AMD Phenom II Review Thread

[Miss Banana]

I saw your explanation and its still off by a mile or 1.6 Kilometers. Kyle made every excuse in the world for AMD Conroe did the AM2's worse than the X2's did the P4. Kyle did the same thing again when 4 X 4 launched. Maybe like Rammsteiner, Kyle is pissed AMD lost yet another round and he's depressed?

Where were the AMD leaning folks claiming to want fair and unbiased benchmarks for AM2 he gave a , 4 X $ he gave a and Conroe he gave yes, a uh? Again, he's revised, hell, rewrote his whole Conroe conclusion when other web sites like Tech Report that you quoted, called him out.

Kyle's whatever but Phenom 2 still did worse in even the most favorable review. You do know that Intel's score were up and down as well, or were you only looking at AMD's?

This was your rather poor defensive post, in which you make excuses.
Kyle once hated intel too, so it's ok for him to make crapreviews about AMD?
Not really Donnie...

[Donnie27]

As I said, we have had some hints. But the current Opteron setup is not totally lost, and price will determine the faith of AMD's server CPUs. AMD will probably stay competitive but margins will decrease.

Donnie27:

Well some seem to give out that label pretty fast. And yeah, those oil dollars will be pouring in for quite some time. Could probably invest so much that Intel would be in serious trouble but highly unlikely, can't see that investment paying off. Better to invest as much to give AMD a chance to be competitive for performance crown (or to be generating positive cash flow) and then let it "manage" itself. There is clearly enough room for more than one CPU manufacturer and AMD is the only option to Intel that's got a chance currently.

Edit: Yeah, the JAS 39 Gripen is quite neat. Was a shame that or neighbour Norway chose the F22 Raptor instead. Would have made a lot of people in Sweden happy... Now here's a sector with A LOT of foul play... Hehe.

I have a friend at Dell. The only thing they're worried about is AMD's Volume and Pricing, not what most reviews are saying. They just like IBM, HP and other do their own benchmarks and review the info for their Boss. AMD is in the right market to make money IMHO and that's mainstream and upper-budget. Even noobs are starting to ask me about Quad Cores. LOL one lady said "Hey Donnie what's the scoop on those new computers with 4 Hard Drives in one?"

F-20 should have beat the F-16 according to the much respected Chuck Yeager.

[Donnie27]

This was your rather poor defensive post, in which you make excuses.
Kyle once hated intel too, so it's ok for him to make crapreviews about AMD?
Not really Donnie...

I don't think Kyle hates AMD or Intel. He loves their ad dollars.

[marten_larsson]

Actually, it is very easy to see understand your point. Please quote me defending Kyle's review? Then you break into rant about "childish" and etc.... I said I DIDN'T LIKE Kyle's review, need to me QUOTE IT? I said "



How in the ^$%& is that defending Kyle? Please Open your eyes and mind, then look and read again!? It is amazing that I'm here talking about Kyle's crappy review. It takes nothing from the overall spanking Phenom 2 got. Even Intel Fanboys know that a better showing by AMD meant they don't have to wait 6 to 9 months for i5 to ship. Many dreaded seeing this and yes there are few loving these review results. Your comment about AMD getting Intel to cut prices is one the most absurd things I've read all day.

My point; I'll read the Phenom 2 reviews and take all of the Pro and Con alike with a big grain of salt, not blindly rely on a few that please me

Seems like you guys are discussing different things without taking notice. You're saying that PII isn't what it needs to be. Miss Banana is saying that the review sucks. Banana isn't claiming that the results are total bogus but the value of this review is close to none since it's not a fair comparison. The same results on equal hardware and there would be no dispute. Since there are plenty of reviews demonstrating a more fair method, why even bother with the HardOCP? This is not fanboyism but a scientific approach. Although the fanboys are more likely to cry "outrage".

