Printable View
Jeeze...Quote:
Originally Posted by Movieman
Who gives a :banana::banana::banana::banana: about whether Conroe is faster than AMD in overall performance?
The point I am trying to make, is that Conroe is slower in 64-bit mode, than in 32-bit mode, while for the K8 it's the opposite.
So, one might say that AMD's 64-bit performance is comparatively superior to Core 2's (despite losing) simply because it GAINS in a 64-bit environment, whilst Core 2 gets a performance hit..
Got it now?!?
Yes, I understand your point. I really do, I just think it is meaningless.Quote:
Originally Posted by Carfax
So, by your logic, if Intel slowed their 32bit performance so that the 64 bit would be faster but it was still faster than AMD's then you would say that Intel had the better cpu?
My point is simple: It doesn't matter how Intels 32 bit scales against it's 64 bit since they are BOTH faster than AMD's.
I think you have settled on the one factor that you could find a chink in Intels armor and have promoted this for whatever reason is beyond me.
Bottom line is that both the 2660 and the 2930 Conroes from Intel are faster than the FX62 in both 32 bit and 64 bit. That is ALL that matters.
This is not a flame at all, this is how I see that diagram and my questioning your interpretation of it.
All of my comments were directed towards the one graph that was shown.Quote:
Originally Posted by informal
No other apps were implied or should be implied from my comments.
Carfax,i don't think they got it.Seems to hard for them to understand that A64 has superior 64 bit implementation.I see that C2D is still faster in that one benchmark,but all in all,A64 gains in 64b while C2D loses or gains very little..
if this really is true... if 64bit is really SO bad... then when vista comes out, ill just get the 32bit version, and continue to beat the AMDs even in 64bit mode...
disappointing, ill admit, if it really is slower in 64bit mode. but just like it has a FSB, i dont care how it works, but it works, and it works well.
Moveman,A64 lost in some benchmarks, while in some it won..
http://akiba.ascii24.com/akiba/colum...ages813899.gif
http://akiba.ascii24.com/akiba/colum...ages813900.gif
Ok, you have a point..If you compare the 2930 conroe vs the FX62 here on the 2 tests: conroe loses to AMD in 64 bit sakura editor while beating it in 32 bit.The 2930 then beats the AMD in both 32 bit and 64 bit in the zip app.Quote:
Originally Posted by informal
So in one of the 4 benchmarks conroe loses.
Thanks for bringing this up.
You're point is clear and consise. Faster = better. But why did they ever go to 64-bit in the first place??? To make things faster. I think its more important to see that AMD is going the right direction with their 64-bit architecture while Intel is STILL having problems optimizing it. All this means is that when 64-bit finally goes mainstream, Intel had better have this issue solved, otherwise this 15-25% gap in conroe vs am2 is going to dissapear.Quote:
Originally Posted by Movieman
I dont' see why you're making a fuss over this. I've seen a few 64bit vs 32bit benches on a A64 that have it end up performing worse (usually marginally) in 64 mode than in 32 bit. Ask me, it's less to do with the chip and more to do with the complier and/or program.Quote:
Originally Posted by Carfax
this is petty at best
Agreed, Here's the reality: AMD is a time tested cpu that has been and is a excellent processor. The conroe is the new kid on the block and looks to be a monster at first glance. We'll need 3 months of testing them in all sorts of envionments to see which is better at what.We may all be wrong on what we think at this point and we may all be right.Quote:
Originally Posted by Eastcoasthandle
My problem is that I love them all,AMD's X2's and Opty's I think are great cpu's. Intels have had issues in my mind since they went from P3 to P4 with the exception of the P4 xeons which i think they did pretty well with.
The new conroes and Kentsfields look to be excellent to me.I've been following the benchmarks of them in Rosetta and they are absolutely huge.
One of the guys just put a kentsfield on Rosetta 24/7 and it looks to make 8300 a day. The top AMD rig was uOpt's quad 875(8 core) at just over 7100 points a day. My DX3600/2mb/800 will pull 3500/day by comparison.
We will see what winds up being the bad boy on the block in short time.;)
These are interesting results, Carfax, thank you for bringing them to our attention. I hope an explanation of them will come forth from Intel in the fullness of time.
Ofcourse not.. You're hung up on Intel vs AMD as competitors, while I am solely hung up on Intel's 64-bit performance.Quote:
Originally Posted by Movieman
The reason why I am concerned, is because 64-bit is increasingly becoming more important. Mostly in the enterprise and HPC sector now, but in a few years, it will definitely be mainstream.
