Intel Core 2 Gaming Performance

[/dev/null]

No good for me. VIA chipset = trash. Give me a sweet 975 or 965 board.

I'm pretty sure my C3 (router) runs on a via chipset. 0 problems. These aren't the KT133 days.

 

[Rasha]

Kyle, i dunno if this has been said or not? but i feel sorry for you, that you have people on this forum that have no fucking clue that you provided a benchmark that reflects reality vs a bullshit benchmark to try to prove one is better then the other. this is just my opinion but if you game at 1024 x whatever you should of been swallowed at birth. I would like to see the SLI numbers, but it was posted why that wasn't the case. this is the most useful review i have seen FOR ME, i know that im not going to "upgrade" to conroe because i wont being doing sli, there is no benifit to me just my 2 cents

 

[FrgMstr]

i think it is nice to finally get some real world testing that is able to cut right through all the marketing hype. i have to admit i was pretty excited about the new intel processors coming out (not excited enough to upgrade but still excited). after reading the article, it is clear to me that i should be in no hurry to upgrade. thanks [H]

No there is no reason to go out and drop big cash on a new Conroe system if you already are into a high end AMD box. Certainly when you go to upgrade, you need to evaluate if it is time to drop the big bucks on a new configuration though.

The one caveat to this is our brothers out there doing video editing at home. The Conroe is simply a monster when it comes to that and yeilds some hugely clear advantages over many mainstream home systems that are out there. If you go AMD for video editing on your next upgrade you are cheating yourself.

 

[HeavyH20]

I read the article through, word for word. One criticism I will have to yield on. The single video card will be the bottleneck in the game testing, so the Core 2 Duo does not really get a chance to shine. If this had been a Crossfire or SLI test system, the Core 2 Duo would have shown its true advantage. And, even if the Intel was just at par with the AMD, that is a big change considering AMD's domination in gaming for the past couple of years.

Bottom line, the article does illustrate what the Core 2 Duo will do (and more importantly, NOT do) for the average gamer, but it does not highlight what the chip would do on the .5% of systems with SLI and Crossfire. That, in truth, can not be considered a failing, but a tempering of the Intel hype that most gamers would experience. How would you feel swapping out your AMD system, CPU, motherboard, and memory to go Intel, then see NO gains? That is the stark reality most gamers would experience.

I, for one, will go with the Intel since I will use SLI, a 30 inch monitor, and eventually go quad SLI. For that, the Core 2 Duo is ideally suited.

 

[Donnie27]

i think it is nice to finally get some real world testing that is able to cut right through all the marketing hype. i have to admit i was pretty excited about the new intel processors coming out (not excited enough to upgrade but still excited). after reading the article, it is clear to me that i should be in no hurry to upgrade. thanks [H]

Please also note that this thread has how many posts compared to the others? This was just one part of testing the Core 2 Duo. Since it offfers a small ray of hope for some folks to cling to, I'm NOT surprised either. Sure it makes sense if you ONLY game. Same thing can be said and the same point can be made about folks who dumpped perfect good Sc-939 and 754 for AM2 and with gains. Now these wild folks are supposed to be conservative now? At least after the games are finished, we see AMD taking a beating.

Yonah was already giving AMD all it wanted before Conroe.

 

[dark_reign]

I'm pretty sure my C3 (router) runs on a via chipset. 0 problems. These aren't the KT133 days.

VIA is still bug-ridden and inferior to Intel's chipsets. I also wouldn't put much faith in a $58 motherboard. I'd only go there if I was building a rig for my grandparents.

 

[FrgMstr]

I read the article through, word for word. One criticism I will have to yield on. The single video card will be the bottleneck in the game testing, so the Core 2 Duo does not really get a chance to shine. If this had been a Crossfire or SLI test system, the Core 2 Duo would have shown its true advantage. And, even if the Intel was just at par with the AMD, that is a big change considering AMD's domination in gaming for the past couple of years.

Bottom line, the article does illustrate what the Core 2 Duo will do (and more importantly, NOT do) for the average gamer, but it does not highlight what the chip would do on the .5% of systems with SLI and Crossfire. That, in truth, can not be considered a failing, but a tempering of the Intel hype that most gamers would experience. How would you feel swapping out your AMD system, CPU, motherboard, and memory to go Intel, then see NO gains? That is the stark reality most gamers would experience.

