AMD Sitch

[hUMANbEATbOX]

dude, you can spout all the math you want, but nothing you have typed so far in this thread has shown that AMD has a killer cpu in its hands. they could very well have, but the proof is in the pudding. none of your math means ANYTHING without actual hardware backing it up.

i can't even bring myself to read all of what you have posted. to me it seems like babble, a smokescreen for reality.

and calling someone a moron is against the rules.

so must i go on?
Havent you people ever Studied Gauss?
cause it seems i´m teaching you people how to read binary.....

you should not be so condescending. not everyone is a mathematician, and it is unfair to expect everyone to be on the same page as you.

i still can't figure out what all of your math is trying to say......i don't see how it relates to real world performance of a cpu, but maybe that's just me.

 

[savageseb]

dude, you can spout all the math you want, but nothing you have typed so far in this thread has shown that AMD has a killer cpu in its hands. they could very well have, but the proof is in the pudding. none of your math means ANYTHING without actual hardware backing it up.

i can't even bring myself to read all of what you have posted. to me it seems like babble, a smokescreen for reality.

and calling someone a moron is against the rules.



you should not be so condescending. not everyone is a mathematician, and it is unfair to expect everyone to be on the same page as you.

i still can't figure out what all of your math is trying to say......i don't see how it relates to real world performance of a cpu, but maybe that's just me.

Fine you right, but i have been insulted as well in very subtle manners.
I am not a mathematician either. i am a layman who reads and knows what he has been taught.
Logic is undeniable, Heart is.
I apologize for calling you a moron Kuroimaho.

you are right beatbox, binary is a 3 dimmensional smokescreen of reality.
and what i am saying is not babble. It sounds like babble cause im using the simplest possible Mathematical Language to explain to you how it works.

If you don´t understand what i am saying you need to read about.

1)Nature of Numbers
2)Vector Spaces. and their relation to Functions on Cartesian Planes.
3)Matrices, Inverse Matrices, Gauss and Determinants
4)Transformations

If AMD goes octal core, and stays with k10 for the end of time then i would say AMD is a **** company, because in order to improve they would need to add cores.
But the raodmap says differently, they are exhausting x86 capacitance, they will pipeline like intel, they will add fusion for graphics, but until we use super efficient Floating Point processing systems this will not change.

once you have integrated this efficient Floating Point system(this is done wtih FPU co-processing untis, at the moment), and once you have bettered it like Intel is doing is planning with Larrabee on the future.

Reality is, Memory Bandwidth, with non uniform memory access over time, this adds scalability... Processing Speed of the Subsystem and Material that you use, are the limiting factors.

These reasons are the ones that make findings like these
http://www.dailytech.com/Scientists...+Quantum+Computer+Advancement/article8238.htm
so exciting.

And im afraid that you are wrong on the hardware backing it up bit.
Capacitors and Pulsing Circuitry help improve efficiency over the work to be Done, or in other words Energy to be consumed over time.
if you dont believe me grab a chemistry book.
this is as far as i can go without going into Discrete Equations, for you cannot draw R^4 without contrasting it with its derivative.
and ive shown you what x^4 looks like.

 

[hUMANbEATbOX]

now that is a more reasoned response.

i see what you mean now, that things will change as AMD adds more cores, and they can finally leverage the extra memory bandwidth they enjoy now. i'm all for that, and i hope that they can realize their goals.

the reality of the situation is that they need to find a way to get there. their capacity is already strained, trying to make dual core k8's. when quads get thrown in the mix, it will only hurt them, as they will have to spend much more making each cpu they ship, and they won't necessarily be getting a lot more money for them (they will have to be competitive NOW, not in 2009, in order to make enough money to even exist in 2009).

 

[savageseb]

now that is a more reasoned response.

i see what you mean now, that things will change as AMD adds more cores, and they can finally leverage the extra memory bandwidth they enjoy now. i'm all for that, and i hope that they can realize their goals.

the reality of the situation is that they need to find a way to get there. their capacity is already strained, trying to make dual core k8's. when quads get thrown in the mix, it will only hurt them, as they will have to spend much more making each cpu they ship, and they won't necessarily be getting a lot more money for them (they will have to be competitive NOW, not in 2009, in order to make enough money to even exist in 2009).

well thats why they arent letting the Real benchmarks out.
Because they are veeery competitive.



