Intel's 65nm Processors: Overclocking Preview

[Nasgul]

http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2578


From Anand Lal Shimpi:
These days, the majority of Pentium 4s are just not very interesting. They are too hot, more expensive and under-performing compared to their AMD counterparts.

Majority not interesting? WTC does he mean? Interesting in what way? Can memory be interesting as well? Can my Network card be interesting as well?.......hhmmmmm!

They're too hot? I guess he doesn't know that Intel's HSF sucks so bad it shouldn't be included in the retail version but there should be a coupon for $25 off with the purchase of a Zalman or Thermalright XP HSF.......Anyhow, quit being so cluless and look at my sig......34c @ 4.0ghz, is that too hot for ya Anand Lal Shimpi? You know what's hot? Curry! now that's hot. So don't get the two mixed up........m'kay?

More expensive? well, I don't mind paying an extra $40 for my CPU (in a low priced range).

Underperforming? oh! I guess you must be refering to the "gaming" factor....well you need to talk to your buddy Kyle Bennett regarding "gaming" benchmarks dude........better yet, read the CPU (Computer Power User) magzine October issue (Anand, you're on page 28) which he describes you in the title of his article on page 30...........by the way, an excellent article.

Finally, let me guess, if Cedar Mill (nor Conroe) score 1fps less in gaming than any AMD, I guess you'll characterize it as: Underperforming and expensive, won't you? http://i6.photobucket.com/albums/y218/Nasgul/icon_smile_double_fingers.gif

Please, learn from Kyle Bennett. :slap:

[adas]

I think that is a pretty good rant...I mean, they are only hot if you are using a stock heatsink...and overpriced...last time I checked, the New M2 FX60 and X2 5000+ were about $1,100-$1,200..Now, I'm not sure, but I don't think that Intel is going to even chearge more then $1,050 for the new EE :stick:
Overpriced my anus...you can get a 650 for like 260$

[saaya]

they do cost more, you admitted it, they do run hot on the default heatsink, you admitted that too, and they are slower in games, you admitted that too... so whats the deal?

sure, its not a big deal, but where does he say it is?
its just small things but its all cons nevertheless.

interesting article, thx for the link :)
i kinda expected more from 65nm, but those are early chips, so its looking pretty good... no positive surprises but no negative ones either :)
curious how amd will respond! :D

[Nasgul]

they are slower in games, you admitted that too... so whats the deal?
its just small things but its all cons nevertheless.


I'm pretty sure there's no CPU Magazine in Germany but in any case then read what I mean by "gaming" benchmarks:

Editors & Benchmarks Suck'05 (http://home.comcast.net/~dvrh10/Kyle_Bennett.PNG) by Kyle Bennet.




I've said before (just can't remember at which forum?): Benchmarks are like "drag racing", sure a car can be faster in a given "TRACK" but take those two cars and put them on a regular road where speed becomes of not importance because there's a SPEED LIMIT!. Which is what Kyle Bennet is saying.........in regards of "gaming" benchmarks and giving the example of Doom3 which the damn games tops at 60fps which is E-X-A-C-T-L-Y what I said in May 29, 2005. Post # 30 & 36 (http://www.xtremesystems.org/forums/showthread.php?t=64110&page=2&pp=25). And you can tell post # 41 is full of crap! massive crap!.

I should've said that the 1fps less would've been in "those" gaming benchmarks and not "real" actual experience playing the game.........

slow? I don't think so, I still play Doom3 at 60fps and UT2K4 "no-regrets" ONS map at 25fps with my GeForce 6600 GT as I did with my GeForce 6800 GT and both systems using a 3.2ghz CPU w/1GB of RAM using the exact same settings in both video cards and system overall.

The "myth" that "X" processor is fater at gaming is pure BS, well, unless you're benching it but in real life, excecuting it, you won't see a difference from "X" processor to "X" processor. One thing is for sure "X" processor will never reach 4.5ghz and beyond! :fact:

Can you say Pentium D 830 @ 5.2ghz? (http://www.ocforums.com/showthread.php?t=416866)

:thumbsup:

[i found nemo]

they do cost more, you admitted it, they do run hot on the default heatsink, you admitted that too, and they are slower in games, you admitted that too... so whats the deal?


pwned

[death metal]

nasgul, can you have a larger pic of that image? i have a hard time reading :)

[saaya]

I'm pretty sure there's no CPU Magazine in Germany but in any case then read what I mean by "gaming" benchmarks:

Editors & Benchmarks Suck'05 (http://home.comcast.net/~dvrh10/Kyle_Bennett.PNG) by Kyle Bennet.




I've said before (just can't remember at which forum?): Benchmarks are like "drag racing", sure a car can be faster in a given "TRACK" but take those two cars and put them on a regular road where speed becomes of not importance because there's a SPEED LIMIT!. Which is what Kyle Bennet is saying.........in regards of "gaming" benchmarks and giving the example of Doom3 which the damn games tops at 60fps which is E-X-A-C-T-L-Y what I said in May 29, 2005. Post # 30 & 36 (http://www.xtremesystems.org/forums/showthread.php?t=64110&page=2&pp=25). And you can tell post # 41 is full of crap! massive crap!.

I should've said that the 1fps less would've been in "those" gaming benchmarks and not "real" actual experience playing the game.........

slow? I don't think so, I still play Doom3 at 60fps and UT2K4 "no-regrets" ONS map at 25fps with my GeForce 6600 GT as I did with my GeForce 6800 GT and both systems using a 3.2ghz CPU w/1GB of RAM using the exact same settings in both video cards and system overall.

The "myth" that "X" processor is fater at gaming is pure BS, well, unless you're benching it but in real life, excecuting it, you won't see a difference from "X" processor to "X" processor. One thing is for sure "X" processor will never reach 4.5ghz and beyond! :fact:

Can you say Pentium D 830 @ 5.2ghz? (http://www.ocforums.com/showthread.php?t=416866)

:thumbsup:

right, so theres no difference between an intel and an amd cpu when it comes to gaming, but the intel rules because its clocked higher...?
you need to chill out a lil, please dont call somebody elses post cr4p, thats close to flaming...

if you dont agree, fine, were not in china and can afford to disagree, but no need to call somebody elses opinion cr4p.

kyle bennet... well let me comment on what he wrote.
numbers and metrics make no sence because we dont know how fast a system is if it scores xx points or fps in some apps? wrong, he probably doesnt know and most people dont know, but the people who play games a lot or bench a lot DO know what those numbers actually mean, it gives them an idea of how fast the system is. and he claims he didnt use metrics? thats a joke right? hes using fps just like any other site, and instead of average and max fps he tries different resolution and detail settings, but those are still all numbers... ;)
its just easier to understand for people who know little about benchmarking and hardware, wich seems to suit the average [h] reader perfectly, but doesnt mean fps and benchmark points are bs.
and dont even get me started on his 20 scrawling colored lines on top of each other wich are supposed to make somebody see what system is faster... hows that an alternative to fps and benchmark results? its totally worthless if you ask me... the idea is great, but the way he does it its in no way an alternative.

do you happen to have the second part of this where he writes about how it should be done in his opinion?

i get what he means, its a real problem, but saying fps and benchmarks are totally useless is wrong...

oh and ill pretend i didnt read that comment about magazines in germany ;)

[cirthix]

ahh, theres a reason i'm going dothan (again)

[Wrench]

It sounds like nasgul doesn't like Anand because he said something that he didn't agree. I think Nas is having an inferiority problem due to the fact he owns an Intel system and most AMD systems will beat his rig.

[Disposibleteen]

I too read that article but Kyle Bennet and almost agree. Yes today's current benchmarks are flawed by nature but it is the best we've got right now. If you dont like it you should go get on the developing team at Futuremark and make a benchmark that actually means something more than just FPS.

[toledo]

It sounds like nasgul doesn't like Anand because he said something that he didn't agree. I think Nas is having an inferiority problem due to the fact he owns an Intel system and most AMD systems will beat his rig.

Man , don`t say that . You just started a flame war here :D .

[saaya]

no, thats not starting a flame war... he just voiced his opinion, and i actually got the same impression... its interestingly many, almost only actually, intel users who suddenly seem to think benchmark numbers dont matter... :D

i dont think we need a new benchmark, we just need better interpretion of the numbers by good editors.

wich is funny, because editors get pushed towards the wall more and more by people ciritcizing them, wich makes them hide themselves behind benchmark numbers, something you cant argue about, number a is higher than number b, period.

in the end its the editors not standing out enough for their opinions, and the fanboys bashing the editors, like you did nasgul, wich drives the editors to be very defensive.

so nasgual, the irony lies very deep within your post... your actually part of the cause of the problem you are so upset about :)

[ibby]

Man , don`t say that . You just started a flame war here :D .

true.. looks at toledos ring and thinks ownage.

every ones got tehir views i just like to read :fact:

[toledo]

i dont think we need a new benchmark, we just need better interpretion of the numbers by good editors.



Saaya i agree with you and also think the interpretion is a user task . As an Intel user i say the benchs do matter , but depends how you adapt those numbers to your daily needs , so a little bit of experience benching is required to figure out what those numbers mean. I guess the editors are afraid of this kind of interpretation because fanboys would kill they , so they prefer to hide behind numbers instead of having good discussions .

[Delirious]

Saaya i agree with you and also think the interpretion is a user task . As an Intel user i say the benchs do matter , but depends how you adapt those numbers to your daily needs , so a little bit of experience benching is required to figure out what those numbers mean. I guess the editors are afraid of this kind of interpretation because fanboys would kill they , so they prefer to hide behind numbers instead of having good discussions .


This is very true, its all about how you use the computer. If X computer is faster for what u do than Y computer or vise-versa then that computer is the one u should be using.

I have no brand loyalty, if one is better for what i need then thats what ill use.

Competition between manufacturers is good cause it gives us better products for cheaper.
Getting caught up in all the competition just causes' threads like this.

[Lithan]

Do you have a better computer if it scores 9562, 8.5 or a C+? Exactly. You have no clue because benchmark numbers mean nothing to you.

9562 > 8.5 ... AhhhhTaaaahs!


Doom 3 won't even allow you to see more than 60fps.


com_fixedTic -1

Hey, I just revalidated what you've essentially built your career on (only to later argue it's uselessness once several places started doing it better and more honestly than you.)

[agenda2005]

Nasgul
Your friend Kyle bennett might learn one of few things from this article http://www.extremetech.com/article2/0,1697,1854726,00.asp.

[i found nemo]

EDIT: flame removed
-saaya

lmao that is going in the sig :rofl:

EDIT: no its not :P
-saaya

[toledo]

EDIT: flame removed
-saaya

:rofl: .....:flame: No comments

[Delirious]

Nasgul
Your friend Kyle bennett might learn one of few things from this article http://www.extremetech.com/article2/0,1697,1854726,00.asp.


Thats what u would call a good article as well as a good way to test performance in games.

[i found nemo]

you are retarted. lithan it is 16 keystrokes ;)

[Mr_Slinky]

Lithan it is 16 keystrokes.
:slap: doh :slap:

[agenda2005]

Thats what u would call a good article as well as a good way to test performance in games.

Can you explain to us how you would perform the so called "GOOD TEST" in games if that is what u would rather not rate as a good article.

[Ugly n Grey]

@agenda
Either you misread that quote or I am confused

[Delirious]

Can you explain to us how you would perform the so called "GOOD TEST" in games if that is what u would rather not rate as a good article.