I, for one, will go with the Intel since I will use SLI, a 30 inch monitor, and eventually go quad SLI. For that, the Core 2 Duo is ideally suited.

And we did exactly point that out as well, not as a fact, but rather a possibility.

When it comes to playing games, the only persons that need to be even a little concerned with upgrading their CPU to a Core 2 processor might be those with high-end SLI, CrossFire, or GeForce GX2 video cards and we have yet to prove that due to the testing limitations we ran into. Then, and only then, you might see an Intel Core 2 processor deliver a performance advantage.

We will surely go back and test SLI and Quad very soon! I can't wait to see the REAL gaming results of that.

 

[Firebat]

I did read the article, and i have mixed feelings about it.

It's generally a given fact that the higher resolution run, the more the GPU will be stressed, and the less the CPU will be. Clearly the point of it was to show you in high res gaming how little the CPU has an effect in performance in current games. Let's be honest here, gamers usually upgrade there GPU more often than there CPU. I certainly do, and when I purchase a CPU I want it to last. Running the games at lower resolutions shows me how the cpu will fare, three years down the road where the GPU is no longer the bottleneck. I don't want to be told that the differences are negligiable between the FX 62 and E6600 in games, only to find out my FX 62 is getting ass raped at 1280x1024 a few years down the road.

Or perhaps [H] could just include a few lower res tests.

 

[FrgMstr]

The thing you should have added to this review is a X2 3800 and pentiumD 5XX which most likely would have shown similar results to all of these chips running the benchmarks you ran in this review.

I hope in the future you add SLI/crossfire setups after these setups hit the market by storm .

Sadly we were not given that much time by Intel. As has been explained here in this thread, pulling real world gaming experiences and docuementing them and running a two minute timedemo are two very different things. We had planned for a much larger comparison and surely with GX2 working, but Intel pulled the rug out from under us and cut our testing time by half after we were 1 day in on testing.

 

[Argion]

I have been browsing and reading this site and others for years now but rarely provide any feedback in the forums. This thread has been the first in a long time that has made me want to provide my opinion.

Some background; I currently have a socket A based system and I am in the market to build a new PC. I was planning on getting an AM2 based system until I saw some of the posts and previews on sites indicating how good Conroe performance would be. Synthetic benchmarks server their purporse, but nothing beats real life performance. At the end of the day there are really three factors I care about:

1. Real Life Performance
2. Price
3. Upgrade Path

The confusion lies in how does one define "Real Life Performance". This difers from person to person, but in all the reviews that I have read so far, the ones on this site match my preferences the most. I spend 90% of the time on my pc gaming. To me, that is the most important when looking at performance. Even on my setup, with a 6600GT I try to run my games at 1600x1200 and I will take a frame rate hit, if I can make the game look a bit better but still be very playable (oh.. and this is on 19" screen). I will not compromise my gaming experience by playing with low settings (why pay 50 bucks for a game that is supposed to be an immersive experience in the first place?).

To me, the article about gaming performance between Core 2 Duo and the A64's was great, because it was targeted at what I am mostly interested in a new pc. Whether you like it or not, we are no longer in the days that one component of the system, namely the CPU, is responsible for 90% of a systems performance. Whether you like it or not, ALL components of the system attribute to this. An awesome processor in a crapy setup is not value added and will do nothing to enchance your day to day computing, and if someone actually puts something like that together ought to have their head checked.

It appears to me, that a lot of the users posting in this thread, have not read the full article. Assumptions and constraints have been clearly defined, system specs have been posted etc. I keep seeing questions, why did you not use this chipset, why did you not use this video card, why did you not use this resolution... the answers to all these questions can be found in the document itself... just read the first page and don't skip to the conclusion.

