Phenom II X4 940 Tested at Stock Speeds

Originally Posted by farlex85

Perhaps I'm reading these wrong but they seem to be performing very similarly to each other (penryn and PII) in this bench. Almost identically in fact (unless that closeness is normal.

there is not a PII bench in this thread with this bench,

the AMD processor you up there is 9850 Black edition

[farlex85]

Originally Posted by wolf2009

there is not a PII bench in this thread with this bench,

the AMD processor you up there is 9850 Black edition

Indeed. Interesting.......

[Woody112]

Originally Posted by farlex85

Perhaps I'm reading these wrong but they seem to be performing very similarly to each other (penryn and PII) in this bench. Almost identically in fact (unless that closeness is normal.

Their going to be close. This bench mark isn't like others where they just blow each other out of the water. A few frames per second is actually a bit more than it seems, if you know what I mean.
But ya the PII is performing exceptionally well. As you can see above I clocked my x9770 to 3.0ghz 7.5x400 and was only a couple frames above in the second run and was actually beat in a few runs on the first test.

Originally Posted by Woody112

Their going to be close. This bench mark isn't like others where they just blow each other out of the water. A few frames per second is actually a bit more than it seems, if you know what I mean.
But ya the PII is performing exceptionally well. As you can see above I clocked my x9770 to 3.0ghz 7.5x400 and was only a couple frames above in the second run and was actually beat in a few runs on the first test.

That is not a PII , thats a 9850 Black edition

[Woody112]

Originally Posted by farlex85

Indeed. Interesting.......

Didn't notice that, crap Now I really want to see that PII in action Go AMD

Originally Posted by Woody112

Didn't notice that, crap Now I really want to see that PII in action Go AMD

yes Go AMD .

x264 acc to me is the most accurate CPU Benchmark and it is not biased or optimized for either processor.

It contains optimizations for both Intel and AMD, if I am not wrong.

The developers were talking about adding Corei7 optimizations into the upcoming builds of x264. I'm sure PII will get some too for its own strengths.

Another thing , x264 doesn't use SSE4 (acc to developers it is crap), so no advantage for Intel there.

[Wile E]

Originally Posted by wolf2009

yes Go AMD .

x264 acc to me is the most accurate CPU Benchmark and it is not biased or optimized for either processor.

It contains optimizations for both Intel and AMD, if I am not wrong.

The developers were talking about adding Corei7 optimizations into the upcoming builds of x264. I'm sure PII will get some too for its own strengths.

Another thing , x264 doesn't use SSE4 (acc to developers it is crap), so no advantage for Intel there.

No, it's not the most accurate bench. No bench is. Every architecture is going to have it's strong points and weak points in different bench types.

To prove my point, when A64 and P4 were the 2 cpus out, the P4 outran the A64s in encoding. Yet we all know that a P4 is not faster than an A64.

[farlex85]

Originally Posted by wolf2009

yes Go AMD .

x264 acc to me is the most accurate CPU Benchmark and it is not biased or optimized for either processor.

It contains optimizations for both Intel and AMD, if I am not wrong.

The developers were talking about adding Corei7 optimizations into the upcoming builds of x264. I'm sure PII will get some too for its own strengths.

Another thing , x264 doesn't use SSE4 (acc to developers it is crap), so no advantage for Intel there.

Well see, the thing is if a program uses SSE4 or something like that, it gets more relevant to me, b/c in "real-world" app your intel proc will be using those instructions. A very good bench isn't one that doesn't use something one camp has and the other one doesn't, it's one that maximizes the strengths of each so that each get a full bench. If it doesn't use SSE4 then I would argue it's skewed.

Also, people like to toss around optimization like they have seen the source code and know what they are talking about (not directed at you), but I don't buy it most of the time. It just depends. Like Darren said earlier a good collection of many benches is needed for the best idea, b/c no one can give a complete picture.

