Interesting article I came across.

Hacp

Lifer
Jun 8, 2005
13,923
2
81
In games such as Wolfenstein, Call of Duty: Modern Warfare 2, Tom Clancy’s H.A.W.X, BattleForge and Far Cry 2 the Phenom II X4 processors were actually faster when clocked up near 4GHz! This is quite amazing as out of the 9 games tested, the Phenom II X4 series was faster than the Core i7’s in 5 of them. Although the margins were very limited, the Phenom II X4 was found to be faster, and had it just managed to match the Core i7 series with the Radeon HD 5970, we would have been impressed.
http://www.legionhardware.com/articles_pages/cpu_scaling_with_the_radeon_hd_5970,1.html

So basically at the highest resolutions with the most powerful video cards, the phenom x4 is faster than the I7. I guess the x6 is the CPU to get if you want to game high end :p
 

Nemesis 1

Lifer
Dec 30, 2006
11,366
2
0
http://www.legionhardware.com/articles_pages/cpu_scaling_with_the_radeon_hd_5970,1.html

So basically at the highest resolutions with the most powerful video cards, the phenom x4 is faster than the I7. I guess the x6 is the CPU to get if you want to game high end :p


Thats a BS article .We have many reviews that show it not to be so. I review doesn't make it so. Looks like the GPU is all done to me . We seen this befor when it first released. Also It be more interesting to see what XF and sli bring to the table . Oh thats right its already been benched . If some of those games don't change FPS as we seen that not CPU thats GPU. The Guy that did test is retarded with his conclusion.
 
Last edited:

Hacp

Lifer
Jun 8, 2005
13,923
2
81
Thats a BS article .We have many reviews that show it not to be so. I review doesn't make it so. Looks like the GPU is all done to me . We seen this befor when it first released. Also It be more interesting to see what XF and sli bring to the table . Oh thats right its already been benched . If some of those games don't change FPS as we seen that not CPU thats GPU. The Guy that did test is retarded with his conclusion.

They're using a dual gpu card. Basically what they showed was that at high enough resolutions, when the GPU becomes the limit, AMD can give you 1 or 2 more frames.

The goal of playing games is to get as much image quality while maintaining ~60 fps for shooters and ~30 fps for RTS/RPGs is it not?
 
Last edited:

IntelUser2000

Elite Member
Oct 14, 2003
8,686
3,787
136
http://www.legionhardware.com/articles_pages/cpu_scaling_with_the_radeon_hd_5970,1.html

So basically at the highest resolutions with the most powerful video cards, the phenom x4 is faster than the I7. I guess the x6 is the CPU to get if you want to game high end :p

I don't know why this happens. Core i7 is a beast in CPU bound scenarios, but can be sometimes slower than Phenom II when the application is GPU bound. The huge lead it has in lower resolutions would decrease in high settings/resolutions, but it should be still there, however minor.
 

Lonyo

Lifer
Aug 10, 2002
21,938
6
81
They missed the important bit: minimum frame rates.
Seriously.

Some guy in the video forum here did a Core 2 vs Core i7 and the Core 2 had terrible minimum frame rates compared to the Core i7 when it was stock vs stock, even when averages were quite close.
Phenom II vs Core i7 minimums would be interesting to see.
 

Fox5

Diamond Member
Jan 31, 2005
5,957
7
81
I don't know why this happens. Core i7 is a beast in CPU bound scenarios, but can be sometimes slower than Phenom II when the application is GPU bound. The huge lead it has in lower resolutions would decrease in high settings/resolutions, but it should be still there, however minor.

Phenom II platform has better pci express bandwidth than i5, and lower pci express latency than i7.
 

BD231

Lifer
Feb 26, 2001
10,568
138
106
Thats a BS article .We have many reviews that show it not to be so. I review doesn't make it so. Looks like the GPU is all done to me . We seen this befor when it first released. Also It be more interesting to see what XF and sli bring to the table . Oh thats right its already been benched . If some of those games don't change FPS as we seen that not CPU thats GPU. The Guy that did test is retarded with his conclusion.

No, he's not. Saying you get competitive performance gaming with an X4 Phenom II has been the case since the inception of the i7 series. It's not a " seen this before " scenario, it's the truth.
 

manimal

Lifer
Mar 30, 2007
13,559
8
0
They missed the important bit: minimum frame rates.
Seriously.

Some guy in the video forum here did a Core 2 vs Core i7 and the Core 2 had terrible minimum frame rates compared to the Core i7 when it was stock vs stock, even when averages were quite close.
Phenom II vs Core i7 minimums would be interesting to see.


I have stopped reading reviews at most places because of this. I simply do not care if one gpu gets 150 fps vs 100 in all the titles I play. What I do care about is the slowdowns in the heavy action sequences and the like. Lets start a petition..
.
 

IntelUser2000

Elite Member
Oct 14, 2003
8,686
3,787
136
Phenom II platform has better pci express bandwidth than i5, and lower pci express latency than i7.

This is not just true for dual PCI Express setups. It's true for single PCI Express as well.

They missed the important bit: minimum frame rates.
Seriously.

Minimum frame rates have much greater variability than average frame rates and are less reliable when measuring.
 

Nemesis 1

Lifer
Dec 30, 2006
11,366
2
0
No, he's not. Saying you get competitive performance gaming with an X4 Phenom II has been the case since the inception of the i7 series. It's not a " seen this before " scenario, it's the truth.


No your incorrect . If you go back and look at reviews when i7 was released you will see whats happening here. AMD was much closer to Intel in games than they are now. What this review was hand picked games that showed a stalled GPU . When i7 came out the 4000 series ATI cards were used . When the 5000 series came out I7 widened its gaming lead almost all here that follow this stuff knows this. Once the GPU has reached its limits the CPUs won't help alot . SO this article hand picked GPU limited FPS. A few times AMD was a couple of frames faster. But so was intel . When the next new GPUs apppear i7 will expand its gaming lead. But To say AMD top end is on par with i7 is pure dream world reality. Heres a good review that shows me everthing I need to see. The cost of parts doesn't matter. What were talking about is some clone comparring AMD to Intel . In this case AMD didn't show up for the fight.

Heres the link . Look at gaming because that what your toting horn About . The i5 dual core slaps AMD around big time and it pushes i7 around . I can do a review and create the winner if I so choose. Thats what you do when you put a gpu against the wall.

http://www.hardwarecanucks.com/foru...-core-i7-875k-unlocked-processors-review.html
 
Last edited:

Idontcare

Elite Member
Oct 10, 1999
21,110
64
91
Minimum frame rates have much greater variability than average frame rates and are less reliable when measuring.

This is true but the world of finance has already developed the statistical analysis tools and methods needed to provide robust characterization of the average values of the tail portions of a distribution.

It is pretty much a global standard (thanks to the Basel Accords) to analyze risk distributions in a manner according VAR (value at risk) and ETL (expected tail loss).

What we enthusiasts have been guided to care about is the median value of the distribution of frame rate (i.e. the average fps). But if one has the distribution of fps (hardocp includes these graphs in their reviews all the time) then computing the ETL (e.g. the average value of low framerates) is feasible and robustly defined as a meaningful metric as demonstrated by its utility in the finance industry (among others).

edit: An example of a practical implementation would be to take the average of the lowest 5% of fps values. While the actual lowest of the low values would be expected to fluctuate greatly from run-to-run we would expect the average of the tail distribution exceeding the 95% to be reasonably static.
 
Last edited:

evolucion8

Platinum Member
Jun 17, 2005
2,867
3
81
No your incorrect . If you go back and look at reviews when i7 was released you will see whats happening here. AMD was much closer to Intel in games than they are now.
http://www.hardwarecanucks.com/foru...-core-i7-875k-unlocked-processors-review.html

Doubtful, when the original Phenom debuted, was considerably behind of the Q6600 and the recently released Q9x00 series. Remember that Phenom II X4 debuted after the Core i7 launch. The gaming performance difference between the Core 2 Quad and the Core i7 isn't that great, even in multi GPU setups, the difference in playability is too little to justify an upgrade, specially if your C2Q is heavily overclocked. I doubt that a difference between 4fps and 45fps can help if the game is running beyond 100fps.

I call a sidegrade, jumping from my current CPU to a i7 920 unless if I overclock the 920 to the same speed as my current CPU, and never will be twice faster. I love to have twice the performance boost between upgrades, only Gulftown can offer me that, and too expensive for my tastes....

Far Cry 2, which is heavily biased to Intel architecture, shows that Phenom II X6 is quite close. Even with WIC, the performance difference in minimum frame rate between the Phenom II X6 and the powerful i7 870 is measly 3fps and 7fps at the maximum frame rate, that will make a difference in playability at all, based on your posted review.

PS: Gulftown is a monster of its own.