Jump to content
Sign in to follow this  
cleadge

AMD Phenom II AM3

Recommended Posts

These new AMD chippies are good! Mind you the Phenom II is what the Phenon I should have been.

Share this post


Link to post
Share on other sites

These new AMD chippies are good! Mind you the Phenom II is what the Phenon I should have been.

 

my thoughts exactly, tho im mighty impress what the BE x3 can do :-)

Share this post


Link to post
Share on other sites

It seems there were some discrepancies which were out of place in the Graphics benchmarks most noticeable in FarCry 2 while using a GTX280.

 

This is FarCry 2 retested in a variety of ways to locate why the i7 lost out to all the Phenom's in FarCry 2. the results are quite interesting.

http://www.tomshardware.com/reviews/geforc...x-280,2156.html

I was going to say something about why a x3 phenom at 2.4ghz was beating an I7 but I figured it must just be toms inaccurate methods of testing.

Share this post


Link to post
Share on other sites

It is kinda disappointing how the GTX280 scales backwards when ran on the i7.

It will be nice to see new benchmarks when the Nvidia issue is resolved.

Share this post


Link to post
Share on other sites

It is kinda disappointing how the GTX280 scales backwards when ran on the i7.

It will be nice to see new benchmarks when the Nvidia issue is resolved.

just go to anandtech for reliable benchmarks when they get AM3 chips.

Share this post


Link to post
Share on other sites

I was going to say something about why a x3 phenom at 2.4ghz was beating an I7 but I figured it must just be toms inaccurate methods of testing.

Hi.

 

Not quite. There is a very valid and logical reason a 3 core CPU can outperform a 4 core, HT'ed CPU, such as a Core-i7.

 

Cache latency and inter CPU communication is an entirely different beast when 3 cores are concerned. It changes the performance metrics of how CPU's work entirely. There are some lovely whitepapers around the traps that talk about just why three cores can be more efficient than 4, at times...

 

z

Share this post


Link to post
Share on other sites

I was going to say something about why a x3 phenom at 2.4ghz was beating an I7 but I figured it must just be toms inaccurate methods of testing.

Hi.

 

Not quite. There is a very valid and logical reason a 3 core CPU can outperform a 4 core, HT'ed CPU, such as a Core-i7.

 

Cache latency and inter CPU communication is an entirely different beast when 3 cores are concerned. It changes the performance metrics of how CPU's work entirely. There are some lovely whitepapers around the traps that talk about just why three cores can be more efficient than 4, at times...

 

z

 

I was more referring to general raw processing benchmarks that the i7 in every other review I have seen slaughters the Phenom CPU's Edited by nesquick

Share this post


Link to post
Share on other sites

I was going to say something about why a x3 phenom at 2.4ghz was beating an I7 but I figured it must just be toms inaccurate methods of testing.

Hi.

 

Not quite. There is a very valid and logical reason a 3 core CPU can outperform a 4 core, HT'ed CPU, such as a Core-i7.

 

Cache latency and inter CPU communication is an entirely different beast when 3 cores are concerned. It changes the performance metrics of how CPU's work entirely. There are some lovely whitepapers around the traps that talk about just why three cores can be more efficient than 4, at times...

 

z

 

I was more referring to general raw processing benchmarks that the i7 in every other review I have seen slaughters the Phenom CPU's

 

 

Avail us!

 

http://www.tomshardware.com/reviews/geforc...x-280,2156.html

Share this post


Link to post
Share on other sites

I was going to say something about why a x3 phenom at 2.4ghz was beating an I7 but I figured it must just be toms inaccurate methods of testing.

Hi.

 

Not quite. There is a very valid and logical reason a 3 core CPU can outperform a 4 core, HT'ed CPU, such as a Core-i7.

 

Cache latency and inter CPU communication is an entirely different beast when 3 cores are concerned. It changes the performance metrics of how CPU's work entirely. There are some lovely whitepapers around the traps that talk about just why three cores can be more efficient than 4, at times...

 

z

 

I was more referring to general raw processing benchmarks that the i7 in every other review I have seen slaughters the Phenom CPU's

 

 

Avail us!

 

http://www.tomshardware.com/reviews/geforc...x-280,2156.html

 

this is what their benchmarks should look like http://www.guru3d.com/article/amd-phenom-i...0-review-test/1

Share this post


Link to post
Share on other sites

On the toms article the i7 smashes the Phenom II in Synthetic, Productivity (3dS Max, AVG, WinRar), and the A/V encoding benchmarks.

