AMD Phenom II 920 & 940 Full Review [UPDATED with more tests]

[Shintai]

4% more performance with AM3

That would be in the best situation. And even 4% is still matched against slower and cooler running Core 2 Quads.

Now the bad thing is...this is what they gonna compete with till 2011. AMD is quickly moving to be the new VIA.

[v_rr]

That would be in the best situation. And even 4% is still matched against slower and cooler running Core 2 Quads.

Now the bad thing is...this is what they gonna compete with till 2011. AMD is quickly moving to be the new VIA.

Predictions again?
2011 is 3 years away, so you are saying that Phenom II 940/945 is going to be the top AMD dog for 3 years

This is how topics get closed. Its with absurd things like that, that the MODs close treads....

[Shintai]

Predictions again?
2011 is 3 years away, so you are saying that Phenom II 940/945 is going to be the top AMD dog for 3 years

This is how topics get closed. Its with absurd things like that, that the MODs close treads....

There might be a slightly faster one. To battle what? 3Ghz Core 2 Quad?

2 (Not 3...) years is alot of time for Deneb/Shanghai to be AMDs top dog. Specially when it aint exactly a speedracer today.

Why close it for this? Because you dont want to hear it? This 945 and maybe 955/965 whatever will also compete against 45nm i5. 32nm i7, 32nm i5 and possible even for a short while 32nm nextgen architecture.

The AMD slide is already wrong in terms of performance. Funny thats how PR is. And thats the most absurd thing that is posted. It reminds me of K10 being 40-50% faster clock for clock than Core 2 Quad.

[herderien]

It is because hyper threading is being used with the core i7 which hampers the performance of that particular game.

Here is a link that shows a performance boost when hyper threading is turned off.

http://techreport.com/articles.x/15818/6

thx for the explanation

the weird thing is that in the tech report article it shows lower performance with ht on then off but in the present review it's when going to high res only where it dramatically falls...i really can't see what's going on

[Glow9]

Did they do a test with all of them at the same speed? Or am I missing it?

[v_rr]

Why close it for this? Because you dont want to hear it?

Witch part of the predictions you didn´t understood? Comments like that in a review of PH2 920/940 are pure flaming post.
Here we are evaluating PH2 performance in aplications, not if it will be the only cpu until 2020.

Open a preditions tread for your self really. And don´t forget the forum best Quote "2008 is going to be a year of Nvidia complete dominance"

The AMD slide is already wrong in terms of performance

Is it? Has far as I know the preview in this tread don´t have PH 1 to compare it to PH2, so it´s impossible to measure if it´s true or not. So here we o again to predition, flaming and a closed tread instead of PH2 performance.

And you have been warned 100 times of that behavior in many treads, and even with a closed tread you keep the same attitude.

AMD is quickly moving to be the new VIA.

This is a pure flamming post, and have nothing to do with PH2. Grow up

[STEvil]

Actually I have to agree with v_rr on this one, Shintai.

Toss the baggage or leave the thread.

edit - that scaling for i7 link only gives dual channel to the skulltrail system (quad channel is optimal). darn.

[amdcian]

phenom 2 consumes only 50w @ 2.6ghz

http://www.hardware-infos.com/news.php?news=2620

massive consumption increase with +0.1v and 1ghz, but we've seen that already (and 120w isnt that much for 3,65ghz)

So great Deneb outperforms Agena double about performance/watt

[Movieman]

Grabs hat, frowns, damn wool itches:
I always loved this line from "Cool Hand Luke"
"What we have here is a failure to communicate"
How do we as a forum stop the pro Intel guys from coming into the AMD threads and vice versa?
Ban all these guys?
I mean all of them?
Now lets add into this mix the crazed fanbois on the ATI/nVidia part of the equation.
Ban them also?
So when we're down to 12 members of the whole forum and someone says the wrong thing we ban them too..
I can see it now: 2 weeks and 3 people left..
I got a better idea:
We take the people that come into the AMD section and bash it and give them a reason to stop. Maybe changing their title to 'I love AMD" for 3 months might do the trick or showing a chopped pick of them hugging Hector Ruis..
Maybe change their sig to " I will not be a fanbois" printed 100 times..
Something has to change to stop this BS for all of us in all sections.
I'm going to kick some of these ideas around with the staff and since logic isn't stopping this maybe shaming someone will make the point.