AMD is way ahead at R&D,
this is what me and cheator have been saying.

 

[FlahsMemory]

Some of you think that ghz is a number that is ultimate and applies equally to all cpus. You fail to realize that a ghz is only applicable to the technology that it represents. For instance company A produces a chip clocked at 2ghz. Company B produces a chip clocked at 3ghz. Company A chip is better than company B clock for clock. But company B at 3ghz is better than company A.

Other senerio, company A release a chip technology 1 year prior to company B. Company A has a set standard for its rating of ghz. Company B can now make a processor with a lower rating in ghz to compete with company A clock for clock giving an illusion of superiority.

This is the scenerio for the last few years. Intel has high clock one year, AMD release chip lower clock which is better, now intel is lower clock and AMD try to compete with intel at higher clock chips with 3ghz and soon x2 6400+. Penryn is core 2 and it will be clocked higher. Phenom lower. Nehalem will probably be clocked lower than barcelona( if AMD makes it next year). And then there is overclocking.
Why is intel allowing so much better overclocking on their chips? Well the reason is because they can make money in the long run if they release a processor underclocked and just better than the top of the line competitors chip. If the competitor has a breakthrough or a secret line of better stepping chips in development then intel can ramp up the chips to what they supposedly can perform.

This is making it difficult for AMD to make a breakthrough technology because they must scrap prior R&D to compete not with what intel has for the market but for what they are able to do later with the technology they are pushing( so phenom should be actually built to compete with an overclocked core2 at 3.6ghz+ not with the 2.6ghz or 3ghz they are releasing on the market. Now with the way intel is moving so fast AMD is force to hold back until they have something brilliant to debut. As of right now with a phenom clocked at 3ghz it only proves they need to ramp there technology up to compete with intels next generation penryn. This is pretty bad as in order to control the market they need to enable room for having faster chips clock for clock and for later releases.(Unless phenom is awesome and it can clock substantially higher than 3ghz.)

If they cannot produce low clocking chips that can compete with high clocking intel chips without head room for better speed later then they are forced to the drawing table again. And this is what I believe the problem is now. Since the window of development is so small because intel has the superior technology on the market and also is going to introduce penryn earlier AMD is forced into hibernation till intel slows down. When intel slows down they can agian gain some room for releasing better technology that can perofrom well. That is why penryn and barcelona are really not the real competitive technology and the real show will begin with who has the better successor in 08 with both to have very unique and different technology compared to their prior generation.

 

[AlabamaCajun]

LOL, this thread got better as the day went on :buhahaha: No acid dropped, nothing smoked or inhaled, just trying to lighten up the p***ing matches.

Please go on, I like to see new ideas born.
While L3 cache is certainly... nice... it is still not necessary. L2 cache is faster (much), easier to access and Intel is planning to have tons of it.
As the number of cores increase so will the number of L2 cache.

And this in no way will get AMD to Intel's level. You need faster CPU's not better memory managed.
Lemme ask you this:
When was the last time YOU made an application that was bottlenecked by the L3 cache?


Quite on the contrary I understand what Savage is trying to say. And also what you are saying. It is a nice idea for increasing FSB to RAM... but... I hate to break it to you... it won't get Intel.
~

FSB on AMD runs at CPU speed, AMD calls it a crossbar. Review the hardware on the use of L3, it's there to bucket L2 cache mostly for reuse and intercore comm. AMD abandoned the FSB years ago in favor of a multipath internal crossbar and separate IO and RAM channels.

 

[savageseb]

now that is a more reasoned response.

but my other responses explain all of that and more. lol

 

[Gautam]

I wonder which one will turn out faster in reality...

 

[hUMANbEATbOX]

but my other responses explain all of that and more. lol

yes, you went into great detail. and you are good at that. but i think the big picture might be eluding you.

like i said before, the big picture is actual hardware running actual software. until i see that, all bets are off, and really, your guess is as good as mine (even without the math ). have you heard of the phrase "The best laid plans of mice and men often fall asunder"? or perhaps, "don't count your chickens before they hatch"?

i honestly hope AMD really has something here, and i hope they can start making money with it ASAP. but without actually seeing any (real) performance numbers to speak of, that's all it is, hope.