Your statement doesnt make sense, re read what i wrote and then read what you wrote. :slap:

I said it was a GOOD test not a BAD test.

[Lithan]

I didn't count enter. DOH!

[agenda2005]

@agenda
Either you misread that quote or I am confused

Dang! I did misfire and totally misread the statement. I agree with Derilious. That was a good real life performance analysis and they did their best to minimize the statiistical error which they admit cannot be overlooked.

[i found nemo]

Can you explain to us how you would perform the so called "GOOD TEST" in games if that is what u would rather not rate as a good article.


in laymens i think he wants an example of a bad review

[agenda2005]

Nasgul,
This is what Intel consider the best way of measuring gaming experience. http://www.silentpcreview.com/article268-page1.html

I beleive my previous link used the same methodology. How did Intel CPU fare in that scenario?

[cirthix]

Nasgul,
This is what Intel consider the best way of measuring gaming experience. http://www.silentpcreview.com/article268-page1.html

I beleive my previous link used the same methodology. How did Intel CPU fare in that scenario?


While that is a move in the right direction and I'd love to see benchmarks show minimum and average framerates, I will not trust any benchmark program provided by a company which makes a product for the market the benchmark program is aimed at. For example, I will not even consider Intel's benchmark in cpu performance, but if they made a hard drive (one of the few markets they don't have a direct interest in afaik) benchmark program, I would pay attention to the results because they do not beneift from the program being skewed towards a certain product or brand.

[Nasgul]

no, thats not starting a flame war... he just voiced his opinion, and i actually got the same impression... its interestingly many, almost only actually, intel users who suddenly seem to think benchmark numbers dont matter... :DLet me guet that part straight. He voices his opinion by saying all that stuff (which I could careless) and then points out about amd stuff and me having Intel which again, I could careless and Toledo says it could start a flame war which I could careless what the post says, but you said it's just voicing his opinion? how come when I "voice" my opinion is close to flaming? Isn't that ironic that one person's opinion can be translated into "voicing" an opinion while the other's into "close to" flaming? That's hypocritical if you ask me.

Just like I did say it back in May 29, benchmarks aren't important to me, they show numbers at which no one could notice the difference between two systems, like you can tell from 50fps to 100fsb? By far anything above 35fps cannot be noticed. I don't have a single issue playing any of the games. I'm just more concern in stability and reliability in my system than "game" benchmark numbers. BTW, did everyone know that Intel is still king and ruler of "gaming" benchmarks? Here look! (http://www.gamepc.com/labs/view_content.asp?id=pm780&page=6)......wow benchmarks matter now....w00t Intel!



wich is funny, because editors get pushed towards the wall more and more by people ciritcizing them, wich makes them hide themselves behind benchmark numbers, something you cant argue about, number a is higher than number b, period.
in the end its the editors not standing out enough for their opinions, and the fanboys bashing the editors, like you did nasgul, wich drives the editors to be very defensive.
Wow! so now it's my fault that they hide behind benchmark numbers...........They have a "job" to do but most important, they get "pay" for that job, either way, they care more about themselves and money. They can print whatever they want if the price is right, if I pay close attention or care to the most part for what they print......shouldn't I be having an amd system then? oh sorry, that's never going to happen, I don't care about reviews nor benchmarks.


so nasgual, the irony lies very deep within your post... your actually part of the cause of the problem you are so upset about :)You're going to have to elaborate where's the irony. I culdn't be any more tansparent than I have already been.

I'm not upset, I just point things out, I think Anand Lal Shimpi is full of it, just like THG but specially Mr. Shimpi, who wouldn't be? specially when he posts the fallowing in his site: "The Asus P5N32-SLI Deluxe is outstanding and deserves serious consideration if you are an Intel enthusiast. It is hard to quantify this statement, but after spending considerable time with this board, it is truly a cut above most other boards in both quality and performance." ..........Sorry but C-19 keeps being a huge pile of crap and I wonder how much nVidia is paying him??


i kinda expected more from 65nm, but those are early chips, so its looking pretty good... no positive surprises but no negative ones either :)
curious how amd will respond! :D

Curious? Let me kill your curiosity, here look: NO CAN DO 65nm (http://www.overclockers.com/tips00852/). Is people gonna have to wait for 4 years? :hehe:

I haven't bought the CPU issue of November, I haven't been to the mall lately.

As for everyone else, at least I have the decency not to post in any amd forum and your opinions mean NOTHING to me which by the way I don't read them 99% of the time, so don't waste your time trying to get my attention :slap: .....and yes, Intel rules!

[FUGGER]

My reply to Anandtech quote original post.

Since the creation of hyperthreading on Intel processors AMD fans have had ample stomping on Intel performance with single threaded applications. If someone were to run two instances of super pi, a HT enabled machine would have scored far better than one without. With several applications running sending data down the pipe, the HT machine is less distracted from the main processes. Those that understood or had the need for this technology adopted it quickly and without complaint.

Now that X2 is out you saw that performance numbers were behind FX 57 and accepted the fact that handling threads was important to that processor. We have had that luxury for years now with Hyperthreading. Trust me, I have had far more use out of my HT than any AMD gamer has had with 64 bit instructions.

Look on PCmark 2005 for FX cpu, look at the P4 non D amoungst the X2's

Intel has a true winner in the mobile line and that is going desktop in 2006, with dual core and dual cpu squeezing in under 32W from both cores that leaves me with a big stick to whack back at the horrid thermal spec of a single threaded FX.

Expect ~4Ghz w/400+ FSB, the chipset was originally designed for 333+

[Delirious]

Just like I did say it back in May 29, benchmarks aren't important to me, they show numbers at which no one could notice the difference between two systems, like you can tell from 50fps to 100fsb? By far anything above 35fps cannot be noticed. I don't have a single issue playing any of the games. I'm just more concern in stability and reliability in my system than "game" benchmark numbers. BTW, did anyone know that Intel is still king and ruler of "gaming" benchmarks? Here look! (http://www.gamepc.com/labs/view_content.asp?id=pm780&page=6)......wow benchmarks matter now....w00t Intel!

Are both the pentium M's overclocked in that test?

They seem to deliver impressive performance but i dont see what that test proves other than an oc'ed cpu is better than a non oc'ed one?

I thought this thread was about accurate testing and benching, and not another Intel vs AMD thread. :rolleyes:

[Lithan]

My reply to Anandtech quote original post.

Since the creation of hyperthreading on Intel processors AMD fans have had ample stomping on Intel performance with single threaded applications. If someone were to run two instances of super pi, a HT enabled machine would have scored far better than one without. With several applications running sending data down the pipe, the HT machine is less distracted from the main processes. Those that understood or had the need for this technology adopted it quickly and without complaint.

Now that X2 is out you saw that performance numbers were behind FX 57 and accepted the fact that handling threads was important to that processor. We have had that luxury for years now with Hyperthreading. Trust me, I have had far more use out of my HT than any AMD gamer has had with 64 bit instructions.

Look on PCmark 2005 for FX cpu, look at the P4 non D amoungst the X2's

Intel has a true winner in the mobile line and that is going desktop in 2006, with dual core and dual cpu squeezing in under 32W from both cores that leaves me with a big stick to whack back at the horrid thermal spec of a single threaded FX.

Expect ~4Ghz w/400+ FSB, the chipset was originally designed for 333+


It's not fair to compair choosing an A64 over a prescott to choosing an x2 over an fx. First off is the price differential. For half of what a fx57 costs, you can grab an x2 that will likely clock within 200mhz of it (and runs 600mhz shy @ stock). So you save money, lose very little in your single threaded situations, and gain an unprecedented (excluding true dual processor rigs) amount in multithreaded apps. Compairatively, Prescott vs A64, you paid more, lost a good amount single threaded, while gaining a good amount (but nowhere near where you're gaining with x2 over fx) multithreaded.

But in all honestly I think the largest thing that has changed is the time. Before X2's and P-D's... you really had to LOOK for situations where ht/dual core would have helped outside of a few apps specially coded to work with HT procs (mostly encoding programs if I'm not mistaked). Now that both manufacterers have made it clear that they wont be able to ramp speeds of individual cores up at the rate needed to keep the market and technology thriving, people buying X2's now are really buying under the expectation that the market will change so that dual cores ARE useful very quickly. Personally, I'm in no rush and certainly am looking forward to the p3/pm-descendant dual cores you talk about, since I was impressed with dothan, and would like to see what it does when intel pays a bit more attention to it.

[saaya]

Let me guet that part straight. He voices his opinion by saying all that stuff (which I could careless) and then points out about amd stuff and me having Intel which again, I could careless and Toledo says it could start a flame war which I could careless what the post says,
can you please rephrase this? i dont get what you mean

but you said it's just voicing his opinion? how come when I "voice" my opinion is close to flaming? Isn't that ironic that one person's opinion can be translated into "voicing" an opinion while the other's into "close to" flaming? That's hypocritical if you ask me.

I think Nas is having an inferiority problem due to the fact he owns an Intel system and most AMD systems will beat his rig.
thats what toledo said, now hows that a flame?

you keep calling people fanboys and make fun of their reading and writing skills, wich is kinda funny since your not spending any time to get your grammar and phrases right either.
please stop calling people fanboys, its just offending people and wont get us anywhere.

Just like I did say it back in May 29, benchmarks aren't important to me, they show numbers at which no one could notice the difference between two systems, like you can tell from 50fps to 100fsb? By far anything above 35fps cannot be noticed. I don't have a single issue playing any of the games. I'm just more concern in stability and reliability in my system than "game" benchmark numbers. BTW, did everyone know that Intel is still king and ruler of "gaming" benchmarks? Here look! (http://www.gamepc.com/labs/view_content.asp?id=pm780&page=6)......wow benchmarks matter now....w00t Intel! if you dont notice a difference between 35 and 100 fps, then get a 6600gt and a celeron d 2.4ghz. if that works for you, fine, but what makes you think you know the ultimate truth and everybody else is horrably misleaded?

and wow, an oced oentium m perfoms better than a stock a64 and pentium 4? amazing! :P


Wow! so now it's my fault that they hide behind benchmark numbers...........They have a "job" to do but most important, they get "pay" for that job, either way, they care more about themselves and money. They can print whatever they want if the price is right, if I pay close attention or care to the most part for what they print......shouldn't I be having an amd system then? oh sorry, that's never going to happen, I don't care about reviews nor benchmarks. right, if you dont care about reviews and benchmarks then why do you keep reading them, keep posting about them, and keep complaining about them? because you dont care about them?
and yes, you bashing anandtech or any other reviewer or site is aiding in editors and reviewers hiding behind the benchmark numbers.
i hope i dont have to repeat this yet again.


You're going to have to elaborate where's the irony. I culdn't be any more tansparent than I have already been.

I'm not upset, I just point things out, I think Anand Lal Shimpi is full of it, just like THG but specially Mr. Shimpi, who wouldn't be? specially when he posts the fallowing in his site: "The Asus P5N32-SLI Deluxe is outstanding and deserves serious consideration if you are an Intel enthusiast. It is hard to quantify this statement, but after spending considerable time with this board, it is truly a cut above most other boards in both quality and performance." ..........Sorry but C-19 keeps being a huge pile of crap and I wonder how much nVidia is paying him??if your not upset then please stop underlining text, use huge letters and multiple !!! and ???, wich usually indicates that a poster is very upset about something :D

well why dont you post your own reviews then, or collect infos about the mianboard from people who have this boards to point out what problems it has? try to build something up instead of destroying something somebody else build up. i have no idea about that mainboard so i cant really comment...

and the irnoy lies within your post because you bash a site and a reviewer, wich in return makes him hide even more behind benchmark numbers, wich is exactly what you are so upset about... thats what i meant with irony.


Curious? Let me kill your curiosity, here look: NO CAN DO 65nm (http://www.overclockers.com/tips00852/). Is people gonna have to wait for 4 years? :hehe:
hahaha, i stopped reading stroglois comments a long time ago... according to him the end of world is just hours away everytime he writes a story hahah :D
sorry, but i cant take that guy serious...
keep reading his blog and sooner or later you will see he just keeps on and on and on and everything just sucks so bad etc :D


Intel has a true winner in the mobile line and that is going desktop in 2006, with dual core and dual cpu squeezing in under 32W from both cores that leaves me with a big stick to whack back at the horrid thermal spec of a single threaded FX
you call the thermal spec of an fx57 horrable? it consumes around 60W i think, an 840 is close to 150W... wich is more than 2x an fx57.

i agree on the ht thing though, its a shame amd didnt go ht earlier... its not a big thing imo, but it would have been a nice add on.

the reason why ht was ignored mostly was not because it sucked though, or because the reviewers and editors sucked. its because microsoft and software and game developers have failed to make use of it. maybe intel didnt support them good enough, or maybe there just didnt get it. its a big shame and it put the software and hardware industry a big step back. because if we had multithreadded games and software and fully multihtreadded os earlier then we would all be running dual core or ht machines already.

but instead the dual core adoption is going sloooooow because theres no real app that shows a big use of two cores or ht... :/

same about games in gerneral btw, theres barely any game wich is using the true potential of todays hardware... any game that gets released nowadays can be played even on the crappiest 2 year old machines, wich is just slowing down the industry and not giving the enthisasts the awesome eye candy their mahines are capable of :/

there really is a big problem with software slowing down hardware atm, and im affriad it will continue to grow because software gets more and more complex but the investmants in software stay the same.

if intel would have supported a game developer team to create a multithreadded game engine wed all be enjoying multithreaded games already and amd would probably have had to adopt ht or something similar, or wed be having dual core cpus already even...

[Nasgul]

please stop calling people fanboys, its just offending people and wont get us anywhere. Like I said, it's hypocritical to view it one way only. The one who called a Fanboy to who it was you to me, so why are you offending me? (see below) Voicing his opinion.....right! Post #11 (http://www.xtremesystems.org/forums/showpost.php?p=1105803&postcount=11)

in the end its the editors not standing out enough for their opinions, and the fanboys bashing the editors, like you did nasgul, wich drives the editors to be very defensive.Now, go back a to page 1 and quote me on who I have called a fanboy.........as a matter of fact go find any of my post during the past 30 days and quote me!........AND in regards of you saying I'm starting to flame someone......remember, I'm just "voicing" my opinion, just like the fella on POST # 9 "voices" his.


if you dont notice a difference between 35 and 100 fps, then get a 6600gt and a celeron d 2.4ghz. if that works for you, fine, but what makes you think you know the ultimate truth and everybody else is horrably misleaded? What makes you think that I use my computer as a gaming console only? If I want that I can get an XBOX for a lot less $$$ and have less problems i.e. unreliable and unstable! Still, you're getting 60fps in Doom3 whether you have a P4 3.0ghz or FX-57, so you tell me where the misleading beggining at?

and wow, an oced oentium m perfoms better than a stock a64 and pentium 4? amazing! Hey! Bechmarks do matter......or suddenly they don't? wich one will it be? or is it because I'm poiting out that Intel processors are way better at gaming than any other?


right, if you dont care about reviews and benchmarks then why do you keep reading them, keep posting about them, and keep complaining about them? because you dont care about them?aI read them because I have to read something while I'm taking a crap and that is how I learned about computer hardware. :D I do multi-taks. I have a few magazines I haven't red yet and I today I bought the Nov. issue of CPU, I'll post Part 2 soon.


and yes, you bashing anandtech or any other reviewer or site is aiding in editors and reviewers hiding behind the benchmark numbers.
i hope i dont have to repeat this yet again.I just pointed out how ridiculous is to call a P.O.S product like C-19 (round 2) "a cut above other boards", in quality and performance? Should I guess that you wouldn't know, since that you may not own a P5WD2-Premium? It's still a fact that any C-19 board is horrible, unreliable and unstable.....people may fall for the crap that he posted and end up posting links like this one: SLi Intel Edition SUX! (http://www.xtremesystems.org/forums/showthread.php?t=70325) and Fugger kills one at 300fsb: SEE HERE (http://www.xtremesystems.org/forums/showthread.php?t=76566).......I could get a couple more from OCForums but I have no desire to show any more facts!


if your not upset then please stop underlining text, use huge letters and multiple !!! and ???, wich usually indicates that a poster is very upset about somethingMultiple !!! & ???, I'm just emphasizing.

well why dont you post your own reviews then, or collect infos about the mianboard from people who have this boards to point out what problems it has? try to build something up instead of destroying something somebody else build up. i have no idea about that mainboard so i cant really comment...I don't know about you but I do know that C-19 does nothing more than just bring more PCI-E lanes to the table, that's it. There's absolutely nothing that has changed so why would I waste time in finding out what's already known? I'm postive that Fugger will get one and kill it, yet againg another C-19 to the grave, the future is not hard to predict.

and the irnoy lies within your post because you bash a site and a reviewer, wich in return makes him hide even more behind benchmark numbers, wich is exactly what you are so upset about... thats what i meant with irony. So, because this is the first time ever I'm "bashing" Anandtech he's been hiding all along before my first time ever "bahsing" him.....Wow! I don't think so pal!.

hahaha, i stopped reading stroglois comments a long time ago... according to him the end of world is just hours away everytime he writes a story hahah :D
sorry, but i cant take that guy serious...
keep reading his blog and sooner or later you will see he just keeps on and on and on and everything just sucks so bad etc :Dhahahahahahahahahahah..........that's hilarious.......now you tell me why I can't take Anandtech seriously?.......real funny, Anand post crap about Intel and I ended up not liking it and you think I'm upset, yet I post something this guy says (bad news) about AMD not delivering 65nm anytime soon and you don't like it?.........hmmm how ironic!


And don't worry about how many watts a Pentium D consumes, when nowadays everone has an OCZ PowerStream 520W or any 500+ watts which are way more sufficient to handle any Pentium D at 5.2ghz. The 65nm CPU will comsume like 18% less power than a current P4 90nm, yet I'm about to buy this Power Supply (http://www.newegg.com/Product/Product.asp?Item=N82E16817194004). Intel rules! :D

[Overconfidence]

And don't worry about how many watts a Pentium D consumes, when nowadays everone has an OCZ PowerStream 520W or any 500+ watts which are way more sufficient to handle any Pentium D at 5.2ghz. The 65nm CPU will comsume like 18% less power than a current P4 90nm, yet I'm about to buy this Power Supply (http://www.newegg.com/Product/Product.asp?Item=N82E16817194004). Intel rules! :D

Damn. I wish I could join this group that you call 'everyone'. Oh, you mean 'everyone who owns a power guzzling pentium D'? Oh yeah, and there's also this thing that people pay... P... P... POWER BILL! Sorry, I forgot, 'everyone' is made of money.

[Nasgul]

Damn. I wish I could join this group that you call 'everyone'. Oh, you mean 'everyone who owns a power guzzling pentium D'? Oh yeah, and there's also this thing that people pay... P... P... POWER BILL! Sorry, I forgot, 'everyone' is made of money.

oh!!!! I'm sorry you're poor, I didn't mean to hurt your feelings but it's just that I'm used to think that everyone can afford an $80 to $120 PSU. I mean com'on! if people can buy a $400 video card, why buy a $20 PSU? lot of people does that and end up ruining their system....not good......

I don't worry about the electric bill, I pay it through my bank automatically, pretty much every bill gets paid through my bank so I never pay attention to Gas, Water nor Electric bill.......as a matter of fact, my phone bill used to be $70 because of all the "features" that we had but now I have Vonage and pay only $28 a month and no long distance charges, flat rate fee, same features as I used to have with Verizon and I don't dial 911, don't need to.
My Pioneer Elite VSX-53TX alone consumes 600W and I couldn't tell you how much power my whole Home Theater (http://img.photobucket.com/albums/v655/dob1224/HomeTheater.jpg) does but I'm sure it's a heck of a lot more than my pc. $175 for the last PSU I'll ever buy doens't seem like a bad idea. So far in my life I have bought 3 PSUs, two were $47 (TT 430W) and the most expensive is the current Antec SP2.0 500W I have which I bought at CompUSA for $60 three weeks ago. The holidays are coming, I'm sure you can find a PT job anywhere, work for two months and then quit and buy whatever you need.........hey! a 500W PSU wouldn't be a bad idea. Welcome to the club!

[Overconfidence]

Yeah, you know, MONEY!? Pshhhh. Who cares, automatic billing. I'll just.. well.. You can buy whatever you want with credit cards, right? Its like, free money. Who cares. Banks let you do loans and stuff, eh?

Hmm, oh, I'm sorry, being 14 years old doesn't allow me to get a job. But then, my parents wouldn't know anything about... paying bills, would they? Nahhh, they probably just charge it to the bank and don't care about it.

Somehow I'd rather buy a much better athlon 64 chip for the difference it costs to get an excessive power supply and pentium D that will cost me more in the end because I (or my parents) will have to pay the bills. Plus an aftermarket heatsink to overclock.

Who says I have a 400 dollar video card?

Please, convince me why, when I'm upgrading out of this XP-M system, I should buy a pentium D, new power supply, motherboard, video card, and DDR2 RAM instead of a much better X2 chip and nf4 motherboard that would end up being cheaper anyway. What about AGP to PCI-e? ULI chipset.

And you're also running your PC for a lot longer every day than your home theatre setup. I'm just going to go with your assuming attitude and say that EVERYONE leaves their computers on 24/7.

Plus, really, who needs to phone 911? Nothing will ever happen to me.

[zbogorgon]

i think fugger just said it all what it is to say about this mater.
Intel has dothan, witch consumes about 25W, on air cooling kicks AMD's FX ass in games, it's three times cheaper and now intel will launch DC dothans. What more can you want from your life. At probably half price, double performance at half consumption, with DDR2.

[Ugly n Grey]

that chip doesn't kick an FX unless you OC the crap out of it and then it's sure not burning 25W....

But the chip is a great design... all credit due.

However, I'm not buying Xeons and can't see why I would when I can buy Optys

[Overconfidence]

To be fair, Dothan chips are fairly expensive (though price drops lately have been nice), but yeah. I personally am looking at p3 derivatives for my next upgrade. Huge-ass pipeline p4s are NOT in my plans though.

[toledo]

that chip doesn't kick an FX unless you OC the crap out of it and then it's sure not burning 25W....

But the chip is a great design... all credit due.

However, I'm not buying Xeons and can't see why I would when I can buy Optys


Yes , it does . I have some game comparisons here .

