AMD, Intel Plotting Six-Core CPU Releases

[matt87_50]

[citation][nom]falchard[/nom]I bet AMD wins this round for a simple reason.AMD 6-core for $300.Intel 6-core for $1200.[/citation]

umm... yeah...

except the amd would be on a par with a $280 intel quad core...

thats all I've been seeing from amd since the core2duos I'm afraid...

 

[Tomtompiper]

But can it play Purble place?

 

[Da_Banig]

I still think my E8400 @ 3.8Ghz on air is godly. And I heard the 6cores will be 4 digits in price.

 

[imageraiter]

Yay, more technology advancements!

Hope I get to live long enough to see Quantum Computers being used at home.

 

[cookoy]

the more the merrier, you just can't have enough

 

[oblivionlord]

"I'd much rather have a 3-4ghz quad than a 2.4ghz six core ANY DAY."

It's still yet to be known what the ocing potential of the six core will be considering your mentioning a 4ghz quad which is obviously oced. Even if we look at 3ghz then you can say that alot of past cpu's that were 2.4ghz could give you simular ocing results as the 3ghz while oced.

In that case if the 2.4ghz six core can do 3.5ghz on air then Id rather have that than the 4ghz quad. Just the same as id rather have my current 4ghz q9550 over a 4.5ghz e8500.

 

[fafner]

I don't worry it because I have every confidence that the skilled programmers (probably not you) will find a way to make software multi-core friendly. Besides, I don't like to do one thing real fast. I'm an extreme multi-tasker, flipping through a dozen screens/tabs back and forth while encoding/burning a DVD.

LMAO. I rest my case.

 

[theubersmurf]

Somehow I doubt intel will release a single 6 core processor for 1k+ USD. Seems like a bad bet marketwise.

 

[fafner]

So that means we'll be getting super-speed processors at 20 and 10 nm, as opposed to the enthusiast-friendly 4 GHz overclocks which seem to be the norm nowadays? If so, I most certainly do stand corrected, and I will be one of the first in line to pick one up.

It is not a matter of we will or not. What I was saying is it is a path they SHOULD take instead of piling core upon core.

What the heck does "enthusiast-friendly 4 GHz overclocks" mean?

How should the burden of progress be distributed? You assert that it is being shifted from the chip makers to the software developers, and that this shifting is unfair to the developers and the public. Does progress lie solely with the chip makers?

Programmers should have to worry as little about the underlying architecture as possible so they can focus on functions and features, unless you are writing in assembly.

Since it is, as you say, a waste of time to multi-thread most applications, does that mean there is no room for innovation or creativity in software development?

Serioiusly, how did you come to that conclusion? You totally lost me there.

and btw, I said "most of the time it is a waste of time."

If the chip makers should be the only ones bearing this burden, how do we go about shifting it back?

Shrink the die. Start experimenting with new architectures.

I'm not against programmers having to optimize for performance. In fact, they always had to. What I'm saying is the consumer has much more to gain if the same effort is put into developing better chips. No matter how hard you try multi-threading is only going to get you so far.

While his comment may be considered immature, he did not use offensive language in his response in an attempt to belittle the author of the original comment.

Thank you, judge 4banger4life. If you are so determined to police this forum why don't you go give the idiot who posted the "ha ha ah you want single core 6GHZ" the same speech?

The fact that you post your comments on a site that is open to the public makes it the business of anyone who visits it. Your words would be taken more seriously if you left the anger out of them.

The value of my post was in is content. If you can't see that it is not my problem and I don't care.

 

[4banger4life]

[citation][nom]fafner[/nom]It is not a matter of we will or not. What I was saying is it is a path they SHOULD take instead of piling core upon core.What the heck does "enthusiast-friendly 4 GHz overclocks" mean?[/citation]

Thanks for the clarification. Tsnor posted a more helpful message describing the power issues the manufacturers are having with the newer processes.

"Enthusiast-friendly 4 GHz overclocks" refers to the fact that most enthusiasts can reach speeds of around 4 GHz with little trouble using current parts.

[citation][nom]fafner[/nom]Programmers should have to worry as little about the underlying architecture as possible so they can focus on functions and features, unless you are writing in assembly.[/citation]

So the burden of progress lies should lie with the chip manufacturers. That answers my question.

[citation][nom]fafner[/nom]Serioiusly, how did you come to that conclusion? You totally lost me there. and btw, I said "most of the time it is a waste of time."[/citation]

Your saying "most of the time it is a waste of time" is what led me to that conclusion. I did not quote you as saying coding multi-threading for all applications is worthless. "...most of the time..." is what led me to think "most applications" . Since I gathered from your response that multi-threaded programming wasn't the answer, I was hoping you would be able to shed some light on what kind of creativity/innovation might be possible on the software development side.

[citation][nom]fafner[/nom]Shrink the die. Start experimenting with new architectures. I'm not against programmers having to optimize for performance. In fact, they always had to. What I'm saying is the consumer has much more to gain if the same effort is put into developing better chips. No matter how hard you try multi-threading is only going to get you so far.[/citation]

That sounds quite reasonable, however, my question was how do we shift the burden back to the chip makers if that's where the focus needs to be?

[citation][nom]fafner[/nom]Thank you, judge 4banger4life. If you are so determined to police this forum why don't you go give the idiot who posted the "ha ha ah you want single core 6GHZ" the same speech?[/citation]

You're welcome, fafner. The reason I did not comment on HundredIslandsBoy's posts is because , so far as I am aware, he did not resort to insults, profanity and name-calling. Whether or not I agree with what he says is besides the point.

[citation][nom]fafner[/nom]The value of my post was in is content. If you can't see that it is not my problem and I don't care.[/citation]

I have no problem with the majority of the content in your posts. I believe that we should all be able to present our knowledge and opinions without having to worry about being called "noob" and being told to "STFU".

That you devalue the content in your posts with profanity and personal attacks is your problem and you should care. Unfortunately, that doesn't seem to be something I can convince you of.

 

[ahslan]

i just hope my AM2+ mobo supports these new processors when they come out...was hoping to upgrade to a quad core, but I might as well wait for these to come out (in any case, quad cores will be cheaper )...

 

[krazynutz]

[citation][nom]falchard[/nom]I bet AMD wins this round for a simple reason.AMD 6-core for $300.Intel 6-core for $1200.[/citation]
Not if a $275 Core i7 quad is still faster than a hexa AMD 😉

 

[descendency]

[citation][nom]krazynutz[/nom]Not if a $275 Core i7 quad is still faster than a hexa AMD[/citation]
Sad but true. If that 6 core chip doesn't out perform hyperthreaded-quad core chips, then it will have to drop price. $250 for a hexa core chip would be nice though.

 

[saran008]

AMD is going to rock once again in the price margin & VFM.

 

[steve sr1]

Though expensive,this kind of computer would be any gamer's dream. And you probably won't need to upgrade it for 4-5 years or maybe even longer till the games catch up.

 

[eyemaster]

I was surprised when I ran Prime95 and my system was still responsive on the Phenom 2. Yes, we need more cores. Everything runs smoothly the more cores you have, as long as you have the memory bus to support it.

 

[soccerplayer88]

I love how people are arguing over these cores. If anything they should be rejoicing over how the quads pricing will drop in the coming months.

Bigger picture fellas.

 

[gnesterenko]

[citation][nom]miloo[/nom]hopefully quad core price will drop a lot once the 6 core comes out~ hehe[/citation]
Of course they will. However, as with all tech, unless you pride yourself as an early adopter, I would stay away from hexa-cores until at least the 2nd stepping of the manufacturing process - same price, a few months wait, but much more headroom to OC (esp with the Black Edition versions)

My own plan is to score a mid-to-high grade ASUS motherboard with the 890FX (SB850) chipset on it when it comes out and drop in an AM3 Athlon X2 for what like $50? Have that hold me over a few months until AMD gives us a native 3GHz hexa CPU and have me a very nice in place upgrade.

I wander though, AMD has said the next socket will be the AM3R2, but its not clear yet whether existing AM3 CPUs will be usable in that socket or vice versa, if AM3R2 CPUs will be usable in AM3 mobos, albeit with limited features (kinda like the AM2+ > AM3 transition). If so, 890FX will be even more of a win, as I'll be able to drop in an even better CPU a year or two down the road without needing to nuke and rebuild my entire rig (are you listening, Intel?!)

"The views expressed here are mine and do not reflect the official opinion of my employer or the organization through which the Internet was accessed."

 

[hundredislandsboy]

That's right people. Keep talking about more cores. The speed war is over. How may times do we have to say the speed war is over? We, as the paying customers, dictate to the chip makers what WE want, not what programmers want. And if the programmers don't like it, fire them and hire some news ones!

How could anyone not like more cores when more cores just makes everything easier?

