Largest Performance Gain

Status
Not open for further replies.
Barcelona is a server processor
which is still a K10 processor.




The only major difference I see is higher HT speed, and the addition of L3 cache.
Then I see you don't know a lot about K10.




You said in your previous post that "Penryn hasn't got a lot of change", which is not true at all.
It is compared to the changes in K10.

Just look at the benchmarks at this website, you can see a comparison between QX6800 and Quad core version of Penryn (Yorkfield)
Welcome to AnandTech.com [ Article: Intel Penryn Performance Preview: The Fastest gets Faster]

In one of those benchmarks tests Yorkfield performed 111% faster than QX6800 !!!!
sure, one benchmark.
extra cache will do that for encoding.

*edit*
also consider the fact that the Yorksfield was clocked higher.
 
It is compared to the changes in K10.

Thats your opinion. But this doesn't mean that your opinion is a fact.

However, Athlon64 is horrible CPU compared to Core 2 Duo, so AMD needs to do a lot of changes for Athlon64 in order to compete with Intel upcoming processors. Intel doesn't need to do a lot of changes for Core 2 Duo


sure, one benchmark.
extra cache will do that for encoding.

And do you think that cache alone will make the processor run 2 times as much as fast in encoding ? !!!

By your logic, E6320 should be at least %100 faster that E4300 in encoding, since it has double the cache. But thats not true.


also consider the fact that the Yorksfield was clocked higher.

It was only 300MHz faster !!!, and it has 1333MHz FSB.

But what is the problem with that ?

Yorksfield has 45nm die shrink, it can achieve higher clocks without consuming any extra power.

..............

Now let us end this debate. Barcelona will come out sometime on June or July, and at that time we will see if it will be able to beat Core 2 Qaud or not.
 
Thats your opinion. But this doesn't mean that your opinion is a fact.
Please read more about K10's architecture.



However, Athlon64 is horrible CPU compared to Core 2 Duo
Wait, this coming immediately after your previous comment?

so AMD needs to do a lot of changes for Athlon64 in order to compete with Intel upcoming processors. Intel doesn't need to do a lot of changes for Core 2 Duo
AMD are making a lot of changes.

And do you think that cache alone will make the processor run 2 times as much as fast in encoding ? !!!
It can, depending on what is being encoded, and in what format. Encoding does require a lot of data to be sent to and from the CPU, so more cache really helps.

The fact that most other benchmarks were in the range of 10-25% improved, while encoding got a much bigger jump, does make my case look more probable.

By your logic, E6320 should be at least %100 faster that E4300 in encoding, since it has double the cache. But thats not true.
Again, it depends on what's being encoded, and in what format.

It was only 300MHz faster !!!, and it has 1333MHz FSB.
And in most tests, it was roughly 10-25% better, which is evidence against the architecture making much of the difference to that performance gain, rather than the frequency.

But what is the problem with that ?
I'm not saying it's a problem. Just that the actual architectural improvements are much easier to compare with them both at the same clock speeds.

Now let us end this debate. Barcelona will come out sometime on June or July, and at that time we will see if it will be able to beat Core 2 Qaud or not.
 
who can u seeing winning the battle in the future?
The future is inherently unpredictable. All we can do is try and make probability statements based on information about what is happening now.

AMD claims K10 to have at least a double figure percent advantage over current Intel quad cores. If so, this will at least put K10 close to Penryn in most applications
 
While C2D beats the current AMD architecture, it doesn't beat it by miles, but by inches. Bleeding edge is where companies want to be, but that razor is a hard one to walk.

I look forward to the new generation of AMD's chips, just to see the further evolution of the species. I seriously doubt I'll be able to buy into it, but I definitely want to see it.
 
While C2D beats the current AMD architecture, it doesn't beat it by miles, but by inches. Bleeding edge is where companies want to be, but that razor is a hard one to walk.

I look forward to the new generation of AMD's chips, just to see the further evolution of the species. I seriously doubt I'll be able to buy into it, but I definitely want to see it.

Well said....the K8 is already only slightly slower than the C2D's really and AMD skipped their K9 and their K10 will most likely be a lot faster than the C2D at this rate. I'm interested to see as well who will come out on top but it doesn't matter as long as both companies are close enough to keep the prices competitive and then the real winners are us.
 
Status
Not open for further replies.
Back
Top Bottom