Also, this article is part of a 4 part piece. If you read them all together, you basically get what other sites include in one review. Some may not agree with this appoach, but it is the way they decided to do it and I was personally fine with it. If you read all 4 articles, you will see that the Core 2 Duo processors do provide with a better although not overwelming performance over the existing AMD products, for a better price. I have been working in the computer business for a long time, and the suggestion as made in some of these posts that Conroe can display 100% performance increase over a similar A64 is highly unlikely. Given that the same A64 probably outperformed a similar P4 based by lets say a couple percentage points, what you are really saying is that this new generation of Intel processors outperforms its previous generation by more than 100% (Moore's law be damned I guess). Take into account that the Core 2 is also heavily influenced by the P3 architecture (older but still better in my opinion than the P4 - we can have another discussion on this) and you can see how this performance increase seems a bit more unlikely (Please note the use of the word "unlikely" instead of "unrealistic").

As a consumer, I like to do my research before buying any product, and that means looking at reviews, reports etc from more than one source. This article and all other reviews on this site, are just a piece of the puzzle I need to assemble before making my decision. I will never choose a product based on what only one person or group of people say, but I will factor these more based on previous experience. If anything, the reviews I read on this site, have always been informative and proven correct almost all the time. It is also a different type of reviewing that contrasts with all the pre caned reviews out there.

I am not going to run out and buy this processor now because of this or other reviews. We can talk about the diferences between chipsets, memory and video cards all day, but at the end of the day, it comes to personal preference on how you would like to build your ideal system. I am not building something today, because I would like to have an nForce 590 motherboard to go alongside my Core 2 Duo so that I can set it up in an SLI configuration(Please no comments on why SLI and not Xfire or Nforce vs Intel chipset). So guess what, I am going to wait just a bit longer (because I can), and wait for all the new reviews that will come out when the motherboards I want will be available and I can see how an SLI setup affects gaming performance. And if the nForce motherboard don't live up to their hype then an Intel chipset becomes an option etc. The existing articles however have gone a long way in convincing me that for now, with existing games and existing *available* hardware, Core 2 Duo is the way to go and for a good price, and at the end of the day, that is all I can ask for.

 

[JetUsafMech]

Here's what I don't understand. There was a great, in-depth article here by [H]ardOcp that drew the conclusion that if you had the highest end CPU, you were GPU limited. [H] could have simply added the Conroe to the mix of those graphs, linked to it, and been done with it. That way people could go "Ohhhhhh.....the bargraphs scale the same everywhere, WTF nVidia/ATI, get your asses in gear, you're bottlenecking us".

And while I understand that Intel cut your legs out regarding the 975 chipset usage, how in the world did every other site get one, not to mention users on various forums, much less the same people that somehow enabled Xfire. (I realize there wouldn't be alot of difference, but using arguably the best AMD chipset vs. arguably a mid-range chipset isn't quite apples to apples).

Thanks for hearing me out.

My .02

Donnie27--agree about the Yonah comment.

 

[/dev/null]

VIA is still bug-ridden and inferior to Intel's chipsets. I also wouldn't put much faith in a $58 motherboard. I'd only go there if I was building a rig for my grandparents.

Dunno..my C3 setup was quite a bit more than $58 but has not yet crashed running openbsd 3.9 with PF, nor hickuped nor really showed much load at all.

All chispets have bugs. It's just which ones do you want to live with ?
may I mention the nforce hardware firewall? UGH. How about sata?

Sure via has their problems but nothing major that I've seen (at least not in the unixes I use)

 

[ecktt]

I am HIGHLY disappointed in this review as it was not a NOT a CPU review but a review of a video card. Kudos to [H]ardocp for a great Video card review but is down right sucked as a CPU review. Even if this review was slanted towards gamers perspective and what the CPU would have done with currently available hardware, it still paint a an incorrect image. I think the numbers would have been same if they used a single core AMD or Intel processor of slightly lower clock. Why? It was clear that the Video card was the limiting factor. All it says is that gamers should spend more money behind a video cards than the CPU (not a bad idea IMHO). Does this go to say that the Conroe isn't worth it? No it doesn't! Strange I didn't seen Quake 4 with some SMP action. I've been a long time fan of Hardocp and hope they learn from this travesty.

