View Full Version : ASRock 4CoreDual-SATA2 R2.0 + Intel Core 2 Quad 9550 = big mistake? >_<

09-09-2010, 02:55 AM
Hi all,

I was hoping I might be able to ask Sean about this directly, but it seems his PM's are currently disabled, so perhaps he will be able to chime in here if nobody else knows the exact answer. :)

I recently received a 4Coredual-SATA2 R2.0 (pre-installed BIOS version 2.20) and a Core 2 Quad Q9550 (E0). Per the CPU support list on ASRock's page (ASRock > Products > 4CoreDual-SATA2 R2.0 > CPU Support List (http://www.asrock.com/mb/cpu.asp?Model=4CoreDual-SATA2%20R2.0)), the Q9550 is supported in all 3 revisions, with only the single caveat that the "FSB may be reduced 5%." My problem is that since installing the CPU, it runs at a 266 MHz bus speed, with a capped 8.5 multiplier (it can be reduced in the BIOS, but not increased), resulting in a CPU speed of 2.26 GHz rather than its potential 2.83 GHz stock speed or the "advertised" 95% speed of 2.69GHz.

As the support for this processor was explicitly added to the R2.0 version of this board, is there a setting somewhere that I might have missed that is supposed to add make-shift compatability for this processor at "correct" speeds, that may not be properly documented in the manual or somesuch? It has been suggested that the only way to try to get any closer to the stock speed of the processor would involve overclocking the fsb, but the purpose of getting a fast "supported" cpu in the first place was to try to achieve good performance without the need for any kind of overclocking and risking stability and undue hardware wear and tear.

I know it might be hoping for too much to think that there might be a "magic" solution to the problem, but if there is not, I would probably not be alone in thinking that this CPU should really not be on the supported list for this motherboard on the ASRock page, or should come with a warning that the FSB (and total cpu speed) would be reduced by 20%, not 5%... After all, I was so excited about the upgrade I thought I would be making, only to be disappointed and have a bit of a bad taste left in my mouth. :( HEEEELLLLLLP! lol

Thanks in advance,

Edit Oct. 2nd, 2010: Additional information has been added to this thread in my last post, after finally hearing back from ASRock. As it stands, they have so far stood by the statement that the processor should run with the equivalent to its normal speed with a maximum of a 5% clockspeed reduction, but have been unable to provide a solution to reach those speeds for my specific case. I would greatly appreciate it if Sean C. or other ASRock sources could help me to get past this dead end.

09-09-2010, 09:09 AM
I have this mobo. They were great for bridging the gap between p4's 472's and c2d's and socket 775's, since they can use agp/pcie ddr1/ddr2 memory, but they're really limited by their slow pcie/agp sockets. I would be really surprised if they could handle a quad core with that amount of power.....

It's really interesting you were able to use one at all though.

Even if it does work, it seems like a waste of a good processor since the pcie socket is only 4x...

09-09-2010, 12:44 PM
... I would be really surprised if they could handle a quad core with that amount of power....


If you had followed the link in my original post, the ASRock CPU Support List includes several Core 2 Quad core CPUs, and that very support is touted as a major feature of a board that is, in fact, named "4Core..." Notably, the R2.0 version's support list (ASRock > Products > 4CoreDual-SATA2 R2.0 > CPU Support List (http://www.asrock.com/mb/cpu.asp?Model=4CoreDual-SATA2%20R2.0)) also includes support for additional 45nm parts vs. its non-R2.0 counterpart (ASRock > Products > 4CoreDual-SATA2 > CPU Support List (http://www.asrock.com/mb/cpu.asp?Model=4CoreDual-SATA2)).

I could see how some might argue that such a CPU might be "wasted" on a motherboard of this kind, but it was purchased with the intention of trying to eke out every last bit of mileage from my existing AGP video card and existing RAM (though alas, having to drop 2 DIMMs and losing 1 gig of my original 2 gigs of RAM killed performance so badly I had to pick up a couple of sticks of DDR2 within the first week) before making the eventual leap to a PCI-E card. My current roadmap for future upgrades looks something like perhaps picking up a GTX 460 or low-end Radeon 58xx card as the next upgrade, and sometime later, perhaps switching to an SLI/Crossfire enabled motherboard and migrating all other components over while adding a 2nd same graphics card.

