• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

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

Page 7 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
http://www.tomshardware.com/stresstest/amd.html

Anybody know where heck the ram is in that picture????

Fan is running on the cpu but no ram sticks populating any of them dimms...they use to be in slots 1 and 2, two of them with silver heatspreaders.....Picture seems to be up-to-date with reported time...Now I am not calling him on this one, but that seems strange!!!
 
They are still there, you can see the memory retention clips disappear about halfway into the module. It's just a poor webcam.
 
I think I see them....They must have moved the webcam because before you could definite see them and know they were heatspreaders and what color....


Anybody want to speculate on why the Intel board is doing about 20+% better in lame encoding then before.....about the same in winrar and now about 7-10% better in gaming, while encoding of Divx appears to be the same???

I am wondering cause a simple mobo change should not have yielded but a few percenatge points if that....check out mobo reviews and you will notice most are almost identical....

I questions toms reviews for just this fact...even he cant seem to duplicate his stuff....I am basing this on the numbers I saw on friday when neither system had rebooted and were running fine and had encoded 25 Cds......At rates shown now the INtel system will have 25 and the AMD will have 21-22...almost a complete flip from before and then some.....
 
Tom's Hardware Bribe at it's best.

First of all is it any wonder they chose 4 tests with a AMD dual core and Intel 4 Core? hmmmmm..

Intel setup had 6-8 crashes depend when you count..They flat out lie when they say only the gigabit gave them trouble because they used an Asus P5ND2-SLI Deluxe afterwards which crashed 3x after that. Then they switched Asus P5WD2 Premium.

Meanwhile they take a AMD system is running solid for 102 hours with no crashes, take it off-line and reset all counters for both systems to 0. What a joke THG is.
 
So in that first 25 cds of friday had it crashed multiple times already cause I am wondering how the big swing in performance....

I got a less swing moving to dual channel memory from single channel memory on a P4 system....I have neever seen from top platform to bottom platform change that much....I went from i845g to i875 at same speed cpu and saw less gain...How do they figure the change unless you have done more then we have seen. More agressive memory settings in the bios of the new board??? I cant imagine that making up 3-5% max.....

Something seems strange other then like I said day 1 no one can seem to reproduce Toms results....now even himself!!!
 
Odd is'nt it😉

I'm on vacation D so I'm not even going to try and justify Toms BS. His benchmarks are not reproducable and contridict not only himself, but every other site on the net which has done multitasking benchmarks suits for these two processors.

EDIT: how did you like how he blamed X2 for stopping farcry when the demo suite has a 1000 limit... duh... paid.😉
 
Originally posted by: Pabster
Originally posted by: AnandThenMan
How about this troll bait. Who's the troll now huh?

http://img.photobucket.com/albums/v295/skynet107/24h_intel-18.png

if you pull your head out of Intel's ass long enough you can read the graph for yourself.
Click to expand...

LOL. You really are an idiot!

I was looking at the 24 hour averages for the Intel rig versus the AMD rig. Which show exactly the figures I listed. I never said anything about the "restarted" rigs (which had only been running for 29 minutes at that time).

And anyone with half a brain can figure out that if the EE were actually at 91C it would not be doing anything. Sheesh, brainpower is in short supply here.
Click to expand...

You labeled the figures you gave as "lower average temp" which is wrong. They were the lowest temperatures. Those are meaningless, you need to take the average temperature which, in this case, was 84.8 for Intel and 52.8 for AMD.

The stats for the benchmark clearly contradict what THG says, so obviously the benchmark is flawed and the results are skewed. I don't even know why you are arguing with hard data.
 
Hey ever seen this one.....The fps for the intel system were in the 30s and now have drop to high 20's and the AMD which was starting to fall behind has blown by the Intel in Farcry.....I have never seen that on a stable system...usually percentages after 30minutes will hold the duration, just that often numbers are too small to get an accurate beat on it for a bit...however I have never seen one speed up like the iNtel did gaining up to a full 2 run lead then fall back to now behind by 8 runs....something has to change for that to happen....
 
Originally posted by: blackllotus
You labeled the figures you gave as "lower average temp" which is wrong. They were the lowest temperatures. Those are meaningless, you need to take the average temperature which, in this case, was 84.8 for Intel and 52.8 for AMD.
Click to expand...

Actually, that figure was the lowest average temperature. But it wasn't for the 24-hour period (ie after the mobo swap and test reset). It was for the week, so it was based on data from the previous test (which has been deemed invalid).

