AMD Phenom II X4 965 vs. Intel core i5-750

[dantoshi]

Which of these processors is the best one?

 

[andy5174]

http://www.lostcircuits.com/
Hanging on one no name web site = fail
His testing methods : For comparison purposes, all benchmarks were run in WinXP 32
Mentions: Intel Turbo is cheating= final fail
Compare to Guru3d, Anand, THG , Tweaktown, PC World FGS !

What the i5 750 did to AMD pricing in 2 months.
http://i482.photobucket.com/albums/rr185/notty222/amd965pricedrop.png

Don't agree with jennyh = Countless fail!

 

[randomizer]

 

[sirheck]

http://images.anandtech.com/graphs/intellynnfieldlaunch_090409000254/19908.png

http://images.anandtech.com/graphs/intellynnfieldlaunch_090409000254/19909.png

http://images.anandtech.com/graphs/intellynnfieldlaunch_090409000254/19910.png

http://images.anandtech.com/graphs/intellynnfieldlaunch_090409000254/19911.png

PII-965 lost in 3 out of 4 games.

In the game PII-965 leads i5, it is 1.5 FPS better which is substantial.

In addition, i5 beats 965 by 4FPS, 18.9FPS and 1.9FPS in other 3 games which are insignificant at all.

LOL thats a good one :lol:

 

[jennyh]

I just dont get why you people cannot look at the Lost Circuits benchmarks and see that something is not adding up.

There is nothing wrong with the i5 system, it performs exactly as it should compared to the i7. The testing methods are good also.

The main difference is the LostCircuits benchmarks aren't the usual fail synthetics. I know intel 'wins' the synthetics...intel has always won at synthetics. Now look at the real apps.

I am also well aware that if excel/photoshop were added that the i5 might even turn it around...but why should they be added? Why not add MainConcept or DIEP chess to the anand/THG benches?

Why trust synthetics over real apps? Did you read what the FTC said about intel and sysmark (and others) btw?

http://www.brightsideofnews.com/news/2009/12/17/why-the-ftc-lawsuit-against-intel-has-substance.aspx

You trust intels synthetic wins or real world apps that intel hasn't corrupted? Think about it.

 

[randomizer]

THG always uses MainConcept in CPU reviews.

@andy: I can take away your veteran status faster than you can get it back so if you've got nothing worth contributing then don't post.

 

[jennyh]

http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=3690&cp=3

False advertising. This includes hiding the compiler changes from developers, misrepresenting benchmark results (such as BAPCo Sysmark) that changed due to those compiler changes, and general misrepresentation of benchmarks as being “real world” when they are not.

http://www.brightsideofnews.com/news/2009/12/17/why-the-ftc-lawsuit-against-intel-has-substance.aspx

"Worse, a number of benchmarks exhibit performance boosts if the CPUID vendor name string is changed to "GenuineIntel".

In any case, it is relatively easy to test for CPUID vendor name string sensitivity. PCMark05, SysMark 2007 Preview, and Everest Ultimate are three benchmarks that have been affected by the vendor name string value that CPUID returns, but others exist.

We know intel wins in synthetics. We can see from LostCircuits that AMD does an awful lot better on real world apps.

What will be interesting to see is how quickly - if ever - THG and Anand ditch the synthetics that the FTC have named as being dubious.

This isn't somebody from AMDZone or me - it's the FTC saying that intel have been cheating at benchmarks for years.

 

[ElMoIsEviL]

The xbitlabs article doesn't have an i5 750. The tweaktown article doesn't have a 965 BE. The hardocp article doesn't have a 965 BE.

You mentioned both the 955 and the 965 so I posted links containing either/or.

As for your claim that the Xbit article doesn't have the i5 750:

I'm pretty sure a Core i5 750 = Core i5 750.

 

[andy5174]

THG always uses MainConcept in CPU reviews.

@andy: I can take away your veteran status faster than you can get it back so if you've got nothing worth contributing then don't post.

Am I the only one who is debating here?! What's wrong with you? all the Implied face plam thing?!

 

[ElMoIsEviL]

And please, everybody knows THG and Anand are intel shills.

Everybody? Are you talking about common sense? (read my sig).

What evidence do you have which proves that THG and Anandtech are Intel Shills? Please... elaborate further 😛


You really ought to not be using words whose meanings are totally lost on you (reasoning and rationality).

 

[randomizer]

Am I the only one who is debating here?! What's wrong with you? all the Implied face plam?!

You're the only one who seems to care about posting nonsense in order to boost your points.

 

[jennyh]

I've just told you what to expect elmo.

If Anand and THG don't ditch the synthetics that intel have been accused of manipulating, then we will know for sure won't we?

And for gods sake how many times are you gonna post a World in Conflict benchmark as 'evidence'. We know how much that game favours intel.