Keep in mind, the recorded performance hit of the PCI-E slot being only a 4x 1.1 slot vs. a 16x 2.0 slot is typically only measured at a maximum of 10-15% in most benchmarks and even less in many real-world gaming applications, so it's not as much as a bottleneck as some would think. Recouping this little bit of extra performance when changing to a dedicated PCI-E motherboard would simply be an added bonus added to the power of a dual-card configuration, following the path above.

Nevertheless, not having things work as expected - most importantly, *as advertised*, simply leaves me quite frustrated. Had that CPU never even appeared on the supported list, it may not have swayed me from buying the board - it would have simply caused me to consider different processor options.

09-09-2010, 11:39 PM
Hmm, what memory are you using? Maybe there's an incompatibility problem there?

09-10-2010, 01:14 AM
Hmm, what memory are you using? Maybe there's an incompatibility problem there?

I started off by trying each of the PC3200 (DDR400) 2x512 mb DDR 1 kits I had here, using both single DIMMs and dual-channel mode testing for both initial stability and to see what settings the BIOS would come up with. Interestingly, it defaulted to 333 MHz (166*2) for both my GEIL and OCZ modules, and wasn't exactly happy running the GEIL kit at 400 MHz in dual-channel mode (though the OCZ kit worked just fine).

Through all of the various memory mixing and matching, the CPU internal clock speed, multiplier, and external FSB remained exactly the same, at 2266 MHz, 8.5 multiplier, 266 fsb. This includes the results of the new RAM that I am using; a Corsair TWIN2X4096-6400C5C kit whose modules are rated compatible with just about every DDR2 motherboard out there, and which I got to work with the stock BIOS using the hand-dandy trick of "install one module, force single channel, cas, tras, speed, ram voltage = low in bios, check stability in os, shut down, install next module, boot to os, confirm stability, restart, enable dual-channel in bios." As the RAM is rated at PC2 6400/DDR2 800MHz, even with the memory linked 1:1 with the CPU's FSB starting from 266 MHz and working upwards, starting from 533 MHz mode would put the RAM at 667 MHz at a 333 MHz fsb, or starting at 667 MHz would put it at 835 MHz at 333 fsb (or 800 mhz at 318 fsb), where it should theoretically work just fine without breaking much of a sweat.

It's funny, though, that the motherboard is either interpreting the CPU details or reporting them directly from the CPU themself in an accurate fashion, before modification of its performance. Anywhere I check (Windows System Properties, CPU-Z, etc.), it shows up as a "Intel(R) Core(TM)2 Quad CPU Q9550 @ 2.83 GHz" followed immediately by an actual speed of "2.26 GHz" lol. I guess those software programmers love rubbing salt in the wound. :P

09-10-2010, 07:49 AM
Hmm, that's interesting. I actually have a similar system with an e6600 processor in it, that reads out as a 1.6ghz. It's actually speedstepping though, although I didn't think this was possible with an e6600. You can see the processor rise and fall in cpu-z as the chip gets more use. I actually have speedstepping turned off in the bios even. That's using an old ecs board. My system benchmarks as comparable to other systems I have with the same processor so I assume it's alright.

Are you sure something similar isn't happening? What does the system benchmark at on geekbench? Is it similar to others with the same processor or significantly lower?

09-10-2010, 01:12 PM
CPU-Z and other monitoring tools show absolutely no variation in the CPU's clock speed - it is absolutely rock solid. Also, the current BIOS for my motherboard does not appear to have any form of support for SpeedStep as an option, let alone active.

3D Mark 06, Sandra, and those tests which ran on PCMark 05 (I didn't have Windows Media Encoder installed, and just found it), showed any other Q9550, Q9450, Q6600/6700 out there spanking my system for the most CPU-intensive/CPU-specific tests - with the performance difference roughly equivalent to the percentage difference in MHz. These same benchmarks also suggest that I should be comparing against 2-2.4 GHz cpus rather than those clocked at 2.6 GHz and beyond. Geekbench also reported that my CPU's correct "peer group" would be in the 2.2 GHz range, as well - but also correctly identified that it *should* be running on a 333 MHz bus.

As I may have inferred before, even at its reduced speed, the Q9550 can handle just about anything I can throw at it that isn't held up by some other part of my system, but getting it up to snuff would reduce the likelihood of it becoming a bottleneck even further - and would help to get an overall better picture of where my system truly stands if/when I make any leaps in the video card department.

