Why are intel mobos so much better than amd?

I read where some people running their intel canterwood at 300mhz which of course gets quad pumped, which means that 1200 million 64-bit batches are sent from the northbridge to the cpu every second.

But the best amd board, an nf7-s, can go to maybe 240mhz double pumped, which results in only 480 million 64-bit batches per second.

Why is it that intel motherboards are able to scream along at almost triple the data rate? I am looking for as much detail as anyone wants to toss out.

Maybe the AMD doesn't have as many CPU cycles as the Intel does, but it has better traction and doesn's spin out in the mud as much. Intel is trying to get more work done by running as fast as they can. AMD tries to get more work done per CPU cycle.

Over all the Intel might edge out the AMD but not by nearly as much as the numbers would suggest.

I totally agree audioguy. I'm just trying to understand why amd chipsets are lagging. (Not that it matters now that the A64 has the memory controller on-die)

Whatever the reason, AMD can run a processor at 1.8 something Ghz (Barton 2500+), and perform similarly to a P4 2.5 ghz.

Try comparing apples to apples - you're asking a question which basically reads "Why don't two completely different designs work the same?"

TC said:

Try comparing apples to apples - you're asking a question which basically reads "Why don't two completely different designs work the same?"

Click to expand...

I don't think so. Each northbridge communicates with a cpu over a 64-bit wide fsb path. I am not talking about the internal design of the cpu, I'm talking about the northbridge's ability to run at high data rates.

johnoh said:

I don't think so. Each northbridge communicates with a cpu over a 64-bit wide fsb path. I am not talking about the internal design of the cpu, I'm talking about the northbridge's ability to run at high data rates.

Click to expand...

They are still two totally different platforms. The CPU is different therefore the NB design is different. Unless you can make a NB that can work for both processors you are never really comparing the same thing.

I'm just looking for some detail as to why there is such a big difference in nb data rates. I realize the cpus are different.

AMD CPUs don't need as many clock/machine cycles as the Intel CPUs do. So the AMD chipset doesn't need to go as fast. AMD double pumps the memory while Intel quad pumps it. Just a differant design philosophy is all.

Why would you consider a more efficiant CPU/chipset as inferior just because it goes slower while accomplishing the same amount of work?

johnoh said:

I'm just looking for some detail as to why there is such a big difference in nb data rates. I realize the cpus are different.

Click to expand...

Because the bus designs are completely different.

Audioaficionado said:

Why would you consider a more efficiant CPU/chipset as inferior just because it goes slower while accomplishing the same amount of work?

Click to expand...

I'm just curious is all.

The memory controller on the northbridge just sends bits to the cpu. It does not execute complex instructions the way a cpu does. The amd system is efficent, the amd cpu does more work per clock cycle, but the nb goes less than half the speed of an intel nb. I was hoping to find out why and its cool if nobody knows, I mean hell I don't know either so I can't complain, but "the design is different" isn't really an explanation.

An amd fsb increase with no change to cpu ghz gives you 50% more throughoput improvement than does an fsb increase on a p4, which says that faster amd chipsets would make a big difference. Here is the evidence regarding the 50% number...

http://hardforum.com/showthread.php?s=&threadid=611788

We just told you why. Intel quad pumps the memory because that's how their engineers designed the Intel platform to run.

AMD only double pumps it's memory because it wasn't designed to quadpump since it's uneeded the way the CPU processes a machine cycle.

Obvously if you quad pump the memory, more bits are going to be running through the NB than if you only double pump it.

If you'd like the techical details, both Intel and AMD have lots of white papers on their designs at their websites. Just google over there and see what you can find on the subject.

Not being an Intel Fanboy, (I was at one time) I think that the AMD based boards do as much "work" as an equivelant Intel board.

I have a TbredB 1700+ (the good one, it's Folding at 2580 MHz), and it will do comparable work to a 3.3 GHz P4 CPU. I don't do benchmarks, I'm going on Folding times, and this one is the fastest rig I have, and the slowest out-of-the-box speed. It even outperformed a 2.4B P4 that I had on an IT7.

steve

Audioaficionado said:

We just told you why. Intel quad pumps the memory because that's how their engineers designed the Intel platform to run.

Click to expand...

Wrong.

Intel quad-pumps the FRONT SIDE BUS. This is totally different than quad-pumping the memory. The front side bus exists between a northbridge and the CPU; the memory bus (which is double-pumped on ANY DDR-based memory platform) exists between the northbridge and the memory.

Any memory bandwidth differences between the two are a result of a) different memory controller designs, b) the speed of the memory in question, and c) the number of channels the memory communicates with the northbridge on.

The front side bus differences are strictly a function of how many times per clock cycle the CPU can transmit to the northbridge and how many cycles per second this transmits. The newest Athlons and Pentium 4s BOTH communicate with their northbridge controllers at 200 MHz; no more, no less (P4C/Athlon 3000/3200.) The main difference lies between the Pentium 4 being able to transmit 4 times per clock cycle for an effective "800MHz FSB." The Athlon is only able to do this 2 times per clock cycle, hence a "400MHz FSB."

Do not confuse the front side bus with the memory bus. They are two independent things. Hence, asynch memory and FSB timings that are possible between a CPU and its memory bus. It doesn't always work efficiently, but it still works.

Now, for the differences in performance that the original poster mentioned. Intel boards are able to scream along at very high effective front side bus speeds because, well, they're designed to! However, most of these transmit cycles are latent; the P4 is figuring out what to send to the northbridge; there's usually not much chance of saturating its own FSB for very long periods of time. Most of the cycles go to waste. The AMD CPUs are able to use more of the bus more efficiently; even if the AMD could quad-pump its FSB, most of these cycles would still be going to waste, anyhow. There's a limit as to how much each CPU can do, and it's usually well below the actual capacity of the memory and Front Side busses.

Now, if you put a 5GHz Athlon in that same place (with the effective 400MHz FSB,) it'd probably have very little trouble turning that previous wide-open bus into a massive bottleneck. The same would go with a 10GHz Pentium 4.

As to the raw clock speeds that the chipsets work at, 250FSB on a Pentium 4 board ("1000FSB," and a reasonable goal) and 230FSB on a good nForce2 board ("460FSB," also within reason) really aren't that much difference. It's raw clock speed that determines a northbridge chip's ceiling; the rare instances where a board can scale up to 300 MHz (as on a few 865/875 boards) is just "one of those chips," usually just an unusually good one out of Intel's batch. Ever notice you never read of the SiS P4 boards getting even close to that speed? Plain and simple, Intel constructs a better northbridge. Nothing against SiS, nVidia, or VIA, but Intel just fabs 'em better...

Thanx for clearing that up theflyingrat

No prob.

Of course, if I managed to screw something up in my [obviously] abbreviated explanation of how this junk works, someone please inform and correct, and reprimand me with the requisite tongue-lashing.

Remember the P4 was designed for rambus, so it is quad pumped, but at 16BITS. The K7's are all 32bits. However, FSB was not the subject of the original question, memory speed was. When I was a student at naval nuclear propulsion school, the instructors used to write in big red letters on our papers "GCE", which means "gross conceptional error". Which is what we have here. The I875P & I865PE are dual channel memory. So is the nVidia Nforce2. In fact, both systems can, in many cases, run the same memory modules. Therefore, what you actually have is AMD @ 225+ and Intel @ 250+. It really doesn't come down to that much of a difference. Intel probably has a few tricks that they have kept to themselves. As demonstrated by the superiority of thier chipsets compared to SiS and Via. AMD, conversly, has had to be more open with thier alternate chipset suppliers. nVidia has had quiet a bit of success designing memory controllers, in case you've never heard of their video cards. All things considered, they do a damn fine job, being a much smaller company than Intel.

MuEagle05 said:

Whatever the reason, AMD can run a processor at 1.8 something Ghz (Barton 2500+), and perform similarly to a P4 2.5 ghz.

Click to expand...

I'm frustrated so I'm sorry if I offend anyone.

It'll run the same programs but not NEARLY as well. I just sold my Intel 2.4C system and bought an AMD 2500+ Barton, and I'm having troubles galore. After RMAing the mobo, CPU and memory, I am still having problems.

Intel is better simply because they make better quality parts. AMD might be able to do mroe per clock cycle, but it seems to me that these chips AMD spurts out can't handle what's going on in their clock cycles, so doing more per clock cycle may be a good thing, but when it's done with so many errors, it sucks like crap.

All my games stutter like crazy, lock up for upto 10 seconds, crash to the desktop, etc. etc. etc. with my AMD system. With my old intel, I've had 0 problems.

That's why Intel boards are better. Simple because they don't have problems.

Amino said:

I'm frustrated so I'm sorry if I offend anyone.

It'll run the same programs but not NEARLY as well. I just sold my Intel 2.4C system and bought an AMD 2500+ Barton, and I'm having troubles galore. After RMAing the mobo, CPU and memory, I am still having problems.

Intel is better simply because they make better quality parts. AMD might be able to do mroe per clock cycle, but it seems to me that these chips AMD spurts out can't handle what's going on in their clock cycles, so doing more per clock cycle may be a good thing, but when it's done with so many errors, it sucks like crap.

All my games stutter like crazy, lock up for upto 10 seconds, crash to the desktop, etc. etc. etc. with my AMD system. With my old intel, I've had 0 problems.

That's why Intel boards are better. Simple because they don't have problems.

Click to expand...

Pardon my opinion, but what you just said goes completely against reason. I don't see why anyone would be offended by it, but I don't see why anyone would take your post as much of a statement about any CPU either.

You obviously don't know what the problem is with your system seeing as how you haven't fixed it yet and there are other people with very similar setups that don't have problems.

For you to lay the blame anywhere is a bit silly, unless you fix things and find exactly where the problem is eminating from. RMA'ing the CPU, memory, and mobo is a bit like stumbling around in the dark and hopeing you stub your toe on what your looking for.

The problem could lie in many places, but even replacing the CPU and finding it solves the problem doesn't mean that it was the CPU's fault - your cooling could have been incorrectly configured or mounted. That's just an example, it likely could not even have anything to do with the CPU.

All I'm saying is this:

When I build an Intel computer, I don't have problems. When I build an AMD computer, I have loads of problems. I think that says plenty.