[Zucker2k]

Dave, while what you're saying is definitely something the powers that be are free to consider, the level of intolerance in this section of the forums is alarming, to say the least. The FACT of the matter is that, PII is reviewed against other processors, and even if it were not, why can't any enthusiast point out comparisons if there are any? Ultimately, the success or not so much of PII is going to be judged against a benchmark, be it PI or other quadcores from the Intel camp. As far as I'm concerned, anybody, and I mean anybody who doesn't have anything nice to say about phenom is automatically labeled in this section of the forums. It's sad indeed. I mean can't we have a discussion based on facts? Take what Shintai said for example, the allusion to Via was not made out of thin air, but rather AMD's competitiveness with reference to its own roadmap. Again, let me say it: the allusion was made based on fact. There may be a little exxageration in there, but the trend lately points to a fast move towards gagging other members with rights to post in this section. I mean they have gone so far as having threads moved from the news section to the AMD section in order to gag anyone with a different opinion, opinion based on test results. It's a sad day in XS!

[knightwolf654]

id just like to say i see criticism as a good thing, id rather see it then non at all, but not at the level shintai takes it to in every amd thread on this forum.

[BenchZowner]

Here's a preview of what's coming soon



All processors clocked at 3.7GHz now.

[Movieman]

Dave, while what you're saying is definitely something the powers that be are free to consider, the level of intolerance in this section of the forums is alarming, to say the least. The FACT of the matter is that, PII is reviewed against other processors, and even if it were not, why can't any enthusiast point out comparisons if there are any? Ultimately, the success or not so much of PII is going to be judged against a benchmark, be it PI or other quadcores from the Intel camp. As far as I'm concerned, anybody, and I mean anybody who doesn't have anything nice to say about phenom is automatically labeled in this section of the forums. It's sad indeed. I mean can't we have a discussion based on facts? Take what Shintai said for example, the allusion to Via was not made out of thin air, but rather AMD's competitiveness with reference to its own roadmap. Again, let me say it: the allusion was made based on fact. There may be a little exxageration in there, but the trend lately points to a fast move towards gagging other members with rights to post in this section. I mean they have gone so far as having threads moved from the news section to the AMD section in order to gag anyone with a different opinion, opinion based on test results. It's a sad day in XS!

To me personally to go into an(X) thread knowing the temperment here, and start to take apart that companies products is essentially trolling because you know the outcome before you ever start.
No one is trying to gag anyone but there has to be a line or a way that the info is expressed so that people do not read it as trolling.
You know I do DC work correct?
I do WCG. Can you see what would happen if I posted in the Folding forum that WCG is better and even proved my point( Just making an example here my folding friends!
They would crucify me and rightly so.
I do go to their threads but only to congratulate people or very rarely to moderate but I tread lightly and go there as a friend.
There has to be some tact involved in all of these sections and the other guys likes and passions understood before you post.
Just courtesy really to make it simple or maybe just good manners.
one should also lose this feeling that " I am always right and will post 40000 times to try and prove the other guy wrong"
Doesn't work.. Just a waste of energy.
I read many threads here. I see some really crazy BS posted as truth but I stay out of it most of the time in the name of peace on the forum.
The reality is that most times in 20-30 mins that BS is buried so far in the forum that it will rarely be read again..

[Frodin]

id just like to say i see criticism as a good thing, id rather see it then non at all, but not at the level shintai takes it to in every amd thread on this forum.

That is because he mixes his one-sided criticism with pure flame baiting. Someone mentions that AM3 will have a 4% performance increase because of DDR3, and he turns it into "AMD is quickly moving to be the new VIA." It is completely off topic, and flame baiting at its finest.

[Hornet331]

Here's a preview of what's coming soon

nice, hope the also include kents and yorkies.

[BenchZowner]

Check the post again Hornet331

[Hornet331]

ah ok n1.

i was 3 min to early

[Brother Esau]

Grabs hat, frowns, damn wool itches:
I always loved this line from "Cool Hand Luke"
"What we have here is a failure to communicate"
How do we as a forum stop the pro Intel guys from coming into the AMD threads and vice versa?
Ban all these guys?
I mean all of them?
Now lets add into this mix the crazed fanbois on the ATI/nVidia part of the equation.
Ban them also?
So when we're down to 12 members of the whole forum and someone says the wrong thing we ban them too..
I can see it now: 2 weeks and 3 people left..
I got a better idea:
We take the people that come into the AMD section and bash it and give them a reason to stop. Maybe changing their title to 'I love AMD" for 3 months might do the trick or showing a chopped pick of them hugging Hector Ruis..
Maybe change their sig to " I will not be a fanbois" printed 100 times..
Something has to change to stop this BS for all of us in all sections.
I'm going to kick some of these ideas around with the staff and since logic isn't stopping this maybe shaming someone will make the point.

"Puts on Special AMD 3D Glasses" and.....Patiently Awaits for Ban Stick to come out and you to Smack Shintai with it////......................C'mon Man Just once for the Home Team


P.S...................Damn Wool Underwear is Driving Me Nuts!
Would it Kill you to Make Me Admin For The Day.....So I Can Take Out The Trash??????