09-10-2010, 01:52 PM
Well, most of the others figure in your graphics cards more than anything. That's why I like geekbench for rating pretty much pure processing power, but not much else. It's telling then if it's low. What did you score? Send a link? It should give a hint if there's a fixable problem maybe.

Nevermind, I found it. Is this your system?


Yeah pretty interesting, proper results seem to be around 4800, yours is almost 1000 points lower.

Here's the current build of my 4core. I downgraded it to a p4 for a machine for my mother. lol


You might also browse the results for similar systems to see if there are any similar builds.


There aren't many quad builds.

Here it looks like someone tried the same thing and got similar results



Yup, not seeing any quads that look like they're getting anywhere near the full potential, so probably a lot of unhappy people out there.

Hmm, this guy managed to oc his q6700, it's actually running at 2.75 and it's a 2.66 chip


American Megatrends Inc. P2.10

His bios revision is a little different from yours, yours is p.2.2

All of the bad ones seem to be running p2.2...

09-10-2010, 02:24 PM
Yup, those are certainly my results you found there in that first link. As I'm sure you've come to notice from the others you found, the guys sitting on the antiquated Q6600's at their 2.3 GHz stock speeds and smaller cache get roughly the same performance as I'm seeing on newer tech with massively more cache and designed for a half-gigahert speed advantage. And the Q6700's actually come out ahead (much to my chagrin...). So technically, they *are* getting just about all the juice they're designed for - not to mention they can be had for half the price, if you don't mind going used/venturing onto auction/private sales sites for a CPU rather than going retail for the peace of mind.

Sure, that guy with the much more modern QX9650 (modern vs. the 6x00's - but the 9550 E0 is the most recent tech afaik) sees a bit of a CPU performance boost from his slightly higher clockspeed (despite the fsb issues), but his RAM subsystem is totally hosed. At least I feel slightly smarter about going the DDR2 route now, lol. Curious about what that one Q6600 user used to edge me out there, though... Anyway, I suppose it will be an interesting benchmark to revisit down the road if I change to a motherboard that lets 'er rip at full speed, huh? :P

Edit: BIOS revision 2.20 is the latest official BIOS from ASRock and adds support for quite a few ATi PCI-E cards without really doing anything to disrupt performance in other ways, that anyone has spotted. It also seems to be much more compliant in allowing users to go the 4 gig DDR2 route (as I have done) than earlier versions. Overclocking on this board has never been ruled as impossible (just very hit and miss) - and other sources have mentioned experimenting with the concept of simply pushing my FSB as far as it'll go without hard locking, since my CPU and RAM should be able to take it, but I don't want to fry the board prematurely, since it might otherwise find a happy home in a family member's system with a different CPU (one designed for a 1066 fsb!) if I do end up swapping motherboards.

09-10-2010, 02:27 PM
q6600's are actually 2.4ghz stock, so they're being impacted as well, but probably closer to the percentage you were expecting.

I'd try an earlier revision bios. My system is actually running 1.5 and been working just fine for years now, although never had a quad in it, or played around with the clocking of it. I used this board to upgrade from socket 478 p4's, which they were great for, and originally had an e6600 in it, but moved that one to a more decent board once I got enough parts together.

You should be able to pick up a decent 775 mobo super cheap at this point.

I have a system based on this board:


It's not horrid. $50.


Early bad drivers gave it a not so awesome rep, but they're fixed now. It's one of my most stable win7 systems.

Actually that one might not work. Mine is actually a p5n-e sli that one's just a p5n sli I think there might be a difference looking at geekbench.

Might have to go with a p5n-d...


here's an e refurb for about $65


09-10-2010, 09:48 PM
Yeah, the only problem with wanting to change motherbaords at this point is that I'm still somewhat tied to my Radeon X1950 Pro AGP card that I'm using, and I'd sort of prefer to wait until I can get a decent PCI-E video card rather than picking up a "disposable" card just to make a quick leap to another motherboard. Of course, I'm also nervous about shooting myself in the foot and there being absolutely *no* 775 boards left by the time I make the leap, so hopefully I'll be able to save up and pull the trigger around christmas or somesuch.

