0.09 micron in 2003 and 10-20 GHz in 2005

[KILLorBE]

Anybody heard/know more bout this?
Intel's new transistors 27.11.2001 @ 18:00 by lynx

Intel has developed a new transistor, called TerraHertz. The transistors are made up of semiconductors using high-k dielectrics, epitaxial wafers, and a real shocker - silicon-on-insulator (SOI) technology. (Which means we are still paying for processed sand) The company will release 10-20 GHz. series with these transistors probably in 2005 but rumour says that they will use pieces of this new technology in 0.09 micron chips in the first half of 2003. http://www.tweakers.net/g/s/kwijl.gif http://www.tweakers.net/g/s/kwijl.gif

[Bmxpunk86pl]

well then we they will need newer ram to be able to keep up with those speeds and like 1 meg of cache on the chip.

[TumbleMonkey]

Sounds like we are all gonna have to use SDRam, spensive, but itll come down.

[EvilCloudStrife]

where where can i buy the intel 10ghz cpu?

[noxqzs]

nice thread revival. :p

[ShadowClock59]

Haha really, nice 4 year bump. :p

[d94]

so 4yrs later and the majority of us intel enthusiasts ppl are @ 4ghz oc'd...hahaha

[Aphex_Tom_9]

hahahaahaha

[Nathan0490]

Gotta love it :)

[Careface]

lol thats a crack up.. like the wholy 'never need more than 16mb of ram' thing :D gotta love company bs

Careface*

[3DFlyer]

Why was another dead thread brought up in this forum? That is old news that you know the technology didn't come to pass.

I keep seeing ancient threads dug up from the archives. All it does is bump current useful information off the board, and litters the board with nonsense.

Please folks, let these dead threads stay buried. When I first read this, I thought it was some spctacular new development only to find it's another worthless out of date thread.

I can see revisting an old post maybe if it had some useful or relevant info in it, but this is worse than spamming. It's totally useless. C'mon guys, you know what I'm talking about here. This stuff makes the forums look messy.

We now have over 60,000 members registered. They don't want to see this kind of stuff.

[matttaylor]

Why dont they want to see this stuff? It's actually kinda funy to see that this never came true. 10ghz? i wish, but at the time this thread was created it was a possibility, but do to heat and money issues these kind of clocks arent around. But hey there is still time, it's only november, so intel has about a month and a half before this thread can be called untrue.

[3DFlyer]

There was never any promise that Intel would release anything with the technology, and the tech was only a rumor to begin with. it was basically dead when it was posted. There's no untruth because this was nothing but a rumor to begin with.

[Aphex_Tom_9]

There was never any promise that Intel would release anything with the technology, and the tech was only a rumor to begin with. it was basically dead when it was posted. There's no untruth because this was nothing but a rumor to begin with.
for nostalgia's sake, it's fun to look at even what consitutes an old rumor. if you dont want to look at this thread, then get out of it. I find it entertaining to look back on what we thought and hoped for, and compare it to how things are now.

[3DFlyer]

On hindsight it is very easy to get out of now, but I shouldn't have had to waste my time reading old rumors that never materialized.

When i saw this, I thought it was something serious that just came out...not something that was buried 4 years ago.

All this ever was was a false rumor.

[orion456]

Those who don't know history are doomed to repeat it.

I enjoyed seeing an ancient rumor dug up and put to ridicule. We should do more of that and not less.

[3DFlyer]

So it's supposed to be fun to redicule a person who posted a rumor, and is obviously no longer here to defend himself? We don't even know why this was posted, or what it was based on. At the time, it might have been credible, but who knows...it's 4 friggen years old!

Sorry, but I'm definitely out of this now. That's definitely not the way we do things here.

I'm out of this and unsubscribing to this nonsense right now.

[EvilCloudStrife]

haha.. just chill out 3DFlyer. Dont have a heart attack. i thought it was funny when i found this and im sure people appricate it.

BTW - no one is making fun of the OP. No one at all. Rumers like this happen all the time. That is the fun of all this new technology speculation. It didnt come true. Sometimes it does. I thought it was interesting and funny.

Not the way we do things here? everyone likes a good laugh now and then. I think its totally relevent because now that it is the year 2005 we can look back and ask ourselfs what went wrong. Why didnt this come to pass? What is in store for the future?

All in all, its not a big deal....

[Blackmage]

lmao, i think intel is holding something back from us.

[Vrykyl]

lol, the rumour forgot that these 10ghz chips would also be 'hot as 3 suns' and require long distance operation to stop your skin from being burned by them :D

[BenF]

lol, the rumour forgot that these 10ghz chips would also be 'hot as 3 suns' and require long distance operation to stop your skin from being burned by them :D

lol! That's awsome

[stang8118]

i don't mind old rumor threads like this being bumped (as long as its not like 100 a day). It's pretty interesting to see the rumors of the past, when now we can put them to rest. Acually sometimes when i am bored i will go back and read some 2+ year old threads, just to see how OCing was back before i was involved in it.

