[solofx7]

I geek bench my i7 iMac when I first got it. I was getting roughly 9600-9700 on my scores. Now a month later I am getting 10k plus on geek bench. I am very happy about this. Yes, I know that I am super dorky I forgot all about burn in. But that does remind me, a windows person asked me if apple does burn in testing before shipping. I am oretty sure it is a yes, but I do not know for sure. Thanks

[Oisín]

Geek bench: a place near a wall at a high school prom, filled with acne-ridden boys with glasses.

[SpaceMonkey]

Why would computers have burn in?

[solofx7]

Originally Posted by Oisín 

Geek bench: a place near a wall at a high school prom, filled with acne-ridden boys with glasses.

I know, its horrible. I do not use my computer for video editing or anything like that, but I just love to see the number it gets
I just bought it so I could see the number

[solofx7]

Originally Posted by SpaceMonkey 

Why would computers have burn in?

I am pretty sure that is the correct term.
Burn in meaning test to near capacity.

[ibook_steve]

I have never heard that term before used in that way. "Burn in" usually refers to "screen burn in" on old CRT displays. Do you mean stress testing?

Steve

[solofx7]

I know why, but it just seems faster out of the box. Though I know the top end will show faster now.

[solofx7]

Hi Steve, Normally that means that they test the hardware of the computer under very high loads. I know that it is a term used a while ago for testing of gaming machines.

[Simon]

I think you might be referring to a break-in phase rather than "burn in".

[solofx7]

Originally Posted by Simon 

I think you're referring to a break-in phase rather than burn-in.

You are probably right, the name has probably changed or I am getting to old to remember the right one

[mduell]

Could be driver improvements in 10.6.3.

[P]

Originally Posted by mduell 

Could be driver improvements in 10.6.3.

That's the most likely answer - OpenGL support got a real boost with 10.6.3. Could also be related to ambient temp and its effect on turbo modes.

EDIT: Wait, geekbench doesn't test graphics, does it? Can't be graphics drivers, then, unless it's putting calculations on the GPU somehow.

[solofx7]

Originally Posted by mduell 

Could be driver improvements in 10.6.3.

I guess thats possible. Please correct me if I am wrong, but doesn't hardware operate more efficiently after a small window of being used i.e. real world out of the box usage, not testing?

[mduell]

No. What, the electrons need to find their way? Snake oil.

[P]

I've heard the term burn in being used of new computers, but that only meant to torture the machine a bit early on (looping benchmarks or something like that), before you put anything important on it. The thinking is to find any potential defects in that initial run, so a bad machine can go back to the OEM directly without risking important data. I've never heard of it as a way to improve performance.

[Big Mac]

Originally Posted by solofx7 

I guess thats possible. Please correct me if I am wrong, but doesn't hardware operate more efficiently after a small window of being used i.e. real world out of the box usage, not testing?

Nope.

[ghporter]

"Burn in" is a valid term for pre-shipping, near-max capacity test of a piece of electronic hardware. It's often used in conjunction with PCs, because they are "integrated" by some builder who has to take for granted that a supplier's specs are really valid...

[solofx7]

Originally Posted by ghporter 

"Burn in" is a valid term for pre-shipping, near-max capacity test of a piece of electronic hardware. It's often used in conjunction with PCs, because they are "integrated" by some builder who has to take for granted that a supplier's specs are really valid...

Thanks, see I did not make that up

[l008com]

My new car got more powerful after it's burn-in period. But a computer? That seems like a prime piece of misinformation to me?

[ghporter]

A computer burn-in is a final, running test (under load) to determine whether it will run when it gets to the customer. That's it. It even used to be a selling point for all the "mail order" PC clone makers; "Our computers are 100% burn in tested before they are shipped!"

[Eug]

Originally Posted by l008com 

My new car got more powerful after it's burn-in period. But a computer? That seems like a prime piece of misinformation to me?

It is, esp. in this context. The speeds are limited to pre-determined levels anyways. Our computer chips are binned, and generally can't run faster than they're set to run, esp. on Macs.

The only place where I've seen people really talk about speeds increasing with burn-in is in the PC overclocking crowd... but we can't manually overclock our i7 iMacs... Maybe he's trying to take the overclocking burn-in (snake oil?) concept a step further to claim that burn-in is somehow making Intel's Turbo Boost more efficient?

