[H] Core 2 Gaming Benchmarks
- Thread starter Skrying
- Start date
eSa said:
It would still not be enough, considering that the 6600 is sitting at a $316 price point. There's almost performance parity, the conroe sucks less power and is probably a better overclocker.
And another big one: The AM2 CPUs require DDR2-800 memory, the conroes only DDR2-533. DDR2-533 and DDR2-666 (CL4) are pretty affordable right now but for two 1GB sticks of DDR2-800 I'd have to cough up €70 more. That basically means that AMD would have to lower the FX-62 price to €250 max. in order to remain even remotely price competitive with a conroe 6600.
And remember that the 6600 isn't even an high-end part, it's middle of the line.
Last edited by a moderator:
Price cut rumours (yes I know the source is dodgy!)
The AM2 cpus don't "require" ddr2-800 (of course, for the absolute best performance, they do, but the difference isn't that huge - I'm not even sure they benefit more than do conroes).L233 said:
Still, you have a point, amd sure will have some trouble offering really competitive cpus. Mainboards are probably going to be a bit cheaper, together with the slightly lower (at this point, still rumoured) price of the X2 3800+ at least at the "low-end of mainstream" market the offerings could be quite competitive (that x2 3800+ doesn't lose too badly against the 1.86Ghz conroe).
Of course, enthusiasts probably really will switch to conroe, as a bonus it even overclocks very well.
As long as intel doesn't sell low-power versions, the Athlon 64 X2 EE versions might generate some sales too, personally I think it's a nice idea. Depending on whose power draw figures you look at today they might indeed draw somewhat less power, which makes it easy to cool them silently (some discrepancies in the power draw figures is due to the used chipsets, especially nf5 is a power hog).
Last edited by a moderator:
Dunno. Availability maybe? If you're looking to build a PC in two weeks time what choice do you have?L233 said:
Those prices don't look sufficiently keen to compete with Conroe in the long-term, but AFAIK Intel have been playing pretty cool on the ramp of Conroe. Presumably AMD are gambling that Conroe will be supply limited for the next few months, and when it really ramps they'll drop the price again.
What's the motherboard situation for Conroe? There were rumours in various places that they may be turning up late too.
Overall, I think maybe AMD are trying to keep prices as high as possible for as long as possible.
Last edited by a moderator:
I did not find Hard's Review that bad (/me runs and hides). Only because they have been cartering their reviews to the gamers and gaming experince using custom demos. As well as judging what they find to be the best experince. And usally most of their users don't run at 800x600, ect. So in that context the review was not that bad as most of us know that faster CPU will not help much when the bottle necks are elsewhere...
That being said after years of buying AMD, my next CPU looks like a Conrore
That being said after years of buying AMD, my next CPU looks like a Conrore
Are there going to be pricecuts on 939pin dual cores? Or did I miss the memo about them being killed off?
A cheap 939 pin dual core (in the same pricing as those listed) would interest me more than the idea of a Conroe, even if the Conroes are lookin' pretty sexy at the moment. I don't particularly like the idea of having to upgrade ram and motherboard, even if the CPUs are priced rather excellently for their performance.
A cheap 939 pin dual core (in the same pricing as those listed) would interest me more than the idea of a Conroe, even if the Conroes are lookin' pretty sexy at the moment. I don't particularly like the idea of having to upgrade ram and motherboard, even if the CPUs are priced rather excellently for their performance.
L233 said:
True, their review is weird to say the least. I have no problem with them getting the point across that CPU isn't the bottleneck in high-res gaming for those less informed, but their manage to totally neglect how impressive a new CPU the Conroe is in the proces. Whether that effect was concious or not, I have no idea.
Fortuneately there are much better reviews from SimHQ, Anandtech and TechReport. I'm very happy with my 19" Eizo LCD and are thus gaming at 1280x1024, so I actually needed to see some benchmarks below 1600x1200 - believe it or not.
Don't you want to know which CPU is faster? If you don't, fine, drift along, but that's definitely not the point here.Skrying said:
The truth is that nobody cares how game X performs at setting B. These are all just representative workloads. The interesting thing is CPU performance, nothing else, and game benchmarks are useful for finding that out, but as a means, not as an end.
