Q6600 isn't real quad?

[reconviperone1]

4 cpu's come up in task mahager, ir runs divx converter faster than a dual core, dont care why or why not it's a quad, just care that it is doing what it's supposed to do.

 

[TechnologyCoordinator]

4 cpu's come up in task mahager, ir runs divx converter faster than a dual core, dont care why or why not it's a quad, just care that it is doing what it's supposed to do.

What? You don't care about marketing jargon but rather care more about results!?

 

[spud]

Could you give some references? Everything isn’t 100% effective as you might know. Switching threads isn’t just a snap, there needs to be memory in the cache for that thread etc.
The L3 cache on Nehalem has a latency of 39 cycles, squeezing in another thread there will not be easy.

L3 is for ondie L2 mirroring so I fail to see your point.

Word, Playa.

 

[kassler]

L3 is for ondie L2 mirroring so I fail to see your point.

Word, Playa.

I think you need to read some about it, how it works. L3 cache is used with the memory controller also.

 

[spud]

I think you need to read some about it, how it works. L3 cache is used with the memory controller also.

It's fundemental usuage is for mirroring L2 with access to the memory controller the core's can make main system memory calls without actually waiting (normal latency), the Core 2's prefecters will have that much more of low latency memory pool to access, again I fail to see your point.

Word, Playa.

 

[BadTrip]

omg welcome back mike

 

[kassler]

It's fundemental usuage is for mirroring L2 with access to the memory controller the core's can make main system memory calls without actually waiting (normal latency), the Core 2's prefecters will have that much more of low latency memory pool to access, again I fail to see your point.

Word, Playa.

English is not my main language so it’s hard for me to exactly understand and explain when there will be technical discussions. You need to exactly explain what you are meaning otherwise I might missinterpretate.

On the core processor and the data isn’t in the cache, then there is something about 250 clocks (not sure about this) to get new memory and the FSB isn’t occupied by other work.
Phenom has just above 100 clocks.
Nehalem that has triple channel DDR3 has lots of bandwidth and can take advantage of very fast memory.

The performance about memory also depends on if it is a 32 or 64 bits application. I know that phenom is design to work optimal on 64 bits applications.

 

[vochtige]

oke folks, this wasn't suppost to get such an interssting topic.

keep on going!

 

[shadowthor]

Even if it isn't real quad, at least the performance is better, and that's all that matters in the end . Yep this topic is definitely interesting, have ppl weighting in from all sides.

 

[jkflipflop98]

For anyone that's kind of "green" and here looking for information, dont listen to a word of that garbage that Kassler is spewing.

He obviously has no idea what he's talking about. He's making claims of superlinear scaling on Phenom. It just doesnt work that way.

 

[kassler]

For anyone that's kind of "green" and here looking for information, dont listen to a word of that garbage that Kassler is spewing.

He obviously has no idea what he's talking about. He's making claims of superlinear scaling on Phenom. It just doesnt work that way.

What am I saying that is wrong? Could you create some facts

 

[Wisecracker]

4 cpu's come up in task mahager, ir runs divx converter faster than a dual core, dont care why or why not it's a quad, just care that it is doing what it's supposed to do.

What? You don't care about marketing jargon but rather care more about results!?

From Tom's CPU Chart
AMD/Intel Quads @ 2.4GHz
(with the e8xxx for good measure)

Divx 6.6.1
2 Minutes DVD Terminator 2 SE (Encoding 720x576 16:9 @ 25 fps)

e8500 = 86 sec
q6600 = 90 sec
e8400 = 91 sec
Phenom 9700 = 96 sec

Don't see a great deal of parallelism across those four cores 😀

 

[TechnologyCoordinator]

From Tom's CPU Chart
AMD/Intel Quads @ 2.4GHz
(with the e8xxx for good measure)

Divx 6.6.1
2 Minutes DVD Terminator 2 SE (Encoding 720x576 16:9 @ 25 fps)

e8500 = 86 sec
q6600 = 90 sec
e8400 = 91 sec
Phenom 9700 = 96 sec

Don't see a great deal of parallelism across those four cores 😀

Post of the day, thank you Wisecracker!

Q: BUT AMD'S QUAD IS REAL-ER! DOESN'T THAT COUNT!?!?
A: No.

 

[kassler]

Don't see a great deal of parallelism across those four cores 😀

Encoding, rendering, singlethreaded games will run very well on Intel. But let the program run alone, if you are working with other applications intel will not like it.

 

[TechnologyCoordinator]

Encoding, rendering, singlethreaded games will run very well on Intel. But let the program run alone, if you are working with other applications intel will not like it.

Fanboyistic misinformation.

Q: But TC, how can I know if something might be fanboyistic misinformation?
A: Have no fear, just look for a few key elements, such as posts that say:

if you are working with other applications intel will not like it.

 

[kassler]

Fanboyistic misinformation.

