Its kinda funny on every forum ,Anyone that seems to understand the differences,gets banned or deleted,if they dont fall for the Intel marketing hype
".. an ounce of honest data is worth a pound of marketing hype." - Spec.Org< answers are there!
Its kinda funny on every forum ,Anyone that seems to understand the differences,gets banned or deleted,if they dont fall for the Intel marketing hype
".. an ounce of honest data is worth a pound of marketing hype." - Spec.Org< answers are there!
Actually my Bribane 2.5 45watt dual core and my 4850 ATI using avivio will slaughter your box encoding. Changing a DVD to DIVX about 2 minutes I think
And its a fact that Intel does very well on single threaded apps, but on muti threads not so well(more than 2 threads), the advantage it has on single thread doesnt scale per core the same, on very multi threaded stuff and heavy loads, it cant because all 4 cores talking over FSB and RAM,not processor cache like Phenom.
Anyway why do you thinK the new Intel copied the AMD Idea ? Can you take a guess? lmao
I would find an Open Source Video Encoder and you might become very unhappy ! Also go look at spec.org standardized tests and bench marks with no cheating
Last edited by Viper666; 08-11-2008 at 09:24 AM.
Actually same clocked Core2 Duo or Quad will wipe the floor with your Brisbane @2.5GHz for the purpose you've mentioned above.
What's the time and the quality you are converting for 2 minutes?Changing a DVD to DIVX about 2 minutes I think
Core2 Duo have better performance scaling than K8. Phenom has better performance scaling than Core2 Quad, but it performs slower at same clock, including the apps that are utilizing more than 2 threads.And its a fact that Intel does very well on single threaded apps, but on muti threads not so well(more than 2 threads), the advantage it has on single thread doesnt scale per core the same
What you are saying is not true. Intel always had and still has much more advanced cache architecture than AMD ever., on very multi threaded stuff and heavy loads, it cant because all 4 cores talking over FSB and RAM,not processor cache like Phenom.
It's core-to-core communication on their Core2 Duo is much more sophisticated than on both K8 and Phenom.
Core2 Quad is a MCM of 2 dualcore chips. The cores on the same chip are communicating 2x faster than the cores on Phenom. For frequencies over 2GHz the cores of the different dies of the MCM, a C2Q with FSB1333 is communicating slower than the cores of a same clocked Phenom.
AMD idea? Intel had a x86 CPU with on-die-memory-controler and on-chip-video 15 years ago, before AMD has designed their first x86 CPU.Anyway why do you thinK the new Intel copied the AMD Idea ? Can you take a guess? lmao
LOL, you are funny. Before you can be so sure, why don't you check for some benches online?I would find an Open Source Video Encoder and you might become very unhappy !
SPEC.org are not measuring desktop performance, and if you haven't noticed the title yet, we are talking about desktop CPUs here. So, face it, on the desktop Intel is owning AMD and that will remain to be the case for sure in the following 2 years.Also go look at spec.org standardized tests and bench marks with no cheating
One hundred years from now It won't matter
What kind of car I drove What kind of house I lived in
How much money I had in the bank Nor what my cloths looked like.... But The world may be a little better Because, I was important In the life of a child.
-- from "Within My Power" by Forest Witcraft
Just link the data..... if this is true, and it is a widely known fact, there must be ample examples.
EDIT: OOOPs, what test have I ran? Hang tight, I will them...
EDIT2: All ran comparing 9600BE@2.3, 9850@2.3, QX9650@2.3, 9850@2.5, and QX9850@2.5 these are all the benches I have ran, some single threaded, some multithreaded.
