Phenom II > i7

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

DrMrLordX

Lifer
Apr 27, 2000
22,507
12,376
136
Originally posted by: aigomorla

The reason why im guessing it gamed worse then the AMD is because of HT ON.

Im hearing in games the HT ON blows.

Could it be more problems with threads bouncing around from real cores to virtual cores?
 

Zstream

Diamond Member
Oct 24, 2005
3,395
277
136
Originally posted by: aigomorla
dude it alway boils down to this.

Unlimited budget: i7
Limited budget: AMD.

A fully decked out i7 machine will spank a fully decked out amd machine but be 3x its cost.

Why don't you try to answer the question instead of marketing Intel for a change?

Anyways, the reason I suspect would be due to Hyperthreading. Many sites who have HT turned off experience an increase in gaming performance.

http://www.bcchardware.com/ind...include=view_photo.php
 

palladium

Senior member
Dec 24, 2007
539
2
81
Yeah I agree with aigo, the site used a single GTX280 which is probably why i7 failed to show its full potential. But I wouldn't expect it to be slower than PII though.
 

Cookie Monster

Diamond Member
May 7, 2005
5,161
32
86
Originally posted by: palladium
Yeah I agree with aigo, the site used a single GTX280 which is probably why i7 failed to show its full potential. But I wouldn't expect it to be slower than PII though.

Since when was performance of the CPU dependent on the GPU in a 3D application??

Unless you mean that these CPUs show more tangible performance increases when multi GPU setups are used.
 

JackyP

Member
Nov 2, 2008
66
0
0
Originally posted by: SlowSpyder
In many ways i7 and Phenom are more alike than i7 and C2D. Any game that can take advantage of i7 strengths, minus HT, should also gain performance on Phenom I would think.
I'm not sure if it is that clear cut and I would bet my money i7 anyway. You have seen the server benchmarks? Both are "sever CPUs" and "similar" as you say, but nehalem completely annihilates shanghai clock for clock in most server tasks, which are generally well-threaded.

I remember from the first wave of reviews that some games benefit from HT while others don't. On average it's a slight improvement, but if you have an i7 set-up and the time to tinker with it you could improve performance if you turn off HT in games where it's detrimental (can you turn off HT via software?).
 

Flipped Gazelle

Diamond Member
Sep 5, 2004
6,666
3
81
Originally posted by: Cookie Monster
Originally posted by: palladium
Yeah I agree with aigo, the site used a single GTX280 which is probably why i7 failed to show its full potential. But I wouldn't expect it to be slower than PII though.

Since when was performance of the CPU dependent on the GPU in a 3D application??

Unless you mean that these CPUs show more tangible performance increases when multi GPU setups are used.

It could be there is a cache latency - or a "bug" in the chipset - that is usually disguised until the GPU is choking on data, in which case even the most minor of blips can cause a measurable dip in fps. More GPU power would alleviate this.
 

Flipped Gazelle

Diamond Member
Sep 5, 2004
6,666
3
81
Originally posted by: OCguy
As soon as developers write games that can actually take advantage of an i7, it wont even be close anyway.

i7-like CPUs are the future.

Yup, I believe I said almost the exact same thing a month or two ago.

Still, I think that GPU power is at the forefront for the high-res gaming crowd.
 

dmens

Platinum Member
Mar 18, 2005
2,275
965
136
for what it's worth, crysis is the worst outlier for i7 when compared with penryn in the data ive seen. so if someone wants to spin data to make i7 look bad, just show crysis by itself... well, you know where im going with that.

p.s. by the worst outlier, i mean it is equivalent to maybe 2% slower. so that german data, i still call bullshit.
 

palladium

Senior member
Dec 24, 2007
539
2
81
Originally posted by: dmens
for what it's worth, crysis is the worst outlier for i7 when compared with penryn in the data ive seen. so if someone wants to spin data to make i7 look bad, just show crysis by itself... well, you know where im going with that.

p.s. by the worst outlier, i mean it is equivalent to maybe 2% slower. so that german data, i still call bullshit.

i'm not surprised if someone from AMDZone is the creator of that site.
 

palladium

Senior member
Dec 24, 2007
539
2
81
Just for the fun of it, I went over to AMDZone and true enough, they have a thread about this. Of all the posts I find this most interesting:

Perhaps certain games will favor i7 at similar resolutions but there is the "real-world effect" that means you have a lot going on in the background where Intel hasn't particularly been successful. perhaps the IMC will help but even Anand admitted that Phenom II was "smoother." I can attest to that at 1920x1080 res.