Its only in the games benchmarks where the Phenom II is on par or slightly ahead of the i7 which is the same results guru3d achieved.

Share this post


Link to post
Share on other sites

On the toms article the i7 smashes the Phenom II in Synthetic, Productivity (3dS Max, AVG, WinRar), and the A/V encoding benchmarks.

Its only in the games benchmarks where the Phenom II is on par or slightly ahead of the i7 which is the same results guru3d achieved.

Its not the phenom2's I am complaining about its the old phenom in particular the triple core 2.4ghz one which somehow manages to pull better scores than the i7 920 in nearly every game and futuremark benchmark they did someone please tell me how the hell they cocked that up because thats like comparing a core 2 e6600 to an athlon 4200+ and saying the athlon did better than the core2 at everything except synthetic benchmarks (with the exception of memory ones) which would indicate that something went wrong because core2 clock for clock is better than the k8's.

 

take for example the following (3dmark vantage does not lie and will use very similar settings no matter how you run it)

 

http://www.tomshardware.com/reviews/socket...nom,2148-6.html

Toms scores using the performance preset which is a universal setting that wont change:

i7

overall-10369

GPU-8298

CPU-41249

phenom x3 8750

overall-12374

GPU-10346

CPU-30025

Posted Image

Posted Image

Posted Image

see anything wrong with those scores? here are what they should look like sourced from Hardware Canucks which is always spot on with their benchmarks (with the exception of no phenom2 which would be near the top)

http://www.hardwarecanucks.com/forum/hardw...-review-15.html

Posted Image

 

All I am saying is as per usual Toms is a load of BS.

Edited by nesquick

Share this post


Link to post
Share on other sites

i didnt actually notice the phenom score in there (probably because i was more interested in Phenom II vs i7), but yes you are right.. there is something wrong with that

Share this post


Link to post
Share on other sites

On the toms article the i7 smashes the Phenom II in Synthetic, Productivity (3dS Max, AVG, WinRar), and the A/V encoding benchmarks.

Its only in the games benchmarks where the Phenom II is on par or slightly ahead of the i7 which is the same results guru3d achieved.

Its not the phenom2's I am complaining about its the old phenom in particular the triple core 2.4ghz one which somehow manages to pull better scores than the i7 920 in nearly every game and futuremark benchmark they did someone please tell me how the hell they cocked that up because thats like comparing a core 2 e6600 to an athlon 4200+ and saying the athlon did better than the core2 at everything except synthetic benchmarks (with the exception of memory ones) which would indicate that something went wrong because core2 clock for clock is better than the k8's.

 

take for example the following (3dmark vantage does not lie and will use very similar settings no matter how you run it)

 

http://www.tomshardware.com/reviews/socket...nom,2148-6.html

Toms scores using the performance preset which is a universal setting that wont change:

i7

overall-10369

GPU-8298

CPU-41249

phenom x3 8750

overall-12374

GPU-10346

CPU-30025

Posted Image

Posted Image

Posted Image

see anything wrong with those scores? here are what they should look like sourced from Hardware Canucks which is always spot on with their benchmarks (with the exception of no phenom2 which would be near the top)

http://www.hardwarecanucks.com/forum/hardw...-review-15.html

Posted Image

 

All I am saying is as per usual Toms is a load of BS.

 

nesquick, I'm very quick to point out a few things here.

 

Here's another review that I've pulled online and I think it's completely unbiased in any way of form...

 

http://benchmarkreviews.com/index.php?opti...3&Itemid=63

 

You're rather quick at jumping the gun here. This review also confirms what Tom's writes about the tri-core Phenom IIs beating the snot out of the i920. But take note on what he says on the subject....

If gaming is the argument for Intel processors, someone needs to go back and re-examine the performance. While it's not entirely unbiased to use a Radeon HD 4870 video card for testing, I felt that ultimately it was more fair than using an NVIDIA product when there are so many games 'tuned' for their hardware over ATI's. Regardless, in our Devil May Cry 4 tests it was apparent that the Core i7-920 doesn't maintain the architectural efficiency to beat out any of the AMD Phenom II processors which cost much less.

And you look in the other gaming benchmark (Far Cry 2), you'll barely notice ANYTHING at the resolutions we're playing at nowadays. So what if you have a $600 CPU which is considered the fastest in the world? If it plays what you want, then be happy with that. What's wrong with something that doesn't match with what you're expecting? I'm more inclined to think that some sites are paid to shut up by the big companies.

Share this post


Link to post
Share on other sites

Large quote plus text

Would you like to have an in house test sora? I have a HD4850 that I can test with my i7 920 at stock or whatever speed you want vs your phenom2 and HD4850, it will put the argument to rest once and for all.

 

edit: that website draws conclusions from 2 game tests.... thats like testing a drug that cures cancer on 2 people and saying yay it works on everyone.

Edited by nesquick

Share this post


Link to post
Share on other sites

Large quote plus text

Would you like to have an in house test sora? I have a HD4850 that I can test with my i7 920 at stock or whatever speed you want vs your phenom2 and HD4850, it will put the argument to rest once and for all.

 

edit: that website draws conclusions from 2 game tests.... thats like testing a drug that cures cancer on 2 people and saying yay it works on everyone.

 

Final EDIT: nesquick, now that I've sat down and thought about it, there is something a bit off. You're not the kind to get that fired up. Seriously, I respect all the points you've brought up and stuff. But having to see both sides of the story isn't a bad thing considering. If you're that fired up about it, then sure. Send all your gear here to my place and we can do a toe-to-toe comparison. I'm not worried as I'm happy with what I've got and I'm not bragging about it. But there is still that something that doesn't feel right about all this. Yes, there will be issues and yes, there will be inconsistencies as we all do.

 

BUT the main point of this edit is to say this. If you're happy about your CPU, fine. BE F**KING HAPPY and just don't get fired up about these sort of things. If you don't, you become just another fanboy. Now, I'm not saying you are one but still. Benchmarks aren't the know-it-all-end-it-all so meh.

Edited by sora3

Share this post


Link to post
Share on other sites

Large quote plus text

Would you like to have an in house test sora? I have a HD4850 that I can test with my i7 920 at stock or whatever speed you want vs your phenom2 and HD4850, it will put the argument to rest once and for all.

 

edit: that website draws conclusions from 2 game tests.... thats like testing a drug that cures cancer on 2 people and saying yay it works on everyone.

 

Final EDIT: nesquick, now that I've sat down and thought about it, there is something a bit off. You're not the kind to get that fired up. Seriously, I respect all the points you've brought up and stuff. But having to see both sides of the story isn't a bad thing considering. If you're that fired up about it, then sure. Send all your gear here to my place and we can do a toe-to-toe comparison. I'm not worried as I'm happy with what I've got and I'm not bragging about it. But there is still that something that doesn't feel right about all this. Yes, there will be issues and yes, there will be inconsistencies as we all do.

 

BUT the main point of this edit is to say this. If you're happy about your CPU, fine. BE F**KING HAPPY and just don't get fired up about these sort of things. If you don't, you become just another fanboy. Now, I'm not saying you are one but still. Benchmarks aren't the know-it-all-end-it-all so meh.

 

Im not fired up, I am just trying to prove a point that an i7 920 should not be getting beaten by something that 1. is lower clocked and 2. less efficient clock for clock.

Don't get me wrong Ill admit here and now an i7 920 will get beaten by a Phenom2 940 BE in games due to its higher clock speed but when its getting beaten by a decent margin by a chip designed for office computers then the alarm bells start ringing.

 

oh and by in house I meant in house atomic forums lol I would never send my gear to someone I have never met.

Edited by nesquick

Share this post


Link to post
Share on other sites

If you want to send it to Atomic forum HQ for a test, I'm happy to do that :D

 

As long as Justin/TheFrunj doesn't break my Phenom II XD

Share this post


Link to post
Share on other sites

If you want to send it to Atomic forum HQ for a test, I'm happy to do that :D

 

As long as Justin/TheFrunj doesn't break my Phenom II XD

lol frunj with my cpu :| I don't like the thought of that :P

 

What I was suggesting is you run test X then I run exactly the same test at stock clocks and we compare the results.

Share this post


Link to post
Share on other sites

Ah. I see. Sorry for the misunderstanding. >.<

 

Anyway, what tests do you want to cover? In order to do this well, I'll lower the clocks to your speed and see if the architecture is what's to blame and you need to tell me what games/apps you'll be using.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×