SCALDED PENTIUM D

[denster]

D stands for DOG

That is how she is running. Like flickin the Light on the Cockroaches in the Kitchen 8O

http://valid.x86-secret.com/show_oc.php?id=79287


SUPER Pi Times:
00.344s 16K
00.688s 32K
01.500s 64K
03.234s 128K
07.016s 256K
16.250s 512K
36.500s 1M
1m 26.375s 2M

 

[denster]

I think either the multiplier is locked on the chip or else the option is not available. Let me double check to be sure.

 

[denster]

No option to change multiplier in BIOS. Any other options? I appreciate your help

 

[denster]

You could go higher if you had Asus P5WD2-E instead and you'll also be Conroe ready.

How do you know this? Did they disclose some new info because I dd not read anything about socket compatibility.

 

[ltcommander_data]

Older i975X chipsets may support Conroe on a case by case basis depending on what VRM the motherboard manufacturer chose to put in it. I think Intel said at the recent IDF that i975Xs produced starting in two weeks time should have guaranteed Conroe compatibility.

 

[ltcommander_data]

Finally. After looking through more than 4 different websites and 20 articles I've found the reference. I was beginning to think I hallucinated it.

http://www.hexus.net/content/item.php?item=4843

The Conroe was running at 1067FSB, but the memory was running at 667 (not at 800Mhz which it will be when it launches). The mainboard used in the Intel system was their D975XBK modified (3 Transistors and a trace cut). The boards which will be shipping in the next 2 weeks should all be Conroe compliant.

I suppose the quote isn't as strong as it could be, but this was the information Hexus provided with their 2.67GHz Conroe vs. 2.8GHz FX-60 benchmarks. I suppose they probably asked the on-site Intel spokespeople or it was in a press kit.

Among all the motherboard makers it seems Albatron is the first to claim Conroe support.

http://www.theinquirer.net/?article=30298

Here's more on the same board. Supposed to be one heck of an overclocker.

http://www.xbitlabs.com/articles/editorial/display/cebit2006-3_5.html

Funny that Albatron also appears to be either one of a few or the only one to demonstrate Broadwater boards as well. I wonder why ASUS and Gigabyte are lagging since they are usually early adopters.

http://www.theinquirer.net/?article=30253

 

[TabrisDarkPeace]

SuperPI only runs on one core right ?
The version I have 1.1e doesn't just only run on one core, the entire application is just 1 thread (not to be confused with isolated threads).

You forgot to add the 1 minute (60 sec) to your 2M score btw

Does the CPU temp get over 70 C ?

Is it throttling as the test progresses ?

There is a tool that can detect CPU thottling, but the name escapes me atm.

Bear in mind you are using a Pentium D 800 series on the 90nm die, which goes to FSB just to keep caches in sync, etc (All intercore communication on the 800 series is treated as 'external' and goes to FSB.... rather wasteful, but it still works).

If you had a 900 series on the 65nm die you could go much higher.

EDITED: "ltcommander_data" reminded me the cores on the 900 series still go to FSB.... Wikipedia confirmed this. (I was 'wishing' they intercommunicated at full processor speed, across a wide bus, ... they don't)

 

[ltcommander_data]

If you had a 900 series on the 65nm die you could go much higher, and the cores can communicate to each other directly, instead of going to FSB.

I don't believe that is correct. The 9xx series is even more separated than the 8xx because while the 8xx was a single die, the 9xx are actually two completely separate dies. There is no direct core-to-core communication and everthing goes through the FSB just like the 8xx. Any performance increase is simply due to the larger L2 cache relieving some of the bandwidth starvation.

In any case, cache-coherency traffic between two cores isn't very significant. I believe the effect is proportional to the square of the number of cores so it really becomes a problem once you get to four cores and higher. That said, because the 800MHz bus would be so clogged under full load, even limited cache-coherency traffic would have a detrimental effect on performance.

 

[TabrisDarkPeace]

Yes, it does.

I foudn it on my HDD, but figured if he searched for it, he'd find it too.

I sold all my Pentium 4 systems to build my current rig before they devalued too much..... so haven't used it for awhile.

 

[denster]

What about this guys???

http://valid.x86-secret.com/show_oc.php?id=79762

Right now Prime 95 has been running for 1.5 hrs... My cpu temp is flexing 62c...

800FSB with a 533Mhz chip. I shouldn't let it run like this forever should I?

I had to raise my VCore to nearly 1.6V which is 200mV above spec max.

Please tell me if you know of any danger.

How long does one run Prime 95 before they can consider a system is "STABLE"???

 

[denster]

You forgot to add the 1 minute (60 sec) to your 2M score btw

Thanks, added! Sorry

Does the CPU temp get over 70 C ?

No. But about 62-65c while running Prime 95 @ 4Ghz
Actually at 3.73Ghz temps are under 50c with light load/Idle

How far/long over 70c would be too much???

Do you need that much speed?
If you do video encoding then faster is always better. For games, above 60FPS will do and that depends more on your gfx card than CPU.

No I don't need that much speed I just wanted to see how far this thing would go! And I am using a Liquidcooler not an HSF on the Pentium D...

And at Idle with STOCK settings, my Cpu Temp is around 32-35c But of course Oc is a whole other ball game.

 

[denster]

Question Wusy: What are your system specs??? Just curious about your signature... What does that mean Prescott=Failure, ASUS A8N Deluxe = Prescott.... ???

(sorry, I am like very technically uninformed)

The reason I ask, well, I'll get to that after you answer

 

[denster]

And another example of me having "Just enough knowledge to be dangerous"...

I have had an ASUS A8N-SLI Prem. for about 6 months and running the AMD X2 4400+ in it. And I have noticed that it will just crash in windows sometimes for no apparent reason. Now usually this doesn't happen during a game or movie, can't ever remember it during a movie, but sometimes when I am just opening from the start menu or whatever, usually in just casual usage, it will just lock up, and no response, no task manager, nothing. No way out but the reset button.

What really sucks is I paid close to $700 for just those 2 items, and My Evga Board and Pentium D 805 costs less than $250, and they have never locked up like that... Anybody have any ideas what is going on???

 

[illicitsc]

my 2.6ghz A64 locks up while im playing wc3...thats like the game with the lightest load EVER... and in cs:s it never loscks up. neither in any benchmark. so wat givs.?

 

[denster]

ASUS A8N Deluxe is a failure because of it's low FSB(HTT) potential, bad voltage fluctuation, 9000RPM chipset fan which fails often, overpriced... etc.

OK. So then it stands to reason that the A8N-SLI Premium is also a failure then? Which may explain why my comp just locks up everynow and then... I had always suspected the board. I do have a Gigabyte board I could try my 4400 in, but I say it is the board and not the processor.

Which is my point in asking you about your signature, perhaps maybe you knew something to confirm my assumption... The premium is just a deluxe wuith an extra pci slot and their "fanless cooling pipe" and maybe the extra gigalan, and usb, 1394 port, but basically the board is the same design huh? Guess in my lack of knowledge I made a poor choice.

And I'll never buy an overpriced CPU again, not after what I see this 805 can do!

 

[denster]

my 2.6ghz A64 locks up while im playing wc3...thats like the game with the lightest load EVER... and in cs:s it never loscks up. neither in any benchmark. so wat givs.?

Well and see here. You are using the A8N-E which is also very similar in structure to the A8N-SLI boards.

So, is it the boards, the processors, or a combination of the 2???

BTW, one day I used the support cd to reflash my BIOS and the dang thing flashed my BIOS to that of the A8N-E!!! And of course I couldn't flash it back to the right one after that. I was pissed. ASUS is something else. I know some people wouldn't agree with me, but are they being honost??? I have suspected my MB from the first time I noticed strange behavoir.

I bought an EVGA board just recently but I think I may have done something to it or else it went bad, so I am RMA'ing it now.

I have an AMD Athlon 3200+ in a gigabyte board in another system in my living room, and it has never locked up once...

Anybody elses thoughts or wisdom/insight on this???

 

[denster]

Nope, ASUS realised the mistake with Deluxe so the Premium was the total fix up for it along with new features like no SLI switch board.

Oh. So I am barking up the wrong tree then...

 

[denster]

So Wusy: Can you tell me, what do you think the best board would be for me to use my X2 4400+ in?


I don't have to have sli, but I am thinking of buying another gfx card to match my 6800 but my other 2 boards are also sli so I could use one of them for gaming.

I just need a good sound board to run my 4400 in.

 

[denster]

I guess I still suspect my MB. But perhaps it is something else causing the lock ups. They are not frequent, but once in a great while, it crashes.

 

[denster]

I think if you remove the 2x512MB pair and just operate with 2x1GB, you might be able to squeeze more Mhz out of RAM using 2.9V

Ok Thanks But I have a question about raising my RAM voltage above spec. the RAM I am using is rated at 2.75V Is it OK to go 2.9V with it? How far should I NOT go????

 

[denster]

You should not go above 3.2V which happens to be the vDIMM limit of your board.
Running 3.2V on 24-7 would be fine for your RAM.

Perhaps I could get a better OC out of my X2 then huh? Maybe it is the RAM that is hindering me not my VCore.

Thanks for the help man

And nice getups you got. Funny and sad to say but everyones treasure quickly becomes an antique in this business.