Sign in with
Sign up | Sign in
Your question

To all Conroe Fanboys!

Last response: in CPUs
Share
July 14, 2006 6:08:45 PM

SH*T; it's a whole step forward, totally another planet...
At leas this was my first impression after reading the benchmarks.

Yes that's unprecedented performance!However It's easy to be caught by enthusiasm (even for the THG's titles, pretty DRASTIC lately) and forget to do some simple arithmetics... That's how the a new, the most ferocious species of fanboys is born: the Conroe Fanboys!

Making some simple points one can see that:
There's nothing substantially new in this architecture. After some deep thoughts I think <<IT'S ALL ABOUT THE 65nm EFFECT>>. Core2 is substantially a K8, boosted with more logic units; here explained the extra performance (because 65nm allow this @ ~ the same die size) and it runs cooler still because of the 65nm process.

Conclusions: When AMD cores shrink to 65 nm they will:
1.BE able to incorporate more arithmetic & logics & wider busses than they currently do, call it K8 or K8L makes no diference @ that point.
2.Run cooler than Core2 because they almost match them on 90nm.
3.Potentially be cheaper after the price war(intel can't make it last

forever) because they do fine even with 512K L2.

AMD's transition to 65 nm (subsequently all the mentioned points) will happen much earlier than the 2008 or 2010 someone has mentioned. IMO, AMD is just round the corner, so keep your feet on the ground people; it's just a step forward not landing on moon :wink: .

More about : conroe fanboys

July 14, 2006 6:13:54 PM

Pssst... delete everything while you're still on time... :lol: 
Related resources
July 14, 2006 6:17:22 PM

Would be glad if you pointed out my missinformation.
July 14, 2006 6:22:36 PM

There's a sharp stick in my backyard...
July 14, 2006 6:23:33 PM

Quote:
SH*T; it's a whole step forward, totally another planet...
At leas this was my first impression after reading the benchmarks.

Terrible spelling and puctuation.


Quote:
Yes that's unprecedented performance!However It's easy to be caught by enthusiasm (even for the THG's titles, pretty DRASTIC lately) and forget to do some simple arithmetics... That's how the a new, the most ferocious species of fanboys is born: the Conroe Fanboys!

Soo even though Conroe is faster people shouldn't be excited by it? What you used to pay $1,000 for we can now get for $320? That makes me a fan!

Quote:

Making some simple points one can see that:
There's nothing substantially new in this architecture.

Incorrect

Quote:
After some deep thoughts I think <<IT'S ALL ABOUT THE 65nm EFFECT>>


Wrong, these things overclock like mothers. If it was all about 65-nm they would be barely overclockable.


Quote:
Conclusions: When AMD cores shrink to 65 nm they will:
1.BE able to incorporate more arithmetic & logics & wider busses than they currently do, call it K8 or K8L makes no diference @ that point.
2.Run cooler than Core2 because they almost match them on 90nm.
3.Potentially be cheaper after the price war(intel can't make it last

Perhaps your right. Where can I buy these AMD chips you speak of? Oh ya, they DO NOT EXIST. Sooo, that's why people will buy Conroe. It will be the best available technology.

Can some other people help me own this guy?
July 14, 2006 6:23:47 PM

Quote:
SH*T; it's a whole step forward, totally another planet...
At leas this was my first impression after reading the benchmarks.

Yes that's unprecedented performance!However It's easy to be caught by enthusiasm (even for the THG's titles, pretty DRASTIC lately) and forget to do some simple arithmetics... That's how the a new, the most ferocious species of fanboys is born: the Conroe Fanboys!

Making some simple points one can see that:
There's nothing substantially new in this architecture. After some deep thoughts I think <<IT'S ALL ABOUT THE 65nm EFFECT>>. Core2 is substantially a K8, boosted with more logic units; here explained the extra performance (because 65nm allow this @ ~ the same die size) and it runs cooler still because of the 65nm process.

Conclusions: When AMD cores shrink to 65 nm they will:
1.BE able to incorporate more arithmetic & logics & wider busses than they currently do, call it K8 or K8L makes no diference @ that point.
2.Run cooler than Core2 because they almost match them on 90nm.
3.Potentially be cheaper after the price war(intel can't make it last

forever) because they do fine even with 512K L2.

AMD's transition to 65 nm (subsequently all the mentioned points) will happen much earlier than the 2008 or 2010 someone has mentioned. IMO, AMD is just round the corner, so keep your feet on the ground people; it's just a step forward not landing on moon :wink: .


Why dont you go back to Gaico commercials?