[solofx7]

Originally Posted by Eug 

It is, esp. in this context. The speeds are limited to pre-determined levels anyways. Our computer chips are binned, and generally can't run faster than they're set to run, esp. on Macs.

The only place where I've seen people really talk about speeds increasing with burn-in is in the PC overclocking crowd... but we can't manually overclock our i7 iMacs... Maybe he's trying to take the overclocking burn-in (snake oil?) concept a step further to claim that burn-in is somehow making Intel's Turbo Boost more efficient?

I truly forgot how to explain this. I will find something that stated this better than I have. I guess the bottom line is that my i7 is testing fast then It was on day one by a bit. When i was on the pc side of computers for years computers seem to run fast after "burn-in" meaning after being used for a while. I do not mean monitor burn in. I guess you would equate this to warming up a tv or car and it performing slightly better. Not a huge difference, but a noticeable one. Thanks for your feedback. I love this board.

[Spheric Harlot]

The speed of all components is governed by clock circuits and clock dividers.

I doubt that "burn-in" can increase clock frequency.

[mduell]

Originally Posted by solofx7 

I guess you would equate this to warming up a tv or car and it performing slightly better.

Actually the Intel chips may allow less turbo when they warm up.

higher temp -> higher TDP @ same clock -> less TDP margin -> fewer turbo boost buckets allowed

[Big Mac]

There are some who claim that an initial burn in period of stressing the components will help optimally bond hardware components, but most experts discount the alleged benefits. solofx, benchmarks may differ in trivial amounts from run to run. I can almost guarantee your computer hasn't gotten any discernibly faster because of any burn in you may have done.

[solofx7]

Originally Posted by Big Mac 

There are some who claim that an initial burn in period of stressing the components will help optimally bond hardware components, but most experts discount the alleged benefits. solofx, benchmarks may differ in trivial amounts from run to run. I can almost guarantee your computer hasn't gotten any discernibly faster because of any burn in you may have done.

I am not saying that I did anything to make it faster. I am just saying the components "wore-in" and got faster

[solofx7]

I give in. You all win.

[Simon]

I've seen high-end audio cables being sold as 'unidirectional'.

[ShortcutToMoncton]

Originally Posted by solofx7 

I give in. You all win.

If you could come up with any plausible explanation for your theory - "electrons travel faster when they've gotten used to the turns in the chip pipeline" - then you might get somewhere.

greg

[solofx7]

Originally Posted by Simon 

I've seen high-end audio cables being sold as 'unidirectional'.

Very funny. No, I do not have " monster cables" on anything

[l008com]

I use special electrons from an alternate universe. They don't heat up nearly as much due to their alternate laws of physics. My computer is so fast, I won't even be making this post until next week.

[solofx7]

Originally Posted by ShortcutToMoncton 

If you could come up with any plausible explanation for your theory - "electrons travel faster when they've gotten used to the turns in the chip pipeline" - then you might get somewhere.

greg

Hey, no need to take it personal or take a shot at me. So you mean evidence other than the actual results I had? Hehe.... In any event, I am just happy with numbers on my iMac that are close to Mac pro type speed.

[l008com]

That didn't sound person at all, that sounded like him repeated what everyone else has been saying, there's no such thing as "burn-in" the way you describe it. Your only evidence was your own test results, which can vary greatly based on many of the things already said in this thread. I wish this thread would die die die.

[solofx7]

Originally Posted by l008com 

That didn't sound person at all, that sounded like him repeated what everyone else has been saying, there's no such thing as "burn-in" the way you describe it. Your only evidence was your own test results, which can vary greatly based on many of the things already said in this thread. I wish this thread would die die die.

I think that I initially said that my description may have been off. Again, I admit that I am wrong.

[Spheric Harlot]

Originally Posted by solofx7 

Very funny. No, I do not have " monster cables" on anything

Nobody with a real interest in audio quality or business ethics has anything Monster Cable in use.

[l008com]

I have a monster in my pants.

Die thread, die!

[solofx7]

Originally Posted by Spheric Harlot 

Nobody with a real interest in audio quality or business ethics has anything Monster Cable in use.

That was the joke I was getting across.