 

[FrgMstr]

I am HIGHLY disappointed in this review as it was not a NOT a CPU review but a review of a video card. Kudos to [H]ardocp for a great Video card review but is down right sucked as a CPU review. Even if this review was slanted towards gamers perspective and what the CPU would have done with currently available hardware, it still paint a an incorrect image. I think the numbers would have been same if they used a single core AMD or Intel processor of slightly lower clock. Why? It was clear that the Video card was the limiting factor. All it says is that games should spend more money behind a video cards than the CPU (not a bad idea IMHO). Does this go to say that the Conroe isn't worth it? No it doesn't! Strange I didn't seen Quake 4 with some SMP action. I've been a long time fan of Hardocp and hope they learn from this travesty.

Thanks for your thoughts, they are noted. We will continue to outline the real gaming experiences powered by the hardware instead of putting hardware in a nonrealistic gaming environment so that then, and only then can differences in performance be assigned.

 

[ciparis]

This article demonstrates the continuance of HardOCP's decline in relevance to me, turning what was once *the* place for an enthusiast to go to first into a place where I am unlikely to learn anything interesting about new hardware.

Why focus an article so tightly on the patently obvious truth that GPU-limited gameplay cannot be improved significantly with a CPU upgrade, regardless of how much faster that CPU is?

Choosing the best-looking, best-performing settings for a card *is* helpful, but it's helpful as an addition to a comprehensive battery of tests, where an editor's experience can help someone find meaning in a sea of data, especially for those incapable of drawing their own conclusions; it's not a substitute for that data, especially for those who can.

This crusade against full testing is resulting more and more in articles almost devoid of substance.

 

[FrgMstr]

Here's what I don't understand. There was a great, in-depth article here by [H]ardOcp that drew the conclusion that if you had the highest end CPU, you were GPU limited. [H] could have simply added the Conroe to the mix of those graphs, linked to it, and been done with it. That way people could go "Ohhhhhh.....the bargraphs scale the same everywhere, WTF nVidia/ATI, get your asses in gear, you're bottlenecking us".

And while I understand that Intel cut your legs out regarding the 975 chipset usage, how in the world did every other site get one, not to mention users on various forums, much less the same people that somehow enabled Xfire. (I realize there wouldn't be alot of difference, but using arguably the best AMD chipset vs. arguably a mid-range chipset isn't quite apples to apples).

Thanks for hearing me out.

My .02

Donnie27--agree about the Yonah comment.

This has already been fully addressed multiple times. Please read the thread.

 

[FrgMstr]

This article demonstrates the continuance of HardOCP's decline in relevance to me, turning what was once *the* place for an enthusiast to go to first into a place where I am unlikely to learn anything interesting about new hardware.

Why focus an article so tightly on the patently obvious truth that GPU-limited gameplay cannot be improved significantly with a CPU upgrade, regardless of how much faster that CPU is?

Choosing the best-looking, best-performing settings for a card *is* helpful, but it's helpful as an addition to a comprehensive battery of tests, where an editor's experience can help someone find meaning in a sea of data, especially for those incapable of drawing their own conclusions; it's not a substitute for that data, especially for those who can.

This crusade against full testing is resulting more and more in articles almost devoid of substance.

Thanks for the support over the years, you will be missed. Your concerns have already been addressed multiple times in this thread, please read it.

 

[/dev/null]

I think alot of people don't understand WHY [H] reviews like this so they take the wrong conclusion that [H] is doing something wrong.

They figure "typical people" game at 1280x1024 with a single card (most people, even [h] people can't go blow $500+ on graphics alone). They went on tos how that spending more on a core 2 duo in gaming when you have a single card at that resolution (typical gaming specs) DOES NOT MAKE SENSE esp if you alredy have an OK setup (e.g.: ath64).

This was fine for me. I dont have a typical setup (I run at 1600x1200) but I will not be going SLI anytime soon (unless 7800GT's hit sub $150 and I come up with some extra money). If you plan to run a SINGLE VIDEO CARD, the cpu currently doesn't matter at TYPICAL resolutions (1280x1024). Sure you can show conroe is faster but I don't think I've gamed at sub 1280 for 2 years now.

If you have $500+ in graphics and a 24" or 30" dell, etc etc this review is NOT for you.

It's not that hard to understand.

I love to read reviews like this, esp when the parts are broken up so I don't have to read about the parts I don't care about.

I have a dual 246 (pretty slow compared to all you guys running 165's @ 2.8ghz) and a single video card and this review shows me I clearly have no reason to upgrade. I'm more than happy to run at 1600x1200 with 8af or 16af and NO aa on my 7800GT.

Is this methodology bad? I don't think so. I'd love to see a poll of how many of you guys run better than the 7900GTX shown in this review.

[H]: Keep up the relevant reviews Good job with the review.

 

[zeekle]

I am not going to run out and buy this processor now because of this or other reviews. We can talk about the diferences between chipsets, memory and video cards all day, but at the end of the day, it comes to personal preference on how you would like to build your ideal system. I am not building something today, because I would like to have an nForce 590 motherboard to go alongside my Core 2 Duo so that I can set it up in an SLI configuration(Please no comments on why SLI and not Xfire or Nforce vs Intel chipset). So guess what, I am going to wait just a bit longer (because I can), and wait for all the new reviews that will come out when the motherboards I want will be available and I can see how an SLI setup affects gaming performance. And if the nForce motherboard don't live up to their hype then an Intel chipset becomes an option etc. The existing articles however have gone a long way in convincing me that for now, with existing games and existing *available* hardware, Core 2 Duo is the way to go and for a good price, and at the end of the day, that is all I can ask for.

I think its really a good idea to wait till more motherboards are on the market. If you are on an AGP system (like me) it might even be worth waiting till late September/Early October for the next gen video cards to start hitting the market.

 

[Donnie27]

Here's what I don't understand. There was a great, in-depth article here by [H]ardOcp that drew the conclusion that if you had the highest end CPU, you were GPU limited. [H] could have simply added the Conroe to the mix of those graphs, linked to it, and been done with it. That way people could go "Ohhhhhh.....the bargraphs scale the same everywhere, WTF nVidia/ATI, get your asses in gear, you're bottlenecking us".

And while I understand that Intel cut your legs out regarding the 975 chipset usage, how in the world did every other site get one, not to mention users on various forums, much less the same people that somehow enabled Xfire. (I realize there wouldn't be alot of difference, but using arguably the best AMD chipset vs. arguably a mid-range chipset isn't quite apples to apples).

Thanks for hearing me out.

My .02

Donnie27--agree about the Yonah comment.

Yea, it's always funny how folks tried to pretend Yonah was a figment of someone's imagination! Many sites didn't even bother testing this Wonderkinderling! I wanted to read reviews that made up the Good, the Bad and the Sucky. In the end, I'm glad I saw how C2D does with Power Savings Mode turned off so I can compare it to others with it turned on to see just how important that feature is.

 

[RooK]

Kyle mostly only did "real gameplay benchmarks" at 1600x1200. Why weren't more tests done at 1024x768 or 1280x1024? Does everyone use 20"+ monitors and have a top end video card?

I'm glad other sites are showing more realistic scenarios with their gameplay benchmarks.

i second that one.

Although I guess for the people who are getting richer and richer thanks to politics, this review is great.

 

[Terra]

Is this methodology bad? I don't think so. I'd love to see a poll of how many of you guys run better than the 7900GTX shown in this review.

[H]: Keep up the relevant reviews Good job with the review.

QFT.
Lets keep [H]ard about real world gaming...then the "timedemo/800x600"-peeps can go somewhere else?

Terra...

 

[RunOrDie99]

You sound like the US Patent office in 1890 : everything that could be invented was invented.

Hear this : in one year you'll have GPUs 2x as powerfull as today.R600 is a monster and is only a few months away.They will unlock Conroe's potential.
Having a Conroe means you are future-proof.

So if I get no benefits until G80/R600, why would I buy a CPU now instead of waiting until R600/G80 comes out and buying a faster CPU for the same money at the same time I upgrade my graphics card?

 

[FrgMstr]

I love to read reviews like this, esp when the parts are broken up so I don't have to read about the parts I don't care about.

Is this methodology bad? I don't think so. I'd love to see a poll of how many of you guys run better than the 7900GTX shown in this review.

[H]: Keep up the relevant reviews Good job with the review.

Thanks, I enjoyed the format as well and I think it is one of the reasons the article is getting so much attention....it is not buried in 30 pags of crap to sift though. We will likely keep the short and sweet format as much as we can.

Thanks for seeing the big picture.