...For the last 24 hour period, yes. That is correct. ~85C Intel, ~53C AMD. But I question the Intel figure, because I don't believe at 83C it would be doing much (if any) processing.
According to the EE 840's data sheet, 69C is where throttling could begin.

The stats for the benchmark clearly contradict what THG says, so obviously the benchmark is flawed and the results are skewed. I don't even know why you are arguing with hard data.[/quote]

Are we to believe that the EE 840 is operating around 85C? If that were the case, it could not possibly be doing any work. At 85C it would be throttled down rather severely. And at 91C (as was reported at least once) it would be doing absolutely nothing.
 
Originally posted by: Pabster
Originally posted by: blackllotus
You labeled the figures you gave as "lower average temp" which is wrong. They were the lowest temperatures. Those are meaningless, you need to take the average temperature which, in this case, was 84.8 for Intel and 52.8 for AMD.
Click to expand...

Actually, that figure was the lowest average temperature. But it wasn't for the 24-hour period (ie after the mobo swap and test reset). It was for the week, so it was based on data from the previous test (which has been deemed invalid).

...For the last 24 hour period, yes. That is correct. ~85C Intel, ~53C AMD. But I question the Intel figure, because I don't believe at 83C it would be doing much (if any) processing.
According to the EE 840's data sheet, 69C is where throttling could begin.

The stats for the benchmark clearly contradict what THG says, so obviously the benchmark is flawed and the results are skewed. I don't even know why you are arguing with hard data.
Click to expand...

Are we to believe that the EE 840 is operating around 85C? If that were the case, it could not possibly be doing any work. At 85C it would be throttled down rather severely. And at 91C (as was reported at least once) it would be doing absolutely nothing.[/quote]

So we can all agree that that specific test is flawed and its data should not be taken seriously?
 
Originally posted by: Pabster
Originally posted by: blackllotus
You labeled the figures you gave as "lower average temp" which is wrong. They were the lowest temperatures. Those are meaningless, you need to take the average temperature which, in this case, was 84.8 for Intel and 52.8 for AMD.
Click to expand...

Actually, that figure was the lowest average temperature. But it wasn't for the 24-hour period (ie after the mobo swap and test reset). It was for the week, so it was based on data from the previous test (which has been deemed invalid).

...For the last 24 hour period, yes. That is correct. ~85C Intel, ~53C AMD. But I question the Intel figure, because I don't believe at 83C it would be doing much (if any) processing.
According to the EE 840's data sheet, 69C is where throttling could begin.

The stats for the benchmark clearly contradict what THG says, so obviously the benchmark is flawed and the results are skewed. I don't even know why you are arguing with hard data.
Click to expand...

Are we to believe that the EE 840 is operating around 85C? If that were the case, it could not possibly be doing any work. At 85C it would be throttled down rather severely. And at 91C (as was reported at least once) it would be doing absolutely nothing.[/quote]

We kinda already know that the temps are being reported incorrectly. Mobo may very well need a BIOS bug fix. So in light of this, can we get beyond the temps now?

 
Originally posted by: blackllotus
So we can all agree that that specific test is flawed and its data should not be taken seriously?
Click to expand...

Agreed, but I don't imagine it'll be too long before an idiot or two (who shall remain nameless) start in with the "whoa man Intel chip is really frying !!!" crapola ...
 
Originally posted by: Pabster
Originally posted by: blackllotus
So we can all agree that that specific test is flawed and its data should not be taken seriously?
Click to expand...

Agreed, but I don't imagine it'll be too long before an idiot or two (who shall remain nameless) start in with the "whoa man Intel chip is really frying !!!" crapola ...
Click to expand...

Well, you can be the official corrections person in the event some idiot does exactly what you said. Look man, we (most of us) know better. We know the CPU would probably be in flames at those temps.. No worries.
It's just a damn shame they cant get that Intel fan to turn faster. That speed seems very slow even for a stock intel hsf. 2100 rpm? Not sure if thats normal especially if this is the fan that shipped with the PentiumD. You would think Intel would ship a fan that could cool this puppy off appropriately. Look at the graph. Every dip in temperature and utilization is a Intel CPU throttle. Add up all that time the CPU throttles and it will severely hurt its marks. That fan is really bugging me.
 
HEY GUYS!!!

THG seemed to have updated their "stupid" stress test pages. First they admit that the AMD system did not reboot and that they would therefore correct their mistake by subtracting 1 from the AMD reboot tally (essentially making it zero). they also admitted that the intel system rebooted twice. but when i looked at the current tally, i was SURPRISED!!!