Thoughts?
 

Flipped Gazelle

Diamond Member
Sep 5, 2004
6,666
3
81
Originally posted by: palladium
Just for the fun of it, I went over to AMDZone and true enough, they have a thread about this. Of all the posts I find this most interesting:

Perhaps certain games will favor i7 at similar resolutions but there is the "real-world effect" that means you have a lot going on in the background where Intel hasn't particularly been successful. perhaps the IMC will help but even Anand admitted that Phenom II was "smoother." I can attest to that at 1920x1080 res.

Thoughts?

Wasn't AT referring to PhII vs C2Q with the "smoother" comment?
 

SlowSpyder

Lifer
Jan 12, 2005
17,305
1,002
126
Originally posted by: JackyP
Originally posted by: SlowSpyder
In many ways i7 and Phenom are more alike than i7 and C2D. Any game that can take advantage of i7 strengths, minus HT, should also gain performance on Phenom I would think.
I'm not sure if it is that clear cut and I would bet my money i7 anyway. You have seen the server benchmarks? Both are "sever CPUs" and "similar" as you say, but nehalem completely annihilates shanghai clock for clock in most server tasks, which are generally well-threaded.

I remember from the first wave of reviews that some games benefit from HT while others don't. On average it's a slight improvement, but if you have an i7 set-up and the time to tinker with it you could improve performance if you turn off HT in games where it's detrimental (can you turn off HT via software?).

I think you missed the point I was trying to make. I'm in no way saying Phenom is better than i7. What I am saying is that if some piece of software is written to take advantage of the i7 architecture, I would think you're more likely to also see gains on the Phenom then on a C2D/C2Q. The i7 and Phenom have some design similarities (IMC/NB on die, smallish L2 cache per core + larger shared L3 cache, the monolithic design allows the cores to communicate with each other more directly vs. the C2D design of two dual core chips on a single package with the dualies communicating with each other via FSB). So if a piece of software was written to take advantage of those design features of the i7 then I would think that Phenom would also likely see a boost on that piece of software compared to C2D/Q. Hope that made more sense.
 

palladium

Senior member
Dec 24, 2007
539
2
81
Originally posted by: Flipped Gazelle
Originally posted by: palladium
Just for the fun of it, I went over to AMDZone and true enough, they have a thread about this. Of all the posts I find this most interesting:

Perhaps certain games will favor i7 at similar resolutions but there is the "real-world effect" that means you have a lot going on in the background where Intel hasn't particularly been successful. perhaps the IMC will help but even Anand admitted that Phenom II was "smoother." I can attest to that at 1920x1080 res.

Thoughts?

Wasn't AT referring to PhII vs C2Q with the "smoother" comment?

Ehh, I couldn't find that comment in the PhII vs C2Q article. What page is it on..?
 

Flipped Gazelle

Diamond Member
Sep 5, 2004
6,666
3
81
Originally posted by: palladium
Originally posted by: Flipped Gazelle
Originally posted by: palladium
Just for the fun of it, I went over to AMDZone and true enough, they have a thread about this. Of all the posts I find this most interesting:

Perhaps certain games will favor i7 at similar resolutions but there is the "real-world effect" that means you have a lot going on in the background where Intel hasn't particularly been successful. perhaps the IMC will help but even Anand admitted that Phenom II was "smoother." I can attest to that at 1920x1080 res.

Thoughts?

Wasn't AT referring to PhII vs C2Q with the "smoother" comment?

Ehh, I couldn't find that comment in the PhII vs C2Q article. What page is it on..?

Ohhhhhhhhhhhhh, the pain........... lemme look......................

Found it. It in this article about PhII X3 720 Xfire performance.

After playing through several levels on each platform, even with the improved performance of the Q9550 after switching to the 9.3 driver set, we thought the Phenom II 940/720BE offered a better overall gaming experience in this title. Are we going to say those three words again? Our therapist advised us not to hold our feelings in as they would eventually manifest and be channeled into a bad review for someone. So without further adieu, Smoother Game Play, Smoother Game Play. That feels better, but all joking aside, we simply experienced better player movement and weapon control during heavy action sequences with our Phenom II processors compared to the Q9550 platform. This was especially true if we were running background applications (IM, File Transfers, AntiVirus, etc.) and especially if CPU usage was over 90%, the Phenom II system never stuttered or gave us a slight pause between level transitions like the Q9550 (editor - Sounds like an SSD review). It might not be noticeable to everyone and at first we thought it was a placebo effect, but doing a blind test with an A/B box always lead us to the Phenom II. The $64 question is if we had the same user experience with the i7 platform. The answer is yes. The i7 offered an improved game play experience over the Q9550 platform based on the same reasons we listed for the Phenom II. Simply put, integrated memory controller and CSI/HTT platform designs perform better than the previous front side bus platforms with these type of system loads.
 

