Tom (Toms Hardware) Caught Lying For Sugar Daddy (Intel)

[firtol88]

Van's Hardware has a Video which tells a different story about AMD chips and how they handle a lack of cooling

Here's a quote

"A video has appeared on the Internet countering a huge dose of FUD my former employer inexplicably dumped on his readership a couple of months back. The video, with AMD credits all over it, is entitled "How an Athlon(tm) MP 1.2GHz Really Copes with Heat Emergencies." The piece demonstrates the AMD Palomino Athlon subjected to brutal circumstances such as heat sink removal while playing Quake III and boot up attempts when a CPU cooler is not attached. In all scenarios, the Palomino comes out unscathed. A similar though much less thorough test came out with unsurprisingly different results at Tom's Hardware. Ouch! Looks like dispensing bad medicine can result in a mouthful of looser teeth. Good job Ben & Joe, perhaps you can also give THG a crash course in analyzing computer technology."

Here's a Link

Just click on the articles title

 

[erub]

unlikely -- wouldn't he have a reason to be bitter about his former employer? And he's a known AMD zealot.

 

[erub]

woohoo go time change

 

[AGodspeed]

 

[Sugadaddy]

Man, I thought you were talking about me...

 

[Mikewarrior2]

And AMD's "video" is worthless since the mb they use isn't available on the common market... it proves one thing... AMD thermal protection only works with a corresponding MB that helps it along... without a mb to help it, the itnernal diode/thermal protection is pretty much ineffective.


Mike

 

[Rand]

AMD's video is worth very little as the mobo used isn't even available on the common market... and incorporates direct support for the on-die thermal diode courtesy of it's own temperature measurement chip.
So long as the XP needs mobo support for thermal protection, and motherboards do not begin supporting that AMD's video is entirely unrepresentable of real world results.

Hence it does absolutely nothing to discredit Tom Pabst's reporting at all.
Tom's own video would have been much more realistic had he done a test in which he simulated a CPU fan failure by unplugging it, rather then completely removing the HSF unit.

Also, it stands to readon that Tom would lie for AMD if anything as he does tend to be somewhat favourable to AMD.
That said, I do not consider Tom biased and think quite highly of him.
He's just extremely vocal about his opinions, and never forgets anything he considers as 'wrong/bad'.

 

[Beowulf]

And the article comes from Van Smith's site and we know he loves to bash others.

 

[highwire]

I tried to understand what this thread was about - no ref at all at the top. This is wierd. The original post(firtol88) of this thread does not appear. That makes it a little hard to sort out. ???

Anyway, Rand has got it about right, I believe.

This is my comment from thread "just fried athlon xp 1800":


<< Taylorm, thanks for reminding us about Murphy's law, and my sympathies.

This fan failure brings up a a closely related issue.

That none of the AMD boards, even the newest, are not able to use the INTERNAL diode on the athlon xp yet, is amazingly dumb.

An external board sensor could cope with a fan failure, but has NO chance of responding to a heatsink failure. Any capable engineer knows what will happens if you hit .5 grams of thermally isolated silicon with 50 watts. Tom's pics were for the trade.

When this is sorted out, I'll feel better about getting a high end AMD chip. >>


I am an AMD fan, but I think that "AMD" video is BS.

 

[ST4RCUTTER]

Van Smith is full of himself, but to be honest, I don't see the big deal with the whole "thermal diode requiring mobo support" issue. I set my BIOS to shut off if the CPU gets to 55C and the problem is taken care of. It takes over a minute to go from idle to a damaged CPU should the HSF fail which is plenty of time for the thermistor to trigger the system shutdown. The Athlons are good well into the 60's. Tom's article would have been much more valuable if he had tested it this way as well. Lets be honest...what are the percentages of HS falling off when compared to HSF failure?

 

[Pabster]

That's just it. A little common sense goes a long ways.

The odds of your heatsink assembly "falling off" -- whilst fragging away in Q3A nonetheless -- are less than you getting struck by lightning three times in your life. (well, that's just a guess ) Really, it just isn't a concern. Much, much more realistic is the fan failure aspect. Fans do go bad -- and with the BIOS thermal monitor and/or software shutdown solutions available, you are easily protected against this (much more) possible occurence.