/
July 14, 2006 6:26:31 PM

Quote:

Articles about the CORE microarchitecture (the basis for Conroe):

http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=274... <-- excellent overview of some of the microarchitectural improvements made in CORE.
http://www.anandtech.com/tradeshows/showdoc.aspx?i=2711 <-- coverage of the official unveiling of the Core microarchitecture at Spring IDF 2006, San Francisco.
http://arstechnica.com/articles/paedia/cpu/core.ars <-- A "big picture" post on ArsTechnica.
http://www.tomshardware.com/2005/12/04/top_secret_intel... <-- the seminal article here at Tom's covering some of the secret plans that lead to Core and some things to come, concerning 45nm.
http://www.realworldtech.com/page.cfm?ArticleID=RWT0309... <-- The original in-depth analysis of the Core architecture. This compares it with Yonah and Netburst. Anandtech and ArsTechnica base their architecture reviews on this one.
http://www.intel.com/technology/architecture/coremicro/... <-- Intel Marketing flash animation about some of the new features in Core. (link goes to page with Flash).


The "So what makes Core so new?" Section:

Here I will try to present some of the differences between CORE and previous architectures (in particular, the Core Duo, Intel's most current architecture you can buy today).

Conroe features a large 4MB cache, which is a first in consumer x86 products. This cache also has ultra-fine-grained power control, and an intelligent sharing mechanism between the two cores (in fact on some motherboards you will see that you can turn off one of the cores and dedicate the entire 4MB to one core, for more stablity in ultra high overclocks for single threaded apps. I can't provide proof for this statement yet, but you'll just have to wait and see if I'm right). That power control and sharing mechanism is not a walk in the park to design, and while many of the principles are present in Core Duo, this was a complete redesign with more aggressive power saving algorithms and more sensors and much more fine grained control, meaning smaller sections of the cache can be turned off more frequently.

Conroe also features memory disambiguation which allows most Loads to be speculatively executed before Stores. This is a HUGE benefit for some workloads. This is also not present in Core Duo.

Conroe has a 4-issue core, which means that 4 (or in some cases 5, see below) instructions are simultaneously pulled from the instruction cache and fed through the pipeline. Core Duo (and every other x86 product on the market) is 3-issue.

Conroe has macro and micro-ops fusion which allows certain combinations of x86 ops to be combined into one as they are pulled from the instruction cache. Macro-fusion allows the chip to save energy and have higher bandwidth, particularly during popular CMP-JMP operation pairs (if/then or switch statements in most programming languages decode into these instructions). It saves energy because it can literally do the same work with less physical transistor switching as the fused instruction moves through the pipeline. Core Duo only has micro-fusion.

Conroe has a completely redesigned 128-bit wide SSE engine that completes all SSE ops in 1 cycle. Core Duo has nothing close to this, in fact nothing on the market has anything close to this.

Conroe is 64-bit, Core Duo is 32-bit.

The SSE units also weren't the only thing expanded. The FPUs were also widened to 128-bit from 64-bit in Yonah, but since the FPUs share some of their resources with the SSE units the two are interrelated. The ALUs have also been widened to 64-bit from 32-bit which should provide a performance boost in integer operations as well.

Other improvements include increases in L1-L2 transfer bandwidth and the near tripling of bandwidth between the L1 cache and the rest of the core. A lot of that has to do with the widening of the internal data paths from 64-bits in Yonah to 128-bits in Core.

Other minor details include SSE4 support, although Intel may be looking to come up with a new marketing name. Related to the 4-wide issue rate is that all the buffers have been widened to handle the increase. The number of ports has also increased from 6 from 5 in Yonah to better organize the additional execution units and the division of capabilities (Vshuffle, etc.) for each execution unit has also changed somewhat to avoid conflicts. The instruction fetch size has also increased from 128-bit to 160-bit (possibly higher) in order to feed the 4 decoders.


Ownage.
July 14, 2006 6:29:39 PM

AMD may not be "far behind" but they are indeed behind NOW, and that is a big difference for any market.

Establishing an advantage is pivotal in technologies, in an industry in which all functions of computer design (including software) tie in some part to a processors capabilities, a two year, or even single year, lull puts one manufacturer at a severe disadvantage in both liscening and R&D.

Even if there's little "fundamentally new" (which, I disagree with, if it wasn't new, hypothetical max. capabilities would be the same as any last-gen processor) about Intel's latest line, but they are still fundamentally more powerful devices and even a small advantage in the computer world can become an unsurmountable gap in little to no time (especially in an industry with significant evolution every three years).
July 14, 2006 6:34:52 PM

You asked for it.
There you go:
Quote:

There's nothing substantially new in this architecture. After some deep thoughts I think <<IT'S ALL ABOUT THE 65nm EFFECT>>. Core2 is substantially a K8, boosted with more logic units;

Following the same logic, then K8 is substantially a 486, boosted with more logic units, thanks to its 90nm EFFECT. :lol: 

Quote:
here explained the extra performance (because 65nm allow this @ ~ the same die size) and it runs cooler still because of the 65nm process.

Nope, in fact Netburst, on the same 65nm process, runs as hot as hell.
NGMA is an architecture which expands on several energy saving features which debuted on NOTEBOOKS (hence, power / heat constrained systems) with the Pentium M.


Quote:
Conclusions: When AMD cores shrink to 65 nm they will:
1.BE able to incorporate more arithmetic & logics & wider busses than they currently do, call it K8 or K8L makes no diference @ that point.