 

[andy5174]

You're the only one who seems to care about posting nonsense in order to boost your points.

Do you think that I really debate because I want the points?

What do I get for becoming a Veteran? Nothing! Nothing different from regular.

I debated as I can't tolerate ignorant fools. That's all.

 

[jennyh]

You seem to have a pretty warped view of 'debate' andy. Perhaps you should read up a bit more on that. Get some debating skills, then come back.

 

[andy5174]

@randomizer:

And stop the face palm thing.

Yeah, of course you can ban me on you own will and that happens all the time and not going to be the last time.

I am not going to tolerate the crap pics from you anymore and I can simply get a new account if you ban me for this.

 

[ElMoIsEviL]

I've just told you what to expect elmo.

If Anand and THG don't ditch the synthetics that intel have been accused of manipulating, then we will know for sure won't we?

And for gods sake how many times are you gonna post a World in Conflict benchmark as 'evidence'. We know how much that game favours intel.

Do you read?

I posted that image to show you that he xbit lab article did feature a Core i5 750. I did not post it for it's benchmarking data.

Your entire world view relies on conspiracies. It's like you think the entire IT realm is comprised of the benevolent AMD against the world. And that somehow the world (and every Corporation and Individual in it) are blind to this conspiracy. You also think that we're all paid Intel shills.

And then you have the nerve to argue with Albert Einstein by using an empty argument such as "Everybody knows blah blah blah". This is going to sound mean JennyH but you're arguing with people who are far more intelligent than you are. The worst part is that you actually think you're making sense. I guess this is like the Katie Couric interview with Sarah Palin. I'm Katie Couric (except I'm male of course) and you're Sarah Palin. Like minded people think you nailed that interview (AMD Fanbois/Right-Wing Rednecks) but anyone with a brain knows you sounded like a... (insert ad hominem of your choice).

You're good for a laugh or two though. At least you keep us all entertained. Not sure THG would be the same without you.

 

[jennyh]

Sigh elmo you are such a child. Do you actually believe you are smarter than me? I mean actually? If so you have got *severe* issues that require medical attention.

It's a conspiracy against intel isn't it? I mean the FTC, AMDZone, the EU etc..it's all a conspiracy against intel.

All those notes from people with nothing to gain...

"Intel failed to disclose material information about the effects of its redesigned compiler on the performance of non-Intel CPUs. Intel expressly or by implication falsely misrepresented that industry benchmarks reflected the performance of its CPUs relative to its competitors’ products.”

Smith said that while on the BAPCo committee he "constantly butted heads with Intel, and some of the correspondence generated during that time demonstrates the extreme and outrageous measures Intel takes to influence that consortium."

It's all a conspiracy against intel and AMDZone has even got the FTC involved. :lol:

Btw yes i can see the xbitlabs article has a 750, the reason I thought it didn't was because it isn't included in the test setup.

http://www.xbitlabs.com/articles/cpu/display/core-i7-870_7.html#sect0

 

[randomizer]

If Anand and THG don't ditch the synthetics that intel have been accused of manipulating, then we will know for sure won't we?

That's pretty poor evidence. But I wouldn't expect any more because there is nothing concrete against either site.

 

[keithlm]

The "beliefs", if you can call them that, on this forum are based on a rational and reasonable interpretation of the evidence. If you have 10 Objective Websites telling you one thing and a single 1 telling you something else... who would you be more prone to believe?

We all know that you consider yourself to be unbiased and a non-fanboy. You keep repeating it like it's a mantra. Since you are part of a majority on this forum you somehow believe that makes you correct in that opinion. You then conclude that you are being rational and reasonable since they all agree with you. But having that majority doesn't automatically make you correct.

NOTE: There are not "10 objective websites versus one link" as you attempt to claim. Not unless you blindly accept summaries of reviews without actually looking at results yourself and doing your own analysis.

===========================================
Let me present an artificial example to illustrate the problem. Below are two examples of what could be summary statements from different reviewers that could be describing the exact same set of benchmark results:

A. "This review had 12 wins and 4 losses for one brand. They take the performance crown and the competition obviously can't compete."

versus

B. "This review had 4 obvious wins, 8 scores close enough to be considered tied and 4 obvious losses. These different chips trade blows and basically perform the same."


The problem is that if a popular reviewer writes the "A" summary statement then many people will blindly accept that as an irrefutable fact. They won't bother looking at the actual data for themselves. A worse problem is that if a few people strongly agree with summary "A" then if they see something like summary "B" they might discount or ignore it. Even though the summaries are the same results worded differently.

