Intel FUD comes to tgdaily

[mforce2]

This is getting boring people . Intel has nothing to show fo itself but promises .

http://www.tgdaily.com/2006/03/07/woodcrest_server_leads_opteron64_in_benchmark/

I mean come on , a story about Intel's own benchmarks of a product they will sell God knows how many months from now . This is just lame .
I though you guys were professionals :roll: .
I realize how desperate Intel is right now but this really is too much . Now maybe if they actually thought of designing a good CPU in the first place instead of the P 4 which was pretty much a failure to begin with ....
Hope to hear some more Intel FUD from their IDF but please people do us a favor and process the damn facts , journalism isn't a copy paste operation you know :lol: .

 

[hergieburbur]

This is getting boring people . Intel has nothing to show fo itself but promises .

http://www.tgdaily.com/2006/03/07/woodcrest_server_leads_opteron64_in_benchmark/

I mean come on , a story about Intel's own benchmarks of a product they will sell God knows how many months from now . This is just lame .
I though you guys were professionals :roll: .
I realize how desperate Intel is right now but this really is too much . Now maybe if they actually thought of designing a good CPU in the first place instead of the P 4 which was pretty much a failure to begin with ....
Hope to hear some more Intel FUD from their IDF but please people do us a favor and process the damn facts , journalism isn't a copy paste operation you know :lol: .

Stop and think, aside from what wusy said, why wouldn't it beat the current Opteron 64s? It was designed from the ground up to counter the Opterons. Intel has made some design mistakes in the past, but they aren't incompetant. They are in this to make money.

Opterons are soon to be replaced by the newer 1207-pin Socket F chips anyway.

 

[mforce2]

I agree with that but I would at least add a comment to it saying that it's not really a real world test .
What I don't agree with is with Intel selling promises .
We don't have anything right now but look what we'll have a few months from now beats what AMD has now in OUR OWN test . Pretty pathetic .
AMD used to do that too though but this should really stop .

 

[hergieburbur]

It definately should stop, but that is what marketing is. Intel has been doing it a lot recently (last 3 years or so), and AMD to a lesser extent. on the other hand, the product would be a total failure if it can't beat whats out there now.

 

[luminaris]

Well, the thing of it is, you have two tech sites performing their own tests. Yes, on Intel benchmarks. However, if you look, you'll see that both of them admit the FEAR tests are not Intel benchmarks.

I for one am not trying to say your totally wrong but, we do have to consider that we are not there at the IDF but yet, Toms and Anandtech were.

 

[ltcommander_data]

please people do us a favor and process the damn facts

What I really don't like is when people pass quick judgement. The fact of the matter is that THG is processing the "damn facts". If you had read the article fully you would have noticed the line:

We're compiling the final figures for you, and will have the raw numbers available soon

If they are sifting through the data themselves, then it is hardly a "copy paste operation". I find it hilarious that the possibility of Intel having any type of lead whether in performance, power, price, etc. seems to aggravate people so much.

 

[mforce2]

OK maybe my tone was a bit harsh but the fact still remains that they published the thing without doing any thorough testing . They should have just waited to have some real facts not some and I quote "graphs shown at IDF" that Intel shows them.
Of course Intel is going to say their new CPUs are the best and they rule but we really shouldn't take their word for it .
What about the title : " Intel Woodcrest server leads AMD Opteron 64 server in performance test " ? If someone hasn't got the time to read the article he'll take that for a fact.
Intel says their future Woodcrest leads over the current AMD Opteron 64 ( in performance tests ) now that would be a good title .
Anyway I don't think the whole thing was biased , it was just negligence IMO . I see it at Anandtech all the time too .
Anyway I think nobody does it better than Apple's Stevie Jobbs . PowerPCs were the best CPUs ever while Apple used them but now they're just a bunch of rubbish and Intel CPUs rule . I'd hate to see Apple switch to AMD . Anybody want to guess what Stevie would say about Intel chips then .... :lol:

I find it hilarious that the possibility of Intel having any type of lead whether in performance, power, price, etc. seems to aggravate people so much.

Just goes to show you how popular Intel is with the techies . Maybe it's because of all that Intel FUD but one thing's for sure , Intel's imgage is not a good one amongst most of the people who know their hardware .
Intel sometimes gives me the impression that they would be willing to sell toilet paper as the best CPU ever if people would fall for it . :roll:

 

[9-inch]

I know exactly what you're talking about.

Toms Hardware has an amusing preference for Intel since the conception of this website.

Just as a few examples: whenever AMD has the lead usually they'll say "AMD's processors are good, but not for long". When Intel leads they say things like "Intel has the best processor on the world and there's no way AMD can touch it!". It's just a matter of reading all their articles, you come with the aforementioned conclusion.

Anyhow, I have some little things to say about Intel's benchmark of Conroe:

1) The review of Conroe vs FX-60 didn't convince me at all. The motherboard's BIOS didn't recognize the processor the way it should and maybe this might hindered the FX peroformance.

2) Let's not forget that the benchmarks were made by Intel.

3) Why didn't they used the Nforce4 platform for the AMD system?

4) The other thing that puzzles me is the posibility of Intel benchmarking their Conroe EE which is supposed to be a quad core cpu. There's NO WHERE in the article that says that the processor tested by Intel was a dual core processor.

5) I'll pay more attention to this topic once the final reviews of both CPUs (Conroe and AM2) see the light of day, untill now everything is just blatant speculation.

 

[ltcommander_data]

Steve Jobs doesn't need to use AMD chips to insult Intel, he's doing just fine while he's using them. I wonder how Intel ever agreed to let Apple advertise that they are finally releasing Intel chips from "dull PCs" and giving them their proper home in Macs.

I'm sure you've seen the benchmarks Anandtech have released with Conroe. They didn't find anything biased about Intel's test systems, and the fact that they are comparing a 2.66GHz Conroe to an overclocked 2.6GHz FX-60 and the Conroe outperforms by such a large margin is amazing. Especially considering the best DDR400 ram was used on the AMD system with 2-2-2/1T timings while the Intel system used conservative DDR2 667 ram at 4-4-4 timings. It would be interesting to see how DDR2 800 with 4-4-4 timings performs or some of that new OCZ PC2 8000 Platinum XTC Extreme Edition stuff at 1GHz with 4-5-4-15 timings. It'd be interesting to see if they can actually get the FSB working at 1066Mhz in a 1:1 ratio with the ram.

http://www.anandtech.com/tradeshows/showdoc.aspx?i=2713&p=1

Conroe isn't released yet, but AM2 is going to have quite a time dealing with Conroe.

I agree that article titles should be more precise, but there's always the consideration that flashy titles are needed to catch readers attention.

 

[mforce2]

Actually I read the first page of the Anandtech review and it was enough for me :
"Honestly it doesn't make sense for Intel to rig anything here since we'll be able to test it ourselves in a handful of months "
Give me a break sure Intel probably din't rig anything too much but maybe jsut tiny little bit , you know a 10-15 % would do just fine . Maybe by the time Conroes launches it will actually have this performance .
This is not an objective test in any way . Intel should have just haded the board and the CPU if they wanted real tests instead of : come see what we got here our unreleased CPU beating AMD's current .
No , I'm sure they din't rig the tests in any way , why would they :roll: . I know I wouldn't rig anything if I had an unreleased product and I was desperate about declining sales .

 

[ltcommander_data]

In regards to your observations:

1) I'll have to concede that the BIOS not recognizing the processor is a bit fishy.

2) The benchmarks were preloaded by Intel, but they were run by Anandtech themselves. I suppose they could have been reconfigured or bugged beyond the ability to readily notice, but then we are dealing with what if scenarios.

3) In this case they were completely justified. You obviously want to remove as many bottlenecks as possible so they were using the fastest graphics rig available right now which is the ATI X1900XT Crossfire. An nForce4 motherboard obviously wouldn't support this. As well, the i975X has support for Crossfire as well making it fair. ATI's current Intel Crossfire boards don't support Conroe anyways (due to the VRMS) so there was no way to use them.

4) I think the quad core theory is very unlikely. At the very beginning of the article they show a picture of the Pentium D, Conroe, and Kentsfield and clearly define in the caption that Conroe is dual core and Kentsfield is quad core. Throughout the rest of the article they mention that it was a 2.67GHz Conroe that was tested. Conroe is a dual core processor otherwise it would have been called Kentsfield. In any case, they ran Quake 4 with SMP disabled and Unreal Tournament 2004 and Half Life 2 aren't multithreaded anyways so a quad core would be no benefit there over a dual core.

5) Of course it's just speculation just as AM2's performance is just speculation. The only benchmarks I've seen of AM2 are from THG have been decried as biased and faked and deliberate anti-AMDism so I've hardly seen any glowing proof on either side. At least Intel's numbers (fake or not) show a performance gain, which makes sense considering the architecture changes that were involved.

 

[luminaris]

F.E.A.R. gets its own page for a couple of reasons:

1) It's the only gaming benchmark that we're using that doesn't use an Intel provided demo. This is the same demo we use in our tests.

That right there says it all man. Those tests were not rigged. Intel is not stupid by any means. They know they've got their backs to the wall. The fact is, Intel came out swinging with horsepower and they are simply showing us that they have the resources, to manufacture a better chip than anyone else.

 

[ltcommander_data]

maybe jsut tiny little bit , you know a 10-15 % would do just fine .

Even if they only rigger 10-15% their lead would seem be there. Just going over the benchmarks the Conroe leads are 28%, 15%, 31%, 20%, 41%, 30%, 11%, and 12%.

our unreleased CPU beating AMD's current

I actually felt it was a nice touch that the AMD CPU was overclocked to 2.8GHz to give Conroe competition from not a current CPU, but something closer to what AMD will be releasing in the future. The fact that AMD was given the best ram while Conroe's was more conservative also made it more fair to AMD.

 

[Dresden]

Hey, it is what it is. Every preview in any market place has the sole intention for creating a bias towards itself. And every preview that will ever be generated by something will showcase highlights.

Compare how many bad movie trailers you have seen compared to bad ones. I actually don't think I have ever seen a bad action movie trailer. I'm just making an analogy. Conroe SEEMS like a good chip. Only time will tell.

 

[9-inch]

I believe those results with a grain of salt. They didn't benchmarked the thing in a 64-bit enviroment where the Athlon 64 shines.

Still, the BIOS issue is still giving me a pain in my b@lls.

DDR 400 with 2-2-2/1T is equal (or a little less performing) than DDR2 667 ram at 4-4-4 timings.

 

[mforce2]

I actually felt it was a nice touch that the AMD CPU was overclocked to 2.8GHz to give Conroe competition from not a current CPU, but something closer to what AMD will be releasing in the future. The fact that AMD was given the best ram while Conroe's was more conservative also made it more fair to AMD.

Yes but don't forget though that there are still other areas where something might have let's say gone wrong .
Giving 1 or 2 sites some real hardware to make their own test would have been nicer and I wouldn't have objected a thing.
I'm not saying Conroe doesn't have a lead . It probably will have one when it's out too . I'm just saying it's not as much as Intel wants us to believe .
A little competition will be a very good thing and I'm looking forward to it but I don't agree with Intel's methods of promoting their stuff . They should launch the DAMN CPU and we'll see how good it is .

 

[luminaris]

Well, we have to take it as it is. If you can't trust Toms and Anandtech to make a good assumption for us then, there's really no use in having them around.

I do agree that they should pre-release at least one or two chips to the sites to perform their own in house testing or at least, Intel should bring them to the sites and let them conduct their own tests.

This is only the beginning and is merely a show and tell show for Intel. Its nothing more than that. We can't be there so we have to rely on Toms and others to be there for us.

 

[mforce2]

Well, we have to take it as it is. If you can't trust Toms and Anandtech to make a good assumption for us then, there's really no use in having them around.

I do trust Tom and Anand , I just don't trust the tests Intel showed them . Maybe they're 100% OK but still it's not fair for somebody to benchmark their own stuff and say it's the best
. I don't care whether Intel or AMD does this .
Last time when AMD wanted to prove they were better they invited Intel for a showdown between their server CPUs if I remember corectly . Intel didn't show up .
I think Intel should now return the favor . Get the Conroe availalbe for people to buy and the politely invite AMD for a public showdown . I wonder if AMD will show up this time . It would only be fair though .

 

[ltcommander_data]

DDR 400 with 2-2-2/1T is equal (or a little less performing) than DDR2 667 ram at 4-4-4 timings.

Precisely. Intel could have just as easily used DDR2 800 at 4-4-4 timings or below to give Conroe an additional performance boost. The new OCZ PC2 8000 Platinum XTC Extreme Edition runs at 1GHz with 4-5-4-15 timings and could probably do 1066MHz with the timings relaxed to 5-5-5. Either would have skewed the result in Intel's favour, but they restrained themselves.

I'm interested in seeing how 64-bit is implemented on this new architecture as well. Since they are starting from scratch having to completely add a non-existant feature to Yonah's framework, they would hopefully be able to do it properly this time.

 

[apache_lives]

This is getting boring people . Intel has nothing to show fo itself but promises .

http://www.tgdaily.com/2006/03/07/woodcrest_server_leads_opteron64_in_benchmark/

I mean come on , a story about Intel's own benchmarks of a product they will sell God knows how many months from now . This is just lame .
I though you guys were professionals :roll: .
I realize how desperate Intel is right now but this really is too much . Now maybe if they actually thought of designing a good CPU in the first place instead of the P 4 which was pretty much a failure to begin with ....
Hope to hear some more Intel FUD from their IDF but please people do us a favor and process the damn facts , journalism isn't a copy paste operation you know :lol: .

Read the article - they mention ICM (as in Integrated Controller Memory???)

woodcrest will kick, we're not talkin bout a netbust processor, this baby is designed by the israle team from P6 tech and they aint FSB hungry, expect results.

 

[ltcommander_data]

Read the article - they mention ICM (as in Integrated Controller Memory???)

Which article? Woodcrest won't have an OMC and it won't for a while. Even I have to admit that the target timeframe for that is sometime late 2007-2008.

 

[AmdMELTDOWN]

Amd Mongrels are somewhat psychotic and are prone to idiotic outbursts.

the next 6 to 8 mos, is going to be painful.

 

[SlitheryDee]

Actually I read the first page of the Anandtech review and it was enough for me :
"Honestly it doesn't make sense for Intel to rig anything here since we'll be able to test it ourselves in a handful of months "
Give me a break sure Intel probably din't rig anything too much but maybe jsut tiny little bit , you know a 10-15 % would do just fine . Maybe by the time Conroes launches it will actually have this performance .
This is not an objective test in any way . Intel should have just haded the board and the CPU if they wanted real tests instead of : come see what we got here our unreleased CPU beating AMD's current .
No , I'm sure they din't rig the tests in any way , why would they :roll: . I know I wouldn't rig anything if I had an unreleased product and I was desperate about declining sales .

Bah. Conroe will outperform AMD's parts by a wide margin, it's designed to do so and it will. I knew that before I saw any benchmarks from IDF. Even if Intel tweaked the results a bit knowing that in 4 months Conroe would be performing at that level the point is moot.

The rule of thumb is that you don't listen to hype from either company and you wait for benchmarks by (hopefully) impartial reviewers. Believe it or not the hype and the advertisement and the wild and often unsupported claims are necessary. It's part of what got computers into almost every home in the first place.

If your competitor is claiming loudly to the masses of thier superior product then you'd do the same to before the masses start believing them. Or in AMD's case, to dissuade a largely convinced consumer market.

 

[levicki]

1) The review of Conroe vs FX-60 didn't convince me at all. The motherboard's BIOS didn't recognize the processor the way it should and maybe this might hindered the FX peroformance.

Yes, missing the brand string printout seriously impacts performance.

2) Let's not forget that the benchmarks were made by Intel.

Yes, they wrote Quake 4 from scratch, right? And HL2 and UT2004 and...

3) Why didn't they used the Nforce4 platform for the AMD system?

Because it doesn't support Crossfire?

4) The other thing that puzzles me is the posibility of Intel benchmarking their Conroe EE which is supposed to be a quad core cpu. There's NO WHERE in the article that says that the processor tested by Intel was a dual core processor.

They said that it was dual core but you haven't noticed. For example Quake 4 scales up 30% on Conroe when SMP is on and it scales 45% on AMD. That is pretty bad scaling for Intel especially because Quake 4 has their SMP patch which suposedly should work better with their CPUs. If the Conroe was quad core, it would have to scale much mure than 30%, don't you think so?

 

[jokersgrin]

Given the fact that Intel has been a "Bully" and their other less than trueful "Benchmarketing" against AMD . I would myself be cautious, given their history to "Fluff" their product. And this fact that they did the test!

Now I'm not saying that Intel has not put out a better chip. But like most people I would like to see independent unbiased testing done and not just by one but by a few. This will validate Intel the crown!

But this fanboyism is gettin me sick people. Weather you like it or not this is a good thing. This competitiveness between Intel and AMD shows progress and better things to come. We the consumer are what they are fighting for ( $$$ ).

I wish Intel and AMD did a show down with their latest and greatest. But I see it like this. Intel has lost alot of ground to AMD! If I was and major investor ( owns a great deal of Intel stock ) into this company, and I start to lose my $$$ because Intel couldn't produce a product that out did everything right away I would expect them to play dirty too, but not to the extent that a Anti-trust suit was filed.

In any sence I hope that Intel does make better this time around, but I hope Amd also comes out with better so on and so forth for both sides.
Because I know that they are fighting to get my $$$$ :mrgreen: And I get the best at that time for my $$$$.