Let's remember that the internal thermal diode for AMD processors is extremely new. Intel didn't have 100% support on all boards from all manufacturers for quite awhile, either.

 

[Mats]

Pabster rules.

 

[Wingznut]

I'm so tired of the "heatsink falling off" crap. The video at Tom's is not about the real world possibilities of a heatsink falling off. It's not an "AMD sucks" video, it's not an "Intel rules" video.

Removing the heatsink is a very quick and extreme way of showing the P4's Thermal Management capabilities. Which personally, I think is very impressive.

 

[JellyBaby]

The old slot 1 cartridges were nice in that you could add a dual-fan HS. If one fan failed the other picked up the load.

I can't think of how a HS could fall off unless UPS or Fedex took your equipment "down town" and you didn't inspect it before booting.

So when will there be AMD motherboard fully supporting the chip's internal thermal checkers? In time for the .13u cpus I would hope.

 

[AGodspeed]

I don't see why people are calling this video crap. Obviously, all AMD needs is a motherboard that supports it's diode, and you're processor won't fry when the one in a million possibility of your heatsink falling off actually happens.

So we'll just have to wait and see when the Athlon XP's get motherboard support, like the dual Tyan boards have.

 

[Pabster]

AGodSpeed wrote:

"I don't see why people are calling this video crap. Obviously, all AMD needs is a motherboard that supports it's diode, and you're processor won't fry when the one in a million possibility of your heatsink falling off actually happens."

Opinions, of course. I thought Tom's test was a bit unrealistic. The heatsink doesn't just come off. Had he instead tested by killing the fan on the respective machines, it would have been a more realistic presentation. I can't help but think that a certain marketing department was behind it. Certainly it showed off the thermal protection of the P4 -- there isn't any arguement about that.

Hopefully, what will come out of the whole debacle is a new stream of mainboards which properly and fully support (and more importantly, utilize) the thermal diode and clock throttling capabilities of Palomino (and future core) processors. But I disagree with those who attack AMD for the current boards -not- supporting them -- the internal thermal management features of the Palomino core are relatively new -- in fact, discounting the MPs which appeared a few months back, you could say they are less than a month old. Intel did not have full support for their diode readings and the like initially either -- it took a while, as will full support for AMD's features.

Now who wants to argue the likelihood of a heatsink falling off? Or tell me "that's not the point"? Tom's test demonstrated the rigs running with the heatsink removed. Exactly when does that happen in the real world?

 

[AGodspeed]

Now who wants to argue the likelihood of a heatsink falling off? Or tell me "that's not the point"? Tom's test demonstrated the rigs running with the heatsink removed. Exactly when does that happen in the real world?

It doesn't happen, but the point is, it could happen. And when it does happen, the Pentium 4 shines very brightly, whereas the Palomino goes the way of the deceased.

Btw Pabster, do you know for sure if your dualie Tyan board actually does support the Palomino's internal diode. I mean, I'm not asking you to test it without the heatsink , but I'm just wondering.

 

[Macro2]

Intel loves Toms video. They have been using in in their corporate presentations since it comes from an "independent" reviewer..(.lol to the last part)

 

[SuperSix]

AMD - Great ideas, horrible half-ass implementation.

Intel - Great marketing, excellent product follow through.


Nowhere does Tom indicate that anyone's HSF unit will simply "fall off". The whole purpose of the video is to show that in the unlikely event of ANY cooling failure, Intel's thermal protection desing is superior. Simple.

...And it is.

 

[AGodspeed]

AMD - Great ideas, horrible half-ass implementation.

Intel - Great marketing, excellent product follow through.

What?!?! So the Athlon was a great idea, but I guess they didn't implement it correctly, right?

I think I know what you're trying to say, and that is that AMD's thermal diode is useless, and so that should be considered "horrible half-ass implementation", as you put it. But no, that's also a misconception.