==========

TO exacerbate the situation what really needs to be done is to compare the results for the same exact benchmark against multiple review sites. (i.e., You can't just lump the end results of "wins" and "losses" together for a single reviewer and think you have something important if you truly want to be objective.) Sure when doing this type of analysis you must consider possible differences between the reviewers. But that is why you would want a larger number of different reviewers to "weed out" the inconsistencies statistically.

When this is done then often what some people originally considered to be "obvious" becomes much less than obvious.

And for gods sake how many times are you gonna post a World in Conflict benchmark as 'evidence'. We know how much that game favours intel.

In a true objective review the person doing the review would remove any anomalous results unless they could definitively explain exactly why there is an anomaly.

Oh but hey... we're not objective; we're living in a fantasy. Unlike Elmo who is the epitome of neutrality.

 

[jennyh]

That's pretty poor evidence. But I wouldn't expect any more because there is nothing concrete against either site.

It might not be great evidence but in the interests of fairness it would make a lot of sense if all of the synthetics named by the FTC were removed from benchmarks.

Is there any pressing requirement to continue to use the same synthetics? I don't see why they can't be replaced by proper apps.

 

[ElMoIsEviL]

Sigh elmo you are such a child. Do you actually believe you are smarter than me? I mean actually? If so you have got *severe* issues that require medical attention.

It's a conspiracy against intel isn't it? I mean the FTC, AMDZone, the EU etc..it's all a conspiracy against intel.

All those notes from people with nothing to gain...

"Intel failed to disclose material information about the effects of its redesigned compiler on the performance of non-Intel CPUs. Intel expressly or by implication falsely misrepresented that industry benchmarks reflected the performance of its CPUs relative to its competitors’ products.”

Smith said that while on the BAPCo committee he "constantly butted heads with Intel, and some of the correspondence generated during that time demonstrates the extreme and outrageous measures Intel takes to influence that consortium."

It's all a conspiracy against intel and AMDZone has even got the FTC involved. :lol:

I don't believe I am smarter than you but I know I am more knowledgeable than you are.

What's this with it being a conspiracy against Intel? Where did I say anything remotely close to that? Last time I checked I was the one who posted the thread about dismantling Intel into smaller companies and I was the one who posted the thread about the charges against Intel.

As for the FTC, these are charges so I'm more apt to wait and see if they're true. The charges AMD had leveled against Intel (with regards to the rebate program) has concrete evidence in the form of email exchanges within both Intel and Dell. Therefore I was more prone to believing that those charges were true.

As for the Intel compiler, I do know that it will insert SSE/2/3/4 optimizations when they are usable and that the compiler will make use of Micro and Macro Ops Fusion as well. These are features which AMD either does not support or supports in part (some of them AMD fully support). AMD has full access to these technologies but choose not to incorporate them into their hardware (see Intel/AMD Technology Sharing Agreements).

So when it comes to the compiler... I would need to see more evidence (something you don't need to see because you already made up your mind long ago that Intel = Evil and AMD = Good).

You just called me a child, but I am not the one who views things as simply being Good or Bad now am I? I like evidence, I find it is useful to side with the evidence rather than create a world view and only cite evidence which sustains that world view (ignoring any contrary evidence a.k.a calling the sources Intel shills).

 

[jennyh]

In a true objective review the person doing the review would remove any anomalous results unless they could definitively explain exactly why there is an anomaly.

Oh but hey... we're not objective; we're living in a fantasy. Unlike Elmo who is the epitome of neutrality.

That is exactly right - anomalous results should be removed until a good explantion is discovered.

Like for example the Last Remnant gaming benchmarks still used by THG which show the gts 250 beating an HD 5850.

 

[randomizer]

Is there any pressing requirement to continue to use the same synthetics? I don't see why they can't be replaced by proper apps.

They are expected in every review for the same reason why 3DMark is put in every video card review. It's just standard practice. I don't know anyone who actually looks at them seriously though.

 

[keithlm]

That is exactly right - anomalous results should be removed until a good explantion is discovered.

Like for example the Last Remnant gaming benchmarks still used by THG which show the gts 250 beating an HD 5850.

Or if when benchmarking two GPU that have very similar results you find that when going to Crossfire vs SLI the results are completely different.

An objective reviewer will at the very least ask "WHY". The shill will mention the difference but not even question it as being anomalous. (Note that I do not say "paid shill" like some people do. I personally don't believe the majority of these shills are paid. You don't have to be paid to be biased.)

 

[randomizer]

Like for example the Last Remnant gaming benchmarks still used by THG which show the gts 250 beating an HD 5850.

The German reviewers (who seem to use it the most) don't know people don't want it I guess. They never check the US site for feedback.

 

[jennyh]

You know that stuff like that hardly helps THG's reputation random. Just once in a while it would be nice for these 'mistakes' to end up in AMD/ATI's favour but somehow they hardly ever do.