Why are Intel CPU's better than AMD's at the same clocks?

[Turbonium]

I have really been also confused by the same thing as to why Intel has a hang over AMD?

But the best answer i have ever heard is that iNTEL IS MORE reliable than AMD due to the fact that in the past AMD processors where known to Overheat alot wich will mean investing in a new processor in a short period of time.

Correct me if i am wrong please.

No clue what you're talking about.

Strictly speaking, it's all about R&D, which comes from a certain % of your budget, which comes from your profits, which come from sales, a percentage of which comes from the consumer segment, where marketing plays a huge role; Intel has always been good at marketing, which coupled with better business (R&D) decisions, puts them in the position they are today.

Intel also makes "GPUs".

I lol'd.

 

[Cerb]

I have really been also confused by the same thing as to why Intel has a hang over AMD?

But the best answer i have ever heard is that iNTEL IS MORE reliable than AMD due to the fact that in the past AMD processors where known to Overheat alot wich will mean investing in a new processor in a short period of time.

Correct me if i am wrong please.

Yes.
1. Whatever the throttling mechanism Intel used for the P4 series was, it did not prevent at least two cases of an overheating processor, that I have personally seen.
2. The Athlon XPs fixed the problem. So, it existed in the wild for a little over a year.
3. They would only overheat if not cooled. It was a rare problem, while it existed.

 

[Turbonium]

Yes.
1. Whatever the throttling mechanism Intel used for the P4 series was, it did not prevent at least two cases of an overheating processor, that I have personally seen.
2. The Athlon XPs fixed the problem. So, it existed in the wild for a little over a year.
3. They would only overheat if not cooled. It was a rare problem, while it existed.

Wasn't the P4 better at throttling than the Athlon? I remember reading posts about Athlons frying themselves if they were run without a HSF for more than a few seconds at load.

 

[Cerb]

Wasn't the P4 better at throttling than the Athlon? I remember reading posts about Athlons frying themselves if they were run without a HSF for more than a few seconds at load.

Athlons did not throttle themselves at all. P4s were supposed to, and most did, but not always. Now, why are you going to run any processor that puts out more than a few Watts without a HSF?

 

[Turbonium]

Athlons did not throttle themselves at all. P4s were supposed to, and most did, but not always. Now, why are you going to run any processor that puts out more than a few Watts without a HSF?

Ikr? I've read about it happening to some people by accident. I've also seen at least one YouTube video of someone doing it on purpose.

 

[NTMBK]

Ikr? I've read about it happening to some people by accident. I've also seen at least one YouTube video of someone doing it on purpose.

Being upset about running a CPU without a HSF would toast it is like being upset that your graphics card catches fire when you try running it without a cooler.

 

[Pray To Jesus]

 

[pelov]

Athlons did not throttle themselves at all. P4s were supposed to, and most did, but not always. Now, why are you going to run any processor that puts out more than a few Watts without a HSF?

You underestimate the stupidity of some people.

I remember going to an acquaintance's house because he needed help with his new PC he'd built. It was an SFF/HTPC in what looked like a receiver type of case. I fired it up, and it would crash after the first few seconds upon boot, just after BIOS posted. OS didn't load, so I'm thinking maybe HDD, mobo or RAM? I open it up to make sure he's wired everything accordingly only to find out he never had an HSF on the CPU...

"It wouldn't fit"

I told him to put his finger on the CPU as I booted up the machine.

"Tell me if it gets hot"

Of course it did, and of course the CPU was hitting its TJmaxx within mere seconds after booting up.

 

[Makaveli]

You underestimate the stupidity of some people.

I remember going to an acquaintance's house because he needed help with his new PC he'd built. It was an SFF/HTPC in what looked like a receiver type of case. I fired it up, and it would crash after the first few seconds upon boot, just after BIOS posted. OS didn't load, so I'm thinking maybe HDD, mobo or RAM? I open it up to make sure he's wired everything accordingly only to find out he never had an HSF on the CPU...

