Intel Core 2 Gaming Performance

Wow, such a major disappointment after all of those pre-launch benchmarks. Guess it goes to show...
 
shungokusatsu said:
That's what I'm saying, but there's about 5 reviews showing the same thing, and what confuses me is how they drastically differ from the [H] review. Perhaps it's because [H] didn't use SLI or better mboard. But I'm seeing more and more 100%+ performance increases on these other benchmark reviews. If this turns out to be true, I'm sold on Conroe without doubt, kinda sucks, I should have waited before building this AMD rig.
Click to expand...

those benchmarks show very little difference at low resolutions but huge gains at high resolutions. that goes against everything we know about cpu benchmarks. i got so sick of people linking to that with no explanation.
 
Rob94hawk said:
As what I expected. Intel doesn't blow AMD away, they just finally catch up.
Click to expand...

I think you'll see when newer GPUs are released that Intel just blew AMD out of the water. That doesn't even cover the price difference nor application performance (which isn't even close).

I'm not a fan boy, I don't care about either brand. From all that I've seen and read, it's clear Intel has a real winner on their hands and is selling them at a very reasonable price.
 
Brent_Justice said:
and how is running at a low resolution going to show you that?

if anything you should be using a high resolution and the latest games, I think Oblivion was a great test, it showed the most difference between the CPUs and it just might be an indicator of the future
Click to expand...
that ends up being more of a gpu test than cpu. i know you know that ;)
 
EternityZX9 said:
Wow, such a major disappointment after all of those pre-launch benchmarks. Guess it goes to show...
Click to expand...


Please post chipset, video card, and cpu of pre-launch benchmarks. Thanks!
 
Joshua_564556 said:
There is no respect for this garbage.

Josh
Click to expand...

Thanks for your thoughts. We will link all the canned benchmarks tomorrow for all the people that need the real story about their gaming experience. :rolleyes:
 
shungokusatsu said:
Let me find them:

http://www17.tomshardware.com/2006/06/05/first_benchmarks_conroe_vs_fx-62/

http://forums.guru3d.com/showthread.php?t=185555

http://www.hothardware.com/viewarticle.aspx?articleid=794&cid=1

I'm not saying these are legit, I'm just totally confused why they are showing such drastically higher scores for Intel.
Click to expand...

Those are far from 100% as you stated before and they are all done on machiens and demos supplied by Intel. As for that forum post, well, that is just too much nonsense to try and make sense of.
 
JetUsafMech said:
All on 975x.
Click to expand...

Do you honestly believe that a CHIPSET is going to give that drastic of an improvement? I dont care if its a mid grade vs an enthusiast chipset... its not going to give results like that. Go find me an A64 and compare it on a NF4 board to some shitter 939 SIS board. I bet if all that changes is the board and chipset, the results will still be damn close.

If you really think these benches are bs based soley on a chipset... your only kidding yourself. If you think there BS because they aren't a crossfire/sli setup like those, then your missing the point of a real world benchmark. A very small fraction of us are using x1900xtx CF or 7900GTX SLI. I'm not, so I don't care how that does. I have a high end single card AMD setup. I wanted to know if switching to conroe would effect my gaming experiance. This review shows it won't. It serverd its purpose.
 
shungokusatsu said:
Those machines are not supplied by Intel from what I read, only the CPU. Could the reason the [H] reviews are so low really be because of you guys not using 975X board? I find that highly doubtful.
Click to expand...


Thanks for your thoughts. I would suggest at this point that go to their boards or start your own thread here on topic and inquire aobut their results as we will not longer discuss them in this thread.

And yes, the THG and HH machines were supplied by Intel with Intel installed OS.
 
Kyle's approach is exactly what I have been looking for.

I own a Dell 2405 like many here on this board. So the naitive resolution that I will be playing at 1920x1200 with Conroe does not make one lick of difference.

