Pentium D's Suck for DC

BadThad

Lifer
Feb 22, 2000
12,099
47
91
I put together a Pentium D 915 system awhile ago and it's giving me HELL with R@H. It was running fine for about a week or so, then it would just randomly stop running R@H.

Full Story on Rosetta Forums

My conclusion: The CPU is throttling. There's a known "bug" with BOINC, when throttling occurs, BOINC will completely halt execution of a project.....and it will NOT restart automatically. :|

Intel didn't design the dasm Pentium D to run under 100% load 24/7. What a piece of crap this CPU is unless you use extreme cooling or special case design (like what Dell uses, i.e. BTX).

Now I really understand Core 2 Duo, lol. It's so superior to P4/PD it's not funny.
 

MadAmos

Senior member
Sep 13, 2006
818
0
76
I have been folding 24/7 on both cores of a Pentium D 915 for several months with no mproblems at all. it is noticably slower than my E6400 but it has been capable of 200+ PPD even on the 2124 WUs. What are your system specs: mobo, case, CPU cooler, Etc ?

Amos
 

Philippart

Golden Member
Jul 9, 2006
1,290
0
0
The Pentium D not only gets extremely hot, it also consumes way to much energy/GHz.

(I don't have one, I just read it in reviews)
 

BadThad

Lifer
Feb 22, 2000
12,099
47
91
Originally posted by: MadAmos
I have been folding 24/7 on both cores of a Pentium D 915 for several months with no mproblems at all. it is noticably slower than my E6400 but it has been capable of 200+ PPD even on the 2124 WUs. What are your system specs: mobo, case, CPU cooler, Etc ?

Amos

Intel D945Gpm mobo
Aluminum Case
Stock Cooler
1024 MB of Corsair DDR2-667
ATI X700 Pro 256MB Radeon
Seagate 7200.10 320.0 GB @ 7200 RPMS
Hauppauge PVR-150 TV Tuner
450W OCZ Modstream PS

The PS has a 120mm fan drawing air in, it's right above the CPU. In the case, I have a 60mm rear fan and a 80mm front fan. CPU is running at default speed/voltage.


 

Wiz

Diamond Member
Feb 5, 2000
6,459
16
81
Can't you disable throttling in the bios?
I have two lga775 based P4's and they both run r@h 24/7.
One is stock 3.8ghz and the other is a 3.0 @ 3.4ghz.
I've disabled throttling on both.

These are not "D" chips - so I don't know if you can do the same with them as I am doing with mine.

Edit:

I should have mentioned I have the Arctic Freezer on the 3.0 - temp at 100% with two r@h processes is 75C.
 

BadThad

Lifer
Feb 22, 2000
12,099
47
91
Originally posted by: Wiz
Can't you disable throttling in the bios?
I have two lga775 based P4's and they both run r@h 24/7.
One is stock 3.8ghz and the other is a 3.0 @ 3.4ghz.
I've disabled throttling on both.

These are not "D" chips - so I don't know if you can do the same with them as I am doing with mine.

Edit:

I should have mentioned I have the Arctic Freezer on the 3.0 - temp at 100% with two r@h processes is 75C.

No, the Intel brand BIOSes suck. I've even updated to the latest BIOS, there's hardly any options. Last night I did DL a monitoring/control utility for throttling. I'll check how it did when I get home from work tonight. Sure hope it killed throttling, that's an annoying feature for a DC'er.
 

MadAmos

Senior member
Sep 13, 2006
818
0
76
BradThad
Are you using the stock intel heatsink/fan? Most DC projects do stress the coolers BTU limits and an aftermarket cooler and using a good thermal interface can make a large difference in thermal stability. If so I would try something like a Arctic 7 Pro like this or similar, there are lots of opinions on the best but most of the better ones will remove heat better than OE.
 

BadThad

Lifer
Feb 22, 2000
12,099
47
91
Originally posted by: MadAmos
BradThad
Are you using the stock intel heatsink/fan? Most DC projects do stress the coolers BTU limits and an aftermarket cooler and using a good thermal interface can make a large difference in thermal stability. If so I would try something like a Arctic 7 Pro like this or similar, there are lots of opinions on the best but most of the better ones will remove heat better than OE.