The Pentium 4 has an on-die thermal "insurance" feature. It does not need a specific motherboard for the throttling function to work.

The Athlon XP has NO on-die thermal "insurance" feature. It seems, the Athlon XP's thermal diode will only work in conjunction with a specific motherboard that will support the thermal diode. Once a motherboard truely supports the Athlon XP's thermal diode, it will perform just the same as the Pentium 4 did in Tom's test when the heatsink was removed (that?s looking at the video from Tom and the one presented in this thread).

End of story, hopefully for a while.

 

[SuperSix]

<< AMD - Great ideas, horrible half-ass implementation.

Intel - Great marketing, excellent product follow through.

What?!?! So the Athlon was a great idea, but I guess they didn't implement it correctly, right? >>



What I mean is that AMD has the best of intentions, but not the infrastrusture to implement their ideas correctly.

Finally putting a thermal diode on? Great! Needs a specially equipped motherboard for it to work? Bad.

Improving on a great desing? (Athlon -> Palomino) Great! Don't add a heat spreader to reduce the fragility of the core. Bad.

Making a great chipset to help the stability of DDR based Athlon solutions? Great! Stopping production because you say that the DDR chipset offerings are fine when it is really because you dropped your ASP's (Average selling prices) so low you can't afford to keep the fab open? Bad.

I could go on, but it's these missteps (and many more) that will keep a superior CPU core design from becoming the dominant CPU. Kind of like football, it's about execution...

And thanks for not calling me an Intel zealot.

 

[Wingznut]

"I thought Tom's test was a bit unrealistic. The heatsink doesn't just come off. "

*Wonders if anyone even bothers to read his posts.* Sigh.

 

[Pabster]

SuperSix wrote:

"What I mean is that AMD has the best of intentions, but not the infrastrusture to implement their ideas correctly."

That's a pretty open-ended statement. Why don't you provide some evidence of this, or is that just an opinion?

"Finally putting a thermal diode on? Great! Needs a specially equipped motherboard for it to work? Bad."

Well, I suspect Intel's thermal solutions wouldn't work properly if the boards didn't, either. Fortunately for Intel, they all do. But it wasn't always that way. Again, it's pretty ridiculous to attack AMD for not having motherboard support when these processors were just released less than a month ago. It takes time.

"Improving on a great desing? (Athlon -> Palomino) Great! Don't add a heat spreader to reduce the fragility of the core. Bad."

Same mantra we've been hearing for a couple years now. The heat spreader theory is a farce. The cores aren't nearly as fragile as some would like you to believe.

"Making a great chipset to help the stability of DDR based Athlon solutions? Great! Stopping production because you say that the DDR chipset offerings are fine when it is really because you dropped your ASP's (Average selling prices) so low you can't afford to keep the fab open? Bad."

Again, do you have evidence to support that, or is it just an opinion? I assume you're referring to AMD761. Considering that AMD762 and 766 are still being produced, I have a pretty hard time believing that any decision to halt production of the 760 north bridge had anything to do with low ASPs or what have you. The fact is, better -performing- solutions are out there now. I'll take the 760 anyday over the unstable and unreliable VIA north bridges, but that's just my opinion. Performance is key these days.

"I could go on, but it's these missteps (and many more) that will keep a superior CPU core design from becoming the dominant CPU. Kind of like football, it's about execution..."

Shall I begin listing the missteps of AMD's largest competitor? Get the point yet?

 

[smp]

Heh heh.. "I set my BIOS to shut off if the CPU gets to 55C "
Mine is happy at 57C under load (genome@home... do it) right now.. during warmer times it's been happy at 61C.. always stable too. Of course, it depends on the motherboard.

 

[ST4RCUTTER]

It should also be noted that it has always been AMD's mantra that they are not in the chipset biz. The only reason the 750 and 760/761 line of chipsets came about was because at the time they had next to zero chipset support. In fact, the first chipsets were intended to be reference designs for the third party chipset makers. The only chipset that AMD will be focusing on in the future will be the 760MP line for their server solutions, where stability is paramount.