Need help, w/ building my computer.

This forum is for informing everyone else about special deals on stuff to buy like DVDs, cheap hard drives, blank CDs on sale, etc.

Postby Scintilla » Tue Apr 06, 2004 2:13 pm

dwchang wrote:More importantly, the Athlon 64's and Opterons "PWN" anything Intel has at the moment both in performance and price.

Okay, gotta ask: what's the difference between the Athlon 64 and the Opteron?

</newb-y question>
ImageImage
:pizza: :pizza: Image :pizza: :pizza:
User avatar
Scintilla
(for EXTREME)
 
Joined: 31 Mar 2003
Location: New Jersey
Status: Quo

Postby dwchang » Tue Apr 06, 2004 2:57 pm

Zarxrax wrote:
dwchang wrote:
Zarxrax wrote:AMDs suck compared to P4s. I dunno what benchmarks you are looking at, but on the one's I see all the time, the althlons loose bigtime, especially at video encoding and such.


You must be reading Tom's Hardware benchmarks. Everyone knows how "subjective" and "fair" they are :lol:.


What's wrong with tomshardware? I mean they aren't biased against AMD or anything... In fact its the site that convinced me to buy an AMD processor in my older system, 600mhz athlon. The tests they perform seem accurate. Whats the problem?


Are you being serious?

Within our industry everyone knows that Tom is a big Intel fanboy and to put it in "non-technical terms"...sucks their dicks.

I can point you to plenty of articles that show this. I remember when there was this huge Athlon 64 news article on nearly every hardware site (Anand, Ars Technica, theInquirer, Van's, etc.) and Tom refused to put it on since he thought it wasn't a big deal...he was wrong.

The biggest one was easily him defending Intel when they went to court against some consumers. Van's wrote a fairly in-depth article about it and Tom just resorted to name calling.

I could go into the nitty-gritty of the case, but it's rather boring (although very applicable given how you're bringing up benchmarks). Suffice to say, don't believe benchmarks. PCMark's address is internal within Intel and it's interesting to see the changes made to PCmark every year and the fact everything AMD does better taken out and everything Intel does inflated further.

For example, Athlons are better for sorting an excel doc with ~1000 entries (I belive this was the #), P4's are better when sorting 60000 entries. You tell me which one is more realistic. Also tell me why it's necessary to put the P4 favoring benchmark all over the place and the AMD one never....hmmm? I can keep going if you'd like, but I guess I already sorta said more than I said I would >_<

Either way, this benchmark thing is a big deal since even educated consumers can't make an...educated choice. Intel OWNS the benchmarks (no not PWNS or whatever..literally owns them) and thus they are very one-sided. It's sad that even benchmarks are filled with lies now.

And no, I'm not exaggerating this. This was the major turning point for me; when I stopped trusting Tom. It was hilarious that he didn't cover such a major case. Then again, everyone in the industry already knows this and knows Tom is an Intel fanboy.

By the same token, I think Anand is pro-AMD, so neither side is truly "subjective."

Scintilla wrote:
dwchang wrote:More importantly, the Athlon 64's and Opterons "PWN" anything Intel has at the moment both in performance and price.

Okay, gotta ask: what's the difference between the Athlon 64 and the Opteron?

</newb-y question>


An Athlon-64 is a consumer-level computer chip with 256 k - 1 M of L2 cache and 1 Hypertransport link.

Opterons are server-level chips with 1 M of L2 cache and 3 HT links.

There are obviously more difference, but for the most part, you just need to know one is the consumer chip and one is for powerful servers.
-Daniel
Newest Video: Through the Years and Far Away aka Sad Girl in Space
User avatar
dwchang
Sad Boy on Site
 
Joined: 04 Mar 2002
Location: Madison, WI

Postby trythil » Tue Apr 06, 2004 3:14 pm

dwchang wrote:For example, Athlons are better for sorting an excel doc with ~1000 entries (I belive this was the #), P4's are better when sorting 60000 entries. You tell me which one is more realistic.


Actually, from the perspective of worst-case analysis of algorithms, the sort test with 60,000 entries WOULD be better.
This is a block of text that can be added to posts you make. There is a 512 character limit.
trythil
is
 
Joined: 23 Jul 2002
Location: Llanfairpwllgwyngyllgogerychwyrndrobwllllantysiliogogogoch

Postby Kalium » Tue Apr 06, 2004 3:18 pm

trythil wrote:
dwchang wrote:For example, Athlons are better for sorting an excel doc with ~1000 entries (I belive this was the #), P4's are better when sorting 60000 entries. You tell me which one is more realistic.


Actually, from the perspective of worst-case analysis of algorithms, the sort test with 60,000 entries WOULD be better.

Yes, but precious few home computers will be doing that. Worst-case isn't always the most useful of tools.
User avatar
Kalium
Sir Bugsalot
 
Joined: 03 Oct 2003
Location: Plymouth, Michigan

Postby trythil » Tue Apr 06, 2004 3:33 pm

Kalium wrote:
trythil wrote:
dwchang wrote:For example, Athlons are better for sorting an excel doc with ~1000 entries (I belive this was the #), P4's are better when sorting 60000 entries. You tell me which one is more realistic.


