Any thoughts on the new Pentium D processor?

Duel core, not HT. That would be good if they had HT on each core.

My opinion, still to new, not tested on crunching much yet. Slow 2.8 Gig on each core, 3.4 HT may be faster, will have to wait and see when others are using it.

---

Pizza@Home Rays Place Rays place Forums

Any thoughts on the new Pentium D processor?

Duel core, not HT. That would be good if they had HT on each core.

My opinion, still to new, not tested on crunching much yet. Slow 2.8 Gig on each core, 3.4 HT may be faster, will have to wait and see when others are using it.

If each core is running at 2.8, then it'd be about the same as a single-core HT processor at 5.6 GHz, and maybe a little faster.

---

Duel core, not HT. That would be good if they had HT on each core.

Yeah, only the high end 840's have HT and those things cost almost $1000 US. Way to much money for anyone I know.

---

7,049 S@H Classic Credits

Any thoughts on the new Pentium D processor?

Duel core, not HT. That would be good if they had HT on each core.

My opinion, still to new, not tested on crunching much yet. Slow 2.8 Gig on each core, 3.4 HT may be faster, will have to wait and see when others are using it.

If each core is running at 2.8, then it'd be about the same as a single-core HT processor at 5.6 GHz, and maybe a little faster.

i don't know about 5.6 but i would bet the time it takes for each would be down vs 3.4 with HT

---

Well I have (2) Pentium D Systems... a 2.8Ghz and a 3.2Ghz. The 2.8 averages 4011 seconds/core and the 3.2 averages 3419 seconds/core. I Figured this by taking one page of results dropping the obviosuly low values from -9 overflow errors and averaging the rest. Not sure if this is the correct way to do it...

so your slower system puts out a WU every 66 mins? thats a WU every 33 mins... man that is smoken... is anyone else pulling these numbers

---

Any thoughts on the new Pentium D processor?

Duel core, not HT. That would be good if they had HT on each core.

My opinion, still to new, not tested on crunching much yet. Slow 2.8 Gig on each core, 3.4 HT may be faster, will have to wait and see when others are using it.

If each core is running at 2.8, then it'd be about the same as a single-core HT processor at 5.6 GHz, and maybe a little faster.

i don't know about 5.6 but i would bet the time it takes for each would be down vs 3.4 with HT

A hyperthreaded processor runs at the clock rate, and divides the CPU so it looks like two different processors, but it does one job at a time.

A dual core processor has, well, two CPUs in one chip. If each CPU runs at the stated speed and clocks are not "shared" or "divided" then it's twice as fast as one that has to divvy up the clocks.

---

I don't own one, but I thought an HT looked like two processors and would do two independent threads at a time. Thus if you have an HT one should set the parameter in boinc to use two processors.

---

May this Farce be with You

On SETI, my Pentium D 840 dual 3.2 ghz (no HT) averaged the following running mikkyo's optimized SETI app YAOSCW-P-r7.zip.

Number of units averaged = 91
Average time per unit in seconds = 3,022.22 per core / 2 = 1511.11 per CPU.
Average time per unit in minutes = 50m 22.22s per core

You can see the actual results at: http://setiathome.berkeley.edu/results.php?hostid=1322967. Also keep in mind that those numbers are an average. Actual times vary by unit.

For comparision my P4 3.4 ghz running with HT delivers
Average time per unit in seconds = 4,639.94 per thread / 2 = 2319.97 per cpu.

---

Team MacNN - The best Macintosh team ever.

I don't own one, but I thought an HT looked like two processors and would do two independent threads at a time. Thus if you have an HT one should set the parameter in boinc to use two processors.

It looks like two processors, but it is still just one core.

---

Those times sound good, and two at a time is great. Like that one with a unit every 33 Min. between the 3 cores. The 840 HT chip is too expensive for anyone here I think. Wow run two SETI and two CPDN at a time with the HT.

Found a Dell system with the 840 chip for US $895 this week, but not out to get a new one, will be looking for a good used P3 or early P4 at a good price. Have to replace wifes old Pentium 133 with something that will run BOINC. Will be slower than this system but able to do it, a few more credits.

