Sign in with
Sign up | Sign in
Your question

G5 absolutely trampled by PCs

Tags:
Last response: in CPUs
Share
October 16, 2003 3:39:42 AM

<A HREF="http://www.pcworld.com/news/article/0,aid,112749,pg,1,0..." target="_new">http://www.pcworld.com/news/article/0,aid,112749,pg,1,0...;/A>
The first comprehensive and truly detailed test putting all the highest contenders against each other. Several FX-51 systems used as well to show any discrepancies in setup differences.
The last page has the G5 tests. Amazingly it simply loses except for a few Photoshop tests, but actually does lose in PS7 against the Dual 2GHZ Opteron (a fairer match anyways?). So now a 2GHZ Opteron Dual system which likely costs less than the G5 system can trample G5's best advantage, and yet it is already taking advantage of the 64-bit plug-in PS7 has. (I wonder if it even does anything, considering Apple's 64-bit does nada other than pure 64-bit extension, compared to AMD64's extra 8 registers!)
Does that spell little performance to be gained by G5s from the 64-bit support?

In anycase, the result is clear: PCs own, and in no way is that an overstatement. They cost thousands less, yes even the FX-51 (imagine if they put in an EE as well just to extend some of the leads in multimedia against G5!), and are clearly better suited for the future, especially for apps which will use AMD64's extra 8 register extension.

Deduct what you will, but I am rather convinced, this thorough test leaves no doubts about the fact G5 is stripped of IPC for higher clock, and is simply NOT a contender to modern PCs anymore. It performs well, but for its price and one-button mouse crazyness (I have to cope with that 3 out of 5 days), this is simply a zealot machine.

--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol:  <P ID="edit"><FONT SIZE=-1><EM>Edited by Eden on 10/15/03 11:40 PM.</EM></FONT></P>
October 16, 2003 8:27:19 AM

I just realised that people bought macs for <i>performance</i>. Thanks for enlightening me.

<font color=blue>
<i><font color=black>Faithless</font color=black></i> is he that says farewell when the road darkens.
J. R. R. Tolkien
</font color=blue>
Anonymous
a b à CPUs
October 16, 2003 10:19:15 AM

I disagree slightly. PC's have always enjoyed a better price/performance ratio than Mac's, that's not different now. However, I give some credit to the PPC970; few if any apps at the moment are recompiled to take advantage of its features. 64 bit support is not the issue (even OSX can hardly be called a 64 bit OS at this point), but the different architecture of the PPC970 is. I've seen a few benchmarks floating around the web where recompilation gained 30% to even 100% speedups. WHile I don't expect 30% to be the norm, I expect a healthy increase nevertheless.

Also, price performance is only one metric; sure an important one, but an existig MAC customer that knows and loves OS-X, has tons of MAC software, should take into account buying new x86 versions of all his apps. Not to mention OS-X is still a great OS .

Maybe those new macs arent the fastest workstations on the planet, but they perform pretty damn well nevertheless, and they make a lot more sense than the old G4's, especially to someone who has been using mac for years. You don't have to be a mac zealot to appreciate one.

As for the one button mouse; you can buy a 5 button USB mouse for a few dollar and hook it up to your mac if you like.

= The views stated herein are my personal views, and not necessarily the views of my wife. =
October 16, 2003 11:53:16 AM

Quote:
As for the one button mouse; you can buy a 5 button USB mouse for a few dollar and hook it up to your mac if you like.


Ah if only you could convince the Colleges and Institutions otherwise.

Quote:
I disagree slightly. PC's have always enjoyed a better price/performance ratio than Mac's, that's not different now. However, I give some credit to the PPC970; few if any apps at the moment are recompiled to take advantage of its features. 64 bit support is not the issue (even OSX can hardly be called a 64 bit OS at this point), but the different architecture of the PPC970 is. I've seen a few benchmarks floating around the web where recompilation gained 30% to even 100% speedups. WHile I don't expect 30% to be the norm, I expect a healthy increase nevertheless

I was maybe a bit ahead I guess. I thought the PowerPC 970 had been already optimized for.

Oh and about OSX, don't get me wrong now, I do like it. I've given my concensus for it in this forum and the POLLS one. I just don't appreciate how they offer these machines for such a price for such limited functions. Even OS X feels limited. I love using it in College work, but anything outside some Internet and image editing, and it gets very annoying. Ever tried word processing? It's hell.

One thing Windows should inherit at least, is the Cascade window mode. That's damn useful on the OS X. Everything scrolls from left to right in Window folders.

--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol: 
October 16, 2003 2:52:08 PM

I just don't see why Macintosh doesn't go to x86 where they can actually compete with Microsoft. To me, it seems Macs are bought for their software and looks, not their internals. Being that most of their internals are so similar to PCs nowadays, why not just go all the way?

If they have their OS operate on Intel or AMD systems, that would drastically reduce the cost of Mac systems, and drastically increase sales as a result. PCs really need competition in the OS part of the market, and with Apples know-how and user base there, I think they can really be in position to be that player.

-------------------------------------------
<font color=blue> "Trying is the first step towards failure." </font color=blue>
October 16, 2003 4:32:34 PM

It's funny because there was once a test involving an Athlon on OS X and apparently it ran some programs faster!
Imagine now the G5 runs Microsoft's OS and IT runs better than on PCs! :lol: 

The problem with MAC OS' in the x86 world in my eyes would be that, while it is excellent in visuals and ease of use, it also is very limited in options and flexibility as well as the ability to tweak it, and when it crashes, it's a true whore. I've seen some crashes on OS X, they take down everything or simply hang the comp. And that's from working in IE and Photoshop.
Granted it could be easy for beginners, but it just isn't very open so you can find problems and fix them.
I guess the one-button mouse design would cater to the newbies hehe. "There's no way you can get lost, it's just one button!"

--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol: 
October 17, 2003 4:32:34 AM

Eden man, i can see that the one button thing realy drives you nuts!!!

I am pretty sure that the OS will hotswop your mouse.. just bring it to the college and plug it in. at the end of class take it with you... i new a few poeple while i was in school that had their own.

yes it would be nice to have the college supply them for you...but they wont. so bring your own... it doesn't solve everyones problem.. but it deos solve yours.

heh... I remeber when my school got in all new iMac's with the new clear Mouse... EVERYONE loved them becasue they were 100 Times better then those godforsaken Pucks!!! oh GOD how i hated those....so people(disrespectedly) destroyed a hand full of all the pucks just so that the school would have to get new mice... hopeing to get the new ones... low and behold we did... mind you we had to wait a few weeks befre every one had mice... but we got what we wanted...(granted some of use wanted more then one button, but i guess we all figured that they HAD to ored the APPLE mice for some garantee purpose i suppose.

anyhow... ill stop now and take alook at that link.

ASUS P4S8X - P4 2.4B - 2 x 512M DDR333 - ATI 9500 Pro(Sapphire) - WD 80G HD (8M Buffer) - SAMSUNG SV0844D 8G HD - LG 16X DVD - Yamaha F1 CDRW - Iomega Zip 250 int.
October 17, 2003 4:56:10 AM

pretty good read.. skimmed alot of it mind you but its pretty much what i would expect.

though apple may not have the fastest... its nice to see that they are finaly contending... now they just need to eather start winning or pricing their machine accordingly...

