Jump to content

AMD - Geekbench scores

AMD benchmark Geekbench OSX Mountain Lion Lion Snow Leopard Hackintosh Legacy Kernel

  • Please log in to reply
86 replies to this topic

#21
chokoboster

chokoboster

    InsanelyMac Protégé

  • Members
  • Pip
  • 7 posts

Do you have two overclocked hacks, chokoboster?

By the way, we use the same wallpaper. In fact, i use it in all my hacks and even recently i started to use it with Windows too. Great desktop background.

Gilles, i trust Cinebench way more than i trust Geekbench. But Geekbench became quite the standard for end user performance benchmarks, so that's the reason i choose it over any other tool for our topic.


yes, I have two overclock hacks, Intel and AMD
the AMD hacks is my old system, cheap system with good performance

#22
gils83

gils83

    InsanelyMac Legend

  • Members
  • PipPipPipPipPipPipPipPipPip
  • 1,467 posts
  • Gender:Male
  • Location:France
  • Interests:le soleil du var et l'informatique
:)

these tests are indicative, not more credible than i7 3770/16000 points 3.5 ghz phenom x4 960T unlocker that x6 to 15000 points.
When we go to the reality of work figures are not all the same, I could make a difference by using the two platforms.
the only thing that is true is that the price / performance ratio is largely in favor to AMD.

Posted Image

#23
theconnactic

theconnactic

    Stubborn AMD user

  • Local Moderators
  • 2,796 posts
  • Gender:Male
Well, long life overclock!

But we must consider the possibility Geekbench is taking advantage of the extra cores of your Phenom IIx6, thus the higher results. However, few real-life applications are already optimized for more than two cores, and the vast majority still run in a single core. So, while tests like Geekbench could show the performance potential of a given CPU, tests that are more task-specific (like rendering benchmarks etc) tend to give results closer to the real user experience.

Anyway, i can tell i agree with you: AMD is unbeatable as long value is concerned, in the short to middle term and not considering scenarios like server arrays and big enterprise usage (where the higher power consumption of most AMD CPUs nullify this advantage).

Best regards!

#24
gils83

gils83

    InsanelyMac Legend

  • Members
  • PipPipPipPipPipPipPipPipPip
  • 1,467 posts
  • Gender:Male
  • Location:France
  • Interests:le soleil du var et l'informatique
:)

yes, this is true for power consumption, it remains on when the configuration is accompanied by a graphic card 500 w! 25w the differences are negligible.
In the workplace, it is quite another, though, when we see wasted energy to cool the building, the rest seems quite ridiculous.

#25
theconnactic

theconnactic

    Stubborn AMD user

  • Local Moderators
  • 2,796 posts
  • Gender:Male
In the workplace, no savings are ever ridiculous ;) But for the home user, i agree with you. In fact, you agreed with me, lol.

Best regards!

#26
gils83

gils83

    InsanelyMac Legend

  • Members
  • PipPipPipPipPipPipPipPipPip
  • 1,467 posts
  • Gender:Male
  • Location:France
  • Interests:le soleil du var et l'informatique
:)

yes and no, the company where I work has 7MW power of 1 MW heating and cooling power with 22 boxes for cooling an area of ​​28 000 m2, we get 30,000 people per year (spa)

powers are not comparable! Intel or AMD, irrelevant ;)

#27
eep357

eep357

    Triple Platinum

  • Supervisors
  • 2,527 posts
  • Gender:Male
  • Location:Dark Side of The Wall
  • Interests:things and stuff

hello , :)


I think you're right, the measured values ​​are wrong in my opinion, the processor remains icy after a battery of tests which is not normal in my opinion, the motherboard In addition to a max TDP of 95 W and a frequency of 4.9 ghtz, I must greatly exceed the 125 w.
I think it should be a synthetic test and physical processor is 3.3 ghz, I'll do a test to see if 5.5 ghz Clover is not a chimera of all these results.


I confirm distorts the Clover, the figures true or false??

FX 6100 is 4.02 and the test cinebench 7500 Geekbench test

As you may have started to suspect after going up to impossible speeds like 5.5ghz, the results are false. Clover has option to set processor speed or turbo mode in it's settings for those rare occasions that the true speed is not being auto detected properly. If you set the speed in clover much higher than the bios and leave system running a while, you'll probably also see the clock in the menubar is a few minutes ahead of what time it should be. Every benchmark uses time in one way or another since PCs are basically big clocks, frequency components run at, clock cycles, everything except storage capacity/file size, revolves around time. Most look at how long it's different tasks take to complete, and generates a score based on this; the faster it gets done, the higher the score, or the more things it can get done in a given amount of time, the higher the score. Even for GPU benches that score FPS, the more frames in a second, the higher the score. So by increasing the clock speed that OSX thinks everything is running at it also thinks everything is happening faster than it really is. So while GPU may only be doing 30 FPS, it thinks it did 60 FPS, or while it really took 2 min to run all the geekbench tests, it thinks they were done in only 1 min. When you do this, truth is your performance actually gets worse since many operations are designed to do things in a given amount of time. Set bios to stock speed and bump clover way up and you'll see desktop animations moving noticebly slower or the chase seen in Cinebench look slow motion, OSX is actually slowing them down because it thinks they are all moving faster than they really are. I too found this out the hard way, changed speed back to default in bios while doing some other hardware testing and seeing all my benchmark scores skyrocket. I was very happy! But had a hunch it was too good to be true :( I wasn't going to post this fearing some people may purposely do this just to cheat benchmarks, but once I saw your comment I could tell you were already suspicious so it would have ended up here sooner or later. No such thing as a free overclock :)

#28
gils83

gils83

    InsanelyMac Legend

  • Members
  • PipPipPipPipPipPipPipPipPip
  • 1,467 posts
  • Gender:Male
  • Location:France
  • Interests:le soleil du var et l'informatique

As you may have started to suspect after going up to impossible speeds like 5.5ghz, the results are false. Clover has option to set processor speed or turbo mode in it's settings for those rare occasions that the true speed is not being auto detected properly. If you set the speed in clover much higher than the bios and leave system running a while, you'll probably also see the clock in the menubar is a few minutes ahead of what time it should be. Every benchmark uses time in one way or another since PCs are basically big clocks, frequency components run at, clock cycles, everything except storage capacity/file size, revolves around time. Most look at how long it's different tasks take to complete, and generates a score based on this; the faster it gets done, the higher the score, or the more things it can get done in a given amount of time, the higher the score. Even for GPU benches that score FPS, the more frames in a second, the higher the score. So by increasing the clock speed that OSX thinks everything is running at it also thinks everything is happening faster than it really is. So while GPU may only be doing 30 FPS, it thinks it did 60 FPS, or while it really took 2 min to run all the geekbench tests, it thinks they were done in only 1 min. When you do this, truth is your performance actually gets worse since many operations are designed to do things in a given amount of time. Set bios to stock speed and bump clover way up and you'll see desktop animations moving noticebly slower or the chase seen in Cinebench look slow motion, OSX is actually slowing them down because it thinks they are all moving faster than they really are. I too found this out the hard way, changed speed back to default in bios while doing some other hardware testing and seeing all my benchmark scores skyrocket. I was very happy! But had a hunch it was too good to be true :( I wasn't going to post this fearing some people may purposely do this just to cheat benchmarks, but once I saw your comment I could tell you were already suspicious so it would have ended up here sooner or later. No such thing as a free overclock :)



hello , :)


exact, Clover plays only on the system clock, which is logical to be, it is not able to change the bios settings without validation parameter changes it.
it's frustrating.

That said the FX 6100 is capable of running at 5 ghz, I've tried. :D

#29
Kikobeats

Kikobeats

    InsanelyMac Protégé

  • Members
  • Pip
  • 34 posts
  • Gender:Male
  • Location:Spain
  • Interests:Programming & Hacks
Posted Image
http://browser.prima....com/user/54544

In Win I have +10000 :(

#30
theconnactic

theconnactic

    Stubborn AMD user

  • Local Moderators
  • 2,796 posts
  • Gender:Male
Do you boot UEFI with Windows?

#31
theconnactic

theconnactic

    Stubborn AMD user

  • Local Moderators
  • 2,796 posts
  • Gender:Male
Lion 10.7.5, Bronzovka kernel, 64-bit mode:

http://browser.prima...kbench2/1730679

Lion 10.7.5, Bronzovka kernel, 64-bit mode:

http://browser.prima...kbench2/1730833

Edited by theconnactic, 08 March 2013 - 04:55 AM.


#32
Kikobeats

Kikobeats

    InsanelyMac Protégé

  • Members
  • Pip
  • 34 posts
  • Gender:Male
  • Location:Spain
  • Interests:Programming & Hacks

Do you boot UEFI with Windows?


What? I don't understand. this is my Win7 score Can you think of any way to improve my score on mac?

:)

#33
theconnactic

theconnactic

    Stubborn AMD user

  • Local Moderators
  • 2,796 posts
  • Gender:Male
No, Kiko: it already should match Win7.

By the way, my score with 10.8.3 is the higher i've ever got with my AMD machine: http://browser.prima...kbench2/1755850

Attached File  Screen Shot 2013-03-15 at 12.15.20 AM.png   293.95KB   15 downloads

Best regards!

#34
burkett375

burkett375

    InsanelyMac Protégé

  • Members
  • PipPip
  • 61 posts
Posted Image

Posted Image

#35
theconnactic

theconnactic

    Stubborn AMD user

  • Local Moderators
  • 2,796 posts
  • Gender:Male
Phenom II x6 1090T?

#36
burkett375

burkett375

    InsanelyMac Protégé

  • Members
  • PipPip
  • 61 posts

Phenom II x6 1090T?


960T unlocked to x6 and overclocked

#37
gils83

gils83

    InsanelyMac Legend

  • Members
  • PipPipPipPipPipPipPipPipPip
  • 1,467 posts
  • Gender:Male
  • Location:France
  • Interests:le soleil du var et l'informatique
:D

simulator Clover ;)


Posted Image


;)

#38
Bitcore

Bitcore

    AMD Fanboy

  • Members
  • PipPipPip
  • 146 posts
  • Gender:Male
  • Location:The Netherlands
  • Interests:Producing Music
Got this right now http://browser.prima...kbench2/1791898

#39
gils83

gils83

    InsanelyMac Legend

  • Members
  • PipPipPipPipPipPipPipPipPip
  • 1,467 posts
  • Gender:Male
  • Location:France
  • Interests:le soleil du var et l'informatique
:)

hello Bitcore

this is normal, the clock of your OS X is not good and influt on any system where the extreme speed of opening and closing page, pass your CPU to 3.1 in your config.plist ;)

#40
Bitcore

Bitcore

    AMD Fanboy

  • Members
  • PipPipPip
  • 146 posts
  • Gender:Male
  • Location:The Netherlands
  • Interests:Producing Music
Yeah i saw your post:)
Ill try it Tomorrow





Also tagged with one or more of these keywords: AMD, benchmark, Geekbench, OSX, Mountain Lion, Lion, Snow Leopard, Hackintosh, Legacy Kernel


0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

© 2014 InsanelyMac  |   News  |   Forum  |   Downloads  |   OSx86 Wiki  |   Mac Netbook  |   PHP hosting by CatN  |   Designed by Ed Gain  |   Logo by irfan  |   Privacy Policy