Intel fanpersons have used the exact same argument for justifying their P4 purchases. It didn't make sense back then and it still doesn't make sense today IMO.
hoom
Veteran
I gotta say paint me impressed as heck
I really hadn't expected Conroe to be as good as its showing (in reviews other than [h]), NZ$600 2.4ghz conroe for the most part matching or significantly bettering NZ$2000 2.8ghz FX62.
I mean, sure it doesn't make that much difference where you are GPU limited but simple fact is that faster performing CPU for cheaper = better & I've been waiting for a long time for a real big performance/$ improvement so I'm thrilled.
I never thought Intel had it in them...
I really hadn't expected Conroe to be as good as its showing (in reviews other than [h]), NZ$600 2.4ghz conroe for the most part matching or significantly bettering NZ$2000 2.8ghz FX62.
I mean, sure it doesn't make that much difference where you are GPU limited but simple fact is that faster performing CPU for cheaper = better & I've been waiting for a long time for a real big performance/$ improvement so I'm thrilled.
I never thought Intel had it in them...
geo said:Wheee. [H] is getting more aggressive in their language on pushing their benchmarking methodology as the One True Faith. "Lie" is getting tossed around there pretty liberally about others.
Wouldn't you want to see multi-gpu results before you came to such conclusions as they come to tho?
The AMD fanboi brigade has their
It's a shame that [H] didnt do things properly. They should have offered a wide variety of benchmarks, such as X1600, X1900 GT, X1900 XT, X1900 CF, 7600, 7900 GTX, 7950, 7900 SLI. Not having run the tests in non GPU-limited circumstances was doing a dis-service to their readers.
BTW: The "
" brings entire new meaning once you read this.You are welcome. :smile:StellaArtois said:
Here are some more:
http://www.extremetech.com/article2/0,1697,1988744,00.asp
http://www.tomshardware.com/2006/07/14/core2_duo_knocks_out_athlon_64/
http://www.trustedreviews.com/article.aspx?art=3161
http://www.behardware.com/articles/623-1/intel-core-2-duo-test.html
http://www.digit-life.com/articles2/cpu/intel-core2-duo-e6600.html
I'm still liking Anand's review the best so far. Its the little things as the 2MB v 4MB L2 Cache comparisons that are quite valuable, atleast to me. Reading the review it looks like Anand had a little extra time over other reviewers for those numbers, maybe he got a headstart with an early press-kit.
I'm late to the party, but what the hell.
This pretty much sums up [H]'s gaming benchmark segment:
This is the angle [H] is going for, and obviously has in the past. You see dear [H] clergy members, other sites aren't actually reviewing the processor, they're spreading hype - while the vanguards of decency and truth here at [H]ardocp will show you the light. Sensationalism that's designed to point out the sensationalism of other sites (that actually test the product in question). Rolleyes doesn't cut it here, I need a barf smiley.
Of course there are benefits to [H]'s approach, manual run-throughs instead of canned demos certainly provide more data than endless Quake4 timedemos a lot of sites engage in whenever a new GPU or CPU hits the streets. So I do appreciate they're doing something different, the problem here is that it's completely unsuited to testing a CPU.
Actually, even outside of CPU testing the problem with [H]'s method as well is that it places so much control in the hands of a reviewer. 40fps average for Oblivion is the target, because....uh, why? There are eye-candy whores and there are also framerate whores. With other sites, I can determine at what settings I'll need to run at to get the framerates I want, I can then make the judgment based on my own personal preferences on whether it's worth the cash outlay. What if a 1152 x 864 setting gives me 50fps? How much FPS will I lose if I go for 4X AA instead of 2X? Is the game CPU-bottlenecked that it can never equate to the refresh rate of my LCD? I'll never know by a Hardocp review.
It's perfectly fine to illustrate that you don't need top-end components to produce a good gaming rig and diminishing returns start to kick in when you move up the $$ ladder, that's definitely somethig I think the general public should be educated on. I fear people think that you absolutely need the $1000 processor and $500 GPU to get anywhere with modern PC action games, evidence pointing to the contrary is good the industry as a whole me thinks.
But this wasn't a "What difference does a CPU make to a high-res modern game?" article - it was supposed to be a review of a frickin' CPU. With Hardocp's logic, the second ATI/Nvidia release their new GPU's and the bottleneck is somewhat alleviated, the Core 2 becomes a better processor due to another product entitely out of the hands of Intel? That's simply idiotic. Heck, Hardocp could just start benching the new GPU's at higher resolutions, after all 1280*1024 is not really the intended target of these new GPU's, that's "last years" res. So nothing changes then as the bottleck will once again be the GPU.
Games this year and next will need more CPU. They will also need more GPU power, HD space, networking bandwidth. Simply because some components of the PC haven't caught up with Intels CPU does not negate the CPU from being an excellent product, you always should look at the price/performance scale compared to the same products in its class when doing a review about a specific piece of hardware. Right now, from the benchmarks most other sites give, it's apparent that a $300 CPU from Intel beats a $900 CPU from AMD. That's better performance for a 1/3rd of the price, which is stunning, and it brings benefits to all - even those not on the high end - because AMD has to drop prices in return.
Somehow I don't think the market would accept [H]'s argument if they came from AMD:"Look, there's no point for us to drop prices, as when you play games on a 30" LCD there's no advantage to Intel!". They're drastically cutting prices because they have to - there's a much better alternative on the market right now. AMD's price cuts are the best review Intel could get.
Edit: Hell, I was being kind to AMD's current fX62 price - it's more along the lines of $1100, not $900.
This pretty much sums up [H]'s gaming benchmark segment:
This is the angle [H] is going for, and obviously has in the past. You see dear [H] clergy members, other sites aren't actually reviewing the processor, they're spreading hype - while the vanguards of decency and truth here at [H]ardocp will show you the light. Sensationalism that's designed to point out the sensationalism of other sites (that actually test the product in question). Rolleyes doesn't cut it here, I need a barf smiley.
Of course there are benefits to [H]'s approach, manual run-throughs instead of canned demos certainly provide more data than endless Quake4 timedemos a lot of sites engage in whenever a new GPU or CPU hits the streets. So I do appreciate they're doing something different, the problem here is that it's completely unsuited to testing a CPU.
Actually, even outside of CPU testing the problem with [H]'s method as well is that it places so much control in the hands of a reviewer. 40fps average for Oblivion is the target, because....uh, why? There are eye-candy whores and there are also framerate whores. With other sites, I can determine at what settings I'll need to run at to get the framerates I want, I can then make the judgment based on my own personal preferences on whether it's worth the cash outlay. What if a 1152 x 864 setting gives me 50fps? How much FPS will I lose if I go for 4X AA instead of 2X? Is the game CPU-bottlenecked that it can never equate to the refresh rate of my LCD? I'll never know by a Hardocp review.
It's perfectly fine to illustrate that you don't need top-end components to produce a good gaming rig and diminishing returns start to kick in when you move up the $$ ladder, that's definitely somethig I think the general public should be educated on. I fear people think that you absolutely need the $1000 processor and $500 GPU to get anywhere with modern PC action games, evidence pointing to the contrary is good the industry as a whole me thinks.
But this wasn't a "What difference does a CPU make to a high-res modern game?" article - it was supposed to be a review of a frickin' CPU. With Hardocp's logic, the second ATI/Nvidia release their new GPU's and the bottleneck is somewhat alleviated, the Core 2 becomes a better processor due to another product entitely out of the hands of Intel? That's simply idiotic. Heck, Hardocp could just start benching the new GPU's at higher resolutions, after all 1280*1024 is not really the intended target of these new GPU's, that's "last years" res. So nothing changes then as the bottleck will once again be the GPU.
Games this year and next will need more CPU. They will also need more GPU power, HD space, networking bandwidth. Simply because some components of the PC haven't caught up with Intels CPU does not negate the CPU from being an excellent product, you always should look at the price/performance scale compared to the same products in its class when doing a review about a specific piece of hardware. Right now, from the benchmarks most other sites give, it's apparent that a $300 CPU from Intel beats a $900 CPU from AMD. That's better performance for a 1/3rd of the price, which is stunning, and it brings benefits to all - even those not on the high end - because AMD has to drop prices in return.