[Woody112]

Originally Posted by wolf2009

yes Go AMD .

x264 acc to me is the most accurate CPU Benchmark and it is not biased or optimized for either processor.

It contains optimizations for both Intel and AMD, if I am not wrong.

The developers were talking about adding Corei7 optimizations into the upcoming builds of x264. I'm sure PII will get some too for its own strengths.

Another thing , x264 doesn't use SSE4 (acc to developers it is crap), so no advantage for Intel there.

Very true, wish it was used more by product reviewers.
Still trying to pick my jaw up off the ground after seeing that it was a 9850 on that bench run. WOW.

alright, I stand corrected

[Wile E]

Originally Posted by wolf2009

alright, I stand corrected

It's still a good benchmark tho. Especially for those who deem encoding important. Everything is relevant to some degree. Encoding is important to me, but my Intel clocks higher, negating the per clock advantage of Phenom.

And that's where it all hinges for me. I need to know 2 things about cpus I am comparing to buy. First, how do they compare clock for clock, and second, how high do they clock?

Now, even if CPU1 is 15% faster per clock than CPU2, but CPU2 clocks 30% higher, guess which one I'm gonna buy. lol.

[Darren]

Originally Posted by Woody112

Still trying to pick my jaw up off the ground after seeing that it was a 9850 on that bench run. WOW.

Indeed.

The Pheonom 9850 and Phenom 9950 are fantastic processors, ashame that Intel fan boys didn't realise that when they contributed to damaging AMDs reputation.

[Woody112]

Originally Posted by Wile E

No, it's not the most accurate bench. No bench is. Every architecture is going to have it's strong points and weak points in different bench types.

To prove my point, when A64 and P4 were the 2 cpus out, the P4 outran the A64s in encoding. Yet we all know that a P4 is not faster than an A64.

Not trying to criticize or anything but why is it that when either AMD or Intel releases benchmarks for upcoming CPU's, their blowing the competitors out of the water. Then we all have to wait for an unbiased review for accurate results, usually done here at TPU.
Can't help but wonder if its like Nvidia and ATI. One game optimized better for one card than the other you know.
I've always felt that encoding was pretty much an unbiased benchmark tool. But as you stated above and have a valid point about the A64 and the P4.

[Woody112]

Originally Posted by Darren

Indeed.

The Pheonom 9850 and Phenom 9950 are fantastic processors, ashame that Intel fan boys didn't realise that when they contributed to damaging AMDs reputation.

Isn't that the truth.
I've owned my fair share of AMD chips but when I made the jump to and E6600 I haven't looked back till now.
I have new found respect for AMD.

[3dsage]

Why doesnt anyone bench a P2 on WPRIME??

[Melvis]

Originally Posted by Darren

Indeed.

The Pheonom 9850 and Phenom 9950 are fantastic processors, ashame that Intel fan boys didn't realise that when they contributed to damaging AMDs reputation.

I second that, i have a m8 that is impossible to change his mind on, he is strictly intel, seagate, and Nvidia, as soon as you mention the other brands he just ignores you and says the horrible, and nasty, and would never touch them, and doesn't respect them at all. I haven't seen many on here that do the same which is good, but some that come close. Its people like my m8 who makes AMD look stupid when there not, and poor performers when they clearly arnt.
How long did it take intel fan boys to realize that the P4 was crap, and that AMD ran alot cooler then intel did, a very long time.

[fullinfusion]

Originally Posted by 3dsage

Why doesnt anyone bench a P2 on WPRIME??

we do!!!
http://forums.techpowerup.com/showthread.php?t=77369

[farlex85]

Originally Posted by Melvis

I second that, i have a m8 that is impossible to change his mind on, he is strictly intel, seagate, and Nvidia, as soon as you mention the other brands he just ignores you and says the horrible, and nasty, and would never touch them, and doesn't respect them at all. I haven't seen many on here that do the same which is good, but some that come close. Its people like my m8 who makes AMD look stupid when there not, and poor performers when they clearly arnt.
How long did it take intel fan boys to realize that the P4 was crap, and that AMD ran alot cooler then intel did, a very long time.

