Back to the question in hand, if I was going to build a complete new system I'd go conroe even if I could only afford a basic chip as that then gives you the option of upgrading to a better processor down the line. I haven't seen or heard of anything in the pipeline from AMD that will make AM2 more desirable in 6 month or even a years time. I'd go with the fresh new platform that intel are offering.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Is the cheapest Conroe still likely to be faster than the fastest AMD?
- Thread starter Participant
- Start date
Soldato
- Joined
- 12 May 2005
- Posts
- 12,631
silversurfer said:
Something to think of, is that its not over the 100 FPS people care about, its other performance.
I for example want to make a workstation, I would rather wait 5 filters to apply a filter on a large PSD, than 8 seconds.
I admit though, I have not used a Conroe, so my oppions are only are from test results and from I have read.
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
easyrider said:How did I post speculation?
I own a E6600 conroe and its faster than my opty 170 @2.8ghz it replaced.
The 6300 can be overclocked to 2.8 ghz making it faster than an FX 62.
You state "super Pi and 3dmark, how very relevent.
Benchmarks from games applications and usual resolutions."
My conroe is faster in games than an FX 62.
And it blows the amd couterparts away in encoding and all other pc tasks.
Gaming benchmarks are just as relevant as super PI scores.
It measures a CPU's performance.
it is wdely dotted around the web that conroe improves Oblivion by 100 % LOL That good enough for you?
If you are a gamer then to state that you are considering AMD now with conroe around is hard to undertand really
again, people have to stop saying this, ONE very very very wrong guy said this, not a single review supports this at any resolution, at high res when its gpu limited where you will end up playing games and the only relevant answer for a gaming pc is that oblivion will run within 5% speed of most ath 64's. at low res the difference is bigger, if you run in low res then buy a x300 and a celeron to match the low res properly. because this is like the 15th time today i've replied to a post like this can you please post any proof behind the 100% improvement in fps claim?
easyrider said:
for a gaming rig its useful to know if while running at the resolution you normally would, would getting a conroe over an ath 64 give you a big boost or not, the answer is not. a lot of people here will upgrade with gaming being the only cpu intensive thing they do, for them a conroe offers nothing. i'd still say buy a conroe over anything else for a new pc if you're coming from something old, but ath 64 up to conroe will offer next to nothing for gaming/ every single review that showed high quality high res setting appropriate to the cards that were used showed no difference in fps(or within such a small range that you would not be able to notice it).
Soldato
- Joined
- 13 Jul 2004
- Posts
- 20,341
- Location
- Stanley Hotel, Colorado
I think on Fear and some other game they showed conroe significantly raised minimum frame rates. Though it would hardly ever fall to 25fps, conroe ensured this figure was 50 instead. I forget where I saw that though
Associate
- Joined
- 12 Jan 2003
- Posts
- 2,034
some sense at last 
Kamakazie! said:some sense at last
hmm.
the sense here is to get a faster cpu for less that being the conroe.
but I'm sure you already know that.
Oblivion is significantly faster on my rig than it was playing it on my opty 170 @ 2.8ghz
drunkenmaster said:
But at high resolutions, the CPU is rarely the limit anyway, any system only runs as fast as the slowest componant. At high resolutions, your slowest componant is your graphics card, not the CPU. Therefore in general you will get very similar results when comparing CPU's if you use high resolutions because (shock horror) you're not really comparing the CPU's at all.
What you should say in your above post is that Conroe offers no benefit with current GFX cards at high resolutions, because it's not the CPU that's holding system performance back. This doesn't prove that Conroe won't offer any benefits, simply that the weakest link in the system is letting it down.
What you say is true, in that it won't offer any benefit for most gamers at average resolutions, but that's not because conroe isn't good. It's the fastest CPU about, and compared to buying an Athlon (even an FX), it will be more likely to still be of use in a while when graphics cards have caught up.
Some interesting benchmarks were posted over at Tom's Hardware (apologies if someone already linked this)
http://tomshardware.co.uk/2006/07/14/core2_duo_knocks_out_athlon_64_uk/
summary of the benchmarks at
http://tomshardware.co.uk/picturegalleries/gallery-200607142.html
Doesn't appear to include some of the lower end chips.
For me Conroe is not just about raw benchmarks though, its about pricing an overall system in terms of VFM.
http://tomshardware.co.uk/2006/07/14/core2_duo_knocks_out_athlon_64_uk/
summary of the benchmarks at
http://tomshardware.co.uk/picturegalleries/gallery-200607142.html
Doesn't appear to include some of the lower end chips.
For me Conroe is not just about raw benchmarks though, its about pricing an overall system in terms of VFM.
Associate
- Joined
- 12 Jan 2003
- Posts
- 2,034
easyrider said:
i do indeed. i was never arguing to the contrary. i was asking for figures that show an E6300 will easily overclock to betwen 2.4-2.6ghz and then outperform an FX-62, as implied by some in this thread.
no ones proved that yet.
Soldato
- Joined
- 4 Aug 2003
- Posts
- 2,636
- Location
- Wicklow, Ireland
This review shows the E6300 (@ 2.4Ghz) outperforming a stock FX62 in multiple tests:Kamakazie! said:
http://www.matbe.com/articles/lire/...-duo/page21.php
More sample benchmarks:
Video Compression: E6300 @ stock = FX62 @ stock (160 seconds)
3DSMax:
No figures can show that a chip will easily overclock, but the evidence so far is that these chips will overclock nicely.
Last edited:
Soldato
- Joined
- 13 Jul 2004
- Posts
- 20,341
- Location
- Stanley Hotel, Colorado
Nicely is putting it mildly. I overclocked a barton by 43% and thought that was amazing. They have got a 67% overclock on air and its still new
silversurfer said:
I thought the 2mb versions where only about 5% slower on average across games n apps at the same speeds with one or two worst case scenarios at 10%? but i've only read some of the reviews so far.
Darren555 said:
Headroom, It'll still be able to keep up with the next gen of gpu's
Not everything is about 1600x1200 completely GPU limited results now.
Dolph said:
Sorry if a little off topic:
So Dolph,
If say in about six months time when newer faster graphics cards are out the gap between the FX62/Conroe should start getting larger at higher resultions?
Thus the Conroe is more future proof and much better value for money than an FX-62.
P.S. On a lot of the high res gaming benchmarks I have seen the Conroe DOUBLES the MINIMUM frame rates when compared to FX-62. For this reason alone I will be purchasing one the rest is a nice bonus. This is where it matters to me most.
Last edited:
DanF said:
I think the review at Anandtech said it was on average 3% slower, at most it was 10%. The conclusion was that the cache doesn't make a huge difference but if you're investing in a PC for the next few years, then go with the larger cache versions.
Kamakazie! said:
I will attempt to answer your questions.
Your post suggests you have two questions: 1) whether the E6300 will "easily overclock" to 2.4GHz 2) Whether an E6300 at 2.4GHz it will "outperform" an FX62.
Q1. To achieve a clock speed of 2.4GHz, the E6300 would need to be accompanied with a motherboard which can handle atleast 343MHz FSB i.e. 343MHz (FSB) x 7 (multiplier) = 2401MHz/2.4GHz (clock speed). All of the popular 975X motherboards can easily handle 343MHz, there's no doubt about that. K.I.T.T.'s post can be used to illustrate two points - firstly, an E6300 can be overclocked to 2.8GHz with the stock cooler; secondly, the popular 975X motherboards (in this case, the Asus P5W) have no problem handling 343MHz FB to give you your 2.4GHz E6300.
RAM is not a problem, most PC2-5300 RAM can run at 343MHz as opposed to its stock speed of 333MHz (i.e. 10MHz overclock). If not, PC2-6400 will definitely be capable of running at 343MHz to give you your 2.4GHz E6300.
Therefore, one can conclude that an E6300 can be "easily overclocked" to 2.4GHz if it is accompanied with a good motherboard and fairly good RAM.
Q2. Now let's turn to the interesting question. We already know that the E6600 at stock speed of 2.4GHz will more or less outperform a stock FX62 in most departments (flick through the benchmarks in Anand's review for confirmation of this). We also know that the E6300 only has 2MB shared L2 cache, whereas the E6600 has 4MB. Therefore, the logical question that follows is whether or not an E6300 overclocked to 2.4GHz is equal to an E6600 running at stock speed of 2.4GHz. Or put in another way, how much difference does the larger L2 cache make? Again, we turn to Anandtech for the answer - Page 4. Anand's review answers this question conclusively. An X6800 with 4MB cache is downclocked to 1.86GHz, and compared to a stock E6300 at 1.86GHz which only has 2MB cache. On average, there is roughly a 3.5% increase in performace by having the larger cache. The main boost in having the larger cache was with DivX encoding, where there was a 10% increase in performance. In games, the performance increase ranged from 1-6%.
So, by deduction, one can conclude that an E6300 overclocked to 2.4Ghz will (on average) be 3.5% slower than a stock E6600 due to the difference in L2 cache. Therefore, the E6300 overclocked to 2.4GHz will "outperform" an FX62 in some cases, whilst in other cases it is more or less "on par" or slightly slower. To prove this, we look again at page 4 of Anand's review. Look at FEAR on page 4. It shows a 1.5% increase by having 4MB cache as opposed to 2MB. Now go to FEAR page 15 and deduct 1.5% from the frames per second of the E6600 to simulate how an E6300 overclocked to 2.4GHz will perform. The result is 106 FPS, which is 5 FPS faster than the FX62. The FX62 has been outperformed, albeit by a small amount (5%). By contrast, Quake 4 showed a 6.7% increase in performance by having larger L2 cache. Now go to Quake 4 on page 14 and deduct 6.7% from the E6600's FPS to simulate how an E6300 @ 2.4GHz would perform. The result is 140 FPS, which is 4 FPS slower than the FX62.
So one can conclude that an E6300 @ 2.4GHz will outperform an FX62 in some cases whilst in others it will be marginally slower or roughly on par.
Taking the above theory one step further, if you overclock the E6300 to 2.6GHz, it is logical to assume that it will outperform the FX62 in all cases, despite only having 2MB cache (it may even be slightly better than a stock E6600, I think)
The above is simply my interpretation of Anand's review, nothing more. I do not present it as fact
Last edited:
the 6300 at 2.4ghz will have a higher FSB if at 1:1 which will give it a little performance boost. So it's probably slightly quicker than the 6600 at stock. At 2.6 it will deffo be quicker than the 6600 and FX62 in pretty much everything.
Of course the 6600 and FX62 can be clocked as well.
Of course the 6600 and FX62 can be clocked as well.