 

[CGR]

Whats really amusing thing about this thread is it hasnt become an AMD vs Intel flame war.. I thought for sure it was going to when I read the first post.

Its more of an AMD vs AMD thread now.

A good majority of items in this thread are all speculation and theory. Why people are arguing on speculation I cannot understand.

Theories are only as good as the solutions that prove them. And since the actual item being talked about doesnt exist yet then a solution cannot be formed.

K10 could very well be a huge Intel killer, only time will tell.

 

[savageseb]

Whats really amusing thing about this thread is it hasnt become an AMD vs Intel flame war.. I thought for sure it was going to when I read the first post.

Its more of an AMD vs AMD thread now.

A good majority of items in this thread are all speculation and theory. Why people are arguing on speculation I cannot understand.

Theories are only as good as the solutions that prove them. And since the actual item being talked about doesnt exist yet then a solution cannot be formed.

K10 could very well be a huge Intel killer, only time will tell.

Can you deny that 2+2=4?
Logic over comes any other argument because it is purely
OBJECTIVE.

The Discrete Mathematical Argument goes like this

Number = Vector (of a Vector Space), the degree to which you Calculate it(Calculus) is upto you,
Because you are structuring data, you decide the degree to which you will structure Sequence of Data(bits)

IF a Number exists and this number is X1=2, and another number exists and this number is X2=2, then 2 times X is equal to x^2.
the derivative of X^2=2X and thus i prove my point.
derivatives work around logarithmic events or Functions. Such as exponents and roots,

This is not speculation, this is basic logical math,
to understand what one says you have to truly understand each one of the words the person is using.

 

[hUMANbEATbOX]

and specifically, HOW does that relate to the performance of K10?

 

[savageseb]

and specifically, HOW does that relate to the performance of K10?

It is as i´ve posted previously throughout this thread, if you missed it, go back to my first post, then to my second, then to my third
(where i bother to actually explain it to people on a mathematical level).

The HOW, is the interaction of Dimmensional Elements in Vector Spaces. but this definition is too worded for someone who hasnt studied these 4 points,
and linked them to Calculus(which is nothing but Functions of time).
a Function Equals work performed OVER time.

1)Nature of Numbers
2)Vector Spaces. and their relation to Functions on Cartesian Planes.
3)Matrices, Inverse Matrices, Gauss and Determinants
4)Transformations

Maybe i should add calculus here...
hell yes,
5) Calculus, for the understanding of Discrete Sequencing.

The Discrete Mathematical Argument goes like this

Number = Vector (of a Vector Space) (this is an AXIOM which becomes a LAW, a preposition, whether the preposition is TRUE or FALSE, you do not know)

the degree to which you Calculate it(Calculus) is upto you,
Because you are structuring data, you decide the degree to which you will structure Sequence of Data(bits)

IF a Number exists and this number is X2=X1=2,
then 2 times X is equal to x^2.
the derivative of X^2=2X and thus i prove my point.
derivatives work around logarithmic events or Functions. Such as exponents and roots,

This is not speculation, this is basic logical math,

to understand what one says you have to truly understand each one of the words the person is using.

 

[CGR]

Can you deny that 2+2=4?
Logic over comes any other argument because it is purely
OBJECTIVE.

The Discrete Mathematical Argument goes like this

Number = Vector (of a Vector Space), the degree to which you Calculate it(Calculus) is upto you,
Because you are structuring data, you decide the degree to which you will structure Sequence of Data(bits)

IF a Number exists and this number is X1=2, and another number exists and this number is X2=2, then 2 times X is equal to x^2.
the derivative of X^2=2X and thus i prove my point.
derivatives work around logarithmic events or Functions. Such as exponents and roots,

This is not speculation, this is basic logical math,
to understand what one says you have to truly understand each one of the words the person is using.

Since you want to fall back on logic then answer this logical question.

If AMD has a K10 chip that will bury Intel's current and possibly their future processors, why wont they release the benches for it?

From a logical point of view, its illogical not to release that information. What does AMD have to gain by not releasing it? They have been beaten up by Intel for a while now. If they had a "KO" chip they would release its numbers in a heartbeat.

Now lets step into theory for a second. My theory is that AMD has no such benches. That they know that K10 may only reach current Intel chip speeds and they can only speculate on how it will compare to future Intel chips. Now that could be reason not to release the benches yet.

"Live long and prosper"

 

[AlabamaCajun]

If a Vector is a point with direction we can determine it's normal or right angle to it's planar direction. If we calculate the movement from that point X1, lets assume is up it's normal must be the X2 or number of cores. If X1 is it's speed we should be able to raise it to the power of it's speed. But you still have to consider the exponential depredation of multiple cores sharing resources. Your variables will have to pull in the latencies of L3 cache, IO channel and Ram bottlenecking. It's more the the growth of the pure exponent that count here. Throw in the chaos of programming and user habits to throw out all of our theories. Intel is beating this horse with L2, AMD hopes to split the horse into pieces with faster RAM channels and L3 victim cache.

 

[ochungry]

Can you deny that 2+2=4?
Logic over comes any other argument because it is purely
OBJECTIVE.

.

By the same token (your logic and mathematical expression of K10)
2+2=4
or 2 to power of 2 = 4
but unfortunately:
root sq of 4 is not always a positive 2 but also a negative 2 (calculus 101)
relating your mathematical logic in regard to K10
Thus
can be a negative or a positive impact on K10 quad core performance.
which hence is true:
as AMD stand right now -
Is that enough mathematical bull for you?

 

[savageseb]

If a Vector is a point with direction we can determine it's normal or right angle to it's planar direction. If we calculate the movement from that point X1, lets assume is up it's normal must be the X2 or number of cores. If X1 is it's speed we should be able to raise it to the power of it's speed. But you still have to consider the exponential depredation of multiple cores sharing resources. Your variables will have to pull in the latencies of L3 cache, IO channel and Ram bottlenecking. It's more the the growth of the pure exponent that count here. Throw in the chaos of programming and user habits to throw out all of our theories. Intel is beating this horse with L2, AMD hopes to split the horse into pieces with faster RAM channels and L3 victim cache.

the problem is you work around the chaos of math,
while understanding math means understanding the assumptions mathematics makes upon reality.

So a Vector in R^2 is a vector pointing in a direction, a Vector in R^3 is the volume of a Sphere
a Vector in R^4 is the Electrostatic volume of a wave.....


To understand this you have to study the mathematical errors found throughout history, starting with Pythagorean Principles.

Science needs to prove itself mathematically, therefore our science is limited upon our math.




FOR all those questioning Logic, Just go back and read everything i have posted, if you dont have a clue of what i am saying, please dont bother to post. im getting tired of repeating myself

 

[savageseb]

Whats really amusing thing about this thread is it hasnt become an AMD vs Intel flame war.. I thought for sure it was going to when I read the first post.

Its more of an AMD vs AMD thread now.

A good majority of items in this thread are all speculation and theory. Why people are arguing on speculation I cannot understand.

Theories are only as good as the solutions that prove them. And since the actual item being talked about doesnt exist yet then a solution cannot be formed.

K10 could very well be a huge Intel killer, only time will tell.

ever heard of anti trust laws, and the many intel has recently violated?
AMD CANT show the market benchmarks without killing intel and getting all the pie.

you are wrong, companies strategize given the R&D developed.
like for example nehalem was planned on 2002 for 2009. seems intel has had to push back the agenda for nehalem half a year.
AMD ironically claims they are half a year behind. so again how wrong am i?




I explained this on another thread, and on this one too, and if you dont believe me (more mathematically), so be it.

It is a logarithmic decrease in the data being bottlenecked by the subsystem of the L3.

The effects are noted through a Scalar Curve( or half wave) technology(A Data Management system) Over the time period the subsystem requires, to handle a certain amount and type of data. This is achieved Through Virtualization.
the time frame limiting the Subsytem to do work is now given by Hertz/S. or cycle/S^2.

1)
say 1 cycle is 32kbits of data processed per second:
Assuming 1 Second = 2π π=pi dont know the symbol key code.
32Kb/x Seconds.
For Versors we can say x=1 in the beginning

now with scalar technology
32kb(x cores)/x(Second squared, s^2)= Maximum Data to be handled.


2) now apply the fibonnacci series sequence, which is an irrational number sequence, of time(i cant say irrational proportion, now can i?).
the irrational number is obtained by taking the square root of the scalar system.

My Scalar system says:
x(set of bits)per cycle( or Maximum Data to be handled) * S^2= Required Processing Power,
this equation exposes itself as a Scalar product
Scalar*[Matrix^2]= Required Processing Power.
now you may take the Square root of Time since this is cycles/Time Squared.
or the square root of the required processing power, the answer will be the same, AN IRRATIONAL INFINITESIMAL NUMBER, you can approximate to this number Thru FPU operations.
this is the major improvement of the core!

3)take this into account http://anandtech.com/cpuchipsets/showdoc.aspx?i=2939&p=10

"Translation Lookaside Buffers, TLBs for short, are used to cache what virtual addresses map to physical memory locations in a system. TLB hit rates are usually quite high but as programs get larger and more robust with their memory footprint, microprocessor designers generally have to tinker with TLB sizes to accommodate."

"Each Barcelona core gets its own set of data and instruction prefetchers, but the major improvement is that there's a new prefetcher in town - a DRAM prefetcher. Residing within the memory controller where AMD previously never had any such logic, the new DRAM prefetcher takes a look at overall memory requests and attempts to pull data it thinks will be used in the future. As this prefetcher has to contend with the needs of four separate cores, it really helps the entire chip improve performance and can do a good job of spotting trends that would positively impact all cores. "
consideration of NUMA

"AMD Virtualization Improvements

The performance-related improvement to Barcelona comes in the way of speeding up virtualized address translation. In a virtualized software stack where you have multiple guest OSes running on a hypervisor there's a new form of memory address translation that must be dealt with: guest OS to hypervisor address translation, as each guest OS has its own independent memory management. According to AMD, currently this new layer of address translation is handled in software through a technique called shadow paging. What Barcelona offers is a hardware accelerated alternative to shadow paging, which AMD is calling Nested Paging.

Supposedly up to 75% of the hypervisor's time can be spent dealing with shadow pages, which AMD eliminates by teaching the hardware about both guest and host page tables. The translated addresses are cached in Barcelona's new larger TLBs to further improve performance. AMD indicates that Barcelona's support for Nested Paging requires very little to implement; simply setting a mode bit should suffice, making the change easy for software vendors to implement."

Do you get it now?

or must i go on?

 

[CGR]

Chaos Theory...

"In mathematics and physics, chaos theory describes the behavior of certain nonlinear dynamical systems that under specific conditions exhibit dynamics that are sensitive to initial conditions (popularly referred to as the butterfly effect). As a result of this sensitivity, the behavior of chaotic systems appears to be random, because of an exponential growth of errors in the initial conditions. This happens even though these systems are deterministic in the sense that their future dynamics are well defined by their initial conditions, and there are no random elements involved. This behavior is known as deterministic chaos, or simply chaos."

Jurassic Park FTW!!

 

[savageseb]

Chaos Theory...

"In mathematics and physics, chaos theory describes the behavior of certain nonlinear dynamical systems that under specific conditions exhibit dynamics that are sensitive to initial conditions (popularly referred to as the butterfly effect). As a result of this sensitivity, the behavior of chaotic systems appears to be random, because of an exponential growth of errors in the initial conditions. This happens even though these systems are deterministic in the sense that their future dynamics are well defined by their initial conditions, and there are no random elements involved. This behavior is known as deterministic chaos, or simply chaos."

Jurassic Park FTW!!

the problem is you work around the chaos of math,
while understanding math means understanding the assumptions mathematics makes upon reality.

So a Vector in R^2 is a vector pointing in a direction, a Vector in R^3 is the volume of a Sphere
a Vector in R^4 is the Electrostatic Potential Volume of a wave.....


To understand this you have to study the mathematical errors found throughout history, starting with Pythagorean Principles.

Science needs to prove itself mathematically, therefore our science is limited upon our math.