• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Official Presler/Cedar Mill Overclocking & Issues Thread

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
proth said:
Intel is gun shy and has been met with failures of Prescott and promises of ultra high clocks. The core of technical staff still believe in ultra high clocks, the management cautious of dissapointing the public again and obsessed with multi-core. This is not my own speculation but comments from a friend or 2 in Oregon. Releasing the new .65um die over 4Mhz does not plug the hole in their roadmap. Nice thread guys, just seen it

EDIT: UFB that guy cracked the lid on the Presler. Now hook up phase right to the die.. :)

Question:
I had a Zalman 9500 in my card and just saw sentinels comment:
"You MUST have adequate cooling (XP120, Ninja Scythe, TT Typhoon and above. Zalman doesnt cut it)"

The Scythe looks really good, but maybe he didn't consider the 9500?.... Any opinions on the BEST HSF out there to date...

thx

Hahah Zalman doesn't cut it, that's almost as funny as the first 2 preslers in his presence have gouged heatspreaders, and their throttling, yet throttle monitor is not showing it. guys are running these @4+ghz on stock cooling.

I'm running the ZALMAN 9500 LED and it's doing a great job cooling this 920@4ghz 1.42v, been runnning 2xPrim95 for damn near 24hr, temps sitting at 59c. The Fatality 7700 is right up there with XP-120, & 9500.
 
30a on the 12v rail?

man i dont wanna sell out 200$ for a new psu...

anyone got a good one that will work thats 150$ or less?(200$ max)
 
remote_username said:
30a on the 12v rail?

man i dont wanna sell out 200$ for a new psu...

anyone got a good one that will work thats 150$ or less?(200$ max)


A little untested, but early reviews are favorable on Sunbeam. You can get a 550w modular with single 12v rail for ~$85.

http://www.newegg.com/Product/Product.asp?Item=N82E16817709003

Some don't like modular plugs because they "theoretically reduce current", but I love modular. For that price, you can silver the contacts. :)

[EDIT] Whoops, it has dual 12v rails. Too bad!
 
Last edited:
OCZ PowerStream 520 cant beat it for dual core with or without SLI
 
Pressler / cedarmill

I have a 631.

How hot is too HOT?
How much vcore is too much vcore?

Folding at 1.38 vcore and cpu temp 54C.

Can I run 1.5 vcore if I water cool it down to 50C? Well I know I can but am I risking trashing the chip. Is spec 1.3 or 1.4 volts +- 5%?
 
yeah, we've had intel folks walking in the door at work with 955XE ES' even before october, looking for boards that will run them.

LOTS of them made the mistake of putting them in P5ND2-SLI's, that of course either didn't post, didnt make it through windows setup, or blew up from sudden Northbridge overheating

~ Gos
 
markodude said:
What do you guys want - 5ghz straight out of the box - seriously if Intel could do that dont you think they would market the chips at that speed? We know the process is doing well since the 960 has been announced at 3.6ghz Dual Core. It is only logical that we will see higher clock frequencies and benchmarks from 9xx than 8xx - and really 9xx is only a stop-gap until conroe.
Basically this is running a pair of 65nm die shrunk Irwindale Xeons on one socket - I almost got the exact same PCMARK score with 9xx I did when I was running Dual Xeons a year ago.....and the 9xx does not cost nearly as much as a pair of 3.6 2Mb Xeons, plus the boards are much cheaper and more overclocker friendly.
Most people have been running single core - dont you think the 9xx will put a lot more strain on your PSU/motherboard/VRMs at the same frequency since it is running 2 cores? Batboy are you considering upgrading your Antec 550w before getting this CPU? Worth considering as I reckon we can pull up to 20A with just the CPU when you get near 5ghz.
And remember 9xx is only officially supported at 1066FSB on the 975x chipset - perhaps there is a reason that we struggle to get the dual cores to such high FSBs, Intel had to do a board spin to support 1066FSB on the dual core even though such a speed was allowed on a single core processor on 955X and 925XE chipset.
Just some thoughts.....perhaps it will take a bit more effort, time, and modifications/upgrades before we see stellar overclocks from this CPU.


I can agree with this reply to a large extent and this is indicative of us being spoiled with respect to Intel single core overclocks. I have a 930D and don't know whether the CPU sucks or the Gigabyte G1975X and PCPOwer&Cooling 850 SSI are unable to meet this chip's power requirements. The ATX 2.01 spec is definately a problem here which why when I opened the Gigabyte box there were plugs in both the 24-pin ATX and 8-pin connectors. If that isn't a hint I don't know what is? It cost money to put little plugs in 24-pin connectors x-many thousands of boards and as cheap as you might think that is there's a reason for every penny spent.

No we don't expect to get 6GHz out of the box, but we are used to exploiting the binning process. I went with Presler for one reason, to have a test sample for my site. I am very dissapointed but it does boot at 300FSB and will run at 4GHz the only problm being stability. because i can run the chip at default Vcore into 15x270FSB and also run the chip at 15x270FSB over-volting to 1.42Vcore and get instablity in both scenarios in the same way, I know it's in the power circuitry.

Responses are right on, each time a new die is released you simply can't expect the binning to be so consistant that there are no failed cores, only locked multipliers from the lowest to highest meaning every low multiplier chip will clock to the highest model's speed and above. Intel has been giving us this on the 90nm process since the first Socket-478 Prescott! Mine overclcked to 4.3GHz and 4.0GHz at deafault. Only problem was (As is happneing now) it cooked mobo CPU power circuitry.

I also suspect there's something amiss when all the ES samples are overclocking like banshees. Then after getting my 930D I found there's no 14X option? How i wished I saved the $75 or so and went with 920D how I wished I listened. If my chip can post at 300FSB then it's not the CPU it's the absence of stable power. What's driving me crazy are the temps, their so dam low so this must be indicative of the CPU being underpowered? I've been relying on water temps which I monitor just before and after the NexXxos XP waterblock I'm using. I'm also using the CORA 642 passive radiator, OASE (made for Alphacool) AP1510 pump with 12V ~ 24V voltage regulator running the pump at 18V. When my water temp at the inlet is 22C the die temp is usally around 30C BIOS F3 reads 25C on F2 BIOS read 5C hehe.
 
remote_username said:
30a on the 12v rail?

man i dont wanna sell out 200$ for a new psu...

anyone got a good one that will work thats 150$ or less?(200$ max)
z0n3 said:
Considering the above linked Silverstone PSU is currently ~10 bucks more at the egg than the current OCZ Powerstream520 (minus rebate), I'd probably just get the OCZ, IMHO.

But, if every penny really counted, I'd consider this Sparkle (Fortron) 550 for ~100 bucks. I've used Fortrons a while ago, in the past, & they've been a pretty decent bang for the buck, but I haven't used this specific one.

Actually, I'd probably just play-it-safe & get the tried-and-true OCZ, since 25 bucks savings on the Fortron is not a huge amount of difference on something as critical as a Dual Core OC'ers PSU, in the big scheme of things. And probably 1/2 the ppl on this forum running Prescotts or 8.xx/9.xx series -D's can attest to their quality & price/performance ratio (compared to the hi-end big boys, like PC P&C, Zippy, etc.).

I'm not saying Silverstone is not good...Oklahoma Wolf knows his PSU stuff...they may indeed be superior to OCZ, for all I know. But I don't. However I do know the OCZ, & it is stellar.

You might want to do a bit more price comparison shopping around on these models ur considering, but a buck-and-a-quarter is a pretty good price on the OCZ, if you don't mind the hassle of mail-in rebates.

If you check my sig rig stats, you'll see how I voted: With my hard earned $$$'s.

JMO

Strat
 
Last edited:
I can't begin to tell you how this all frustrates me. The Intel influenced ATX 2.01 18A limit completely side-tracked the evloution of the PSU. 12V rails were approaching the 50A as they should. I understand three times 22A = 66A yet it's still 22A on the CPU rail and that's not going to cut it for Presler overclocking with motherboard using the incorrect Inductor arramgments (cost conscious) and Multi stage VRMs. Of course amperes alone can't overcome low quality power conversion on mobos, but technology going in the right direction was derailed costing us three years of R&D and lower prices because UL wantes to protect some idiot whom checks his PSU by sticking the Molex in his mouth.