I am a little anctious to see what panther has to offer...but i wouldn't expect it to make such a leep as to keep...er sorry, 'cause' apple to take 1st place in the 64 Bit world. but will just have to wait and see... its only going to be a week till its release...

so i guess we'll have to wait a month or more before we can find out how it stands up to compitition...and again the compition will have all its new stuff out by then to say that they are faster...

you know its like a game of Leap frog...but PC always seems to say were the bast louder.

ASUS P4S8X - P4 2.4B - 2 x 512M DDR333 - ATI 9500 Pro(Sapphire) - WD 80G HD (8M Buffer) - SAMSUNG SV0844D 8G HD - LG 16X DVD - Yamaha F1 CDRW - Iomega Zip 250 int.
October 17, 2003 10:19:50 AM

Guys Remember this. The Twits that read Mac Crap sites that beleave that there PCs are faster are reading One Website. People Who are reading One website are DUMB. It was funny when I read Mac sites and then started reading 20 Other sites. The more test done. I love it When Imac website was almost @ faster. yet when I seen other sites. that where doing test it was Like even. But facts where people in Apple company disable stuff on the Intel or Apple to make there cpu faster. Like HT. But when other sites did the same test with every thing running on intel and Amd they seem to run closer to apple.

But everyone Know this. Never go to manufacture for the test. That means Amd for amd test or Intel for intel test or Apple for apple test. If anything write down the high scores and check other sites and see if there right.

But for me I hit around 20 websites to try to make ture statment about a hardware. At least that gives me % of being right.

One thing I dont care if it amd or intel or apple. I m looking into a Intel laptop and looking into a New computer. The laptop mainly for games and Muisc when I m not home. and for a company from home. And Looking into a Athlon 64 for Games at home that deal with 64 bit but that will be a while. I will wait for the prices drop. And it will also help with my company. And I wait untel my Laptop payed off
October 18, 2003 4:55:05 AM

The G5 is currently infantile in its release, give the OS time to mature on its new CPU.

Panther is due to be released October 24th

"The 1.8-GHz single-chip G5 ($2999)" www.apple.com dual 2Ghz for $2999

Jaguar has no ability to handle 64 bit extensions in its current rev. Its comperable to running dual 2Ghz G4's on a 1Ghz bus. None of the current shipping G5's have a 64bit version of OS X with them (Unless you are a Apple developer or attended WWDC). It will be shipped later on the release date. All G5 buyers get Panther for free, and G4 owners can get panther for $100 and a 5 pack license for $199. Not a bad deal for OS licensing.

Photoshop CS will have the 64 bit OS X advancements not photoshop 7. goes back to the dual G4 2Ghz speed equivelance equation with Photoshop 7. Yes we are already running Photoshop 7 on the G5 dual 2Ghz

8 slots DC DDR 400 is a good start.

<A HREF="http://fugger.netfirms.com/g5/g51.jpg" target="_new">System config page</A> I installed 2x 512 PC2700 DDR sticks into the DDR 400 slots and had no problems so far.

<A HREF="http://fugger.netfirms.com/g5/g5cpuout.jpg" target="_new">I pulled out a CPU before I ever plugged it in, looking down on the CPU socket</A>

<A HREF="http://fugger.netfirms.com/g5/g52.jpg" target="_new">This is what came out</A>

<A HREF="http://fugger.netfirms.com/g5/g5b.jpg" target="_new">A closeup of the bottom of the CPU, showing pins and how fragile the socket is</A>

<A HREF="http://fugger.netfirms.com/g5/g5cpu.jpg" target="_new"> Another shot of the bottom, you can see heat pipe on left side</A>

<A HREF="http://fugger.netfirms.com/g5/g5cpu1.jpg" target="_new">Side view, heat sink installed</A>

<A HREF="http://fugger.netfirms.com/g5/g5sink.jpg" target="_new">The heatsink is huge w/heat pipes</A>

<A HREF="http://fugger.netfirms.com/g5/g5multis.jpg" target="_new">A close up of the multiplier settings on the bottom of the CPU</A>

<A HREF="http://www.adobe.com/products/photoshop/main.html" target="_new">http://www.adobe.com/products/photoshop/main.html&lt;/A>

I want to take one to 2.6Ghz and run some super pi

Shipped from Hong Kong...

Oh yeah, <A HREF="http://fugger.netfirms.com/4764.jpg" target="_new">I moved up to #1 in the world non-LN2/CO2</A>I am above all CO2, only 2 LN2's above me. R507 modded Mach1 cooling.

<b>"Granted I dont own a P4. But I read enough stuff and waste enough time on forums newsgroups IRC and computer news sites that I proberly know more then if I DID own a P4." -vk2amv</b>
October 18, 2003 5:46:53 AM

Quote:
Jaguar has no ability to handle 64 bit extensions in its current rev. Its comperable to running dual 2Ghz G4's on a 1Ghz bus. None of the current shipping G5's have a 64bit version of OS X with them (Unless you are a Apple developer or attended WWDC). It will be shipped later on the release date. All G5 buyers get Panther for free, and G4 owners can get panther for $100 and a 5 pack license for $199. Not a bad deal for OS licensing.


How are they able to ship G5 systems with 8GB of memory? And unlike x86-64, the 64-bit extension to the PPC ISA does not involve more registers or any other enhancements other than just being 64-bit. So unless you're talking about addressing more than 4GB of memory, there'd be no speed improvements.
Compilers may get better for the PPC 970 as IBM continues to improve upon them and in time, performance will, of course, grow, but it will have nothing to do with whether the OS is "64-bit" or not.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.
October 18, 2003 6:00:51 AM

Well said.

If there is anything it will do, is remove the OS limits on applications. But that means jack for performance.

I'd really like to see where the heck are Mac freaks coming up with "64-bit optimizations will help!"

--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol: 
October 18, 2003 8:13:59 AM

Other than the fact that Apple uses a different OS and infrastructure I don't see what's so advantageous in owning a MAC in the first place other than the feeling of exclusivity.

I'd rather put that money into either a dual XEON or OPTERON system. My opinion.

<A HREF="http://www.anandtech.com/mysystemrig.html?id=23810" target="_new"><font color=blue>My System </font color=blue></A>
<b><font color=blue>VAGABOND<font color=blue></b>

<b><font color=blue>veni,vidi, and ended up in THGC<font color=blue></b>
October 18, 2003 10:03:38 AM

I can imagine you completely overlooked who makes the G5 processor, the bus technology and altavec.

Let me answer mr I dont have one to know what I am talking about's question on installing 8gb ram into a G5 with jaguar and Ill get back to your misconceptions.

<b>"Granted I dont own a P4. But I read enough stuff and waste enough time on forums newsgroups IRC and computer news sites that I proberly know more then if I DID own a P4." -vk2amv</b>
October 18, 2003 10:21:19 AM

The G5 architecture can adress up to 8GB of memory but in reality with a 32 bit OS there is a problem. There is a bug or a barrier at 1.5GB ram to install applications.

