Hardocp core 2 duo article

[Stelly]

well i think its better going for conroe regardless, but that's besides the point lol. All i want is for somebody to review the cpu's using the same settings thruout and lay it all on the table for us to see. Whats the point in claiming they're going to do it fairly if they then review like that? there's no need for skewing the benchmarks.

I'm in total agreence with you mate

Stelly

 

[Stelly]

Hi Guys... I know that this is a little off topic but can anyone tell me what game is being advertised with the product at the bottom of this page??

http://www.hothardware.com/viewarticle.aspx?page=10&articleid=845&cid=1

Stelly

 

[fish99]

Battlefield 2142 I think.

 

[drunkenmaster]

And also half the price. /thread

Keep in mind we have already seen lots of independant benchmarks of games at resolutions of 1600x1200, this includes games like F.E.A.R and Oblivion, and minimum framerate was concluded to be double in some cases with the Conroes.

please stop quoting that, its one unofficial forum stuff from one guy which nothing else, now 20 odd reviews, 5 odd previews or any other forum people have backed up.

the [H] review shows oblivion to have almost identical max, min and average framerates on the two different cpu's, there has been nothing, absolutely nothing to show that that guys results he posted were true, every review site, from respectable to shady show the same numbers that utterly trash those "double minimum fps" claims. please please realise those numbers were completely and utterly wrong/lies.


as you'll see from [h]'s review(first time i've seen a review point this out), the conroe is faster, without a doubt, no question, but for gaming its ALL gfx card, at 1600x1200 pretty much full quality if not full quality with a 7900gtx, max/min/ave are almost identical on a x6800, a e6700(266mhz slower and 1fps slower pretty much) and a fx62. also at 1280x1024 with NO aa/af the numbers are still almost identical, on a single 7900gtx you still see very little difference, even at your standand resolution for a 17/19" tft you still won't see a difference based on cpu as long as you get a decent-ish one (talking not a lowly clocked celeron or the worst semprons).

the point to notice is this, price of the fx 62 makes no difference, the 266Mhz drop from x6800 to E6700 made NO difference either, even a E6600 wouldn't have dropped much at all, likewise a 4400+ or something would have been marginally(less than 5%) slower than the fx62.

i challenge anyone to find numbers, forum users, reviews, previews, aliens that support those ridiculous oblivion/fear double fps numbers.

 

[drunkenmaster]

There is one factor thats not represented in any of the reviews that I have read so far, but is a key factor...

If comparing conroe to the AMD FX62, based on many results (admitidly using ES chips) conroe overclocks far better than the AMD.

Take the top of the range X6800. These seem to do between 3.5 and 4ghz on air where as an FX62 will just top 3.1ghz. It is rare to find anyone with a Conroe of any type that does not do at least 3.2Ghz.

When you take the better overclock into account the gap between Conroe and the AMD64 grows quite a bit wider and therefor those 4-6 frames per sec in games will increase quite a bit.

Just my thoughts

i wish, i mean i really do, but the problem is the fps gap is less than 4-6 fps, and they are gpu limited, more cpu power doesn't help render more frames. all overclocking the cpu will do is mean theirs more juice to be used for a background application. as the review shows the 266Mhz jump from E6700 to the x6800 gave about 1fps increase, so 3.8Ghz is likely to do nothing extra for gaming. the review also shows that, based on numbers theres no reason for a card like that to be playing as low as 1280x1024, as you just get fps in all games way way beyond the "playability" numbers, you're just sacrificing quality. 1600x1200 with lots of eye candy gives you very playable rates and the best eye candy the card can give, is utterly gpu limited and any cpu from a say 2.2Ghz up will give basically the same numbers.

 

[BubbySoup]

From Anandtech ;

The goal was to reach the highest possible speed that was benchmark stable. Super Pi, 3DMarks, and several game benchmarks were run to test stability. The 2.93GHz chip reached 4.0GHz on air cooling in these overclocking tests. That represents a 36% overclock on air with what will likely be the least overclockable Core 2 processor - the top line X6800.

