If you were going to spend $1000USD on a AMD X2 you would be interested in this thread.
Also the information coming out is important to understand what we can expect out of the two main Window XP desktop CPU makers.
.
If you were going to spend $1000USD on a AMD X2 you would be interested in this thread.
Also the information coming out is important to understand what we can expect out of the two main Window XP desktop CPU makers.
.
I must be an uber-genius then.
If a cpu thread is 100%, that doesn't mean an applciation thread is getting 100%. Other operating system routines could also be taking cpu time.
.
In intel's case:
28.4%+21.5%+25.8%+21.9% = 97.6%
Leaving 2.4% for "other operating system routines".
In amd's case:
43.3%
Leaving 56.7% for "other operating system routines"
What are these routines, and how can they eat up so much cpu time? I wanted an explanation, not a vague generalization.
You posted the stats, inconsistencies and all, now explain them.
<a target=_blank class=ftalternatingbarlinklarge href="http://www.tomshardware.com/cpu/20050603/stresstest-01.html">Sixth Day
Our observations suggest that the AMD system can't distribute the CPU's computing power as well among the four single applications as the Intel system does. So the big question here is: Why?
What are the reasons for the AMD dual core CPU being apparently unable to ideally distribute the four applications on the two cores, even though the load on the single cores is at maximum? Is this an issue of the integrated memory controller and its memory allocation/controller logic? Or perhaps the integrated memory controller of the X2 produces more overhead, resulting in lower performance?
</a>
Wasn't I warning people about the bad signs coming from the AMD X2's lack to do the tasks properly?
If you know people contemplating buying a AMD X2, warn that person about the design faults!
.
Our observations suggest that the AMD system can't distribute the CPU's computing power as well among the four single applications as the Intel system does. So the big question here is: Why?
What are the reasons for the AMD dual core CPU being apparently unable to ideally distribute the four applications on the two cores, even though the load on the single cores is at maximum? Is this an issue of the integrated memory controller and its memory allocation/controller logic? Or perhaps the integrated memory controller of the X2 produces more overhead, resulting in lower performance?
</a>
Wasn't I warning people about the bad signs coming from the AMD X2's lack to do the tasks properly?
If you know people contemplating buying a AMD X2, warn that person about the design faults!
.
It's probably called a crash. The AMD system has crashed if you go by the dialog on the Farcry window.
A crash can result in the code's intruction flow being looped.
.
I've seen unexplained dialog boxes before, yet tom's never reported a crash.
Since you're citing tom's 6th update as a relevant source of information then this should hold water right?
Unless they are faking the values then the information I read is the X2 is having problems and it's a major concern for anyone serious about using the X2 for business or professional use.
AMD X2=LEMON
.
AnandThenMan
Diamond Member
- Nov 11, 2004
- 3,991
- 627
- 126
Porkster=attention whore asswipe with no real knowledge of anything. For you to even quote this:
pretty much proves you are clueless. The above statement from THG is meaningless and flat out wrong. You are the laughing stock of the hardware community. U proud of yourself?
In the real world, the X2 and the Opteron are the envy of the processor business. It has won awards all over the word for best CPU, best design, best innovation, etc. etc. etc. It is being endorsed by basically every computing sector and is very highly regarded.
So for the last time, learn to comprehend above a 2nd grade level or STFU.
edit: BTW, everyone PLEASE ignore this vacant troll once and for all.
pretty much proves you are clueless. The above statement from THG is meaningless and flat out wrong. You are the laughing stock of the hardware community. U proud of yourself?
In the real world, the X2 and the Opteron are the envy of the processor business. It has won awards all over the word for best CPU, best design, best innovation, etc. etc. etc. It is being endorsed by basically every computing sector and is very highly regarded.
So for the last time, learn to comprehend above a 2nd grade level or STFU.
edit: BTW, everyone PLEASE ignore this vacant troll once and for all.
<a target=_blank class=ftalternatingbarlinklarge href="http://www.overclockers.com/tips00794">Just one snippet
Sales of servers with AMD's Opteron processor accounted for 5.7 percent of the 1.6 million unit x86 market. This was a slight improvement from the 5.4 percent market share Opteron held during the previous quarter, but still far from AMD's stated goal of 12 percent market share by year's end.
</a>
.
what a nice way to get besides the point, anyways, THG couldnt conclude anything relevant, I'm waiting for them to disable HT and then we'll see what happens.
Disabling HT is pointless. It makes the test totally biased towards AMD because if you wanted a 3.2GHz dual core Pentium vs an Athlon X2, you should use the MUCH MUCH cheaper 840 which is 3.2GHz with no HT, and also less than half the price of the X2 4800+.
I'm sorry, but this test is using 2 equal processors (the top end from both AMD and Intel) so disabling HT is bloody pointless and would only server to make the test more rediculous than it already is.
As far as determining how each "CPU" handles the load though, hopefully it will show Intel having the same problems as AMD when it comes to doing 4 threads on 2 cores while the OS can't handle it.
Did anyone bother to inform THG "engineers" (yes, the air-quotes are there for a good reason) that the divx process is being preempted by the other, CPU bound, higher priority processes, which results in thread starvation, and that and that alone is the true reason behind the AMD's ridicously low scores in the movie encoding test. It seems they are ready to blame the X2's memory controllers for what is clearly an OS scheduling problem.
well, i was up in the air, and wasn't sure if i wanted to upgrade or not. after seeing intel's attempt at what seems to be damage control, by using "THG" to, in my opinion, trying to hide it's flaws and make it shine brighter than it is, and also from the NUMEROUS forums around the net that have intel "supporters" coming out of nowhere with an attempt to skew the results in their favor. thanks to threads like this and another one under this forum, i will definitely be buying one of these X2's as soon as possible. the performance is staggering, and after what seems to be attempts on THG's part with using an unmatched HSF to make the X2 fail, they couldn't do it, although the same cannot be said about the DC from intel which neaded just the right mixture of hardware to make the thing work. that bodes well for AMD IMO. apparantley intel's hardware is not ready for the public yet, again, in regard to the results from THG and from the other reason i mentioned above. 99 percent of the reviews favor AMD's solution by a large margin.
anyway, despite these shanananigans by some posters and THG, in fact partially due to, i know many people including myself that are eager to get there hands on one of these gem's!
anyway, despite these shanananigans by some posters and THG, in fact partially due to, i know many people including myself that are eager to get there hands on one of these gem's!
now, I dont want HT to be disabled as a means of lowering performance or leveling the field. but mainly as a way to get to the bottom linne wether or not this is an OS scheduling issue. if they disable HT and suddenly about as much divX gets done as the AMD it'd be damnd hard to blame the AMD memory controller. that is why I want HT disabled, the performance numbers are rather worthless anyways. the only thing I'm intrested is in knowing where the issue lies. now IF they redo the test and the numbers are similar as now, (but lower maybe because of no HT) then we can conclude there is something wrong with the AMD processor in terms of thread handling, I dont see how that could be, and my money is on a similar behavior as AMD's dual cores. and those morans at THG still didnt mention at what thread priority the encoding runs. because the percentage of CPU time is about what i'd expect when it was set to a lower priority than the rest.
and as said before, if it's set to a lower priority there is no issue with the AMd chip, then that one is behaving just as it should, but then there is an issue with the intel chip, more precisely with windowsXP which would then be unable to handle a dualcore HT chip.
so a test with HT disabled will give me insight on the thread priority of the divX thread, something THG apparently does not want to reveal, so it's the only way to get that info.
and as said before, I'm much more intrested in the whole scheduling issue than in any performance numbers, because there is most likely a winXP sheduling issue and i'd think most people would like to know what one it is and with what proc it happens (intel or AMD).
Porkster STFU already!!!!!! You are making an A$$ out of yourself.
The following is from a new "Industry Update article Q2/2005" at this site, looks like people are finally starting to get educated and catch on, Intel's days of CPU domination may be numbered!!!!
----------------------
Seven months ago, on a trip to Taiwan, we went around asking all of the motherboard manufacturers that we encountered what their split was between AMD and Intel boards being shipped. At that time, we noted that despite what had been happening in the enthusiast community, motherboard manufacturers were still shipping mostly Intel based platforms. In fact, the split between AMD and Intel motherboards was similar to the 80/20 market share split between Intel and AMD - obviously, in Intel's favor.
This time around, the tune was very different. Note that only 7 months have passed since my last Industry Update, but a lot has changed in the market. Whereas the largest percentage of AMD motherboards shipped (that we heard) 7 months ago was 30%, this time around, it was 65% for desktop motherboards. Most motherboard manufacturers we talked to claimed that between 40 and 65% of their motherboard shipments were AMD platforms, not Intel.
