GuitarDaddy
Lifer
- Nov 9, 2004
- 11,465
- 1
- 0
AnandThenMan
Diamond Member
- Nov 11, 2004
- 3,991
- 627
- 126
The Intel system rebooted again, by my count that is 5 even though it only shows 2. Also the Intel has 0% CPU usage again. But despite all the idle time it manages somehow to encode video at 25 times the rate of the AMD system.
What a load of CRAP from THG. They have wroten some crappy articles , bu this is really a new level of retardation and intel biasing crap.
Anybody have any proof that while the cpu sat there for 9 hours idle (which I saw that) the DIVX encoding time increased??? This would prove he is a liar....
Funny how none of the iNtel fanboys will come in and support Tom.....
LIke I have always said that no matter what a person says the settings (main ones) are they can still skew any results with many setting swith dont always think about.....The best thing to do is to look for similiar results across may sites....Problem is Toms stuff is usually never the same as other more reputable (appear less biased) and seems difficult to ever reproduce....
Funny how none of the iNtel fanboys will come in and support Tom.....
LIke I have always said that no matter what a person says the settings (main ones) are they can still skew any results with many setting swith dont always think about.....The best thing to do is to look for similiar results across may sites....Problem is Toms stuff is usually never the same as other more reputable (appear less biased) and seems difficult to ever reproduce....
AnandThenMan
Diamond Member
- Nov 11, 2004
- 3,991
- 627
- 126
The problem is, the "live" stats seem to update when ever they feel like it. For example, the Intel box was idle for 5-6 hours (as of now it just started processing again) but I just saw the Divx encoded minutes jump up even though the cpu was idle. It is not possible to 100% correlate the idle time with any increases in stats because the live stats are not live. But it sure LOOKS like the Intel stats increase even during 0% CPU usage. Here's another png of several hours of downtime.
Link
In total, the Intel system has been idle for 30% of the time.
Link
Duvie, i think the proof that the intel CPU continued "working' at idle is that currently, all the charts comparing the athlon x2 to the intel EE are increasing at a generally proportionate rate. If the intel CPU was idle for 9 hours, there's no way that all the charts would show the 2 CPU's always being at the same proportion performance-wise; either the athlon would be a given amount above the intel in performance in addition to the given proportion, or the intel would be making a comeback if it is faster. yet none of the graphs appear to show this.
If i remember correctly the athlonx2 has had about 2x the # of archives compressed in comparison to the EE since the start, yet this should be impossible given 9 hours idle......although I'm not 100% certian on this. but I'm still pretty sure
Originally posted by: screech
Duvie, i think the proof that the intel CPU continued "working' at idle is that currently, all the charts comparing the athlon x2 to the intel EE are increasing at a generally proportionate rate. If the intel CPU was idle for 9 hours, there's no way that all the charts would show the 2 CPU's always being at the same proportion performance-wise; either the athlon would be a given amount above the intel in performance in addition to the given proportion, or the intel would be making a comeback if it is faster. yet none of the graphs appear to show this.
If i remember correctly the athlonx2 has had about 2x the # of archives compressed in comparison to the EE since the start, yet this should be impossible given 9 hours idle......although I'm not 100% certian on this. but I'm still pretty sure
That isn't true though...I watched it as the P4 went down Friday evening and the CD and winrar were not increasing...The game was not increasing in runs but the FPS always held 36.....I cant say for certain bu the encoding time may have been moving.....
When it was frday afternoon it was 24CD for Intel and 26CDs for AMD then by friday evening it was 55CDs for Intel and 72CDs for AMD.....that is not proportional....
This is beside the point though...the test and results are borked irregardless...
What I wonder is how is it Intel has only 2 reboots...It is clear there were 2 definite downtimes for the intel system yet AMD has none....AMD and INtel system from the moment I saw them early friday morning each said 1 though I couldn't ascertain if they counted initial boot as a reboot...If the did they are fvcking morons, but at this point it appears that is more then evident enough..,...
Also NOtice how INtels average and high cpu temp is now 2-3c lower then inital start of test on firday and portions of saturday..yet AMDs temp is within less then 1c of how it is always been.....Did they due some tweaking of the temps when it was down for soooooo many hours??? Who knows...Is this the normal AS or thermal paste breakin?? if so why does AMD not gain any???
ah, my bad then. it just seems that it has been proportional form what i've seen lately with winrar, i guess im wrong
Also ppl list how they think the P4 has rebooted even more then the obvious 2 since friday froom the week chart??? I see some sharpe red line dips that hit 0% usage bottom but without a zoom in I cant tell for sure....Also the downtime of the reboots suggest that Toms has no batched setup and so a reboot may not start the apps upon restart. In those cases it looked like it would have started the apps fine...or maybe just those times a lab person was there when it happened.....
It is a sad test no doubt...
It is a sad test no doubt...
Originally posted by: screech
ah, my bad then. it just seems that it has been proportional form what i've seen lately with winrar, i guess im wrong
NO problem...
However with the encoding test would someone link me to something that showed it was still crunching along while being down that first time for some 9 hours.....
The thing to notice is how AMD cpu usage stay so firm for such a long period of time and intels looks like a heart monitor....
So what do you call what happened at 3:00pmto the INtel chip??? Both cores were at 0% usage for 6-10 minutes....You can also verify that by slowing of fan speed, raising of vcore levels and drop and hold in cpu temps....
Is that a hard lock....Is it the computer just confused itself and took a break??
Is that a hard lock....Is it the computer just confused itself and took a break??
IN the review they were only running the encoding app....This is a mutlitasking test of 4 cpu intensive apps...all 4 apps if ran single would at least hold 100% for one core....NOw winrar is not multi-threaded so it would only take one core...the gaming is only single threaded as well...The lame encoder can be multithreaded, but I dont know if that is the one he is running....Divx is but to a degree and depends greatly on the application...Some Divx programs are extremely friendly to Ht and Dual cpus....
So the test are not what is in question...
It is the descrepancies in the reboot count....reboot is not being automatically counted cause it took a day to count the one that happened on Friday while the 2nd one has never been counted yet....Both cpus had 1 reboot counted as of the first time I saw the thread on Friday morning..For the life of me looking at the charts I can not see where the AMD rebooted...though it is very clear where the iNtel did as well as periodically the P4 seems to sit idle for 5-10 minut spurts....
Other discrepancies are rumored the divx test was continuing to count even while it was (the system) done for nearly 9 hours on Friday....That is suspicious....
So the test are not what is in question...
It is the descrepancies in the reboot count....reboot is not being automatically counted cause it took a day to count the one that happened on Friday while the 2nd one has never been counted yet....Both cpus had 1 reboot counted as of the first time I saw the thread on Friday morning..For the life of me looking at the charts I can not see where the AMD rebooted...though it is very clear where the iNtel did as well as periodically the P4 seems to sit idle for 5-10 minut spurts....
Other discrepancies are rumored the divx test was continuing to count even while it was (the system) done for nearly 9 hours on Friday....That is suspicious....
The AMD system now gone from 1 to 0 reboots. At this rate AMD will finish with at least -1 reboots and Intel 4+.
Actually, what makes me wonder the most is how the Intel rig is getting results with a 25%+ downtime. The results don't seem to be the most relevant thing in this test though. They can't be easily weighted and compared. Running each test seperately would show which cpu is faster (and has). For multi-tasking you don't need a 7 day duration to do a benchmark. It adds in other problems that THG is apparantly facing and makes the results suspect.
Why can't THG also do just one run of the script and see which finishes first? They did a video comparison of hyperthreading before. It clearly showed which rig loaded fastest. A performance interpretation without the times of one run would be extremely dishonest in my eyes.
I think in the end, the Intel and AMD camp will find reasons to proclaim their superiority from the results THG will come up with.
Actually, what makes me wonder the most is how the Intel rig is getting results with a 25%+ downtime. The results don't seem to be the most relevant thing in this test though. They can't be easily weighted and compared. Running each test seperately would show which cpu is faster (and has). For multi-tasking you don't need a 7 day duration to do a benchmark. It adds in other problems that THG is apparantly facing and makes the results suspect.
Why can't THG also do just one run of the script and see which finishes first? They did a video comparison of hyperthreading before. It clearly showed which rig loaded fastest. A performance interpretation without the times of one run would be extremely dishonest in my eyes.
I think in the end, the Intel and AMD camp will find reasons to proclaim their superiority from the results THG will come up with.
- Jan 16, 2003
- 21,219
- 54
- 91
Why would Intel fanboys come in and support Tom? Everybody is seeing the same thing you are Duvie, who in their right mind would defend them now after seeing this farce? In the beginning of this post, I saw people immediately hammering Tom's site, and I kind of defended them a little. Geez, give them a chance, but not now. They really f*****d this up. I know now that Anand will never conduct such a test as too many variables can go wrong and AT would get beaten down for it just like Tom's is now. AT doesn't want that headache.
I make it seven reboots. 4 up to last night when it rebooted but the counter reset to 3 (actually 5) then twice today (7). I also reckon it's on it's 3rd motherboard (one Gigabyte and two Asus).
I have a feeling, though, that the reboot counter doesn't include THG switching off to change hardware.
I have a feeling, though, that the reboot counter doesn't include THG switching off to change hardware.
That'll be 8 reboots now....
Addition...
Looks like they're swapping more hardware out, currently the HSF, CPU, RAM and one VGA card have been removed from the Intel box.
Interesting to see that while the Intel machine is off the total power usage is now at 345watts, rather than the normal 855watts (if I remember rightly) that pegs the Intel power usage at 510watts!
Addition...
Looks like they're swapping more hardware out, currently the HSF, CPU, RAM and one VGA card have been removed from the Intel box.
Interesting to see that while the Intel machine is off the total power usage is now at 345watts, rather than the normal 855watts (if I remember rightly) that pegs the Intel power usage at 510watts!
PeteRoy
Senior member
- Jan 16, 2003
- 21,219
- 54
- 91
They should abort the test and start over if they are going to change the damn mobo.
They should restart it anyway. The AMD system is doing very nicely.
the dudes hand has not moved in the last 1/2 hour or so. the Intel image is no longer refreshing.
They should restart it anyway. The AMD system is doing very nicely.
the dudes hand has not moved in the last 1/2 hour or so. the Intel image is no longer refreshing.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 22K
-
News NVIDIA and Intel to Develop AI Infrastructure and Personal Computing Products- Started by poke01
- Replies: 387
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
Facebook
Twitter