I think the word fanboy is tossed around too much here, people seem to jump on the defensive or offensive when someone (like myself ) criticizes something or questions the brand. I am easily swayed by objective results, as are most, but I am skeptical when I see numbers like those on the first page, just as some are elated when they see other numbers I would consider meaningless. It's a perspective thing. Your m8 there is irrational, and that sort of thinking could be called fanboyish, but I see that term extended to people all the time and it sort of ruins discussions imo. We can objectively debate and discuss the results (as I think we have done fairly well here) w/o putting on a red or blue shirt.

[fullinfusion]

Originally Posted by Wile E

It's still a good benchmark tho. Especially for those who deem encoding important. Everything is relevant to some degree. Encoding is important to me, but my Intel clocks higher, negating the per clock advantage of Phenom.

And that's where it all hinges for me. I need to know 2 things about cpus I am comparing to buy. First, how do they compare clock for clock, and second, how high do they clock?

Now, even if CPU1 is 15% faster per clock than CPU2, but CPU2 clocks 30% higher, guess which one I'm gonna buy. lol.

Yeah but where's the fun in clocking such a easy proc Eg: (intel)

Wouldn't you get a much better feeling of accomplishment having a challenge going for a high AMD clock? (just playn)

[farlex85]

Originally Posted by fullinfusion

Yeah but where's the fun in clocking such a easy proc Eg: (intel)

Wouldn't you get a much better feeling of accomplishment having a challenge going for a high AMD clock? (just playn)

Depends on how fruitful the efforts are.

[Darren]

Originally Posted by fullinfusion

we do!!!
http://forums.techpowerup.com/showthread.php?t=77369

I'm I seeing that right,

E8200 @ 6229 MHz scores 12.152 sec

and

Phenom 9850 @ 3318 MHz scores 11.545 sec


In WPrime

[ShadowFold]

Originally Posted by andy_usa

Reality check... that's not going to help with those numbers. AMD is going down. Intel's just pwning them year after year. I think they should just focus on there Graphic chip business and drop the CPU competition all together. It's not like they have a chance of fighting back. They're just really so behind and there's no point catching up.

Troll alert, do not feed.

Originally Posted by Darren

I'm I seeing that right,

E8200 @ 6229 MHz scores 12.152 sec

and

Phenom 9850 @ 3318 MHz scores 11.545 sec


In WPrime

phenom has 2 extra cores

[Darren]

Originally Posted by wolf2009

phenom has 2 extra cores

So I would be right to presume that WPrime supports multiple cores, why dont review sites use WPrime more often? opposed to Super Pi

Originally Posted by andy_usa

Reality check... that's not going to help with those numbers. AMD is going down. Intel's just pwning them year after year. I think they should just focus on there Graphic chip business and drop the CPU competition all together. It's not like they have a chance of fighting back. They're just really so behind and there's no point catching up.

You must be completely spasticated.

If AMD gave up on CPUs what is going to stop Intel rising the prices on their existing CPUs which are already stupidly expensive. Think boy, think.

Secondly, these reviews are previews, knowone can judge AMD until they are on the shelves and tested with DDR3 memory on a AM3 motherboard otherwise the tests are bias in comparison to Intels DDR3 on the i7.

Year after year?

I can remember when AMDs slowest Durons, and when AMDs slowest Semprons were taking out Intels fastest and most expensive chip.

[farlex85]

Originally Posted by Darren

So I would be right to presume that WPrime supports multiple cores, why dont review sites use WPrime more often? opposed to Super Pi

Good question. Things like that and Sandra seem to be used very seldom, instead reviewers seem to prefer to use games, superpi, and 3dmarks and the like for benches. I can understand a fair amount of the community invests into gaming, but these really aren't very good tests of cpu's and how they are used today.