Socket 939 Sempron found........

[Keysplayr]

Here

It would seem from these early results that the Pentium system only runs on average 10C warmer than the AMD system. Note the fan speeds as well. Intel fan rotation speed it max of 3590 and AMD fan rotation speed is minimum 4441rpm and averages 5114 and maxxed out at over 6000 rpm. must be pretty loud.

Discreditors: (Those that say "Pffft!!" )

irwincur: "THG - Tom's "Intel Supplied and Paid" Hardware Guide"
****Believes THG has been paid off by Intel****

Amplifier: "Seriously"
****Agrees with irwincur****

Lonyo: "For an 850W PSU with 4 12V rails, one of them sucks, since it's at 11.6V for the AMD system, while then Intel system has 12V @ 12V. "
****Believes the AMD system's PSU must be borked****

brandon: "Smell like BS to me. Since when is the Pentium EE 9x faster than the 4800+??? "
****Either believes the test is Bullsh!t, or does not understand about HT****

Duvie: "AMD fan is substandard albeit stock. AMD PSU is reporting low 12V rail voltage.
Task Manager should be running and also Throttlewatch"
****Cheesy AMD fan, Bad/faulty PSU on AMD, Task Manager & Throttlewatch****

PetNorth: "We can't tell if this thing is throttling or not."
****Should run throttlewatch****

Zebo: " The Intel system crashed. The uptime hours dropped back down and still says 1 Reboot"
****Something is borked there. Software not reporting correctly****



Just keeping a little record in case Anand wanted to run the same kind of testing. At least he will know what to look out for and what he will be up against. Very Political.....

 

[irwincur]

I think the site doing the testing really needs to be taken into account.

THG - Tom's "Intel Supplied and Paid" Hardware Guide

 

[Keysplayr]

Originally posted by: irwincur
I think the site doing the testing really needs to be taken into account.

THG - Tom's "Intel Supplied and Paid" Hardware Guide

Don't even go there. The specs for each system are clearly listed. Unless somebody outright proves that THG is intentionally showing Intel in a brighter light over AMD, don't go there.

And more importantly, why doesn't ANANDTECH do something like this?

 

[Amplifier]

Originally posted by: irwincur
I think the site doing the testing really needs to be taken into account.

THG - Tom's "Intel Supplied and Paid" Hardware Guide

Seriously.

 

[ZobarStyl]

Well, even if Tom's leans towards Intel, this test seems fair (for anyone who wants to do 4 things at once)...and the 4800+ seems to be winning at everything except the video encoding, but it looks to be the 4 total threads on the EE are helping that...being that they are tied on the CD's, AMD has a lead on the WinRAR and has a 50% higher framerate (24 vs 37) in FarCry. EDIT: It just jumped back up to 36 on the EE, methinks it's fast but has trouble with FPS drops.

 

[Keysplayr]

Originally posted by: Amplifier

Originally posted by: irwincur
I think the site doing the testing really needs to be taken into account.

THG - Tom's "Intel Supplied and Paid" Hardware Guide

Seriously.

Ignored.

 

[Lonyo]

For an 850W PSU with 4 12V rails, one of them sucks, since it's at 11.6V for the AMD system, while then Intel system has 12V @ 12V.

 

[imported_brandon]

Smells like BS to.

Since when is the Pentium EE 9x faster than the 4800+???

 

[Keysplayr]

Originally posted by: brandon
Smells like BS to.

Since when is the Pentium EE 9x faster than the 4800+???

Where the heck did you get that from?

 

[imported_brandon]

I may have misread it, but, it still seems like BS to me...

Encoded Time on Intel System: 100 Minutes
Encoded Time on AMD System: 10 minutes

Make that 10x faster, unless I read it wrong.

 

[rise]

cool link, thanks keys.

 

[Keysplayr]

Originally posted by: brandon
I may have misread it, but, it still seems like BS to me...

Encoded Time on Intel System: 100 Minutes
Encoded Time on AMD System: 10 minutes

Make that 10x faster, unless I read it wrong.

No, you read it right. It is due to the extra threads the HT Intel chip can run. The X2 can only run a maximum of 2 threads at any given time. The Pentium 840EE can run up to 4 at any given time. The X2 has its hands full and cannot address the Divx encoding thread nearly as often as the intel chip can. Unless I am mistaken or if someone can explain it better.

 

[Keysplayr]

Originally posted by: rise4310
cool link, thanks keys.

My Pleasure brudda.

 

[Duvie]

ONe does in fact need to question Toms...Unless you have forgotten these morons a few months back ran a test that was badly implemented and they had to correct many things. Same type of test and they really had very little clue what the heck they were doing then. I think anybody can question them. I wouldn't question on them being biased as much as incompetent.

The fact is their last test showed little 10c difference yet most ppl in the real world were seeming to have a bit more like 14-15c with same type of HSF's.....

He probably has the jist, but one needs to realize this is the bottom line speed of single core chips (600mhz lower), whereas AMDs dual core offering is only 200mhz below its flagship chip.

Anyway you look at it 10c is a heck of a lot, especially when you need OCing headroom....

 

[Keysplayr]

Originally posted by: Duvie
ONe does in fact need to question Toms...Unless you have forgotten these morons a few months back ran a test that was badly implemented and they had to correct many things. Same type of test and they really had very little clue what the heck they were doing then. I think anybody can question them. I wouldn't question on them being biased as much as incompetent.

The fact is their last test showed little 10c difference yet most ppl in the real world were seeming to have a bit more like 14-15c with same type of HSF's.....

He probably has the jist, but one needs to realize this is the bottom line speed of single core chips (600mhz lower), whereas AMDs dual core offering is only 200mhz below its flagship chip.

Anyway you look at it 10c is a heck of a lot, especially when you need OCing headroom....

Who needs OCing headroom? Everyone? 10C is a lot? Not thinkin so. Its actually less than 10C quite often. goes down to 7C difference sometimes. So maybe the Pentium does consume a lot more power than the AMD, but its not that much hotter. At least nowhere near as hot as AMDwees make it out to be. And when you say real world. Are you meaning that THG exists in an imaginary world? Are they not in the real world?

It appears to me that they went to extra lengths to utilize the closest testbeds they could find. Both are GB NF4 SLI boards. How much closer can you get cross platform?

I mean, I'm really trying to maintain a neutral attitude here but you guys really make it difficult sometimes to do this.

 

[Duvie]

OK had a chance to check it out....

AMD vs Intel things...

1) The fan is substandard and not even a 80mm fan. High rpms suggest that and I bet it is like a standard cooler. That is fine since he is comparing retail models....A cheap 20 dollar 80mm fan will do a bit better IMO...Intels lower rpm suggest a higher cfm fan....AMD needs to go to higher cfm, larger mm size fans cause 5000rpms is a bit to whiny for most....

2) The power supply is weak....How come the amd system has a power supply reporting so weak on the 12v rails...I would question any reboots with reading like this. f he has checked them to be bogus then stop showing them but any person with half a brain would be worried about a 11.46v on the 12v....Intel chip is 11.99 or higher the whole time....



Unless he shows me a cpu task manager showing all 4 cpus loaded continuously I may question if this is actually happening....

I also think Intel cpus should run with throttlewatch so we can see nothing is happening 71c max id quite high IMO....


Also does his 1 reboot mean both systems restarted on its own or his initial startup??? I read where that would be a correct assumption but they crashed at near identical times then based on online time.....

 

[Duvie]

Originally posted by: keysplayr2003

Originally posted by: Duvie
ONe does in fact need to question Toms...Unless you have forgotten these morons a few months back ran a test that was badly implemented and they had to correct many things. Same type of test and they really had very little clue what the heck they were doing then. I think anybody can question them. I wouldn't question on them being biased as much as incompetent.

The fact is their last test showed little 10c difference yet most ppl in the real world were seeming to have a bit more like 14-15c with same type of HSF's.....

He probably has the jist, but one needs to realize this is the bottom line speed of single core chips (600mhz lower), whereas AMDs dual core offering is only 200mhz below its flagship chip.

Anyway you look at it 10c is a heck of a lot, especially when you need OCing headroom....

Who needs OCing headroom? Everyone? 10C is a lot? Not thinkin so. Its actually less than 10C quite often. goes down to 7C difference sometimes. So maybe the Pentium does consume a lot more power than the AMD, but its not that much hotter. At least nowhere near as hot as AMDwees make it out to be. And when you say real world. Are you meaning that THG exists in an imaginary world? Are they not in the real world?

It appears to me that they went to extra lengths to utilize the closest testbeds they could find. Both are GB NF4 SLI boards. How much closer can you get cross platform?

I mean, I'm really trying to maintain a neutral attitude here but you guys really make it difficult sometimes to do this.

Anwering bold in order..

With that fluctuation I am thinking that all 4 cpus may not be staying loaded constantly. I would like to see a cpu task manager for both and also have throttlewatch running...In the past ppl with even 6x series chips have said the P4 will throttle before 70c....I wonder if it hits 70c and throttles down until it cools to like 68c and then heats up again and thorttles again. I would like to see more proof this may not be happening. It seems like a logical request.


NO...I am saying some tried to dup his last test using retail models and fans and quite larger differences between same 2 cpus....One can question anything when ppl cannot duplicate...that is all I am saying....


Finally I have never heard an AMD chip going up at 60c yet I have heard Intel chips throttling at 70c so trust me 10c can make a huge difference....

It does show you one thing though...The first stress test was between 2 of the more flagship chips of Intel and AMD and the difference was similar to this...Now we have 2 cpus at different ends of each companies current flagship speeds yet the results are sill similar. It definitely shows that 3.2ghz may be all they could do to keep it in line.


AMD just needs to sell it with a better (80mm or even 92mm zalman style cpu and they still have ample room to grow with the X2s.....

 

[Keysplayr]

Originally posted by: Duvie
OK had a chance to check it out....

AMD vs Intel things...

1) The fan is substandard and not even a 80mm fan. High rpms suggest that and I bet it is like a standard cooler. That is fine since he is comparing retail models....A cheap 20 dollar 80mm fan will do a bit better IMO...Intels lower rpm suggest a higher cfm fan....AMD needs to go to higher cfm, larger mm size fans cause 5000rpms is a bit to whiny for most....

2) The power supply is weak....How come the amd system has a power supply reporting so weak on the 12v rails...I would question any reboots with reading like this. f he has checked them to be bogus then stop showing them but any person with half a brain would be worried about a 11.46v on the 12v....Intel chip is 11.99 or higher the whole time....



Unless he shows me a cpu task manager showing all 4 cpus loaded continuously I may question if this is actually happening....

I also think Intel cpus should run with throttlewatch so we can see nothing is happening 71c max id quite high IMO....


Also does his 1 reboot mean both systems restarted on its own or his initial startup??? I read where that would be a correct assumption but they crashed at near identical times then based on online time.....

Being that they both started at the same time, it was either a power failure or it was done intentionally to start the testing.

 

[PetNorth]

Notice red and purple lines (CPU Load).

Dual core X2 is 100% load constantly.
Dual core PEE has tremendous down & up.
I ask me if this is caused by throttling or not. I don't know for sure. Someone?

 

[Lonyo]

Originally posted by: Duvie
OK had a chance to check it out....

AMD vs Intel things...

1) The fan is substandard and not even a 80mm fan. High rpms suggest that and I bet it is like a standard cooler. That is fine since he is comparing retail models....A cheap 20 dollar 80mm fan will do a bit better IMO...Intels lower rpm suggest a higher cfm fan....AMD needs to go to higher cfm, larger mm size fans cause 5000rpms is a bit to whiny for most....

2) The power supply is weak....How come the amd system has a power supply reporting so weak on the 12v rails...I would question any reboots with reading like this. f he has checked them to be bogus then stop showing them but any person with half a brain would be worried about a 11.46v on the 12v....Intel chip is 11.99 or higher the whole time....



Unless he shows me a cpu task manager showing all 4 cpus loaded continuously I may question if this is actually happening....

I also think Intel cpus should run with throttlewatch so we can see nothing is happening 71c max id quite high IMO....


Also does his 1 reboot mean both systems restarted on its own or his initial startup??? I read where that would be a correct assumption but they crashed at near identical times then based on online time.....

12V rails thing is odd, especially as both are using the same PSU.
The heatsink thing? If you read the accompanying article, both systems are using stock heatsinks.

The temperature issue? Well, the measurements could be from mobo sensors, which could mean they numbers are affected by that.
There seems to be no suggestions of manually places tem psensors, so I would guess they are using the build in stuff (do both CPU's now have on die sensors?), which coul dbe why there is "only" 7 degrees difference.

 

[Capt Caveman]

Originally posted by: PetNorth
Notice red and purple lines (CPU Load).

Dual core X2 is 100% load constantly.
Dual core PEE has tremendous down & up.
I ask me if this is caused by throttling or not. I don't know for sure. Someone?

Just a guess, possibly when the PEE finishes a task?

 

[Lonyo]

Originally posted by: brandon
I may have misread it, but, it still seems like BS to me...

Encoded Time on Intel System: 100 Minutes
Encoded Time on AMD System: 10 minutes

Make that 10x faster, unless I read it wrong.

Yes, Intel is winning on one of the parts making up the test.
Both systems are doing AL of the tasks listed at the same time, so the AMD machine is playing a game faster, doing RAR archives faster and ripping CD's at the same speed, while encoding video slower than the Intel system.
This just means that the AMD system is swloer at one of the 4 tasks than the Intel system, due to being able to handle less threads. But the encoding test on its own means nothing, and says nothing about the speeds of the two systems.

 

[Amplifier]

Originally posted by: keysplayr2003

Originally posted by: Amplifier

Originally posted by: irwincur
I think the site doing the testing really needs to be taken into account.

THG - Tom's "Intel Supplied and Paid" Hardware Guide

Seriously.

Ignored.

Another phanboi pwned.

 

[PetNorth]

Originally posted by: Capt Caveman

Originally posted by: PetNorth
Notice red and purple lines (CPU Load).

Dual core X2 is 100% load constantly.
Dual core PEE has tremendous down & up.
I ask me if this is caused by throttling or not. I don't know for sure. Someone?

Just a guess, possibly when the PEE finishes a task?

hmmmm... but if it would be the reason, I imagine we would see some down in X2 load, wouldn't we?

 

[Capt Caveman]

Originally posted by: PetNorth

Originally posted by: Capt Caveman

Originally posted by: PetNorth
Notice red and purple lines (CPU Load).

Dual core X2 is 100% load constantly.
Dual core PEE has tremendous down & up.
I ask me if this is caused by throttling or not. I don't know for sure. Someone?

Just a guess, possibly when the PEE finishes a task?

hmmmm... but if it would be the reason, I imagine we would see some down in X2 load, wouldn't we?

I said it was just a guess. I was thinking it was b/c the hyperthreading spreads the load, when a task completes, the load gets redistributed and creates the fluctuation among the four cores(physical and logical).