They restarted everything and now AMD is behind intel in CD encoding, only barely ahead in winRAR, and still far behind in DivX??? this is really some sort of biased THG article again.

Few points to the "geniuses" at THG:

1) they should change the farcry to RUNS and not just FPS because, the faster system clearly is able to make more RUNS and the graph should reflect that. not that stupid FPS only thing.

2) why did they restart the count to zero again?? why blame the northbridge??? sorry THG, but since the P840EE uses the external bus to shuttle data between the CPUs maybe that's why the intel system is freaking HOT. maybe you need to watercool the NB too in an intel system. lol

3) Hey why not mention that the P4 system is running at 90C!!!! (AMD at 55C)

4) again, why restart the count back to zero??? is it because your beloved intel lost in 3 of the 4 benchies by a mile??? plus the fact that no one would believe that the EE is 20x faster than an X2 anyway on DivX???

5) sigh

3)
 
So are we sure it is throttling now??? I see the cpu dips correspond still to the usage dips so which one is it??? The app is just in between runs or is there actually cpu throttling going on???

I wouldn't think so cause I dont see a build back up of heat...It would keepp throttling at much closer level, IMO. It appears to be happening in quite regular 20 min intervals....I think there is a simple explanation in its usage by the apps to explain the drops....Could be the HT and when an app finishes for that brief moment maybe things are being reassigned and actually slows the system as it reschedules loads....Just a guess but it seem to regular and the heat builds back fast and then tops out for the 20min periods...

Though I do have to admit AMD doesn't seem to show same signs in its graph....

 
3 boards later and still reboots...

I could tell something was getting flaky...When you see scores building and then reverse it is odd....

I wonder why they are not smart enough to right an auto bat file to launch all the apps at the beginning...now instead we have another downtime where the Intel sits idles and the performance numbers are shite again.....

I hope they use an INtel board next and when that reboots they pick the whole thing up and throw it out the window....The damn thing can only run 4 core load for like 10-12 hours and it will reboot.....This is about the 4th official reboot but the 5th downtime.....
 
WTH? Intel system is clearly idle, yet current Farcry FPS is 27. How is that possible? 😕

Current Frame Rate on Intel System: 27 FPS at 178 runs
Current Frame Rate on AMD System: 36 FPS at 244 runs
Click to expand...
 
The runs are not increasing it is just holding the last successful run's average.....None of the scores have increased including the runs.....
 
hmmm well the idle temps are about the same as with the other motherboards, but the load temps are nearly 90 degrees. I find it hard to believe the processor would get to that level without throttling or being damaged.

What is the deal anyway? Is THG just not good at putting together systems? Can't blame the motherboard this is like the 4th unit they have tried. Maybe the processor is just defective or they damaged it when installing it. The Intel rig went idle again, wierd. This doesn't look good for Intel. To be honest I've never had the issues with a build like Tom is having. I just don't know what to think.

Originally posted by: AnonymouseUser
WTH? Intel system is clearly idle, yet current Farcry FPS is 27. How is that possible? 😕

Current Frame Rate on Intel System: 27 FPS at 178 runs
Current Frame Rate on AMD System: 36 FPS at 244 runs
Click to expand...
Click to expand...

The "live" charts are not live actually. I and several others thought something shady was going on because the Intel stats sometimes would go up even when the system was idle, but it's only because the stats update at what seems to be random.
 
It happens....

I think they should try another chip, intel mobo, new ram and another power supply..and ofcourse star the test again....If it does it again like I said throw the POS out the window.....
 
Originally posted by: Duvie
It happens....

I think they should try another chip, intel mobo, new ram and another power supply..and ofcourse star the test again....If it does it again like I said throw the POS out the window.....
Click to expand...


True, it does happen. But 4 motherboards? I've screwed up system builds but nothing like that.

It would be so much nicer if we really did have a live feed on the monitors to actually watch what is going on.
 
4 mobos are not bad when it could have been a faulty ram stick which errors out at what would be equivalent to a 20th pass on memtest...For an average user they may not find the error...

It also could be a a power supply issue, the cpu like already stated, and or other component.....


Who kow just make sure this time we have that heater like across the room from the AMD machine...remember last tieme when the INtel cpu was heating up the AMD cpu???
 
They are using a PC Power and cooling 850, unargueably the best PSU on the market, 4 or 5 mobos and good ram. I say call it a failure on the Intel side and quit ! And I think the AMD is winning on performance also !
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top