New Intel Destroys AMD

Has AMD release any details on their competiting architecture? Wikipedia has some info on Conroe, I read it first before checking out the graphs, and it said something about intel highlighting power consumption over performance.

Would be nice to see what AMD are capable of- competition is nice for the market :)
 
I had all my things picked out, and now I see this....confuses things!!!
 
Oh boy, it starts agian.


well Bman, these things are made to confuse people.

but my thing with this is why do they test software that is made for a single core on dual core hardware.

if you make the software for dual core it will perform better. so all this is really saying is that at the current moment with a single core OS and a Single core app, the intel is faster but what happens when vista and a proper dual processing app is run on it???
 
BMAN

PC upgrade rules:
1) Don't pick hardware until the day you are going to buy.
2) Don't check prices after you hit the enter key on the order.
3) After assembling your system and being happy with it do not look at benchmarks on any other hardware.
4) Don't compare benchmarks on your system with other peoples benchmarks at 3dMark.

These rules were created to keep people from torturing themselves.
 
Good set of rules...
 
Thanks, damn shame I can't force myself to follow them, lol.

Back to the anandtech article:

The test was set up by Intel which means the MB's, chipsets, etc were selected to show the maximum difference between the AMD and Intel setups. If they weren't the VP of marketing would have the VP of engineering fired. Other MB's may yield different results

They used systems based on the ATI RDX200 and SB450 southbridge for a crossfire compliant setup. RDX200 was last years AMD crossfire chip. I'd like to see the same tests run on an NF4 board with an SLI rig (and VIA chipsets too) before I made any assumptions.

The Quake 4 and UT2004 demo were run using an Intel demo recording, this means optimizations for their test setup could have been applied. The FEAR demo was stock.

The AMD system was crippled with DDR 400 RAM. Performance with DDR2-800 RAM will be the benchmark for this years top systems, not the DDR2-667. AMD is holding off the AM2 release until the DDR2-800 is widely available because their integrated memory controller benefits greatly from it compared to -667 RAM.

The FEAR and encoding tests were impressive and less likely to doctoring than the video setup. But remember we're talking DDR2-667 vs plain jane DDR 400 on the AMD.

I'm surprised the Conroe could produce that large a difference in video performance. The CPU should not have that much impact unless some feature of the MB chipset is choking the video card performance.

But don't forget this is Intels big shot for the year. AMD has not launched theirs yet. Stay tuned for the independent tests of AM2 vs Conroe this summer.

PS - BMAN, unless you feel like waiting 9-12 more months for product to stabilize and prices to drop off the initial price peaks there is no reason to reconsider your buy.
 
Last edited:
Very good to know, thanks...can always upgrade again later
 
did anyone try comparing benchmarks with the same cpu (amd x2 2.8 was it?) on a setup that was in amd's favor? (ie nforce 4 sli setup)
im sure benchmarks would be greatly increased
like leejend very well pointed out, these benchmarks were setup and ran on a system that tailored to intel cpus. i'm sure the amd cpu was choked somehow or some way...but as leejend also pointed out, this is a new chip vs old chip, and amd hasnt yet done these tests on their new lineup. can't wait to see new vs new or current vs current or however you want to put it. this is like comparing soldier to super soldier...i would hope that newer technology would be more effecient than older technology! typical. intel has a very f'ed up way of doing their pr, and if it isnt pointed out by a trained eye, consumers will walk blindly into it. intel targets amatuers. smart but tasteless
 
LeeJend said:
Thanks, damn shame I can't force myself to follow them, lol.

Back to the anandtech article:

The test was set up by Intel which means the MB's, chipsets, etc were selected to show the maximum difference between the AMD and Intel setups. If they weren't the VP of marketing would have the VP of engineering fired. Other MB's may yield different results

The motherboard in question is a GOOD motherboard.

The new marchitecture from Intel is quite competitive. Remember what I posted a few months back about the upcoming procs from intel and the multi-core solutions?

The performance is actually pretty stellar and as more mature hardware comes out to support the procs, you will see better performance still :smoker:

They used systems based on the ATI RDX200 and SB450 southbridge for a crossfire compliant setup. RDX200 was last years AMD crossfire chip. I'd like to see the same tests run on an NF4 board with an SLI rig (and VIA chipsets too) before I made any assumptions.

This should not make an iota of difference since the video cards used were the same. If you used an SLi rig, there should perhaps be a closer performance delta but not a massive closing. Conroe is a GOOD processor.

The Quake 4 and UT2004 demo were run using an Intel demo recording, this means optimizations for their test setup could have been applied. The FEAR demo was stock.

It COULD have been optimized. Given that FEAR was run stock, it COULD also be just that powerful.

The AMD system was crippled with DDR 400 RAM. Performance with DDR2-800 RAM will be the benchmark for this years top systems, not the DDR2-667. AMD is holding off the AM2 release until the DDR2-800 is widely available because their integrated memory controller benefits greatly from it compared to -667 RAM.

You realize that there are higher latencies involved with DDR2 and you also realize that the A64 proc has an on-die memory controller right? How then do you consider it to be crippled. Yes, theoretical bandwidth is better but this is not apparent in all situations.

The FEAR and encoding tests were impressive and less likely to doctoring than the video setup. But remember we're talking DDR2-667 vs plain jane DDR 400 on the AMD.

This hasn't mattered so far for Intel v/s AMD benches, it should still not matter.

I'm surprised the Conroe could produce that large a difference in video performance. The CPU should not have that much impact unless some feature of the MB chipset is choking the video card performance.

Efficiency is much better.

But don't forget this is Intels big shot for the year. AMD has not launched theirs yet. Stay tuned for the independent tests of AM2 vs Conroe this summer.

You'd again be surprised at what Intel is currently working on besides conroe.
 
Are these the same benchmarks that were put up on Anandtech earlier? I was talking to someone and he told me that they were comparing a next-gen Intel proc with a current gen AMD one.

If what I just said has nothing to do with the topic, ignore this post :p
 
I've seen mention to this, and am left with the impression that a next gen Intel part is being compared to the current gen AMD part. Of course we don't know what the next gen AMD part will be, what it will be capable of, how it will stack up, etc... Time will tell. In the end, it's good for each company to have a time one upping each other. It keeps them from ending up in a rut, where much progress isn't made.

R&D of course costs money... And if one didn't have to spend money on R&D, to try to increase sales, there could well end up being less insentive to do so...

Course, speaking of Vista, there is one other consideration. Most of the benefits we had seen were from the on-die memory controller. If Vista-64 is released with decent 64-bit driver support, and especially if one starts seeing 64-bit compiled apps; we haven't necessarily had a good test to see what our own procs can do. There's a lot of hardware in there, we simply aren't using, because of making our OS run 32-bit OS, drivers, and software. And no reason to mention the reasons for this, as I stuck with XP-32 myself.

There are registers and other hardware on the A64s however, which is only accessible from within long mode however; so in the mode we're running them in, we don't make use of...
 
I'm still jaded from when NVidia trumpeted that the 5800 would blow any existing video cards (9700 Pro) out of the water. Sometimes they even claimed it would double performance. My current stance on any tech company's claim is to take it with a spoonful of salt.

I want to believe this is true though. I want to see great performance, but I'm nobody's fool. I'll believe it when I see it; hopefully Intel won't let me down.
 
Disregard what I formerly said about not seeing how A64 would perform with actual 64-bit software support; because from what I'm finding out and hearing, Conroe will be a 64-bit CPU. So any benefit AMD will see wrt this, Intel will see also.

But besides that, this is even better news for the consumer. Given the release of Vista in both 32-bit and 64-bit flavors, once both Intel and AMD comes out with 64-bit CPUs; there won't be much reason for devs to hold back too much longer on 64-bit software and driver support. On the Vista drivers, these companies have to make it anyway, so...

Course, at first it will be expected that some older 32-bit CPUs are still floating around. But in time, and after x86-64 procs have been sold by Intel and AMD for awhile, they're won't be much reason left for devs to not develop/compile their software to 64-bit. In the end, we win wrt the software side of things as well...
 
Holy crap, lots of information!! It's crazy how far we are going ahead with technology. I can't wait to get my hands on some of these things. I'm glad I'm of the newer generation, so I can live my whole life testing out all these crazy new things that will come out lol
 
why is everyone spending so much time justifying why a next gen chip is better than a current gen chip? leejend hit the nail on the head...it's going to get good when am2 is compared to the conroe, and my money is on amd!

we're just wasting time comparing an old chip to a new one...whats the point in all this?
 
I am waiting for the "waaah waaah, Dell should use AMD" crowd to chime in soon :)

Also Matt, it should be Madimux Mattimus no?

:smoker:
 
wind3D said:
why is everyone spending so much time justifying why a next gen chip is better than a current gen chip? leejend hit the nail on the head...it's going to get good when am2 is compared to the conroe, and my money is on amd!

we're just wasting time comparing an old chip to a new one...whats the point in all this?

Why not?

Conroe is imminent, AM2 is on its way and the procs have not been demoed like this.

Remember when the 9700pro was similarly debuted? This gives us a lot of things to consider and the performance is very good. It shows that Intel has moved away from the p4 configs and gone with a different/multi-stage proc which has a far more efficient design.

Further, considering the massive impetus on TDP, the Merom/Conroe configs will likely also have low thermals comparative to the older prescott era procs as well as having better bandwidth throughput to the memory.

The northbridge design is revamped. Intel kept the mem controller off the proc itself in an effort to reduce overall TDP which implies that AMD will have thermal issues with the on-die mem controller vis-a-vis Intel in coming iterations.

Not a really big issue since AMD has done a good job with their thermals, but as Intel gets cooler, you will see more and more wins for them in different segments.

Discussing this is good because it gives us an understanding of what is going on :) Not discussing it simply because your preferred IHV/OEM or whatever else is not being shown in the best possible light is counter-productive.
 

Members online

No members online now.

Latest profile posts

Also Hi EP and people. I found this place again while looking through a oooollllllldddd backup. I have filled over 10TB and was looking at my collection of antiques. Any bids on the 500Mhz Win 95 fix?
Any of the SP crew still out there?
Xie wrote on Electronic Punk's profile.
Impressed you have kept this alive this long EP! So many sites have come and gone. :(

Just did some crude math and I apparently joined almost 18yrs ago, how is that possible???
hello peeps... is been some time since i last came here.
Electronic Punk wrote on Sazar's profile.
Rest in peace my friend, been trying to find you and finally did in the worst way imaginable.

Forum statistics

Threads
62,015
Messages
673,494
Members
5,621
Latest member
naeemsafi
Back