I know that is sensitive to say anything good about AMD, those who have taken Intel to their hart believes that Intel is best on almost everything. Presenting technical information apart from review that they might have problem to understand, on what conditions the review is done etc is often the proof that is the root of their knowledge.
Presenting information about AMD’s strong points and you will be called fanboy immediately. I wonder who is the real fanboy…

 

[TechnologyCoordinator]

I know that is sensitive to say anything good about AMD, those who have taken Intel to their hart believes that Intel is best on almost everything. Presenting technical information apart from review that they might have problem to understand, on what conditions the review is done etc is often the proof that is the root of their knowledge.
Presenting information about AMD’s strong points and you will be called fanboy immediately. I wonder who is the real fanboy…

Q: Who is the real fanboy?
A: The person spreading FUD about Intel multi-tasking WITHOUT backing it up with any factual information such as benchmarks.



And since you like Wikipedia so much, here you go: http://en.wikipedia.org/wiki/Fear%2C_uncertainty_and_doubt

 

[uguv]

I know that is sensitive to say anything good about AMD, those who have taken Intel to their hart believes that Intel is best on almost everything. Presenting technical information apart from review that they might have problem to understand, on what conditions the review is done etc is often the proof that is the root of their knowledge.
Presenting information about AMD’s strong points and you will be called fanboy immediately. I wonder who is the real fanboy…

Back up your statements with benchmarks and people will not dismiss them as fanboi-ism. Just throwing out statements like "Intel processors can't multi-task well" without backing it up with some facts doesn't help to strengthen your argument.

 

[kassler]

http://www.warrenuncensored.com/images/THG/Theo.png

Q: Who is the real fanboy?
A: The person spreading FUD about Intel multi-tasking WITHOUT backing it up with any factual information such as benchmarks.

here is one sample
http://connexitor.com/blog/pivot/entry.php?id=191

 

[TechnologyCoordinator]

here is one sample
http://connexitor.com/blog/pivot/entry.php?id=191

Server. I have no negative comments about AMD's server product, I feel it is competitive with Intel's product.



However, we're talking desktop quad cores, Phenom and Core 2 Quad. Not Opteron and Xeon.

 

[kassler]

Server. I have no negative comments about AMD's server product, I feel it is competitive with Intel's product.

However, we're talking desktop quad cores, Phenom and Core 2 Quad. Not Opteron and Xeon.

Do you know what the main differences are between serves and desktops?
If you know that and remember what have written about scaling, could you see what I am talking about?
There was some shortage on 45nm C2Q because Intel is prioritizing the server market, much more money there and they needed to get all advantage they can to compete with AMD.

Xeon and Opteron is almost (if not the same) same processors as C2Q and Phenom.

 

[thunderman]

The Phenom does have a more advanced design…four cores on one die will always be better than duck tapping two relic Dual cores together. Each core on the phenom has a data path…where as the Intel will hit bandwidth problems and hit a wall before it’s full potential has been reached. As more software Is multi-core enabled and more than one demanding application is being used…the Phenoms true potential will be realized and Core2Cheeseburger will be forgotten. Nehalem is an Intel panic solution…. Intel followed AMD as AMD are more skilled. Intel know Native Quad is the way to go….AMD is leading the way as usual.

Nehalem is AMD technology in disguise! Future proof yourself with Phenom!

AMD4Life!

 

[vochtige]

i agree with thunderman, amd always shows the way, but jsut has badluck in performance! you see it in cpu and gpu.

i love the germans 😛

 

[TechnologyCoordinator]

Do you know what the main differences are between serves and desktops?
If you know that and remember what have written about scaling, could you see what I am talking about?
There was some shortage on 45nm C2Q because Intel is prioritizing the server market, much more money there and they needed to get all advantage they can to compete with AMD.

Xeon and Opteron is almost (if not the same) same processors as C2Q and Phenom.

Yes. I do know the difference between a desktop and a server. Do you?

The article you quoted was about multi-socket systems, only available in the server market (exception QuadFX Abandoneware). AMD scales better to multiple sockets.

However, any Intel quad-core desktop processor is simply better than ANY AMD desktop processor in nearly every benchmark.

The Phenom does have a more advanced design…four cores on one die will always be better than duck tapping two relic Dual cores together. Each core on the phenom has a data path…where as the Intel will hit bandwidth problems and hit a wall before it’s full potential has been reached. As more software Is multi-core enabled and more than one demanding application is being used…the Phenoms true potential will be realized and Core2Cheeseburger will be forgotten. Nehalem is an Intel panic solution…. Intel followed AMD as AMD are more skilled. Intel know Native Quad is the way to go….AMD is leading the way as usual.

Nehalem is AMD technology in disguise! Future proof yourself with Phenom!

AMD4Life!

 

[kassler]

Yes. I do know the difference between a desktop and a server. Do you?

The article you quoted was about multi-socket systems, only available in the server market (exception QuadFX Abandoneware). AMD scales better to multiple sockets.

Why?