Sure they WILL BE ABLE.
But IF and WHEN this is going to happen, it's anybody's guess.

Quote:
2.Run cooler than Core2 because they almost match them on 90nm.

Might be, but again, this is just your speculation.

Quote:
3.Potentially be cheaper after the price war(intel can't make it last
forever) because they do fine even with 512K L2.

Why do you think AMD is at any advantage in a price war?
Intel is a bigger company (hence it has more capital), has way more production facilities, and a much much better and cost effective process.

Quote:
AMD's transition to 65 nm (subsequently all the mentioned points) will happen much earlier than the 2008 or 2010 someone has mentioned.

Actually it should happen end of 2006 / beginning of 2007.
But this is just another point proving that you're misinformed.
However, a transition to 65nm does not mean the debut of a new architecture, unlike what you may think.
July 14, 2006 6:35:48 PM

Quote:

Articles about the CORE microarchitecture (the basis for Conroe):

http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=274... <-- excellent overview of some of the microarchitectural improvements made in CORE.
http://www.anandtech.com/tradeshows/showdoc.aspx?i=2711 <-- coverage of the official unveiling of the Core microarchitecture at Spring IDF 2006, San Francisco.
http://arstechnica.com/articles/paedia/cpu/core.ars <-- A "big picture" post on ArsTechnica.
http://www.tomshardware.com/2005/12/04/top_secret_intel... <-- the seminal article here at Tom's covering some of the secret plans that lead to Core and some things to come, concerning 45nm.
http://www.realworldtech.com/page.cfm?ArticleID=RWT0309... <-- The original in-depth analysis of the Core architecture. This compares it with Yonah and Netburst. Anandtech and ArsTechnica base their architecture reviews on this one.
http://www.intel.com/technology/architecture/coremicro/... <-- Intel Marketing flash animation about some of the new features in Core. (link goes to page with Flash).


The "So what makes Core so new?" Section:

Here I will try to present some of the differences between CORE and previous architectures (in particular, the Core Duo, Intel's most current architecture you can buy today).

Conroe features a large 4MB cache, which is a first in consumer x86 products. This cache also has ultra-fine-grained power control, and an intelligent sharing mechanism between the two cores (in fact on some motherboards you will see that you can turn off one of the cores and dedicate the entire 4MB to one core, for more stablity in ultra high overclocks for single threaded apps. I can't provide proof for this statement yet, but you'll just have to wait and see if I'm right). That power control and sharing mechanism is not a walk in the park to design, and while many of the principles are present in Core Duo, this was a complete redesign with more aggressive power saving algorithms and more sensors and much more fine grained control, meaning smaller sections of the cache can be turned off more frequently.

Conroe also features memory disambiguation which allows most Loads to be speculatively executed before Stores. This is a HUGE benefit for some workloads. This is also not present in Core Duo.

Conroe has a 4-issue core, which means that 4 (or in some cases 5, see below) instructions are simultaneously pulled from the instruction cache and fed through the pipeline. Core Duo (and every other x86 product on the market) is 3-issue.

Conroe has macro and micro-ops fusion which allows certain combinations of x86 ops to be combined into one as they are pulled from the instruction cache. Macro-fusion allows the chip to save energy and have higher bandwidth, particularly during popular CMP-JMP operation pairs (if/then or switch statements in most programming languages decode into these instructions). It saves energy because it can literally do the same work with less physical transistor switching as the fused instruction moves through the pipeline. Core Duo only has micro-fusion.

Conroe has a completely redesigned 128-bit wide SSE engine that completes all SSE ops in 1 cycle. Core Duo has nothing close to this, in fact nothing on the market has anything close to this.

Conroe is 64-bit, Core Duo is 32-bit.

The SSE units also weren't the only thing expanded. The FPUs were also widened to 128-bit from 64-bit in Yonah, but since the FPUs share some of their resources with the SSE units the two are interrelated. The ALUs have also been widened to 64-bit from 32-bit which should provide a performance boost in integer operations as well.

Other improvements include increases in L1-L2 transfer bandwidth and the near tripling of bandwidth between the L1 cache and the rest of the core. A lot of that has to do with the widening of the internal data paths from 64-bits in Yonah to 128-bits in Core.

Other minor details include SSE4 support, although Intel may be looking to come up with a new marketing name. Related to the 4-wide issue rate is that all the buffers have been widened to handle the increase. The number of ports has also increased from 6 from 5 in Yonah to better organize the additional execution units and the division of capabilities (Vshuffle, etc.) for each execution unit has also changed somewhat to avoid conflicts. The instruction fetch size has also increased from 128-bit to 160-bit (possibly higher) in order to feed the 4 decoders.


Ownage.