Actually, from the perspective of worst-case analysis of algorithms, the sort test with 60,000 entries WOULD be better.

Yes, but precious few home computers will be doing that. Worst-case isn't always the most useful of tools.


Well, I'd like to correct one thing -- "worst-case" wasn't supposed to be in there. Input size is always "for some large N".

Also -- yeah sure, most home systems won't be sorting large data sets. However, I don't see how that makes the 60,000-item test any less or more valid than the 1,000-item test.

I mean, the benchmark is ludicrous. Nobody outside of Microsoft and some Shared Source (TM) partners know how Excel's sorting routines operate; that alone is enough to disqualify the benchmark, IMHO.

I was just pointing out that larger input sizes can sometimes give a better picture of an algorithm's performance.
This is a block of text that can be added to posts you make. There is a 512 character limit.
trythil
is
 
Joined: 23 Jul 2002
Location: Llanfairpwllgwyngyllgogerychwyrndrobwllllantysiliogogogoch

Postby dwchang » Tue Apr 06, 2004 5:19 pm

trythil wrote:I was just pointing out that larger input sizes can sometimes give a better picture of an algorithm's performance.


But as Kalium stated, benchmarks should benchmark realistic things that a consumer will actually do. Basing it on something unrealistic and saying that this product is the best FOR THE CONSUMER is outright lying. That was what the entire case was about and I believe they settled (i.e. Intel -> $$$ -> people). That seems to be the current trend for both Intel and Microsoft.

Perhaps I should go find that article. This one isn't even the most ludicrous. Then again, the most ludicrous thing is the fact the benchmark is done by Intel people. I believe we went to court over and now have ONE AMD person on the 20+ person board that decides these things.

BTW in case someone finds out I'm wrong, it *MAY* have been SysMark and not PCMark. Either way, my point is you shouldn't trust benchmarks. Sadly the next question is "what should I trust?" and sadly unless you work in the industry or have a degree in this, sadly...nothing. Crappy ain't it?
-Daniel
Newest Video: Through the Years and Far Away aka Sad Girl in Space
User avatar
dwchang
Sad Boy on Site
 
Joined: 04 Mar 2002
Location: Madison, WI

Postby Scintilla » Tue Apr 06, 2004 5:51 pm

dwchang wrote:Either way, my point is you shouldn't trust benchmarks. Sadly the next question is "what should I trust?" and sadly unless you work in the industry or have a degree in this, sadly... <b>nothing.</b>

What about <i>you?</i> Image
ImageImage
:pizza: :pizza: Image :pizza: :pizza:
User avatar
Scintilla
(for EXTREME)
 
Joined: 31 Mar 2003
Location: New Jersey
Status: Quo

Postby dwchang » Tue Apr 06, 2004 6:05 pm

Scintilla wrote:
dwchang wrote:Either way, my point is you shouldn't trust benchmarks. Sadly the next question is "what should I trust?" and sadly unless you work in the industry or have a degree in this, sadly... <b>nothing.</b>

What about <i>you?</i> Image


I thought we already went over this ;). I have a bachelor's degree in Computer Engineering and work for AMD as an engineer. Obviously biased, but at the same time, I'd like to think I work for the lesser or two evils :P.
-Daniel
Newest Video: Through the Years and Far Away aka Sad Girl in Space
User avatar
dwchang
Sad Boy on Site
 
Joined: 04 Mar 2002
Location: Madison, WI

Postby trythil » Tue Apr 06, 2004 7:18 pm

dwchang wrote:
Scintilla wrote:
dwchang wrote:Either way, my point is you shouldn't trust benchmarks. Sadly the next question is "what should I trust?" and sadly unless you work in the industry or have a degree in this, sadly... <b>nothing.</b>

What about <i>you?</i> Image


I thought we already went over this ;). I have a bachelor's degree in Computer Engineering and work for AMD as an engineer. Obviously biased, but at the same time, I'd like to think I work for the lesser or two evils :P.


So long as you touch x86, you are a bearer of evil.

:P
This is a block of text that can be added to posts you make. There is a 512 character limit.
trythil
is
 
Joined: 23 Jul 2002
Location: Llanfairpwllgwyngyllgogerychwyrndrobwllllantysiliogogogoch

Postby dwchang » Tue Apr 06, 2004 7:25 pm

trythil wrote:
dwchang wrote:
Scintilla wrote:
dwchang wrote:Either way, my point is you shouldn't trust benchmarks. Sadly the next question is "what should I trust?" and sadly unless you work in the industry or have a degree in this, sadly... <b>nothing.</b>

What about <i>you?</i> Image


I thought we already went over this ;). I have a bachelor's degree in Computer Engineering and work for AMD as an engineer. Obviously biased, but at the same time, I'd like to think I work for the lesser or two evils :P.