Somehow I don't think the market would accept [H]'s argument if they came from AMD:"Look, there's no point for us to drop prices, as when you play games on a 30" LCD there's no advantage to Intel!". They're drastically cutting prices because they have to - there's a much better alternative on the market right now. AMD's price cuts are the best review Intel could get.
Edit: Hell, I was being kind to AMD's current fX62 price - it's more along the lines of $1100, not $900.
Last edited by a moderator:
I have no problem with [H] approach as part of the mosaic. I'll go farther and say I see value in it, as part of a larger mosaic of reviews. I don't even have a problem with them explaining why in their minds it is superior and to be preferred to competing approaches. Everyone has that right, even that responsiblity, to explain why they make the choices they do. But they've crossed over into "One True Faith" territory, with everyone else going to hell, and I don't like that one little bit.
Exactly. A post from Kyle himself illustrates this attitude perfectly:geo said:
Bloody nauseating. Conrone does give you gaming benefits, and it was not "hyped" by Intel in terms of any dishonesty with regards to performance - Intel simply had testers on systems with SLI/Crossfire to remove the GPU as the bottleneck, which frankly most expected as it would be pretty stupid to demonstate any superiority of your hardware when another manufacturers hardware that you have no control over is holding yours back. Actual benchmarks that test the speed of Conroe with crossfire (or at resolutions low enough where the GPU is not a bottleneck) back up Intel's early gaming performance claims.
Luckily I turned off the irony-meter before reading that "agenda" comment. The attitude displayed by Kyle is incredibly sensationistic and needlesly defamatory to other sites (aside from the need of getting page hits - advertisers don't care if people are visiting Hardocp to praise or flame, they just care about clicks), while claiming that every other review is doing exactly what he is. Absolutely fucking shameless.
Interestingly enough, Hardocp's video card reviewer stated that they would have included a 7950GX2, but they couldn't get it working on their motherboard. Fair enough, but if they were planning to use a more powerful video card and couldn't do it, why not try and demonstrate what a more poweful GPU would do your benchmarks by alleviating the GPU bottleneck some what? You know, by uh...LOWERING RESOLUTION? Which btw, is exactly what Hardocp did in their FX62 review. So is Hardocp saying they have an "agenda" back then, and now they're the one unbiased source?
Best sum-up of the review on Hardocp's forums:
Last edited by a moderator:
I can't wait until, I don't know, car magazines adopt this perfectly logical and relevant testing mythology. Can't you see it already?
"Don't let other publications lie to you - only we give you the real story! In this much anticipated showdown of Bugatti Veyron with 1996 Ford Taurus SE, we used only the real-world conditions - a 20 mile stretch of 65 mph highway with moderate traffic - and of course we obeyed the speed limit. The results were shocking!
Average speed, 20 miles
Veyron: 64.1 mph
Taurus 62.3 mph.
You see it right. Under the real-world conditions, the ~$1 million Veyron is only 3% more powerful then $1200 Taurus. Don't let the Bugatti cronies pull the wool over your eyes - there is no evidence that it better then Ford by any objective measure"
It is also worth noting that during their last Pre-Conroe CPU article, AM2 review, HardOCP made a heavy use of low-resolution testing in the gaming section of the article. As they say, ladies and gentlemen, draw your own conclusions….
"Don't let other publications lie to you - only we give you the real story! In this much anticipated showdown of Bugatti Veyron with 1996 Ford Taurus SE, we used only the real-world conditions - a 20 mile stretch of 65 mph highway with moderate traffic - and of course we obeyed the speed limit. The results were shocking!
Average speed, 20 miles
Veyron: 64.1 mph
Taurus 62.3 mph.
You see it right. Under the real-world conditions, the ~$1 million Veyron is only 3% more powerful then $1200 Taurus. Don't let the Bugatti cronies pull the wool over your eyes - there is no evidence that it better then Ford by any objective measure"
It is also worth noting that during their last Pre-Conroe CPU article, AM2 review, HardOCP made a heavy use of low-resolution testing in the gaming section of the article. As they say, ladies and gentlemen, draw your own conclusions….
Similar threads
