AMD vs. Intel Architectural Debate

[bioduken]

Heres a little math for you, superpi 1m while all 4 cores are loaded 100% crunching with BOINC. If you can beat that with your AMD show me then I will stop BOINC and rerun it...

 

Are you using your computer to calculate superpi ?

 

If Intel can use it's fast cache and don't need to talk to hardware or threads doesn't speak with eachother, then your are right. it is fast!

Edited July 1, 2008 by bioduken

[bioduken]

Both have their strong suites. Unfortunately for the kinds of apps and in most games the Intel will win out Period. Try video encoding. Yes the Phenom coupled with a Crossfire X setup on the Spider Platform is fast at video encoding but it is done by offloading the bulk of the work from the CPU and letting the GPU do the work. Clock for Clock the Intel Quad core is faster than a phenom in my book. I have no hardware allegiance. I will buy what performs best with what I do.

 

AMD runs smoother, it responds faster and applications start faster. also browsing on the internet is a bit faster with amd. if you encode and leave the computer then intel is fast. but if you are working with other tasks as the computer encodes then amd will be smoother. intel doesn't multitask as amd does.

 

running games in 1920x1200 or 2560x1600 then amd is better. as new games arrive amd will also increase the lead and may be better on lower resolutions.

 

When ray tracing games comes out at the end of this year och at the start of next year it is good bye with the current intel processors. then it is time to buy nehalem for intel users

Edited July 1, 2008 by bioduken

[Waco]

AMD runs smoother, it responds faster and applications start faster. also browsing on the internet is a bit faster with amd. if you encode and leave the computer then intel is fast. but if you are working with other tasks as the computer encodes then amd will be smoother. intel doesn't multitask as amd does.

 

running games in 1920x1200 or 2560x1600 then amd is better. as new games arrive amd will also increase the lead and may be better on lower resolutions.

 

When ray tracing games comes out at the end of this year och at the start of next year it is good bye with the current intel processors. then it is time to buy nehalem for intel users

You're so hopelessly wrong it's almost hilarious except that people may read your posts and buy inferior hardware thinking that it's better.

 

I'm one of the biggest AMD/ATI fanboys on here (ask anyone who I've argued with in the past) but there's no way in hell I'd agree with anything you've posted except in the theoretical realm and we all know that means exactly dick in the real world.

[road-runner]

AMD runs smoother, it responds faster and applications start faster. also browsing on the internet is a bit faster with amd. if you encode and leave the computer then intel is fast. but if you are working with other tasks as the computer encodes then amd will be smoother. intel doesn't multitask as amd does.

 

running games in 1920x1200 or 2560x1600 then amd is better. as new games arrive amd will also increase the lead and may be better on lower resolutions.

 

When ray tracing games comes out at the end of this year och at the start of next year it is good bye with the current intel processors. then it is time to buy nehalem for intel users

Do you own a new Intel Quad? Nehalem would be the death of AMD if they didnt have ATI, and it may still be there death...

 

Edit: Can your AMD do this on water cooling?

 

[bioduken]

Do you own a new Intel Quad? Nehalem would be the death of AMD if they didnt have ATI, and it may still be there death...

 

Edit: Can your AMD do this on water cooling?

 

Yes I own one Intel. I don

[Waco]

Yes I own one Intel. I don’t buy intel and hasen’t for about one year but that is because the new socket. I usually buy about 4-5 computers every year. Now with the AM2 socket this number has decreased to 2-3 (it is possible to switch parts). When Nehalem arrives and if that performs well for the price, then I may buy intel again.

And no, AMD is slower when it comes to raw processor performance. Nehalem is in fact slower also when it comes to raw processor performance then the core generation.

I agree with all of you that say that intel is a faster processor. The problem is that the total package buying a Intel computer has some bottlenecks. These bottlenecks hurt performance. Also intel isn’t good when applications are scaling to a lot of threads, it isn’t built for that.

You can claim it hurts performance all you want and you can claim that pigs can fly...in reality it doesn't matter in the slightest.

 

Intel is the top dog right now and if you can't admit that then you've blinded yourself. Imaginary bottlenecks don't hurt performance.

 

As for Intel processors not scaling well when using multiple threads...that's just total rotten rutebegas.

 

 

(edited by mod for language)

[bioduken]

As for Intel processors not scaling well when using multiple threads...that's just total stinky diapers.

Do you know why AMD is strong on servers?

[jammin]

It's all very well to proclaim the advantages of AMD processors, but in this case numbers mean more than words.

If you want to convince people then back up what you are saying with some data from a reliable source.

[bioduken]

It's all very well to proclaim the advantages of AMD processors, but in this case numbers mean more than words.

If you want to convince people then back up what you are saying with some data from a reliable source.

I am not here to convince people of anything. You can believe what you want.

I am just telling you how they work because this is something that I am interested in and knows a bit about.

One thing about this review that was very good was that it was the first review that I have seen where they had one phenom compared with intel processor and tested games in different resolutions. I have never seen one review like that before. What also was very good about it was that It confirmed what I have read about the two architectures. How Intel and AMD works.

When reviews are done for processor speed they normally test games at very low resolutions. They never test I/O performance. This test didn’t test I/O but if you know what to look for you can draw some very interesting conclusions from the test.

Edited July 1, 2008 by bioduken

[jammin]

I am not here to convince people of anything. You can believe what you want.

 

That's fair enough but statements such as:

 

AMD runs smoother, it responds faster and applications start faster. also browsing on the internet is a bit faster with amd. if you encode and leave the computer then intel is fast. but if you are working with other tasks as the computer encodes then amd will be smoother. intel doesn't multitask as amd does.

 

running games in 1920x1200 or 2560x1600 then amd is better. as new games arrive amd will also increase the lead and may be better on lower resolutions.

 

Need to be backed up with evidence if they are to be taken seriously.

Most people (I would hope) are already aware of the limitations of FSB architecture but are less inclined to care if performance differences aren't clearly measurable.

An yes, the Phenom processors may well scale to multiple threads more effectively in certain applications when compared with Intel processors, but in the above statement you make it sound like there is a huge difference when really it is relatively marginal in most cases.

[bioduken]

eed to be backed up with evidence if they are to be taken seriously.

 

Usually I don

[jammin]

It's true that most people won't know very much about processor architecture and how it works (ask me about it and I'm unlikely yo be able to tell you much in detail).

 

They just start one application, runt it for some time and then gets some sort of number.

 

Is that maybe a bit dismissive of using applications to test performance?

Applications after all, are what everyone has to use to get things done, so in the case where they measure differences in tangible concepts such as how much time it takes to complete a task (where the user will actually notice the difference in real terms providing the differences aren't in fractions of a second) they are entirely relevant regardless of architecture.

Many use multi-threaded applications in testing, but I can't recall many that use multiple applications running at the same time to measure multi tasking performance.

 

Anyway, I feel this discussion is veering away from the original topic of the thread somewhat.

I you want to talk about processor architecture further and how a majority of testing methods may be flawed in assessing performance, the feel free to start a new thread, I think it would be interesting.

 

As for getting into the server market where Opterons take on Xeons, it's a little different than comparing consumer desktop CPUs (despite the architectural similarities in the CPUs themselves).