To provide some idea of overclocking abilities with other Core 2 Duo processors, we ran quick tests with E6700 (2.67GHz), and E6600 (2.4GHz). The test E6700 reached a stable 3.4GHz at default voltage and topped out at 3.9GHz with the Tuniq Cooler. The 2.4GHz E6600 turned out to be quite an overclocker in our tests. Even though it was hard-locked at a 9 multiplier it reached an amazing 4GHz in the overclocking tests. That represents a 67% overclock.

 

[FrankJH]

undoubtedly the Core is a better cpu than anything AMD - but consider how much you would have to invest first - just to get level 5 shadows.......just cant see it myself

Admitedly there will be games out in the near future even more taxing than oblivion etc but even so I cant see it really being worth losing money on a decent s939 system to upgrade cpu , mobo and ram - and thats if you are happy with everything else

including vat you are talking nearly £600 for a half decent Core Duo system (for the three main parts) and thats before you even go to premium ram or cpu lol - infact according to reports that wouldnt even be Conroe would it - would be Allendale or whatevr its called

 

[Digital Punk]

Thought the Hard OCP gaming review was good.

In the real world tests there doesn't seem to be much gain, and just affirmed that if you already have a decent gaming rig, then switching to conroe for a few fps more might not be the best option.

 

[fish99]

Tests against some slower CPUs might have been more use to the average punter. Not everyone already has an FX-62.

From my point of view sitting here with an intel 875P board and Skt 478 P4 3.2 Prescott, thinking about whether to go A64, A64 X2 or Core 2 Duo for my next platform, this article tells me nothing. In fact this statement by HardOCP is down right misleading isn't it?

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 even prove that due to testing limitations we ran into. Then and only then, you might see an Intel Core 2 processor deliver a performance advantage.

All the other articles out today show Core 2 is significantly faster than A64, A64 X2, P4 etc, in games, and not just at low resolutions and low settings, even in 1600*1200.

I know that's not the whole picture because all the articles have used super fast video cards, which not everyone has (like me with an X1800 XT), but still, HardOCP would have us all believeing there's nothing to gain from Core 2.

Or am I missing something.

 

[drunkenmaster]

nope, the only review that shows gaming at 1600x1200 is at low quality and with crossfire, 1x x1900xt benchmarks at 1024x768 are useless and 2x x1900xt at 1600x1200 are useless, neither are remotely gpu limited. admittedly i haven't checked around much for other reviews, we already know what the chips are and the reviews show nothing new at all but i've only seen one review with gaming at a resolution and quality setting appropriate for the card(s) used and it specifically says the upgrade will give you nothing more than a couple fps and is not gonna help.

yes we don't all have an fx62, but the difference between the X6800 and the E6700, 266mhz apart, is non existant, dropping another 266Mhz won't show much either, dropping down to a 2.2Ghz ath 64 wouldn't make more than 3-4 fps difference either if that. the cpu doesn't really do any more work at all if you're gaming at 1600x1200 than at 1024x768, the physics, the placements of the characters its all the same. the cpu works that out and sends the data to the gpu to tell it where to draw what. only gpu load changes at diff resolutions, so if your game can give you 115fps as 1024x768 thats your same fps limit at 1600x1200, so if your pc is only throwing out 70fps its your gpu, no where near your cpu limit and a much slower cpu will still be fine for giving that 70fps anyway.

 

[Darkwave]

please stop quoting that, its one unofficial forum stuff from one guy which nothing else, now 20 odd reviews, 5 odd previews or any other forum people have backed up.

Did I or did I not say "independent review"?

Also, look here: http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2795&p=1

Plenty of Conroe mopping the FX-62 all around, even the E6300 is very close which is astonishing for the price.

 

[drunkenmaster]

Did I or did I not say "independent review"?

Also, look here: http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2795&p=1

Plenty of Conroe mopping the FX-62 all around, even the E6300 is very close which is astonishing for the price.

cpu limited, x1900xt's in crossfire, 1600x1200 with no aa/af, not high quality settings is no where near the realms of gpu limited, therefore cpu limited, therefore useless. still one review that i've found and one preview benchmarked 1600x1200, high quality with one gfx card and both showed within 2% fps for min, max and average. like i said, find a single review that shows otherwise, i can't.

http://www.xbitlabs.com/articles/cpu/display/core2duo-e6300_11.html

this shows, low res big difference, higher quality(doesn't say the res for some reason, 1.83Ghz stock cpu gets 150fps, increasing to 173 when overclocked by 1.1Ghz. just over a 10% improvement for over a 50% overclock, and sorry thats only at medium quality. high quality quake 4 showed a 15fps boost when at 1024x768. high quality and low res and a 10% boost for what a 55% ish overclock. if this doesn't show again that games are still 90% gpu limited i don't know what does.

hell, even in the anandtech review with crossfire at not that high a res at all a 530Mhz overclock gave them a sub 10% performance boost so even with crossfire its still not "that" cpu limited at all.

 

[Darkwave]

cpu limited, x1900xt's in crossfire, 1600x1200 with no aa/af, not high quality settings is no where near the realms of gpu limited, therefore cpu limited, therefore useless. still one review that i've found and one preview benchmarked 1600x1200, high quality with one gfx card and both showed within 2% fps for min, max and average. like i said, find a single review that shows otherwise, i can't.

LOL? It's a benchmark to show real-world gaming performance instead of synthetic benchmarks. Show me any CPU that doesn't bottleck Crossfire and I will give you a pound. If you had bothered to check page 17, you would have seen it has pure CPU-bounding tests with the Conroe pulling ahead by no less than 22.8% and up to 55% across the board.

Sheesh, you're in serious denial mate. I love AMD as much as anyone, but let's just face it, Conroe rocks and for the price it is a serious bloody (or should I say broken with the bridge smashing into the brain) nose for AMD.

Anandtech sum it up best on the last page with this quote; "Intel's Core 2 Extreme X6800 didn't lose a single benchmark in our comparison; not a single one."

 

[drunkenmaster]

LOL? It's a benchmark to show real-world gaming performance instead of synthetic benchmarks. Show me any CPU that doesn't bottleck Crossfire and I will give you a pound.

If you bothered to check page 17, by the way, it has a pure CPU-bounding test with the Conroe pulling ahead by no less than 22.8% and up to 55% across the board.

Sheesh, you're in serious denial mate. I love AMD as much as anyone, but let's just face it, Conroe rocks and for the price it is a serious bloody nose for AMD.

Anandtech sum it up best on the last page with this quote; "Intel's Core 2 Extreme X6800 didn't lose a single benchmark in our comparison; not a single one."

you seem to not be listening, i've said it in many threads, i'm for conroe, i'm getting conroe its faster simple fact. i'm getting it because i do a lot of encoding and graphical design work, for gaming it will not improve my experience one little bit.

a few people with crossfire here run in 1600x1200, a higher percentage use crossfire to run 1920x1200, or even higher, at that point its gpu limited again if you use high quality settings. thats the point. i've never once said its not faster, i'm trying to stop 100's of people thinking that just buy going from a 3500+ ath 64 to any conroe they will get more than a few fps increase in gaming performance.

the very point is that 99% of people run 1600x1200 full quality settings with a single x1900xt, 7900, 7800, x1800xt and its fine, that is real world performance and every benchmark shows most cpu's giving equal performance. 1600x1200 is a low res for crossfire that most people won't be using, its not real world performance as in the real world very few people will set up like that.

http://enthusiast.hardocp.com/article.html?art=MTEwOCw1LCxoZW50aHVzaWFzdA==

read that and the pages before and after, this is a real world benchmark, as in, a setup that you would use in the real world, a single top end gfx at the resolution its best at performing identically on 3 different clocked cpu's(ok not identically, within what 2%).

a lot of people don't encode, don't use very intensive cpu programs except for gaming, and i'm trying to stop the idea that getting a conroe will suddenly jump your framerates up by a big margin, because they won't at all, and people will be wasting there money.

 

[Darkwave]

Put it this way... Even my little X1800XT is bottlenecked by my CPU, a Conroe would go a long way to boosting performance in Oblivion in particular (by a huge amount on the minimum FPS especially.)

Everyone can look forward to this sort jump in CPU power, we've been in need of it for some time.

 

[drunkenmaster]

http://www.extremetech.com/article2/0,1697,1988803,00.asp

well blimey me if this doesn't say it all, the low res low details, and high res high details. for instance, at low res low detail the overclocked x6800 gets almost double the fx60's fps but at high res high detail, exactly the same number, exactly, utterly gpu limited, one gpu, this is real world this is what you will see by upgrading. all the other games show very close fps numbers, a few of them, like hl2 show how well an engine can perform, looks purdy but still runs blisteringly at high detail, you could infact go even higher res on hl2, have playable framerates and as the fps drops they different cpu's will be even closer percentage wise.

 

[drunkenmaster]

Put it this way... Even my little X1800XT is bottlenecked by my CPU, a Conroe would go a long way to boosting performance in Oblivion in particular (by a huge amount on the minimum FPS especially.)

Everyone can look forward to this sort jump in CPU power, we've been in need of it for some time.

prove it, really, thats all i'm asking, as i've said, one, unofficial "preview" guy on a forum showed these numbers with not a single other review supporting them. [h], extreme tech both show them to be wrong, even anandtech isn't showing such huge differences with crossfire running so please, find anything to support the one guy who has said double min fps. the link to [h] shows what, 2fps drop in minimum at a playable framerate and resolution.


please read this http://www.hothardware.com/viewarticle.aspx , then view the next page, this is with a 7900gtx sli setup and both quake 4 and fear at only 1600x1200(low for crossfire) but very high qual settings show almost no difference. also this is the first review i've seen compare 965/975 performance. at low res its slightly behind on fear and ahead on quake, decent chipset then?

http://www.trustedreviews.com/article.aspx?page=7475&head=0

another review which shows low res low qual vs high res high quality, single 1900xtx being used, at the right settings when at 1600x1200, showing almost identical performance across ath 64/conroe's. and read the test setup what it says about the resolutions tested.

 

[Doppleganger]

Having read a few reviews it does seem to me that someone who mainly uses their pc for gaming and has a fairly high-end set-up already, won't find it worth their while to upgrade to Conroe. The Conroe is easily the best cpu out there currently, but without an increase in gpu power gamers are not going to see most of the extra cpu grunt in the real world.

I'm gonna wait until DX10 cards come out and then look at moving to Conroe then. Or even hold on to see what AMD can do with KL8.

 

[drunkenmaster]

Having read a few reviews it does seem to me that someone who mainly uses their pc for gaming and has a fairly high-end set-up already, won't find it worth their while to upgrade to Conroe. The Conroe is easily the best cpu out there currently, but without an increase in gpu power gamers are not going to see most of the extra cpu grunt in the real world.

I'm gonna wait until DX10 cards come out and then look at moving to Conroe then. Or even hold on to see what AMD can do with KL8.

thats exactly my advice, unless however you are on like an ath xp system and want/have to upgrade then it really would be silly buying a am2/939 system new now, though they will make attractive 2nd hand systems right now. tbh even with dx10 cards the cpu won't make more of a difference, dx10 cards will bring some newer effects to the fold, meaning dx10 games will have more effects to do, prettier graphics but using a lot of the power. generally games will become harder and again the new hardware will have a certain resolution where it maxes out. probably for the first few months of a new generation old games and a couple new games don't push the hardware, but some big games will come out and again for a single card 1600x1200 max quality becomes optimum and most cpu's will cope with that.

for gaming upgrading from a 939/am2 won't show anything, for proper cpu usage, encoding and stuff you'll see a clearly big increase in performance(for most things).

 

[Digital Punk]

Yeah, I think there's a few people more than a little ****** off that they're not going to get the huge fps increases with their core duo's that they were expecting.