3DMark01
3DMark03
3DMark05
3DMark06
Aquamark3
Blender
Cinbench2003
CinbenchR9.5
CinbenchR10
Company of heros 1.4
Commanche4
crysis
crystalmark
CPU rightmark
CPUMark2.1
Doom3
DVDShrink3.2
Euler3D
Everest (benchmark suite version 4.5)
FarCry
FEAR
Fritz Chess (1-4 cores)
Geekbench2
H264 Bench Tech ARP version 1 and 2
HexusPiFAST
HL2-LC
LameMP3 Multthreaded both MS and Intel compiled
LostPlanet (which I have linked above)
MainConcept (1 and 2 Pass HD264, MPEG2 render 4 movie files)
MaxiPI (weird behavior on Phenom here, one core is always 10% slower than the other 3)
Multitask suite 1 (4 single threaded apps)_
Multitask suite 2 (4 multithreaded apps)
n-bench3 (this is AMD's benchmark, runs faster on Intel quads)
Ortho CPU benchmark
Paint.NET (benchmark suite)
PCMark04
PCMark05
Povray3.7B23 (Phenom does very well here)
Prey
QW-Enemyterritory
RightMarkMA
SciencMark2.0 (again Phenom does very well here)
SeriousSam2
Unreal Tournament 3
Windows Media Encoder 9
World in Conflict
WPrime
Last edited by JumpingJack; 08-11-2008 at 04:42 PM.
One hundred years from now It won't matter
What kind of car I drove What kind of house I lived in
How much money I had in the bank Nor what my cloths looked like.... But The world may be a little better Because, I was important In the life of a child.
-- from "Within My Power" by Forest Witcraft
Here is one:
http://connexitor.com/blog/pivot/entry.php?id=191
There are some links to scaling samples in the thread alsoCode:Clients Opteron 875 Opteron 2347 Xeon 5345 1 3.1s 2.9s 1.6s 2 6.7s 6.5s 12.6s 3 12.1s 10.5s 20.5s 4 18.1s 14.9s 44.6s
EDIT: Let me add..
Ohhhhh, dual socket server... hmmmmm, sure. Ok.... that's fine, this I agree with... it really has nothing to do with ability to multithread, but really just gross overall bandwidth. There are a few server and many HPC applicaitons where the working sets are huge and the raw throughput is tremendous.... AMD's BW design really shines through here.
However, how does that transcend the bulk of the market? By this, I mean, where desktop and single socket clients... you know, the ones we use.
Last edited by JumpingJack; 08-11-2008 at 05:04 PM.
One hundred years from now It won't matter
What kind of car I drove What kind of house I lived in
How much money I had in the bank Nor what my cloths looked like.... But The world may be a little better Because, I was important In the life of a child.
-- from "Within My Power" by Forest Witcraft
Last edited by JumpingJack; 08-11-2008 at 04:59 PM.
One hundred years from now It won't matter
What kind of car I drove What kind of house I lived in
How much money I had in the bank Nor what my cloths looked like.... But The world may be a little better Because, I was important In the life of a child.
-- from "Within My Power" by Forest Witcraft
I am doing a lot of server programming. Could you please inform me about the differences? How much more communication do you think there is between cores on servers compared to desktops?
If you take one Web application for example, each request is usually handled in one single thread. The communication between threads is not that big, each thread is managing its own state. It is much more difficult to design applications if threads are talking to each other in different ways.
Keep your eyes open for games optimized for Playstation 3. The cell processor is a bandwidth monster. But you need to take advantage the cores (if you could call them that) to get speed. More and more game engines seem to have learned to optimize for this. XBOX works like a PC and if the game is optimized for PC then they will focus on Intel.
On servers there is often applications developed in Java and .NET. These are using a lot of memory; C++ is normally not that big in memory use.
Why not testing applications that use memory? Databases exist on desktops also. running more than one applications will use more memory, running five will use more than two. Web browsers etc it using a lot of memory.
If you only test single threaded applications or applications that are tiny in memory use you don’t need to test, you can predict the result and save some time. High frequencies and one BIG L2 cache will be faster.
This I cannot do, i do not have the data for server.... what i do have is benchmark data from several different applications common to desktop, clock for clock AMD does not win one, single or multithreaded, or even multitasking.
Also, second request can you explain why doubling the FSB BW does not show any appreciable difference in lost planet, if this is not convincing for you then I can do what ever game you want....
EDIT: Hang tight, I think I can link up a publicly accessible paper on this topic.... it should be informative
Jack
Last edited by JumpingJack; 08-11-2008 at 05:20 PM.
One hundred years from now It won't matter
What kind of car I drove What kind of house I lived in
How much money I had in the bank Nor what my cloths looked like.... But The world may be a little better Because, I was important In the life of a child.