[BenchZowner]

Keep things civilized guys, unless you want to get this thread locked up again and miss new info & benchies

[azraeel101]

When the clock for clock @3.7 tests are done could they be posted in another thread so we can discuss in peace? This thread is already derailed...

[amdcian]

Seems Kensfield has become the legend

[Gener_AL (UK)]

Dave, while what you're saying is definitely something the powers that be are free to consider, the level of intolerance in this section of the forums is alarming, to say the least. The FACT of the matter is that, PII is reviewed against other processors, and even if it were not, why can't any enthusiast point out comparisons if there are any? Ultimately, the success or not so much of PII is going to be judged against a benchmark, be it PI or other quadcores from the Intel camp. As far as I'm concerned, anybody, and I mean anybody who doesn't have anything nice to say about phenom is automatically labeled in this section of the forums. It's sad indeed. I mean can't we have a discussion based on facts? Take what Shintai said for example, the allusion to Via was not made out of thin air, but rather AMD's competitiveness with reference to its own roadmap. Again, let me say it: the allusion was made based on fact. There may be a little exxageration in there, but the trend lately points to a fast move towards gagging other members with rights to post in this section. I mean they have gone so far as having threads moved from the news section to the AMD section in order to gag anyone with a different opinion, opinion based on test results. It's a sad day in XS!

Seems for all your wisdom you stil havent realised that you are not discussing performance of the Phenom II. Intolerance you say .... Ok move along. See the link saying "log out" ? Thats intolerance.

See your signiture ? Totlerating that is one thing.
See your text ? Tolerating that is another.

What to say about Phenom I or II.

Its Cheap upgrade for AMD owners.
Its is competatvie.
Its platform performance is quite decent.
But we know this already.







Look fowared to the Deneb Vs Agena Reviews BenchZowner
Any power tests (Watts consumed) Undervolting test would be interesting also. I Plan on doing some myself. My 9850 happily runs at 1.25v stock speed and 1.05v @ 1.2ghz
Finding infomation regarding deneb & Agena comparing power consupmtion Overclocked and underclocked is hard to find.

One thing i hope deneb will change is the current attitude of larger hardware review websites, to focus more on various platforms rather then review all the latest and greatest on Intel Hardware. I cant really remember the last time game reviews were done on copetiting platforms with idetical GPUS for the majoirty of reviews.

[leoy]

Seems kentsfield isn't jealous of a q9550....
Nice scores for the phenom.

(off topic)Why do i have the impression that q6600 clocks higher on average than q9xxx??

[mAJORD]

There might be a slightly faster one. To battle what? 3Ghz Core 2 Quad?

2 (Not 3...) years is alot of time for Deneb/Shanghai to be AMDs top dog. Specially when it aint exactly a speedracer today.

Why close it for this? Because you dont want to hear it? This 945 and maybe 955/965 whatever will also compete against 45nm i5. 32nm i7, 32nm i5 and possible even for a short while 32nm nextgen architecture.

The AMD slide is already wrong in terms of performance. Funny thats how PR is. And thats the most absurd thing that is posted. It reminds me of K10 being 40-50% faster clock for clock than Core 2 Quad.

While I honestly agree AMD will be a pretty ordinary situation on a performance front by late 2010 in the void between intels 32nm trans and AMDs ), but that is a relativly short period of time. I find it amusing that despite Intels domination, you find it nessersery to exagerate your arguments..

You seriously think a 965 will be the top dog? The model numbering should be a hint... 3.4Ghz 985, or maybe even a 3.5Ghz 995 for all we know is a well truely a reasonable prediction for the quads to top out at on 45nm next stepping, which should at least leave them in a position to compete with mid-range i5's, something that's much more important to the company than i7.. (though maybe not to those preying for AMD domination).

Must i remind you again..

Brisbane G1 dec 06 2.6ghz
" G2 q1 08 3.1Ghz

windsor F2 mid 06 2.8ghz
F3 mid 07 3.2ghz

they can ramp the frequencies up pretty quickly if under pressure

add to that the possibilty (given AMD's vague and volitile roadmaps) of a 6core 'istanbul-on-the-desktop' to give some MT clout to the lineup if the market demands it, and your predictions of total misery look a whole lot less dramatic.

There's many things wrong with that vague amd slide you're basing your predictions on, one of which being the indication of a new m-arch on a new process.. so i wouldn't take it as gospel, good or bad.,

DO yourself a favour and put your crystal ball on performance predictions away.. stick to product release times or something. Looking at Intels 2009-2011 roadmap, then applying it to AMDs current lineup and saying they're doomed as if it's a fact is ridiculous..

[OverClocker_gr]

Look fowared to the Deneb Vs Agena Reviews BenchZowner
Any power tests (Watts consumed) Undervolting test would be interesting also. .

Just got a watt meter so yes,we will include a power consumption but without agena since we send it back to his owner