Can't you figure out it's just commercial S-H-I-T; You can find tons of it about the Pentium Pro & up!
July 14, 2006 6:40:14 PM

If you think it's just speculation, remember the transition from 18 to 13 microns that the Willamette to Northwood brought: 2 to 3.06GHz running even cooler. This is how you can make an idea of what the 90-65nm brings :wink:
July 14, 2006 6:45:46 PM

:roll:
.. i leave you in the hands of the real Intel fanboys.. :lol: 
Have fun!
July 14, 2006 6:46:56 PM

normal sane socialized people buy whats best for the money.

If those AMD's you speak of actually existed people would buy those. Thats competition. People buy whatever is the better product for the price.

You're hypocrisy is astounding though. You title a thread "all conroe fanboys" in a attempt to bash fanboys but yet here you are being a fanboy.

Children like you really need to consider going outside more. i would recommend basketball if your tall. Maybe try rollerblading. Go outside more you really need the vit D.

Quote:
SH*T; it's a whole step forward, totally another planet...
At leas this was my first impression after reading the benchmarks.

Yes that's unprecedented performance!However It's easy to be caught by enthusiasm (even for the THG's titles, pretty DRASTIC lately) and forget to do some simple arithmetics... That's how the a new, the most ferocious species of fanboys is born: the Conroe Fanboys!

Making some simple points one can see that:
There's nothing substantially new in this architecture. After some deep thoughts I think <<IT'S ALL ABOUT THE 65nm EFFECT>>. Core2 is substantially a K8, boosted with more logic units; here explained the extra performance (because 65nm allow this @ ~ the same die size) and it runs cooler still because of the 65nm process.

Conclusions: When AMD cores shrink to 65 nm they will:
1.BE able to incorporate more arithmetic & logics & wider busses than they currently do, call it K8 or K8L makes no diference @ that point.
2.Run cooler than Core2 because they almost match them on 90nm.
3.Potentially be cheaper after the price war(intel can't make it last

forever) because they do fine even with 512K L2.

AMD's transition to 65 nm (subsequently all the mentioned points) will happen much earlier than the 2008 or 2010 someone has mentioned. IMO, AMD is just round the corner, so keep your feet on the ground people; it's just a step forward not landing on moon :wink: .
a b à CPUs
July 14, 2006 6:47:03 PM

Conroe will be the best tech available. Intel has really stepped up their game. There isn't really any reason why Intel should be 2nd here. They are the chip giant. Intel competed quite nicely with their inferior prescott chip. Intel is going to have a one year advantage in the 65nm process over AMD.
July 14, 2006 6:49:55 PM

Quote:
There's a sharp stick in my backyard...


Mind if I borrow that for a minute DeusEx?

Quote:
Can some other people help me own this guy?


I just shovel the stick up his ass, I hope I assisted you in owning him lol
July 14, 2006 6:53:19 PM

I think he may not be the only one "misinformed". Why would AMD continue to market a slower (How much slower? I'll leave that up to you decide, :wink: ) processor at a much higher premium? I'm still very confused why AMD would try to do this. It doesn't seem like a smart move.

So, AMD will just sell at the same, albeit outrageous, prices even after Intel brings a faster processor to the table?

You know good and well AMD will lower prices to match the price/performance of Intel. Intel did with their dual-cores when they saw the speed of the X2's. They saw they had a slower product and realized the only way for people to even think about buying them would be if the prices were lower.

~Ibrahim~
July 14, 2006 6:57:09 PM

Lol ... so you say the best cpus are the ones we can expect in the future ... tell me something new :roll:

AMD are behind now, that's a fact just look at all the benchmarks around the globe. Maybe AMDs transistion to 65nm might do some tricks, but there are absolutely no indications of this, at this point, it's pure speculations taken out of thin air. Nothing I can base my selection of hardware on, there's always something better in the horizon, but I want something better here and now.

And conroe running cooler just bcs of 65nm, that's garbage talk. Everyone with a little knowledge to semiconductors and electronics knows that the internal design are just as important as the size of technology used for making them.

For the record, I have a 3800+ atm and have had AMD since my k4-450, but one thing's sure, Im migrating to Intel platform once summer holidays are over ... that's also a fact.
July 14, 2006 7:02:08 PM

Ikaj, yes they will lower prices, and they are known and significant if on their own, but even at the new prices they are grossly disproportionate when compared to competing Intel chips.
It does mean that they are still worthy to consider, but most buyers willing to fork over hard cash for a gaming or production rig will be hard pressed not to select Intel if price/performance and thermal outputs are considered.
July 14, 2006 7:03:43 PM

Quote:
I think he may not be the only one "misinformed". Why would AMD continue to market a slower (How much slower? I'll leave that up to you decide, :wink: ) processor at a much higher premium? I'm still very confused why AMD would try to do this. It doesn't seem like a smart move.

So, AMD will just sell at the same, albeit outrageous, prices even after Intel brings a faster processor to the table?

You know good and well AMD will lower prices to match the price/performance of Intel. Intel did with their dual-cores when they saw the speed of the X2's. They saw they had a slower product and realized the only way for people to even think about buying them would be if the prices were lower.

~Ibrahim~


Where did I say that AMD wouldn't lower their prices? They will, but not enough. In order for a new system builder (that is sane) to consider AMD as an option they would have to lower the FX-62 down to the price of $300 for it to compete with it's rival the E6600 (which is Intel's midrange, not top dawg product).

AMD won't keep it's prices the same, but it sure as heck won't sell an FX-62 for $300.
July 14, 2006 7:05:12 PM

Quote:
normal sane socialized people buy whats best for the money.

If those AMD's you speak of actually existed people would buy those. Thats competition. People buy whatever is the better product for the price.

You're hypocrisy is astounding though. You title a thread "all conroe fanboys" in a attempt to bash fanboys but yet here you are being a fanboy.

Children like you really need to consider going outside more. i would recommend basketball if your tall. Maybe try rollerblading. Go outside more you really need the vit D.

SH*T; it's a whole step forward, totally another planet...
At leas this was my first impression after reading the benchmarks.

Yes that's unprecedented performance!However It's easy to be caught by enthusiasm (even for the THG's titles, pretty DRASTIC lately) and forget to do some simple arithmetics... That's how the a new, the most ferocious species of fanboys is born: the Conroe Fanboys!

Making some simple points one can see that:
There's nothing substantially new in this architecture. After some deep thoughts I think <<IT'S ALL ABOUT THE 65nm EFFECT>>. Core2 is substantially a K8, boosted with more logic units; here explained the extra performance (because 65nm allow this @ ~ the same die size) and it runs cooler still because of the 65nm process.

Conclusions: When AMD cores shrink to 65 nm they will:
1.BE able to incorporate more arithmetic & logics & wider busses than they currently do, call it K8 or K8L makes no diference @ that point.
2.Run cooler than Core2 because they almost match them on 90nm.
3.Potentially be cheaper after the price war(intel can't make it last

forever) because they do fine even with 512K L2.

AMD's transition to 65 nm (subsequently all the mentioned points) will happen much earlier than the 2008 or 2010 someone has mentioned. IMO, AMD is just round the corner, so keep your feet on the ground people; it's just a step forward not landing on moon :wink: .


I'm not saying "Don't buy Conroe", Just don't put it on the Black & White saying AMD is dead.
July 14, 2006 7:05:28 PM

Guys just leave m25 alone. He dosent know what he's saying, he's smoking dope and he's in his own little world right now.
July 14, 2006 7:06:16 PM

you've got him back peddling TC.

AMD is not dead, why is it thought that one architexture will destroy a company?

im pretty sure our slogan isnt "one architexture to rule them all" or is it 8O
July 14, 2006 7:06:51 PM

Quote:
Conclusions: When AMD cores shrink to 65 nm they will:


You got that right. When.
I like AMD processors, but intel have the upperhand for now.
The fact of the matter is, that if you are about to buy a new cpu / pc you want core2duo.
July 14, 2006 7:08:17 PM

Quote:
There's nothing substantially new in this architecture.

Dude, you have issues. Conroe is a totally new architecture. 65nm has little to do with it.

Plues, Intel went with lower clock speeds, so it's not a Netburst :lol:  If the architecture was not different, we would see 5GHz+ (at least) CPUs.

Read the whole thing I stated @ the beginning; It's just K8 with larger busses & more units. If this is really new & sensational?!!!...
July 14, 2006 7:12:13 PM

Quote:
That's how the a new, the most ferocious species of fanboys is born: the Conroe Fanboys!

Fanboy is a term for someone with a lack of knowledge, typicaly representing a brand. Conroe is not a brand it is a product, there is a HUGE difference. There is nothing negative to like and represent what is the best, Conroe.

Quote:
Making some simple points one can see that:
There's nothing substantially new in this architecture.
This is not true. Why don't you read more before giving such rediculous statements!

Quote:
After some deep thoughts I think <<IT'S ALL ABOUT THE 65nm EFFECT>>.
Another BS! What about 65nm Netburst chips? :roll:

Quote:
Core2 is substantially a K8,
Another BS! You have a great lack of knowledge about the things you are talking about. Don't shame your self with such statements.
Quote:
boosted with more logic units; here explained the extra performance
The improved execution units and the added one are boosting Core2 performance, but there are another architecture technologies that are boosting it also and without them the extra execution unit will be useless.
Quote:
(because 65nm allow this @ ~ the same die size) and it runs cooler still because of the 65nm process.
Yes it runs cooler becouse 65nm. It runs cooler also becouse of its architecture. See the TDP of Netburst before you make another stereotypic conclussion.

Quote:
Conclusions: When AMD cores shrink to 65 nm they will:

Your logic is absolutely incorrect:
Quote:
1.BE able to incorporate more arithmetic & logics & wider busses than they currently do, call it K8 or K8L makes no diference @ that point.

No, they can make the same with 90nm or 130nm, but the price of their chips will be much more higher.
Quote:
2.Run cooler than Core2 because they almost match them on 90nm.

No, we don't know if 65nm K8 will run cooler than Core2. 90nm SOI2 K8 is not even close to 65nm "bulk" Core2. AthlonFX-62 2.8GHz has TDP 125W, Core 2 Extreme 2.93GHz(5% clocked higher) has TDP 75W.
Quote:
3.Potentially be cheaper after the price war(intel can't make it last forever) because they do fine even with 512K L2.

No, 300mm 65nm SOI3 waffers are more expencive than 300mm 65nm "bulk". AMD are at the beggining of their 65nm production process and their yields needs to be improved. That is not the case with 65nm Intel "bulk" production process, they have their 65nm production matured.
With 512kB or with 1MB L2, Athlon64 is being owned by same clocked Core2 in every benchmark known to mankind.

Quote:
AMD's transition to 65 nm (subsequently all the mentioned points) will happen much earlier than the 2008 or 2010 someone has mentioned.
At the same time Intel will have 32nm.
Quote:
IMO, AMD is just round the corner, so keep your feet on the ground people; it's just a step forward not landing on moon :wink: .
BS, you are FANBOY
July 14, 2006 7:14:27 PM

Quote:
SH*T; it's a whole step forward, totally another planet...
At leas this was my first impression after reading the benchmarks.

Yes that's unprecedented performance!However It's easy to be caught by enthusiasm (even for the THG's titles, pretty DRASTIC lately) and forget to do some simple arithmetics... That's how the a new, the most ferocious species of fanboys is born: the Conroe Fanboys!

Making some simple points one can see that:
There's nothing substantially new in this architecture. After some deep thoughts I think <<IT'S ALL ABOUT THE 65nm EFFECT>>. Core2 is substantially a K8, boosted with more logic units; here explained the extra performance (because 65nm allow this @ ~ the same die size) and it runs cooler still because of the 65nm process.

Conclusions: When AMD cores shrink to 65 nm they will:
1.BE able to incorporate more arithmetic & logics & wider busses than they currently do, call it K8 or K8L makes no diference @ that point.
2.Run cooler than Core2 because they almost match them on 90nm.
3.Potentially be cheaper after the price war(intel can't make it last

forever) because they do fine even with 512K L2.

AMD's transition to 65 nm (subsequently all the mentioned points) will happen much earlier than the 2008 or 2010 someone has mentioned. IMO, AMD is just round the corner, so keep your feet on the ground people; it's just a step forward not landing on moon :wink: .


You will all have to forgive me, but this has to be said.. "ARE YOU QUIET POSSIBLY MENTALY RETARDED?"

Here is an article that compairs K8 vs Core:
http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=274...

Read up and return once you actually know what the hell your speaking about.


Btw: From what ive read the last few months AMD will do no such thing as add more FPUs... AMD needs this shrink to enable higher clocks to compensate for the fact that SOI is utter garbage when you overclock. AMDs cpu`s suffer from other major fault, the memory controller only likes low latency ram. Hence the move DDR2 didnt give any benifits to speak even if the bandwith increased. The only thing AMD can do to compensate for the higher latencys is to increase clocksspeeds wich in turn raises the heat... go figure
July 14, 2006 7:16:00 PM

Quote:
I think he may not be the only one "misinformed". Why would AMD continue to market a slower (How much slower? I'll leave that up to you decide, :wink: ) processor at a much higher premium? I'm still very confused why AMD would try to do this. It doesn't seem like a smart move.

So, AMD will just sell at the same, albeit outrageous, prices even after Intel brings a faster processor to the table?

You know good and well AMD will lower prices to match the price/performance of Intel. Intel did with their dual-cores when they saw the speed of the X2's. They saw they had a slower product and realized the only way for people to even think about buying them would be if the prices were lower.

~Ibrahim~


while i think this statement is true, Intel could and did afford to drop prices buy ALOT, I dont think amd can afford to drop prices to low for to long of a period specially for your price for performance senario (that will make an fx62 HOW MUCH ?". I however am not a fanboy of either company as long as i get the "spoils" of this war.
July 14, 2006 7:18:22 PM

Nice Flame.

i like how you use facts instead of just running your mouth and not your brain.
July 14, 2006 7:29:35 PM

Quote:
..., but most buyers willing to fork over hard cash for a gaming or production rig will be hard pressed not to select Intel if price/performance and thermal outputs are considered.


Anybody who games knows that the CPU is almost insignificant compared to the GPU...

CPU Chart

If you look, the processors vary slightly. Yeah, a faster processor will get you an extra 1-30 FPS, but nothing compared to going from one GPU to another. The money spent on a faster CPU is easily spent on better cooling for overclocking and investing in a faster GPU.

Intel's enormus captial must've been a large player when they lowered the Pentium D prices. I wonder how well and how AMD will cope.

Yes, I agree Intel is moving forward and AMD has a tough few months ahead of them, but they will pull through, sooner or later.

(I'm trying to figure out this 65nm objectively)

Didn't we see lower cooling and power requirements when GPU's moved from 130/110nm to 90nm? Wouldn't CPU's have a similar pattern?