So long as you touch x86, you are a bearer of evil.

:P


This is coming from Mr. MIPS the obsolete?
-Daniel
Newest Video: Through the Years and Far Away aka Sad Girl in Space
User avatar
dwchang
Sad Boy on Site
 
Joined: 04 Mar 2002
Location: Madison, WI

Postby Mr Pilkington » Tue Apr 06, 2004 11:39 pm

dwchang wrote:....and work for AMD as an engineer....

And intel is the devil.


I say go with the VIA C3 :P
Image
<Orwell> WTF are they doing upstairs. Almost sounds like construction
<[Kristyrat]> Orwell, you see / <[Kristyrat]> when a man really likes a woman / <[Kristyrat]> or has a bottle of chloroform
<inthesto> He takes her stuff / <inthesto> And then poops in her vagina
User avatar
Mr Pilkington
 
Joined: 09 Apr 2002
Status: Stay outa my shed

Postby Zarxrax » Wed Apr 07, 2004 12:09 am

dwchang wrote:
Zarxrax wrote:What's wrong with tomshardware? I mean they aren't biased against AMD or anything... In fact its the site that convinced me to buy an AMD processor in my older system, 600mhz athlon. The tests they perform seem accurate. Whats the problem?


Are you being serious?

Within our industry everyone knows that Tom is a big Intel fanboy and to put it in "non-technical terms"...sucks their dicks.


Well I don't know about athlon 64 stuff, but I got proof here that back in the day, Tom's Hardware published data that said Athlons seriously kicked Intel's ass: http://www4.tomshardware.com/cpu/19990823/index.html

And as far as benchmarks go, i could care less about bullshit "benchmarking programs"... i don't even look at those scores. I likewise dont look at stuff like excel or autocad or whatever. I look at the stuff that pertains to me. Video encoding, audio encoding, 3d rendering, games (though I don't play games anymore I still like those benchmarks for some reason :p).
User avatar
Zarxrax
 
Joined: 01 Apr 2001
Location: Concord, NC

Postby dwchang » Wed Apr 07, 2004 12:29 am

Zarxrax wrote:
dwchang wrote:
Zarxrax wrote:What's wrong with tomshardware? I mean they aren't biased against AMD or anything... In fact its the site that convinced me to buy an AMD processor in my older system, 600mhz athlon. The tests they perform seem accurate. Whats the problem?


Are you being serious?

Within our industry everyone knows that Tom is a big Intel fanboy and to put it in "non-technical terms"...sucks their dicks.


Well I don't know about athlon 64 stuff, but I got proof here that back in the day, Tom's Hardware published data that said Athlons seriously kicked Intel's ass: http://www4.tomshardware.com/cpu/19990823/index.html

And as far as benchmarks go, i could care less about bullshit "benchmarking programs"... i don't even look at those scores. I likewise dont look at stuff like excel or autocad or whatever. I look at the stuff that pertains to me. Video encoding, audio encoding, 3d rendering, games (though I don't play games anymore I still like those benchmarks for some reason :p).


Dude this is durnig the PIII days ;). This is nearly 4 years old :P. Things are different now.

And I'm somewhat happy to see that you sorta judge things yourself. Most consumers *should* try and just see for themselves, but you must admit that most usually just read those benchmarks and dont' question.
-Daniel
Newest Video: Through the Years and Far Away aka Sad Girl in Space
User avatar
dwchang
Sad Boy on Site
 
Joined: 04 Mar 2002
Location: Madison, WI

Postby madmag9999 » Wed Apr 07, 2004 10:38 am

what is a program i could run to run benchmarks on my 2 comps. i have a p4 and a amd 2800+ and id like to see the differences. i went on what everyone was telling me that athlon is better and dicieded to get one for my new comp and it dose seem to be better then the p4 but id still like to run some benchmarks and see for myself
Moonslayer's Guide to a-m-v.org | AD & ErMaC's Guides to Audio & Video
"I'm sorry but i don't trust anything that bleeds for 5 days and doesn't die."
User avatar
madmag9999
 
Joined: 10 Aug 2003
Location: Pennsylvania
Status: Engaged

Postby dwchang » Wed Apr 07, 2004 11:10 am

madmag9999 wrote:what is a program i could run to run benchmarks on my 2 comps. i have a p4 and a amd 2800+ and id like to see the differences. i went on what everyone was telling me that athlon is better and dicieded to get one for my new comp and it dose seem to be better then the p4 but id still like to run some benchmarks and see for myself


You could download stuff like Sysmark, PCmark and SiSandra. Just google for them. However, again, I warn you that such things aren't an accurate indicator of performance as I already explained.
-Daniel
Newest Video: Through the Years and Far Away aka Sad Girl in Space
User avatar
dwchang
Sad Boy on Site
 
Joined: 04 Mar 2002
Location: Madison, WI

PreviousNext

Return to Heads Up!

Who is online

Users browsing this forum: No registered users and 1 guest