FX57@3.16Ghz been kicked by a Dothan at 2750Mhz in games . Usually you need 300Mhz more on an A64 to mach a Dothan in games . And 2700Mhz is a average air overclock for a Dothan .

[Vapor]

Yes , it does . I have some game comparisons here .

FX57@3.16Ghz been kicked by a Dothan at 2750Mhz in games . Usually you need 300Mhz more on an A64 to mach a Dothan in games . And 2700Mhz is a average air overclock for a Dothan .Hmm....but when you consider that you can actually use SLI with the FX and not the Dothan it opens up a whole new can of worms.

Either way, I must say I'm very impressed with the potential of Presler as a cruncher.

[toledo]

Hmm....but when you consider that you can actually use SLI with the FX and not the Dothan it opens up a whole new can of worms.

Either way, I must say I'm very impressed with the potential of Presler as a cruncher.

I agree .....without SLI Dothan is only the single card 3D Bench king .

[Lithan]

Yes , it does . I have some game comparisons here .

FX57@3.16Ghz been kicked by a Dothan at 2750Mhz in games . Usually you need 300Mhz more on an A64 to mach a Dothan in games . And 2700Mhz is a average air overclock for a Dothan .

Can you link us? I've seen a lot of assumptions about where an overclocked dothan competes with an overclocked FX/Opty/SD clock:clock, but no actual testing of two directly compairable systems.

[cirthix]

this thread compares dothan, prescott, and athlon64, all overclocked.
http://xtremesystems.org/forums/showthread.php?t=75074
look at PcCI2iminal's results

there is a reason why people dislike intel's current desktop offerings and will go to great lengths to use pentium-Ms in their desktops.

[Overconfidence]

FX57@3.16Ghz been kicked by a Dothan at 2750Mhz in games . Usually you need 300Mhz more on an A64 to mach a Dothan in games . And 2700Mhz is a average air overclock for a Dothan .

That makes sense though.. isn't the P-M a 10-12 stage pipeline and the a64 a 14 (maybe 16, I forget) stage pipeline?

Also good to note that Dothan is a bit erratic in what apps it's good in and which it is not... At least from what I remember..

[saaya]

Like I said, it's hypocritical to view it one way only. The one who called a Fanboy to who it was you to me, so why are you offending me? (see below) Voicing his opinion.....right! Post #11 (http://www.xtremesystems.org/forums/showpost.php?p=1105803&postcount=11) where did i call you a fanboy?
and where did anybody in that quote call you a fanboy?

and you havent called anybody a fanboy?
Obviously, your reading and comprehensions skills are not your forte, then again, car sales man act the same way: They offer you what you don't want, in this case the "fanboys" have to hijack INTEL related threads. :nono:

When you recommend only amd and have an avatar like that........Sherlock has cracked the case.

Hey! I want either hot dogs or nachos to watch the movie..........(amd fanboy response is) no get pizza. :clap:
those were the posts you linked to in your second post in this thread. and i replied to that asking you to please calm down as your posts are close to flaming people.

Now, go back a to page 1 and quote me on who I have called a fanboy.........as a matter of fact go find any of my post during the past 30 days and quote me!........AND in regards of you saying I'm starting to flame someone......remember, I'm just "voicing" my opinion, just like the fella on POST # 9 "voices" his.[/COLOR]there you have your quotes, and no, its not the same thing as post 9 in this thread. and i didnt call you a fanboy, i said you bash the editors and reviewers like the fanboys do it. different thing.

[QUOTE=Nasgul]What makes you think that I use my computer as a gaming console only? If I want that I can get an XBOX for a lot less $$$ and have less problems i.e. unreliable and unstable! Still, you're getting 60fps in Doom3 whether you have a P4 3.0ghz or FX-57, so you tell me where the misleading beggining at? what "misleading" are you talking of? and please asnwer my question, if it works for you, fine, but what makes you think you know the answer to what everybody else wants to buy and add to his rig?

Hey! Bechmarks do matter......or suddenly they don't? wich one will it be? or is it because I'm poiting out that Intel processors are way better at gaming than any other? ermmm are you thinking out loud? your the one whos switching all the time and cant decide.
i dont get what your trying to say here...

I read them because I have to read something while I'm taking a crap and that is how I learned about computer hardware. :D I do multi-taks. I have a few magazines I haven't red yet and I today I bought the Nov. issue of CPU, I'll post Part 2 soon. right, so you really dont care about this but your even spending money on reading the latest reviews wich you hate so much because they dont mean anything?


I don't know about you but I do know that C-19 does nothing more than just bring more PCI-E lanes to the table, that's it. There's absolutely nothing that has changed so why would I waste time in finding out what's already known? I'm postive that Fugger will get one and kill it, yet againg another C-19 to the grave, the future is not hard to predict. oh, if everybody knows about this already then why do you keep raving about it? telling everybody what they already know...


So, because this is the first time ever I'm "bashing" Anandtech he's been hiding all along before my first time ever "bahsing" him.....Wow! I don't think so pal!.thats quite cynical and thinking in a very limited and short term action-reaction model.

hahahahahahahahahahah..........that's hilarious.......now you tell me why I can't take Anandtech seriously?.......real funny, Anand post crap about Intel and I ended up not liking it and you think I'm upset, yet I post something this guy says (bad news) about AMD not delivering 65nm anytime soon and you don't like it?.........hmmm how ironic!
one is based on real hardware we have right now, even though you might not agree with it and he might be biased and misleaded, hes basing his opinion on facts. that stroglio guy is all about speculation. pure speculation...
and the difference is i laugh about him and dont take it serious, i dont start a thread about it and rave about that guy sux so bad and his site sucks and they are all biased and liars and get paid by company x etc etc, second difference :)

And don't worry about how many watts a Pentium D consumes, when nowadays everone has an OCZ PowerStream 520W or any 500+ watts which are way more sufficient to handle any Pentium D at 5.2ghz. The 65nm CPU will comsume like 18% less power than a current P4 90nm, yet I'm about to buy this Power Supply (http://www.newegg.com/Product/Product.asp?Item=N82E16817194004). Intel rules! :Dpressler is what prescott should have been back then... running at ok temps and consuming an ok amount of power and ocing nicely and performing nicely. thats at least what ive heard of it.

i think fugger just said it all what it is to say about this mater.
Intel has dothan, witch consumes about 25W, on air cooling kicks AMD's FX ass in games, it's three times cheaper and now intel will launch DC dothans. What more can you want from your life. At probably half price, double performance at half consumption, with DDR2.
yonah will consume around the same as an fx57, cost almost the same and will perform around the same. you dont really gain any performence with the second core, its speeding up windows and multithreadding but not games and benchmarks. and yonah wont be clocked a lot higher than dothan, and wont perform a lot higher than dothan from what ive heard.

and dont forget that you dont need an fx57, there are opteron 939 146 chips that oc around the same as fx57 chips on air and cost only 200$ ;)
you cant beat that with a dothan setup. board plus adapter plus dothan cost the same or more as an opteron 146 939 setup.
and actually the only board for a dothan with pciE was never available in the us and has just been sold out in europe, i bought the last 20 boards for xs members on thursday last week ;)

dothan is nice, no doubt, but intel screwed up big time. they didnt allow mobo makers to build compatible mainboards for the high end market wich limited the desktop pentium m market highly, and made it very hard for enthusiasts to go dothan. intel shot itself in the foot not supporting dothan...

i hope they do better with yonah!
yonah will pretty much be an fx57 for a little less money. but by then the fx57 will cost less than it does now, and there will be an fx60 dual core... but i agree, its shaping up to be an interesting match. :)

[Jessfm]

I can't beleive i just read this thread thinking there was a point to it - what a waste of time and bandwidth.

Only the fool supports one who wishes to take his money..........

PERIOD

[toledo]

this thread compares dothan, prescott, and athlon64, all overclocked.
http://xtremesystems.org/forums/showthread.php?t=75074
look at PcCI2iminal's results

there is a reason why people dislike intel's current desktop offerings and will go to great lengths to use pentium-Ms in their desktops.


@ Lithan - The results are from PcCI2iminal . He will make a more complete review soon , i have some of the results with me but not all . But the link that cirthix posted is a good reference .

[saaya]

very nice results in that thread, i cant believe the dothan trash talk in that thread... a 6xx would have performed better... sorry but thats an excuse to feel not so bad about seeing a pressie stumped by a dothan.

id like to see some numbers though to get it proven, do you know anybody who can run a 5xx vs a 6xx at the same clocks at high speeds and test if theres a diference and how big it is?

[Nasgul]

where did i call you a fanboy?
:stick: In Post # 12 (http://www.xtremesystems.org/forums/showpost.php?p=1105864&postcount=12)
in the end its the editors not standing out enough for their opinions, and the fanboys bashing the editors, like you did nasgul, wich drives the editors to be very defensive.

and you havent called anybody a fanboy? Cryin' out-loud, that was me "saying" fanboy back in May, not October. As you can see, I'm "saying" the word fanboy, I'm not directing it to anyone in particular, like you did Saaya. :slapass: the fanboys bashing the editors, like you did nasgul.


those were the posts you linked to in your second post in this thread. and i replied to that asking you to please calm down as your posts are close to flaming people.Nah, lately I don't "try" to flame people, I just rather ignore their senseless posts (thoughts) and it doesn't bother me one least bit whatever they have to say about what I post.


and i didnt call you a fanboy, i said you bash the editors and reviewers like the fanboys do it. different thing. See! you're calling me a Fanboy again! Now, what if I were to say?: Saaya, you're like an amd fanboy, would you take it as me calling you an amd fanboy or saying you're like one? What exactly do I mean then? That you're not but you're like one? Now, are you or are you not?


right, so you really dont care about this but your even spending money on reading the latest reviews wich you hate so much because they dont mean anything? I really don't, I pay $12 for subscribing to this magazines, which it helps me relax and feel better while doing a #2. I would take the newspaper but the Washington Post is not quite easy to handle while you're sitting down in such small area, specially when I have a wall hitting my kneee. There's a few things here and there that you can find out through the magazines that I haven't through "review" sites, for example, did you know OPERA is now free? I didn't know that till last night.



one is based on real hardware we have right now, even though you might not agree with it and he might be biased and misleaded, hes basing his opinion on facts. that stroglio guy is all about speculation. pure speculation...
and the difference is i laugh about him and dont take it serious, i dont start a thread about it and rave about that guy sux so bad and his site sucks and they are all biased and liars and get paid by company x etc etc, second differenceRrriggght! we all hear and read what we want to hear and read, I can't argue with that. Except this time being somewhat vocal seems like it bothers people which it shouldn't, will you get mad if I said I hate Pepsi and it sucks and I'll never drink a can of it? Why would it bother you? if I'm bashing Pepsi. We all have differences of opinion, we all reserve our own right to critize what it seems to be "idiotic" (to the individual), you don't like what that guy says about amd not delivering 65nm anytime soon..........n' I love it! hear-no-evil, see-no-evil.


and I'd just like to point out once again that: Intel rules!

Don't be hatin' (http://www.overclockers.com/tips00852/)

[toledo]

very nice results in that thread, i cant believe the dothan trash talk in that thread... a 6xx would have performed better... sorry but thats an excuse to feel not so bad about seeing a pressie stumped by a dothan.

id like to see some numbers though to get it proven, do you know anybody who can run a 5xx vs a 6xx at the same clocks at high speeds and test if theres a diference and how big it is?