That's the one I just ordered, lol.
 

Wiz

Diamond Member
Feb 5, 2000
6,459
16
81
There's only one PITA about the AF7 - to be aware of:
Those little black feet (connectors) tend to be a bit too fragile, be careful with them.
Also, this thing is very large - hopefully you will have enough room for it.
 

BadThad

Lifer
Feb 22, 2000
12,099
47
91
Originally posted by: Wiz
There's only one PITA about the AF7 - to be aware of:
Those little black feet (connectors) tend to be a bit too fragile, be careful with them.
Also, this thing is very large - hopefully you will have enough room for it.

Thanks for the head's up. If it doesn't work out in the PD system, I can always try it out in my C2D system.
 

apriest

Senior member
Apr 25, 2002
237
0
0
www.aaronpriestphoto.com
I completely agree! I have a number of Pentium D 8xx and 9xx systems running R@H with stock Intel coolers and Intel motherboards. They are stable as can be, but I have to set Rosetta to 50% or the fans go nuts within 10 minutes. They are dang hot!

I've got a couple dual Xeon 5130 servers (based on the Core 2 architecture) and they run full throttle with no cooling issues whatsoever. I'm building a couple Core 2 workstations right now and I'm curious to see how big an improvement it will be over the Pentium D and especially the Pentium 4.

I set Rosetta to 10% on my Pentium M and Core (2) laptops, just to avoid any heat or fan issues.
 

networkman

Lifer
Apr 23, 2000
10,436
1
0
Well that explains the problems I was having with some HP brand P4D boxes. :|

It got to the point where I just gave up and don't even run DC on 'em any more.

 

dawks

Diamond Member
Oct 9, 1999
5,071
2
81
Hmm I am running a Pentium D 920 (2.8Ghz Dual Core) in an HP box (has a huge HSF) and its been running Rosetta@Home 24/7 since May (2006) without problems. I've also been getting about 500-520 PPD from it.

Never had it crash and haven't seen it error out.
 

Assimilator1

Elite Member
Nov 4, 1999
24,125
508
126
Hi BT

What spec is a P4 D 915?

Btw I'd be very wary of disabling throttling without at least upgrading your HSF ,remember they throttle them so they don't cook
Your exhaust fan could do with an upgrade ,say to 92mm, if you can fit it ,that alone can make a big difference.What speed does your 60mm fan spin to?
 

BadThad

Lifer
Feb 22, 2000
12,099
47
91
Originally posted by: networkman
Well that explains the problems I was having with some HP brand P4D boxes. :|

It got to the point where I just gave up and don't even run DC on 'em any more.

Poor design (thermally)! I have a Smithfield PD Dell that runs R@H just fine, I attibute that to the superior BTX case design. Too bad BTX didn't catch on, it's really a much better design than ATX.

This past October I attended a tech conference. Intel guys were there and they officially told everyone that BTX is dead due to "lack of interest".
 

BadThad

Lifer
Feb 22, 2000
12,099
47
91
Originally posted by: Assimilator1
Hi BT

What spec is a P4 D 915?

Btw I'd be very wary of disabling throttling without at least upgrading your HSF ,remember they throttle them so they don't cook
Your exhaust fan could do with an upgrade ,say to 92mm, if you can fit it ,that alone can make a big difference.What speed does your 60mm fan spin to?

The 915 is the Presler core (65nm) at 2.8GHz. I bought this over the Smithfield (90nm) because I figured it'd run a lot cooler. For the case this system is in, I cannot run a larger exhaust fan, it's designed for a 60mm and that's it. The machine got so bad, it has corrupted WinXP and all the WU's returned were garbage (client error message). Either I've completely toasted this CPU or the heat became unbearable for the system. Sure hope Intel will replace it if it's bad, it is a retail boxed CPU.

My plan now for the box is to just remove the mobo/cpu and replace it with a C2D E6300/865 I have laying around. I think I'm just going to either chitcan this combo or try it in a different case with the new cooler I have coming. At that point, it will be nothing more than a full time Rosetta cruncher sitting my basement.
 

Assimilator1

Elite Member
Nov 4, 1999
24,125
508
126
lol ,you have a C2D& mbrd just lying around!??:Q ,got a spare one for me?

