Sl7e5 Vs Sl7kc: The New 3.2e D0 War!!!

[woutertal]

Jepp. but does it run as hot as the C0 and D0 ? ..if it does then watercooling is the way to go for 24/7 stable overclock. I get around 68celcius at full load with artic silver 5 + Zalman 7000AlCU + 4 case fans (case open) + a huge outside fan blowing in. I have to look for better cooling. Hell not even a Zalman Reserator seems to cut it with the prescott core.. Going to do the vcore mod on monday first and then i have to buy a watercooling system.

Watercooling is the only way to go if you want to reach stable 4.0 in my opinion. Although Jszent runs 4.2+ stable with a Zalman 7000, modified with a huge Panaflo fan.

The PN runs hotter than the DO steppings if you can believe the Intel spec. It is not always right: for example the CK is quoted to have a lowe Vcore than it has in reality. So.. who knows? The guy in the link above might have one or not. The problem is that he only knows the order code which now is the same for E5 and PN.

 

[Vflux]

Watercooling is the only way to go if you want to reach stable 4.0 in my opinion. Although Jszent runs 4.2+ stable with a Zalman 7000, modified with a huge Panaflo fan.

The PN runs hotter than the DO steppings if you can believe the Intel spec. It is not always right: for example the CK is quoted to have a lowe Vcore than it has in reality. So.. who knows? The guy in the link above might have one or not. The problem is that he only knows the order code which now is the same for E5 and PN.

If anyone would like to compare their stable 4Ghz clocks here are my runs:
3DMark05
3DMark01
PCMark04
3DMark03
Aquamark3

Everything is run via side panel closed and no artifacts on my screen.

Vflux

 

[Sucka]

If anyone would like to compare their stable 4Ghz clocks here are my runs:
3DMark05
3DMark01
PCMark04
3DMark03
Aquamark3

Everything is run via side panel closed and no artifacts on my screen.

Vflux

I don't know what it is with you Vflux, but you've got me benching like crazy lately

These were done on my secondary rig once again, SL7E5, BH5, ect. Most of my ORB slots are filled up with my new 6800 results. But here are some screenies of mine at "daily settings". Looks like once again the BH5 makes a big difference in most of these benches as we're almost identical on all of them despite 200MHz difference in clock. I am using a 9800XT, however I ran it at default speeds for these tests to make it closer to your system stats. I'm gonna bump my CPU up as high as my BH5 will go sometime this week and see what i can get then. Once again fun to throw our benches out there. And i'm sorry to say, but i'll be moving to AMD next week for the first time, but i'll be keeping my intel around for fun. Not my top benches, but just a comparison.

 

[Sucka]

Oh yeah, system stats for benches:

Abit IC7-G - No mods, air cooled
3.2E @ 238FSB @ 1.5v - SL7E5 (3.8GHz)
2 x 256 Mushkin lvl 2 pc-3200 BH5 @ 3.3v
ATI 9800 XT @ Default clocks
WD Raptor 36GB
Antec True Power 550
Audigy 2 ZS
Windows XP Pro SP2

 

[DirtMerchant]

I just fried a 3.2E! Trashed it.... and that was with water cooling. These things are nasty little beasts and they will rape your mobo. Right now I am working on replacing all the caps on my motherboard so I can get a more stable vcore.

To be honest with you all, I hate these friggin Prescotts. I should have just gotten another Northie, but I figured the deeper pipeline would get me a higher OC. I wound up trashing my CPU in the process.

It is definitely not the PSU that is the problem, it is that these motherboards are NOT made to handle the raw power and leakage that these CPUs drain. I agree with Hawk. I want something that will run stable 24/7. I was hoping to hit 4Ghz atleast, but so far I am unimpressed, in many ways I wish I got a 3.2 Northie instead.

I have a RMA'ed 3.2E coming to me soon. I might even be willing to sell it cheap to someone here if they want it. Brand new unopened 3.2E.

Anyone interested hit me up in a PM. I think I might just stick with the Northies for now, unless my "droop" mod thread gets a good amount of response.

 

[woutertal]

I just fried a 3.2E! Trashed it.... and that was with water cooling. These things are nasty little beasts and they will rape your mobo. Right now I am working on replacing all the caps on my motherboard so I can get a more stable vcore.

To be honest with you all, I hate these friggin Prescotts. I should have just gotten another Northie, but I figured the deeper pipeline would get me a higher OC. I wound up trashing my CPU in the process.

It is definitely not the PSU that is the problem, it is that these motherboards are NOT made to handle the raw power and leakage that these CPUs drain. I agree with Hawk. I want something that will run stable 24/7. I was hoping to hit 4Ghz atleast, but so far I am unimpressed, in many ways I wish I got a 3.2 Northie instead.

I have a RMA'ed 3.2E coming to me soon. I might even be willing to sell it cheap to someone here if they want it. Brand new unopened 3.2E.

Anyone interested hit me up in a PM. I think I might just stick with the Northies for now, unless my "droop" mod thread gets a good amount of response.

Nobody can have missed all the warnings in this thread about mosfet and caps cooling, general mobo cooling etc. I am sorry it happened to you dude, but if you would have done yr research it wouldn't have happened. Sorry to put it so blunt, but it's true that there's now quite some info available that anybody tempted to OC a 3.2E tp 4.0 GHz should read.

 

[hawtrawkr]

I just fried a 3.2E! Trashed it.... and that was with water cooling. These things are nasty little beasts and they will rape your mobo. Right now I am working on replacing all the caps on my motherboard so I can get a more stable vcore.

To be honest with you all, I hate these friggin Prescotts. I should have just gotten another Northie, but I figured the deeper pipeline would get me a higher OC. I wound up trashing my CPU in the process.

It is definitely not the PSU that is the problem, it is that these motherboards are NOT made to handle the raw power and leakage that these CPUs drain. I agree with Hawk. I want something that will run stable 24/7. I was hoping to hit 4Ghz atleast, but so far I am unimpressed, in many ways I wish I got a 3.2 Northie instead.

I have a RMA'ed 3.2E coming to me soon. I might even be willing to sell it cheap to someone here if they want it. Brand new unopened 3.2E.

Anyone interested hit me up in a PM. I think I might just stick with the Northies for now, unless my "droop" mod thread gets a good amount of response.

are you sure it was the cpu that was killed and not the mobo?

did you sink all the mosfets up? Im sceptical on how much of a gain replacing your caps would get you stability wise id think you would get more of a gain from sinking the mosfets and possibly adding a fan that blows on the upper part of the mobo where the processors power regulating fets and caps are located.

even with fairly high end watercooling when id run cpumsr it would show my cpu overheating when running dual prime 95s over 3.9ghz and 1.5v (and yes i reset the waterblock a couple times to make sure that wasnt what was wrong.)

 

[Vflux]

I don't know what it is with you Vflux, but you've got me benching like crazy lately
These were done on my secondary rig once again, SL7E5, BH5, ect. Most of my ORB slots are filled up with my new 6800 results. But here are some screenies of mine at "daily settings". Looks like once again the BH5 makes a big difference in most of these benches as we're almost identical on all of them despite 200MHz difference in clock. I am using a 9800XT, however I ran it at default speeds for these tests to make it closer to your system stats. I'm gonna bump my CPU up as high as my BH5 will go sometime this week and see what i can get then. Once again fun to throw our benches out there. And i'm sorry to say, but i'll be moving to AMD next week for the first time, but i'll be keeping my intel around for fun. Not my top benches, but just a comparison.

[/QOUTE]

You havent even vmodded you're board yet and your switching already? I think theres a new S939 board from Epox you might want to check it out. I wont be switching anytime soon.

BTW, my 470W noisetaker crap out on me last night while i was trying to bench at 4.1Ghz. I havent even pushed my 9800 Pro yet. Im going to need a new 600W psu to everything out including the Tornados. Cooling is no longer an issue for me, winter is coming....I have a cheapo 300W here to power some of the drives n fans etc....But im waiting patiently for an OCZ 600W powerstream............I got to get me a 1GB set of Adata/Gskill/OCZ EB for christmas and im done...........

Again thx for having fun and everyone else feel free to include your 4Ghz benchies......Have fun n enjoy your rig before they become xtinct.


Vflux

 

[woutertal]

Im sceptical on how much of a gain replacing your caps would get you stability wise id think you would get more of a gain from sinking the mosfets and possibly adding a fan that blows on the upper part of the mobo where the processors power regulating fets and caps are located.

I disagree completely. A leaky cap can completely ruin the stability of yr pwr to CPU supply. Replacing them is good, BUT you have to find the right, High temp Low ESR, and correct voltage replacements. It is recommendable to add caps to the back of yr mobo, doubling or tripling the capacitance by adding caps of same or double value in parallel. This mod GREATLY reduces the heat in the mobo Vcore pwrplant. Look at my mods in this thread.

 

[Vflux]

I disagree completely. A leaky cap can completely ruin the stability of yr pwr to CPU supply. Replacing them is good, BUT you have to find the right, High temp Low ESR, and correct voltage replacements. It is recommendable to add caps to the back of yr mobo, doubling or tripling the capacitance by adding caps of same or double value in parallel. This mod GREATLY reduces the heat in the mobo Vcore pwrplant. Look at my mods in this thread.

I agree...IMO adding caps paralleled to the ones already onboard help and reduces the tension from the cpu during load. I think it's one of the reason i was able to push my pressy up to 3.9Ghz+ without adding a notch a vcore onto it. Let's just say all the 3Dbenchies i did from previous state didnt need more than 1.43V to get them all running 3D stable, Then psu starts to crap another story, another loot down the drain......Cooling the mosfets help them to output their particular rating at same state.....e.g. PSU. Cooler is always better.......

I'm still wondering how does these pressy run under subzero cooling? Waiting patiently for hawtawkr's vapo results.

BTW, here are a couple of my runs i took a snap on while i was looking for safe clocks from previous testing.........

GL
Vflux

 

[woutertal]

I'm ordering some high power fans for my BIX.. it's running on low noise Papsts. Perhaps I can shave off a few degrees and add another notch of FSB

 

[hawtrawkr]

I disagree completely. A leaky cap can completely ruin the stability of yr pwr to CPU supply. Replacing them is good, BUT you have to find the right, High temp Low ESR, and correct voltage replacements. It is recommendable to add caps to the back of yr mobo, doubling or tripling the capacitance by adding caps of same or double value in parallel. This mod GREATLY reduces the heat in the mobo Vcore pwrplant. Look at my mods in this thread.

this is true if you assume he has a leaky cap in which case he could just rma the board since as he stated he hasnt done a droop mod to it yet.

adding more caps would help to stabalize vcore and reduce the temps but he didnt say he wanted to add more just replace the ones he already had

 

[Vflux]

Next bench: Sucka! Ya listening bro..

Test Bench: HexusPiFast 3.3
Full Modded P4C800-E Dx Rev.2
3.2E@4Ghz 5:4 Cas2.5-3-3-7 BL8 PAM-Standard
Vcore:1.38V
Vdimm:2.75V
Vagp:1.6V
Sindepanel On, Closed case scenario......

It's all in the caps!!!

And!!!
Droop mod!!!

Enjoy,
Vflux

 

[maxwedge]

I'm at 236 fsb and 1.5125v with this cpu on my IC7 and it's been a royal pain in my A!! I keep running into reboot problems! And I don't mean bsod's because my box is set to stop at bsod's but the cpu just reboots. I'm even running 2 pwr supplys....550 antec for the board and 380 antec for everything else. I've sinked the pwm mosfets, added 2 side pannel fans (120mm and 92mm) and put the torronado on the sp94.

I've run P95 23 1/2 hr and then got a reboot. I got 12hr of P95 last night, at 240fsb/1.5375, and then it rebooted. It may work with more volts but the problem is this sucker starts getting really hot. I'm comming to the conclusion that temerture isn't really the issue (reason is below), but it's the fact that the board just can't supply enough juice to power this cpu at high overclocks. Some boards are just better than others (even the same brand and model). I not about to mod my IC7 as I've killed one before, lol!

With my system I really don't feel the 2 power supplys have helped and I got the same problems when I moved this box out to the garage (15F cooler air!). This cpu has passed 12hr's of memtest86+ at 245 fsb 1.4625 vcore, with 100 loops of every test...so it's gotta be a power issue (of the board).

My ram is simpletech pc4000 and I've had it to 276 fsb 1/1 w/no problems(2.8v). So it's not the ram causing my issues.

Best wish's to all of you having success with this chip.
I think I'll prob go with that dfi canterwood socket T board.

 

[woutertal]

Maxwedge what are yr temps?

 

[KRich]

Well guys, I decided to dive right in and try a 3.2E which I ordered on Friday last week and I received today.

It is a SL7E5, Week 22, 2004 Costa Rican chip and I've played with it for an hour or two just to get a feel for it.

I'm running it on an ABIT IC7-G Rev 1.1 BIOS 26 Motherboard, using 1GB (2x512) of Mushkin PC3200 Special DDR RAM, cooling the CPU with a SP-94 w/ Vantec Tornado 92mm and powered by an OCZ Powerstream 520 Watt PSU.

So, first things first, I roughly tried to determine the highest FSB I could get on stock voltage (1.385.. in BIOS) and which tested as Dual Prime stable (Only ran it for an hour to test roughly) and it seems happy at 225FSB on stock - I can get it up past that to near 235FSB and run 3DMark2001 and so on but Prime fails instantly.

I've tried to get it stable at 240FSB with up to 1.5485 V in BIOS but failed so far - It runs Dual Prime 95 alright for 30 mins at 1.54xx V in BIOS or so and then the system freezes - Below that VCore, it fails instantly - Could be temps I reckon - More on that later.

I'm currently Dual Priming as 235FSB with 1.4625V in BIOS which under load is showing as about 1.41V - So far it seems alright (been running nearly an hour but of course I need to let it run for atleast 12 hours to know for sure).

I did try 235FSB at 1.4375V in BIOS and it failed one instance of Dual Prime on 21 minutes so as above, i've upped the VCore to 1.4625V and am Dual Priming right now.

I did get the system to boot up in to XP at a magical 4Ghz with 1.58xxV in BIOS but it wasn't stable and those Mosfets were scorching hot so that is a no go for now.

I guess though that the fact that it does atleast bootup and get into Windows at 4GHz is not a bad sign

Now for temps, here is a comparison for you (I've listed previous temps with the 3GHz Northwood I've just replaced and please note the system setup, cooling and so on is the same in both instances)

P4 3GHz Northwood @ 3.6GHz 1.525V (Dual Priming - 100% load)

CPU: 54 degrees celsius (44 to 47 degrees celsius having deducted the extra that ABIT boards read)

PWM: 36 degrees celsius

P4 3.2GHz Prescott @ 3.6Ghz 1.3875 (Dual Priming - 100% load)

CPU: 59 degrees celsius (49 to 52 degrees celsius having deducted the extra that ABIT boards read)

PWM: 40 degrees celsius

P4 3.2GHz Prescott @ 3.76Ghz 1.4625 (Dual Priming - 100% load)

CPU: 64 degrees celsius (54 to 57 degrees celsius having deducted the extra that ABIT boards read)

PWM: 44 degrees celsius

My board is all stock, no mods at all or anything, case cooling is good as well.

So guys, what do you think? I only use air cooling and I am thinking of trying a XP-120 with a 120MM fan.

So, those are my very very early results with the CPU and I am going to be taking time over the coming days to get proper results and test it thoroughly and will post my observations and experiences here.

Please do note that as I have only tested the chip for a matter of hours, the results might change or differ but i'll be sure to post back when I have more to tell after further testing.

Comments and feedback much appreciated and cheers again guys for all the input and efforts.

 

[maxwedge]

Originally posted by woutertal
Maxwedge what are yr temps?

These temps are at full load running 2 instances of F@H. 70f in my house right now. They'll go up a little with P95 but not much and I've checked all the voltages at the board and there right on, even though their reported low.

You can see the vcore drops a little under load as its set at 1.5125 in bios at this moment.

I think the temps look pretty good but in the summer when it's 95f in here I may have a different story, LOL!

max

 

[Vflux]

Here's my 8M SuperPi run just for kicks...

Maxwedge, You might try looking for the the truepower sense line mod for your 12V/3.3V/5V rail at xtremesystems. You need to up them a bit. BTW use speedfan to measure your voltages. MBM5 isnt accurate. Either it reads it too high or too low. My mm reading is right on the dot with speedfan/asusprobe......


Vflux

 

[maxwedge]

Originally posted by Vflux
Maxwedge, You might try looking for the the truepower sense line mod for your 12V/3.3V/5V rail at xtremesystems. You need to up them a bit. BTW use speedfan to measure your voltages. MBM5 isnt accurate. Either it reads it too high or too low. My mm reading is right on the dot with speedfan/asusprobe......

I only posted that one because it's a small pic. All those programs read the same...speed fan, Hardware doctor, MBM ect. and I've checked those voltages, with a dvom, at the atx pwr header at the board (under load) and their all good. At that point the 12v rail is right on 12.0v even though the software says it's low. I think the sensors on the board arn't calibrated correctly, but????? Maybe if I tweak the pwr supply to bring that up it may help me out.

I'll go look for that mod and thanks.

max

 

[maxwedge]

I guess I'm working on my power supply tomorrow!
Thanks, I'll post back with the outcome.

http://www.xtremesystems.org/forums/showthread.php?t=37574&highlight=truepower+sense+line+mod