Is AMD X2 Reliable for Business Use?

[Porkster]

"Is AMD X2 Reliable for Business Use?"

With Tom’s live CPU testing of the AMD X2 and Intel 840EE, I noticed that the X2 fan is running at extreme speeds. I personally feel 6000 RPM is far too high for reliable use and probably noise creation.

Now we all can agree that if a CPU fan dies, that all systems these days will close down and risk causing data damage and hard-drive failure, but is there an extra risk of data damage on the AMD X2 series? The extreme rotation speed of the AMD X2 CPU's is bound to wear and burn out the fan's coil and bearing much faster, resulting in high risk.

It could be argued that if a fan slowly weakens that a CPU could be strained without clear signs from the system bios monitoring of a failure.

Tom, it would be interesting to see what the AMD X2 does when a slower or decreased fan speed has on the temperature of the CPU.

Also what is the wattage of the AMD X2 fan? If the fan is like over 20watts then that value should be noted when comparing power consumption to Intel. AMD seems to be using older chemistry and mineral bases for their silicon so using extreme cooling as an off-the-shelf standard to get a dual core product, to me seems improper when there is a risk of the fan making business use a question.

.

 

[pat]

Ok, then how reliable is a system that could go as high as 71C for cpu temp in a business environnement..

Maybe Intel sould have make their fan going faster!!!

Wow.. you must really be an Intel fanboy trying to find arguemnt to bash AMD...

As a fanboy post, your's beat em all!!

Hey, do you really think that either AMD or Intel would have put a product on the market if they wouldnt be reliable? Do you really thing that either AMD or Intel would offer a 3 years waranty on their CPU/fan?

AMD dont use older chemistry. Right now, their silicon is much more eficient than Intel one and the reason thay use a faster fan is because the cooler and fan are smaller than the Intel one.

So yes, I would trust either AMD or Intel for business use, leaning for AMD because they put less heat in the case and, if you add the heat from the HDD, the chipset and the video card, make up for an easier to control environnement and a reliable environnement too.




<font color=red>Sig space for rent. make your offer.</font color=red>

 

[Schmide]

The fan is just an integrated piece of the cooling solution. AMD just built a solution that may run at a faster RPM but most likely has a less aggressive attack on its blades. In the end, it is most likely a similar cooling solution and has an equal runtime rating. A HSF is a total package and is generally rated by C/W (celsius per watt). You can look at the <A HREF="http://www.overclockers.com/articles373/socketA.asp" target="_new">heatsink roundup</A> to get a gage of different fan speeds and cooling performance. As for power, fans of that size and power use a minor less than 5 watt and the difference between fans should be less than a watt. The rotation speed has little to do with the amount of power used, more likely the CFM (cubic feet per minuet) is usually directly correlated to the power used.

In the end the Intel processor uses much more power than the AMD processor.

See <A HREF="http://www.techreport.com/reviews/2005q2/athlon64-x2/index.x?pg=15" target="_new">http://www.techreport.com/reviews/2005q2/athlon64-x2/index.x?pg=15</A>

Dichromatic for your viewing plesure...

 

[edurm]

"Is AMD X2 Reliable for Business Use?"

Yes, it is 100% reliable

Much more reliable than a INTEL PRES<b>HOT</b> Core


Plz, check out this link: http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2249&p=13

It shows power consumation when CPU is Full Load

 

[Starfishy]

<A HREF="http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2249&p=13" target="_new">http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2249&p=13</A> - I can't stand them not being clicky

 

[Porkster]

"Plz, check out this link"

You are missing the point. Intel CPU's maybe using more power but the fan required to cool the CPU is under less strain than the AMD X2, making Intel a lower risk in the Business Environment.

If the AMD X2 fan is anything like the older models then it would be using over 12 watts. Add that to the AMD CPU wattage and any extra fans need to balance the case airflow and there you have a problem.

I'm asking Tom to report on the fan's quality and how the AMD X2 will cope in a fan failure or quality loss.

I personally believe the AMD X2 is a risk for business use and critical data management.

.

 

[Starfishy]

You are missing the point. Intel CPU's maybe using more power but the fan required to cool the CPU is under less strain than the AMD X2, making Intel a lower risk in the Business Environment.

You are missing the point. If you were really using the money to invest in one of these systems you are not going to use stock heating. If you knew better you would upgrade the cooling. Thus it means that the power consumption of the fan is not an issue (not that it would be significant anyway). Furthermore, it makes the temperature measurement a little arbitrary as well, nobody is going to cool these with the stock cooling. However, it does give a semi-good measure of heat production using the assumption that both of the stock cooling solutions are approximately equal in function. This might be an incorrect assumption but I'm not sure if it can be avoided.

 

[Porkster]

Starfishy Quote : "You are missing the point. If you were really using the money to invest in one of these systems you are not going to use stock heating."

Sorry, but you are overlooking the fact that AMD is already incorporating a fan running at crazy speeds of 6000RPM. There aren't too many fans that go over that speed that are anything that are acceptable for noise and efficiency. Also since the two CPU’s that are being reviewed are premium range and new releases, it’s unlikely non proprietary fans have not been tested with these CPU’s. Passive cooling solutions may also be non existent.

Why are people so blinded regarding AMD. This fan issue is clearly a risk for those in professional business. I would sack IT staff if they purchased these AMD systems for critical type applications.

.

 

[fishmahn]

It seems to me you are making 1 very dangerous assumption.

"Both Intel and AMD use the same quality fan."

I wouldn't make that assumption if I were you. A fan that is made to spin at 6k can be just as reliable as a fan that is made to spin at 3k. In fact, I would (in my non-technical way) be more inclined to trust the fan rated at a higher rpm because *of necessity* higher quality parts are used. I can see the other side however, because motorcycle engines (high rpm) don't last as long as car engines (low rpm)... but that's another bad assumption, and it's making a correlation where no correlation exists.

Another point...

Both AMD and Intel will shut down the CPU on thermal overload. So, in the case of a thermal shutdown, both CPUs have the same potential for data loss and corruption. Yes, Intel will throttle before it shuts down, but in the case of a catastrophic heatsink failure, even a throttled P4 will overheat and likely have to shut down in a matter of seconds, so the throttling part is pretty useless if you ask me.

I think the best way for you to determine the validity of your claim is to have an electrical and mechanical engineer evaluate the fans of Intel and AMD processors and have him/her/it/other tell you which is more likely to fail over time.

I have no problem using AMD in a business environment. The company I currently work for has a thing for Dell, so I buy what they sell (bread-box sized heatsink and all), but in prior jobs, and at home, I've used both AMD and Intel, all of which run 24/7/365, and one of them the only fan in the machine for over 2 years was the CPU fan (and it was an Athlon Thunderbird - nice and hot...) as the PSU fan locked up (cheap PSU). That system is still going fine, and it's 4+ yrs old now.

Mike.

 

[Schmide]

Dude the HSF is a total knockoff of a <A HREF="http://www.thermaltake.com/coolers/venus/rs/a1770.htm" target="_new">http://www.thermaltake.com/coolers/venus/rs/a1770.htm</A>
The fan on that model runs at 3500 to 6000 rpm and uses 1 to 5 watts. You're making a mountain out of a molehill.

Dichromatic for your viewing plesure...

 

[RichPLS]

___________________________________________________________________
Hey, do you really think that either AMD or Intel would have put a product on the market if they wouldnt be reliable? Do you really thing that either AMD or Intel would offer a 3 years waranty on their CPU/fan?
___________________________________________________________________

LOL Pat remember the 1.13GH P3 flop! LOL!