Example: If you had 2 GB memory in your G5 loaded with Jaguar you could not install "FileMaker Pro" or "Corel Knockout" without taking out 512MB of memory or 1 GB to keep DC DDR. You can re install the memory after the software install and the application will work normally. There are a few other applications that are effected by this memory check.

Everything over 2GB is ignored until Panther it seems.

modified 10/14/03
http://docs.info.apple.com/article.html?artnum=107338

We are running with 2GB until Panther is released.

<b>"Granted I dont own a P4. But I read enough stuff and waste enough time on forums newsgroups IRC and computer news sites that I proberly know more then if I DID own a P4." -vk2amv</b>
October 18, 2003 8:24:15 PM

Erm I am well aware of the 2GB limit.

That however has no relation to "Better performance". What exactly do you fathom getting more in performance because you can run more RAM?

You haven't proven anything that once 64-bit support comes out, it will perform even better.

--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol: 
October 18, 2003 9:00:31 PM

"You haven't proven anything that once 64-bit support comes out, it will perform even better."

End of discussion for me, it appears you are trolling with the lamest of intentions. You might want to work on better angles if you plan of trolling with any sucess.

Thanks for the new signature.

<b>"Granted I dont own a P4. But I read enough stuff and waste enough time on forums newsgroups IRC and computer news sites that I proberly know more then if I DID own a P4." -vk2amv</b>
October 18, 2003 9:13:34 PM

It may be able to work with bigger files, but what does that have to do with performance itself?

Imgod2u even stated there shouldn't be any performance increases because the 64-bit technology in it isn't like AMD64 with the extra registers. So why am I the "troll" here?

Fugger your attemps at judging people are truly not putting you in the place to argue. If you had proof that there were optimizations from 64-bit and that it actually would increase performance even when the OS doesn't need to calculate 64-bit integers for the particular application, then there wouldn't be a discussion.

--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol:  <P ID="edit"><FONT SIZE=-1><EM>Edited by Eden on 10/18/03 05:16 PM.</EM></FONT></P>
October 19, 2003 5:29:37 AM

Nice look into that machine there thanks...:) 

as for that Stick of Ram... Strange that Apple would say that there parts are only but the best... but is that CL 3.0 RAM? isn't 2.0 or lower supposed to be considered better? Just wondering.

wonder if there would be anydiff if you put in better RAM?

ASUS P4S8X - P4 2.4B - 2 x 512M DDR333 - ATI 9500 Pro(Sapphire) - WD 80G HD (8M Buffer) - SAMSUNG SV0844D 8G HD - LG 16X DVD - Yamaha F1 CDRW - Iomega Zip 250 int.
October 19, 2003 5:36:55 AM

I thought that this might shed some light on to what kind of perfomance boosts one could find with upgradeing to Panther. ( note that G4's are not 64bit and they too gain boosts for this update)

Startup times greatly improved: Not only does the system itself start up faster (particularly on multiprocessor machines, compared to Jaguar), but logging in and out of user accounts is also nearly instantaneous on most systems. If one automatically logs in, for some systems boot times are cut by as much as half.

Application launch times: These have also vastly improved over Jaguar. Even under moderate to heavy loads, a dual 867MHz PowerMac G4 (MDD) will launch most applications in less than one full "bounce" of their Dock icons -- almost instantly -- and even on older systems the difference is like night and day. Panther just plain scales performance by an order of magnitude compared to Panther across the board.

Mail: The new mail filters are much more effective, even "untrained," compared to those in Jaguar's version of Mail.app. Performance, particularly when viewing HTML messages, is greatly improved.


ASUS P4S8X - P4 2.4B - 2 x 512M DDR333 - ATI 9500 Pro(Sapphire) - WD 80G HD (8M Buffer) - SAMSUNG SV0844D 8G HD - LG 16X DVD - Yamaha F1 CDRW - Iomega Zip 250 int.
October 19, 2003 6:44:45 AM

Quote:
End of discussion for me, it appears you are trolling with the lamest of intentions. You might want to work on better angles if you plan of trolling with any sucess.

That's just about borderline fanboy'ism as it comes. He's not bashing or putting down macs at all.

Quote:
You haven't proven anything that once 64-bit support comes out, it will perform even better.

How is that trolling at all? It's just a statement asking whether or not you can prove that 64 bit extentions will give the G5 a boost or not. Look at the Hammer, even just judging by it's poor performance using it's SSE2 extensions, one could say that it still has some driver development to go before it really gets a boost. And even under betas of 64 bit windows, all the code needs to be recompiled, drivers and such, and even with that, it performs worse in some cases. Read up on tech documents and sites like Anandtech that have already done testing with the betas.

:cool: I run my AthlonXfx at 7.65 Exahertz :cool:
October 19, 2003 1:17:46 PM

Didn't you use to be an intellite?

<b><font color=red>They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.</font color=red></b>
October 19, 2003 1:23:48 PM

G5s are good, but not as good as apple make them out to be.

One button mice are bad, but I've been hearing rumours that apple are developing a really good mouse with Ipod style scrolling. But the problem is I've been hearing that for a while.

I like OSX as it is good to the user. But it does have a [-peep-] load of compatiblity problems, for hardware AND software. And someone here mentioned people with OSX having lots of software. What about people with OS9 and LOTS of software. No, classic mode doesn't work. Well hardly anyway, and when it does, its too slow. You can't boot into OS9 with these G5s so they're screwed.

All in all, apple ain't as bad as they used to be.

<b><font color=red>They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.</font color=red></b>
October 19, 2003 6:21:52 PM

Quote:
Startup times greatly improved: Not only does the system itself start up faster (particularly on multiprocessor machines, compared to Jaguar), but logging in and out of user accounts is also nearly instantaneous on most systems. If one automatically logs in, for some systems boot times are cut by as much as half.

Application launch times: These have also vastly improved over Jaguar. Even under moderate to heavy loads, a dual 867MHz PowerMac G4 (MDD) will launch most applications in less than one full "bounce" of their Dock icons -- almost instantly -- and even on older systems the difference is like night and day. Panther just plain scales performance by an order of magnitude compared to Panther across the board.

Mail: The new mail filters are much more effective, even "untrained," compared to those in Jaguar's version of Mail.app. Performance, particularly when viewing HTML messages, is greatly improved.


CD, although I'd ask for proof here, I can already tell you some of those claims are absurd if they're "generalized", especially for the G4 ones. It's nice to see improved startup times, but anyone who even complains OSX takes time on G4s is nuts. We work with simply uni-867MHZ G4s and it's never been a problem. But where it gets absurd is the program startup times. I can understand if IE or Safari launches fast, but Adobe Photoshop?
Yeah right!

Quote:
( note that G4's are not 64bit and they too gain boosts for this update)

I think this is what gives it away, in that (no offense of course) this has nothing to do with proving G5s with 64-bit will do better.
It's great to hear though. I hope the college buys the licenses for upgrades. Productivity helps. I guess if Duals get better, it's because Apple has figured out how to optimize for that. I hope it works for uni-processor systems too.

But still, what is interesting is Fugger's blatant attack where he doesn't prove one bit that G5s get optimized for 64-bit. It's a given they will run more RAM and bigger files with 64-bit, where the performance comes is beyond me. And then he comes and calls the skeptics like me, trolls.

--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol: 
October 19, 2003 6:23:33 PM

Oh I know what you mean. Although until we see a performance ratio that actually conforms to its price premium, I don't think it is worth it. Future Dual Opterons at 2.2GHZ and above will continue to up the performance delta and still cost less. So one is left wondering just where will the performance come from other than compiler and OS optimizations.

--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol: 
October 19, 2003 6:34:07 PM

Out of curiosity though, does the OS X have a quick self-config of hotswappable hardware like USB mice?
If it allows me to put mine in for example, and just work right off with even right-clicking, that'd rock.

--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol: 
October 19, 2003 9:07:15 PM

I don't realy sure, but my fried used to use his at the school, but that was on OS 9.x, Another friend of mine uses a 4 button mouse and say he used all 4 buttons in PS and Illustrator. He is running OS X right now but he might have been doing that under 9 at the time he told me...

I take a look into that for you though...

ASUS P4S8X - P4 2.4B - 2 x 512M DDR333 - ATI 9500 Pro(Sapphire) - WD 80G HD (8M Buffer) - SAMSUNG SV0844D 8G HD - LG 16X DVD - Yamaha F1 CDRW - Iomega Zip 250 int.
October 24, 2003 3:48:59 AM

yes Eden it does, I asked asked around, could get a hold of my friend, but knew a guy that has a Mac and has a Forum (UgL) so i posted the Question there for you.

<A HREF="http://u-g-l.board.dk3.com/2/viewtopic.php?t=124" target="_new">As i susptected</A>




ASUS P4S8X - P4 2.4B - 2 x 512M DDR333 - ATI 9500 Pro(Sapphire) - WD 80G HD (8M Buffer) - SAMSUNG SV0844D 8G HD - LG 16X DVD - Yamaha F1 CDRW - Iomega Zip 250 int.
<P ID="edit"><FONT SIZE=-1><EM>Edited by cdpage on 10/23/03 11:50 PM.</EM></FONT></P>
October 25, 2003 1:14:38 AM

Thanks!

I also have a similar mouse as his, an optical plus scroll Logitech 2-button.

--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol: 
Anonymous
a b à CPUs
October 25, 2003 11:06:57 AM

>Imgod2u even stated there shouldn't be any performance
>increases because the 64-bit technology in it isn't like
>AMD64 with the extra registers. So why am I the "troll"
>here?

Maybe not a troll, but clueless as usual. Its funny how everyone has to prove their claims to you, while you never provide half a shred of evidence of your points.

>Imgod2u even stated there shouldn't be any performance
>increases because the 64-bit technology in it isn't like
>AMD64 with the extra registers.

The extra registers in AMD64 have in se nothing to do with 64 bits. In theory, I think you should be able to write a 16 bit app under DOS that makes use of them, much like old DOS extenders (emm386 anyone ?) used 386 protected mode instructions on a 16 bit OS (Dos, windows 3.x) by switching back and forth between CPU modes. You could hardly call such a piece of software "64 bit", just like you couldnt call win 3.11 a 32 bit OS because it supported a "386 enhanced" VM feature on 386 class cpu's.

Now, the G5 is quite a different beast from the G4. Optimizing for the G5 *will* increase performance by better exploiting its enhanced altivec units, cache structure, better branchprediction and its revised µops dispatching, scheduling and issueing. You might not call these optimizations "64 bit" but who cares ? Its not like you'd call any K8 optimized libraries 64 bit either.

Finally, integer math code will simply speed up (up to 4x) by using 64 bit registers on any 64 bit cpu. So yes, you would increase performance by recompiling your apps for 64 bit operation, even though this will not make a huge difference in most cases (except things like encryption algorithms).

= The views stated herein are my personal views, and not necessarily the views of my wife. =
October 25, 2003 5:28:49 PM

Wow BB, you are growing more and more with the desire to create scenarios to be able to accuse me of being clueless. When I question 64-bit performance strictly, I question THAT and THAT only. Where did you see me question the entire CPU optimization?
In fact, since when did I ever express that CPU-specific optimizations don't help?
I questionned Fugger like imgod2u, but then I get the usual reply grasping for attention: "Maybe not a troll, but clueless as usual. Its funny how everyone has to prove their claims to you, while you never provide half a shred of evidence of your points.
"
At least prove me that I did say something wrong.

Quote:
Finally, integer math code will simply speed up (up to 4x) by using 64 bit registers on any 64 bit cpu. So yes, you would increase performance by recompiling your apps for 64 bit operation, even though this will not make a huge difference in most cases (except things like encryption algorithms).

I think I'll leave that to imgod2u to see how true that is.

--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol: 
Anonymous
a b à CPUs
October 25, 2003 5:56:14 PM

>At least prove me that I did say something wrong.

Oh dear.. havent I proved that often enough by now ? I mean, really, you *still* think NUMA speeds up a single CPU machine ? You still believe a process' memory footprint can grow up to 4 GB under windows ?

>I think I'll leave that to imgod2u to see how true that is.

It doesnt require a whole lot of knowledge to imagine 64 bit integers are faster to work with when you don't have to split them up. Not that 64 bit or even longer integers are common in anything but encryption, or things like folding, but still. here is a nice illustration:
<A HREF="http://www.intel.com/design/security/rsa2000/itanium.pd..." target="_new">http://www.intel.com/design/security/rsa2000/itanium.pd...;/A>

Quote:
Approximate number of CPU
instructions per decrypt: instructions per decrypt:
– 32-bit architecture 32-bit architecture
1.5 million multiplies 1.5 million multiplies
1.5 million adds 1.5 million adds
– IA-64 architecture IA-64 architecture
375K multiplies 375K multiplies
375K adds


4x less instructions, like I said, in theory up to 4x faster performance. BTW, this part has nothing to do with IA-64, it holds true for any 64 bit cpu. It was about the only type of bench intel could show during the Merced days :) 

= The views stated herein are my personal views, and not necessarily the views of my wife. =
October 25, 2003 6:39:14 PM

Quote:
It doesnt require a whole lot of knowledge to imagine 64 bit integers are faster to work with when you don't have to split them up. Not that 64 bit or even longer integers are common in anything but encryption, or things like folding, but still. here is a nice illustration:
http://www.intel.com/design/security/rsa2000/itanium.pd...


In reply to:
--------------------------------------------------------------------------------

Approximate number of CPU
instructions per decrypt: instructions per decrypt:
– 32-bit architecture 32-bit architecture
1.5 million multiplies 1.5 million multiplies
1.5 million adds 1.5 million adds
– IA-64 architecture IA-64 architecture
375K multiplies 375K multiplies
375K adds



--------------------------------------------------------------------------------


4x less instructions, like I said, in theory up to 4x faster performance. BTW, this part has nothing to do with IA-64, it holds true for any 64 bit cpu. It was about the only type of bench intel could show during the Merced days :) 

I find it funny you used THAT to prove a point. I know and wouldn't deny the facts that when you DO need the extra calculations, it would help, but if the apps don't need over 32-bit integers, where do you actually see a boost coming?
I don't deny whatsoever applications that utilize 64-bit will benefit, but to just claim "optimizing for 64-bit will make G5 go faster", when you don't prove us most apps on Macs need over 32-bit integers or consume too much memory, is what I want to see proof on. Again how is that cluelessness?