I take it Smithfield & Presler are dual core variations of the Prescott core?
 

BadThad

Lifer
Feb 22, 2000
12,099
47
91
Originally posted by: Assimilator1
lol ,you have a C2D& mbrd just lying around!??:Q ,got a spare one for me?

I take it Smithfield & Presler are dual core variations of the Prescott core?

LOL! Well, I did the Intel Retailers Edge program again this year so I had these just sitting here.....sorry, no spares brother, lol.

Yea, it's basically a couple Prescott cores in one......HOT HOT HOT, lmao.
 

Smoke

Distributed Computing Elite Member
Jan 3, 2001
12,650
206
106
I've started to post to this thread several different times and each time I decided against it for a couple of reasons. One, just sort of lazy and didn't feel like composing a lengthy tome. And secondly, I really don't like posting anything that may be taken as a negative toward DC. :frown:

I have noticed over the course of the last couple of years that most (not all) systems that run DC (specifically BOINC) 24/7 tend to start running hotter and hotter. Now I know a big reason for that is the accumulation of dust and dirt. The first thing I always do when confronted with one of these hot running systems is to open them up and give them a thorough cleaning. That means brushing all the tight dirty spaces and fan blades and then giving them a thorough vacuuming and finally a high pressure blow out. This usually results in an immediate improvement but never as much as I would expect. Never do the temps return to the original ranges when the systems were new.

That has led me to believe that the electrical and electronic parts start wearing away over time. I know that aged and loose AC connections in a home are responsible for heated circuits and blown breakers. There must be a similar condition taking place in these computers. This is strictly my opinion and not backed up with any referenced source though I bet if I took the time to look I'd probably find articles stating just this fact. I do recall the topic of bulging capacitors being discussed many times before and heat was often mention as a possible cause.

So the old saw that running DC projects don't shorten the life expectancy of computers (very much) may be an exaggeration. For geeks like us who want a new system every six-months or so this doesn't really cause a problem. But a lot of people (like the "little folks" on TeAm Enterprise) tend to keep their computers for 3 - 5 years and even longer. I have had to shut down BOINC on most of these older systems because they just can't handle the load any more resulting in computer freeze ups, sometimes corrupted operating systems, etc. After I removed BOINC, all of these systems started running just fine and are now providing their owners continue good service. I remind you I'm not trying to knock DC but I'm just reporting my experiences.

I've been keeping up with temperatures of fairly new computers right here in my home and have been surprised to see the quick deterioration. Let me give a few examples:

I have a two-year old Dell SC400 with a 3.2GHz Pentium4 that now (after a thorough cleaning) while running BOINC has a CPU temp that hovers between 72C - 74C. :Q When BOINC is not running this system runs around 47C - 49C. I have moved this machine into my wife's office where she keeps the room temp at 70F (21C). This machine was previously in my work shop where the room temperature was about 3 - 5 degrees F higher. I had to stop running BOINC on this machine in my workshop. I observed CPU temps while running BOINC (in the workshop) between 78C - 82C. I made an interesting observation that I have not seen reported elsewhere. It seems that computers have a spike in CPU temps when BOINC if finishing a WU (or two), transmitting the results, downloading fresh WUs, and is getting started on the new WUs. I have observed this on many different systems. This particular computer (while running in my workshop) would spike up to 85 - 90C and then freeze up and crash. Normally I wouldn't discover the problem until after several days went by. After awhile I gave up. That's when I decided to move it to my wife's office and replace her Dell 430SC, 2.8GHz P4 that I moved on to my grandson. I have just started running BOINC again on this machine on a trial basis. I have it set to use the BOINC Preferences meaning BOINC stops upon any use and does not startup again until 3 minutes of non-use. So the jury is out.

A machine that BadThad is thoroughly familiar is about one-year old. It is a custom built machine with GOOD cooling that is running a PentiumD 2.8GHz on an Intel Motherboard and therefore NOT overclocked. I am actually using it as my primary workstation. Using the BOINC Preferences also I have observed this computer while running BOINC has CPU temps between 63C - 65C. When BOINC is IDLE the CPU temps range between 46C - 48C. The increase in temperatures when running BOINC is amazing!