Lower powered processors will be tested when we get them. Some sites enjoyed samples that were not shared with us. As stated, wish would have had more time to cover a broader scope and certainly put SLI into the mix. Next month...

 

[/dev/null]

i second that one.

Although I guess for the people who are getting richer and richer thanks to politics, this review is great.

Did you guys even read the review? Most where done at 1280x1024!

 

[coldpower27]

Yea, it's always funny how folks tried to pretend Yonah was a figment of someone's imagination! Many sites didn't even bother testing this Wonderkinderling! I wanted to read reviews that made up the Good, the Bad and the Sucky. In the end, I'm glad I saw how C2D does with Power Savings Mode turned off so I can compare it to others with it turned on to see just how important that feature is.

I can justify not testing Yonah due to the fact it wasn't directly marketed as a desktop processor, it's exemplary as a performance/watt processor but it also had fairly limited motherboard support.

Though the sites I usually go for Hardware Review, are Anandtech and TechReport did benchmark Yonah so I was satisfied nonetheless.

 

[FrgMstr]

Actually all of our applesto-apples gameplay was done at 1280x1024. Funny I see most folks glossing over the releative small differences in REAL gameplay there.

http://enthusiast.hardocp.com/article.html?art=MTEwOCw5LCxobmV3cw==

A lot of folks still simply do not understand that real gameplay and timedemos do NOT show the same results...again this is why we stopped using time demos.

You can sit down and game with Intels X6800 and E6700 and FX-62 and I will gurantee that you will find no discernable differences in the gameplay experience.

 

[cronic007]

I hope you guys got plenty of traffic off of this completely amd biased review, because your credibility as a impartial review site is gone. Good Luck!

 

[Terra]

I hope you guys got plenty of traffic off of this completely amd biased review, because your credibility as a impartial review site is gone. Good Luck!

*chough*
Click me

Terra - Reality bites

 

[Riptide_NVN]

Kyle you have more patience than I do to put up with all this idiotic nonsense.

I for one am happy that you didn't perform the review as a strict CPU benchmark. If Conroe won't benefit me much at higher resolutions w/AA and AF turned on then is there supposed to be something wrong with me because I want to know that?

Contrary to what some people say I have seen multiple posts around various forums from people that were saying Conroe would provide significant benefits even in 16x12 or higher w/AA,AF turned on. I was skeptical, glad to know I've been for the most part vindicated there. But someone had to do the tests and for that I am thankful.

Jesus Christ there are a million other sites out there that did everything in 1024x768 and people still want to whine and complain because you guys did it differently. I wish they would shut their traps, quit complaining, and go read the reviews that were done in the way they wanted. There's only what? 25 or more out there like that?

 

[cronic007]

*chough*
Click me

Terra - Reality bites

I am refering to the game performance review, the other reviews are decent, but the game performance review is totally wacked. anybody heard of sli or crossfire here?

 

[dad11345]

*chough*
Click me

Terra - Reality bites

People who have already made up their minds to buy Conroe are going to react to someone pointing out reasons why they may not need it. Look in cronic007 signature for a clue on his thought process.

 

[digitalfreak]

Crossfire on a 975 board should have been done, plain and simple. If you wanted to prove that the C2D is no faster than the FX-62 with a single GPU, that's fine. That's really only half the story though.

I really hope you're not including sites like AT in your "BS & liars" comment either.

 

[HOCP4ME]

Okay, I didn't have time to read through all 23 pages, but I would just like to say that I think the [H] did a good job with this review. It was meant to show what improvement someone upgrading from S939 to Conroe right now could see. Had the review beeen like all the others and forcasted a 100% improvement, there would have been a lot of users very dissappointed after buying a new processor, mobo, and RAM only to find out there is little increase in their framerates. Rather than let that happen and then tell them "oh, but your games are GPU-limited, so you need to get SLI and play at 800x600 with no AA or AF to see the improvement", [H] decided to tell those people right away that upgrading to Conroe would not improve their gaming experience. Yes, Conroe is faster. No, you won't see a difference in today's games.

Having said that, however, I do think some things could be improved:
1. Do another review in the future with SLI or Crossfire if you can. There are some people (by the looks of this topic, many people) who would like to see what Conroe can do when given all the GPU power it wants. As Anandtech showed us, with Crossfire, there is still nowhere near a 100% improvement, but it is a noticeable one anyway.

