9500 Pro to 9800 Pro: yawn..

  • Thread starter Thread starter Wblane
  • Start date Start date
W

Wblane

Talk about an underwhelming increase in performance. I've got a Barton on an
NF7-S v2.0 running at 2300Mhz. 1 GB of DDR 2100 running at 150Mhz. Asynch CPU
FSB and memory (200/150). At least the bugs in the Catalyst 3.10 remained the
same.
-Bill (remove "botizer" to reply via email)
 
-
Wblane stood up at show-n-tell, in
(e-mail address removed), and said:
Talk about an underwhelming increase in performance. I've got a
Barton on an NF7-S v2.0 running at 2300Mhz. 1 GB of DDR 2100 running
at 150Mhz. Asynch CPU FSB and memory (200/150). At least the bugs in
the Catalyst 3.10 remained the same.

Let's see.... you are running subpar cpu and memory and are yawning,
because the new graphics card did not 'wow' you? Check your
bottlenecks/settings. I think you have something wrong.


WAIT. I just realized who you are....

TROLL.

*plop*
 
Wblane stood up at show-n-tell, in
(e-mail address removed), and said:


Let's see.... you are running subpar cpu and memory ...

I wouldn't call an OC'd XP3200+ subpar
 
Something must be wrong.
I went from a overclocked 9500 pro. (flashed Bios)
To a 9800. There is a substantial increase.
Now. I will add to that comment. My 9500pro when overclocked could handle
anything that I could throw at it. The payoff is with newer games like "Call
of Duty".
I also noticed the increase in demanding flight sims like "IL-2" etc.
I am running a xp3200 with 1g ram.
So be patient...
The 9500 pro is currently still a good card.
Dan
 
Barton on an NF7-S v2.0 running at 2300Mhz. 1 GB of DDR 2100 running
I wouldn't call an OC'd XP3200+ subpar

You are right that intel fans have trouble figuring out that his cpu in
isolation more than equals a 3.2ghz P4, but on the other hand Athlons are
severly memory bandwidth-limited to begin with, and the OP is exacerbating
this problem by running at a low memory busspeed.

He would see more improvement by investing in pc3500 and running at 1:1
ratio 220fsb or so.

rms
 
I was always curious about this. Since the Abit NF7s v2.0 has a 128-bit wide
memory bus in dual-channel mode (something Intel doesn't have does it?)
wouldn't it be more than capable of keeping up w/a 203Mhz FSB at 150Mhz? If in
dual-channel mode it xfers 128-bits per cycle it would be functionally running
at 300Mhz, even without considering DDR.
Or at least run the memory and cpu in sync to avoid the wait states.












Subject: Re: 9500 Pro to 9800 Pro: yawn..
Path:
lobby!ngtf-m01.news.aol.com!ngpeer.news.aol.com!feed2.newsreader.com!news reader.com!newshosting.com!news-xfer1.atl.newshosting.com!63.223.6.99.MISM
ATCH!c03.atl99!sjc70.webusenet.com!news.webusenet.com!nf3.bellglobal.com!u
rsa-nb00s0.nbnet.nb.ca!53ab
2750!not-for-mail
From: "Mike P" (e-mail address removed)
Newsgroups: alt.comp.periphs.videocards.ati
References: <[email protected]>
<[email protected]>
<[email protected]>
<[email protected]>
Lines: 25
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 6.00.2800.1158
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Message-ID: <[email protected]>
Date: Mon, 05 Jan 2004 02:54:37 GMT
NNTP-Posting-Host: 156.34.212.238
X-Complaints-To: (e-mail address removed)
X-Trace: ursa-nb00s0.nbnet.nb.ca 1073271277 156.34.212.238 (Sun, 04 Jan 2004
22:54:37 AST)
NNTP-Posting-Date: Sun, 04 Jan 2004 22:54:37 AST
Organization: Aliant Internet


-Bill (remove "botizer" to reply via email)
 
I wouldn't call an OC'd XP3200+ subpar

NO SHIT!

SUB PAR = Celeron P4 core... where 2.6Ghz = SLOWER than a 1.4Ghz AMD..
bottom end...

Only area of AMD's weakness is the VIDEO encoding (making) and 3D
rendering - because these are SINGLE tasking jobs DUE to the design of
the Intel P4 CPU... which was made FOR Mhz more than actual
Performance! DUH


Oh... never mind the AMD64... smokes the P4s easily... at 1200Mhz
Lower on the clock... and thats with 32bit Windows.
 
You are right that intel fans have trouble figuring out that his cpu in
isolation more than equals a 3.2ghz P4, but on the other hand Athlons are
severly memory bandwidth-limited to begin with, and the OP is exacerbating
this problem by running at a low memory busspeed.

AMD CPUs don't use the bandwidth of a P4, but a P4 is VERY memory
efficent with all the garbage IN<>OUT with its branch prediction....
so it takes a big performance HIT without the memory.

SO... with work such as Office and gaming, the AMD with its less
memory bandwidth is STILL FASTER...

But lets compare HyperTransport to Intel's memory bus for SMP,
HyperTransport Is state of the art allowing 2-way bandwidth for data.
 
Back
Top