-- from "Within My Power" by Forest Witcraft
One hundred years from now It won't matter
What kind of car I drove What kind of house I lived in
How much money I had in the bank Nor what my cloths looked like.... But The world may be a little better Because, I was important In the life of a child.
-- from "Within My Power" by Forest Witcraft
Well, that is true ... part of the OS. I haven't tried to install Oracle and query my house hold expenses.
So what do I need to do to benchmark the registry?
I kinda think the XP registry (or even Vista registry) is a bit small to overwhelm a processor, considering even a Sempron can run it .
One hundred years from now It won't matter
What kind of car I drove What kind of house I lived in
How much money I had in the bank Nor what my cloths looked like.... But The world may be a little better Because, I was important In the life of a child.
-- from "Within My Power" by Forest Witcraft
Isn't it clear, you linked the data.
You don't need to see the code for the data, if the FSB is a problem, then why if I double it it has no effect? you keep going back to this code thing, what is it in the code that makes you think you are going to find an answer?
So again, if per your hypothesis, the FSB is saturating at high resolutions ... why at high resolutions can I effectively double (or 1/2) the effective FSB BW and not change the result?
One hundred years from now It won't matter
What kind of car I drove What kind of house I lived in
How much money I had in the bank Nor what my cloths looked like.... But The world may be a little better Because, I was important In the life of a child.
-- from "Within My Power" by Forest Witcraft
This is normally what matters for users in normal use. Quick responses and fast load times, switching between applications etc. There are few computer users that run applications with tasks that last for a long time. But if they do that then they don’t sit and wait, they are probably surfing or doing something else while the application that needs time is doing its work.
I don’t know about you but if I am extracting some big file from some compressed format and that takes ten minutes. I don’t sit and wait for that. I do other tasks.
No, I think you don't know what you are talking about ....
Look, your claim in this thread is that in high res gaming, that the FSB cannot handle all the little threads poking around and that Intel's CPU performs poorly because of the FSB, thus Phenom is best.
I will not argue that high res gaming shows them neck and neck based on the data. But if your FSB argument is true, you do not need to see the code to make an emperical observation. Based on your hypothesis, the FSB is saturated, so if I lower the FSB speed I should take a performance hit....
What you are really observing is the GPU is limiting the FPS, you have hit the GPU limited regime.... the load variation on the GPU is deterined by the resolution you select and the oversampling chosen for the aliasing. You reject this notion, yet I showed you data that proves this notion.... so what I am asking you to do is explain the data.
My suspcion is that any time someone provides you with the data that refutes your claims you will fall back on "I need to see the code", which is in and of it's self code for "I don't know".
jack
One hundred years from now It won't matter
What kind of car I drove What kind of house I lived in
How much money I had in the bank Nor what my cloths looked like.... But The world may be a little better Because, I was important In the life of a child.
-- from "Within My Power" by Forest Witcraft
One hundred years from now It won't matter
What kind of car I drove What kind of house I lived in
How much money I had in the bank Nor what my cloths looked like.... But The world may be a little better Because, I was important In the life of a child.
-- from "Within My Power" by Forest Witcraft
If you are talking about games you should test a game that is communicating with the GPU and is using memory at the same time (there will be conflicts in the FSB that increase latency, amd has no conflicts there). Then you will see big improvements for AMD. More and more advanced games will run better and better on AMD. If the game is using more than one thread to render data then AMD will perform better.
I haven’t read what you have written about Lost Planet. I need to check some tests before I can tell you something about that game. How it scales, how it is using threading etc
EDIT: Checked two tests on Lost planet
Here is one: http://techreport.com/articles.x/14424/4
Comparing QX9650 and QX6850, isn't the main difference cache size? That would mean that Lost Planet likes a big cache (is there other differences between these two processors?)
Last edited by gosh; 08-11-2008 at 06:35 PM.
One hundred years from now It won't matter
What kind of car I drove What kind of house I lived in
How much money I had in the bank Nor what my cloths looked like.... But The world may be a little better Because, I was important In the life of a child.
-- from "Within My Power" by Forest Witcraft
Bookmarks