Sign in with
Sign up | Sign in
Your question

Athlon64 kills P4 in Doom III

Tags:
Last response: in CPUs
Share
August 4, 2004 6:45:01 AM

<A HREF="http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=214..." target="_new">Click</A>


------------
<A HREF="http://geocities.com/spitfire_x86" target="_new">My Website</A>

<A HREF="http://geocities.com/spitfire_x86/myrig.html" target="_new">My Rig</A> & <A HREF="http://geocities.com/spitfire_x86/benchmark.html" target="_new">3DMark score</A>
August 4, 2004 8:56:41 AM

Thats very interesting, and being pro-AMD, its good, <b>but</b>, whats up with the test.

Wondering why the 2500+ XP chip beat the 2700+ and 2800+ chips. Also I expected the 2800+ XP chip to do better than the 2700+ mainly due to the 2800's increased L2 cache.

Anyways, what'd you expect? :smile:

My Desktop: <A HREF="http://Mr5oh.tripod.com/pc.html" target="_new">http://Mr5oh.tripod.com/pc.html&lt;/A>
Overclocking Results: <A HREF="http://Mr5oh.tripod.com/pc2.html" target="_new">http://Mr5oh.tripod.com/pc2.html&lt;/A>
Related resources
August 4, 2004 9:12:11 AM

one thing ill note that will make Kenavit happy is how suprisingly the Prescott is faster than the Northwood. but this is an exclusive thing to Doom3, just about every other situation/game is the opposite


VERY interesting link


man i cant wait until A64's scale up to 2.6ghz+. all those in that review were 2.2ghz. it wont be that long either :) 

-------
<A HREF="http://www.albinoblacksheep.com/flash/you.html" target="_new">please dont click here! </A>
<A HREF="http://www.subhi.com/keyboard.jpg" target="_new">This is you, interweb junky</A>
August 4, 2004 9:20:30 AM

Quote:
Wondering why the 2500+ XP chip beat the 2700+ and 2800+ chips. Also I expected the 2800+ XP chip to do better than the 2700+ mainly due to the 2800's increased L2 cache.


the 2500+ has 512kb cache, the 2700+ has 256kb. dont knwo about the 2800+ tho, i thought they were all 512kb L2, except some very early models



its also nice to know that my $140 Barton @ 2.5ghz will probably be on par or a tad faster than a 3.4ghz Northwood

-------
<A HREF="http://www.albinoblacksheep.com/flash/you.html" target="_new">please dont click here! </A>
<A HREF="http://www.subhi.com/keyboard.jpg" target="_new">This is you, interweb junky</A>
August 4, 2004 9:39:33 AM

I get more and more sure that my next rig will be an A64-based one...

What I find more interesting is the comparison of the 2 Semprons & the Celeron - the improvement from the on-die memory controller is incredible!

The performance of the ol' t-bred chips (like mine :frown: )is pretty lame though. Looks like I might have to overclock mine back up somewhat for DoomIII, even though I'll be primarily limited by my 9800Pro.....

I can hear Kanavit running here to gloat about his prescott actually being faster than a Northwood for <i>something</i> :lol: 

---
Epox 8RDA+ V1.1 w/ Custom NB HS
XP1700+ @200x10 (~2Ghz), 1.4 Vcore
2x256Mb Corsair PC3200LL/1x512Mb Corsair XMS PC4000 2.5-3-3-7
Sapphire 9800Pro @412/740
August 4, 2004 1:18:18 PM

And no one mantion it yet, but Quake 3 based games were faster with P4 based system.

So, this new id engine is no more favoring the Intel platform. This means that in the next years most (even all) games based on the Doom 3 engine will be faster on the AMD K8 Architecture. This confirm to me what will be my next CPU, an Athlon 64.

AMD really gave their CPU one of the best thing ever done in CPU architecture : ON-DIE MEMORY CONTROLLER! At first, this feature looked a bit weird, since the memory controller performance would be tied up to AMD cpu revision! But, this really pay off. Even the slowest K8 (Sempron 3100+) really shines, this look very promising and this leave Intel buyer with only one thing : HT. And this feature is only good in some situation...