"It wouldn't fit"

I told him to put his finger on the CPU as I booted up the machine.

"Tell me if it gets hot"

Of course it did, and of course the CPU was hitting its TJmaxx within mere seconds after booting up.

i've seen this also.

Noobs will be noobs you have to expect the unexpected with them lol.

I witnesses a guy bend on the pins on his P4 trying to install it then wonder why it wasn't working and why I wouldn't let him touch my pc lol.

He was unaware that the chip only fits in the socket one way and decided to push it in.

 

[BallaTheFeared]

Being upset about running a CPU without a HSF would toast it is like being upset that your graphics card catches fire when you try running it without a cooler.

They don't catch fire though, they shut down just like modern cpus.

Well at least my 470s do

 

[steddora]

The p4 was a joke in my opinion. Rhambus ram was a good idea at the time but the cost wasn't. I had an Athlon XP 1700+ that more than kept up with the p4@2ghz. Problem was that three memory timing was horrible on those older platforms. And my xp ran 30 % cooler in the same room with comparable hsf's.

Either way, Intel learned their mistakes and worked back into r&d and came to the core 2 stage. And since amd has been falling in its paces. I hate seeing amd falling like they have. I don't want to know my next system is going to be Intel.

Either way, watching the market tells the tale, amd is doing poorly and if they don't start getting in the game, they may just fade our entirely.

 

[anongineer]

Short answer: money. Slightly longer answer: throwing money at process tech.

Micro-architecture is important and all, but transistors make them possible. You can draw diagrams for massively high throughput, low latency ALU's, but they'll only be awesome on paper if your transistors happen to be slow, and can't hit target clock frequency.

From a business standpoint, transistors could also
- have too much variability, which you could recover by binning, but it hurts margins and annoys OEM's, who might be expecting more supply of high value parts instead of bargain bin stuff
- tend to have mysterious holes or other defects in them, which really hurts yields (hope you only paid for known good die!)
- high leakage, in an age where people care about battery life and their electricity bill, to a lesser extent

 

[masteryoda34]

I have really been also confused by the same thing as to why Intel has a hang over AMD?

But the best answer i have ever heard is that iNTEL IS MORE reliable than AMD due to the fact that in the past AMD processors where known to Overheat alot wich will mean investing in a new processor in a short period of time.

Correct me if i am wrong please.

NO

 

[Saylick]

This is how I believe the story unfolded.

Like everyone else said, money is a huge contributor to the gap you see between the two companies, both from a market share perspective and from a performance perspective. Money allows Intel to keep the ball rolling while a lack of money limits the amount of progress AMD can realistically make in a given amount of time.

Ever since the Netburst fiasco, Intel has stuck to its tick-tock cadence which has resulted in the production of an updated evolution of its pre-Netburst microarchitecture every 2 years: Haswell is an updated version of Sandy Bridge, Sandy Bridge is an updated version of Nehalem, Nehalem is an updated version of Core, and Core was based on P6. In the foundation of each new microarchitecture lies sections of the old, but it is updated to fit the needs of its time. I wouldn't exactly say their post-Netburst architectures were the result of "out of the box" engineering in the sense that they've basically found a winning strategy, ran - or should I say is currently running - all the way to the bank with it, and never looked back.

AMD, on the other hand, must be constantly innovative and produce highly radical designs in hopes of catching an unforeseen trend if they ever want a chance at catching back up to Intel. In my mind, because Bulldozer is such a different microarchitecture compared to K10, it is an attempt at going outside the box in order to catch up to Intel... and we all know how that played out. Realistically, of course, the odds of catching up are nil to none at this point. Intel's ball is continuing to roll at a very quick pace (and speeding up) while AMD's ball is slowing down (layoffs). Perhaps AMD should begin rolling its ball in a different direction before it ends up with no ball to roll.

For the sake of being punny, long story short AMD dropped the ball.