Also, re: ebay - I'm in Canada, so those listings that ship only to the U.S. are out for me, and sometimes even if you *can* get something shipped to Canada, if the seller only ships by UPS, they add $25-45 in additional "brokerage" fees any time they cross the border. :( My dad once bought a "$50" motherboard that ended up costing him over a hundred bucks by the time it got here, after all the rediculous shipping charges... Had a decent experience with newegg.ca, though; the price displayed (with local taxes, etc. added automatically in the shopping cart) was the final price - item was delivered with no "SURPRISE! We need more money before we hand it over!" Tiger Direct also has some stores in our area which sometimes have decent stock, as well, and they tend to price match their website.

09-11-2010, 01:47 AM
It sounds like you have another processor for this board though, and extra memory. I'd just get another graphics card cheap and build a new system and keep that old one the way it is, and put the quad in something decent. You can get msi gtx 260 refurbs cheap these days and those are really decent cards that will be at least twice as powerful as that agp. You could probably put together a whole new system for less than $150 or so at this point.

09-11-2010, 06:14 AM
It sounds like you have another processor for this board though, and extra memory. I'd just get another graphics card cheap and build a new system and keep that old one the way it is, and put the quad in something decent.

Extra memory, yes. An extra cpu, no; the suggestion of putting in a CPU that was content to run on a 1066 fsb when passing the board along would be contingent on actually being able to find one of said processors at the time, lol.

And like I was saying about the video card, with the 9550 running at its full potential, any sub-$400 video card would probably still be the bottleneck for most games, so I figured there's no real sense in dropping $100 or so on a video card right now and another $150 later after the GTX 460's or whatever come down that far, when I could just be a little more patient and maybe pick one up (or even newer technology) for less total expense in a few months - and grabbing a corresponding motherboard at the time or shortly thereafter.

Note also that quite a few older/cheaper 775 motherboards, such as the P5N-SLI may use older chipsets that do not support a 1333 MHz fsb (which may be one of the major differentiations between the nforce 500 and 700 series chipsets, for example), so I would be quite wary about buying anything off of ebay unless I could get decent information from the manufacturer, first - and quite a few manufacturers' pages have been culled of many older products. Just try searching for the P5N-SLI on ASUS' site, for example.

I'm still hoping that Sean or another user with good ties to ASRock might reply here with an explanation of how the Q9550 is supposed to function with my existing motherboard as advertised on their website, to save the extra future hassles. :)

09-11-2010, 07:05 AM
You can double check by looking at geekbench. I'm 100% sure a p5n-e, or d will support your processor because there are examples of it. You can get refurbed gtx260's for like around $100, or even less..so they're not a bad card to upgrade to right now.

09-13-2010, 11:13 AM
You can double check by looking at geekbench. I'm 100% sure a p5n-e, or d will support your processor because there are examples of it. You can get refurbed gtx260's for like around $100, or even less..so they're not a bad card to upgrade to right now.

I took a look around to see what 200-series nvidia cards might be compatible with my motherboard since I don't have access to the resources to justify a GTX 460 atm, and it seems that all of the GTX 200 series have 10.5 inch PCB's (check the specifications at GeForce GTX 260 (http://www.nvidia.com/object/product_geforce_gtx_260_us.html), for example) in reference/standard format - which is a problem for the 4CoreDual-SATA2 boards. They are only 9.6" across, with the IDE connectors along the same horizontal plane as the PCI-E slot, so any cards with an overall length of >8.5" just plain will not fit with IDE cables present, and may still cause rubbing or interference issues on/with the sockets even using an all-SATA setup (I currently have all parallel ATA devices, no SATA).

Basically, that means the only nvidia solutions in that price range would be the GT 240 and GTS 250, both of which received quite lackluster reviews (and justifiably so) for their performance vs. ATi counterparts in the same price range, or even other nvidia products' overall "bang for the buck." Now, I know these ASRock Dual boards constitute the very definition of a "niche market," but I find it extremely interesting that the GTX 460 just so happens to have an 8.25" overall length and is reported as being compatible with PCI-E 1.0/1.1 slots by Zotac, Palit, and probably other manufacturers, on their sites, making them seem almost custom-tailored to accomodate the motherboard's key "flaws" in graphics card support. Not to mention that (depending on whose benchmarks you believe) a GTX 460 rivals the GTX 285 in performance, which in turn compares with dual GTX 260's in SLI, while consuming less power and generating less heat and noise. Oh, and did I mention it also has DirectX 11 support?

And thus the logic behind the GTX 460 being my object of future graphics card lust. The X1950 Pro I am using now is the first ATi card I have used, and through several OS reinstalls and multiple versions of Catalyst drivers, it has just never had the kind of stability and reliabilty my nvidia parts always seemed to bring to the table, and has me leery of going the ATi route again, when you still frequently see their cards being described as "a great piece of hardware for the money... But the drivers still suck."

You would think living an hour away from ATi headquarters would make me an automatic fanboy, but all it does is make me wanna go over there and smack someone. :)

10-02-2010, 12:57 PM
To update this thread:

I finally received a reply from ASRock CSD from the submitted web form. After a couple of emails back and forth with the CSD rep stating that the fsb should only be reduced by 5% as per the CPU compatibility page, this was the last message I ended up receiving:

"There is no any special setting for 1333FSB processor, You simply just drop in and bios will automatically detect it. From all the hassle I see. I suggest you contact the vendor and get the different mainboard, that fully support your Q9550 processor. Sorry about the issue you are having."

I replied back:

"It sounds like you are saying that my CPU should be able to run at the proper speed, but that the auto-detection is not setting it up correctly in the BIOS. Could you please tell me the proper manual BIOS settings to enable the correct speeds for the Q9550 CPU? If it is simply a problem with the 2.20 BIOS auto-detection, manual settings should be much less dangerous to test than a BIOS flash down to version 2.10."

Since that message, I haven't heard any reply for the past week. I am almost tempted to try to see if BIOS 2.10 would "magically" find proper settings to get my CPU up to speed, since the rep seems to believe that the 1333 MHz fsb support for the Q9550 is not an error in the listing on the website.

I apologize for what might seem like thread necrosis, but based upon the "official" position that my CPU is not performing as expected by an official ASRock rep, I am hopeful that perhaps I may either receive another email at a future date, or hopefully receive some feedback from Sean C. here that might be useful both to myself and for anyone else looking to equip one of these motherboards with what could well be the top-performing CPU on the support list.

10-15-2010, 06:53 AM
Hello just got ASRock 4CoreDual-SATA2 R2.0 + Intel Core 2 Quad Q6600 .

The long story sort , I got the board before three days , originally I got an E7500 CPU,
that turned bad , and I had to return for replacement on warranty hopefully.
The point is that this could take lots of time , so I just got today another (used) CPU
the Q6600.

Every one who gets that Quad Q6600, the first thing that he reads on Google are how well it overclocks at the Toms hardware article.

I am trying to let behind an truly great system, the P3.2 with ASUS P4C800-E DLX ,
just because my needs about video editing does not get covered by it.

I am very happy wit my ATI HD3850 AGP, that mostly for it, I had choose the 4CoreDual-SATA2.

About the 4CoreDual-SATA2 -5% at quads , I was aware of it , and truly I do not care.
Even the Quad Q6600 in comparison with the latest E7500 that is the most powerful core 2 duo that the board supports , looks to be killed from the Q6600 at every benchmark.

I managed just to test this board with an Pentium D so to confirm that it works,
just an 10 minutes test.

Yes I am dying to learn if its possible to push this CPU just a bit more than the default.

Currently I own OCZ platinum SDRAM 3200 2-2-2-5 and ordered and DDRII OCZ value pro PC-4200 ( 266) 4-4-4-12 so at 1:1 it will be good enough .
The same DDRII gets out as DDR 667 (333) with 5-5-5-14 , and so I bet that my modules can run in both speeds by setting the proper parameters.

I have all ready load the official Bios 2.20 , and I have also and the unofficial Bios ,
released by an German site of people that tweak the 4CoreDual-SATA2 R20 ,
and by the tweaked bios it can use even 4GB of ram.

Unfortunately , I did not manage to find if this hacked bios offers more tweaking options than the standard.
And at what extend .

I vote too over stability, and I do not like to risk the motherboard too.

And so I will follow only safe for the motherboard steps, the Q6600 has known potentials, and the 3G looks sweet , but I can live even with something in between the default and the 3G ..

At the Bios looks that the motherboard can lock the PCI and AGP , and this is nice.
My question are ... how much is the stablest FSB that this board can offer.

If it can work at 333 1:1 , I will fly my hat ..
That's an 9X333= 2997 - 5% = 2.847 GHz that works nice for me.

10-15-2010, 07:14 AM
Additional info about the stability issue that the board faces...

At the most threads that I had read , the issue in all are the memory voltage.
The latest OCZ value Gold that I got are 1.8V 100% compatible with this board that at the memory voltage setting ( Bios High ) it can only supply 2.03 volts Max .
And so the most of the PC-800 DDRII runs out of specs !! They need 2,1 or 2,2V .

The hacked bios offers at list more voltage at the memory , and funny enough , after loading the hacked , by setting the memory voltage to Low ( The hacked setting gives more voltage to the modules ) ...

Here is an report of an user that uses the hacked Bios ...

.................................................. .................................................. ................
<!--[if gte mso 9]><xml> <w:WordDocument> <w:View>Normal</w:View> <w:Zoom>0</w:Zoom> <w:PunctuationKerning/> <w:ValidateAgainstSchemas/> <w:SaveIfXMLInvalid>false</w:SaveIfXMLInvalid> <w:IgnoreMixedContent>false</w:IgnoreMixedContent> <w:AlwaysShowPlaceholderText>false</w:AlwaysShowPlaceholderText> <w:Compatibility> <w:BreakWrappedTables/> <w:SnapToGridInCell/> <w:ApplyBreakingRules/> <w:WrapTextWithPunct/> <w:UseAsianBreakRules/> <w:DontGrowAutofit/> </w:Compatibility> <w:BrowserLevel>MicrosoftInternetExplorer4</w:BrowserLevel> </w:WordDocument> </xml><![endif]--><!--[if gte mso 9]><xml> <w:LatentStyles DefLockedState="false" LatentStyleCount="156"> </w:LatentStyles> </xml><![endif]--><!--[if !mso]><object classid="clsid:38481807-CA0E-42D2-BF39-B33AF135CC4D" id=ieooui></object> <style> st1\:*{behavior:url(#ieooui) } </style> <![endif]--><!--[if gte mso 10]> <style> /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-parent:""; mso-padding-alt:0cm 5.4pt 0cm 5.4pt; mso-para-margin:0cm; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.0pt; font-family:"Times New Roman"; mso-ansi-language:#0400; mso-fareast-language:#0400; mso-bidi-language:#0400;} </style> <![endif]--> ASROCK 4COREDUAL-SATA2 R2.0 BIOS: R2.20A
- I’ve had to reset the bios 10+ times (Overclocking).
- No post with 2nd stick of ram without bios 2.20A
- CPU-z showed that the board guessed wrong on ram timings
- Manually set timings (see below)
- Pretty sure it’s been driven to the limit now
Bios Readings:
BIOS L2.20 REV: a
Total Memory: 4096MB
Bios Ram Settings:
DRAM Freq: 333 (ddrII667)
Flexibility Option : Disabled
Trp: 4T
Tras: 12T
Trcd: 4T
DRAM Voltage: Low
DRAM BUS Selection: Dual Channel
Bios CPU Settings:
CPU Host Frequency: Manual
Actual Frequency: 222
Ratio Actual Value: 14 (222Χ14=3.1ghz)
Bios Hard Drive Settings:
PIO Mode: 4
S.M.A.R.T.: Enabled
32Bit Data Transfer: Enabled
Boot-Up Readings:
3.10 Ghz
3328mb Ram
CPU-Z (Version 1.55):
Core Speed: 3107 Mhz
Bus Speed: 221.9 Mhz
Rated FSB: 887.7 Mhz
Memory Type: DDR2
Memory Size: 4096 MB
DRAM Frequency: 369.9 Mhz (x2=ddr2-739.8!!)
CL:4 tRCD:4 tRP:4 tRAS:12
Command Rate: 2T
Passmark Performance Test:
Memory Mark: 981.6(top score) 959(bottom score)
Mem – Allocate Small Block: 4227.5 Mbytes/sec
Mem – Read Cached: 2736.7 Mbytes/sec
Mem – Read Uncached: 2432.5 Mbytes/sec
Mem – Write: 1513.9 Mbytes/sec
Mem – Large Ram: 1283.9 Operations/sec
CPU Mark: 2603.4
Disk Mark: 683.8
Total Physical Memory: 3327 MB RAM
Total Available Memory: 2740 MB RAM
Bottom Line: If you want to overclock, grab ram that can handle it!
-Am still playing with settings, if anyone has suggestions, post away, even though this is an old thread, figured someone may benefit from this somewhere sometime!