 

[jeffunit]

[citation][nom]viper_11[/nom]the benchmarks of this cpu showed that those six cores could harldy take over Intels quad core Xeon (with the new core i7 like).So i hope those new 6 core CPU have faster cores or at least have a price around i7 920 = $200-300[/citation]

I don't know what 'benchmarks' you are talking about, but for my real application my amd 940 is significantly faster than an i7 920. Perhaps you have drank too much intel optimized kool-aid?
If you let me pick the compiler and optimizations I can make almost any chip look faster or slower than another.

For my benchmark, I was using gcc 4.3.2 and 4.4.1 and the machine specific optimizations and -O3. What compiler was your
'benchmarks' using? Or do you even know?

 

[oblivionlord]

Lets not look at jst compiling... The i7 920 is signifigantly faster than the amd 940 is soo many aspects. This is very well common knowledge in many forums with even results showing how much faster the i7 920 is over the amd 965.

Perhaps with your optimizations, your 940 shows more results than the i7.

 

[dextermat]

Well i dunno but i feel like these guys not gonna last long enough in the market:

I'm waiting for 8 core + since my 2 core does the trick right now.

(then again i hope games/application will turn to multi-core soon)

 

[fafner]

Tsnor posted a more helpful message describing the power issues the manufacturers are having with the newer processes.

Thanks for the diss.

"Enthusiast-friendly 4 GHz overclocks" refers to the fact that most enthusiasts can reach speeds of around 4 GHz with little trouble using current parts.

Then how is it not better if they can give you a stock 6Ghz?

Your saying "most of the time it is a waste of time" is what led me to that conclusion.

I didn't think it would be necessary but let me me break it down for you. What you said implies that every program that is worth anything has already been written and the only way programmers can be creative is to optimize performance. Now, really, do I need to point out how ridiculous your statement is?

I was hoping you would be able to shed some light...

If that is really your intention maybe you should consider being less judgemental.

That you devalue...

What I wrote make sense. That's all I care. What you think is your problem. Feel free to continue being an idiot.

 

[fencer55]

I have an AMD Phenom II X4 940BE @ 3.0ghz, and that chip was much faster than my AMD 6500+ dual core at 2.9ghz. I noticed a huge increase in speed and my FPS in games doubled with that upgrade, even though the speed of the chips was nearly identical. Granted I play games that ARE optimized for multiple cores. I can also easily up the multiplier and run it my Phenom at 3.5ghz. If the six core cpus make my stuff even faster, and have an unlocked multiplier, I will upgrade again.

 

[4banger4life]

[citation][nom]fafner[/nom]Thanks for the diss.[/citation]

I merely pointed out that his comment was more helpful in that, after making his assertion, he wrote two sentences explaining his reasoning.

[citation][nom]fafner[/nom]Then how is it not better if they can give you a stock 6Ghz?[/citation]

I don't believe that I ever wrote that having high stock clock speeds was bad if it was practical. I did write that I thought very high stock speeds weren't currently practical for most people. If they do come out with stock 6GHz chips that are practical and affordable, that would be terrific.

[citation][nom]fafner[/nom]I didn't think it would be necessary but let me me break it down for you. What you said implies that every program that is worth anything has already been written and the only way programmers can be creative is to optimize performance. Now, really, do I need to point out how ridiculous your statement is?[/citation]

I did not write anything definitive concerning performance optimization. What I wrote was an explanation of how I arrived at a conclusion based upon what you wrote, in response to your question as to how I arrived at said conclusion.

[citation][nom]fafner[/nom]If that is really your intention maybe you should consider being less judgemental.[/citation]

The full sentence was "Since I gathered from your response that multi-threaded programming wasn't the answer, I was hoping you would be able to shed some light on what kind of creativity/innovation might be possible on the software development side."

I asked you if you could expand upon the possibility of software creativity/innovation. I did not deliver an ultimatum, so do not feel obliged to respond if you do not want to.

[citation][nom]fafner[/nom]What I wrote make sense. That's all I care. What you think is your problem. Feel free to continue being an idiot.[/citation]

What you wrote makes sense to yourself, and it is apparent to everyone that it is the only thing you care about.

While you make a lot of assertions, there is nothing wrong with that. However, if you do not back up those assertions with thorough explanations or solutions, you invite a higher possibility of questions from other readers. The more you explain your reasoning, the higher the chance others will understand or accept what you are trying to say.

What I think is indeed my problem, and has been as far back as I can remember. (^_^)

While it is very generous of you to grant me control over my state of being, you can keep the name-calling to yourself.