Edit:

Ofcourse Dell and the others don't care about reviews. They take the numbers, add them up, check the price. If it has performance on par with price it's a deal. Otherwise it's not. They want to stay competitive too, therefore fooling their costumers too much won't do them any good. Mainstream and upper budget was were RV770 was aimed. It turned out to be a successful move. To gain halo they brought R700. Unfortunately AMD won't be able to smack two PII in one and call it a day but the first part is covered for now.

[AbelJemka]

Seems like you guys are discussing different things without taking notice. You're saying that PII isn't what it needs to be. Miss Banana is saying that the review sucks. Banana isn't claiming that the results are total bogus but the value of this review is close to none since it's not a fair comparison. The same results on equal hardware and there would be no dispute. Since there are plenty of reviews demonstrating a more fair method, why even bother with the HardOCP? This is not fanboyism but a scientific approach. Although the fanboys are more likely to cry "outrage".

Edit:

Ofcourse Dell and the others don't care about reviews. They take the numbers, add them up, check the price. If it has performance on par with price it's a deal. Otherwise it's not. They want to stay competitive too, therefore fooling their costumers too much won't do them any good. Mainstream and upper budget was were RV770 was aimed. It turned out to be a successful move. To gain halo they brought R700. Unfortunately AMD won't be able to smack two PII in one and call it a day but the first part is covered for now.

Exactly!

[Rammsteiner]

This was your rather poor defensive post, in which you make excuses.
Kyle once hated intel too, so it's ok for him to make crapreviews about AMD?
Not really Donnie...

Kyle sounds like the average 'I go with the best and the rest is retarded' guy. Whatever, shows only the credibility he, his review and his policies lacks, his loss, no one else's.

Also Ive to agree with Miss that it's a rather poor post for naming my name Feeling hurt because PhII lost another round The only thing AMD just lost is a bad '08

[savantu]

Yeah just hold those palms on your ears and sing "nanananana".
FYI ,they already have the tools for 32nm process node.FYI, 6 core Deneb is just 4 core deneb with 2 more cores(duh),not a new design.FYI 45nm was delayed one quarter (Q4 for Shanghai instead of Q3).FYI,6 core deneb on 32nm will be just a shrink with more cache(again not a hard thing to do,AMD done it countless times).

You can live in your intel only world and think AMD will have 4 core deneb next 2 years,or you can use facts and logic and burst the repetitive dream you have.

Really ? 2 years in advance ? Even Intel is only now starting to ramp 32 at its development FAB and they're due to launch Westmere in 11 months.

Aren't you forgetting who owns the FABs now ? What's their incentive into moving so fast?

[savantu]

...Miss Banana is saying that the review sucks. Banana isn't claiming that the results are total bogus but the value of this review is close to none since it's not a fair comparison. ..

You get the cookie award for posting a contradiction as big as Everest.

[marten_larsson]

You get the cookie award for posting a contradiction as big as Everest.

What's your problem? The issue isn't in the results but how they were obtained. The results are most probably accurate but that doesn't mean they're comparable. I could do a review with PII at 3,8GHz, comparing it to a i7 965 at 1,2GHz, look at the results and claim PII a winner. The results would be correct but my conclusion wouldn't because of a flawed method. How is that not obvious? Where's the contradiction? Please, show me a scientific study were practices as Kyle's has been used.

[Zucker2k]

What's your problem? The issue isn't in the results but how they were obtained. The results are most probably accurate but that doesn't mean they're comparable. I could do a review with PII at 3,8GHz, comparing it to a i7 965 at 1,2GHz, look at the results and claim PII a winner. The results would be correct but my conclusion wouldn't because of a flawed method. How is that not obvious? Where's the contradiction? Please, show me a scientific study were practices as Kyle's has been used.

If one were to look at your result by itself, then you're right, but if other results corroborate yours then you're wrong.

[Miss Banana]

If one were to look at your result by itself, then you're right, but if other results corroborate yours then you're wrong.

Exactly, and since kyle used trisli and noone else did, noone can say whether or not his results are the same as other results.
Geeees, why are we still talking about this???


So anyhow, the tomshardware review was not bad I think, it concluded that Phenom II has better power consumption than whatever intel is offering at the moment.
Which is a bit.... positive for AMD of course, but I do wonder where all the differences in power consumption conclusions came from.
http://www.tomshardware.com/reviews/...0,2114-24.html

[Shintai]

I think the real question is who marten_larsson other account is or was

[Miss Banana]

You get the cookie award for posting a contradiction as big as Everest.

And you sir, should really take another look at the word contradiction and contemplate on its meaning.

[Rammsteiner]

If one were to look at your result by itself, then you're right, but if other results corroborate yours then you're wrong.

Now what does it matter!?

The point here is not how to read reviews, the point is that the way H's review was done is highly unfair in combination with comparisons against other platforms. If he used 2GB DDR800 on Agena and Deneb and compared those, that's a whole different story than 2GB DDR800 Deneb vs 4GB DDR1600 C2. Could be me, I just can not understand how others dont seem to understand this.

The basics of any review/test is to make sure all conditions are the same so that the results reflect the reviewed/tested object only. That's stuff you learn at school, it doesnt take Einstein to define that theory. And yes, I even think using DDR1600 for C2 is a bit of a gray area, I even think that is wrong since that is in no way comparable with DDR800.

[marten_larsson]

If one were to look at your result by itself, then you're right, but if other results corroborate yours then you're wrong.

No, it's not right. The method is flawed and even if the results are the same it doesn't justify using different setups. Not if you want to draw conclusions and include it in a "mass" of studies. Small variations are of course allowed (otherwise there wouldn't be a spectrum). Sure, one study that's flawed can be included without that much impact but if 50% of the reviews were done in the same manor it would have had a great impact. (At least since 5% performance difference can be the judgement of "total domination" in some people's minds.)

His method is better if you wanted to study something like "How much i7 has to be crippled to match a PII".

[T_Flight]

I tried to tell you that the reviews don't matter right now. When one doesn't like the results, the best way to try and negate them is to attack the reviewer. Of course that doesn't work with those who already know that trick, but it's the only thing left to do during damage control.

[Miss Banana]

I tried to tell you that the reviews don't matter right now. When one doesn't like the results, the best way to try and negate them is to attack the reviewer. Of course that doesn't work with those who already know that trick, but it's the only thing left to do during damage control.

Did you read it?

[roofsniper]

AMDs roadmap doesnt agree with you. Nor does their 32nm capabilities.

don't trust amd's roadmaps. its pointless that they even create them. let me just say this there is no information out at all right now about bulldozer. don't trust what you have seen online. theres a big difference between amd's roadmaps and the company's roadmaps. the company's roadmaps are a lot fuller and include everything they have planned out for this year.

I tried to tell you that the reviews don't matter right now. When one doesn't like the results, the best way to try and negate them is to attack the reviewer. Of course that doesn't work with those who already know that trick, but it's the only thing left to do during damage control.

i don't think any one is arguing with any reviews except for the one at [H]. that one is horrible. even tho everyone says anandtech is intel biased i really liked their phenom II review.

[Miss Banana]

i don't think any one is arguing with any reviews except for the one at [H]. that one is horrible. even tho everyone says anandtech is intel biased i really liked their phenom II review.

Yeah I liked anandtech's review, allthough he as well used 800 mhz ram for some reason. I really do wonder what the performance difference is, did anyone find a review that looks at ram speed yet?
Or does anyone have a phenom II somewhere and some ram so we can give it a try?

[Zucker2k]

don't trust what you have seen online. theres a big difference between amd's roadmaps and the company's roadmaps. the company's roadmaps are a lot fuller and include everything they have planned out for this year.

So investors and OEMs don't care one way or the other what AMD's publically declared roadmaps are? And what's the difference between AMD road map and "company's" roadmap? I think they're one and the same and so the point is moot. Until AMD comes up with another roadmap that is different from what is already known you and Informal are just grabbing at straws, as usual. I'm surprised he even managed to emerge from that rock of embarassment after all his predictions fell by the way side Incorrigibility thy name is Informal.

On a more serious note: I asked a question that hasn't been answered: since none has tried what Kyle has done, how do you know if it's possible to run that configuration and with DDR2 1066 ram? I mean seriously, it wasn't as if he didn't try; he stated clearly that he run into problems, and since the results wouldn't have been affected significantly, then the conclusion still remains the same. Are you guys arguing that he should have crippled the Intel system in order to make it fair? How does that reflect real world IF the AMD system can't handle that configuration fully with 1066 ram and the Intel could with DDR3 at 1600?

[panfist]

did anyone find a review that looks at ram speed yet?

You did. And you didn't like the results. If you're going to ignore that review you might as well do us all a favor and ignore [H], too.

[marten_larsson]

On a more serious note: I asked a question that hasn't been answered: since none has tried what Kyle has done, how do you know if it's possible to run that configuration and with DDR2 1066 ram? I mean seriously, it wasn't as if he didn't try; he stated clearly that he run into problems, and since the results wouldn't have been affected significantly, then the conclusion still remains the same. Are you guys arguing that he should have crippled the Intel system in order to make it fair? How does that reflect real world IF the AMD system can't handle that configuration fully with 1066 ram and the Intel could with DDR3 at 1600?

Yeah, it's a possibility that the issue is with the AMD hardware not functioning with the DDR2 1066 but it could just as easily be just his particular set of hardware. And yes, he should have crippled the C2Q just to show the difference in performance with the same setup. Why? To prove that it was the CPU and not the RAM that made the difference.

And Shintai; I'm assuming you were just kidding about me being someone else. I'm just me, no one else...

[Miss Banana]

On a more serious note: I asked a question that hasn't been answered: since none has tried what Kyle has done, how do you know if it's possible to run that configuration and with DDR2 1066 ram? I mean seriously, it wasn't as if he didn't try; he stated clearly that he run into problems, and since the results wouldn't have been affected significantly, then the conclusion still remains the same. Are you guys arguing that he should have crippled the Intel system in order to make it fair? How does that reflect real world IF the AMD system can't handle that configuration fully with 1066 ram and the Intel could with DDR3 at 1600?

Enough reviewers managed to get it working, so really what's the deal here?
Maybe the nvidia chipset was being difficult, maybe he was using incompatible ram, maybe he was using wrong timings, how can we possibly know?

[roofsniper]

So investors and OEMs don't care one way or the other what AMD's publically declared roadmaps are? And what's the difference between AMD road map and "company's" roadmap? I think they're one and the same and so the point is moot. Until AMD comes up with another roadmap that is different from what is already known you and Informal are just grabbing at straws, as usual. I'm surprised he even managed to emerge from that rock of embarassment after all his predictions fell by the way side Incorrigibility thy name is Informal.

On a more serious note: I asked a question that hasn't been answered: since none has tried what Kyle has done, how do you know if it's possible to run that configuration and with DDR2 1066 ram? I mean seriously, it wasn't as if he didn't try; he stated clearly that he run into problems, and since the results wouldn't have been affected significantly, then the conclusion still remains the same. Are you guys arguing that he should have crippled the Intel system in order to make it fair? How does that reflect real world IF the AMD system can't handle that configuration fully with 1066 ram and the Intel could with DDR3 at 1600?

zucker i am sorry but seriously every time i read one of your posts i feel as if i am in kindergarten or something. you need to grow up. where do you think i am getting my information from? hmmmm tell me? because you seem to think that i make it all up. last time i checked businesses don't tell the public every single detail of what is going on. they don't say exactly when stuff is coming out and they don't list everything. everyone was arguing over when deneb was coming out for a long time but i have known that they would come out on january 7th since may of last year. you act as if you know everything but you don't seem to have the slightest clue. and im pretty sure the amd system can handle 1066 ram. i have not seen one issue with it before and mine runs it fine. the fact that he says he has problems with running 1066 shows more of a user error than an error of an entire company.

[gallag]

So some people do not even believe AMD about AMD if the news is not overly positive. Unbelievable.