This P4 2.4 runs around 1 HR 40 Min with the optimized SETI App. not to bad.

Ray

---

Pizza@Home Rays Place Rays place Forums

The other question is will a


Intel Pentium D 830 (3.0Ghz/Smithfield/2x1MB/800FSB/EMT64/XD/Dual Core)

be equal to

Intel Pentium 4 531 64bit(3.0Ghz/PRESCOTT/1MB/800 FSB/0.09u/LGA775/HT/EM64T)

in performance.

Both will use the 32bit windows platform. and the 531 will have HT enabled.

Therefore both units will run 2 work units at a time.

Can someone post some results if they use these types of CPU's.

---

You can not beat them; P4 840ee 3.2 HT does 4 Seti optimized WUs in about 1 1/2 hrs, as I have two boxes, I do 8 WUs every 1 1/2 hrs.....

The other question is will a
Intel Pentium D 830 (3.0Ghz/Smithfield/2x1MB/800FSB/EMT64/XD/Dual Core)

be equal to

Intel Pentium 4 531 64bit(3.0Ghz/PRESCOTT/1MB/800 FSB/0.09u/LGA775/HT/EM64T)

in performance.

The Pentium D830 will do much better then be equal to. Look at my numbers for the D 840 3.2 ghz posted earlier versus a P4 3.4 ghz. You can extrapolate out from there. The D830 will be substantially faster overall.

---

Team MacNN - The best Macintosh team ever.

You can not beat them;

Unless you go AMD...
:-)

---

Grant
Darwin NT

A hyperthreaded processor runs at the clock rate, and divides the CPU so it looks like two different processors, but it does one job at a time.

A dual core processor has, well, two CPUs in one chip. If each CPU runs at the stated speed and clocks are not "shared" or "divided" then it's twice as fast as one that has to divvy up the clocks.

The new Pentium dual cores still share a memory bus and a few other components (Intel's design is not quite as elegant as AMDs dual core/HyperTransport implementation) so there is still an issue with contention and each core fighting the other for memory access.
Potentially they could be as fast as 2 seperate CPUs, but when running independent tasks in the real world this is rarely the case.

---

so your slower system puts out a WU every 66 mins? thats a WU every 33 mins... man that is smoken... is anyone else pulling these numbers

=========

It's also confusing what you said too, (puts out a WU every 66 mins? thats a WU every 33 mins...) but I think I know what you meant to say ... ;)

Okay, now that I have gotten some work units uploaded from my AMD 64 4200+ dual core I can compare it against my Pentium D 840. Here are the results.

AMD 64 4200+ 4089.55 seconds per work unit average per core.
68.16 minutes or 42.25 work units per day with both cores running.

Pentium D 840 3378.40 seconds per work unit average per core.
56.30 minutes or 51.15 work units per day with both cores running.

Both are respectable. However, on Einstein which favors the AMD the tables are reversed.

AMD 64 4200+ 20498 seconds or 8.44 work units per day.

Pentium D 840 26513 seconds or 6.52 work units per day.

---

Team MacNN - The best Macintosh team ever.

Okay, now that I have gotten some work units uploaded from my AMD 64 4200+ dual core I can compare it against my Pentium D 840. Here are the results.
AMD 64 4200+ 4089.55 seconds per work unit average per core.
68.16 minutes or 42.25 work units per day with both cores running.
Pentium D 840 3378.40 seconds per work unit average per core.
56.30 minutes or 51.15 work units per day with both cores running.

On the Pentium is that with HT enabled or without? Toms Hardware said you can make that chip think it actually has 4 processors, don't know about the AMD.

---

On the Pentium is that with HT enabled or without? Toms Hardware said you can make that chip think it actually has 4 processors, don't know about the AMD.

There are two different Pentium D840 chips. The one Tom's Hardware was referencing is the Pentium D 840EE. The only difference is the EE has HT. Mine doesn't have HT. The D 840EE might show a 10-15% improvement from the HT. Remember that it is still just 2 3.2ghz cores with 1 meg of L2 cache per core. The HT will make it a little more effecient. The AMD's do not have Hyper Threading as that belongs only to Intel.

---

Team MacNN - The best Macintosh team ever.