Quote:
I mean, really, you *still* think NUMA speeds up a single CPU machine ?

Granted I made a small error on that, but if I recall correctly, to my defense, I did mention Opteron. I should have stated multiple CPU processors but I didn't, my bad. However I do know how Opteron does well in CPU scaling. That's not the subject though.
Quote:
You still believe a process' memory footprint can grow up to 4 GB under windows ?

Again with the creating scenarios?

Look, I'm starting to think you have seen tons of things I said which were foolish. I can point my initial 64-bit claim, and this recent NUMA thing which was a mental crapout error, so now, where else did I screw up?
You make it seem as if every thing I say is absolute BS, and now, I calmly ask you to point it out.
In all of my time on THG, whenever I said something wrong, I was corrected with no insults, mainly because I can do errors and usually I try not to say foolish stuff, but you on the other hand have no respect whatsoever, are cynical and will go lengths on only ME above all, like an obsession, to make me sound like I am the most wrongful person here.
Christ drop it already, you're seriously getting annoying with your insults. You want to correct me, tell me I made an error and point it out.


--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol: 
October 26, 2003 12:15:37 AM

Yup there a bunch of big scrap around the boards with good flame.I guess time for little explaintion again on data precison and 64 bitness stuff.

Again for IA-64,PPC970,K8 ,USIII.

64 bit precision or 64 data size are use if there a number that exceed 4 billion or 2 power 32 -1 as 0 count.If exceed 2 power 64 a even greater bit size will be use.Multiple instruction/multiple data can give you larger data precision if need.
In a case of opteron and PPC970 if encounter a 80 bit precision need from software RAX and RBX will be use for calculation and storage will need RAX and RBX.In 32 bit mode 3 register will be use.?4 bit is useful only when dealing with massive number as it reduce the number of reg use and instruction use as if you multiply 8*124 *6*15 will be split in many instruction if you split them by crunch of 64 bit thing will go faster that split them in 32 bit chunk.This case will happen rarely.Pointer now are allwayse 64 bit so they are bigger.In computer hardware bigger mean slower not faster a 16 bit cpu will been much faster for a ot of stuff, P4 ALU rapid excution engine work soly on 16 precision as they are smaller.Others concept 64 bit path vs 1 bit this exsemple is extreme but make thing clear.On a mothers trace the little ligne that cant be see bettween componant are never straight as clock delay have to be very close to each others for a syncronisation bettween ligne if not the clock will go wild for the rest.To get to the point a 1 ligne will allwayse straight having a smaller delay or faster as the data are smaller while the 64 bit will go slower.Inside a cpu is the same to prevent degradation of 32/16 bit performance mode are set as any 64 data will take more time to excute.The word ""mode"" should not as been use but i dont have find a better way to explain it.

64 bit give you a edge for large number FPU or ALU.Again there disaventage flag and pointer are moving to 64 that put extra pressure on systemes and resource also File allocation table must move to 64 bit also call FAT for windows 9X ntsc for NT kernel.SO memory adresse range will move to 64 bit and space will move to 64.That lead to larger code as there will be lose space many EXE have gain 2X or more
^^(do not take this as rule also read well EXE)^^
by the way some IA-64 have gain more that 4 time the size.


In short most software will lose from moving to 64 bit exception to sientific stuff that most of the time use number on float like 10/3=3.33333333333333333333333333 but precision is need.Or to save in number of instruction.


Approximate number of CPU
instructions per decrypt: instructions per decrypt:
– 32-bit architecture 32-bit architecture
1.5 million multiplies 1.5 million multiplies
1.5 million adds 1.5 million adds
– IA-64 architecture IA-64 architecture
375K multiplies 375K multiplies
375K adds


That marketing stuff comparing a IA-64 mpu to IA-32 mpu is no sense in any way.


when you don't prove us most apps on Macs need over 32-bit integers or consume too much memory, is what I want to see proof on. Again how is that clueless

They will increase from the lose space in the new FAT and bigger flag,pointer.

I dont like french test<P ID="edit"><FONT SIZE=-1><EM>Edited by juin on 10/25/03 11:37 PM.</EM></FONT></P>
October 26, 2003 2:33:40 AM

Interesting read, but again, I ask you: Do you feel 64-bit optimization is possible for even regular applications which don't call for 32+ bit integers or over 32-bit addressing, for the G5? Or in general, where do you see 64-bit support on G5s standing?

Quote:
They will increase from the lose space in the new FAT and bigger flag,pointer

What will increase? And it's supporting who's stance on the G5 64-bit technology?

--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol: 
Anonymous
a b à CPUs
October 26, 2003 11:20:07 AM

>it would help, but if the apps don't need over 32-bit
>integers, where do you actually see a boost coming?

If the app doesnt use large INTs, and doesnt use 64 bit memory addressing, how exactly could it be considered a 64 bit port ??

>I don't deny whatsoever applications that utilize 64-bit
>will benefit, but to just claim "optimizing for 64-bit will
>make G5 go faster",

So, its semantics again hu ? You agree pretty much *any* app can benefit from optimisations for the PPC970 renewed architecture, AND you agree that any app using large INTs or large memory addresses will benefit (substantially) from being ported to 64 bit. Yet your point is that "optimizing for 64-bit will make G5 go faster" is wrong. Good point Einstein.

>Granted I made a small error on that

Small ?? Well, maybe small given your track record, but its still a major blunder. What's worse though, is that I *really* can not understand why you would throw in the NUMA post in a discussion on the FX' clock scaling when you obviously don't (didn’t ?) know what NUMA is ! If you knew anything at all about NUMA, you'd know it can not possibly affect performance of a single CPU FX, and can never be the reason. So why post that then ? Why ?? Trying to show off or what ?

> but you on the other hand have no respect whatsoever, are
>cynical and will go lengths on only ME above all, like an
>obsession, to make me sound like I am the most wrongful
>person here.

I can not stand the "I don't know, but I'll just pretend" attitude. I can not stand people making opinions and posting them on just about any subject while mostly none of these are founded on even a superficial understanding of the matter at hand. And what makes it all worse, is your inability, or rather unwillingness to absorb knowledge handed to you by others (not just me) and change your POV. Nah, your typical response would be something like "its my POV and I’ll stick to it" or "I don't know a thing about this, but I won't take anyone's word for it without undeniable proof". And even then, if you don't understand the proof given, you'll just dismiss it. Its that attitude that will make me pull your pants down every chance I get. If nothing else, maybe it will make you think twice before posting about something you don't have a clue about

>where else did I screw up?
>You make it seem as if every thing I say is absolute BS,
>and now, I calmly ask you to point it out.

You really want this ? you sure ? Okay.. I just wasted some time going over 2 or 3 threads, and came up with the following quotes that show just how much you know about the topics you so much like to discuss. You probably won't realise the irony or silliness of all of them though:

Quote:
<font color=red>Prove with a real link that the memory controller isn't the very same. (athlon FX and A64)

</font color=red>

64 bit versus 128 and unregistered versus registered memory isn’t a difference apparently ? When someone else points this out you reply:

Quote:
<font color=red> That's true, I knew about the differences, but essentially they are both the same on-die memory controllers with DDR400, in that they run very much at the same speeds. If it were DDR333 I'd give him credit

</font color=red>

LOL. So it’s the speed that defines whether the controller is different then? Guess what, any K8 controller works at core frequency (anywhere from 1.4 to >2.2 GHz). So following your argument, the opteron 144 has a different MC as the opteron 146 ! and what you claims, the DDRxxx speed is just a divisor that only affects the modules, not the controller itself !!