So why on Earth should I go out and buy a new processor, new mobo and new ram for the exact same game performance that I am getting now? Especially with K8L coming around the bend? (and yes the AM2 version is coming around the bend and not in 2008 :rolleyes: )

Thanks Kyle for saving me a shit load of money.
 
So let me get this straight.

All these tests are so similar beacsue the CPUs are too powerful for the 7900GTX, and the 7900GTX is perfroming at its absolute best? What would this mean for my X1900XT and PD930 OCed at 4GHz?
 
Brent_Justice said:
For those complaining about the gameplay evaluation tell me, will you be buying these high-end CPUs with low-end video cards and playing at low resolutions for gaming?

This evaluation tells you exactly what you need to know when comparing gaming performance between the platforms. It tells you exactly what the differences are when playing the games as me, you and everyone else playing games on such a system.
Click to expand...

I agree with you totally Brent but here is the problem. When you benchmark the games at these higher resolutions where you are more GPU bound and there is little difference between the processors it does show what real world performance is like for the average user with that setup at that resolution TODAY. It however does NOT give him a feel for how much faster processor A is compared to processor B when a faster video card comes out. When you benchmark at lower resolutions like 800x600 and 640x480 is allows the CPU's to shine more and show a difference between each other. No, people dont play at those resolutions but it shows which processor is faster in that game and by how much. Today the E6700 may be around the same speed as the FX-62 at those resolutions you benched with a 7900 GTX but what happens to the person with a SLI/Crossfire setup or the guy that upgrades to a G80/R600 down the road but keeps the same processor?

|CR|Constantine said:
Thanks for the review Kyle and I will be sticking with my current AMD setup.

As for all you Intel yahoo's congrats on finally catching up to AMD, I guess the only problem is ...AMD K8L response is still to come..oh so close yet so far.

:eek:
Click to expand...

Caught up and blew right by. I dont know if you've looked at any benchmarks outside of just games but Conroe has a big advantage in a lot of areas. The Pentium 4/D processors can keep up with the current AMD processors at those resolutions benched here.

Chris_B said:
Near sure i read somewhere it got pushed back to 2008.
Click to expand...

Yeah i saw that article as well. AMD slides show K8L in 2007 but a few reports are saying 2008 and only 65nm K8 processors for 2007.

EternityZX9 said:
Wow, such a major disappointment after all of those pre-launch benchmarks. Guess it goes to show...
Click to expand...

Doesn't go to show anything. Just need to learn how to interpret a benchmark :). These same results can be said for the Penium 4/D vs the Athlon 64.

|CR|Constantine said:
Kyle's approach is exactly what I have been looking for.

I own a Dell 2405 like many here on this board. So the naitive resolution that I will be playing at 1920x1200 with Conroe does not make one lick of difference.

So why on Earth should I go out and buy a new processor, new mobo and new ram for the exact same game performance that I am getting now? Especially with K8L coming around the bend? (and yes the AM2 version is coming around the bend and not in 2008 :rolleyes: )

Thanks Kyle for saving me a shit load of money.
Click to expand...

Maybe because you might have a little more powerful GPU setup driving that 2405fpw like a SLI/Crossfire setup? A single 7900 GTX doesn't fair that well in heavy GPU games like Oblivion at 1920x1200.
 
|CR|Constantine said:
So why on Earth should I go out and buy a new processor, new mobo and new ram for the exact same game performance that I am getting now? Especially with K8L coming around the bend? (and yes the AM2 version is coming around the bend and not in 2008 :rolleyes: )
Click to expand...

You make it sound like K8L will make any difference in gaming at such high resolutions, silly !!!!!!.
 
shungokusatsu said:
Those machines are not supplied by Intel from what I read, only the CPU. Could the reason the [H] reviews are so low really be because of you guys not using 975X board? I find that highly doubtful.
Click to expand...

Plus Tech Report set up theirs, so did Digi Times via one of the XBits guys, Even Anand had got his hand on some. The links are in this forum. IMHO, E6700 stole the show even if it were the exact same speed as FX-62 since it is half its price, uses less power (if features are turned on) and etc.. Once you overclock it, the value jumps off the chart, even with a weak overclock to 3GHz it becomes an even better deal.
 
Joobot said:
So let me get this straight.

All these tests are so similar beacsue the CPUs are too powerful for the 7900GTX, and the 7900GTX is perfroming at its absolute best? What would this mean for my X1900XT and PD930 OCed at 4GHz?
Click to expand...
You might get faster framerates than [H] gets with the faster CPUs if the X1900XT runs the game faster than a 7900GTX.
 
Joobot said:
So let me get this straight.

All these tests are so similar beacsue the CPUs are too powerful for the 7900GTX, and the 7900GTX is perfroming at its absolute best? What would this mean for my X1900XT and PD930 OCed at 4GHz?
Click to expand...


You would probably be interested in this article that we published months ago, but got buried and ignored by almost every hardware site on the planet. It basically showed that all the hype about Intel totally owning AMD were all bullshit.

http://enthusiast.hardocp.com/article.html?art=MTAwMiwsLGhlbnRodXNpYXN0

New month, same story. High end gaming is no longer CPU dependent and the fastest CPU in the world will not change that. That is exactly what we showed. Whodathunkit?
 
pxc said:
You might get faster framerates than [H] gets with the faster CPUs if the X1900XT runs the game faster than a 7900GTX.
Click to expand...

The X1900 series pretty much own the 7900 series in Oblivion so yeah i would say he would have a good chance at pulling better fps with his 930 @ 4GHz then the FX-62/6700/X6800 results in Oblivion here.
 
|CR|Constantine said:
Kyle's approach is exactly what I have been looking for.

I own a Dell 2405 like many here on this board. So the naitive resolution that I will be playing at 1920x1200 with Conroe does not make one lick of difference.

So why on Earth should I go out and buy a new processor, new mobo and new ram for the exact same game performance that I am getting now? Especially with K8L coming around the bend? (and yes the AM2 version is coming around the bend and not in 2008 :rolleyes: )

Thanks Kyle for saving me a shit load of money.
Click to expand...


But for someone looking to upgrade soon (such as myself), you don't only want to know what your cpu is capable of today. Since the vid card is, admittedly, the bottleneck in today's games... it will most likely be the bottleneck in tomorrow's games also. So, the next logical step for an upgrade would be a new video card (not to mention, upgrading the CPU could mean a new mobo and ram also). Then, your cpu could be the bottleneck... so don't you want to know which CPU has the most raw power (thus giving you better performance when it IS the bottleneck) so you can buy the one that'll last the longest??

I DO appreciate what [H] does though... everywhere will have the other types of benchmarks so it's still interesting to see these numbers as well. Just trying to show why some people care more about the other numbers. :) :)


Jason (hope that all made sense)
 
It seems like people are intent on assigning the [H] crew some nefarious motive for the way they did the testing, and I don't understand it. They seem to have proved over and over that they're gamers who like the hardware that gives the best gameplay experience. I certainly didn't read anything in the article or its conclusion that told people the new Intel chips aren't worth the money or that they aren't potentially faster in non-gaming apps. All they did was show us gaming benchmarks taken at the kinds of resolutions that people with top-flight gaming systems play at all the time.

For the heck of it, here's how I would interpret the results I've seen so far (here and other places) - all depending on upgrade timing, of course:

If you're a gamer with a socket 939 or AM2 system, the jump to Intel probably isn't necessary (and, it should be noted, that the evaluation of 939 versus AM2 on this site made clear that the difference was marginal at best).

If you're a gamer with an older system looking to do a complete system overhaul, Core 2 would probably be the way to go if you're shooting for "top of the line."

If you're a gamer with a socket 939 or AM2 who also does a lot of CPU-intensive non-gaming stuff, you might want to look at a Core 2 for your next big upgrade.

If you're a gamer with an older system who also does a lot of CPU-intensive non-gaming stuff, the Core 2 is probably your best bet.

