1MHz

[HaMMuS]

I'm currently working with a microwave detector which oscillates at 2.5THz. The fact is that as technology progress, such numbers will only become more commonplace, and in some ways the terms used to describe them will lose their meaning, as people will not immediately think "1GHz = 1 Billion Hz". Certainly in computing terms, it is often only the number infront of the MHz/GHz which is most important to people eg 4GHz>3GHz.

 

[Wipster]

I imagine that there is a theoretical maximum clock speed? Surely there is a point where the electrons just cannot move fast enough?

Yeh I guess there would be, if the clock takes longer to propagate through the chip then the period and settle time that would be it

 

[p4radox]

I imagine that there is a theoretical maximum clock speed? Surely there is a point where the electrons just cannot move fast enough?

That's countered by using smaller transistors, ie: less far to travel. The limit is therefore the theoretical minimum transistor size.

Anything below 16nm tech will probably be inpracticable due to weird quantum tunnelling effects.

 

[Big.Wayne]

Off Topic Insight:

No offense intended Wayne, but how do you have 11'000+ posts on an overclocking forum and not know what Hz are etc?

Hi Amraam,

It's a fair question and no offence taken! . . . I am a complete non-academic and entered the computing scene in 1995 as a photographer who wanted an Apple-Mac to learn about digital imaging . . . however the £1500 price tag was prohibitive so a good friend of mine who was a fuzzy-logic/neural network expert suggested he could build a PC for just £350 . . . he dragged me around a London computer fair and bought a few hundred quid of new and used components before we travelled back to his flat to get the thing built . . . . sadly after two/three hours both himself and an other friend had failed to get the newly constructed budget PC to boot! . . . I then reluctantly had no choice but to dive in an get hands on as a trouble-shooter because my £350 had turned into a large beige box that didn't do very much . . . there was no internet, no overclockers.co.uk forums etc . . . so much harder back then!

I have always acquired knowledge on a Need-To-Know basis, I'm not one to spend time learning something if I cannot then *apply* that newly acquired knowledge to a situation . . . I've never needed to know how a MHz subdivides and my only grasp of mathmatical figure are directly related to money . . . £100/One Hundred Pound, £1000/One thousand pound etc

My first computer was a Compaq 66MHz so I have always thought of the minimum denomination in MHz (Megahurtz) and have never encountered a setting or one situation that deal with Hz (Hurtz) . . .

The overclocking arena is filled with people from all walks of life and not soley just computer-science academics . . . what I lack in theory I hope is offset against 15 years of hands-on, practical experience!

I never studied photography either, it's just another one of those vocational things . . . back in 1983 while my friends were working hard at school on their maths O-Levels (no computer in schools back then) I was out starting a career as a Musical-Press Photographer travelling around London and the U.K taking pictures of as many rock-bands as possible! . . . I didn't know what silver-halide was or how it worked, no clue how depth-of-field related to aperture etc . . . all I knew was how to load 35mm film into a camera, how to focus and how to blag myself photo-passes! . . . I got a *huge* archive of work from the 80's that is waiting to be digitized and hopefully made into an exhibition and a book, kinda lots of black & white photos like this:

Moral of the story is, You can study more and do less . . . or you can study less and do more, knowledge for knowledge sake is overrated IMHO . . . i just think you need enough to be confident in what you do . . . keywords being *confident* and *do* . . .

The only reason I made this thread was because some information is reaching me about possible health-implications to people that live/work in close proximity to modern computers caused by some *Vibrations*, I like to make my own mind up about these things so I needed to know how many *actions* were taking place inside a processor, even if they are on a tiny, tiny level I consider a machine that Pulsates/Oscillates at up to 16 *BILLION* times a second to be something that may well have an effect on the Human Body . . . if there is any truth in this it's not gonna be made public knowledge is it? . . . at least not through the conventional media (BBC News, Custom PC etc! ) . . .

Anyway, that's all for now, just thought I'd paint a picture for you and although I have not studied computer sciences I would like to think i have earned my place in the OcUK community with hands-on technical support and trouble-shooting advice which is a SkillSet you could never acquire from a book or calculator! . . . as individuals we all have our weak spots but we often don't like to admit it, at least not in public . . . it's am ego thing I think but sadly that blocks the path to actually learning stuff as people will be scared to ask what is obviously a really simple question as others may scoff or snigger! . . . I think the best teachers in life are the humble ones . . . and humility is a great but very misunderstood virtue . . . I think this quote sums it up nicely!

"The reason why seas and oceans are great, is because they're lower than the rivers and streams" - Ksmail

 

[MikeHunt79]

That's countered by using smaller transistors, ie: less far to travel. The limit is therefore the theoretical minimum transistor size.

Anything below 16nm tech will probably be inpracticable due to weird quantum tunnelling effects.

I see, so eventually the ever increasing clock speed will hit a wall, and so will the ever shrinking die's... It makes you wonder how many years it will be before this happens, but I guess now we have multicore setups with software taking advantage or more than 1 core, then clock speed will hopefully be less of a limitation than it once was.

 

[DavyBoy]

What kinda effects are we talking about Wayne?

Got me all paranoid now, thinking I'm going to get cancer from my PC!!!!!

 

[MikeHunt79]

^ I would not worry unless you're using your PC without a case.

Anyway, that's all for now, just thought I'd paint a picture for you and although I have not studied computer sciences I would like to think i have earned my place in the OcUK community with hands-on technical support and trouble-shooting advice which is a SkillSet you could never acquire from a book or calculator! . . . as individuals we all have our weak spots but we often don't like to admit it, at least not in public . . . it's am ego thing I think but sadly that blocks the path to actually learning stuff as people will be scared to ask what is obviously a really simple question as others may scoff or snigger! . . . I think the best teachers in life are the humble ones . . . and humility is a great but very misunderstood virtue . . . I think this quote sums it up nicely!