JackyP

Member
Nov 2, 2008
66
0
0
Originally posted by: SlowSpyder
Originally posted by: JackyP
Originally posted by: SlowSpyder
In many ways i7 and Phenom are more alike than i7 and C2D. Any game that can take advantage of i7 strengths, minus HT, should also gain performance on Phenom I would think.
I'm not sure if it is that clear cut and I would bet my money i7 anyway. You have seen the server benchmarks? Both are "sever CPUs" and "similar" as you say, but nehalem completely annihilates shanghai clock for clock in most server tasks, which are generally well-threaded.

I remember from the first wave of reviews that some games benefit from HT while others don't. On average it's a slight improvement, but if you have an i7 set-up and the time to tinker with it you could improve performance if you turn off HT in games where it's detrimental (can you turn off HT via software?).

I think you missed the point I was trying to make. I'm in no way saying Phenom is better than i7. What I am saying is that if some piece of software is written to take advantage of the i7 architecture, I would think you're more likely to also see gains on the Phenom then on a C2D/C2Q. The i7 and Phenom have some design similarities (IMC/NB on die, smallish L2 cache per core + larger shared L3 cache, the monolithic design allows the cores to communicate with each other more directly vs. the C2D design of two dual core chips on a single package with the dualies communicating with each other via FSB). So if a piece of software was written to take advantage of those design features of the i7 then I would think that Phenom would also likely see a boost on that piece of software compared to C2D/Q. Hope that made more sense.
I got what you mean; but if software is optimisied I'd assume it will scale to n threads (as well as possible). If it uses n threads the gap will surely widen because of HT. Other than that (at 4 threads) both systems should benefit more or less the same, I agree.
 

aigomorla

CPU, Cases&Cooling Mod PC Gaming Mod Elite Member
Super Moderator
Sep 28, 2005
21,019
3,480
126
Originally posted by: Zstream
Originally posted by: aigomorla
dude it alway boils down to this.

Unlimited budget: i7
Limited budget: AMD.

A fully decked out i7 machine will spank a fully decked out amd machine but be 3x its cost.

Why don't you try to answer the question instead of marketing Intel for a change?

Anyways, the reason I suspect would be due to Hyperthreading. Many sites who have HT turned off experience an increase in gaming performance.
http://www.bcchardware.com/ind...include=view_photo.php

How am i marketing anything? Didnt i say this WAYYYYYYYYYY back?

Originally posted by: DrMrLordX
Originally posted by: aigomorla

The reason why im guessing it gamed worse then the AMD is because of HT ON.

Im hearing in games the HT ON blows.

Could it be more problems with threads bouncing around from real cores to virtual cores?

Isnt that what HT ON means? <--- not directed at you DrMxLord

And yes i think its from jumping from physical cores to virtual cores.
And when you SLI or TRI SLI or even QUAD-FIRE, the i7 has an advantage ove amd.

This IS NOT FUD. The i7 is not for everyone, just like a porsche will never be driven by everyone.
 

ShawnD1

Lifer
May 24, 2003
15,987
2
81
Originally posted by: DrMrLordX
Originally posted by: aigomorla

The reason why im guessing it gamed worse then the AMD is because of HT ON.

Im hearing in games the HT ON blows.

Could it be more problems with threads bouncing around from real cores to virtual cores?

I'm thinking this. HT was never a problem with the Pentium 4 because all of the logical cores were still on the same physical core. This is not true on the i7. If a game is trying to use 2 hardware threads and they are both put on the same physical processor core, what kind of effect will that have on the performance?

We need more info about how hyperthreading works and how Windows deals with it. This seems like one of those things where different operating systems could have very different performance when using the same hardware simply because they manage threads differently.
 

SMPEngineer

Junior Member
Apr 18, 2009
5
0
0
Originally posted by: jones377
Yeah this can't really be denied anymore as it shows up not on just one or two sites but in lots of places. And more importantly, no one has yet offered a good explanation for this.