It's very interesting to see that the K8 architecture is getting better with the time (like a good red wine).

<b>Intel, please wake up and fight!</b>

--
A7N8X / <font color=green><b>AMD Sempron 2800+</b></font color=green>
Kingston DDR333 2x256Megs
<font color=red>Built by ATI Radeon 8500 128Megs</font color=red> @ C:275/M:290
August 4, 2004 1:22:36 PM

>And no one mantion it yet, but Quake 3 based games were
>faster with P4 based system.

Q3 was, but for instance, RTCW which was based on the same engine, was considerably faster on Athlon than P4 for some reason.

= The views stated herein are my personal views, and not necessarily the views of my wife. =
August 4, 2004 2:46:43 PM

Quote:
And no one mantion it yet, but Quake 3 based games were faster with P4 based system


Q3 loves memory bandwidth, something that AMD chips lacked before .. not that they really needed them, it was just a quirk of the game engine


Doom3 also loves memory bandwidth, but A64's have tons of that now (and at a MUCH reduced latency as well)

-------
<A HREF="http://www.albinoblacksheep.com/flash/you.html" target="_new">please dont click here! </A>
<A HREF="http://www.subhi.com/keyboard.jpg" target="_new">This is you, interweb junky</A>
August 4, 2004 4:46:07 PM

LONG LIVE AMD!!!!
:lol: 
August 4, 2004 5:01:13 PM

Quote:
Wondering why the 2500+ XP chip beat the 2700+ and 2800+ chips. Also I expected the 2800+ XP chip to do better than the 2700+ mainly due to the 2800's increased L2 cache.

Could it be memory bandwidth?
I got a xp2700+, I've tried difference clock speed just by changing multiplier, but for some reason, at 166x11=1833MHz, it has <b>greatest</b> bandwidth efficiency <i>(tested using AIDA32, the memory read benchmark)</i>
and the order goes like this
93% efficience @ 1833MHz 166*11
90% efficience @ 2000MHz 166*12
87% efficience @ 1917MHz 166*11.5
89~90% effience@ 2166MHz 166*13
65% efficience @ 833MHz 166*5

all of them are based on 166FSB with memory@2.5/2/2/5, but why only at 1833MHz it has highest bandwidth? could it be a reason why 2500+ beat 2700+ and 2800+?

<P ID="edit"><FONT SIZE=-1><EM>Edited by Nights_L on 08/04/04 05:51 PM.</EM></FONT></P>
August 4, 2004 8:35:36 PM

Well this is definately one bandwidth hungry game. :o 

<sarcasm hat> <i>Big surprise there.</i> </sarcasm hat>

One thing that did brighten my day though is that my P4C 2.6GHz beats an AXP3000+. :)  Boy am I glad that I didn't go AMD when I bought my system a while back. :o  Though It <i>would</i> be nice to have an A64 <i>now</i>. :\ Oh well. Too many more important things to spend money on at the moment. Oh for the days of being single and having a full wallet...

<pre><b><font color=red>"Build a man a fire and he's warm for the rest of the evening.
Set a man on fire and he's warm for the rest of his life." - Steve Taylor</font color=red></b></pre><p>
August 4, 2004 9:53:51 PM

Only thing worse than a Nvidia vs. ATI slugfest is an AMD vs. Intel donnybrook. And I thought Ford vs. Chevy was bad. Well, I got my sasparilla and I'm steppin' out of the line of fire. By the way, I'm running an FX-53 right now and waiting on my X800XT PE from Excaliber-HIS.

Da Worfster

If you don't stand for something, you'll fall for anything.
August 4, 2004 10:30:40 PM

And this is a surprise to whom? :lol: 

Abit IS7 - 2.8C @ 3.4 - Mushkin PC4000 (2 X 512) - Sapphire 9800Pro - TT 420 watt Pure Power
Samsung 120gb ATA-100 - Maxtor 40gb ATA - 100
Sony DRU-510A - THAT'S MORE LIKE IT!
August 4, 2004 10:56:20 PM

The supprise is that an A64-3200+ would beat a P4EE 3.4. No, that's a major shock.
August 4, 2004 11:08:21 PM

We'll see shortly. I'm going head to head against a A64-3000 with my old OCZ memory in FarCry. The truth shall be known!

Abit IS7 - 2.8C @ 3.4 - Mushkin PC4000 (2 X 512) - Sapphire 9800Pro - TT 420 watt Pure Power
Samsung 120gb ATA-100 - Maxtor 40gb ATA - 100
Sony DRU-510A - THAT'S MORE LIKE IT!
August 5, 2004 3:28:51 AM

i don't think i'm going to notice 5fps between AMD and intel . LOL!!

but i did make the right choice, my prescott is faster in doom3 than the northwood. Maybe because D3 is new and uses SSE3.

guys remember, Intel isn't as memory dependant as A64. If i got an A64 3400+, i had to find good ram which would be very expensive. The P4 isn't as dependant on memory timings as AMD chips. So , getting a 3.2e prescott was actually more effective, than finding a new A64 mobo, A64 cpu, and good low timing ram.

------
Prescott 3.2E 1MB L2 HT
1GB PC 3200 Dual channel(PAT)
Asus P4P800 Bios 1016
PNY Geforce 6800 GT 256MB DDR3
60,823 Aquamarks <P ID="edit"><FONT SIZE=-1><EM>Edited by Kanavit on 08/04/04 11:30 PM.</EM></FONT></P>
August 5, 2004 4:24:22 AM

from what I know, A64 doesn't need good low timing ram, due to its ondie memory controller no?
if that's the case, then I ccould say P4 is dependant on memory bandwidth? TheRod just posted a link that shows A64 does not even need DDR400 to run, even with DDR333/266, it could run at least 90% as fast as with DDR400
but you can't do that with P4? if you put dual channel 266 with a 800fsb P4, it's gonna break the P4?

and Prescott is faster than Northwood in D3 due to its larger cache..if i remember correctly what I read :tongue:
August 5, 2004 6:39:27 AM

You really have no idea what in the f*ck you're talking about... I'm glad you showed up in this thread though, just to review your stupidy to everybody.

P4 doesn't need low timing rams? you have go to be f*cking kidding me, well this comment isn't as bad as HT has immediate impact on performance from one of your other posts, that was hilarious i couldn't stop laughing.

And do you even know what SSE3 is?

<A HREF="http://www.ebaumsworld.com/celebgay.html" target="_new">click here now!!</A>

<A HREF="http://www.ebaumsworld.com/epilepsyr.shtml" target="_new">DON'T CLICK HERE!!</A>
August 5, 2004 6:57:57 AM

> don't think i'm going to notice 5fps between AMD and intel
>. LOL!!

The difference is closer to 15 FPS. And if you don't notice that, maybe you'd notice $500 difference between two similary performing cpu's.

>but i did make the right choice, my prescott is faster in
>doom3 than the northwood.

LMAO, you werent going to notice 15 FPS, but you do care about 3.3 FPS ?

= The views stated herein are my personal views, and not necessarily the views of my wife. =
August 5, 2004 7:04:34 AM

Quote:
LMAO, you werent going to notice 15 FPS, but you do care about 3.3 FPS ?

Good one, but so far the thing that makes me love Kanavit
so much that i want to gouge out his eyeballs and skullf*ck him is this, don't remember exactly but somewhere around those lines, "Athlon64's useless since microsoft's never gonna release Winxp64, Hyperthreading on the other hand has immediate impact on performance."

I couldn't stop laughing after reading that, now another one to add on, AMD loves low latency while P4 doesn't.

Oh yeah another one, "Why pay for useless 64bit technology, a64 sucks, it's costing the same and it's 1ghz behind." Yes please explain to me then why Athlon64 is outperforming the P4 in games when it's 1Ghz behind and with half the cache.

<A HREF="http://www.ebaumsworld.com/celebgay.html" target="_new">click here now!!</A>

<A HREF="http://www.ebaumsworld.com/epilepsyr.shtml" target="_new">DON'T CLICK HERE!!</A>
August 5, 2004 7:46:10 AM

I love how you are trying to justify your purchase of an inferior CPU and generic RAM....

ROFL

</font color=red><i><font color=red>GOD</font color=red> <font color=blue>BLESS </font color=blue><font color=red>AMERICA
August 5, 2004 9:58:36 AM

<useless post>
Picky Picky. I use whatever works. 15fps faster when its past 35fps is unknoticeable. BLAH! We all know what a P4 is designed to do. We all know what an Athlon is designed to do. Whats the problem again? I forgot? Where is that processor that beats both in all kinds of aplications.
I know!! I'll design one... MO HA HA HA! Clobosessor! 1 MHZ 500 megs of Cache! 20Ghz FSB!!
</useless post>

"If youre paddling upstream in a canoe and a
wheel falls off, how many pancakes fit in a doghouse? None! Icecream doesn't have bones!!!"
Anonymous
a b à CPUs
August 5, 2004 2:42:10 PM

Quote:
So , getting a 3.2e prescott was actually more effective, than finding a new A64 mobo, A64 cpu, and good low timing ram.

Effective!!! what about getting a 3.0e to upgrade your 2.8b and then getting a 3.2e!!!now thats effective allright and you did have to find a mobo for your P4 and some new ram. Anyhow who cares your post always make my day =)

Out of curisosity, if you dont mind telling us how old are you and what are your incomes¿?

Asus P4P800DX, P4C 2.6ghz@3.25ghz, 2X512 OCZ PC4000 3-4-4-8, Leadtek FX5900 w/ FX5950U bios@500/1000, 2X30gig Raid0
August 5, 2004 3:01:35 PM

<b>OMFG</b>, i can't believe you guys didn't know that AMD are more sensitive to ram timings than Intel chips. If i bought A64 3400+, and used generic DDR400 3-4-4-8 timings it would be slower than my 3.2E prescott at same timings. Because it uses onboard memory controller , the A64 is even more memory dependant than a 3.2e chip. Honestly..

------
Prescott 3.2E 1MB L2 HT
1GB PC 3200 Dual channel(PAT)
Asus P4P800 Bios 1016
PNY Geforce 6800 GT 256MB DDR3
60,823 Aquamarks
Anonymous
a b à CPUs
August 5, 2004 3:16:45 PM

It is more timming sensitive and the P4 is more bandwith sensitive but honestly your P4 is still gonna end up getting raped by the A64 even if you cripple it!

lile read for you
<A HREF="http://www.tomshardware.com/motherboard/20040119/index....
" target="_new">http://www.tomshardware.com/motherboard/20040119/index....
</A>

And uhhh intersting quote

Quote:
Things start getting untidy when you combine compute-intensive tasks with large quantities of data such as file compression. In such categories, the memory timings make or break performance - the Pentium 4 processors either take the lead or bring up the rear, depending on whether the memory timings are fast or slow. We were duly impressed by the Athlon 64 FX-51's scores, which maintained its ranking no matter what kind of memory it was given. This steadfastness is largely due to the integrated memory controller.


Asus P4P800DX, P4C 2.6ghz@3.25ghz, 2X512 OCZ PC4000 3-4-4-8, Leadtek FX5900 w/ FX5950U bios@500/1000, 2X30gig Raid0
August 5, 2004 3:17:24 PM

If you are spending that much on a processor and trying to get the best out of it, why would you buy generic ram anyway? Its like getting a getting a Porshe but buying second hand tires...

Kanavit, why don't you just admit the the A64 beats the room warmer.
August 5, 2004 3:36:46 PM

Quote:
If you are spending that much on a processor and trying to get the best out of it, why would you buy generic ram anyway? Its like getting a getting a Porshe but buying second hand tires...

Kanavit, why don't you just admit the the A64 beats the room warmer.

i never complained about prescott heat. In fact, i may need it this winter. Heat is not a problem, its air cooling noise that i cannot bear. thats why i bought a silent Zalman cooler.
i hate spending money on RAM for some reason, like hard drives. Thats why i bought generic ram, and hooked it up with a Prescott. low cas Corsair XMS 2-2-2 high performance DDR400 ram costs like $300+ for 1gb Dual channel kit. As long as my system boots i'm happy. LOL

------
Prescott 3.2E 1MB L2 HT
1GB PC 3200 Dual channel(PAT)
Asus P4P800 Bios 1016
PNY Geforce 6800 GT 256MB DDR3
60,823 Aquamarks
August 5, 2004 3:37:13 PM

Quote:
i don't think i'm going to notice 5fps between AMD and intel . LOL!!

but i did make the right choice, my prescott is faster in doom3 than the northwood. Maybe because D3 is new and uses SSE3.

guys remember, Intel isn't as memory dependant as A64. If i got an A64 3400+, i had to find good ram which would be very expensive. The P4 isn't as dependant on memory timings as AMD chips. So , getting a 3.2e prescott was actually more effective, than finding a new A64 mobo, A64 cpu, and good low timing ram.

...........

Intel isnt as memory dependant as A64? ahahahahaha

SSE3? you keep mentioning this. those big numbers and catchy marketing phrases really get to you dont they! did you even read those reviews?

made the right choice, because your CPU comes out faster than the northwood in ONE game ? you built your system for one game?

dude and you wonder why we make fun of you. i used to try to restrain myself but you earned it

-------
<A HREF="http://www.albinoblacksheep.com/flash/you.html" target="_new">please dont click here! </A>
<A HREF="http://www.subhi.com/keyboard.jpg" target="_new">This is you, interweb junky</A>
August 5, 2004 3:42:38 PM

whate else could explain why Prescotts were faster than northwoods in Doom3 , when it was slower in every other game, 1mb cache maybe?? Doom3 could be optimized for SSE3 and advanced hyperthreading. Doom3 uses AI+ sound from CPU resources.

------
Prescott 3.2E 1MB L2 HT
1GB PC 3200 Dual channel(PAT)
Asus P4P800 Bios 1016
PNY Geforce 6800 GT 256MB DDR3
60,823 Aquamarks
August 5, 2004 3:45:52 PM

im sure ID would have let the reviewers know if it uses SSE3, its not like it should be kept secret. i dont know if it is or it isnt, but it was never mentioned once.

and doesnt the O/S have to support SSE3 before the applications can use it?

it was pretty much proven that the increase in cache is what made the difference

-------
<A HREF="http://www.albinoblacksheep.com/flash/you.html" target="_new">please dont click here! </A>
<A HREF="http://www.subhi.com/keyboard.jpg" target="_new">This is you, interweb junky</A>
August 5, 2004 3:49:56 PM

thats the thing, the prescott uses higher latency L2 cache, so the increase in size negates any performance gains because of the higher latency compared to half size northwood L2 cache. So that reason has been eliminated possibility why prescott is faster than northwood in D3.

------
Prescott 3.2E 1MB L2 HT
1GB PC 3200 Dual channel(PAT)
Asus P4P800 Bios 1016
PNY Geforce 6800 GT 256MB DDR3
60,823 Aquamarks
August 5, 2004 4:07:58 PM

Quote:
i never complained about prescott heat. In fact, i may need it this winter. Heat is not a problem, its air cooling noise that i cannot bear. thats why i bought a silent Zalman cooler.
i hate spending money on RAM for some reason, like hard drives. Thats why i bought generic ram, and hooked it up with a Prescott. low cas Corsair XMS 2-2-2 high performance DDR400 ram costs like $300+ for 1gb Dual channel kit. As long as my system boots i'm happy. LOL

I mentioned the prescott being hot in one word and you spend a paragraph telling me how YOU personally don't mind it.