"The reason why seas and oceans are great, is because they're lower than the rivers and streams" - Ksmail

Nice post, forums like this one can be a great place in order to share knowledge, no matter what your background is. I'm still learning new stuff everyday, and it's nice to see a thread like this one as opposed to the usual 'spec me a gaming router' type of thread.

In my limited experience computer science degrees tend to gloss over much of the hardware side. I think this is because many people taking the course would be going on to a software development or (god forbid) a project management role as opposed to a hardware based role. While it did cover some low level stuff tho like assembly language and binary, but I imagine an electronic engineering degree would be more relevant to the low-level hardware side.

I find the whole electronics/hardware side more interesting personally, and most of my knowledge in this area is self-taught or the results of a few home experiments, right now my current project is building a device that can resurrect dead car batteries!

 

[kppanic]

I see, so eventually the ever increasing clock speed will hit a wall, and so will the ever shrinking die's... It makes you wonder how many years it will be before this happens, but I guess now we have multicore setups with software taking advantage or more than 1 core, then clock speed will hopefully be less of a limitation than it once was.

We hit the clockspeed wall full force head first with pentium 4s as they reached 4ghz at that litho level the TDP was just mentally high. So far rather than the clockspeed the limiting factor has been the amount of heat the transistors generated as a result of switchng them on and off.

 

[Big.Wayne]

Got me all paranoid now, thinking

No mate!

It wasn't anything terrible like that . . . it was to do with electromagnetic fields? and how it effects the operator on a Energetic level, this would be things like a weakened Immune System, Fatigue, Chronic Fatigue, problems sleeping, depression/moods etc . . .

if 1MHz is One-Million Cycles then I've got a Triple-Core sitting here now Oscillating 10.8 *BILLION* every single second . . . in the time it has taken me to write this post alone I have been exposed to the fallout from 648 *BILLION* ripples of electromagnetic radiation . . . ouch!

 

[Big.Wayne]

I'd be tempted to round it up to 2 Teracycles.

I really like that idea btw, I am not familiar with that word but I think it sounds cool!

Is it possible to knock-up a spread sheet to work that out!

Core Number = [Manual Input here]
Core Frequency = [Manual Input here]
Teracycle Duration= [Auto Answer Here]

Prime95 Teracycle Stability Test lol . . . there will be a race to see who can complete it the quickest! . . . the person with the most cores running at the highest frequency would win right? . . . so the faster your computer was the shorter duration you would need to test it?

Weird Science!

 

[weird_dave]

if 1MHz is One-Million Cycles then I've got a Triple-Core sitting here now Oscillating 10.8 *BILLION* every single second . . . in the time it has taken me to write this post alone I have been exposed to the fallout from 648 *BILLION* ripples of electromagnetic radiation . . . ouch!

Not quite, having 3 cores does not increase the number of clock cycles per second, as has been explained earlier.
However, it is the clock edges that produce most of the EM field (for the clock signal), and since there are 2 every cycle, you'd actually have 7.2 billion bursts of EM radiation per second. Ripples wouldn't be correct since there could (will) be lots of super fast ones at every clock edge due to the small inductance and capacitance in the chip.

 

[Tilluss]

This thread was an interesting read, kinda opens you eyes as to how fast even basic computers really are.

 

[Kribby2006]

Not quite because Hyperthreading only kicks in when there is available execution units on the cores.

But you do have to consider that Core 2 processor cores (and i7's) are 4 issue cores, which means each core can execute 4 different instructions at the same time, and a large number of x86 instructions can be processes in a single clock cycle. The point of hyperthreading is to try and make sure that all 4 execution units (in each core) are as heavily utilized as possible.

So on a Quad core, with hyperthreading and the right software, your looking at an maximum instruction execution rate of 4(cores)x(4 execution units) x clock speed.

In reality, most applications use a broad range of instructions which may take as low as 1 clock, but some instructions take multiple clock ticks to complete.

Lol how many times hahah it seems everyone missed the bit where wayne said imagine it was eight actual cores. I know HT doesnt work like that See below:

Ok for the purposes of this example lets just say it's a full core . . . so eigth cores running at 4GHz basically . . . how many light bulbs flick on and off!

What is the figure?, I can't work it out myself and I need the answer expressed in words as I do not understand large numbers!

 

[Amraam]

With the amount of random stuff you post you should have about that many posts soon too !


Really who cares...
And we have a different forums section here too if you haven't noticed yet, some of them have completely nothing ( or almost nothing ) to do with IT or overclocking at all.

Have a look below the hardware section and you'll find them

Sorry, my post did sound offensive, I didn't intend that way

I was just shocked hehe. I'm very much a noob with OC stuff anyway, so I can get away with being a bit cheeky

My random posts are always because I'm too scared to spend money haha. I ask what something is like hoping for someone to chose for me, because I'm too indecisive (and stingy) to spend money quickly

 

[Amraam]

@ Big.Wayne,

Good post mate, I agree. As I said above, after rereading my post it did come across as offensive, but I really didn't mean that way - I should have added in smileys

I'm similar. I work in IT and I've done a basic computing degree - I learn what I need to, or more importantly, what interests me

Please don't think I was mocking you, or scoffing/sniggering as you put it I honestly wasnt, my post was in jest alone

 

[tomos]

That's countered by using smaller transistors, ie: less far to travel. The limit is therefore the theoretical minimum transistor size.

Anything below 16nm tech will probably be inpracticable due to weird quantum tunnelling effects.

i'm sure that intel already have to take some things like this into consideration nowadays. think it was on Anandtech a post by IDC.