Quote:
<font color=red> Can you prove the Xeon's core is the same as the P4's, as in the server component circuits are intact and that the bridges on the chip are the same?

</font color=red>

“Intact server components circuits” ?? bridges (that don’t even exist on P4’s) as a proof the core is the same ?

Quote:
<font color=red> If 64-bit solves some address virtual memory limit, then that is far from because it's 64-bit integers, that's almost an indirect fix

</font color=red>

An instant classic !

Quote:
<font color=red> I simply don't grasp why a 32-bit register, and addressing, which entitles you to 4GB of addressing and 4 billions of integers (or was it milions?) to work with, cannot get around some virtual memory issue, and that a processor with extended register size suddenly can.

</font color=red>

Let me tell you why: cause you don’t understand a thing about registers, addressing or virtual memory., hell you can’t even do binary math!

Quote:
<font color=red> why hasn't Microsoft so far simply just made their OS to 3GB physical and 1GB virtual in the first place?

</font color=red>

Yeah that would solve it LOL

Quote:
<font color=red> Can't they properly manage the file sizes, optimize, compress and make sure the video RAM buffers are entirely used before RAM Aperture is used?

</font color=red>

Mail Carmack ! You have some revolutionary ideas young Einstein !

Quote:
<font color=red> I was indicating if texture loads and graphical loads were directly loaded from HDD into memory buffer and not on the RAM.

</font color=red>

Mail nVidia and ATI ! Ignoring small details such as the fact that “buffer” is also RAM, and usually amounts to a enormous 64 or 128 Mb, surely making 64 bit addressing useless, its yet again a terrific and realistic idea ! All ATI and nVidia would have to do is add a SCSI, ATA and SATA controller interface with support for all known RAID setups and controllers on their gfx cards, add support for FAT, NTFS, EXT2, SFX,.. file systems, somehow bypass the driver, locate the files on disk and load them straight into VRAM ! Oh, don’t forget to put intel in CC:, so they can modify AGP and PCI express to allow this.

Quote:
<font color=red>
They must have changed something on the packaging die [of the Athlon MP] because otherwise it would always work on normal Athlon XP mainboards

</font color=red>

Every Athlon MP works on any Athlon XP board… and a “packaging die” ?

Quote:
<font color=red> As a dumb Joe sixpack back then, whenever I'd read MMX I'd tout it!
I was always reporting my P2 350 as 350MMX rather than 350MHZ MMX. It just sounded cooler.
<b>Now I'm "just a tad bit" smarter.</b>

</font color=red>

No you’re not :p 



= The views stated herein are my personal views, and not necessarily the views of my wife. =
Anonymous
a b à CPUs
October 26, 2003 11:40:49 AM

I have a hard time understanding most of what you are saying; however I think you are confusing floating point precission with the 32/64 bit discussion at hand. I think I already pointed this out to you.. floating point precission has nothing to do with 32 or 64 bit architectures. Even a 486DX had 80 bit VPR's. Neither Athlon 64 not PPC970 change anything relevant AFAIk to the FP units over their 32 bit predecedors.

>That marketing stuff comparing a IA-64 mpu to IA-32 mpu is
>no sense in any way.

Its comparing integer math on a 64 bit platform versus a 32 bit platform. It applies *exactly* as quoted on a G4 versus G5 running DES encryption software.

>In short most software will lose from moving to 64 bit
>exception to sientific stuff that most of the time use
>number on float like 10/3=3.33333333333333333333333333 but
>precision is need

This on the other hand has nothing to do with the discussion at hand. An opteron doesnt handle 10/3 any different than an Athlon, neither does a G5 versus a G4.

>File allocation table must move to 64 bit also call FAT for
>windows 9X ntsc for NT kernel.SO memory adresse range will
>move to 64 bit and space will move to 64

What are you trying to say here ? That a 64 bit cpu will gain susbstantially because it would be faster when using 64 file systems ? If so, you're out of your mind. Not even Apple is crazy enough to claim this.

= The views stated herein are my personal views, and not necessarily the views of my wife. =
October 26, 2003 12:45:03 PM

On PPC 970 my guess is no there wont be gain neither for opteron or K8

I dont like french test
October 26, 2003 12:51:11 PM

What will increase? And it's supporting who's stance on the G5 64-bit technology?

Cache ligne or FAT use space adresse in 32 bit mode you use 32 fat on 64 bit 64 and so on.The same happen in cache ligne but cache must be close full or not.Someone might have a better explaination.

As pointer are the size of the fat they are now bigger and also they is totaly no need for larger pointer advantage 64 adressing 4+GB of ram.

I dont like french test
October 26, 2003 1:01:47 PM

If the app doesnt use large INTs, and doesnt use 64 bit memory addressing, how exactly could it be considered a 64 bit port ??

You just figure it out they wont be recompile.

So, its semantics again hu ? You agree pretty much *any* app can benefit from optimisations for the PPC970 renewed architecture, AND you agree that any app using large INTs or large memory addresses will benefit (substantially) from being ported to 64 bit. Yet your point is that "optimizing for 64-bit will make G5 go faster" is wrong. Good point Einstein.

Architecture yes 64 bitness no does there app that need more that 5GB of ram on MAC.If they need 4.4gb they will shrink the code not making it move to 6 GB with 64 bit mode.

Core from P4 XEON gallatin are the same.

MCH from all K8 are the same or might there are small change between opteron/athlonfx vs amd64 but when i see DDR 400 support on athlon FX it make me think they are the same



I dont like french test
October 26, 2003 1:18:45 PM

have a hard time understanding most of what you are saying; however I think you are confusing floating point precission with the 32/64 bit discussion at hand. I think I already pointed this out to you.. floating point precission has nothing to do with 32 or 64 bit architectures. Even a 486DX had 80 bit VPR's. Neither Athlon 64 not PPC970 change anything relevant AFAIk to the FP units over their 32 bit predecedors

What the differece between Int number with float number nothing.You mix SSE/2 reg have ability to run 64 bit precision in 1 pass.

Its comparing integer math on a 64 bit platform versus a 32 bit platform. It applies *exactly* as quoted on a G4 versus G5 running DES encryption software.

Comparing absolute performance of a Vliw big cache systemes low clock speed vs a P4 CISC low cache high clock speed is not significant at all.

This on the other hand has nothing to do with the discussion at hand. An opteron doesnt handle 10/3 any different than an Athlon, neither does a G5 versus a G4.

Handling nothing as that allwayse X86 opcode or Power pc code.The number of 3 that can be use without major performance drop will be seen after 32 bit on GPR/FPR on any 32 platform exception to SSE/2 DP again if you can use SSE-2.

What are you trying to say here ? That a 64 bit cpu will gain susbstantially because it would be faster when using 64 file systems ? If so, you're out of your mind. Not even Apple is crazy enough to claim this.

No gain will be see but Fat must move to 64 bit and adresse range to 64 bit that a thing you have no choice if you want to lift 4GB limitation.In short FAT or little square in the memory will to 64 bit.Ram are made in matrix with collumm and adresse to use 64 adressing you will need 64 case and 2 power 64 -1 possibility.

I dont like french test
October 26, 2003 4:28:38 PM

Both the pc and apple (64)is in their small age,so the benchmarks results will change in future.One thing is clear ,spending big money for an apple is becoming not reasonable.there are very good things in apple but is it worth the price?Steve Jobs have to optimize his G5 so that it should show almost double perfomance compared to pcs in the graphics area if he wants some success(anybody can point out if the PPC is capable for that?).Or he can cut the price to half.
October 26, 2003 6:13:04 PM

Quote:
What's worse though, is that I *really* can not understand why you would throw in the NUMA post in a discussion on the FX' clock scaling when you obviously don't (didn’t ?) know what NUMA is ! If you knew anything at all about NUMA, you'd know it can not possibly affect performance of a single CPU FX, and can never be the reason. So why post that then ? Why ?? Trying to show off or what ?

Considering I conceded to having made a small mind error, I think it's pretty low on your part Bobby, to actually use this to your advantage when I just accepted the error. I know what NUMA is, the basics, I don't need to spell it out. I made an error, conceded, but of course you will work on that to your advantage. You're low Bobby, very low.
Quote:
And what makes it all worse, is your inability, or rather unwillingness to absorb knowledge handed to you by others (not just me) and change your POV.

Oh yeah I forgot, you had extremely convincing arguments why 2GB is going to be a limit for most apps next year. Slvr Phoenix disagreed, but hey, it's gotta be Eden who founded the error.

Furthermore, YOUR inability to dig anywhere else than where you were involved in a thread, shows that you don't know me very well to accuse me of not accepting proofs or good explanations from credible people. Congrats again, judgeman Bobby.

Quote:
its my POV and I’ll stick to it

Ah so it's wrong to believe we as home users won't be needing 64-bit any soon, if not even next year until around the end?
Quote:
“Intact server components circuits” ??

Xeon systems DO have extra features. The pin amount isn't there for nothing. It does have extra features for multiprocessor configs, often amounting to server purposes.
The latter, was, ok, an error. Whooptee do, we got now 3. Being reasonable, thanks for correcting me.

Quote:
An instant classic !

You think you're that smart, to actually reuse once more a very low form of argumenting, where you take back a quote where it was explained afterwards it was a semantics issue. But you needed it to prove a point. Bobby, you're LOW, you have little to prove if you use things that have been cleared.

Quote:
Mail Carmack ! You have some revolutionary ideas young Einstein !

How that is a blunder, I wouldn't know. Slvr as well agreed we don't have proper graphics programmers, but of course, you will use it against me.

Quote:
and a “packaging die” ?

Once again reusing a cleared argument . My god you truly are grasping at straws.


It has occured to me you corrected me on 4 issues: Not knowing how the OS works with the 4GB it has and the 2GB limitation, thinking there were bridges on the Xeon core, 2 errors based on personal error and not misinformation (The NUMA reference, and the packaging die thing). If it's any consolation, I could include the memory controller thing. It may be different, but not THAT different.

Now what's even worse: you actually hold it against me things I clarified, or conceded to but used it again. You're the kid who will use a reason to protect his cause when the parent explained him what happened, but he'll use it the next time he's angry to actually think he has a point. You don't accept people do errors.

Cry me a river BB, you are a low form of human, you use the above methods, very low in any kind of case, to prove something. I did errors, learned them, but you will continue going lengths to try to ruin my pleasure here. Guess what, you didn't. I'm now actually turning the tide here, laughing at the fact you would be lower than I.

Get over it already and move on man. Because from now on, whenever in a thread you jump on me, I ain't replying.

--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol: 
Anonymous
a b à CPUs
October 26, 2003 9:21:08 PM

>Considering I conceded to having made a small mind error, I
>think it's pretty low on your part Bobby, to actually use
>this to your advantage when I just accepted the error.

The point is not that you made an error.. the point (or my question) is why you posted it in the first place. I can't see any other reason as pretending to know something, so showing off. Sorry for ruining that

>Oh yeah I forgot, you had extremely convincing arguments
>why 2GB is going to be a limit for most apps next year.

That wasnt exactly the only point that was discussed.

>Slvr Phoenix disagreed, but hey, it's gotta be Eden who
>founded the error.

Slvr Phoenix has everyright to disagree with me,just like anyone. But unlike you, I can only recall one factual error from him recently (his claim some collegue would have exceeded 3 GB on 32 bit windows). And unlike you, I've never seen him pretend to know something he didn't.

>Furthermore, YOUR inability to dig anywhere else than where
>you were involved in a thread, shows that you don't know me
>very well to accuse me of not accepting proofs or good
>explanations from credible people

A good proof or explanation stands on its own, no need to judge the people behind it. Furthermore, did you really expect me to read another gazillion posts of you to make up my mind ? I've wasted enough time as it is.

>Ah so it's wrong to believe we as home users won't be
>needing 64-bit any soon, if not even next year until around
>the end?

No, the "its my POV regardless" attitude applied to much more then that issue.

>Xeon systems DO have extra features.

This would be a good time to stop trying to prove you where right.. really. I don't think you want me to screen other threads for your posts as well. See, its not only about being "right" here, using a vocabulary like "Intact server components circuits" really does say something about the knowledge of the poster. Maybe not to you, but it does to me and countless others. Sorry if you can't see that, maybe you will in a few years.

>You think you're that smart, to actually reuse once more a
>very low form of argumenting, where you take back a quote
>where it was explained afterwards it was a semantics issue.

see comment above.

> you have little to prove if you use things that have been
> cleared.

Hey, you asked for it. Why should I now exclude errors on which I spent an ungodly ammount of time correcting you already ?

>You don't accept people do errors.

I sure do. read the threads again, I started out very politely and patiently correcting your errors. However, I already outlined what I don't accept, no need to repeat myself.

>you are a low form of human

Yeah I probably am, arent I. Ruining the pleasures of a kid that seems so desperate to get any form of respect around here, that had already achieved a "forum master title", that loves pretending to know something, and then I ruin all that by pointing out his errors. And then upon his own request, I regroup a small dozen of those in this post. I'll probably burn in hell for that, I mean I should really.

I'll grant you one point though; I have been focusing on you lately, for several reasons; before I came back here, I started reading for a while, and saw an enormous ammount of childish bickering, flame wars, tons of pseudo technical nonsense discussion that often only served as arguments in the fanboy arguments, not discussion about the technical issues themselves. Not everyone of course, but a large majority. So I gave myself the mission impossible of trying to talk at least some sense into these forums, and to pick a few balloons.

Since you seem to spend half your life on this forum, have some "resident forum fixture" or whatever title, craving for recognition and being right while at the same time displaying an astonishing level of cluelessness, you are an obvious and easy balloon to pick. And by never just accepting a correction, but instead desperatly trying to prove you where right in the first place while constantly making new factual errors in the process, you brought this upon yourself really.

a few illustrations to go along with that:
<A HREF="http://www.winternet.com/~mikelr/flame16.html" target="_new">http://www.winternet.com/~mikelr/flame16.html&lt;/A>
<A HREF="http://www.winternet.com/~mikelr/flame38.html" target="_new">http://www.winternet.com/~mikelr/flame38.html&lt;/A> (those shoes are obviously Omid's :) 

To end with a positive note and to your credit, sometimes you also remind me of this:
<A HREF="http://www.winternet.com/~mikelr/flame64.html" target="_new">http://www.winternet.com/~mikelr/flame64.html&lt;/A>

= The views stated herein are my personal views, and not necessarily the views of my wife. =
October 27, 2003 12:35:06 AM

Quote:
This would be a good time to stop trying to prove you where right.. really. I don't think you want me to screen other threads for your posts as well. See, its not only about being "right" here, using a vocabulary like <b>"Intact server components circuits" </b>really does say something about the knowledge of the poster. Maybe not to you, but it does to me and countless others. Sorry if you can't see that, maybe you will in a few years.

Don't twist my words BB. I asked if the Xeon's server components therefore the circuits constituting them were intact transitioning to a P4 packaging. So now you twist words huh?
Quote:
see comment above.

Yup, you didn't get what I meant indeed.
Quote:
Hey, you asked for it. Why should I now exclude errors on which I spent an ungodly ammount of time correcting you already ?

So to prove that Spud doesn't deserve to come back to this forum, we will use 2 year old quotes from a time where he committed his own errors but learned from them, just to send it against him huh?
When I ask errors, I don't ask for little things I miswrote but knew what I meant. The packaging die was a result of a phrasing error. Using it against me doesn't prove I don't know what I was talking about, especially when I admitted I didn't write it right. Forgive me for not living up to your image BB, for I have sinned. :eek: 

Quote:
Yeah I probably am, arent I. Ruining the pleasures of a kid that seems so desperate to get any form of respect around here, that had already achieved a "forum master title", that loves pretending to know something, and then I ruin all that by pointing out his errors. And then upon his own request, I regroup a small dozen of those in this post. I'll probably burn in hell for that, I mean I should really.

I'll grant you one point though; I have been focusing on you lately, for several reasons; before I came back here, I started reading for a while, and saw an enormous ammount of childish bickering, flame wars, tons of pseudo technical nonsense discussion that often only served as arguments in the fanboy arguments, not discussion about the technical issues themselves. Not everyone of course, but a large majority. So I gave myself the mission impossible of trying to talk at least some sense into these forums, and to pick a few balloons.

Since you seem to spend half your life on this forum, have some "resident forum fixture" or whatever title, craving for recognition and being right while at the same time displaying an astonishing level of cluelessness, you are an obvious and easy balloon to pick. And by never just accepting a correction, but instead desperatly trying to prove you where right in the first place while constantly making new factual errors in the process, you brought this upon yourself really.


Ok this ends now, we're done here. I'm tired of an obsessed person, you're like a stalker here. You won't be proving anything to me as far as I am concerned. I can learn and be corrected by many others here, and can behave rightfully with others. Don't expect me to change for you, nor expect me to respond to you. When I make my own errors I'll notice and refine myself. As far as I am concerned, an obsessed freak on the other side doesn't bother me. I continue to hold respect for the good people here, and will convey my opinion on something, whether you find it misinformed or not. Others will be free to debate my point or point out a flaw in it, you won't. I'll be glad to change and learn from others, but certainly not from Bobby Bee who is obsessed.

C'est fini ici mon cher!

--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol: 
October 27, 2003 4:38:54 AM

Ok... i hope that It HAS ended! Im not going to make any comments...only that i will try to get this Poll/topic back on track.

Optimizations on the G5 should be somewhat more previlant now. <A HREF="http://www.apple.com/macosx/" target="_new">Panther</A> is now available.

If anyone out there had a G5 I Highly recomend this Update!

Saw this once when the G5 was first anounced!
<b>THIS HAS TO BE THE BEST THING I HAVE EVER SEEN!!!</b>
<A HREF="http://www.apple.com/macosx/features/expose/#" target="_new">Exposé</A>
<b>click on the TRY IT OUT! you have to see this</b>

I am am curious however as to what perfomance improvments are noted, as i am sure there are many.(not neccessarily becasue of this 64/bit responsibility)

I will admit, i do not know very much about wether or not 64 bit will play a major roll in any perforamce increases or not at this time. However, it will in the future i am quite sure. as programs do inherently become more and more complex. this complexity is largely do to one thing, the user interface. This is important, because that is what Apple is all about...and having apple opening this up for themselfs, the user will likly reep the benifits, NOW and Later.



ASUS P4S8X - P4 2.4B - 2 x 512M DDR333 - ATI 9500 Pro(Sapphire) - WD 80G HD (8M Buffer) - SAMSUNG SV0844D 8G HD - LG 16X DVD - Yamaha F1 CDRW - Iomega Zip 250 int.
October 27, 2003 10:57:18 AM

I have to admit, Exposé is pretty nice! Part of what I liked about OS X is just how the animation is smooth and not very cluttery. The dock's a nice idea, although I am not sure if it works well on Windows environments what with a Start bar. In fact, the more I think about it, the more I feel that if Apple had Windows' amount of software, the Dock idea would likely fail on account of having so many things. I think they really designed OS X around the idea that it's going to be limited. Just a wild guess hehe.

I am well aware however that there are add-on Dock-like programs for Windows, just in case you were to suggest it.

Not too sure if the complexity of a user interface is how they should shoot to convince people we need 64-bit though, that's a bit of an odd choice there. :eek: 

--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol: 
October 27, 2003 3:41:16 PM

About the Dock... i can't understand what you see wrong with it...you think it has limited space?

from what i remember the last machine i was on had about 15 programs all lined up along the bottom and they all fit very comfortalby( they shrink to fit)...you can add or remove any quick start program to... it is vertualy the same as the start bar in windows, just with out the start button.

you have your quick start programs(as many or as little as you want) and all you current running pragrams.

just as with the windows bar it can auto hide or not.

however unlike windows you can choose the size you want it and as well the zoom thing you can adjust that too(smaller or larger) not too sure if you can get rid of it all to gether though...

windows has the title of the document labled for each running program... can remember off hand if OSX does or not.


ASUS P4S8X - P4 2.4B - 2 x 512M DDR333 - ATI 9500 Pro(Sapphire) - WD 80G HD (8M Buffer) - SAMSUNG SV0844D 8G HD - LG 16X DVD - Yamaha F1 CDRW - Iomega Zip 250 int.
Anonymous
a b à CPUs
October 27, 2003 4:24:22 PM

>Don't twist my words BB

It was a straight quote, as where all the others, no need to twist anything.

>You won't be proving anything to me as far as I am
>concerned.

Yeah, yeah, you're pretty good at that already, just carry on.

>I'll be glad to change and learn from others,

Hey Juin, he's all yours now..

>but certainly
>not from Bobby Bee who is obsessed.

Oh, don't flatter yourself so much. Gee, I sure hurt your feelings, now didnt I ? Grow up kid, and get some fresh air from time to time.

= The views stated herein are my personal views, and not necessarily the views of my wife. =
!