I am hoping that this performance issue is due to software, and not Intel's 64-bit hardware.
For the sake of competition, you're right. But this thread is about Intel's 64-bit implementation (as it relates to AMD's), and not Intel vs AMD as two companies who are trying to best one another..Quote:
My point is simple: It doesn't matter how Intels 32 bit scales against it's 64 bit since they are BOTH faster than AMD's.
Yes, how dare I criticize Intel :rolleyes: I should just sit back like a good little droid and ignore Core 2's 64-bit performance degradation..Quote:
I think you have settled on the one factor that you could find a chink in Intels armor and have promoted this for whatever reason is beyond me.
There are people out there that actually USE 64-bit for work you know.. Not everyone is a gamer/enthusiast.Quote:
Bottom line is that both the 2660 and the 2930 Conroes from Intel are faster than the FX62 in both 32 bit and 64 bit. That is ALL that matters.
I posted this for them..
If you actually read my first post, you'll see that I wasn't making a fuss over it, and that I even considered the discrepancy may be due to sub-optimized compilers/program.Quote:
Originally Posted by Pongi
I only got aggravated because I had to clarify my stance numerous times, aswell as my intentions.
Apparently, people think I posted this as some dig against Intel, which is completely and utterly ridiculous..
I have praised Core 2 as much as anyone else, but it is not a perfect cpu and it does have it's faults.
However, some people just can't seem to rise above their petty AMD vs Intel obsessions and look at the technical aspects.. :rolleyes:
Well.. then I'm confused about your point. Your premise seems to be that every app should be faster when compiled for any given 64-bit CPU than the same app on the same CPU in 32-bit mode. Empirical evidence would seem to prove you wrong. This theory (every app is faster when compiled for 64-bit mode) would seem, at first, or every, glance to be wrong.Quote:
Originally Posted by Carfax
So, either the world is wrong (Conroe, being the fastest x86 CPU proves you wrong) or you are wrong (your premise is simply incorrect).
Now. Tell me. Is Conroe not faster in pretty much every meaningful CPU benchmark than any AMD CPU (including 64-bit)? Or, dare I say, is perhaps your premise simply _wrong_?
I vote (durr) your premise is simply wrong.
But assuming that they use a FX-62, why wouldn't they be able to afford a E6700 which retails for much less? Theyd probably be able to afford an X6800 even :shrug:Quote:
Originally Posted by Carfax
Maybe intel will release some kind of 64bit drivers when it's released???
Maybe I should have just posted this as my orginal reply and it would have been clearer and more to the point:Quote:
Originally Posted by Carfax
As long as the Conroe shows that it is clearly superior, as it did in the three benchmarks that you and the other guy provided(3 benchmarks,32 bit and 64 bit in each) and of the 6 "tests" the AMD scores better in only one of the six, those charts made my point for me.
In 2 of the 3 64bit benches, the conroe was superior.
In all 3 of the benches at 32 bit the conroe was superior.
Why would one care if the scaling between the Intels 32 bit and 64 bit wasn't as good as the AMD's percentage wise since the conroes 64 bit performance is still showing as faster than the AMD'sin 2 of the 3 tests.
You mention that you didn't do this for the gamers. I assure you that if the gaming companies had to depend on me they would starve. Most of what I do is video encoding and it's the processor that does it the fastest is all that matters.
You also say that although 64 bit is not of great importance now it will be in years to come. I totally agree with that and I am sure that when that time comes in a few years both Intel and AMD will have made great strides in 64 bit.
The thing is that we are not living in the future, we are living in the present and in the present, the conroe looks to be the faster cpu in both 32 bit and in at least 2 of the 3 apps that have been used as examples tonight.
There are no petty obsessions on my part, I assure you. I looked at the chart you provided and it said the conroe was a faster chip and based on that I really didn't see the purpose of your post except to try to find a reason to trash the conroe.
We can debate this forever but faster is faster period. If both the 32 bit times and the 64 bit times for a benchmark are both held by the conroe what does it matter is the slower AMD cpu has faster times in 64 bit than it's own 32 bit scores. When one is buying, all things being equal, one buys the cpu that will do what ever your particular usage is in the least amount of time.
You say look at the technical aspects, I say you should look at the reality of why people buy what they do and faster is better..Plain and simple..If you can't understand that, then there's nothing I can say to you.
I assure you that if the AMD and the Conroes performace were reversed, I would be making this same statement in favore of the AMD cpu. I have no brand loyalties at all. None, I own: AMD,Intel,ATI,nVidia,WD,Maxtor,Seagate,Fortron,Ant ec,etc
I bought what was the fastest and fit my budget at the time.
Listen:
FX62 @ 64 bit is JUST AS FAST As x6800 is (32 or 64 bit) at least in cinebench.
So yes, the difference is large. I dont have a conroe to test with, so I cant comment on intel 64 bit performance (yonah is 32 bit only :| )
<3 yonah as a daily rig though, more than the FX :)
Don't forget to play state of the art games at 640*480 too, so your beating on AMD's will be even bigger lolQuote:
Originally Posted by AndrewZorn
Wow, the fear of C2D is with you. Look! C2D come with a dull IHS I guess now A64 is far superior now :fact: :woot: :eleph: :cheer: :party: :worm: :bananal: :banana3: :hump:
Give me a freaking break with this whaa, whaa over C2D :rolleyes:
To anyone reading some of the garbage in this thread...all this propaganda servers to do is to cause you to doubt yourself. That is the major agenda to these self proclaim truths that are petty, inconclusive and stupid. If you read the title of this thread it clearly gives away their ill willed, bias attempts of self indignation...that's all they are trying to achieve here. They want you to be just as afraid as they are. Don't let these individuals cause you to doubt what you intend to do regarding your next AMD or Intel Purchase stick to your guns. C2D overall is a good CPU and from what I have seen in 32 vs 64 bit OS there is nothing to fear in some cases 1/2 the cost of their competitor.
Here are more results remember, you can't relay on just 1 source alone. Also remember that we are not fully engulfed in 64 bit OS as of yet so more tests will be needed once everything is mainstream. Also note, that 64 bit OS operations will require better optimized drivers, etc. This is something we simply haven't seen yet because we are not fully 64 bit yet...get it?
could this be that the benchmarks are not compiled for the c2d? does the PentiumD line gain from 64 bit in the benchmarks?
i think this will be taken care of by intel if its software related.
Could be that they made some changes to the architecture wich some, or lack, of optimization compiler-wise impact 64-bit performance.
They can fix it, but in windows you will probably never see the changes, because software companies will use very little optimization so the software can be compatible with all the cpu's.
Can also be some defect in the chipset's drivers.
Can't someone here in the forum run some benchs? Just to clear this out.
EDIT:
http://www.planetx64.com/index.php?o...1&limitstart=6
Hmmm, these benchs show normally performance increases for c2d but, in this bench the gain is too low compared to amd's. And from what i've seen pentium4/D showed more gains than amd cpu's, now it's the opposite.
The move to 64bit was for greater addressable memory and 64bit precision, speed was a byproduct. If you recall, Intel is bad about NIH and will and have shot themselves in the foot before for spite(really for PR/image but it boils down to the same). They tried to kill off AMD's 64bit extension method and then later claimed AMD stole it from them after Intel invented it first. Now they come out with a chip that does meh to poor in that mode. Considering that by the end of the year 40% of Intel's production will be Core chips and that 20% of Intel ='s 100% of AMD production, it means they'll soon be more Intel Core chips than AMD so if the majority of chips out there don't do well performing a certain task, what do you think programmers are going to do? They're going to do as they've always done and cater to the majority. Intel is using their size to squeeze out anything non-Intel, in this case AMD 64bit extensions, out of consciousness. Intel is more than capable of producing a fast chip which leads me to believe the poor showing in 64bit is political. AMD have often come up with a better play but it's always been Intel's game(x86, mmx, squash 3dnow for sse). AMD 64bit extensions is AMD's game and Intel isn't having that.Quote:
Originally Posted by dr_sharp
wrongQuote:
Originally Posted by fhpchris
http://www.techreport.com/reviews/20.../index.x?pg=13
So, one might say that AMD's 64-bit performance is comparatively superior to Core 2's (despite losing) simply because it GAINS in a 64-bit environment, whilst Core 2 gets a performance hit..
Got it now?!?
you do realize that depending on how it was compiled, can make a huge difference in performance.
http://www.techreport.com/reviews/20...2/index.x?pg=6
They are both good CPUs. Period.