The following is quoted from Wikipedia. Not sure if it helps or not. Basically try HT on and off and see which is best for you.

Intel claims up to a 30% speed improvement compared against an otherwise identical, non-SMT Pentium 4. The performance improvement seen is very application-dependent, however, and some programs actually slow down
slightly when Hyper Threading Technology is turned on. This is due to the replay system of the Pentium 4 tying up valuable execution resources,
thereby starving the other thread. However, any performance degradation is unique to the Pentium 4 (due to various architectural nuances), and is not characteristic of simultaneous multithreading in general.

While this is true for general computing, I have made the following observation about BOINC and HT:

Running SETI@Home as a single CPU may yield an average benchmark score of 2200 points (using a 3GHz CPU as a baseline). Enabling HT does slow down the CPU's performance, by scoring a simply 1400 per virtual core, making the total between the two 2800, meaning more work gets done overall.

If you'd rather use WU time as an example: An average WU may take a 3GHz non-HT CPU about seven and a half hours to complete. With HT enabled, each virtual core may take up to eight and a half hours to complete, but it's getting them done simultaneously (i.e. two every eight and a half hours), whereas it would take the non-HT CPU fifteen hours to do the same amount of work.

The above estimates are not scientific, but have generally proven very accurate and seem to hold true every time (the numbers may be different, but the logic and premise still holds true).

I have two 840ees and I will back your numbers OzzFan1. I even ran all four for CPDN and seasonal.

I've had the same experience once I turned on both processors on my 3.2Ghz HT rig. It almost doubled my throughput.

---

Calm Chaos Forum...Join Calm Chaos Now

The following is quoted from Wikipedia. Not sure if it helps or not. Basically try HT on and off and see which is best for you.

Intel claims up to a 30% speed improvement compared against an otherwise identical, non-SMT Pentium 4. The performance improvement seen is very application-dependent, however, and some programs actually slow down
slightly when Hyper Threading Technology is turned on. This is due to the replay system of the Pentium 4 tying up valuable execution resources,
thereby starving the other thread. However, any performance degradation is unique to the Pentium 4 (due to various architectural nuances), and is not characteristic of simultaneous multithreading in general.

While this is true for general computing, I have made the following observation about BOINC and HT:

Running SETI@Home as a single CPU may yield an average benchmark score of 2200 points (using a 3GHz CPU as a baseline). Enabling HT does slow down the CPU's performance, by scoring a simply 1400 per virtual core, making the total between the two 2800, meaning more work gets done overall.

If you'd rather use WU time as an example: An average WU may take a 3GHz non-HT CPU about seven and a half hours to complete. With HT enabled, each virtual core may take up to eight and a half hours to complete, but it's getting them done simultaneously (i.e. two every eight and a half hours), whereas it would take the non-HT CPU fifteen hours to do the same amount of work.

The above estimates are not scientific, but have generally proven very accurate and seem to hold true every time (the numbers may be different, but the logic and premise still holds true).

I have two 840ees and I will back your numbers OzzFan1. I even ran all four for CPDN and seasonal.

---

And the beat goes on
Sonny and Cher

BOINC Wiki

The following is quoted from Wikipedia. Not sure if it helps or not. Basically try HT on and off and see which is best for you.

Intel claims up to a 30% speed improvement compared against an otherwise identical, non-SMT Pentium 4. The performance improvement seen is very application-dependent, however, and some programs actually slow down
slightly when Hyper Threading Technology is turned on. This is due to the replay system of the Pentium 4 tying up valuable execution resources,
thereby starving the other thread. However, any performance degradation is unique to the Pentium 4 (due to various architectural nuances), and is not characteristic of simultaneous multithreading in general.

While this is true for general computing, I have made the following observation about BOINC and HT:

Running SETI@Home as a single CPU may yield an average benchmark score of 2200 points (using a 3GHz CPU as a baseline). Enabling HT does slow down the CPU's performance, by scoring a simply 1400 per virtual core, making the total between the two 2800, meaning more work gets done overall.

If you'd rather use WU time as an example: An average WU may take a 3GHz non-HT CPU about seven and a half hours to complete. With HT enabled, each virtual core may take up to eight and a half hours to complete, but it's getting them done simultaneously (i.e. two every eight and a half hours), whereas it would take the non-HT CPU fifteen hours to do the same amount of work.

The above estimates are not scientific, but have generally proven very accurate and seem to hold true every time (the numbers may be different, but the logic and premise still holds true).

---

This is a perenial question .IIRC it only helps to the extent of about 20% more speed

---

Old enough to know better(but)still young enough not to care

The following is quoted from Wikipedia. Not sure if it helps or not. Basically try HT on and off and see which is best for you.

Intel claims up to a 30% speed improvement compared against an otherwise identical, non-SMT Pentium 4. The performance improvement seen is very application-dependent, however, and some programs actually slow down
slightly when Hyper Threading Technology is turned on. This is due to the replay system of the Pentium 4 tying up valuable execution resources,
thereby starving the other thread. However, any performance degradation is unique to the Pentium 4 (due to various architectural nuances), and is not characteristic of simultaneous multithreading in general.

---

"none so blind as those who will not see" (John Heywood 1546)

Don't drink water, that stuff rusts pipes!


You are making Proof out of Logic, by just being dubious! {Bluestar to me)

I doubt that hyperthreading in itself doubles throughput. It hasn't in the past; maybe it had increased it by about 25 percent (with Classic). Several months ago I saw that four-threaded processors (before the Ds) were actually doing worse than 2-threaded ones, but this may not be true today.

---

Hi all,

just wanna celebrate the double of my credits in the last 7 days.....

btw i saw in the stats of the computer that the dual xeon with ACTIVATED HT got more credits.....

Recent averagecredit Total
250.36 3,197.21

Why are you hiding your computers from us ?
No way we kan help you or validate your statements - if you are hiding your hosts.

Kiva

---

Greetings from Norway

Crunch3er & AK-V8 Inside

Hi all,

just wanna celebrate the double of my credits in the last 7 days.....

btw i saw in the stats of the computer that the dual xeon with ACTIVATED HT got more credits.....

Recent averagecredit Total
250.36 3,197.21

---