Saaya , the only major change in performance i notice between 6xx and 5xx P4s is the minimal fps in games . The 6xx family seems to keep it higher but it also does not clock as hight as the 5xx . I don't believe that talk either .

[railer]

um this thread is just getting stupid. Wasn't it to comment 65 nm overclock and not AMD vs Intel war, i mean i see those who's better threads all the time and i think they are just waste of my time.

But honestly A64 got nothing to match Dothan in gaming if Dothan sys is properly configured. Sure there is better support for AMD like nice SLI mb's and other but if you realy want you can make ur litle dothie do amazing things.

And aboutg Yonah oh boy thats a dark horse guys. Here is some intresting facts about it. I got this from Intel article :D

A single core Dothan would have required 4 clock cycles to perform a read-out when hitting a L2 cache line. With Yonah's Bandwidth Adaptation Buffer, it is possible to perform it in 2 clock cycles. As such, the available bandwidth stays the same, but access time is shortened. Yonah will also feature optimized pre-fetchers and a deeper write output buffer to further increase cache performance.


I like this man. Dothan eats games cause of it's low latency cache. Not as i understand this statement it will get reduced access time to 1/2 what Dothan has. Fast read acces and transfer is all that needed for games. And more powerfull cpu then Dothie in calc's sence. Could be great setup for gaming platform. The competiton must pullout something more serios then higher clocked cpu. Cant wait to overclock this with 2x sli's :slap:

And let me post some numbes how Dothie compares in games to A64 and Intel p4. That Doom3 timedemo was done 800/600 no AA no AF so VC would not be a show stoper , just pure cpu performance. Can your Opteron do that even @3.1g? From what i see its 400mhz clock per clock dif at least.

Be careful of Yonah Amd fanboys :stick: cause in 2 month's overclocked a64 just wont cut it, we might be talking 30% or more faster gaming performance then Dothan

http://www.xtremesystems.org/forums/attachment.php?attachmentid=39238&stc=1
http://www.xtremesystems.org/forums/attachment.php?attachmentid=39241&stc=1
http://www.xtremesystems.org/forums/attachment.php?attachmentid=39240&stc=1

[JoeBar]

I too believe that this thread is going way out of topic...

[saaya]

:stick: In Post # 12 (http://www.xtremesystems.org/forums/showpost.php?p=1105864&postcount=12) i already explained this in my last post, but here we go again. i didnt say fanboys like you. please ask you english teacher about the meaning of commas :P

Cryin' out-loud, that was me "saying" fanboy back in May, not October. As you can see, I'm "saying" the word fanboy, I'm not directing it to anyone in particular, like you did Saaya. :slapass: the fanboys bashing the editors, like you did nasgul.be carefull who you swing that stick at ;)
and let me say it once again:
i didnt say fanboys like you. please ask you english teacher about the meaning of commas :P

Nah, lately I don't "try" to flame people, I just rather ignore their senseless posts (thoughts) and it doesn't bother me one least bit whatever they have to say about what I post. right, you dont care about any of our senselss posts, thats why you quote all of them and spend 30 minutes to reply to them :D
just like you dont care about benchmarks but even pay for magazines printing benchmarks, right? :D
if you really dont care about peoples replies then why do you post in the first place? i get the feeling your just abusing xs as a place to post your daily blog.


See! you're calling me a Fanboy again! Now, what if I were to say?: Saaya, you're like an amd fanboy, would you take it as me calling you an amd fanboy or saying you're like one? What exactly do I mean then? That you're not but you're like one? Now, are you or are you not?
right, if saying somebody acts like xxx it means he is xxx? again, please ask an english teacher of your choice to explain this to you. :D

I really don't, I pay $12 for subscribing to this magazines, which it helps me relax and feel better while doing a #2. I would take the newspaper but the Washington Post is not quite easy to handle while you're sitting down in such small area, specially when I have a wall hitting my kneee. There's a few things here and there that you can find out through the magazines that I haven't through "review" sites, for example, did you know OPERA is now free? I didn't know that till last night. yes i knew that for weeks, we posted about it on the front page btw :D
your still not making much sence... you dont care about reviews but pay money to get them printed out. you dont care about replies but you quote and discuss them in long posts. intel and amd cpus are about the same and theres no big difference but intel is better.
your contradicting yourself repeatedly with everything you say.

Rrriggght! we all hear and read what we want to hear and read, I can't argue with that. Except this time being somewhat vocal seems like it bothers people which it shouldn't, will you get mad if I said I hate Pepsi and it sucks and I'll never drink a can of it? Why would it bother you? if I'm bashing Pepsi. We all have differences of opinion, we all reserve our own right to critize what it seems to be "idiotic" (to the individual), you don't like what that guy says about amd not delivering 65nm anytime soon..........n' I love it! hear-no-evil, see-no-evil.why you shouldnt bash anything you dont like? because people who do like it will get upset and if you keep provoking them your going to cause a flamefest. if you dont like it or even hate it then just say so... its not like anybody will give a lot... but you should realize that whats true for you isnt true for other people.

and amd wont get 65nm in time? who cares? :lol:
if they dont get their 4ss moving and intel beats them in performence and price then we will all be getting intel cpus... whats there to be upset about?
i dont like if one company dominates a market, so id probably always try to support the underdog, be it intel or amd or ati or nvidia, as long as they dont have any majoy flaws in their products like via or xgi.


and I'd just like to point out once again that: Intel rules!
right, not that anybody cares, but good for you ^^

Saaya , the only major change in performance i notice between 6xx and 5xx P4s is the minimal fps in games . The 6xx family seems to keep it higher but it also does not clock as hight as the 5xx . I don't believe that talk either .
well the 6xx is faster than the 5xx, but its minimal, less than 1% from what ive seen, and in some benches its even slower than the 5xx.
but id like to see somebody with a 5xx and 6xx to compare them at high clockspeeds... maybe somebody can lend this guy who compared a dothan and a 5xx a 6xx to compare?

railer, dothans dont perform that well in games because of the low latency cache, its about the same latency as an 64s afaik. its the cpus architecture, not the cache that makes it perform so well i think.

yepp, i dont see where this thread is going either. i dont wanna be an 4ss so ill leave it open for nasgul to reply and post a final rant and will probably close it then :D

[Vapor matt]

well ive just removed a 570j and replaced it with a 660, and at the same clock speed 320fsb 4480mhz this was the max speed i could the 570j 100% stable without throwing out strange benchys!

the 660 out performed the 570j, in 3dmark 2001 it made a diffrence of 1988 3d marks! I even put 570j back in to check and at lower clock speed the 660 still out performed the 570j.

that is just one bench mark i know but quite a diffrence wouldnt you say. I dont want to kick off a big argument over which is faster at what, but i know what i saw and i can get more speed from the 660 than the 570J

matt

[Nasgul]

i already explained this in my last post, but here we go again. i didnt say fanboys like you. please ask you english teacher about the meaning of commasYou're a real piece of work. So you didn't call me a fanboy but said I'm like one......you think the comma seperates the meaning of not calling me one but you're actually saying I'm like one, right! you're not calling me one, I'm just like one.

i didnt say fanboys like you. please ask you english teacher about the meaning of commasright, you think the comma seperates the meaning of not callimg me one.........I'm just like one. right!


if you really dont care about peoples replies then why do you post in the first place? i get the feeling your just abusing xs as a place to post your daily blog.Well, I'm just trying to prove how wrong you are in your thinking, other than that everyone else can post what they want. I'll stop posting till you realize that did in fact call someone a "fanboy". Unless you keep thinking once again that a comma means NO.....



right, if saying somebody acts like xxx it means he is xxx? again, please ask an english teacher of your choice to explain this to you. Shouldn't you be looking at the English dictionary on that?
Again, if you're saying somone is "like" one you're trying to say he "is" one......what part of that don't you get? Don't you know what the word "compare" means?
Here in America when someone says: That person is acting or being like an idiot, just like you.......it means that you're being called one also, whether or not you are one, you're still being called one. Now I'm not sure how the English in germany works but my guess it's completly different than American English.

and the fanboys bashing the editors, like you did nasgul, wich drives the editors to be very defensive.K, let's see, the fanboys bashing editors like I did, who's bashing editors? fanboys, and whose being like one?.......right I'm not being called a Fanboy...Rrrriiggghtt! Saaya, you din't call me one, no-no-no ;) you just think I'm one.


and amd wont get 65nm in time? who cares? :lol:Everybody. It's called competition. You think people doesn't want to see a 24-pipe ATI card?

i dont like if one company dominates a market, so id probably always try to support the underdog, be it intel or amd or ati or nvidia, as long as they dont have any majoy flaws in their products like via or xgi.Now, you know Intel, ATI and nVidia are not the underdog here. People will still support the "underdog" no matter what, even if Intel were to put out the best chip ever, Pentium M is by far the best CPU in terms of why people buy A64s "only"......games, yet I don't see them making the switch.

Supporting the "underdog" is a nobel cause but I'll pass on that.........Intel rules! If you think that not "anybody" cares, think again. This is the Intel Forum and it's an Intel world......amd, nah! I don't think so http://i6.photobucket.com/albums/y218/Nasgul/vommit.gif.

[Overconfidence]

Why should 'fanboys' 'watch out'?

Bah. I'm here for the best chips possible. Should people be called 'AMD Fanb0izz' just because they recognize that current AMD chips are generally accepted as better than p4s? I'm not gunna go 'oh noes' because Intel is coming out with a good chip, I'm happy. If they come out any sooner I'll definately go for the Mobile chip derivatives.

And just because he COMPARED (Simile: A comparison using like or as :P) you to a fanboy doesn't mean he's calling you a fanboy. REALLY.

EDIT: Actually, he said that you were bashing the editors. Fanboys also bash editors. Therefore you have some likeness to a fanboy. Most Fanboys have hair. Therefore you are probably like a fanboy. (Unless yer bald. In which case I'm sorry).

[Carfax]

railer, dothans dont perform that well in games because of the low latency cache, its about the same latency as an 64s afaik. its the cpus architecture, not the cache that makes it perform so well i think.

The large low cycle L2 cache in Dothan contributes heavily to it's gaming performance.. The A64 has a higher cycle L2 cache than Dothan aswell..

The L2 cache in Dothan isn't the only thing that contributes to it;s gaming performance. Other things like the really advanced prefetch also helps, but I think it's mostly the cache.

[saaya]

well ive just removed a 570j and replaced it with a 660, and at the same clock speed 320fsb 4480mhz this was the max speed i could the 570j 100% stable without throwing out strange benchys!

the 660 out performed the 570j, in 3dmark 2001 it made a diffrence of 1988 3d marks! I even put 570j back in to check and at lower clock speed the 660 still out performed the 570j.

that is just one bench mark i know but quite a diffrence wouldnt you say. I dont want to kick off a big argument over which is faster at what, but i know what i saw and i can get more speed from the 660 than the 570J

matt

hmmm thats a really big boost there... same cpu speed? so the 660 had a higher fsb, did it also have a higher mem speed?

You're a real piece of work. So you didn't call me a fanboy but said I'm like one......you think the comma seperates the meaning of not calling me one but you're actually saying I'm like one, right! you're not calling me one, I'm just like one. my apologies if thats how it sounds, i just meant to say that your doing what all the fanboys are doing. in this situation, not in everything you do. i guess you are an intel fan since you keep posting intel rules in almost every post, so i dont really get why you would be upset about beeing called an intel fan. other people dont like it because they dont want others to think they are biased about manufacturers. you keep explaining that you are biased and love intel, so i dont get how its upsetting you. please explain.


Well, I'm just trying to prove how wrong you are in your thinking, other than that everyone else can post what they want. I'll stop posting till you realize that did in fact call someone a "fanboy". Unless you keep thinking once again that a comma means NO.....
no, i didnt call you a fanboy, if thats how the sentence sounds to you or sounds to everybody im sorry, thats not what i meant to say in that sentence. right, so how am i wrong in my thinking?
im listening.

Shouldn't you be looking at the English dictionary on that?
Again, if you're saying somone is "like" one you're trying to say he "is" one......what part of that don't you get? Don't you know what the word "compare" means?
Here in America when someone says: That person is acting or being like an idiot, just like you.......it means that you're being called one also, whether or not you are one, you're still being called one. Now I'm not sure how the English in germany works but my guess it's completly different than American English. its not really an english thing actually, its the same in all languages. if you say somebody acts like xxx it doesnt mean he is xxx. otherwise there would be no need for those 2 sentences.
that dog looks like lassie. does it mean it is lassie?
that guy talks like BA barracus, is it mr t? probably not... :D
that guy acts like a retard, is he a retard? probably not, if he were then why would anybody state this anyways? would be like telling your friend "oh look, that thing over there is a tree"
if you say it looks like a try though your saying its probably or eventually not one but it seems to match some cirterias wich we define as "tree".

wich is why i said you are acting like the fanboys.
the fact that i have to explainthis basic grammar/rhetoric to you makes me wonder. i guess you were just kidding and trying to make me waste 10 minutes, well you won :D

Everybody. It's called competition. You think people doesn't want to see a 24-pipe ATI card? right, but nobody cares whether there will be a 64 pipe card in 2007. its not like it will influence anybodys upgrade decisions...

Now, you know Intel, ATI and nVidia are not the underdog here. People will still support the "underdog" no matter what, even if Intel were to put out the best chip ever, Pentium M is by far the best CPU in terms of why people buy A64s "only"......games, yet I don't see them making the switch. because upgrading to dothan is a pita?

Supporting the "underdog" is a nobel cause but I'll pass on that.........Intel rules! If you think that not "anybody" cares, think again. This is the Intel Forum and it's an Intel world......amd, nah! I don't think so http://i6.photobucket.com/albums/y218/Nasgul/vommit.gif.
did you ever get curious and tried an amd rig?


EDIT: Actually, he said that you were bashing the editors. Fanboys also bash editors. Therefore you have some likeness to a fanboy. Most Fanboys have hair. Therefore you are probably like a fanboy. (Unless yer bald. In which case I'm sorry).hahaha :D

The large low cycle L2 cache in Dothan contributes heavily to it's gaming performance.. The A64 has a higher cycle L2 cache than Dothan aswell..

The L2 cache in Dothan isn't the only thing that contributes to it;s gaming performance. Other things like the really advanced prefetch also helps, but I think it's mostly the cache.

http://www.x86-secret.com/articles/cpu/dothan/dothan-3.htm
dothan vs banias, pretty much identical cache latency, yet dothan performs better in games...

[saaya]

The large low cycle L2 cache in Dothan contributes heavily to it's gaming performance.. The A64 has a higher cycle L2 cache than Dothan aswell..

The L2 cache in Dothan isn't the only thing that contributes to it;s gaming performance. Other things like the really advanced prefetch also helps, but I think it's mostly the cache.

heres my a64 754 newcastle at only 1.6ghz, now please tell me how the dothans perform so well in games because of the cache latency...
its not the cache...
stride 4 8 16 32 64 128 256 512
size (Kb)
1 3 3 3 3 3 3 3 3
2 3 3 3 3 3 3 3 3
4 3 3 3 3 3 3 3 3
8 3 3 3 3 3 3 3 3
16 3 3 3 3 3 3 3 3
32 3 3 3 3 3 3 3 3
64 3 3 3 3 3 3 4 3
128 4 6 8 16 17 12 12 14
256 4 6 8 16 17 12 13 13
512 4 6 8 16 17 12 13 13
1024 4 6 11 25 44 81 83 89
2048 4 6 13 25 44 81 83 89
4096 4 6 13 25 45 82 84 92
8192 4 6 13 25 45 84 84 92
16384 4 6 13 25 45 84 86 91
32768 4 6 13 25 45 83 86 91

2 cache levels detected
Level 1 size = 64Kb latency = 3 cycles
Level 2 size = 512Kb latency = 12 cycles

[Astral7]

nasgul or coppi or what ever you are calling yourself this week, please just stop! You are an embarrassment to all intel users, I wish I could get my cpu idle temps to your credibility level(-100 K).


The reasons for Dothan's performance in games is short pipelines, but more importantly short *pipeline segments*, this in addition to optimized queueing and providiing hot bypasses when queues are empty. Then there is the low latency cache as well as cache coherency, and micro operations fusion.

Jonah will offer a performance increase, but not in the manner Merom/Conroe will bring.

[saaya]

you think conroe will be faster? hmmm you mean per clock? but its comming at only 2,9ghz i heard, so it better have a better ipc hehe :D
well if its as efficient as a pentium m or even more than that, and clocks at 2.9ghz that should be impressive :D

[wwwww]

heres my a64 754 newcastle at only 1.6ghz, now please tell me how the dothans perform so well in games because of the cache latency...
its not the cache...

Dothans have a 9ns L2, it gives them an advantage. Also IIRC, cache latency=pipeline length, the shorter pipeline also gives them an advantage, then again if it was all cache latency the p3s would be owning (7ns), they lack bandwidth though.

[oohms]

i cant wait till the conroe/yonah/merom come out :D
i hope they will have ironed out their poor performance in media encoding and that, but even if they havent, im still gonna buy one for their killer gaming performance ;)

[saaya]

Dothans have a 9ns L2, it gives them an advantage. Also IIRC, cache latency=pipeline length, the shorter pipeline also gives them an advantage, then again if it was all cache latency the p3s would be owning (7ns), they lack bandwidth though.
are you sure its ns? i dont think its ns because it depends on the clockspeed of the cpu. higher clockspeed lower latency. thats why cache latency isusually in clockcycles. its 7 for p3, 9 for banias, 10 for dothan and 12 for a64, maybe the venice/sadys have 11, dont know.

can anybody run the cache latency test on his p4 northwood prescott and prescott 2m? those would be interesting results! :)

the cache latency test is included in the cpuz zip btw :)

just run the app, then rightclick the dos console that shows the result and click "mark all" then ctrl+c and copy to a txt file.
when you post it here use code tags or it wont show up correctly, thx :)

[K.I.T.T.]

When is the ETA for yonah though? 2 months? :stick:

im fed up of waiting!! :stick: :stick:

[olfen]

this is my 2.4ghz xeon

stride 4 8 16 32 64 128 256 512
size (Kb)
1 2 2 2 2 2 2 2 2
2 2 2 2 2 2 2 2 2
4 2 2 2 2 2 2 2 2
8 2 2 2 3 7 2 3 2
16 3 3 7 15 22 18 19 19
32 3 4 8 17 19 18 19 19
64 3 4 9 18 18 22 18 30
128 3 4 9 18 18 19 19 19
256 3 4 9 18 19 19 19 20
512 3 4 10 20 24 39 41 43
1024 3 10 23 35 67 301 300 275
2048 3 10 18 36 67 254 252 306
4096 3 11 20 39 79 261 263 274
8192 5 11 21 40 76 267 255 275
16384 3 10 25 40 76 276 257 273
32768 3 11 21 36 68 258 292 279

3 cache levels detected
Level 1 size = 8Kb latency = 2 cycles
Level 2 size = 256Kb latency = 19 cycles
Level 3 size = 512Kb latency = 41 cycles

[wwwww]

are you sure its ns? i dont think its ns because it depends on the clockspeed of the cpu. higher clockspeed lower latency. thats why cache latency isusually in clockcycles. its 7 for p3, 9 for banias, 10 for dothan and 12 for a64, maybe the venice/sadys have 11, dont know.

can anybody run the cache latency test on his p4 northwood prescott and prescott 2m? those would be interesting results! :)

the cache latency test is included in the cpuz zip btw :)

just run the app, then rightclick the dos console that shows the result and click "mark all" then ctrl+c and copy to a txt file.
when you post it here use code tags or it wont show up correctly, thx :)

oops meant clock cyckes.
Northwood I think is 21 and prescott is 31. Read in someone's sig that prescott 2M is higher but I've never had one but it would explain why they clock better without any lesser power usage.

Just tested my dothan and yes its 10 clock cycles. Got 9 from my 1.6 banias in my laptop. Thought they would be the same.

[saaya]

check the xeon latency, its 19 and thats a northwood, a prescott must be really bad... like 25 :o

the chace isnt limiting prescotts, thats why a high latency on cache doesnt mean they oc any higher. the reason intel limited the cache latency has to do with the way cache gets addressed i think, and i heard something about yields too... maybe its hard to get 2mb cache that can be addressed quickly? or maybe its like with a memory controller where you need 2t for a certain amount of ram? dunno, but im pretty sure the cache isnt limiting prescotts oc speed. isnt there an option in bios to disable l2 cache? at least for a64s there is on the dfi boards. this doesnt make any sense, but it could be used to see if the cache is limiting the cpus max oc speed :)

now all we need is the cahc elatency readout from a prescott and prescott 2m :)
thx a lot olfen!

[K.I.T.T.]

Cache latency computation, ver 1.0
www.cpuid.com

Computing ...


stride 4 8 16 32 64 128 256 512
size (Kb)
1 5 5 5 5 5 5 5 5
2 5 4 4 4 4 4 5 4
4 4 5 4 4 4 4 4 4
8 4 4 4 4 4 4 4 4
16 4 4 4 4 5 5 5 4
32 6 8 11 20 31 28 28 28
64 6 8 11 21 28 31 28 33
128 6 8 11 21 28 36 28 35
256 8 8 11 20 29 29 29 29
512 9 8 12 21 29 30 31 40
1024 6 8 12 33 32 36 59 45
2048 6 14 15 27 65 156 306 292
4096 6 9 15 29 60 131 323 312
8192 6 8 15 29 56 166 300 301
16384 6 8 15 29 56 133 288 292
32768 6 8 15 28 73 132 288 359

2 cache levels detected
Level 1 size = 16Kb latency = 4 cycles
Level 2 size = 1024Kb latency = 33 cycles


1M Prescott (P4 540) :)

Now, someone answer my question: :slap:

When is yonah expected to arrive? january? If it does, would you get it, or wait for conroe? (conroe is LGA775 right?)

Also, would the next mobile chip after yonah be on the same socket?

:toast:

[saaya]

thx! :D
i cant really answer your questions though, sorry...
i think yonah will be out for christmas...
i can only point you to sams site, x86-secret, its all in french so trnslate it with babelfish or something else. he has a bunch of interesting news about new cpus usually.

LOL just saw your sig hahaha :D

[Zebo]

I'm pretty sure there's no CPU Magazine in Germany but in any case then read what I mean by "gaming" benchmarks:

Editors & Benchmarks Suck'05 (http://home.comcast.net/~dvrh10/Kyle_Bennett.PNG) by Kyle Bennet.




