No, I disable clocking down on both chips. I'm not even sure if it works on the dothan with the adapter.
By idle I mean desktop work, a'la net surfing.
Printable View
No, I disable clocking down on both chips. I'm not even sure if it works on the dothan with the adapter.
By idle I mean desktop work, a'la net surfing.
8th update.
http://www.tomshardware.com/cpu/20050603/index.html
That is true, but i still see it as twisting the truth..Quote:
There's nothing negative to report about the stability of either system. Both have run without a hitch since we restarted the stress test, thus giving us a great opportunity to compare their performance.
Perhaps they are restarting the test so many times that everyone forgets what happened in the first 2 rounds ?
Ok, now thats just...Quote:
Although our initial results showed the Intel HT system breezing past the AMD system at 13 times its speed, AMD pulls ahead of the HT-free Intel processor at roughly three times faster. Even when running the Lame encoder - where Intel had also been faster - the AMD system pulled into the lead.
WHAT ?Quote:
With our tests so far, it seems that hyperthreading is better than having separate CPUs at distributing and balancing the load on the overall processor(s). We aren't sure if this is due to the design of the memory that we are using, the individual CPU controllers or bus architectures, or something else that we haven't tested. We are investigating this further.
I think zebo might be in trouble, because I suspect this means Intel will somehow release lower clocked p4D's with HT and then He'll have to buy one.
Have I mentioned yet how I was completely right earlier?
Love how they say intel was 13x faster, but don't mention that it was because they were running 4 heavy load apps at once and that AMD won every other benchmark before their... adjustments. Essentially cloaking the important finding here, that the ONLY situation Intel beats AMD in is running 4x apps at once with HT enabled. And that even then it can only win 1/4 of the programs. They try to make it sound like The EE is really 13x faster than the X2.
I'm going to apply for a job at THG once I remove the part of my brain controlling that which done do my thinking... and bowel control.
Actually it does not matter if U have the best product, if only you have the best PR PPL in your staff that can convince buyers that you have the best......intel have lived pretty long on that now...thx to DELL among others.
Quote:
Originally Posted by Lithan
No doubt Dothan is a sweet chip brudda.. but what are we talking here? 15W, maybe, separates a dothan from a venice (both same Mhz)??? Not frekken 100W+ differnece. That's huge! And gets exponetial with dual cores and overclcoking... not touching a prescott...You got that right.:) And definity not two of them slapped on same package. :nono: <crazy?>
Quote:
Originally Posted by Lithan
Buy one?
:confused:
You must be crazy or joking.:D I already recognised and predicted intels HT dual core will beat a X2 in four intensive tasks all at once. SO??? BFD! Who runs 4? Show me someone besides Toms bias testing. And is 30 FPS acceptable to anyone under those conditions? Hell no. Not to mention X2 totally wipes the floor of any Dual core pentium in everything else. Two intensive operations, every single threaded app, every multimedia app, every multithreaded app. Everything!!! It's not even close, the lowest end X2 beats the 840XE like a red-headed step child. Then heat, power, and instability of pentium duallies aint cool at all dude.
Overclocking? Forgettaboutit.
Other than willamette this is probably the worst CPU intel has ever released.
I wonder is it good thing if CPU ignores application priorities set up by programs/os ?
the CPU doesnt.
The OS sets the priority based on what the program asks for and how many cores are available.
or thats what I read somewhere a long while back.. maybe it has changed, but doing differently would potentially waste resources..
I didnt read the whole toms thing, what was toms setting the priority level to when they were modifying it by hand for the X2? All of them should have been at the same priority, otherwise one task will have hogged the entire CPU and not balanced at all.
Taskmanager is not the way to handle application priorities, it just wont work properly (thats why THG failed by changing loadbalance).
Those priorities are built in the apps, and based on this info OS scheduler handle things.
But the hyperthreading seems to bypass priorities by giving equal amount of CPU cyckles to applications, - regardless of priorities - and that is bad IMO.
It's a hack, it makes OS believe that there is more CPU's, and in single CPU systems scheduler waits less resulting in smoother system, which i think is good.
But servers running database (or any app) should give higher priority to that application, and not to give equal priorities to lower priority apps (background tasks).
I just read Linux DB test at anandtech.com and it seems that HT is not good in servers.
http://www.anandtech.com/IT/showdoc.aspx?i=2447
So actually that THG testing has provided more understanding for me about HT and how it handles applications.
Im surprised to say that last one. :D
Dont reckon much to THG's Load Balance software it can vary from 50% to actually over 100% sometimes (all the while the overall load being near 100%).
I registerd here just to be able to post on this topic.
I sent THG an e-mail asking them why do they seem to 'help' intel and why had the DivX encoding slowed down on AMD platform.
Here's what they replied. I know it's old but my account required activation.
Quote
We don't help anybody...And the DiVx test runs on the AMD machine, but
slowly. We believe this happens due to the fact P4's HT-feature, but we
will disable HT within the next days to see if that is the reason why
the P4 is that much faster in this discipline.
they basically told you to go screw yourself Razi :D
sooo... didja do it? ;)
btw, welcome to XS, i'm pretty new around here myself, and the THG crap sorta aided in my decision to join here.
Quote:
Originally Posted by Zebo
Wasn't it you who said you'd buy a p4-d if their low end ones had HT?
Oh it's MUCH worse than Willy. You practically have to use nonstandard cooling just to run the damn thing. Willy's were slow but at least they worked.
Now for fugger who's gonna cascade the damn thing, sure it'll perform allright, though he could almost certainly get better performance if he put the same effort/money in an AMD setup. But for Joe Everyman who sticks a P4 Dual core in a 100*+ server room on stock or a 1/2u cooler stacked with ten others, well let's just say Intel will be lucky to get away without a lawsuit.
I kid, I kid.Quote:
Originally Posted by Zebo
Yay, I'm helping!Quote:
Originally Posted by SlicerSV
Well, no, i didn't do it. Wasn't in the mood. :DQuote:
Originally Posted by SlicerSV
They might have told me to go and screw myself but they sure said it with style, didn't they?
The sad part is when they announce their falsified results, many people will actually believe them. Why not lynch them in public for saying such crap.
:slapass:
:banana: :banana: :banana:
ok i dont know about you guys, but nice guys finish last. wehre is that intel, if it runs liek that, its means is one buh-had @$$ chip. its rebel without a cause. i too stray form the norm, and i need a chip that does the same thing. you only THINK its a performance device then BAM you get slapped in the face wiht a lazy 3 hour lunch break and the rest of the after noon is shot down. that isthe kinda no fear attitude i want MY new rig to have. you can have yoru numbers you silly over clockers, my new mahcien aint pretty or fast, but its got IN YORU FACE ATTITUDE.
--intel &THG new add caimpaign
--how do you spell campaign?
Aww this thread was too good to die. Did THG ever apologize for the crap they pulled?
lol what a resurrection....
no they didnt why they should have????? they never made mistake xD LOL
why should they apologize for being paid?
There is another update.
http://www.tomshardware.com/cpu/20050714/index.html
Yeah right.Quote:
The AMD system featured the Dresden-manufactured 90 nm Athlon 64 X2 4800+, which consists of two FX-53 cores each running at 2.4 GHz with 1 MB L2 cache.
Are they trying to say that AMD X2 power consumption is equal to 2 FX-53 CPU's ?
:rolleyes:Quote:
Four applications ran in tandem on both platforms over a period of 14 days. In this case Intel's top model - the Pentium 840 EE - easily outperformed its rival from AMD.
What a bunch of whores.
Quote:
Originally Posted by Hans.Gruber
Toms hardware is whom this article is talking about:
http://www.theinquirer.net/?article=22332
To a tee.
And when you think about it, Intel, bribing HW sites is small change compared to how they bribe OEMs which is factual.. called a rebate after excluding enough AMD chips... so I would'nt put it past them... Business can be dirty sometimes chasing the bucks.
BTW Hans Die Hard is like my favorite movie... Top 80's action flick!
going back in time a bit, I recall buying my first P4 1.8 and watching my 1.3 Tullie beat it...at everything (except memory bandwidth I had DDR after all ). I was more than a little upset. I relagated the P4 to a web browsing box for the old lady and switched teams.
Then I read that article at Tom's, how it was set up, what the nonsense was all about. And I figured out I still didn't want an Intel and bought a x2, so the article was helpful.
I'm still a little warm behind the ears about this, the article is still on the front page links there last I checked. It's still bad journalism. Twits.
It was hilarious at first. Then it just got sad. I thought of an analogy that I think works pretty well to explain the situation, but it was about children being sold in prostitution, so it probably wouldn't be appropriate.
jesus resurrection of thread lol
they shouldnt make another stress test of cpu if they want to keep some readers LOL