You keep claiming that the Prescott is faster(marginally) than the northwood in Doom III, and that justifies you to deal with the heat and power issues that come along with it. Yet you like to cipple your system with generic ram. Doesn't make any sense.

Who said you MUST use Corsair cas 2-2-2 dual channel kit for an AMD. You are just trying to justify buying a lower valued heat warmer by inflating the price of the AMD.
August 5, 2004 4:58:38 PM

How about 40fps? personally i'm quite happy with 40fps+..
August 5, 2004 5:02:12 PM

Lads, at least let HIM believe what he says makes sense ;)  Cos i the end, you all know what the truth is, and I bet the number of pprescott owners (kana....) around here highlights that truth perfectly ;) 

XP2000, 512 ddr 2700ram, GF4 MX440, XP Pro
August 5, 2004 5:07:50 PM

Quote:
whate else could explain why Prescotts were faster than northwoods in Doom3 , when it was slower in every other game, 1mb cache maybe?? Doom3 could be optimized for SSE3 and advanced hyperthreading.

Obviously you didn't read the entire article..
Here is the reference for you to <A HREF="http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=214..." target="_new">click</A> if you are lazy to find it :lol: 
August 5, 2004 5:11:20 PM

Umm DUH thats a no-brainer

Mobile Barton 2500+ @ 2420mhz 11x220 1.7v
Asus A7N8X Dlx 440 FSB
1gb Geil GD pc3500 Dual Channel (2-3-3-6)
Segata 80gb SATA 8.5ms seek
ATI Radeon 9800 Pro to XT(463/795)
August 5, 2004 5:19:26 PM

$500.00 difference?
Who would pay $500.00 more for the "status" of having "Intel inside"?
I guess that makes my next upgrade a no-brainer.
Bob
August 5, 2004 6:16:17 PM

Quote:
thats the thing, the prescott uses higher latency L2 cache, so the increase in size negates any performance gains because of the higher latency compared to half size northwood L2 cache. So that reason has been eliminated possibility why prescott is faster than northwood in D3.


eliminated? its been spefically proven to be the exact reason why!

even tho the Prescotts L1/L2 cache is slower, its still much faster than having to pull that data from system ram


wheather a program benifits from it depends on how its coded. Carmack's engines always seemed to like fast ram/big caches, must be his style

-------
<A HREF="http://www.albinoblacksheep.com/flash/you.html" target="_new">please dont click here! </A>
<A HREF="http://www.subhi.com/keyboard.jpg" target="_new">This is you, interweb junky</A>
August 5, 2004 7:10:24 PM

Realy. I thaught I was being generous by saying 35. My peepers have never knoticed a difference once the image is repeated fast enought to be smooth.

"If youre paddling upstream in a canoe and a
wheel falls off, how many pancakes fit in a doghouse? None! Icecream doesn't have bones!!!"
August 5, 2004 7:10:49 PM

Quote:
"Kanavit seriously stop being such a Intel loyal limped dick pussy."



ROTFLMBBAO! (Rolling On The Floor Laughing My Big, Black, A** Off) You guys are killin' me!!!! Stop it. Play nice or I'll laugh so hard I'll git fired!!!

Da Worfster


If you don't stand for something, you'll fall for anything.
August 5, 2004 7:19:02 PM

not trying to brag or anything, but i can tell the difference up to about 85fps.. of course the closer it gets to this, the harder it is to tell.


but i can DEFINATLY tell the difference between 40fps and 60fps.

-------
<A HREF="http://www.albinoblacksheep.com/flash/you.html" target="_new">please dont click here! </A>
<A HREF="http://www.subhi.com/keyboard.jpg" target="_new">This is you, interweb junky</A>
!