With lower nm we see smaller dies and thus lower temperatures and less power needed, right?

Intel's 65nm technology roughly doubles transistor density compared to the previous generation, and delivers industry-leading performance and power-reduction features.


That is direct from Intel's site. Note the last words. They don't directly attribute this to 65nm, but it is mentioned in the same sentence as 65nm leading the consumer to believe that 65nm is part of the equation.

Intel also mentions this, as a bullet in a presentation describing the logic of 65nm:

65 nm derivative technology for ultra-low power products.

In this same presentation, they DO attribute power-reduction, but not to 65nm, but to smaller transistor gate length which are now at 35nm. The transistors have also improved leakeage performance, the major player of those hot and heavy Prescotts.

~Ibrahim~
July 14, 2006 7:31:42 PM

Hey, you copied my reply point by point! :evil: 
:lol:  :wink:
July 14, 2006 7:32:30 PM

Quote:
Anybody who games knows that the CPU is almost insignificant compared to the GPU...

HA!!! Did he really just say that? Let me double check:

Quote:
Anybody who games knows that the CPU is almost insignificant compared to the GPU...

HA! HE REALLY DID!!! ROFLCOPTER!!!

July 14, 2006 7:42:52 PM

Hi,
I'm an electronic engineer and i've studied microprocessor architecture, and i'm impressed when some of you talk about architectures of microprocessor with tha ease.
Conroe is proving that is a good CPU and i always prefered AMD because in my opinion and in very other engineers that i know AMD had (until now) better design and solutions for better performance at lower clocks.
Maybe you (some) are very wiseman... or just people that do not know what are talking about. sorry, but is true.
July 14, 2006 7:44:53 PM

I think I can organize some facts, objectively hopefully:

1. Conroe, Core 2 Duo, C2D is currently faster than AMD's current lineup. How much faster is harder to answer.
2. Conroe has achieved lower wattage and heat claims. This cannot be directly attributed to 65nm, but we cannot ignore 65nm either. This is because of bits and pieces that have been stolen from the Mobile Architecture, low leakage and reduced transistor gate lenghths being major players.
3. AMD will reduce prices similar to what Core will debut at, yet it is still undetermined what the actual street price will be and whether it will still be competetive to a degree.
4. Intel has already debuted it's 65nm lineup with Yonah and the 9xx series. And those were not mobile friendly components, in realtion to heat and power. The Xeon Dempsey CPU's are also 65nm, but those are intended for the Server Market.
5.AMD is set to release their 65nm lineup in the dying months of this year.

If I am wrong, please correct me. If I'm favoring a side, please correct me.

I may have made a small hyperbole on that statement, but changing GPU's will certainly hold a MUCH MUCH MUCH higher FPS increase in gaming that switching CPUs. Yes, CPU's play a part, but I would much rather spend my money on a faster GPU than another CPU.

~Ibrahim~
July 14, 2006 7:50:27 PM

Quote:
Hi,
I'm an electronic engineer and i've studied microprocessor architecture, and i'm impressed when some of you talk about architectures of microprocessor with tha ease.
Conroe is proving that is a good CPU and i always prefered AMD because in my opinion and in very other engineers that i know AMD had (until now) better design and solutions for better performance at lower clocks.
Maybe you (some) are very wiseman... or just people that do not know what are talking about. sorry, but is true.



noob...


study more: Hardwaresecrets.com
sorry, even you're an Electronic Engineer, i disagree in your opinion.[/i]
July 14, 2006 8:08:24 PM

The HTML code is good, just a bad website. They've either removed the link or moved it without a redirect link...

~Ibrahim~
July 14, 2006 8:14:52 PM

Quote:

noob...

[/i]

yeah you know a lot...
July 14, 2006 8:18:37 PM

lol...didn't notice that. The HTML software was trying to be helpful and close that HTML tag that it thought you didn't close, but you did. And it puts it right before you click Submit and your cursor is almost always at the bottom of your post by then thus nullifying the need.
~Ibrahim~
July 14, 2006 8:25:09 PM

And you do? You have not proved a single thing in this thread. Just because you come here saying that you are an engineer we should believe what you say? You come here with no information at all, just saying I am an engineer! Listen to me! Sure, did you know that I am an engineer and a brain surgeon on the weekends? At night I am a pimp too. Don't say what you are, try to prove it next time.
July 14, 2006 8:29:44 PM

Quote:
And you do? You have not proved a single thing in this thread. Just because you come here saying that you are an engineer we should believe what you say? You come here with no information at all, just saying I am an engineer! Listen to me! Sure, did you know that I am an engineer and a brain surgeon on the weekends? At night I am a pimp too. Don't say what you are, try to prove it next time.

Did i say i know something?
Architecture is so complex that you have no idea and u throw away your opinions just as they were the truth.
Do you know what is forwarding in a pipeline? This is the most basic thing to improve performance in a pipeline technology and it was introduced in pentium (ups lol, in 486 DX2/3/4). I give u some weeks to study that and then you can say what you want.
July 14, 2006 8:32:29 PM

Bottom line =

Whether X2 or Conroe, your going to get what you pay for.
AMD will still win the performance per dollar war as they always have.
If Conroe 2.4 is $400 and matches or beats the performance of the FX-62, then the FX-62 will be $350. Watch and weep kiddies.
July 14, 2006 8:32:33 PM

nah AMD isn't dead otherwise intel would have a monoply and they would get split again forming AMD2 anyway.

But people buy whats best for their money. Conroe is appealing to me because of reduced noise and increased performance like the amd for less. I was going to get an AMD x-2 4600+ but then i heard about conroe and the prices. Then i saw the performance of it.

wouldn't it be logical for me to buy the better processor for less?

Quote:
normal sane socialized people buy whats best for the money.

If those AMD's you speak of actually existed people would buy those. Thats competition. People buy whatever is the better product for the price.

You're hypocrisy is astounding though. You title a thread "all conroe fanboys" in a attempt to bash fanboys but yet here you are being a fanboy.

Children like you really need to consider going outside more. i would recommend basketball if your tall. Maybe try rollerblading. Go outside more you really need the vit D.

SH*T; it's a whole step forward, totally another planet...
At leas this was my first impression after reading the benchmarks.

Yes that's unprecedented performance!However It's easy to be caught by enthusiasm (even for the THG's titles, pretty DRASTIC lately) and forget to do some simple arithmetics... That's how the a new, the most ferocious species of fanboys is born: the Conroe Fanboys!

Making some simple points one can see that:
There's nothing substantially new in this architecture. After some deep thoughts I think <<IT'S ALL ABOUT THE 65nm EFFECT>>. Core2 is substantially a K8, boosted with more logic units; here explained the extra performance (because 65nm allow this @ ~ the same die size) and it runs cooler still because of the 65nm process.

Conclusions: When AMD cores shrink to 65 nm they will:
1.BE able to incorporate more arithmetic & logics & wider busses than they currently do, call it K8 or K8L makes no diference @ that point.
2.Run cooler than Core2 because they almost match them on 90nm.
3.Potentially be cheaper after the price war(intel can't make it last

forever) because they do fine even with 512K L2.

AMD's transition to 65 nm (subsequently all the mentioned points) will happen much earlier than the 2008 or 2010 someone has mentioned. IMO, AMD is just round the corner, so keep your feet on the ground people; it's just a step forward not landing on moon :wink: .


I'm not saying "Don't buy Conroe", Just don't put it on the Black & White saying AMD is dead.
a b à CPUs
July 14, 2006 8:34:59 PM

To be honest, I encourage strong competition between AMD and Intel. It doesn't matter which company is "on top," just that they are in close competition. This encourage both strong market forces to keep prices down as well as competitive forces to make better products. We stand to see better development because there is a strong competitor willing to take the lead for the next leg of the race if the other guy should happen to fall slightly behind (hey, what can I say? I'm a bit influenced by the Tour de France).

I will continue to buy BOTH products, and will remain processor-agnostic as long as they both work in concurrence with regards to compatibility. Granted, my next desktop processor purchase will be a Conroe, but not long after that, I will probably need a new server, and may be more than happy to look at this 4x4 option that AMD is developing (Hypertransport is key here). If I get cheap, quality hardware, all the better for me (or any consumer). And AMD and Intel are encouraged to develop better hardware when competing for my money.
July 14, 2006 8:39:12 PM

to bad ppl dont understand that, its about watever product has better performance at the moment, as of rite now core 2 duo does, thats all that matters
July 14, 2006 8:43:57 PM

Exactly. I agree completely with the last two posts...

~Ibrahim~
July 14, 2006 9:04:19 PM

Your idiocy is second only to BM.
July 14, 2006 9:09:43 PM

Did I say anything myself about CPUs? Look through every one of my threads. You will find out one thing, I have NEVER said anything about the architecture at all. I don't know about it. My point is you come in here saying that you are an engineer and that AMD has a better architecture yet provide no facts for it other than you are an engineer and studied it.

Forwarding reduces stalling.
July 14, 2006 9:13:59 PM

All I can say about this guy's post is WAAAAAAAAAAAAAAAAAAAA!!!!!

Listen, I was prepared to buy AMD on my next build until news of the Conroe came out. I may still buy AMD if they lower their prices enough. BUT and a big BUT, how can AMD lower the price of the FX-62 lower than the $316 of the 2.4 ghz Conroe? You're talking $700 lower.

I am a fan of any processor/manufacturer that can come up with a product to meet my needs and make it for a price I can afford.

I don't want to see either Intel or AMD fail. Failure of either one means higher prices for the products we enjoy.

Face it, I don't see how Intel can justify the high price of their 9XX extreme processor or how AMD can justify the high price of their top FX processors when the Conroe comes out.

So, to the baby who started this post, accept the Conroe for what it is, "competition" and delight in the lowering of prices for the AMD chips you love. Go out and buy one of them even.

hball
!