A different take on the 760MP and nForce from Hardware Analysis...no grinning over there

[NFS4]

With all the positive light swirling around the nForce and 760MP, this article is in sharp contrast from Hardware Analysis (formerly Hardware Central):

http://www.hardwareanalysis.com/content/article/1284/


<< For AMD to give Intel a run for their money and make their x86 SMP architecture the de-facto standard in the workstation market they should not have released their 760MP chipset with two 1.2GHz Athlon MPs but rather 1.7GHz or even 1.8GHz versions of the CPU instead. They should've rather waited for these CPUs to become available and in the meantime focus on the consumer market. The workstation market isn't one that is forgiving, but is one that puts performance, stability and compatibility first, before any price considerations. If the 760 MP and Athlon MP CPUs had been introduced at 1.7GHz or 1.8GHz clockspeeds they would have swept the ground from under Intel's offerings, probably both in price and performance. From the very start AMD would then have had a product that is not only an alternative, but also an obvious choice for anybody wanting the highest performing x86 workstation. Now they unfortunately have to face another clockspeed battle and prove themselves in the workstation market that they could've had in their pocket from day one. >>



Ummm, aren't they already spanking the Xeons in price and performance????


<< Another story is nVidia, the same company that more or less persuades you to buy a new graphics accelerator every six months or so. Apart from designing graphics accelerators and diligently working with Microsoft on the upcoming X-box gaming console they've announced their Athlon DDR chipset, the nVidia nForce. Not to be outdone by anyone, the specifications and prospective performance figures show us a chipset that will give the upcoming Intel i845, AMD's 760 or VIA chipsets a run for their money. For a company that has never before designed a chipset or any core-logic other than graphics accelerators, I think that that is a rather bold claim to make. >>


I don't think they give NVidia enough credit

 

[gunf1ghter]

Sounds like sour grapes to me...

We'll just have to see what the real world performance and stability of the nForce is when it has been released and people like Anand that tend to be impartial have a chance to test it.

 

[Davegod75]

stupid author is going on Mhz alone. Little does he konw that the athlon 1.4's will smoke any p4 1.7Ghz or Xeon

 

[NFS4]

<< Sounds like sour grapes to me...

We'll just have to see what the real world performance and stability of the nForce is when it has been released and people like Anand that tend to be impartial have a chance to test it. >>


Seems kinda funny for them to be making statements about the 760MP performance/price-wise when they don't even have any benchmarks or a review up on their site of the platform.

 

[Shmorq]

<< stupid author is going on Mhz alone. Little does he konw that the athlon 1.4's will smoke any p4 1.7Ghz or Xeon >>

But most consumers don't look at benchmarks and stuff so they'll think the 1.7GHz is automatically faster. Maybe AMD should go to the Pentium equivalent rating...

 

[bigbootydaddy]

he does have a point, his first statement was how to completely crush intel in workstations, it sounds like a reasonable statement...

intel is not going to roll over and die just cuz whats out on shelves right now, amd zealots need to realize if they say enough is enough, enough will be enough.

 

[NFS4]

<<

<< stupid author is going on Mhz alone. Little does he konw that the athlon 1.4's will smoke any p4 1.7Ghz or Xeon >>

But most consumers don't look at benchmarks and stuff so they'll think the 1.7GHz is automatically faster. Maybe AMD should go to the Pentium equivalent rating... >>


&quot;Regular&quot; consumers won't be buying MP's. IT professionals and corporations will be buying 760MP's. Those folks tend to know more about the performance of the different platforms than regular consumers.


<< he does have a point, his first statement was how to completely crush intel in workstations, it sounds like a reasonable statement...

intel is not going to roll over and die just cuz whats out on shelves right now, amd zealots need to realize if they say enough is enough, enough will be enough. >>


In the CPU world, nobody absolutely CRUSHES anyone in performance. It's normally a marginal victory at best. The original Athlon didn't &quot;crush&quot; the PIII. The Thunderbird doesn't &quot;crush&quot; the Pentium 4, and Dual Athlon MP's don't &quot;crush&quot; dual Xeons. But the difference is noticeable.

For them to expect AMD to come out with 1.7 and 1.8GHZ processors NOW is asking too much. Hell, the 1.5GHz Palominos aren't even here yet.

And as for the statment of waiting til those CPU's to be available to come out with 760MP. What's the point of that? If AMD had waited, Intel would have 2.1GHz Xeons and AMD would have 1.8GHz Athlon MP's. We'd still be in the same situation. You've got to start somewhere.

 

[Shmorq]

<< IT professionals and corporations will be buying 760MP's. Those folks tend to know more about the performance of the different platforms. >>

I wish this was the case at work...

Those guys are saying how everything that nVidia has put out so far is only in writing. But Hardware Analysis is doing the exact same thing. (Like someone mentioned, no benchmarks)

 

[Nitemare]

Isn't it amazing how the uninformed can state their opinions as facts? But, what is more amazing is most people swallow it hook, line and sinker. I think AMD needs to abandon the Mhz rating and go to something else. All the ignorant people, including some in IT, just see the higher number and the higher price tag and &quot;know&quot; it is superior. I work with othe engineers on a daily basis and some of the ignorance really surprises me.. There should be some super type test that combines business applications, graphics, database and other tests into one rating..and rates it according to other CPU's kind of like sisoft sandra. One super test, with Intel CPU's scores matched against AMD's...and...umm Cyrix's..hehe. This should help somewhat. If Intel is the only one using Mhz and both AMD and everyone else switches to this other test, Intel's precious Mhz advantage will mean nothing.

 

[GL]

The author of that article - Sander Sassen. I swear I've seen his name on crappy articles before. Time to do research. You know, Sony just got caught with a fake reviewer pimping their movies. How do we know Intel isn't doing the same (only in a negative fashion against their competitors)?

AHA!
This is the same Sander Sassen that denied the PIII 1.13GHz had a problem - Linky Linky!



<< Tom,
I must say you've made a total joke out of yourself with your latest 'update.' Quite frankly, I have had no problem whatsoever with the 1.13GHz Pentium III and I'm using the exact same motherboards (Asus, Via and i820 chipset as well as the old P2B) as I presume you're using. If you cannot refrain from posting erroneous and obviously biased reviews or updates or whatever you like to call 'em, then at least make sure your lab is correctly setup and all benchmarking and system assembly operations are performed by a person capable of doing so. If I had a faulty CPU I would check back with the manufacturer in the first place and have it replaced BEFORE telling the world of your 'mishaps' or posting any benchmarks. You're mis-informing your audience as well as making your own testing procedures as well as your ethics a joke! >>



This is the same Sander Sassen that said Rambus was God and DDR SDRAM would never make it Read this article about how he says DDR-SDRAM boards will have to be 6-8 layers (cough nForce 4 layers, cough) - Linky Linky


<< However, due to the growing demand in memory bandwidth, the arrival of GHz CPUs and the ever-growing demands of today's software, SDRAM seems to have run into bandwidth limitations. Whereas DDR SDRAM and VCDRAM might be able to hold off the introduction of a new memory standard for a while, it is inevitable. While DDR SDRAM might promise increased memory bandwidth, it will run into severe timing, latency and propagation delay problems due to the wide databus and ever increasing clockspeeds. Memory may then be cheap to produce, but motherboards will then need to have six or even eight PCB layers to be able to run these memory modules at such high data rates and clockspeeds, thus increasing motherboard costs substantially. >>



Need I go on further...I've only skimmed the Google surface

 

[Noriaki]

The problem is that IT professionals still won't consider AMD. It's all intel all the time.

Right now I work in the Seismic processing division of a large oil company. We are using some x86 machines for some of our work.

A guy across from me just had to buy a Gig of Rambus for $2300 (canadian, and yes it's still a bad price, but he has to buy from a specific vendor), and use a P4 1.5Ghz....

Ok....clue time: This Seismic work is absolutely FPU itensive. The Athlon would DESTROY a P4 in this work (if it was SSE2 optimized that would be different, but it's not, just x87 FPU). But still he can't convince his superiors to even try and Athlon machine. IBM has AMD machines available (IBM is our main PC vendor), but we won't even look at them.

IT professionals are NOT that smart.

Though you are right, the consumer/smp market is tiny.

AMD has an uphill battle either way.

As for the nForce...well I think people are getting just a little hyped up over an integrated chipset. Good for the OEM market yes. Good for nVidia's pocket book, definately. But good for the enthusiasts? I doubt it....what does it offer us?

Wow...GeForce2MX integrated graphics...wow!
Some weird onboard sound card that does some sort of DD encoding...hmmm 6 high quality analog outputs with something like A3D applied would be better for gaming I think...

I'm sure that the nForce will be on par with the other AMD platform DDR chipsets....but I don't know why people think it's so great...

Hardware pre-fetch is really the only useful piece, and that only when you have a TBird or Duron...becuase the Athlon4 does it already...

Is it just becuase nVidia makes it that everyone thinks it's goin to be the greatest thing for PCs since memory that isn't hard soldered to the motherboard?

I hope that the nForce works out...the AMD760 always gets stuck with that awful 686B...and while I got mine working fine, I'm still leary to suggest it...Via has screw up after screw up after screw up.
The MVP3 Socket7, the Pro133A that can't perform with a BX, the 686B...and so on...I don't really trust them. ALI's DDR chipsets perfromance is pretty poor...and Sis...well it looks hopeful but they don't have a very good reputation in the north american market and I don't think mobo manufactureres really like them either....

So really the nForce could be quite helpful, a good solid DDR chipset for the Athlon at last....but it's hardly as revolutionary as everyone makes it out to be.

 

[GL]

<Cough> Even more drivel. Rambus is using Sander Sassen in its own press releases!

Linky Linky

This is almost too good. The next time I try to sell snake oil I might just hire this guy:

 

[Duvie]

Even if we had mp athlon chips of 1.7 to have a direct mhz to mhz comparison Intel zealots would cry &quot;you can't compare them since they are different architecture&quot;. You can't win witht hese people regardless of what the numbers say.

If intel, being embarrassed by getting beat to the 1 gig mark, hadn't incorporated their leatest marcketinh scam (I mean scheme) we could have a level playing field...level in the sense that the average sheep the consumer equate higher mhz directly with performance...Intel jumping from the first failed 1.13 p3, late downgraded to 1gig p3 to 1.3 gig p4 is where it all began. That is about the current difference 300mhz yet chips are nearly equal.

 

[Texmaster]

<< stupid author is going on Mhz alone. Little does he konw that the athlon 1.4's will smoke any p4 1.7Ghz or Xeon >>



Now exactly a fair comparison since AMD doesn't make a 1.4. The highest as far as I know is 1.33 and that is about even with a 1.7 P4.

And I'm curious how AMD is going to attack the heat question with 2 of thse hot tamales in there.

 

[Texmaster]

<< Even if we had mp athlon chips of 1.7 to have a direct mhz to mhz comparison Intel zealots would cry &quot;you can't compare them since they are different architecture&quot;. You can't win witht hese people regardless of what the numbers say.

If intel, being embarrassed by getting beat to the 1 gig mark, hadn't incorporated their leatest marcketinh scam (I mean scheme) we could have a level playing field...level in the sense that the average sheep the consumer equate higher mhz directly with performance...Intel jumping from the first failed 1.13 p3, late downgraded to 1gig p3 to 1.3 gig p4 is where it all began. That is about the current difference 300mhz yet chips are nearly equal. >>



Obviously you dont like Intel but I dont understand why you think its &quot;crying&quot; to point out that the 2 processors are of different archectectures? Thats just fact.

 

[Noriaki]

<< And I'm curious how AMD is going to attack the heat question with 2 of thse hot tamales in there. >>

With the Delta45 obviously :disgust: (50dbA 45CFM 60mm fan if you don't know).

 

[Muerto]

I dont' think this guy quite knows what he's talking about. He said that AMD should have released an SMP chipset when the orignal Athlon was released. The reason why they didn't is that they coudn't afford to be making Athlons, 750 chipsets, and SMP chipsets at the same time (as well as the rest of their product line). Plus, and SMP chipset has to tbe near perfect. There can be NO stability issues or other bugs or else it wont get any attention at all.

The nForce chipset has just been released and already he's trashing it. It's not even in any motherboards yet! True Nvidia hasn't designed a chipset before but he could at least give them the benefit of the doupt. I didn't see him running any tests or benchmarks on it, nor on the 760MP for that matter.

 

[Charles]

<< Ok....clue time: This Seismic work is absolutely FPU itensive. The Athlon would DESTROY a P4 in this work (if it was SSE2 optimized that would be different, but it's not, just x87 FPU). But still he can't convince his superiors to even try and Athlon machine. IBM has AMD machines available (IBM is our main PC vendor), but we won't even look at them. >>



Intel and Pentium are famous brand name, and the Pentium processors been around for many years. And that is Intel's strong point.

AMD has been known for producing a low cost processors for many years, and only until 1999 they release their high performing x87 processor, Athlon. And in order to gain market share, they have to price their product lower than their competitors, and it might take them a while to promote Athlon as a high performing chip.

AMD needs people like us to prove that Athlon is not bad (I don't want to start a flame, so I don't want to say that they are superior). Guess what, I've succesfully convinced about 10 friends of mine to buy Athlon/Duron system. It turned out that they really like it, and those 10 people convinced their other friends to buy AMD system rather than Intel's.

 

[Noriaki]

Oh I know charles, I understand completely why they use Intel only. I was saying to NFS4, that these are &quot;IT professionals&quot; and they won't even look at AMD. He said AthlonMP is for the IT professionals...but in my experience that market is ignoring AMD altogether.

 

[ElFenix]

texmaster - the 1.4GHz athlon was released today, as well as the duron 950.

nvidia does have more experience making DDR controllers than anyone else, you'd think if anyone could make a good one they could.

 

[NFS4]

<< Oh I know charles, I understand completely why they use Intel only. I was saying to NFS4, that these are &quot;IT professionals&quot; and they won't even look at AMD. He said AthlonMP is for the IT professionals...but in my experience that market is ignoring AMD altogether. >>


I was mearly stating that IT people would be the ones buying 760MP's (if anybody at all buys them), not the &quot;regular&quot; consumer (and WE make up a very small percentage of the buying public). Nothing more, nothing less. Those IT guys typically know more about the industry than your average Joe walking into CompUSA regardless of whether they would by an AMD server or not for their company.

 

[Noriaki]

Ahh ok fair enough. I misunderstood what you meant.

 

[Texmaster]

<< texmaster - the 1.4GHz athlon was released today, as well as the duron 950.

>>



Oh! My sincere apologies. Been on vacation

 

[GL]

Sander Sassen, Intel Hired Assasin

Looks like the site's own readers have handed this guy a bone to chew on.

 

[Texmaster]

<< Sander Sassen, Intel Hired Assasin

Looks like the site's own readers have handed this guy a bone to chew on. >>




Thats the glory of forums. Everyone can comment.