I've said before (just can't remember at which forum?): Benchmarks are like "drag racing", sure a car can be faster in a given "TRACK" but take those two cars and put them on a regular road where speed becomes of not importance because there's a SPEED LIMIT!. Which is what Kyle Bennet is saying.........in regards of "gaming" benchmarks and giving the example of Doom3 which the damn games tops at 60fps which is E-X-A-C-T-L-Y what I said in May 29, 2005. Post # 30 & 36 (http://www.xtremesystems.org/forums/showthread.php?t=64110&page=2&pp=25). And you can tell post # 41 is full of crap! massive crap!.

I should've said that the 1fps less would've been in "those" gaming benchmarks and not "real" actual experience playing the game.........

slow? I don't think so, I still play Doom3 at 60fps and UT2K4 "no-regrets" ONS map at 25fps with my GeForce 6600 GT as I did with my GeForce 6800 GT and both systems using a 3.2ghz CPU w/1GB of RAM using the exact same settings in both video cards and system overall.

The "myth" that "X" processor is fater at gaming is pure BS, well, unless you're benching it but in real life, excecuting it, you won't see a difference from "X" processor to "X" processor. One thing is for sure "X" processor will never reach 4.5ghz and beyond! :fact:

Can you say Pentium D 830 @ 5.2ghz? (http://www.ocforums.com/showthread.php?t=416866)

:thumbsup:
You need to look harder...Real Gaming Challenge: Intel vs. AMD (http://www.extremetech.com/article2/0,1697,1854788,00.asp)

You should be thankful for "drag racing" because the picture gets even worse for intel when we are not "drag racing" and under sustained load and for user enjoyment.

The results speak for themselves. The average frame rate across all six games for the Athlon 64 system is 61fps, while the Pentium 4 averaged 54fps. That's a 13% difference—not tiny, but not large enough to bowl us over. What is more important, we feel, is how often a game runs slowly enough that you can feel it. This methodology is consistent with the one used by a new performance analysis tool in the works at Intel. We picked arbitrary performance thresholds, but these are numbers based on years of game playing experience. We picked frame rates at which you actually notice an impact on how the game feels, not the absolute minimum required to play and enjoy a game. This is where the Athlon 64 really kicks the Pentium 4 in the teeth. Our P4 system spent almost a third of the time, across all games, beneath our target minimum FPS. The Athlon 64 system, on the other hand, spent only 14% of its time there. This is a difference of a whopping 121%!

[Nasgul]

my apologies if thats how it sounds, i just meant to say that your doing what all the fanboys are doingI do what a fanboy does? and that would be? Telling Mr. Shimpi that he's full of garbage? well he is and I just can't see how you can rate a person being a fanboy for telling an editor how full of garbage he is. Do we have a "you're a fanboy" chart in here?.

I think it's very idiotic to say things such as the one he posted: "These days, the majority of Pentium 4s are just not very interesting. They are too hot, more expensive and under-performing"......... If so how come there are tons of people getting the Asus 955X and P4s and Ds and overclocking the nutts of them? I'm yet to hear/see anyone blowing one up because of heat, and as far as underperforming? underperforming at what? Is there such software and programs currently on the market that a P4 or PD lags at?

Then he says: "Most exciting to us was the 4.25GHz overclock that we saw on Presler, as a 4.25GHz Pentium D will truly be a formidable opponent to AMD's Athlon 64 X2". <-----Wrong again Shimpi. Look at this THREAD PCMARK05 (http://www.ocforums.com/showthread.php?t=418591) top 10. This generation of Pentium Ds (8 series @ 4,005mhz) have enough ponies to square off with an X2 4800+ @ 2,880mhz. Damn! I can't wait to get my hands on a Presler.

How's my level of "fanboyism" so far?



please stop calling people fanboys, its just offending people and wont get us anywhere. i guess you are an intel fan since you keep posting intel rules in almost every post, so i dont really get why you would be upset about beeing called an intel fan. other people dont like it because they dont want others to think they are biased about manufacturers. you keep explaining that you are biased and love intel, so i dont get how its upsetting you. please explain.
My whole beef here is if you're going to "preach" then practice what you preach, I honestly don't care if anyone call me or labels me as a fanboy as you're doing it yet again!. Just because I like something and dislike the other, doesn't mean I'm a fanboy, does it?
biased: adjective;
showing an unreasonable preference or dislike based on personal opinion.
I'm transparent, not biased. I have many reasons why I like Intel the most and not "other". Everyone has their reasons, there's no such thing as being unreasonable and everyone can like and dislike whatever they want to, if you ask me I think that term is redundant but everyone here loves to use it at free will.


right, but nobody cares whether there will be a 64 pipe card in 2007. its not like it will influence anybodys upgrade decisions...Oh! but it will to some.


did you ever get curious and tried an amd rig? Yes I did get curious but that curiosity was killed with the P5WD2-Premium, 8000ULs and 650 @ 4.0ghz and I'm 100% positively sure that my gaming (if any, barely) experience would've not changed absolutely ONE bit by getting an AMD based system.

As miniscule performance differences become the new standard, I'd rather stick to Intel as their Core Logic are second to none with it comes to: easy to overclock, reliability, stability and compatibility......too bad not many can't say that about nforce 4 chip-set in either camp. Intel based system don't need that super-low cas latentices memory modules nor stinking "command rate" to be fast. Put any brand, any size and any latency memory modules and off you go, full speed ahead.

Intel rules!

P.S. Keep dreaming Zebo, you're still getting 60fps in Doom3. :hehe:

[IluvIntel]

Cache latency computation, ver 1.0
www.cpuid.com

Computing ...


stride 4 8 16 32 64 128 256 512
size (Kb)
1 5 5 5 5 5 5 5 5
2 5 4 4 4 4 4 5 4
4 4 5 4 4 4 4 4 4
8 4 4 4 4 4 4 4 4
16 4 4 4 4 5 5 5 4
32 6 8 11 20 31 28 28 28
64 6 8 11 21 28 31 28 33
128 6 8 11 21 28 36 28 35
256 8 8 11 20 29 29 29 29
512 9 8 12 21 29 30 31 40
1024 6 8 12 33 32 36 59 45
2048 6 14 15 27 65 156 306 292
4096 6 9 15 29 60 131 323 312
8192 6 8 15 29 56 166 300 301
16384 6 8 15 29 56 133 288 292
32768 6 8 15 28 73 132 288 359

2 cache levels detected
Level 1 size = 16Kb latency = 4 cycles
Level 2 size = 1024Kb latency = 33 cycles


1M Prescott (P4 540) :)

Now, someone answer my question: :slap:

When is yonah expected to arrive? january? If it does, would you get it, or wait for conroe? (conroe is LGA775 right?)

Also, would the next mobile chip after yonah be on the same socket?

:toast:


2M Prescott - cheap as chips (pardon the pun !) 630 cpu.
Cpu-Z latency test:

[cirthix]

wow, this thread turned nasty fast. lets just say that at the moment, here's the situation...

we have three competing cpus for single core and two for dual core, i'll lay it out for you (when i use > signs, i mean better than)
dothan+prescott+san diego(venice is nearly the same, just less cashe).

in games, dothan>sandiego>prescott
when overclocked...
in games, dothan>sandiego>prescott

who cares about encoding performance? set it to run overnight, if it finishes before you wake up, its fast enough :) games are the only real stressors of computers now where performance is needed.


heat-wise,
dothan>sandiego>prescott
when overclocked,
dothan>sandiego>furnace>prescott

simply put, prescotts put out too much heat.

price wise, buying new:
sandiego>dothan>prescott
buying used:
dothan>sandiego>prescott

low end dothans sell for very cheap on ebay, which turns the tables a bit.

when things go to dual core, we have p-d and x2

for games,
x2>pd
when oc'd
x2>pd

heat-wise
x2>pd
when oc'd
x2>pd

price wise
pd>x2
when used
not sure

the price difference between x2 and pd is about 100 bucks new at the lowst end, but its certainly worth it to upgrade. i'm talking about the lowest end here, if we wanted comparable processors in terms of model numbers, the x2 would be cheaper (3800+ vrs 3.2g dual core)

sure, the dothan is great, but it has some problems. the a64 is the next best way to go, and i'm not seeing these slightly-better-than-prescott 65nm chips changing that.

[Nasgul]

you still don't have a clue.

Clue! (http://home.comcast.net/~dvrh10/Kyle_Bennett.PNG)

When you've finished a game, do you also count the frames per second you did get during your gaming session as well or count the ones that are more than enough to run a game?


A64 the best way to go? 65nm not slighty better than Prescott?

hhhmmm..........Can you said 4.5ghz 250fsbx18 (http://images.anandtech.com/reviews/cpu/intel/65nmOCpreview/cedarmilloc.jpg) or 250FSBx17 (http://images.anandtech.com/reviews/cpu/intel/65nmOCpreview/presleroc.jpg)? and using stock "voltages" (& cooling).....This setting will definitely put the scores a lot closer than the puny 18% overclock we saw back there. Now we're in business and will have a fair comparison.

How much overclock percentage you can do on an A64 or FX with stock voltage? Certainly can't get 250fsb x full multi and using 1.3v, can it?

[Rocket]

Damn. I wish I could join this group that you call 'everyone'. Oh, you mean 'everyone who owns a power guzzling pentium D'? Oh yeah, and there's also this thing that people pay... P... P... POWER BILL!

Out of all the stupid posts in this thread, this has to be the funniest!

Same people that pay a $1.30 for coffee @ TimHortons complain about hydro bills...........lol

[zabomb4163]

why is a troll like Nasgul still a member of these forums?

-thats right i'm calling you a troll and to prevent any confusion about what that means ill paste a definition from the forum code of conduct in here .

"Anyone entering the forum with the express intent to cause trouble or harm"

and you know what......I use a p4 in my main system and a pentium M in my laptop and i still think your post on this forum are just an attempt to insite arguments from others. LEAVE. go to [H]

[AkXb70]

After seeing this thread and noticing a few things, i HAD to register...

Someone mentioned the FX57's TDP as 60w??? i find that hard to believe as AMD states the TDP is at 104w...i would think they know their own processors
http://www.amdcompare.com/us-en/desktop/Default.aspx <check there

also, someone else was comparing pipeline lengths...well heres the breakdown:
K7 = 10
K8 = 12

P6 = 12 (this includes banias and dothan)
Netburst 180 and 130nm = 20
Netburst 90nm = 31
New Arch = 14

dothan vs banias, pretty much identical cache latency, yet dothan performs better in games...

well...there are other enhancements that make dothan faster. it has double the L2 (2MB vs 1MB), and enhanced prefecth instructions in addition to better micro-ops fusion

as for the p4vsk8 debate...well, netburst wasnt always bad, remember the P4C's vs k7? the AXP's couldnt touch the p4's...both companies have their high and low points. this is intel's low point (although highly clocked p4's can still be compeititve even agaist o/ced a64's). i think around 2H 06 things should become more interesting. I honestly dont think yonah is what intel fans are waiting for...but we'll see

for those who dont remember:
http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=1834

[IluvIntel]

why is a troll like Nasgul still a member of these forums?

-thats right i'm calling you a troll and to prevent any confusion about what that means ill paste a definition from the forum code of conduct in here .