[Myhre]

I think this thread is awesome. I think it's a credit to the guy that started it and the two responses he got.

well then we they will need newer ram to be able to keep up with those speeds and like 1 meg of cache on the chip.

This guy was pretty close to the truth, we did end up having 1mb+ cache on the chip.

Sounds like we are all gonna have to use SDRam, spensive, but itll come down.

I really like looking back and seeing were we thought the industry was heading. Remembering when SDram was expensive is awesome.

I guess I'm just a sucker for nostalgia.

[AdvanS13]

at least the soi part is right

[macklin01]

Notice that Intel didn't say they were going to come out with 10 GHz CPU's, but rather that they could most likely get individual transistors to switch at that speed. Having fast transistors is a step on the way to fast CPU's, but not the end step. It's also a good way to develop new technologies that will eventually find their way into future products. This isn't all that different from the concept cars that are seen at car shows. They're probably impossible to mass-produce and have problems of their own, but elements from their designs will trickle into mainstream models over time. Think of these as "concept transistors." More updated information is available here (http://www.intel.com/technology/silicon/micron.htm#tera).

Google for Intel TeraHertz transistor, and look at the first hit (on intel.com). You'll find a very reality-aware quote:
"Our research has shown that we can continue to make smaller and faster transistors, but there are fundamental problems we need to address around power consumption, heat generation, and current leakage," said Intel Labs director of component research Gerald Marcyk in a prepared statement. "Our goal is to overcome these barriers and produce chips that have 25 times the number of transistors of today's microprocessors at ten times the speed with no increase in power consumption." Even in 2001, Intel had an inkling of the problems it was about to encounter. (I'm sure that AMD and IBM were also aware.)

There may be cases where you'd want faster transistors in certain key parts (e.g., if a part of the CPU coordinates the work of other parts) of a CPU design, although I don't know the engineering aspects well enough to comment further on that. In such a case, however, one could see very fast transistors used in a CPU with an apparently lower clock speed. In fact, does anybody here know what speed all the transistors inside a 3 GHz CPU actually switch at? I'd imagine that they switch fairly faster than at 1/(3000*1000) second speeds.

Also, IBM, AMD, and Intel were all caught off guard by the new difficulties of 0.09um. Sometimes, you find unexpected physics affect your long-term plans. Nobody was immune to these surprises.

In these respects, there's nothing all that ridiculous about the "rumor" brought up in the original post a few years ago. (Not to mention that they weren't rumors, but rather misunderstood or misrepresented research findings.)

As for bumping up old threads, if it's for the purpose of a laugh (e.g., with a contentless post), rather than being posted with a call for retrospection (http://www.m-w.com/cgi-bin/dictionary?va=retrospection), then the intent of that bump is suspect. -- Paul

[deathman20]

Holy cows lol. Now this is a post from the past. But its definatlly funny to see old news like this brought to the top, expecially 4 years later showing you how much and what the direction of the companies changes.

BTW: I want a 10-20Ghz CPU also. Giveme giveme

[AdvanS13]

Holy cows lol. Now this is a post from the past. But its definatlly funny to see old news like this brought to the top, expecially 4 years later showing you how much and what the direction of the companies changes.

BTW: I want a 10-20Ghz CPU also. Giveme giveme
there are 8-10ghz chips testing in the RnD plant here...hmm 9.8ghz merom :)...to bad it won't happen to retail :eh?: ...cooling would be a pain

[deathman20]

there are 8-10ghz chips testing in the RnD plant here...hmm 9.8ghz merom :)...to bad it won't happen to retail :eh?: ...cooling would be a pain

Evil pure evil, teasing us like this hehe.

[AdvanS13]

yea i told them to give it to me and they laughed ;)

[deathman20]

You could always say you where borrowing it :)

[Sentential]

Oh the irony of life

[Chaos_Being]

lol, the rumour forgot that these 10ghz chips would also be 'hot as 3 suns' and require long distance operation to stop your skin from being burned by them

I can see it now, joe sixpack buys a computer and lets the cooling fan get clogged with dust, and then wonders why his computer melted into slag and burned through his floor :D

I thought it was humorous to see such an old (and outlandish) boast as well. It makes me remember an old intel roadmap I saw a few years ago- there were definitely 7, 8, 9 and above ghz processors listed on there ;)

[AdvanS13]

although possible, that was the direction back then. obviously the ghz race is over

[deathman20]

Back in the day when things where simple, 1 CPU and maybe HT. Now Dual Cores that don't clock at the same speed (wish we could just clock 1 at a time, that be sweet). Next Quad Cores... Man can't wait to try and OC that hehe.

[Stratcat]

Ahhh...don't let the current Intel "Can't do 4GHz" smoke screen deceive you!

It's just a huge multi-corporate, multi-industry conspiracy.

The new 20GHz Intel chips are actually alive & doing well, and nearly ready for distribution.