2. I'm glad you will be doing some overclocking later, but when you do, don't say "we overclocked the Conroe to 4Ghz, but didn't see an improvement in any games because they were GPU-bound", because that would be pointless. I'm sure you guys would never do anything like that though.

3. Make sure you make readers understand that most of the games were GPU-bound, and that Conroe did indeed have more power, it just wasn't being used and may be used with the next generation of video cards.

Also, I think making different articles for different categories was a great idea. Good job with that.

Overall, it's a great review, and really shows what the real-world benefits of upgrading now would be. In a CPU review, however, you need something else too. You need a clear indication of which CPU gives you more power, whether it's used or not, because it could be used in the future. Your CPU-intensive applications article covers this well, but the gaming article doesn't. If you were to apply the same techniques to overclocking, you would end up saying "don't bother overclocking your CPU from 3GHz to 4Ghz, because the games are GPU-bound anyway, so you might as well underclock to 2GHz for increased stabillity". That obviously doesn't make any sense, which is why you always use CPU-inensive benchmarks for CPU overclocking. Do the same for CPU-reviewing. The two parts of any CPU review are "how much raw power does it have?" and "how much real-world difference will this power make in my applications?" You are perhaps the only site that takes the long amounts of time required to answer the second question. In doing so, however, you have somewhat forgotten about the first question, and I think that's why so many people are complaining about it.

But seriously guys, please stop yelling at Kyle and the others for doing a real-world test instead of a raw-power test. Do you have any idea how long it takes to do a "highest playable" gaming test? They have to try a setting, then run through a level in the game, then raise a setting, then run through the level again, then raise another setting, then run through the level, then lower another setting so they can raise another setting, run through the level, and continue until they find the optimal settings, and repeat that on every system for every game. Whereas for a raw-power benchmark, you just have to pick some settings and run through the game once on each system for each game. We should be very glad that the [H] takes the time to do these benchmarks for us, and even take extra time to do less popular things such as widescreen, quad-SLI...etc.

You know what I think it is? I think that the [H] crew is so used to doing their super-awesome-blows-every-other-site-out-of-the-water video card reviews that they tried to apply the same approach to this CPU review. So thanks for taking the time to be the first ones to do that.

 

[cronic007]

Take the personal stuff to PM - Kyle

 

[Riptide_NVN]

I do agree with those that say it would've been nice to have had the review done with GTX SLI or X1900XT Crossfire. But you can go elsewhere for that if you wish.

 

[HOCP4ME]

I do agree with those that say it would've been nice to have had the review done with GTX SLI or X1900XT Crossfire. But you can go elsewhere for that if you wish.

Most people don't realize that they tried to do just that. Their 7950GX2 didn't work with the mobo. And when they realized that, there wasn't enough time left for them to go out and switch everything to Crossfire.

 

[Donnie27]

I can justify not testing Yonah due to the fact it wasn't directly marketed as a desktop processor, it's exemplary as a performance/watt processor but it also had fairly limited motherboard support.

Though the sites I usually go for Hardware Review, are Anandtech and TechReport did benchmark Yonah so I was satisfied nonetheless.

Neither was Mobile AthlonXP Barton, but did that stop the Flood of reviews, just trying to be fair here. Funny thing was Northwood C still kicked the crap out of it.

 

[dad11345]

I could give a rats a++ whose hardware I buy. I buy what is fastest, period. I have owned 8 different dual core AMDs over the past 2 years, so I can't see how I would be biased towards intel. The last Intel I owned was a 2.4c, because that was the best ocer at the time. I just hate to see a review that shows, at best, half of the story.

cronic007, would you not at least give some thought to the effect you having already ordered your Conroe has on why you feel the way you do. I am glad your getting the processor as I feel the 6700 is the best buy for gamers who need to buy a processor or just want too.

Think about it this way, if 25 other sites are looking at it the same why not have someone else look at it differently? Who wants to read 25 reviews which are the same? To approach a review differently is not a credibility issue. It would take a misrepresentation of test result or a flawed conclusion to warrant such a comment. I do not see either.

 

[dekard]

nice article... thanks Kyle and [H] Crew!