X CPUs.com - Phenon vs. Yorkfield gaming teaser using Lost Planet

[brentpresley]

http://*****/forums/mo...n-cpu-gpu-collide.html

Small teaser of Jack's difinitive 60+ game comparison of a Phenom and Yorkfield (both at 2.5GHz).

Looks like as interconnect traffic jumps, the Phenom makes up a lot of the ground it loses to the Yorkfield at lower resolutions.

Be interesting to see if this holds true for more than just this one game.

 

[ShawnD1]

Originally posted by: brentpresley
Be interesting to see if this holds true for more than just this one game.

Probably not. Bit-Tech is one of the few sites that benchmarks processors at high resolution as well as low resolution, and the Phenom loses in both Crysis and World In Conflict.
Crysis
World In Conflict

 

[Qbah]

All nice and dandy but... who in their right mind runs a QX9650 at 2.5GHz? The discussion is in my eyes purely academic as it has no real world usage. The QX runs at 3GHz natively, right? You could argue that you can probably OC the Phenom but you can do the same with the QX (alot higher actually and annihilate the fastest Phenom).

As a technology comparison, this article works ok, I guess. Bottom line is, the Phenoms have substantialy lower clocks can process less data at the same frequency.

 

[brentpresley]

Originally posted by: Qbah
All nice and dandy but... who in their right mind runs a QX9650 at 2.5GHz? The discussion is in my eyes purely academic as it has no real world usage. The QX runs at 3GHz natively, right? You could argue that you can probably OC the Phenom but you can do the same with the QX (alot higher actually and annihilate the fastest Phenom).

As a technology comparison, this article works ok, I guess. Bottom line is, the Phenoms have substantialy lower clocks can process less data at the same frequency.

If you don't clock them the same, then you get people complaining it isn't a fair comparison. If you DO, then you get people (like you) complaing as well.

Damned if you do . . .

 

[Qbah]

Originally posted by: brentpresley
[If you don't clock them the same, then you get people complaining it isn't a fair comparison. If you DO, then you get people (like you) complaing as well.

Damned if you do . . .

You can always find something to complain about However, after looking around it seems that the test is pretty valuable, since a 2,5GHz Quad from Intel - Q9300 costs similar to that Phenom. Though they could use it for the comparison then Yeah yeah, I'm just saying ! Though I remember reading that the Q9300 will get dropped and a bit higher clocked version with more cache will be put in that price point from Intel. Dunno where I read that though, was a few days ago.

 

[ShawnD1]

Originally posted by: brentpresley

If you don't clock them the same, then you get people complaining it isn't a fair comparison.

Who gives a shit what fanboys think? They'll buy from company X regardless of what the benchmark says.

Either test both processors at stock speeds or test both at the highest stable overclock.

 

[brentpresley]

Originally posted by: Qbah

Originally posted by: brentpresley
[If you don't clock them the same, then you get people complaining it isn't a fair comparison. If you DO, then you get people (like you) complaing as well.

Damned if you do . . .

You can always find something to complain about However, after looking around it seems that the test is pretty valuable, since a 2,5GHz Quad from Intel - Q9300 costs similar to that Phenom. Though they could use it for the comparison then Yeah yeah, I'm just saying ! Though I remember reading that the Q9300 will get dropped and a bit higher clocked version with more cache will be put in that price point from Intel. Dunno where I read that though, was a few days ago.

Yes. I saw that as well. Think it was HKEPC (sp?).

 

[Sylvanas]

I just skimmed the article but this is what I gather. Firstly, this should probably be in the CPU forum. Secondly Lost planet is a terrible console port and should not be used for benching. On that note it should be said that you can't come to a useful conclusion on the results of ONE game....they need to test many more- each engine performs differently.

 

[brentpresley]

Originally posted by: Sylvanas
I just skimmed the article but this is what I gather. Firstly, this should probably be in the CPU forum. Secondly Lost planet is a terrible console port and should not be used for benching. On that note it should be said that you can't come to a useful conclusion on the results of ONE game....they need to test many more- each engine performs differently.

Yeah, it's pretty obvious you skimmed it and didn't READ it.

1) my OP stated that this was a PREVIEW of a full 60+ game comparison on the C2Q vs Phenom, and not the only comparison that one should consider

2) the ENTIRE point of the comparison is to find at what resolutions, and for each chip, that the game goes from being CPU bound to GPU bound. And just imagine the work required to do this for EACH of the 60 games.

3) please read before commenting.

 

[Sylvanas]

Originally posted by: brentpresley

Originally posted by: Sylvanas
I just skimmed the article but this is what I gather. Firstly, this should probably be in the CPU forum. Secondly Lost planet is a terrible console port and should not be used for benching. On that note it should be said that you can't come to a useful conclusion on the results of ONE game....they need to test many more- each engine performs differently.

Yeah, it's pretty obvious you skimmed it and didn't READ it.

1) my OP stated that this was a PREVIEW of a full 60+ game comparison on the C2Q vs Phenom, and not the only comparison that one should consider

2) the ENTIRE point of the comparison is to find at what resolutions, and for each chip, that the game goes from being CPU bound to GPU bound. And just imagine the work required to do this for EACH of the 60 games.

3) please read before commenting.

I was in a hurry . If 60 pages of this article are as detailed as this one item then I look forward to reading the full article when it is released.

 

[Wreckem]

Originally posted by: Sylvanas
I just skimmed the article but this is what I gather. Firstly, this should probably be in the CPU forum. Secondly Lost planet is a terrible console port and should not be used for benching. On that note it should be said that you can't come to a useful conclusion on the results of ONE game....they need to test many more- each engine performs differently.

Well, lost planet is pretty much the only game that really takes advantage of multi-core/quad core CPUs.