with your wallet. Everyone has difference experiences and you are across the "pond" (haha love that phrase) so who knows if it was a batch issue etc.

 

[nfiniti9]

about a year ago I got into a great conversation with a rep at Newegg that said to me that 93% of the components RMA'd were components that were functioning perfectly well.
-buying incompatible parts
-not looking at QVL's
-taking the PSU out of the box, plugging it into the wall, and returning it because it didn't start up (thats my favorite):haha:
-various other not knowing what they are doing , or reading instructions
- buyers remorse

...etc etc

add a few more percent for damaged in shipment and you start to get the idea where a lot of these "Brand X Sucks man!" comments come from.

I don't use generalizations for 1-2 experiences, I've built many many machines so I try to hold my opinions until I get enough experience to justify my opinion.

 

[red1776]

I don't use generalizations for 1-2 experiences, I've built many many machines so I try to hold my opinions until I get enough experience to justify my opinion.

okaaayyy, thats a good thing to do.....I have no idea what it has to do with my post however. I was relaying what I thought was an amusing conversation with a rep from newegg.

That wasn't aimed at you. Hang around a while and watch the visceral fanboy responses to stories on new product releases, and Op's asking things like "what Graphics card should I get" and such....you will know what I meant.

 

[nfiniti9]

okaaayyy, thats a good thing to do.....I have no idea what it has to do with my post however. I was relaying what I thought was an amusing conversation with a rep from newegg.

That wasn't aimed at you. Hang around a while and watch the visceral fanboy responses to stories on new product releases, and Op's asking things like "what Graphics card should I get" and such....you will know what I meant.

another case of me not taking the time to read it properly ) read it on my phone responded on the computer. dont think ill ever learn that one. i try to do too much at once. anyway its definitely true. when i test stuff, especially network equipment, that is supposed to be not working it almost always is. the one exception is usually ram. sometimes its compatibility but its usually bad.