They are going to be used in the ECU of next year's cars, which will run on tap water, get >300 MPG, and produce absolutely zero emmisions.

Really.

Strat

[Shelnutt2]

Ahhh...don't let the current Intel "Can't do 4GHz" smoke screen deceive you!

It's just a huge multi-corporate, multi-industry conspiracy.

The new 20GHz Intel chips are actually alive & doing well, and nearly ready for distribution.

They are going to be used in the ECU of next year's cars, which will run on tap water, get >300 MPG, and produce absolutely zero emmisions.

Really.

Strat



20GHz chips. AMD would be left in dust. Then agian they probaly would require their own power plant and vacum cooling setup to run them.

[Stratcat]

20GHz chips. AMD would be left in dust. Then agian they probaly would require their own power plant and vacum cooling setup to run them.AMD is countering w/an Athlon512 24000.

Concerning this, Michael Dell released this statement:

"Dell is constantly evaluating new technologies, and at the present time, we don't have AMD processor-based systems in our portfolio. AMD's Athlon processor is an unproven computing platform."

Strat

[Avg]

Although we didn't reach 10ghz plus, all prescott owner are close to it. Wait don't throw any stones yet... For those who don't know there are parts in the prescott that runs at twice the cpu's clockspeed, so those who have reached 4ghz there are parts in the cpu running at 8ghz, and that is one of the reasons why the prescott runs so hot.

[deathman20]

Although we didn't reach 10ghz plus, all prescott owner are close to it. Wait don't throw any stones yet... For those who don't know there are parts in the prescott that runs at twice the cpu's clockspeed, so those who have reached 4ghz there are parts in the cpu running at 8ghz, and that is one of the reasons why the prescott runs so hot.

And I thought it was just all the leakage and power hogging issues causing them to run so hot :)

[KILLorBE]

So it's supposed to be fun to redicule a person who posted a rumor, and is obviously no longer here to defend himself?
According to his Public Profile his last visit was 11-06-05 at 05:20 PM...actually his last visit is right now :D

And why would I need to defend myself?

[ronaldo]

I like that Intel/Amd changes the way of thinking, and i like the future multi-core (when Intel say around 100 cores on cpu by 2015) more than i would liked more ghz...

[macklin01]

According to his Public Profile his last visit was 11-06-05 at 05:20 PM...actually his last visit is right now :D

And why would I need to defend myself?

I couldn't have said it better, myself. Actually, I've really enjoyed reading the original responses, as they have proven to be pretty insightful. (DDR, etc.) And actually, as we've read in the newer responses, some aspects of what you posted also came to pass.

Not bad! :) -- Paul

[3DFlyer]

According to his Public Profile his last visit was 11-06-05 at 05:20 PM...actually his last visit is right now :D

And why would I need to defend myself?

Becasue folks were laughing at the stuff you posted, and you were not here. Now you are, so that's great. I just don't like it when people make jokes about others who aren't around, and may not be because the post was so buried probably unsubscribed from it 3+ years ago.

[Avg]

And I thought it was just all the leakage and power hogging issues causing them to run so hot :)
That is why I said it was one of the reasons, because it runs at twice the speed it leaks a lot more then the rest of the cpu. Intel themselved reconized this as a problem themselves. :D

[lowfat]

AMD is countering w/an Athlon512 24000.


Strat

haahahaha. that made my morning.

[Sjaak]

512 bit..yummy. That's gonne be hell and high water for programmers.

[orion456]

So it's supposed to be fun to redicule a person who posted a rumor, and is obviously no longer here to defend himself? We don't even know why this was posted, or what it was based on. At the time, it might have been credible, but who knows...it's 4 friggen years old!

The fun isn't directed at any poster for Pete's sake. It's all about how wrong most future prediction are and a great warning for those caught up in the excitement.

As for people predicting we won't see 10 ghz very soon....think again.

"[HP] has published a paper in the Journal of Applied Physics [2005] demonstrating a crossbar latch which gives signal restoration and inversion computers need without using transistors. That, it explained, could result in computers thousands of times more powerful than existing machines." The crossbar latch uses nanometre scale devices that are cheap and easy to build, said HP. The technology could replace silicon technology, which is facing fundamental problems in physics and challenging scientists at Intel"

http://www.theinquirer.net/?article=21020

http://www.hpl.hp.com/news/2005/jan-mar/crossbar.html

[Chaos_Being]

Interesting links!

"We are reinventing the computer at the molecular scale," said Stan Williams, HP Senior Fellow and QSR director, and one of the authors of the paper. "The crossbar latch provides a key element needed for building a computer using nanometer-sized devices that are relatively inexpensive and easy to build." A nanometer is one billionth of a meter or 39/1,000,000,000 inch.

[Stratcat]

512 bit..yummy. That's gonne be hell and high water for programmers.
Probably not as much hell & highwater as programming single apps as multi-threaded, for the proposed upcoming quad-cores! :rolleyes:

Strat