"Anyone entering the forum with the express intent to cause trouble or harm"

and you know what......I use a p4 in my main system and a pentium M in my laptop and i still think your post on this forum are just an attempt to insite arguments from others. LEAVE. go to [H]

"Anyone entering the forum with the express intent to cause trouble or harm"

Where does it say "Troll" in that statment?

Nasgul has the right to express his opinion like anyone else here, if we could'nt do that, then XS.org would go out of business.

The words "Trouble" and "harm" can have different meanings to different people. To clear this up, it is necessary to be pedantic about meanings of words. After all, that is what the legal industry does as a cornerstone in its conduct of business.

[mdzcpa]

Thats quite enough. Keep it on topic and keep it civil please. Last nice warning.

Thanks.

[PetNorth]

After seeing this thread and noticing a few things, i HAD to register...

Someone mentioned the FX57's TDP as 60w??? i find that hard to believe as AMD states the TDP is at 104w...i would think they know their own processors
http://www.amdcompare.com/us-en/desktop/Default.aspx <check there

yeah, but Intel & AMD TDP measurement is different:

Intel TDP: tipical power consumption. It can be higher (and it is).
AMD TDP: max power consumption in the worst of the circumstances. It can be lower (and it is, in fact it is much lower).

Example:

http://www.lostcircuits.com/cpu/amd_x2-3800+/prime95.gif

http://www.lostcircuits.com/cpu/amd_x2-3800+/10.shtml

[AkXb70]

yeah, but Intel & AMD TDP measurement is different:

Intel TDP: tipical power consumption. It can be higher (and it is).
AMD TDP: max power consumption in the worst of the circumstances. It can be lower (and it is, in fact it is much lower).

Example:

http://www.lostcircuits.com/cpu/amd_x2-3800+/prime95.gif

http://www.lostcircuits.com/cpu/amd_x2-3800+/10.shtml
i know their measurements are different...which is why i didnt dispute the 150w+ claim on the P-D's (intel rates them at 130w)

and that is the ONLY graph i have seen that rates a64's power consumption figures so low (i mean really, if a 2.4GHz venice consumed only 30w load, you could run it with a little NB cooler just like a 27w dothan)...most other reviews show power draw much closer to the amd specs ;)

[zabomb4163]

you could run it with a little NB cooler just like a 27w dothan)

-what about a BBA ram cooler?
http://www.xtremesystems.org/forums/showthread.php?t=76107

http://www.overclocking-masters.com/Photos-Screens/undercool-2.wmv
(sorry, couldnt find somone using a chipset cooler)

[AkXb70]

-what about a BBA ram cooler?
http://www.xtremesystems.org/forums/showthread.php?t=76107

http://www.overclocking-masters.com/Photos-Screens/undercool-2.wmv
(sorry, couldnt find somone using a chipset cooler)
thats at 1.1GHz and undervolted, so it doesnt count, as i said 2.4GHz :p:

there was a french site that ran a ULV dothan without a hs or fan at all

look, im not saying that amd chips run hot...i know they run rather cool (look at turion for example...although they use slower swithching transistors for even less heat at the same vcore/clocks) but they dont run 30w full load @ 2.4GHz

[zabomb4163]

thats at 1.1GHz and undervolted


didnt watch the video did we? initially he wanted to see how low he had to clock it to run without a heatsink. after that he put the BBA ramsinks on and tried to see how far it would overclock.

go back and watch the video this time. that is not an underclocked amd64 3000+.

[AkXb70]

didnt watch the video did we? initially he wanted to see how low he had to clock it to run without a heatsink. after that he put the BBA ramsinks on and tried to see how far it would overclock.

go back and watch the video this time. that is not an underclocked amd64 3000+.
i did watch it...it ended right after he showed the cpuz shot at 1.1GHz :confused:

EDIT: cpu-z looked like 1066Mhz to me at first...i looked at it again though, and youre right, its at 1.866 ;)

hmm, well the 3000+ is still in the lower part of the thermal envelope, (ie its still over 500MHz away from the 2.4GHz chip) but its still remarkable

do you have any more reviews/people with power draw figures?

[zabomb4163]

http://www.overclocking-masters.com/Photos-Screens/undercool-2.wmv

click that link. go to the 17 second point. press pause. in the bottom left hand part of the picture you will see 18 followed by 65 in red letters.

now go to the 25 second mark and press pause. you will see the mulitiplier as being 9 and the HTT as being 207. core speed 1856

33 second mark again we see 1865.

*edit* i made this post before you edited yours

[PcCI2iminal]

Can you link us? I've seen a lot of assumptions about where an overclocked dothan competes with an overclocked FX/Opty/SD clock:clock, but no actual testing of two directly compairable systems.

here we go

http://img34.imageshack.us/img34/8929/imagem0412ml.jpg
http://img34.imageshack.us/img34/8472/imagem0290ci.jpg

Cinebench
FX57@ 3160mhz
http://img220.imageshack.us/img220/1656/cinebenchfx571tx.th.jpg (http://img220.imageshack.us/my.php?image=cinebenchfx571tx.jpg)
Dothan @ 2750mhz
http://img220.imageshack.us/img220/7453/dothancinebench4cu.th.jpg (http://img220.imageshack.us/my.php?image=dothancinebench4cu.jpg)

F.E.A.R. MP Demo
FX57 @ 3160mhz
http://img220.imageshack.us/img220/6391/fearminimo8vx.th.jpg (http://img220.imageshack.us/my.php?image=fearminimo8vx.jpg)
Dothan @ 2750mhz
http://img220.imageshack.us/img220/5403/dothanfearmpminimo8qb.th.jpg (http://img220.imageshack.us/my.php?image=dothanfearmpminimo8qb.jpg)

DVShrink3.2 4.7gb from 8.7gb
FX57 @ 3160mhz
14m47s no screen
Dothan @ 2750mhz
http://img220.imageshack.us/img220/4946/dothandvdshrink3eh.th.jpg (http://img220.imageshack.us/my.php?image=dothandvdshrink3eh.jpg)

the review is finished n i'm workin with Toledo to rebuild the review in english,js wait a bit

sry for the out topic.

[agenda2005]

PcCI2iminal
You must really really hate FX-57. Why calling the FX-57 a CRIMINAL. Are you an Intel employee or one of those rebate/Intel exclusive dealers?

[PcCI2iminal]

PcCI2iminal
You must really really hate FX-57. Why calling the FX-57 a CRIMINAL. Are you an Intel employee or one of those rebate/Intel exclusive dealers?


man leave me out of this things i like Dothans and A64 and i dont't care about who make the chips

Edit -Criminal is my nick.

[zabomb4163]

man leave me out of this things i like Dothans and A64 and i dont't care about who make this chips

thanks for being unbiased. too bad more members wont take that viewpoint. whoever makes the best chips go for them.

the dothan is an incredible processor but the platform makes it a hard choice for some. if it were dual core and on the 775 platform i think more members on xtreme would go for it.

[PcCI2iminal]

thanks for being unbiased. too bad more members wont take that viewpoint. whoever makes the best chips go for them.

the dothan is an incredible processor but the platform makes it a hard choice for some. if it were dual core and on the 775 platform i think more members on xtreme would go for it.

well Dothan is not a great choice but u need to know what do u want,
to play games and not heavy tasks a cheap p4p and ct-479 can be the best for u,if u are looking for the heavy tasks,games and 64bits OS prescotts with emt64 or a64 are the way to go.

dothan/ct479 =2 more years of life from now,same for 939/754 and same for 775/i925/i955

[AkXb70]

thanks for being unbiased. too bad more members wont take that viewpoint. whoever makes the best chips go for them.

the dothan is an incredible processor but the platform makes it a hard choice for some. if it were dual core and on the 775 platform i think more members on xtreme would go for it.
didnt some motherboard company say they were going to make an i975x solution for yonah? i think it was aopen...but i dont remember atm

i really think it sucks that intel didnt put pentium-ms on the same socket as p4's. that would really have let them shine

[Peen]

Why does everyone go so worked up about this stuff? Seriously what do computers do thats so special? Im looking at my PC right now, alll that money into it, but it still does nothing! the CPU is just sitting there in its socket all quite doing nothing. When is it going to pop out of there and be useful and make me some damn dinner!

[PetNorth]

i know their measurements are different...which is why i didnt dispute the 150w+ claim on the P-D's (intel rates them at 130w)

and that is the ONLY graph i have seen that rates a64's power consumption figures so low (i mean really, if a 2.4GHz venice consumed only 30w load, you could run it with a little NB cooler just like a 27w dothan)...most other reviews show power draw much closer to the amd specs ;)

Well, this is the only site I've seen where only cpu consumption is measured (for example, Tech Report, Anandtech etc measure the whole system consumption).

[AkXb70]

Well, this is the only site I've seen where only cpu consumption is measured (for example, Tech Report, Anandtech etc measure the whole system consumption).
proably because total system consumption is easier to measure accurately ;)

after seeing that 1.87GHz venice running on that BGA hs....i think its possible that the lower echelon 67w envelope prcessors consume far less than 40w....but i still dont think a 2.4GHz venice at rated vcore consumes less than 60ish w (this is the top teir of the 67w thermal envelope...after this it moves to 89w at 2.6GHz, but that is for san diego...at 2.8 it is rerated again to 104w)

[PetNorth]

I suppose it's difficult to insolate the cpu consumption only, but it's useful to have some approximate idea about real power consumption, because TDP isn't a valid measure for this purpose.

I've found another site measuring cpu power consumption only, XbitLabs. They use S&M to stress the CPU (this program stress it much more than Prime):

http://www.xbitlabs.com/articles/cpu/display/athlon64-fx57_3.html

http://www.xbitlabs.com/images/cpu/athlon64-fx57/cons.png

[AkXb70]

I suppose it's difficult to insolate the cpu consumption only, but it's useful to have some approximate idea about real power consumption, because TDP isn't a valid measure for this purpose.

I've found another site measuring cpu power consumption only, XbitLabs. They use S&M to stress the CPU (this program stress it much more than Prime):

http://www.xbitlabs.com/articles/cpu/display/athlon64-fx57_3.html

http://www.xbitlabs.com/images/cpu/athlon64-fx57/cons.png
thats alot close to what i would have thought (on both sides) ;)

[The Stilt]

are you sure its ns? i dont think its ns because it depends on the clockspeed of the cpu. higher clockspeed lower latency. thats why cache latency isusually in clockcycles. its 7 for p3, 9 for banias, 10 for dothan and 12 for a64, maybe the venice/sadys have 11, dont know.

can anybody run the cache latency test on his p4 northwood prescott and prescott 2m? those would be interesting results! :)

the cache latency test is included in the cpuz zip btw :)

just run the app, then rightclick the dos console that shows the result and click "mark all" then ctrl+c and copy to a txt file.
when you post it here use code tags or it wont show up correctly, thx :)

Saaya,

Actually the clockspeed doesn´t affect cache latency at all.
Dothan @ 600MHz (6x100MHz): L2 Latency = 10, L1 Latency = 3
Dothan @ 2000MHz (20x1000MHz): L2 Latency = 10, L1 Latency = 3

[Carfax]

Saaya,

Actually the clockspeed doesn´t affect cache latency at all.
Dothan @ 600MHz (6x100MHz): L2 Latency = 10, L1 Latency = 3
Dothan @ 2000MHz (20x1000MHz): L2 Latency = 10, L1 Latency = 3

It does affect latency, but not the latency as per clock cycles, only in absolute time ie nanoseconds.