Intel, of course, did not have much faith in our findings, stating that they are in direct conflict with widely reported market share numbers that have been made public in the past.
Our feeling is that the truth is somewhere in between; Intel continues to lead in OEM sales; however, AMD has made some significant gains across the market. Not to mention that our sources for these figures have little reason to lie blatantly about them, and not a single manufacturer that we talked to offered a vastly different story.
With the exception of the recently released dual-core CPUs, we have not recommended the Pentium 4 in over a year - it's not too surprising to see some of this reflected in motherboard shipments. To think that Intel has lost absolutely no ground to AMD in recent history is absurd in our opinion. AMD hasn't taken over the market, but they've surely grown to be far more than just a thorn in Intel's side. Just five years ago, motherboard makers were afraid to display AMD boards at their booths at Comdex or Computex, but today, AMD based motherboards are the most interesting and proudly displayed.
Although the discussion above applies to desktop motherboard shipments, AMD also appears to be doing quite well in the server market. We spoke to one server motherboard manufacturer whose current shipments are 90% AMD platforms and only 10% Intel platforms. We were absolutely shocked by these statistics, but it seems that most of the Intel server motherboards are being shipped by Supermicro with manufacturers like iWill and Tyan focusing much more on AMD.
-----------------------
http://www.anandtech.com/tradeshows/showdoc.aspx?i=2444
blah, the AMD fanboys make this duscussion just as hard as the intel trolls. so I'd cut the crap about AMD/Intel running the universe, they are buisnesses, my next proc is going to be an AMD but most likely the one after that will be an intel again. it doesnt matter, they are buisnesses they want your money, and we are clients we want the best for our money. so act like it and not like it is your long lost love.
Jeff7181
Lifer
- Aug 21, 2002
- 18,368
- 11
- 81
What a complete and total moron. The Athlon 64 doesn't schedule processes, the OS does. Hyper-Threading in the Pentium 4 takes over process scheduling from the OS... we knew that years ago when HT was introduced. Changing thread priority in Windows does nothing, because Hyper-Threading handles process scheduling on the Pentium 4.
It has nothing to do with memory controllers or bus speeds. It has everything to do with Hyper-Threading. Damn... you'd think the "engineers" at THG would be smart enough to realise that if I can.
OK update...It was the newtrok connection obviously...
Inyeresying now is Divx has not done any work in the last 12 hours...CD encoding too a mjor drop from 82 cd lead from INtel down to only 56....Winrar jumped back up to near 1500 lead and the only thing Intel has gained on is FarCry....It has widdled the lead down to 393....
Very funny all of a sudden it wants to be a gaming computer and sacrifice its leads in the areas it was winning by a safer margin...I think we have found AU...(rolling my eyes)
Inyeresying now is Divx has not done any work in the last 12 hours...CD encoding too a mjor drop from 82 cd lead from INtel down to only 56....Winrar jumped back up to near 1500 lead and the only thing Intel has gained on is FarCry....It has widdled the lead down to 393....
Very funny all of a sudden it wants to be a gaming computer and sacrifice its leads in the areas it was winning by a safer margin...I think we have found AU...(rolling my eyes)
Originally posted by: bullsh!tter
Hi everyone in this great forum.
First of all, I want you guys know that I come from the Toms hardware forum and I'm amazed of the pile of BS that "Porky" is posting in this respectable site (and in other sites like firingsquad, xtremesystems and many more). I think this guy is a loser named Fugger (if you go to the xtrmesystems forums you'll notice a pattern in the way this guy is posting. I have a little suspicion that Fugger (or Porkster or whatever his damn name is) is the same guy at Toms Hardware that posted the 6th. update stating that the poor peroformance of the X2 in DivX is due to the memory controller. Only a dumb ass like Porkster could say something like that and it's a coincidence that "Porkster" and the guy at THG are stating the same conclusion.
What do you guys think of this??
P.S: And for those "idiots" that still thinks that AMD processors are not reliable, here's something that will shut your mouths for good.
I thought that was weird too.
That says 2 things to me:
1.) AMD handle games better than Intel (Which we knew).
2.) SLI is pointless.
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
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
News NVIDIA and Intel to Develop AI Infrastructure and Personal Computing Products- Started by poke01
- Replies: 411
-
Facebook
Twitter