Just throwing some ideas out there..

1) Video driver related
2) QPI latency vs HT latency
3) Hyperthreating
4) Crappy chipset?

I don't claim to know the answer but I am getting really interesting in this now. Like I said, it can't be explained away anymore.

None of the above.

1. This occurs with either nVIDIA or ATi hardware but ONLY when the Graphics hardware becomes the bottleneck. It is not driver related.

2. In order to find out the answer one has to be familiar with QuickPath. QuickPath is a 2HOP design while HyperTransport is a 3HOP design. In short this simply means that Quickpath is a lower latency design when compared to Hypertransport therefore this is also NOT the reasoning.

3. One could look to Hyperthreading but when faced with a CPU bottleneck the same is not observable. This of course indicates that it has to do with the Graphics sub system.

4. You're close but no cigar.

The reasoning is quite simple. It has to do with how PCI Express Header information is passed through the I/O Hub to the CPU. Generally speaking, PCI Express is naively not compatible with the QPi link. Intel engineers devised something called "Tunneling" which you can see described in this patent from Intel: http://www.patentstorm.us/pate...10000/description.html.

All this means is that there is a slight amount of latency associated with the exchange of information from the PCI Express bus to the QPi Link. This is the cause for what you're seeing. As for why the 965 is slower than the 920.. that remains a mystery. I don't believe this is related with the Phenom II issue we're seeing here.

It's worth mentioning that this ONLY occurs when the GPU becomes the bottleneck for performance. This generally means that when using SLI, Crossfire or Multi-GPU cards.. this phenomenon does not rear it's ugly head.

As for which CPU is more powerful... simple.. Core i7 by an extremely large amount.
 

ShawnD1

Lifer
May 24, 2003
15,987
2
81
Originally posted by: SMPEngineer
It's worth mentioning that this ONLY occurs when the GPU becomes the bottleneck for performance. This generally means that when using SLI, Crossfire or Multi-GPU cards.. this phenomenon does not rear it's ugly head.

As for which CPU is more powerful... simple.. Core i7 by an extremely large amount.

The shitty part is that GPU is the bottleneck at least 90% of the time. I'm using a ~3ghz E6600 and I still haven't really had any problems with CPU bottlenecking. At 1440x900 resolution with 0xAA and 8xAF (forced driver settings), every single game I have is GPU bottlenecked by my 8800GTX. I have a feeling that would still be true even if I had something like a GTX 280.
 

SMPEngineer

Junior Member
Apr 18, 2009
5
0
0
My experience has been that when using Dual GPU solutions, one is not GPU bottlenecked as often in the current titles therefore the behavior is not present.
 

Zstream

Diamond Member
Oct 24, 2005
3,395
277
136
Originally posted by: ShawnD1
Originally posted by: DrMrLordX
Originally posted by: aigomorla

The reason why im guessing it gamed worse then the AMD is because of HT ON.

Im hearing in games the HT ON blows.

Could it be more problems with threads bouncing around from real cores to virtual cores?

I'm thinking this. HT was never a problem with the Pentium 4 because all of the logical cores were still on the same physical core. This is not true on the i7. If a game is trying to use 2 hardware threads and they are both put on the same physical processor core, what kind of effect will that have on the performance?

We need more info about how hyperthreading works and how Windows deals with it. This seems like one of those things where different operating systems could have very different performance when using the same hardware simply because they manage threads differently.

That is wrong, HT on the P4 was turned off on many many applications. The same technology used in many Xeon servers.
 

Atechie

Member
Oct 15, 2008
60
0
0
I guess nobody here can read german?

The OC'ed the 940 to 955 speeds, they didn't have a 955 CPU and since the multiplier is not unlocked in the 940, they scewed the results...

Nice job, not understanding the article in question...and make bad assumptions based on that
 

mhahnheuser

Member
Dec 25, 2005
81
0
0
Originally posted by: SlowSpyder
Originally posted by: OCguy
As soon as developers write games that can actually take advantage of an i7, it wont even be close anyway.

i7-like CPUs are the future.

In many ways i7 and Phenom are more alike than i7 and C2D. Any game that can take advantage of i7 strengths, minus HT, should also gain performance on Phenom I would think.

Amen to that. I've waited a long time to read this comment. Agree 110% maybe now we will finally get better multi-core and 64 bit optimised code. U-bloody-reka!!!
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |