is the E0 prescott worth waiting for?

[brad84cnc]

Also when are your guesses about when the will become available to buy?

 

[klah]

last week

 

[Gnoad]

No

 

[THUGSROOK]

Originally posted by: Gnoad
No

 

[Zebo]

Originally posted by: THUGSROOK

Originally posted by: Gnoad
No

 

[kmmatney]

Link

The thing runs at 77C in an open case!

 

[brad84cnc]

thanks for your input. i forgot to add that i was mostly concerned with the 530 and 540 versions of the prescotts.

 

[Zebo]

Do yurself a favor and get a northwood 3.x, I just read that article kmmatney posted and combined with being slower clock for clock it won't even run it rated speed. Probably negating any Mhz advantage and save yourself a bit of cash.

 

[Auric]

E0 is a stepping for the whole range. While the 3.8 may have issues with the basic Intel cooler, the lower wattages should be a boon for o'erclocking the 2.8, 3.0, etc. variants with a better cooler.

 

[AkumaX]

Originally posted by: Zebo

Originally posted by: THUGSROOK

Originally posted by: Gnoad
No

 

[LTC8K6]

Check out XBIT and Firingsquad reviews of the 3.8. They both got it to 4.2 on air stable.

 

[AristoV300]

I would like to see how the S478 E0 chips do for overclocking. I am very pleased with my D0 3.2E, it runs cooler than my previous 3.0C, and will hit 3.8 on stock vcore:thumbsup:

 

[RalfHutter]

FWIW, I've been playing with a 570J (3.8 EO) for the past few weeks. It's been running on a Soltek 915GPRO board . I'm using a Thermalright XP-120 with a 120mm Panaflo L1A running at 5 volts. Temps sensors have been calibrated using this method.

At an ambient temp of 21°C, idle temps are in the very low 40°C range. Load temps (2xCPUBurn) are in the low 60°C range. 2xPrime95 gives me load temps in the high 50°C range. That doesn't seem too bad to me, especially with a very low CFM cooling fan, albeit on a kick-butt heatsink.

 

[Vee]

Originally posted by: kmmatney
Link

The thing runs at 77C in an open case!

Both TM1 and TM2 throttling features are highly undesirable and bad things, from a user stand.

Throttling just hides a heat problem, with poor performance. And in the case of TM1, you don't know what's going on either.

TM1 is maybe responsible for the widespread erroneous belief that Intel is less plagued by heat than AMD.
I prefer the AMD way. If there is a problem, you get a firm feedback. And if it seems to work, it actually does work for real, all the time.

TM1 is also the main reason I got so terribly pissed off by the Pentium 4. The P4 does a lot of things worse than the old PIII (per clock) or AMD. But I have lately realized that throttling was my greatest problem.

E0 now goes another disgusting step down this road. Buy a 3.8GHz PC for lot's of money, and it may never really execute heavy stuff better than a 3.2GHz. (which in turn, for many things, due to FSB/clock ratio and prescott pipe, is not really better than the old 2.8P4C.) - Great, eh?

Intel are NOT solving their problems. Only hiding them from clueless customers, or helping PC builders hiding them. Disgusting!
And yeah, I know. Put a big enough heatsink on. But you know, all that throttling is intended for ready made brand name PCs.

Both the old 2.8P4C and 3.06P4@533FSB were pretty great CPU's. I can't believe the years have passed, and Intel have not done ANYTHING yet, that is actually significantly better for the desktop.

 

[THUGSROOK]

Originally posted by: Auric
E0 is a stepping for the whole range. While the 3.8 may have issues with the basic Intel cooler, the lower wattages should be a boon for o'erclocking the 2.8, 3.0, etc. variants with a better cooler.

i wouldnt care how high it clocks if its gonna run 60*C+ (let alone 70*C+) !!!!!

 

[LTC8K6]

Ralfhutter, would you D/L Throttlewatch and tell us what it says about your CPU? Even though you don't have a stock cooler, maybe we'll learn something.

 

[LTC8K6]

TM1 is maybe responsible for the widespread erroneous belief that Intel is less plagued by heat than AMD.

What cave have you been living in?

The widespread belief is the other way around.

 

[Vee]

Originally posted by: LTC8K6

TM1 is maybe responsible for the widespread erroneous belief that Intel is less plagued by heat than AMD.

What cave have you been living in?

The widespread belief is the other way around.

Well. The cave you have been living in, is maybe hardware techsite' forums?

Seriously, most IT posers coming by my place, seeing my A64s chugging silently, cooly, flawlessly along, usually remarks that they personally don't like AMD, since they're so hot and unstable and software incompatibe and whatnot...

 

[DanRydell]

Vee is right. And I'm sure LTC8K6 meant the opposite of what he said. I had to read it twice to get it straight.

 

[RalfHutter]

Originally posted by: LTC8K6
Ralfhutter, would you D/L Throttlewatch and tell us what it says about your CPU? Even though you don't have a stock cooler, maybe we'll learn something.

OK, I got Throttlewatch and ran 2xCPUBurn for an hour. At a max temp of 63°C I had zero throttling.

Then I tried a different experiment. I replaced the 5V Panaflo L1A with a 120mm Globe fan running on a Zalman Fanmate so I could adjust the fan speed in small amounts. I again ran 2xCPUBurn while monitoring the Throtlewatch application. I adjusted the fan speed of the Globe fan to give higher load temps. And wouldn't you know it, when the load temps approached 70°C the CPU started throttling!. It seems to start around 68°C and if I tried running it up around 75°C I was getting 75%+ throttling.

So it seems like my little test backs up the throttling results published in the Sudhian article, FWIW. I guess the lesson is: "If you're using an EO Prescott, keep it's temp under about 65°C to prevent throttling".

 

[Zebo]

So in other words at stock you ran a $50 HS with a $19 a 120mm fan with a deep throw, which growns and you did'nt trottle. Great.

Try Prime 95, torture and blend, cpu burn is'nt near as intense.

 

[AdamK47]

I'm interested to see how the 478 pin E0 steppings overclock. I have an external water cooler with a secondary internal radiator, so heat and noise are not a problem with most any reasonable voltage.

 

[RalfHutter]

Originally posted by: Zebo

Try Prime 95, torture and blend, cpu burn is'nt near as intense.

FWIW, 2xPrime95 gives me full load temps about 3°C cooler than 2xCPUBurn.

 

[Zebo]

Originally posted by: RalfHutter

Originally posted by: Zebo

Try Prime 95, torture and blend, cpu burn is'nt near as intense.

FWIW, 2xPrime95 gives me full load temps about 3°C cooler than 2xCPUBurn.

That's interesting I experiance just the opposite.

Arn't you a silent PC guy? WTF you using a processor that uses twice the power, has twice the leakage as the AMD counter parts? At least get a dothan with DFI board if you just tilt twards Intel. (nothing wrong with that, they are a repected company)

 

[RalfHutter]

Originally posted by: Zebo


Arn't you a silent PC guy? WTF you using a processor that uses twice the power, has twice the leakage as the AMD counter parts?

It was sent to me as a review sample.

Originally posted by: ZeboAt least get a dothan with DFI board if you just tilt twards Intel. (nothing wrong with that, they are a repected company)

I've got a 2.0 Dothan ("Intel 755") plus the AOpen and DFI 855 boards that I'm playing with. This is much more up my alley, but it's still fun playing with the real fast, real hot and real inefficient CPUs too!