IDF: Core i7 Will Feature A Turbo Mode

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Status
Not open for further replies.
This is great news. Older applications (like music software etc) and the recent games would greatly benefit from increased performance on a single core. I have to disagree with gamerk316. Altough it's true that windows runs alot of processes in the background, your single core should never go beyond 5-8%, unless you are using a really old processor. No game/app will run slower because you have a 8% loss in processing power. the requirements for most apps/games are much lower than what the average user is running these days.
 
bah, my 3.11 has turbo mode, eat that. also wasnt C2D supposed to have a simlar thing, just not off, but really really down. although i fear that it will often be hard to get it to go into that mode, because of all the background threads, and say if your playing music. ect.
 
Interesting I just hope it works like they intend it to. Also it will be neat to see how much less heat will be generated with no voltage going to the turned off cores.
 
Interesting that fans applaud the 0.2 or 0.4 Ghz dynamic oclox. The same figures on AMD-Phenom are considered inadequate! The double standard says it's good if it's spintel. ACC (advanced clox calibration) manipulates Phenom cpu and allows much HIGHER oclox.

Spintel has a great trick to impair quad = off via "turbo". So that's oldtek rehash. This "marketing hype" is designed to trigger interest. Eventually more oldtek rehash will be revealed too, like in Larabe being a rehash of old P90's or whatever. I think we are seeing a rehash of 486 - and the joke is on all of you who blindly follow this.

Think about it - 4x dieshrink 486 cpu's - comes with built-in turbo mode. So if I connect 4x486 in a server farm, I could...? This is really getting funny. Just add mmx!

Let's see some UNBIASED test reports - o really - good luck - it will look wonderful until the real testers reveal the hidden bias.

AMD, meanwhile, is making innovative improvements and is preparing Shanghai server and Deneb desktop.
Turbo is not expected; although AMD's 486 did a great job in those days. But if AMD did a turbo, it would make 4cores like 8cores - for those times when you need that option.

I feel very comfortable with AMD's latest and planned releases; and the current position is practical in desktops. And they are the leader in servers w Opteron.

If you trust an7i7rus7 you should feel good about this too - I guess. (?) I am not sure. Does this make you feel comfortable? Do you trust what they say? How can you argue with speculation? - you need facts.

So I hear there is a new cpu - the rest is marketing hype - have you bought in yet? Not me! I just wish I could trust that the test reports had no built-in bias; and the reviewing site was not on the bribe payrole.

Let truth be known.
There is nothing to fear in 'truth'.
There is everything to fear in 'covert'.

Enough tall tales and hype and bedtime stories about turbo multithread cores and no oclox. Tell me the TRUTH. SHOW me.

cos I don't trust your rep. - netburst rehash? give it up!
 
programmers, (unless its assembly) dont write to individual cores or the cpu at all. The compiler does all that, so its all pointless... until programmers step up and start writing difficult code. .net, c/++, csharp, vb, etc dont let programmers pick a core to write to, so its pointless and will be for another 10 years.
 
[citation][nom]otheos[/nom]Great stuff! Too bad Core i7 will cost an arm and a leg (why not, AMD is lagging so much behind).[/citation]

AMD is lagging in only highend cpu's - and only a very few, for now.

AMD is leading in grafx.
AMD is leading in integrated video.
AMD is leading in Hi-Def support.
AMD is leading in the server market.
AMD is leading in platform concepts for pc's.
AMD is leading in on-cpu memory controller.
AMD is leading in HT.
AMD is leading in price/performance - that's all for you, btw!
AMD is leading in "innovation" - this is a foreign concept to spintel.
AMD is leading in being the beautiful underdog that everybody loves.
AMD is leading in becoming the new leader (again) that everybody loves.

Shanghai and Deneb 45nm are due very soon, and are reported on schedule.

Any news from the AMD camp at IDF?
 
stridervm 08/20/2008 7:07 AM:

Wierd, I think.... By the way Windows currently uses multiple cores, I think this will very much either cause :

- Negatively affect performance.
- Lots of BSODs.
- Will not work.

So how will that work? Windows already does try to divide the single applications performance between cores. Which means every core will be active. Wierd.
Click to expand...

Italics above are mine.

General Comment - not specific to stridervm:

Man! Unbelievable how many people just assume and criticize while not fully understanding how it will work. My idea is not that it will be without problems, but just how irrelevant prejudices should not even be posted. Just watch and see it coming on stage, THEN fire at it really is crappy.

 
Step 1: buy Nehalem
step 2: turn off turbo to overclock and permanently have higher clocks...

Awesome idea for the average user who will never overclock (97% of the population).
 
Really, this isn't any form of a special idea. They're basically saying "your CPU will run at the max we will allow and clock down when underused" same as always. Saying that it would "overclock" means its technically running at the max they will allow anyway. This isn't any form of a change at all, it's just marketing/advertising.
 
[citation][nom]ZootyGray[/nom]AMD is lagging in only highend cpu's - and only a very few, for now.AMD is leading in grafx.AMD is leading in integrated video.AMD is leading in Hi-Def support.AMD is leading in the server market.AMD is leading in platform concepts for pc's.AMD is leading in on-cpu memory controller.AMD is leading in HT.AMD is leading in price/performance - that's all for you, btw!AMD is leading in "innovation" - this is a foreign concept to spintel.AMD is leading in being the beautiful underdog that everybody loves.AMD is leading in becoming the new leader (again) that everybody loves.Shanghai and Deneb 45nm are due very soon, and are reported on schedule.Any news from the AMD camp at IDF?[/citation]
LOL!
Holy buddy, your wife work for AMD or something?
 
[citation][nom]jeb1517[/nom]So a TURBO feature turns OFF cores....am I the only one that sees the irony?[/citation]

Hmmm. Ironic? Perhaps someone can explain.

I can see how it's really useful for a nooby user trying to justify how and why they bought into a 386 concept that would cripple 3/4 of what they paid too much money for in the first place - when all they wanted was some good grafx performance in gaming - but didn't know that all they needed was a 2x-core with a FATI-Grafx card? That's called 20-20 wisdom.

Irony? It's ironic that oldtek gimmickry is used to hype a supposedly all new cpu - esp. when NOTHING is known about it. I think we are being set up.

Could it be they have energy issues with all cores?
Or perhaps you mean 'Why buy 4 cores to turn them off?
Or maybe those 4 cores don't exist - all just hypethread. Only exist in software? Virtual core? So turbo software switch? That would be ironic. Or something.

Please explain the irony part as you, or others, see it.
 
The problem with this feature in a dual-core is that the second core will never shut off. It's not like a game is the only thing happening. The drivers and OS still need to do work too, and that will either keep the second core busy or send us back 5 years to fast single-cores.

The technology only makes sense in a more than two core setting, not dual as implied by the author.
 
[citation][nom]designerfx[/nom]Really, this isn't any form of a special idea. They're basically saying "your CPU will run at the max we will allow and clock down when underused" same as always. Saying that it would "overclock" means its technically running at the max they will allow anyway. This isn't any form of a change at all, it's just marketing/advertising.[/citation]

kinda like...
o yeh, and my head is made of chocolate!

keep that hype spinnin', employboy
 
Those that read the articles written by "Anandtech.Com" probably read the one that said that "Nehalem" was first and foremost aimed to improve Multi-Threaded performance and Memory bandwidth wich is great for the server market were Intels current processors faces some stiff competition from AMD.

To me, this sounds like a perhaps simple attempt for Intel to improve "gaming" performance whit the "Nehalem" architecture and make it appear much stronger than the Penryn for the "average joe" that doesn't care how the CPU works.
As long as that "Core i7" is much faster than an equally clockrated "Core 2" it's all good, ignoring the fact that the speed is gained by secretly boosting the clockrate.

- Tywald
 
[citation][nom]grieve[/nom]LOL!Holy buddy, your wife work for AMD or something?[/citation]

Negative - not at all. But I wonder about the spintel employboys at certain obviously biased websites; or in test reports that do not present the entire story, and mislead.

I call for UNBIASED testing on a neutral test bed with no-slight-of hand trix, like changing drivers, specs, hardware, as a cover up to fudge results, or impair competition, or sway the market.

I support TRUTH and not market HYPE and hypnotic mob control. If I cause you to think, that will be fine. But their is nothing known about any of THIS - it's all spin. Nice job, if you can get it, and live with your conscience.

"I am a self-confessed AMD fanboy for ethical and personal reasons".

Drop into www.AMDzone forum threads and find out what's going on in the background.

Or buy into whatever you are told. I don't support monopoly and antitrust. Those work against us, no matter anything else. It is not news; it's old news. How long has this been going on? haha, sad story.

And the list I made, I believe is true. Prove me wrong, because I do not fear truth, and I can make mistakes. Do your own research at an unbiased source.

Truth will set you free. And AMD is not as far behind as you think - or have been led to believe.
 
hmm, I notice that on my Q6600 in some apps the core usage plays hot potatoe in a way that one core will shoot up in usage while another will drop to almost nothing and then they will trade. Isn't that almost as efficient in that no one single core creates a high hot spot on the die but rather the thermal is distributed? With 3 cores shut down and one doing all the work I would think the one core might reach its thermal limit quicker. Just an observation.
 
ZootyGray, I just think it's ironic because a feature being marketed as TURBO is actually cutting off cores and making the processor slower overall. It's speedstep 2.0 basically but for marketing reasons they decided to call it TURBO. Maybe hibernate or something would've been more appropriate?
 
[citation][nom]ZootyGray[/nom]Negative - not at all. But I wonder about the spintel employboys at certain obviously biased websites; or in test reports that do not present the entire story, and mislead. I call for UNBIASED testing on a neutral test bed with no-slight-of hand trix, like changing drivers, specs, hardware, as a cover up to fudge results, or impair competition, or sway the market.I support TRUTH and not market HYPE and hypnotic mob control. If I cause you to think, that will be fine. But their is nothing known about any of THIS - it's all spin. Nice job, if you can get it, and live with your conscience."I am a self-confessed AMD fanboy for ethical and personal reasons". Drop into www.AMDzone forum threads and find out what's going on in the background. Or buy into whatever you are told. I don't support monopoly and antitrust. Those work against us, no matter anything else. It is not news; it's old news. How long has this been going on? haha, sad story.And the list I made, I believe is true. Prove me wrong, because I do not fear truth, and I can make mistakes. Do your own research at an unbiased source. Truth will set you free. And AMD is not as far behind as you think - or have been led to believe.[/citation]
OMG DAMNED MONOPOLISTIC GAME PLAYERS OMFG.
No,they're not at the moment,but in general,I don't see the real advantage-as a gamer.I once owned a high end AMD platform based on the Athlon,but those days are of old.Truth be told,you are a fanboy.
Only thing I would buy from AMD is an ATi graphics card,atleast for the moment.Not because of hate,but because simply reality which you seem to lack.
[citation][nom]jeb1517[/nom]ZootyGray, I just think it's ironic because a feature being marketed as TURBO is actually cutting off cores and making the processor slower overall. It's speedstep 2.0 basically but for marketing reasons they decided to call it TURBO. Maybe hibernate or something would've been more appropriate?[/citation]
Does not make it slower.It depends on the program,slower single threaded applications are still there,so why not have the feature? My only real concern is how long it will take to activate the cores and use them,and if it will actively deciding this...or perhaps you can manually fine-tune(which I doubt).
 
What exactly is a "speed bin"? Do we have to oc in "bins" now? What if it falls outside a "bin"? What exactly does "jumping from 3 GHz to 3.2 GHz or even 3.4 GHz" mean? 400mhz is hardly noticable of an oc. What happened to q6600's 50+% 1.2+ghz oc through bios? In trying to implement those gemmicks, all they're doing in reality is tie down the hardware and limiting what you can do.
 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom