And so it begins again, this time with Intel's new Core2Duo

[raju]

well you know where I learnt my trade don't you????

 

To be serious though, I guess the industry is in a little disarray, I just felt some issues should be presented over there in a diplomatic fashion that's all. especially because I am in the process of a review and have found a few issues....

 

I have to work hard to keep all of my posts balanced and I never aim to upset or hurt anyone, I supose some users respect that. Ultimatley I hope some of the posters in all forums will learn to be a little more open minded and approach both methods with a little respect and understanding.

 

I'm glad you thought it was ok...

 

I'll keep working, I can't see many people sending me stuff as it'll get dissected to some extent. I tend to overlook at lot of things, but 2 things seems to be the driving force behind product marketing (by users that is) these days and thats overclocking for stability and overclocking for benchmarks, so anything failing to perform in those areas cannot be overlooked....

 

I know no-one else has presented results like mine with the Asus, I do hope they can and will. Trouble is most of the guys I know personally have different goals (unfortunatley), but I respect their hobby regardless. The more laymen overclockers I know don't have the cooling capacity to push the board sufficiently ( which is a kind of paradox anyway:lol).

 

regards

Raja

[red930]

Whew, I finally got my silver TDX and processor properly lapped (sorry no pictures, my buddies camera was stolen and I don't have one).

 

Went ahead and just entered in 375x10 at 1.575 volts, went all the way through 3dmark01 with no issues, so I might have to push it further tonight before I try to run orthos on it.

 

I'm idling at 34C right now, though it is slightly colder in my room I believe it was idling at 40-42C before, so it's looking good so far...

 

 

Though that news about nvidia vs. ati cards on this is kinda disturbing...might have to 'steal' my roommates 7900GT...

 

Edit: trying Orthos 'blend' now, it's reaching just under 60C where before it'd be touching over 70C.

[red930]

Ugh.

 

After about an hour and a half of Orthos my computer completely froze, and I'm not entirely sure it was due to any instability in the processor (though it could be).

 

Anyways, one of my Raptors appears to be down for the count, the Bios doesn't see it and neither does Windows. I just got back up and running after reinstalling x64 on my other one just now...it's been a while since I formatted but man a single Raptors feels godawful slow.

 

That upsets me.

 

Ah well, might as well try to take down the other one.

[Polizei]

That sucks. My 7200rpm drive is just fine for what I do, unless I want to rip a DVD and then everything bogs down.

 

I noticed youre just shoving the voltage into that E6700... do you really need that much to keep it stable? I was 5 hours into Orthos with 370x9 @ 1.45v before my AC bit the dust and my temps shot up.

[red930]

I noticed youre just shoving the voltage into that E6700... do you really need that much to keep it stable? I was 5 hours into Orthos with 370x9 @ 1.45v before my AC bit the dust and my temps shot up.

 

Anything above 3500-3600 or so on both the E6600 and E6700 (not counting ES processors) seems to require a lot of juice to attain. I believe Oldguy used 1.65 on his E6600 to get 3800 Orthos stable.

 

And temps aren't really an issue right now. At 1.475 volts and 350x10 (the only stable speed I remembered the settings for so I could install windows) I'm idling at 19C on one core and 20C on the other. It's nice 'n cold in here, can't wait for winter to really get into full effect. Last year my Opteron 170 would hit around 15C full load...I miss those days.

 

But anyways, I'm tryin for me as I type. I'm windering if a potentially bad hard drive and just a blown Windows install in general have been giving me problems. That last install went through at least 6 processor swaps (between a Pentium D 805 and my E6700) and all kinds of crashes on overclocking both of them. Add in some Pi runs on that poor 805 at nearly 1.8 volts and I'm just asking for some bugs.

 

Anyways, hopefully I can get priming tonight. I think I'll stick with trying to get 370 stable then move on from there.

[Polizei]

Yeah, I need to do something about my 60C idle chipset... AS5 didnt bring the temps down compared to the stock stuff, they are actually up about 12C. I think Im going to buy the HR-05 tomorrow and see what I can do.... I dont know.

[OldGuy]

Anything above 3500-3600 or so on both the E6600 and E6700 (not counting ES processors) seems to require a lot of juice to attain. I believe Oldguy used 1.65 on his E6600 to get 3800 Orthos stable.

 

Check again... I used 1.625 and I might have gone just a tad above that to try 3850. When it rebooted during the first couple of minutes of orthos I decided I was done.

[bone450]

I just read the first post, and only the first post. I must say that it was some of the funniest stuff out there. Wait, hold on, my neighbor is burning down my dog house with his overclock.

[raju]

Ugh.

 

After about an hour and a half of Orthos my computer completely froze, and I'm not entirely sure it was due to any instability in the processor (though it could be).

 

Anyways, one of my Raptors appears to be down for the count, the Bios doesn't see it and neither does Windows. I just got back up and running after reinstalling x64 on my other one just now...it's been a while since I formatted but man a single Raptors feels godawful slow.

 

That upsets me.

 

Ah well, might as well try to take down the other one.

 

as an experiment if you're willing, try using crystalcpuid and drop the multiplier to 9x, and use clock gen to see if you can get over 400fsb Orthos stable. In truth you probably won't based on the fact that the chipset latency gets set according to the default cpu multiplier (tony and fcguy's work). But it will be interesting never the less. If using a single gpu try fiddling with that pcie even more..

 

regards

raja

[Guest LithoTech]

As usual, another fantastic rant to read, and it doesn't matter how much I agree with it or not, that is beside the point. Good read, quite enjoied it!

 

Had a little spare time this morning to read the first post over coffee. I had left the page open from last night, figured I could savour it more today rather than skim-rush it with sleepy eyes last night.

 

I peg myself in that majority of users, although I would describe myself as a conservative overclocker. Extreme for me is where I have to add voltage to the cpu. I like to push the system to it's limit on stock voltage, achieve the highest possible clock in that scenario, verify it is 100% stable for 80 or 800 hours prime because anything less is not a perfect config and something is amiss. I'll do the standard 24 hours prime as routine, then proceed to use the system for weeks on end without rebooting, always folding when I'm not hogging the resources, opening and closing a myriad of apps and games -- if I can go a month like that, or to the next MS update that requires a reboot, I'm satisfied. This doesn't mean I don't add voltage and see where or how far the system will go, it's just that it isn't my primary goal and doesn't impress me as much as a high frequency clock on stock volts that can have several weeks uptime without issues. I still enjoy pushing the system to voltage limits in some sessions, especially for ram, but usually not until after I've locked down a stable stock volts config.

 

We have some very interesting times ahead of us. These new conroe cpu's are very attractive no matter what mobo you put them in.

 

I've been running the numbers for several different configurations using C2D's, and putting an E6300 in the even cheapest mobo available (ASRock 775Dual-VSTA comes to mind), will net a very responsive slick system for a budget orientated price. Yes that mobo is very limited in how far the FSB will go, but this application of components would not be for overclocking, it would use previous ram and video, and end up costing only the mobo and cpu for an upgrade that would seriously squash it's previous configuration. Some of my clients would like this very much.

 

There is just no end to the combinations of use and abuse we could put these cpu's and mobo's through. I can't wait to get my grubby hands on one lol.

 

The Infinity board was initially attractive to me personally for the same reasons HG states here. I was reading through the various intel forums and posts here when these boards and cpu's first appeared, and was only discouraged by the cost of the board in my area -- and only because in the comparative price range there were other, perhaps more attractive, options. However, these other optional boards were not 975x chipsets, and do not come with the rest of the DFI branding, features, support and sarcastic reviews.

 

I'm still a ways off of getting into one of these systems for myself, but should hopefully be building one for a client in the next weeks. I'll certainly be taking another look at the Infinity, and revisit local prices. This system will need to be a multimedia station, and jack into a home theater with a large plasma. Initially using just TV out, I expect we'll be putting in a TV Tuner as soon as everything else is working good. So I'll probably be looking for a board for it's features more than it's overclocking prowess, and in fact may still go with an AM2 system for the huge selection of cpu parts to upgrade to and their respective currently slashed prices, and larger gamut of motherboards to choose from. I have pretty much creative license here, and would listen to any suggestions on this, just send a PM.

 

I'll also have to revisit this thread to follow the rest of the replies as I've no time now and my coffee is done. I'll have to stop in the intel section again and spend an evening there soon too.

[red930]

as an experiment if you're willing, try using crystalcpuid and drop the multiplier to 9x, and use clock gen to see if you can get over 400fsb Orthos stable. In truth you probably won't based on the fact that the chipset latency gets set according to the default cpu multiplier (tony and fcguy's work). But it will be interesting never the less. If using a single gpu try fiddling with that pcie even more..

 

regards

raja

 

Gave that a shot and still couldn't get over 389, so apparently it does have something to do with the latencies used.

 

My overclock at 350x10 was stable through 120 mhz on PCI-e, I didn't try any higher but I suppose I could. But at the same time I I don't think it's necessary to go over 101 either, as increasing it didn't bring me any more stability at 370x10.

 

I think I'm gonna back off and go a bit slower. Either I'm missing something or I've reached the limits of the board itself or the processor.

[raju]

I must admit, the E6700's don't seem to be clocking any better than the E6600's atm. On phase there is a margin due to fsb limits on the 975 boards, which I will soon discover. My friend Johann has a E6700 on the way which I may ask to test in return for a mutual exchange. WIll be interesting to see if the PRO, can crack that fsb barrier on an E6700. We are frightfully close to the rd600 now, Oskar managed a whopping 530fsb out of that beast, unkown to the actual tests he ran but that should at least eliminate chipset latency limits on the e6700's. The rest is down to cooling.

 

regards

Raja