Of course, all of the above is meaningless if all you want is for someone to tell you which chip is "#1," but that's never been the style I've seen at the [H]...
 
Bottom line is this. [H] is compairing a mid-level 965 chipset to a flagship 590MCP, and not one sentence in the article explaines this. A Intel 975X chipset needs to be used to be fair. Yes indeed the AMD FX-62 system loses in every test by 1FPS to the X6800 XE. The reason the difference is not 15+FPS is because 965 is no where near as good as 975x.

As Kyle stated in this thread, a 975X board was not available, so what was available is what was used(mid-range 965). To a person who does not know that 975x exist, these benchmarks are VERY mis-leading.

I hope down the line [H] will write a review of X6800 XE on 975x, and put out a balanced article.
 
cr0w said:
You make it sound like K8L will make any difference in gaming at such high resolutions, silly !!!!!!.
Click to expand...

Your exactly right odds are it wont make a big dent either. But for people like me who have gaming as primary function for their rig and other apps second K8L will bridge the gap or surpass them.

So for people like me playing at high resolutions (even though I believe 1600x1200 should be the new standard) it makes absolutely no sense in upgrading or jumping ship for people like me with high end AMD solutions that cater only to gaming.
 
RangerXLT8 said:
Bottom line is this. [H] is compairing a mid-level 965 chipset to a flagship 590MCP, and not one sentence in the article explaines this. A Intel 975X chipset needs to be used to be fair. Yes indeed the AMD FX-62 system loses in every test by 1FPS to the X6800 XE. The reason the difference is not 15+FPS is because 965 is no where near as good as 975x.
Click to expand...

It has little to nothing to do with the chipset. It has a lot to everything to do with the fact the games were GPU limited. The 7900 GTX couldn't pull high enough frames at those higher resolutions to descern a real difference in performance. If you would of thrown in a high-end Pentium 4/D you would of seen the same kind of results you see with the FX-62/E6700 here. Just look at the other sites that have benchmarks at 800x600. No that isn't a resolution that most people game at but its a resolution that takes the video card out of the equation and shows you which processor is faster and by what factor. This will come into play in the real world when you start gaming with SLI/Crossfire or a new G80/R600 video card down the road. Graphics cards today just aren't fast enough to keep up with Conroe at high resolutions because the GPU has to do a lot more work in a game then the CPU does.
 
visaris said:
You people have no respect. You should be thanking Kyle for the hard work and contribution to us all instead of your current pathetic display.
Click to expand...

Dude, the review is flawed. Does he owe it to people to fix it? No. Does it seem like he even cares? No. Are people getting accurate results as to the performance of the new Intel chip? No. I'm an AMD guy, but, really, I would not even do this to Intel.
 
Babbster said:
It seems like people are intent on assigning the [H] crew some nefarious motive for the way they did the testing, and I don't understand it. They seem to have proved over and over that they're gamers who like the hardware that gives the best gameplay experience. I certainly didn't read anything in the article or its conclusion that told people the new Intel chips aren't worth the money or that they aren't potentially faster in non-gaming apps. All they did was show us gaming benchmarks taken at the kinds of resolutions that people with top-flight gaming systems play at all the time.

For the heck of it, here's how I would interpret the results I've seen so far (here and other places) - all depending on upgrade timing, of course:

If you're a gamer with a socket 939 or AM2 system, the jump to Intel probably isn't necessary (and, it should be noted, that the evaluation of 939 versus AM2 on this site made clear that the difference was marginal at best).

If you're a gamer with an older system looking to do a complete system overhaul, Core 2 would probably be the way to go if you're shooting for "top of the line."

If you're a gamer with a socket 939 or AM2 who also does a lot of CPU-intensive non-gaming stuff, you might want to look at a Core 2 for your next big upgrade.

If you're a gamer with an older system who also does a lot of CPU-intensive non-gaming stuff, the Core 2 is probably your best bet.

Of course, all of the above is meaningless if all you want is for someone to tell you which chip is "#1," but that's never been the style I've seen at the [H]...
Click to expand...


Look we're not stupid, i don't need to have someone lay out for me which one is going to give the best "experience", "experience" sounds like marketing; aka bullshit. I've been doing this long enough to know that when i'm given numbers for parts a,b,c....x,y,z, that its going to mean a whole alphabet when put together.

all that i need to be told is that CPU a is 1.5 times better than CPU b, and then i'll know that when combined GPU 1 or 2, i'm going to get certain results. I and most people on this board, don't need it dumbed down.

I do not believe that the [H] has any motive other than to do the best job that they think they can. I think just about every other review that the [H] has done i have greatly appreciated so i really have ZERO intention of diminishing their hard work.

But It's frustrating to read a review that dosen't really tell me anything other than the canned benches are bullshit. wow, a company fudging marketing materials whoda thunkit? :rolleyes:
 
Will be interesting to see what the other people's reviews say:

found this 1 so far:

Performance on the X6800 and E6700 Core 2 Duo products was stellar. In nearly every single benchmark we saw here in our review at PC Perspective, both the X6800 and the E6700 came out ahead of the Athlon FX-62 processor. While the $999 X6800 CPU beating the $1000 FX-62 CPU isn't a big story on its own, the $530 E6700 processor beating it is. The X2 4800+ CPU obviously couldn't keep up with the E6700 either, and that is the most price competitive part AMD has for it. AMD is going to be in more than a bit of trouble come the end of July if they don't have an answer to Intel's Core 2 Duo product line. The E6700 sample we tested with here was able to trounce the FX-62 in many cases, and came out the leader in nearly every test we threw at it.

They benchmarked w/ 3DMark 06, and other stuff that realworld playing situations do , At least I do - 1280x1024 , and the other gaming tests were at 1024x768 to push the CPU they said, so it wasnt a bottlenecked GPU situation.

As for my personal opinion - its interesting to see what others find - I have a x2 4400 right now, and a 7800 GTX 256, I see alot of gain from the e6600/6700 , w/ a 7950. I just need to find them in stock (the processor) to get things going, lol. Plus overclocking at 3.5-4ghz+ makes me wonder what the benchmarks would be w/ 3dmark06 - since that gives you a better idea of what your getting out of everything (gpu, cpu, etc). I dont think the Fx62 overclocks that well, and stable - and definitely not on purely air cooling.
 
visaris said:
You people have no respect. You should be thanking Kyle for the hard work and contribution to us all instead of your current pathetic display.
Click to expand...

Kyle makes his living running this site. For all intensive purposes, [H] is a business and should be treated as such. Kyle should be held up to professional standards and expectations from consumers, including legitimate criticisms. Kyle is not running this site out of the pure goodness of his heart, he's doing it because he's very passionate about it and he can make a buck doing it.
 
Joshua_564556 said:
The fact is you are leading your readers astray --- you should also put along side this BS review an 805D performance at the same resolutions/settings. You showed nothing more than a GPU bottlenecking every game -- as you allude to but do not fully explain.

Users who purchase LCDs and are unaware of the biased nature of this so called review of a CPU will think wow 1600x1200 when they are actually natively getting 1280x1080. Some, HARDCORE gamers will actually run down rez in order to appropriately frag your bud.

This review will ultimately look like trash when all other data confutes what you present here in 'hidden' light. What happens when someone rig's up with an AM2 FX-62 with your setup, then when the next gen cards come out -- find that they are indeed throttled by the CPU where as this would not have otherwise been the case.

There is no respect for this garbage.

Josh
Click to expand...


He does fully explain. You just did read the darn review. You're the type who'll call any review BS if it doesn't say exactly what you want to to say - in this case that we should all go out and buy a Conroe the second it comes out. I find it a pathetic that you spend your time trolling forums flaming people when they don't hype your favorite company enough.
 
That was a nice GPU review. I love how it brought all the kooks out saying it proved Intel somehow "rigged" their "fake" results.

It's easy to defend a pointless review by saying that you were showing that a GPU is your bottleneck for most modern games at a high resolution. What does this have to do with a CPU, again?

The simple fact is that people upgrade GPU's more than they upgrade CPU's When you upgrade your 7900GTX to a new vard next year, suddenly your CPU becomes the bottleneck on current games.

It's interesting, by the way, that no high end SLI systems were tested. Would have been interesting to see some e.g. Counter Strike: Source benchmarks. Of course, obviously that wouldn't have supported the obvious attempt to draw viewership with "controversial" results.

In the end, I didn't find the results controversial at all. There are high end video cards for a reason - if you didn't know that they are the bottleneck on latest-generation games then maybe this was a revelation for you.
 
shungokusatsu said:
Let me find them:

http://www17.tomshardware.com/2006/06/05/first_benchmarks_conroe_vs_fx-62/

http://forums.guru3d.com/showthread.php?t=185555

http://www.hothardware.com/viewarticle.aspx?articleid=794&cid=1

I'm not saying these are legit, I'm just totally confused why they are showing such drastically higher scores for Intel.
Click to expand...
Remember the other reviews had more powerful video subsystems, so they aren't GPU bottlenecked like this review is at 16x12 High Quality with a Single 7900 GTX

Either that or they used lover quality settings then what HardOCP is using hence the larger difference as the load shifts toward the CPU.
 
CodeWaste said:
Look we're not stupid, i don't need to have someone lay out for me which one is going to give the best "experience", "experience" sounds like marketing; aka bullshit. I've been doing this long enough to know that when i'm given numbers for parts a,b,c....x,y,z, that its going to mean a whole alphabet when put together.

all that i need to be told is that CPU a is 1.5 times better than CPU b, and then i'll know that when combined GPU 1 or 2, i'm going to get certain results. I and most people on this board, don't need it dumbed down.
Click to expand...

Do you really think that doing a set of benchmarks at 640x480 or 800x600, or changing the chipset, will really quantify which processor is [x] percent better than the other?

You may dislike the word "experience," but it's supposed to be what a "gamer" is looking for when playing a game. If two different processors provide the same gameplay results as one another in a standard situation (situation meaning resolution and detail level), isn't that good information to have? At least as important as which one pumps out the most fps in artifical, low-resolution tests?

When the [H] changed their benchmarking criteria oh those many moons ago, I was one of those who wondered exactly what the point of it was. Having experienced the way they do things ever since, I've learned that they're looking for exactly what I would look for in their hardware: A good gaming "experience." It makes perfect sense to me, but I suppose it's obvious that YMMV.
 
The review showed us the results of his testing.

His question under test was "How will these CPU's affect gameplay with current games at popular resolutions". The answer was "not much".

Any other conclusions drawn from this data are pure extrapolations (i.e. "C2D barely caught up with AMD"). With these kinds of games you'd expect to be limited by the video subsystem, so the results aren't surprising.


And thank you Kyle, even though the results are in-line with what I'd expect, it's still good to see that someone took the time to run the numbers. Sure, 640x480 tests would show off the raw CPU power of the chips, but you're correct, nobody actually uses that resolution.
 
Guys, take a look at all the reviews you are posting. The results at similar resolutions being shown here are basically the same. I dunno about you but 100pts in 3dMark06 is pretty much a wash to me. The fact is that if you are looking for straight gaming performance then, a cheap AMD or Intel is a great solution. If you have a decent AMD system already, there is no reason to run out and grab Intel.

Is the Core 2 DUO faster.... YES! The [H] review does not show anything skewed at all in their performance indication on the gaming side as it shouldn't. I am sure that if they were running SLI there would be slight improvements but then Kyle did mention that there were reasons they did not do so.

This makes it pretty clear what is going on here.. Intel finally caught up to AMD in gaming performance. Its about fucking time now ! Do they want a cracker or something? :p
 
You must log in or register to reply here.
Back
Top