Here's a list both Dual and Single rail PSUs I'd consider by Brand and Amperes. 99% are between $100 and $200 and the "Egg-on-my-face...come back to momma for more power..." SR = Single Rail all others Dual.

A little note on my favorite brands. 1.) PCPowwer but you need the 1kW because it's the only semi-SR design. I have the 850 multi rail and she's struggling with Presler. 2.) Tagan (I believe to be second best on the planet but they just don't seem to want to boost the power) they also design ePower and the company is dedicated to Power Conservation, so your handicapped by their philosophy. They designed and still do every OCZ Powerstream (not modular) 3.) Zippy 700W (or 800W) only but it's almost $300 and impossible to find. 3.a) The upcoming unit whch may compete w/PCPower as #1 is AOpen's new supply which hasn't been reeased yet. 4.) FSP or Fortron 6.) XG's (MGE) Vortec 600W looks very interesting with a 37A 12V rail, oddly their 500W had a 38A rail, but the internals are so small?

1.) Sparkle FSP550-60PLG SR = 36A

2.) Aspire ATX-PFC680 12V1 = 22A, 12V2 = 24A

3.) SilverStone SST-ST56ZF SR = 38A

4.) XG Vortec PSVO-600 SR = 37A

5.) Enermax Liberty ELT620AWT 12V1 = 22A, 12V2 = 22A

6.) FSP (Fortron) FX-700GLN 12V1 =15A ~ 12V4 =15A (60A total)

7.) ePOWER (Tagan design) Silent Engine Tiger 650 EP-650P5-T1 12V1 = 20A, 12V2 = 20A

8.) OCZ PowerStream 520W SR = 33A (Tagan design as I wrote in the very first Powerstream review back in June 2004) THIS is what makes the OCZ's so good.

9.) The highest 12V rails I've found in in this unit but's it's $289 Athena Power AP-P4ATX85FE SR = 48A (850W PSU)

10.) PCPower&Cooling offers an entire list of Refurbished PSU's I just donated a 6-year old Silencer 400W which I got refurbished about 4-years ago, and it's still going strong!
 
Last edited:
Nicely done Liquid3d. Something not obvious to peeps is the dual 4 pin CPU connector common on most mobos can be powered by 2 seperate rails. Having (2) 20A rails is considered worse than a single 34A rail but in fact it's much better because there's a full 40A available for the CPU, 20A for each of the 4 pin CPU connectors. Beware there are cheapo supplies that have the same rail going to both 4 pin CPU connectors albeit 30+ Amps. My OCZ600SLI-ADJ has (2) 20A rails V1 and V2 which both rails blug into the (2) CPU connectors. The problem with this power supply is video shares current with V2, while accessories share power with V1.

What we really need is (5) isolated 12v rails:
V1 @ 20A for CPUa
V2 @ 20A for CPUb
V3 @ 20A for VIDEOa
V4 @ 20A for VIDEOb
V5 @ 20A for CD, Fans, Accessories


LOL...1200w only for the 12v
 
I decided to go back through the list and look for 550W > with Mutiple 12V Rails (a few high current budget units) in addition to the 20 + 4-pin, checking for a 8-pin connector will help .

EDIT: by multiple rails I meant greater then 3 because only these designs may have an isolated 12V for the CPU

1.) SILVERSTONE ZEUS SST-ST65ZF ATX12V 650W +12V1@13A, +12V2@18A, +12V3@16A, +12V4@8A

2.) FSP Group (Fortron Source) FX700-GLN ATX12V/ EPS12V 700W +12V1@15A,+12V2@15A,+12V3@15A,+12V4@15A

3.) FSP Group (Fortron Source) FX600-GLN ATX12V/ EPS12V 600W +12V1@15A,+12V2@15A,+12V3@15A,+12V4@15A

4.) COOLER MASTER Real Power RS-550-ACLY ATX12V/ EPS12V 550W +12V1@18A, +12V2@20A, +12V3@10A

5.) SILVERSTONE SST-ST56ZF ATX12V/ EPS12V 560Watts (SR = 38A 12V) highest 12V rail for $137

6.) ASPIRE Chameleon ATX-AS550W-GN ATX12V v2.03 550W SR = 36A

7.) SinTek 500 & 600 SLI PSUs
(these are absolutely beautiful with relatively conservative power ratings) they have external memory voltage adjustment knob, and PCIex adjustment with knob 34A on the 12v rail. they each have LCD Temp-indicators

So far people have found the OCZ Powerstream 520W to be an excellent solution for Presler Overclocking. What might be useful is to determine what attributes allow this PSU to succeed where others fail?

We know the design is by Tagan, it's a single 12v rail 34A flow rate 24-pin connector and 8-pin connector. Findning a similair desing should help, renmembering PFC may work against us between Presler's demands and current (no pun intended) mobo power conversion circuitry.
 
Last edited:
Multiple rails just made things worse in our testing.
That is why I believe the single rail OCZ520W had no problem, while the twin rail OCZ600 tripped the internal over current.
 
PFC (Power Factor Correction) is another name for "Circuit breaker" and while trying to pinpoint where Presler does the most damage it confounds the issue. With PFC your unsure whether your system just re-booted because of VRM overload, inductor strain, mosfett overheating, or the PSU just shut-off for a millisecond because that's better then the alternative.

PFC (Please Feel-free to Correct (me))

On another note (vent)... what really burns my stove-top pudding, are the superfluous Intel standards. Does Intel sit at the position of Chairman of the board at JEDEC, UL, etc...?

PSU-makers should have been re-designing the moment they knew the ATX 2.01 was slated for re-vision. Design something better, like dedicated, isolated CPU-rail? What's so freaking hard about that? Then place multi-rails for SLI/CF and they can still ride the SLI Marketing wave. Problem is, market PSU sales have increased dramatically sonce SLI and the ATX 2.01 were announced. They've never made so much money as PC-users run frantically from room to room smashing into stupid walls trying find SLI ready units. SLI in itself is a silly iudea since it only doubles card sales while giving consumers half the benefits for twice the price plus the cost of the SLI board and new PSU. the double the cooling as you double the GPU heat etc.

All the while this lethal combination of multiple rails for SLI mated with an anemic 12V rail, is like a bad day on he 405.. As the market continues to be flooded with multicore CPU's and PSU makers riding the profits of the SLI multi rail marketing wave we're all going to empathize with LA commuters. Guess which lane below is closest to that 12V rail?

"Here's your pre-ATX 2.01/EPS system..."

405-105.jpg


"Here's your post-ATX 2.01/EPS system on marketing drugs..."

10-5-101.jpg
 
Last edited:
Liquid3D said:
PFC PSU-makers should have been re-designing the moment they knew the ATX 2.01 was slated for re-vision.


Liquid3D,

I'm in the market for a 600-700W PS, and I don't want to buy something that will become obsolete in 6 months. Can you point me to any well-designed tests that demonstrate that different PS's have an effect on overclocking stability?
 
crotale said:
Multiple rails just made things worse in our testing.
That is why I believe the single rail OCZ520W had no problem, while the twin rail OCZ600 tripped the internal over current.
Crotale..With a comment like that, maybe my thinking is flawed. I know you really know your stuff and have been there. The dual rail hype may be another trick by the supply companies, or maybe small differences between the rails have a negative affect when combining them in the core. Such small differences may be negligable for lightly loaded systems, but amplified during high current loads. Not sure which mobo you used and can't remember if you benched presler which has dual core. Total +12v isolation would be necessary for dual rail to be effective, 2 seperate CPU requlation circuits. My feeling is ASUS simply brought in the 2 rails and tied them together within the power planes of the board with a choke or something. Asus engineering of power systems is very poor in my book.

Liquid3D said:
PFC (Power Factor Correction) is another name for "Circuit breaker" and while trying to pinpoint where Presler does the most damage it confounds the issue. With PFC your unsure whether your system just re-booted because of VRM overload, inductor strain, mosfett overheating, or the PSU just shut-off for a millisecond because that's better then the alternative.
Lq3d. I hate to disagree with people on a public forum, but I believe PFC is compensation for A/C line capacitance. Depending on the A/C transmissions losses and loads within your home, there may be a lag between Voltage and Current rise within the A/C sine wave. The bigger the "lag", the more inneficient A/C powered devices are, generating more heat than if current and voltage followed the exact sine wave profile. PFC compensates for this lag, making the supply more efficient. Quoting this from some stored grey matter that has been clouded by many years of digital ASIC design and coding, and too freakin lazy to look it up. PS: Love your analogy
 
Last edited:
I have no problem being corrected publically, privately so long as I remain open to criticism I can expand my mind. I appreciate your sensitivity however. It's when people insult each other simply because one member has knowledge another does not and uses his wit to berate others I find that offensive, your obviously much more healthy. In fact by correcting me publically you not only help me you help people I have mislead in my analogy.

I've only just started to delve into power supply circuitry and motherboard power conversion circuitry, and everything I know has resulted from reading articles such as PSU basics or searching down articles such as ths on the Intersil core controller. Electronics arefamilair to me from the days I was into gih-end audio tube amplification electrostatic speakers etc. but I discovered my passion for computers through studying Neurophilosophy. The computers is perhaps the best tool neuroscience has for studyig the brain and how we think or represent reality internally.

There's obvious applications such as Parallel Distributed Processing which mimics strength rations between neural nodes or synapses, however; topics such as Intentional Systems (Intentionality) and thought representation to a certain extent can be modleed as software. Most interesting are the metaphysical similarities between the human brain and a computer. Both are binary systems the PC uses 1/0's the brain uses synapes which are either firing or sedentary. Both have long and short term memory, both are dependent upon periperials (eye, ears, etc. On the PC camera, mocrophone, keyboard etc.) to gather information from the outside world yet "represent" (hence the Representation in philsophy) in binary language. This also where the term comes "...the Language of Thought." Both are electrical devices, bother essentially organic (silica).

OK enough off topic. Thnak you for some reason i was associating PFC with an article i red where I thought they claimed PFC was shutting down the PSU, where others were more stable. I now understand a bit better, the quote below alos links to a easy to follow PFC article;

Multiplying a reactive load's RMS voltage and RMS current will give you the circuit's "volt-amps" (VA, which you may remember seeing on the spec sheets for uninterruptible power supplies) rating. This is its apparent power, but not its real power. Power equals VA times power factor, and power factor is the cosine of the phase angle between voltage and current
 
proth said:
Crotale..With a comment like that, maybe my thinking is flawed. I know you really know your stuff and have been there. The dual rail hype may be another trick by the supply companies, or maybe small differences between the rails have a negative affect when combining them in the core. Such small differences may be negligable for lightly loaded systems, but amplified during high current loads. Not sure which mobo you used and can't remember if you benched presler which has dual core. Total +12v isolation would be necessary for dual rail to be effective, 2 seperate CPU requlation circuits. My feeling is ASUS simply brought in the 2 rails and tied them together within the power planes of the board with a choke or something. Asus engineering of power systems is very poor in my book.
I'm far from being an expert at PSUs, and it would be wrong to draw any general conclusions from my experiences.

What happend to us was that when the Presler was heavily overclocked our OCZ600W with dual 12v rails tripped the over current protection (just shut down). This happend at "mild" load and "mild" vcore at around 6GHz. The OCZ520W, on the other hand, had no problems even with heavy load from the CPU at 6.3GHz and an X1900XTX in 3DMark.

Our guess, since we don't know for sure, is that OCZ has split the PCI-express 12v from the mobo + CPU 12v rail and divided the accepted load 50/50 to PCI-e/mobo+cpu. The 12v rail to the mobo became too weak in this case.

However, I do not believe this would be a great problem to less extreme systems, but the Presler do draw a serious amount of current and I recommend known good PSUs.
 
Back