Interesting Intel guy "View", Is it true?

[oskarfr]

Hey everybody....

Know I just met a "Intel" person today. Intel person, meaning no "AMD on my table, sir!!" He was explaining to me the reason why they (his company) did not build machines with AMD processors.
The answer was, AMD has to many errors and does not run all programs, well is not guaranteed to run all programs anyways.
Also he said that overclocking was useless, now that he knew how processors work. And AMD was a "overclocking" processor. Well the reason for this, he said, was that there is a error checking circuit, that checked if an operations worked out or not and by overclocking, errors became more often and therefor take longer time to finish instructions. He gave me an exemple: If you overclock a processor 20%, then you would maybe get 25% error and therefor actually get 5% drop in performance.

In short, I don't buy this. It doesn't seem to fit with Tom's benchmarks and.. well I have never heard this before...

Well what is your opinion, and if you really know your stuff, is there anything to this??

Oskar

 

[tfbww]

As you pointed out, EVERY review by Tom's (or if you went to one of the overclocking sites) shows that OC'ing improves performance. OC'ing is a great way to make more of out of less. Tell him to do a little more research next time. (Caveat being that you didn't over-OC, i.e. that you still have system stability.)

 

[bdaley]

Every benchmark I've done after OCing (successfully) a processor showed that performance increased.

Where did that numbnuts get his information? Probably from an Intel or AMD sales rep. who doesn't want you to OC, just spend more $ for a faster proc.

"There's no such thing as gravity, the Earth just sucks"

 

[slvr_phoenix]

Actually, I seem to very vaguely remember something about overclocking like that many years ago. I think it wasn't so much a degredation in system speed (obviously a successfully overclocked chip would perform faster than one that isn't), but there was the chance of your chip(really old school chips here) coming up with an error every so often because the signal just wasn't clear enough. And that could cause stability problems and/or just plain old system lock-ups.

But again, this is both VERY old overclocking and just a vague memory. Surely by now the only time you'd run into this is when trying to severely overclock a slow chip and are sitting right on the edge of it's capabilities.

Like say you had an Athlon 600. You know that Athlons can run at 1.2GHz. But despite all your effors your computer will only boot up with it at 900MHz. And even then programs occasionally crash for no reason or your system locks up every so often when it was a rock stable system at 600MHz.

That's just an example of when something like what was described might happen. But it is pretty much just theory. For all the luck involved it might run stable even at 900MHz.

- Sanity is purely based on point-of-view.

 

[AmdMELTDOWN]

I don't know I would like to see cold hard facts if I were to believe that, It seem lately ppl are testing virtual cpu's against imaginary cpu's, weird benchmarks from disgruntled ex-employees, 4way cpu's that are not really there. etc...

be it intel or amd if you use what you have in a productive manner and it helps you out then what's all the big deal about! end of story.

it's like you're sitting on one of the fastest chip on the planet and all ppl can do is battlezone? post 1000 msg per wk? d/l porn?

 

[Grizely1]

Go tell the guy he's a [-peep-] lamer and he sucks and shoulnd't be in the computer business.

 

[Guest]

Interesting, Does this fellow work for a electronic superstore? Cause those people usually do not know what they are talking about.

 

[Mordy]

>>also he said that overclocking was useless
the fact that there is a lot of people that using overclocked computers,and they like it, can proove othervice .

>>AMD has to many errors
All the test proofs it is sometimes opposite.
I agree with Grizli opinion about that person.
p.s. I use overclocked buses and CPU's since 386.



All i can offer you is the truth. Nothing more.

 

[oskarfr]

Actually he builds computers, Intel based of course. This is a computer company that sells computes to firms and indeviduals.
I thought it was interesting, that a person in the biz didn't actually know more about AMD and overclocking. maybe it's some kind of brochure from intel, that states all the "facts".

As you can maybe hear I lean towards AMD, but I would buy Intel processor if it was better and cheeper.

Oskar

 

[gong75]

When I OC'd my old K6-2, the benchmarks grew almost linearly with the MHz.

It is common that you know more than the sales person in a computer store. Once I bought a video card and wanted to test it before taking it home. Obviously he knew how to install it but couldn't find out how much video memory it has.

 

[lakedude]

If a chip is underrated by the manufacture and other chips with the exact same design are running much faster then overclocking almost always works very well. Being an overclockable chip is always a good thing even if you don't overclock because a chip running way slower than it could run will last and be stable. In the old days only Intel made chips that had any real headroom and Intel was the "overclocking" company. Now AMD also makes chips not clocked to their full potential so they are overclockable as well. Overclockable is a good thing.

 

[Guest]

"...I would buy Intel processor if it was better and cheeper."

If you want better and cheaper, buy the Athlon (or Duron). The guy's full of it.

 

[Guest]

Now there's an intelligent comment. Geez Grizely, how about engaging a neuron or two before spewing...

Here's a more useful comment:
The 'Intel guy' that confided (oooh) with the author of the 1st post here is a fake, in my opinion. Every time I see an overclocked system tested, I see increased numbers (no exceptions).

A processor either can deal with the MHz, or it freezes or generates hard drive errors (can be tragic) due to under voltage. It turns into a crispy critter if it has too much voltage, revs too high at the clock setting, and doesn't have a thermal overload protection circuit.

 

[Grizely1]

Thanks. Despite insults my message was brief, clear and to the point. If the guy thinks that then he doesn't deserve to own a business that has anything to do with computers.

Or maybe he did just read the manual that came with an Intel product....... Who knows.

----------------------
I hate intel