A third example is a six-month old PentiumD system in a case with VERY GOOD Cooling parameters. This is a 2.66GHz PentiumD that I have overclocked as high as 4.0GHz when it was new. I have it running BOINC 24/7, always on. Over the course of the last few months I have had to slow this computer down more and more as the CPU temps started climbing. It is currently running overclocked at 3.65GHz and its CPU temps range between 62C - 65C. There is a drastic drop in CPU temp when BOINC is turned off. At idle this machine's CPU temp run between 42C - 43C. That is a 20C climb when running BOINC.

Maybe this worsening temperature situation I have observed is unique to Intel Pentium 4/D CPUs? I don't think so because many of the "little folks" computers that I have had to remove BOINC were AMD systems. I never did a thorough examination of the temperatures on those old computers but I bet the worsening temperature situation over time was the culprit there also. This temperature problem might also be the reason Intel has created the newest Core2 Duo type systems to run at lower speeds and also lower power consumption levels. It would be interesting to see if any AMD users have observed similar experiences. Or maybe this is a BOINC issue? Does BOINC use CPU more heavily than other DC Projects?

(If you read all the way through this, you deserve a gold star.) :laugh:
 

BadThad

Lifer
Feb 22, 2000
12,099
47
91
Wow...give me the gold star, lol. I agree with you 100% my friend. Oddly, it doesn't seem to affect every system at the same pace, but my experiences are similar.
 

Smoke

Distributed Computing Elite Member
Jan 3, 2001
12,650
206
106
Just got a suggestion from an old friend. I'm ready for bed right now but it will be my project for tomorrow. I'll give you all an updated report then.

Night All :moon:
 

Smoke

Distributed Computing Elite Member
Jan 3, 2001
12,650
206
106
Tried to go to sleep but I just had to try Freewolf's suggestion.

Jim asked if I had tried cleaning off the Artic Silver and applying a fresh new coat. The thought had crossed my mind but I don't recall ever hearing this was necessary.

So I opened up my overclocked PentiumD 2.66GHz system and did what he suggested.

For ease of reading I will copy what I wrote about this system in my super long post a few messages back:

A third example is a six-month old PentiumD system in a case with VERY GOOD Cooling parameters. This is a 2.66GHz PentiumD that I have overclocked as high as 4.0GHz when it was new. I have it running BOINC 24/7, always on. Over the course of the last few months I have had to slow this computer down more and more as the CPU temps started climbing. It is currently running overclocked at 3.65GHz and its CPU temps range between 62C - 65C. There is a drastic drop in CPU temp when BOINC is turned off. At idle this machine's CPU temp run between 42C - 43C. That is a 20C climb when running BOINC.

I have not changed the FSB so the system is still running at 3.65GHz. With BOINC/Rosetta running the CPU temperature is now holding between 57.5C - 58.5C. That is a drop of 4.5C - 6.5C! Tomorrow I'll start edging the FSB up a little.

Thank you Freewolf! :wine:

So it looks like I may be doing a little refreshing of Artic Silver on quite a few machines and might possibly end up getting a few more systems back in the crunch.
 

BadThad

Lifer
Feb 22, 2000
12,099
47
91
I'm surprized it made that big of a difference. Sure hope it helps!

I've had quite a few OC'ed systems where I had to keep backing down the FSB over the months/years until I was eventually back at stock speed. These systems didn't have any temperature issues, I think it was just general degradation of the mobo components.....I still think most systems gradually go down when pushed to the extreme 24/7/365.
 

Wiz

Diamond Member
Feb 5, 2000
6,459
16
81
I had very similar experiences with putting on new thermal compound, it does seem to lose it's mojo after a couple years.

I thought I had posted the same kind of thing somewhere...

So I went over to the Team Enterprise forum and sure enough, there it was:

Well I thought I would try something before giving up on my Arctic Cooler.
I spent $5.99 & got a tube of AS-5. (Arctic Silver 5)

Here are the numbers before & after:

Before AS-5
6/24/06 14:34hrs
76c load
61C idle
38C mobo

After AS-5
6/24/06 15:47hrs
62C - 65C load
47C - 49C idle
40C mobo

This is with a small OC - 3.0 >> 3.15ghz
I will bump it up a bit & see what happens.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |