THG Intel vs AMD stress test disaster

Thg started a stress test of the latest amd and intel plattform on june 3rd, and so far it has been a disaster. thg hardly documented their methods of testing and the problems they found and how they solved them.
for me its sad to see such a well known site do such a bad job, and have removed the old logs and performence stats making it look like they are trying to hide things.

UPDATE: it looks as if THG is using a BTX heatsink in an ATX case wich results in those ridiculous load temps of 70-90°C! :doh:

UPDATE: THG finally posted some more detailed infos about the problems they had, but still not about all the problems that were reported by xs members who checked the results and live feeds of the systems! http://www.tomshardware.com/cpu/20050603/index.html



i hope this gets enough attention to change the quality of their site in future!

here is the article:
http://www.tomshardware.com/cpu/20050603/index.html

here you can see the systems running on a webcam live feed that gets updated every few seconds.
http://www.tomshardware.com/stresstest/index.html

and here you can see the results of the different apps they are stressing both systems with wich gets updated all the time as well.
http://www.tomshardware.com/stresstest/charts.html


amd system:
a64 x2 4800+ with boxed HSF
gigabyte nf4 sli
2 x 512mb ocz pc3200 2-2-2-5
2x msi 6800U 256mb
2x WD 160GB SATA
PCPNC 850W PSU
Case with 2 fans, in front of the hdds and exhaust below the psu.

intel system:
840EE with boxed HSF
gigabyte nf4 intel edition sli
2 x 512mb ocz DDR2 667 4-4-4-5
2x msi 6800U 256mb
2x WD 160GB SATA
PCPNC 850W PSU
Case with 2 fans, in front of the hdds and exhaust below the psu.


thg log:
i made a log of all the events thg posted about in their article
xs log:
i made a log of all the events xs members posted about that thg hasnt mentioned


06-03 test starts
06-04 at 8:45am paciffic time the gigabyte nf4 intel edition board crashes
06-04 at 5pm paciffic time the gigabyte nf4 intel edition board crashes
06-04 bios flashed to increase the chipsets fanspeed, system restarted
06-04 the wattmeter dies :caution:
06-06 a bios flash kills the gigabyte nf4 intel edition board :caution:
06-06 an epox nf4 intel edition is beeing used
06-06 the epox board suddenly dies as well :caution:
06-06 an asus nf4 intel edition is beeing used
06-06 Crucial DDR2 667 is beeing used since they couldnt get the ocz memory to work on the news board
06-06 the reboot counter gets changed from 3 to 1
06-06 the reboot counter gets changed from 1 to 4
06-07 the intel system crashed again
06-07 an intel 955x board is beeing used, amd still running sli
06-07 the 955x board from intel doesnt work with the 840EE and died? :caution:
06-07 second 955x board from intel doesnt work with the 840EE either and died? :caution:
06-07 05B boxed heatsink gets replaced with a 05A boxed heatsink because of a termal pad??
06-07 enermax 480W psu beeing used instead of pcpnc
06-07 both systems now running with only one card
06-08 an asus 955x board is beeing used
06-08 different boxed HSF with bigger copper base from intel beeing used 05A heatsink removed
06-08 amd is no longer running sli, the crashcounter and the performence stats get reset AGAIN!
06-08 the webcam showed the intel system on the desktop yet the far cry fps of the intel system were increasing and increasing
06-08 the fps in far cry in the intel system dropped and the cd encoding speed jumped up significantly, looking as if the priority of the benchmakrs had been changed
06-08 the amd performence stats show that the amd suddenly encoded cds much slower, only 1 cd in one hour for at least 2 hours
06-08 since the asus 955x board has been used the system has crashed at least twice
06-08 the fanspeed of the intel fan jumped from 2000rpm on the old heatsink to 3600rpm on the new heatsink, so its not just a bigger copper base that makes the difference. the temps of the intel rig went down significantly from 90°C to 67°C and the intel system is suddenly faster in far cry than the amd system in the stats
06-08 the vcore on the intel system is now lower than it used to be in the earlier phase of the stress test
06-10 the fanspeed of the intel heatsink is up to 4300rpm!!!
06-10 the network connecting the test systems seems to be down, yet the intel systems score keeps increasing while the amd systems score froze
06-10 the persformence stats changed drastically all the sudden:
Lame Encoder Intel = 301 AMD = 256
Winrar Intel = 1634 AMD = 2072
Farcry Intel = 27FPS AMD = 35FPS
Divx Intel = 1380 minutes AMD = 80 minutes

About 10 minutes later the following:
Lame Encoder Intel = 303 AMD jumps up to 311
Winrar Intel = 1643 AMD jumps to 2321
FarCry Intel = 27FPS AMD drops to 15FPS
Divx Intel = 1390 AMD jumps to 130
looks like the priorities for the different tasks have been changed manually, no comment from thg...

06-13 ht gets disabled on the intel system?
06-14 final results get posted, intel is now faster in mp3 endocing compared to the previous results, no comment from thg. also no mention of the intel system crashing several times
06-14 intel mainboard changed because the asus board doesnt allow ht disabling?
06-10 ht disabled on the intel plattform to find out why divx is slower on the amd system
06-10 Priorities for Divx get set manually, making the step of disabling ht to find out about windows automated thread prioritizing useless?


at this monent the reboot counter of the intel system shows 3 reboots, wich havent been documented by thg, at least not yet? :confused:
further more thg claims the intel system was running steadily under load, wich is not true, not even by their own documentation.

Quote:

---

Originally Posted by THG

Monday, June 6, 2005: Our stress test has now been going for more than 60 hours. The systems from AMD and Intel are running steadily under full load, and a lot has happened in the Munich THG lab during this time

---

here is a roundup of what was documented by our forum members:

Quote:

---

Ok you guys need to view these with internet explorer since they are mht archive files.

First off we have the inital gigabyte board with already one re-boot

Next Toms switches boards to Asus NF4 and still shows 3 reboots

Can't have that!!! so he re-sets the boot counter back to 1!!!

Crashes 4 more times as shown here before finally upgrading to Asus 955x board. That's at least 7 crashes so far!!!

Meanwhile, as shown here AMD is rock solid 97 hours straight!!! (actually more like 102 but I don't have an archive of that)

Toms does'nt like Intel 7-9 crashes so he resets counter yet again to Zero for the intel system and whipes out AMD rocksolidness restarting all testing


Intel Crashed 3x since last night with the new testing on Asus 955x

Count them up to 12 crashes for intel, Zero for AMD.

---

and heres a funny link you gotta check out :D
http://users.pandora.be/compas/stresstest/intel.html


thx a lot to Mr. Tinker and Zebo for their help :toast:
im really curious about thgs conclusion of this :D

heres the discussion about the whole thg disaster on our forums:
http://www.xtremesystems.org/forums/...ad.php?t=64950

I think this would have been much better if THG had given specs to intel and AMD and just plugged the rigs in and let them run.

I would bet that both would have run flawlessly throughout.

nice work saaya you have to put that in the front page

done :)

Wow. Its common knowledge that they are fairly pro-intel, but I didn't realize it went quite that far
edit: is that smoke I see in the Intel system pic? :eek:

its a fun link :D

So, much for the need to light a cig. during tests anymore. :ROTF:

http://www.xtremesystems.org/forums/...chmentid=32203

Looks like a photochop as the black plastic fan is glowing red.

look at the CPU temps for the intel - they are going off the charts

Ah, they now admit that the current Intel page is a spoof

LOL its not a site from THG its a joke!
that last link i posted is a joke somebody posted in the discussion thread in the amd section :D as i said, its just a fun link :D

Quote:

---

Originally Posted by FUGGER

Looks like a photochop as the black plastic fan is glowing red.

---

it just got that hot ;) :p:

Doing stress hardware test on top of line hardware is easy, i would like to see a rig with pcchips, a regular case, with just one fan in the back, and running those programs, because it is the reallity of many Pc´s around the world....

Quote:

---

Originally Posted by Dautonico

Doing stress hardware test on top of line hardware is easy...

---

apparently not

THG is the biggest trash on the interwebnet. Man I prefer reading Theinq.com then THG for hardware news.

:with: :owned: i prefer The Inq and just skip over THG, theyre so bias its not even funny, everybody is bias to an extent, but theyre a whole different level.

they were not always like that... i hope they finally realize that everybody is making fun of their bias and the quality of their articles and the management wakes up.

heres another funny image a thg fan made :lol:
http://sylphys.ddo.jp/upld2nd/pc/img...0609184911.jpg

this whole stress test is drawing a lot of attention... negative attention though, and i hope the thg management knows that this could be positive attention and make their site more successfull if they would increase the quality of their site :D

Quote:

---

Originally Posted by saaya

they were not always like that... i hope they finally realize that everybody is making fun of their bias and the quality of their articles and the management wakes up.

---

true, but for the last while it has been completely bias, maybe theyll see some of these threads around the net and realize something.

Quote:

---

Originally Posted by saaya

heres another funny image a thg fan made :lol:
http://sylphys.ddo.jp/upld2nd/pc/img...0609184911.jpg

this whole stress test is drawing a lot of attention... negative attention though, and i hope the thg management knows that this could be positive attention and make their site more successfull if they would increase the quality of their site :D

---

yea, no doubt that this is a very big test between the two companies, but it would be awefully hard for them to admit their bias and the current outcomes of the tests because of their standpoint on AMD and Intel. they would have to do a complete 180 and show the reality of their tests and articles without their extreme bias

updated the thread with more infos, and updated the log :)

:toast:

OMG!!!!!!!!!!1

Quote:

---

Friday, June 10, 2005: Both systems have been running without any problems so far, AMD since June 7, 8 pm, Intel since June 8, 3 pm.

---

HHhaHAhhaHAhahahHahahAhAhAh :rofl:

http://www.tomshardware.com/cpu/20050603/index.html

Heh. THG surely sux on this one. In my opinion to make a real stress-test they ought to use 955x and nF4ultra chipsets. And use apps like prime and something like that: so it could run 24/7(not 3d games). Now that would be the stress-test.

Quote:

---

Originally Posted by saaya

OMG!!!!!!!!!!1



HHhaHAhhaHAhahahHahahAhAhAh :rofl:

http://www.tomshardware.com/cpu/20050603/index.html

---

What's up with the blatant lies coming from THG? The Intel system has been riddled with heat issues, the sky is falling and the word we get from THG is, "don't look up, everything is fine."

THG is for nubs.

Awww... I always liked THG... this is heartbreaking for me!!

I used to love Sharky Extreme but then they got bought out and became pants so I converted to THG. For the past three or four years I've checked THG every day in fact so much so that my girlfriend goes on about 'Who is this Tom with all the hardware and why do you spend so much time looking at his site rather than looking at me'. Since I've come across this site I've found that I look over at THG less and less. Not because of any bias or anything (because I didn't know about that) but because news seems to be *made* here rather than *reported* there.
Take the X2 chips for example, THG have barely got any stuff about them now whereas members on here have had them for a good month or two and have already showed how well they overclock, perform etc. I used to read THG for their articles on new hardware but people on here write up their findings months before any online review sites.

So what is the actual bias of THG? I always thought they were a bit... kind... to Intel, ever since I read their article on the P4 when it first came out where although they did say that it was a bit pants (c'mon, it was compared to the P3 and the Athlon) but they spun it in the most pro-Intel and 'nice' way that they possibly could have done.
I get the feeling that they're pro Intel, pro ATI. Is that true? Or are they pro nVidia?

I'll openly admit that I'm pro AMD and pro Nvidia and that I particularly dislike ATI. Not that they don't make good chips, its just like BMW and Mercedes... its difficult to like them both, you pick a side and you stick to it!!

from what ive seen thg used to like nvidia in the nv30 days, but is now rather biased to ati, but its minimal and i could be wrong. and it probably varies from article to article...

i stopped reading thg finally after they posted a review of a fan from gigabyte. the fan had a special design wich minimized the dead spot behind the center of the fan where no air movex. the tard reviewer at thg who reviewed it used it on a heatpipe heatsink, one of those twoer heatsinks with heatpipes on the edges and copper fins connecting both forming a small tunnel where you can mount fans on both sides.

and he trashed the fan because it didnt show any benefits over other fans... of course not because those heatsinks dont care about the dead spot behind the center of the heatsink like normal heatsinks where air gets pushed down onto a heatsink that sits right on the cpu :slap:

he trashed the fan and the fan was a flop and never really made it to retail, even though the deisng was nice and innovative! i was so mad at thg back then... but that only lasted for a day, now i only roll with my eyes if i hear of their artiles again and most of the time laugh about them... its sad though, thg used to be a good site a long time ago...

Wow... Restarting the reboot meter... Talk about bias... :slapass:

Usama aka Ferrari Freak

3 times iirc!

Good thread saaya..man you really got a lot more stuff than I realised.. Forgot about the Intel boards and EPOX's..

Bottom line: Toms HW is clueless on getting a system running, that any HS student can do.. In light of this there is no way I trust thier benchmark programming or tabluation methods, especially in light of the "no network" issues. I've stopped replying to the "other" thread in AMD forum because really what's the use? there's like 10 other "good" sites out there Im busy reading instead...techreport, anandtech, xbit etc..:D Seriously, not only is TOMs HW one of the ugliest sites on the net with 1992 HTML markup and yellow, thier reviews are always questionable to me.

im documenting the results of their live charts... the fps is jumping up and down for the intel system... from 15fps to 28fps to 38fps back to 15fps over 1-2 hours....

i bet they somehow crashed the intel system and are trying to fake the results now to hide it.

i wouldnt be surrpised if they would reset the reboot counter yet again... seriously...

If memory serves me correctly... this is similar to something that happened when Quake 3 came out (I'm not totally sure as this was a while ago, but this is how I remember the events). THW got ahold of a beta of Quake 3 and they were using the Quake 3 benchmarking tools to post some results. The problem was that at the time id didn't have the benchmarking tools working yet with the game! But THG were swearing up and down that the results were real and meaningful even though that wasn't even possible. So I believe that it turned out they were using Quake3 to generate meaningless framerates and posting that as if you could tell anything by them.

Unfortunately,THG has always posted questionable results... :nono:

saaya, you kick ass. this thread is officially l337! thanks for the welcome in the other thread too btw. my favourite quote of this entire thing:

Quote:

---

Friday, June 10, 2005: Both systems have been running without any problems so far, AMD since June 7, 8 pm, Intel since June 8, 3 pm.

---

:bounces:

thx :)

and yeah, that quote is insane :lol:

hes taking what very little credit he had from me, and flushing it down the toliet. HOW CAN ANYONE TRUST THIS GUY NOW? :slapass:

I always thought that THG was for n00bs. Considering how it is a bit biased to whichever hardware brand it is leading...

I'd rather much prefer a site with fair, unbiased reportings compared to THG. This shows how badly it does benchmarks. I mean honestly, BTX = ATX? I thought it was BTX != ATX. Apparently, THG doesn't seem to apply a bit of logic... :(

All in all, it still is a respected site to me.

http://www.tomshardware.com/cpu/20050603/index.html

update, they had an intel engineer over and they claim the problem were related to the memory. "surprisingly" the intel amd amd systems results in the peformence charts are matched, both win in 2 categories, intel in mp3 and movie encoding (what a surprise) and amd in the gaming and archiving category.

they disabled ht on the intel system to see if the low divx priority has anything to do with windows' thread priorities. and they switched the mainboards on the intel rigs because the asus board didnt let them disabled ht? 0_o
and the new board they use doesnt support any vcore or fanspeed readings?
ok, this is an intel es board, but the other intel es boards did allow fanspeed and vcore readings... this sounds a bit weird to me...
and they changed the thread priorities of divx now, why they did that doesnt make sense to me though... i thougt they wanted to see if windows would automatically give divx encoding a low priority? so now they disabled ht but at the same time set a higher manual priority to divx? hows that going to answer any questions?

oh and no word on how the intel system was now suddenly faster in mp3 encoding than the amd system, in the previous results they made the amd system was def faster in mp3 encoding, now they dont match but intel beats amd significantly...

Tom's credibility and respect went down in 2000... When AMD released the K7 and it dethroned the Pentium 3, Tom had a hard-on for them ever since. I didnt know people still went to his site :confused:

if you mess with the test and you rig the results why do the test in the first place?
why even do the stress test and post it live on the web?
its just too dumb to be true.

I knew they were biased towards Intel in the past. It seemed a little more subtle though. Now it is just blatantly obvious.

And they just restarted the reset clock again.... no good :nono:

now the intel system is running without ht, i think thats why.

lol. I did enjoy thg no2 movie though

i used to still look at them occasionally regarding monitors, cases, and power supplies, but i'm not even going to do that anymore.

THG has officially earned my complete scorn.

reboot meter was reset a few days ago again if it hasn't been posted yet.

Once @ 90 hours of stresstesting, now at 180 hours or something now they are at 60 hours IIRC ...

thats 3!!!!!!!!!!!!! stresstest, inwhich AMD didn't even cough.

Rocksolid, Intel ..... I counted 6 reboots total for the first 2 stresstest runs.

This run .. lemme check .. 0 still @ 64 hours into the third stresstest.

THG has lost ALL respect from me

Anyone know if those live cams are still on and a link to them?

Wonder if THG has been noticing the negative feedback here @ XS. :stick:

AMD system
Intel system
overall tests results

I'm glad there are sites like xtreme systems to debunk Tom's work... it'll make it harder for Tom to hide the real results. Sad really. Tom is an Intel lacky, fanboy and he is screwing Intel over with this little comparo... :rolleyes:

Tom might not be biased, he probably just gets payed a ton of cash to say the things he does. That site is pretty damn big in the computing world, and I am sure Intel wants to take advantage of that. But honestly I hate Intel and THG, so LONG LIVE AMD!!!

Quote:

---

Originally Posted by uscfan

Tom might not be biased, he probably just gets payed a ton of cash to say the things he does. That site is pretty damn big in the computing world, and I am sure Intel wants to take advantage of that. But honestly I hate Intel and THG, so LONG LIVE AMD!!!

---

A couple weeks ago I too would've said I hated Intel, but now I prefer saying I hate Pentium 4's (Especially Prescotts), Xeons, Celerons, and anything else made by Intel except the 533FSB Dothans :D

Why do you hate Xeons?

And true, dothans are the best thing made by intel to thisd day

Looks like THG finally realise how to write intresting article.

Tom's hardware guide update

;)

Quote:

---

Originally Posted by Hans.Gruber

Looks like THG finally realise how to write intresting article.

Tom's hardware guide update

;)

---

LMAO

Usama aka Ferrari Freak

This just unbelievable, Final update.

Quote:

---

When multiple applications are running, the clear conclusion is that the Intel Pentium 840 Extreme Edition is superior to the AMD Athlon 64 X2 4800+.

---

:censored:

Absolutely bull:banana::banana::banana::banana:.

First, you have to be able to run the system stable, which seemed to be a several day chore for the Intel rig. Yet, the AMD rig barely so much as hiccuped.

Secondly, you have to have a ridiculous amount of air flow to keep that Intel cooled. To think that it gets so hot at stock voltage that it throtles itself is just pathetic.

Its hard to believe that a site that used to be very reputable could deny the obvious and lean towards the poorer performing chip. Thousands of us checked up on this daily. I guess these silly germans at Tom's have no common sense. (no offense to any germans here, just to those at TomsHardware;))

Quote:

---

Originally Posted by Hans.Gruber

This just unbelievable, Final update.

[/b]
:censored:

---

:lol:

I thought Dr. Thomas Pabst didn't do drugs :confused:


Unless he based this conclusion on time taken to encode his :banana::banana::banana::banana: collection into AVI format :wasntme:

... just another reason to avoid THG. Lame!

It would be nice to see the same test done by a reputable site.

Quote:

---

Originally Posted by Frisch

The moment i have a hard time seeing the articles, because of advertising making it impossible to wacth without getting blinded by massive banners, i'm off.
I left THG many years ago.

---

Quote:

---

Originally Posted by Frisch

It's not the ads in it self, it's the sign of money going between the "neutral" tester and the advetiser, that tells me , it's time to leave.

---

Repeat peat.