<pre><font color=red>°¤o,¸¸¸,o¤°`°¤o \\// o¤°`°¤o,¸¸¸,o¤°
And the sign says "You got to have a membership card to get inside" Huh
So I got me a pen and paper And I made up my own little sign</pre><p></font color=red>

 

[apesoccer]

Ok

Intel:
avg rpm 3590
avg temp 67c

AMD:
avg rpm 5500
avg temp 60.5c

A 7 degree difference in pretty directly proportional to the speed difference in the fans on the two chips. I'm fairly certain that they run at about the same temperature because of how close they are now (and i have a multispeed fan on my cpu at home that runs between 2500-7000 rpm...the difference in temps isn't as great as you would think for around 2000+- rpm). I think they both need better cooling in my opinion, but i don't know what they're rated to either. In any event, to be sure, we would definately need to decrease the speed in the fan on the amd processor and increase the speed on the intel processor and see what temps they run at those speeds.

So the question isn't whether it does a better job of cooling, it's a question of whether running at higher rpms is going to kill the fan faster. Typically a fan running at higher rpms has a better chance of dieing then a lower rpm one, assuming similar quality control. So i'm agreeing with you that a faster fan is more likely to go dead then a slower one.

That said, the amd processors are doing 15-30% more work then the intel processors.

If i were a company who is rated in the billions for revenue, i would probably limit my risks as much as possible, ie i would put higher quality control on my fan. So i imagine that they have already thought of the fact that they're going to be attacked for having a higher rpm fan and choosen one that is both quieter and of a better quality then most. Obviously in stating that, it's just a thought, not a fact.

However, Starfishy, you kinda have missed the point in your most current response =(. In a buisness solution (say HP for instance, since they've agreed to put amd processors in their computers), if HP decides to buy 1 million processors with fans, they aren't going to replace the hsf, and they're not going to expect the people who buy their product to either. So your last statement is irrelevant. We're used to responding to people who want an individual computer, rather then a buisness, i'm sure you see what he's saying now. In a buisness environment, you don't go replacing parts as soon as you get a product, that could end up voiding your warrenty, not to mention adding a cost to your end cost.

I don't consider the fan running 5500rpm to be extreme, just my opinion. I imagine it creates extra noise, but how much is the key. It certainly makes sense that there would be extra noise though, otherwise they would have gone with a slower fan. So on this, i'm going to just wait to see how loud they really are and save my argument for later.

Hrm, a cpu fan dieing has never caused a hard drive to fail (corruption from the cpu crashing because [edit: the cpu] got too hot YES, but that doesn't cause physical damage to a hd). If your computer has been properly set up, as it should be in a Buisness environment, the computer will shut down (via the bios settings) if the cpu fan dies, and will not boot until it's been replaced (see new Dell computers). That said, if a fan isn't designed to outlast the useful life of the computer (3-5 years in a buisness/organisation), then it isn't designed for the right computer in my opinion.

One last thing, to address the comment on overall extra heat build up. If intel's chip is hotter by 7c then that extra 7c is added to the inside of the case faster then the amd processor, which means that, because the intel chip doesn't have as adeqate cooling as the amd chip does, the fans in the intel setup are going to have to pick up the slack, not the other way around.

F@H:
AMD: [64 3000+][2500+][2000+ down][1.3x2][366]
Intel: [X 3.0x3][P4 3.0x2][P4 2.4x5 down][P4 1.4]

"...and i'm not gay" RX8 -Greatest Quote of ALL Time<P ID="edit"><FONT SIZE=-1><EM>Edited by apesoccer on 06/03/05 04:53 PM.</EM></FONT></P>

 

[Schmide]

it would be using over 12 watts

One of the fastest most powerful 84.1 cfm fan runs at approxitmately 9w.

Sunon 80 x 38mm fan, model #PMD1208PMB1-A rated at 84.1 cfm @ 5700 RPM, 55.2 dBA, 9.1w

Dichromatic for your viewing plesure...

 

[jim552]

There was a time that I would have the same concern as yourself over the fan issue. Much of what you bring up I used to believe as "true" as well.

It made sense then, and even now, it seems to make sense as well.

Now though, this would no longer be a concern.

We are an AMD shop for ALL desktops. We have been heading this direction, since the K-5 introductions.

The K-5 processors were a concern because their thermal protection was, I believe, at first non-existent and then later not that good.

There was much concern over how "fan issues" would become problems. This was an extra-strong concern for the PC's that were out of state.

We did loose a couple of K-5 CPU's do to heat stress, but nothing that was of any real concern.

When the Athlon's came out the need for better cooling was a concern with them as well, but use has shown that this isn't a real problem.

I do have some fans that are running at 6000 RPM already. Mostly I was curious about extra wear-and-tear, dust collection, and what afffect it had on existing CPU's. (Basically, I have been assuming fans may need to run at higher speeds with the upcoming greater power dissipation.)

These "test fans" that I have actually running on desktops have had no problems for a couple of years.

I, personally, believe that your concerns are ones that I would have agreed with in the past, but now they are not concerns of mine.

Actual use and deployment may bring about a different opinion but I am not expecting that.

Mostly dual cores will be used on Servers and higher end systems. Our Servers are all in nice little cool-offed rooms, as well as having extra fans in the Server cases.

Also I should note that initially we tended to purchase a mix of about 50-50 AMD stock fans and third party fans of arguably better quality.

Generally, the third party fans kept things a bit cooler, but we haven't noticed any increase in fan life, or decrease in CPU life.



edited: Corrected a couple of issues. ("show" to "shown", "not" to "note".)<P ID="edit"><FONT SIZE=-1><EM>Edited by jim552 on 06/03/05 05:20 PM.</EM></FONT></P>

 

[Schmide]

But is nothing compared to this.

<A HREF="http://www.subzeropcs.com/delta_tfb1212ghe_extream_high_speed_220_cfm_fan_120mm_x_38mm.html" target="_new">http://www.subzeropcs.com/delta_tfb1212ghe_extream_high_speed_220_cfm_fan_120mm_x_38mm.html</A>

That might hover with the proper voltage.

Dichromatic for your viewing plesure...

 

[RichPLS]

At 65dBa, would not take many to sound like at an airport!

<pre><font color=red>°¤o,¸¸¸,o¤°`°¤o \\// o¤°`°¤o,¸¸¸,o¤°
And the sign says "You got to have a membership card to get inside" Huh
So I got me a pen and paper And I made up my own little sign</pre><p></font color=red>

 

[RichPLS]

I understood Prescot stock HSF's were fine quaulity if not overclocking.

<pre><font color=red>°¤o,¸¸¸,o¤°`°¤o \\// o¤°`°¤o,¸¸¸,o¤°
And the sign says "You got to have a membership card to get inside" Huh
So I got me a pen and paper And I made up my own little sign</pre><p></font color=red>

 

[Kelledin]

Now we all can agree that if a CPU fan dies, that all systems these days will close down and risk causing data damage and hard-drive failure, but is there an extra risk of data damage on the AMD X2 series?

An *extra* risk? Not really.

The extreme rotation speed of the AMD X2 CPU's is bound to wear and burn out the fan's coil and bearing much faster, resulting in high risk.

Apparently 6000 RPM is what that fan is designed for. Compare by MTBF.

Also, one can generally expect that a fan (and any part prone to wear) has a probability of failure related to the operating lifespan in a "bathtub curve" shape. The probability of failure starts out high (because of the chance of manufacturing defects, improper initial setup, etc), then quickly drops after the the first few days or weeks of operation (the burn-in period). Eventually the probability will level out and start to gradually increase as the part reaches the point where normal wear starts taking its toll.

It could be argued that if a fan slowly weakens that a CPU could be strained without clear signs from the system bios monitoring of a failure.

Doubtful. Temperature monitoring generally handles slow overheats better than near-instantaneous overheats. To the system BIOS or monitoring hardware, overheat temperature is overheat temperature, no matter how long it took the CPU to reach that temperature. Sudden temperature spikes are far more dangerous; they cause more stress and are more likely to not get handled in time.

Also what is the wattage of the AMD X2 fan? If the fan is like over 20watts then that value should be noted when comparing power consumption to Intel.

It's rare for a CPU fan to be over 10W. A few Delta fans sometimes draw more than that--enough to overload some motherboard fan headers--but they're a very extreme case. And they seem to last about as well as other fans; some of my 1U systems are equipped with top-end Delta blowers.

AMD seems to be using older chemistry and mineral bases for their silicon so using extreme cooling as an off-the-shelf standard to get a dual core product, to me seems improper when there is a risk of the fan making business use a question.

The need for extreme cooling is a direct result of the CPU drawing more power and having less die surface area through which to dissipate the heat. In both aspects, Prescott-based CPUs have the greater need for extreme cooling.

Temperature tolerances might make a difference, but both AMD and Intel CPUs are specced for about the same temperature limits: 65-72 Celcius.

Obviously Intel is content to let thermals ride closer to borderline than AMD. That will likely cause the CPU to not last as long, but it's up to you whether to like it or lump it. FWIW consumer CPUs are typically not designed to last forever, only to outlive their usefulness.

<i>"Intel's ICH6R SouthBridge, now featuring RAID -1"

"RAID-minus-one?"

"Yeah. You have two hard drives, neither of which can actually boot."</i><P ID="edit"><FONT SIZE=-1><EM>Edited by kelledin on 06/03/05 08:56 PM.</EM></FONT></P>

 

[endyen]

Boy, are you ever smart. I guess I better not by one of those 7200rpm hdds. I guess those raptors that run at 10k must only last a week. Thanks so much for your help.

 

[RichPLS]

You forgot those disposable 15,000 rpm drives, you get a five year warranty, but probally spend all your time RMA'ing for replacements.

<pre><font color=red>°¤o,¸¸¸,o¤°`°¤o \\// o¤°`°¤o,¸¸¸,o¤°
And the sign says "You got to have a membership card to get inside" Huh
So I got me a pen and paper And I made up my own little sign</pre><p></font color=red>

 

[RichPLS]

Did you not get the sarcasm intended memo! Cheeze Wiz
If you only get one drive if 5 years, you get screwed, I get at least 12 new drives under warranty cause they spin so fast at 15k till I overclock them. I have some up to 30,000RPM before the platters themselves break apart.

<pre><font color=red>°¤o,¸¸¸,o¤°`°¤o \\// o¤°`°¤o,¸¸¸,o¤°
And the sign says "You got to have a membership card to get inside" Huh
So I got me a pen and paper And I made up my own little sign</pre><p></font color=red>

 

[RichPLS]

ohhh the ultra-DeathStar...Make them pay Darth, make them pay for thier inadequateness!

<pre><font color=red>°¤o,¸¸¸,o¤°`°¤o \\// o¤°`°¤o,¸¸¸,o¤°
And the sign says "You got to have a membership card to get inside" Huh
So I got me a pen and paper And I made up my own little sign</pre><p></font color=red>

 

[RichPLS]

Practice, you need. Training proper, spin disk plenty. Use the force, the dark side.

<pre><font color=red>°¤o,¸¸¸,o¤°`°¤o \\// o¤°`°¤o,¸¸¸,o¤°
And the sign says "You got to have a membership card to get inside" Huh
So I got me a pen and paper And I made up my own little sign</pre><p></font color=red>

 

[function9]

What kind of business? I don't see the X2 or Intel's DC chip being widely and quickly adopted by corporations/businesses. Last time I checked, it doesn't take 2 cores to run Word and Outlook simultaneously.

 

[endyen]

You are missing the point. Intel CPU's maybe using more power but the fan required to cool the CPU is under less strain than the AMD X2, making Intel a lower risk in the Business Environment.

Let's be serious for a moment here. The Intel chip has a 20%+ higher heat output than the Amd chip. For the intel fan to keep that chip as cool? as it does, it must be moving considerably more air than the Amd fan. Since it is moving at a lower speed, it must have more torque.
If you did not know, it is the stress on the "bearing" that generally causes failure. Since the high torque fan exherts a good deal more fore on the "bearing" it is far more likely to fail.
This holds true through most motors. While high torque motors are far more expensive than high speed motors, they tend to fail more quickly than a high speed motor.
If anything, the Intel fan is under a good deal more stress than it's Amd counterpart.