Sign in with
Sign up | Sign in
Your question

Can someone explain AMD CPU's to me?

Tags:
Last response: in CPUs
Share
March 30, 2003 7:58:59 AM

I've read the articles, I've searched the archives and all that happens is I get INCREDIBLY confused. I'm buying a new computer and the last time I did that was 6 years ago when I bought my P2 450. That's when buying a 450 meant you were getting a 450. Now with this AMD stuff, supposedly when I buy a Barton 2500+, it only runs at 1.83GHZ. So why would they call it a 2500 when it doesn't go that fast? I am building a computer to play games, that's all I want. I won't be using Cad, Photshop or anything but playing games and maybe converting some movie clips. So what should I go with? I was set on the Barton 2500+ becuase I thought I was buying a 2.5GHZ processor but I guess I was wrong... P.S. I'd prefer not to have a reply saying "Get a P4 3.06 GHZ" because that's WAY out of my price range, I was looking in the range of a Barton 2500+ or an XP 2700+. Thanks, I await someones wisdom.

<P ID="edit"><FONT SIZE=-1><EM>Edited by Kizane on 03/30/03 04:20 AM.</EM></FONT></P>

More about : explain amd cpu

March 30, 2003 8:30:04 AM

that "2500+" is actually called Pentium Rating or Performance Rating. This all starts from the competition between Intel and AMD when they lauched Pentium IV and Athlon XP. Due to technological differences, an Athlon XP is able to outperform an Pentium IV (or at least performs like a P4) even clocked at slower frequency. Therefore, because of marketing strategy, AMD came out with the idea of PR. As example, the Barton 2500+ is called so because AMD claims that even clocked at 1.83GHz it can outperform a Pentium IV 2.53GHz so it's rated at "2500+". (this is a bad example I known because Barton 2500+ is unable to beat P4 2.53GHz)
So now if you are going to buy an AMD CPU, Barton 2500+ is not a good idea. You should get Thoroughbred-B 2700+ instead.
Hope I'm speaking English =)

You never know how stupid you are until you have done something stupid enough for you to realize it.
March 30, 2003 8:38:52 AM

Thanks for some of the clarification. So what does it mean that the Barton runs at 1.83GHZ? I'm looking at the XP2700+ now just because it's faster. If it indeed is on par with the P4 2.7GHZ.. You've been very helpful.
Related resources
March 30, 2003 8:50:39 AM

It means that the Barton 2500+ runs at 1.83Ghz..... Its just that a barton of 1.83Ghz can perform the same as a P4 at 2.5Ghz - The Athlon does more work per clock. The same goes for the 2700+

<A HREF="http://service.futuremark.com/compare?2k1=5467618 " target="_new">Almost Breaking 12k!!</A>
March 30, 2003 12:01:57 PM

Just FYI:
Intel is the market leader so they pretty much set the standards for how processors are measured. It has been common for CPUs to be measured in MHz or GHz, but this is not the only thing that determines a processors "processing" speed. Thus a processor with less cycles per second (Hz) can outperform a processor with a higher Hz rating if counting instructions per each of these cycles. Think of it this way: If I type 60 words a minute with one hand and you type 60 words a minute with two hands, my wpm per finger is twice that of yours even though we put the same amount of words on the screen.
Intel's new mobile processor (pentium-M) is very slow as far as Hz goes, but a 1.6GHz pentium-M will perform better than or equal to a Pentium 4 2.4GHz or an XP 2400 (1.93 GHz).

Hope this helps a little,


Poor windows was not a brave soul. Threatened by the force of General Protection and Major Problem the little OS committed suicide by hanging.
March 30, 2003 3:03:04 PM

Actually, it's not based on Intel's procs, or so AMD claims. What happens is that in order to keep up with the massive Mhz leaps of the P4, AMD had to pull some marketing trick. Compared to the ThunderBird, Palamino was like 15% faster per clock, so it had a higher rating. As cores improve and higher efficiency is gained, the difference between actual Mhz speed and PR rating grows considerably (though I must admit AMD seems to tout its horn a little too much with certain new processors).

Hilbert space is a big place.
March 30, 2003 6:23:25 PM

You've just landed square on the reason why I've pulled all AMD based products from my availability list. AMD LIES to it's customers.

Yes, their CPUs will outperform a Pentium IV at the same clock speed. This is a good thing... but in trying to capitalize on this they have created so many differing core designs, so many different ways of measuring performance and so many different part numbers that no sane person can keep track of them. It's just doo damned confusing.

Be smart... go Intel. With Intel the decision is between Celeron and p4 and you buy as much speed as you can afford. If you are doing typical homeowner stuff look at the new Celerons (1.7ghz and up). If you are doing a lot of gaming or cpu intensive stuff the extra bucks for the P4 is well justified. Intel also does a much better job of dealing with heat... their stock coolers are just fine, they have on-chip heat spreaders and there are on-chip safeguards to prevent disaster. Plus, with Celeron and the P4 now mounting on the same motherboards upgrading is easy.

AMD was the way to go... until they started this "core of the week" crap and went to "Performance Ratings" instead of clock speeds... it's confusing the hell out of their customers.


--->It ain't better if it don't work<---
March 30, 2003 6:54:25 PM

unless you only do some internet and e mail nver go with a celeron in that case AMD's are better you right however on the fact that AMD is getting a bit awkward with its ratings the 3000+ overall gets nowhere nere the performance of the PIV 3.06

SL6EF OC's GOOD
March 30, 2003 8:21:49 PM

The Celeron is not to be underrated. For those doing the basic home owner stuff --banking, a little investing, messing about on the web, a little word processing, maybe a couple of games etc.-- the Celeron is a perfectly reasonable choice. At 1.7ghz (or better) the new generation Celeron provides more than enough git up and go for most general purpose applications.



--->It ain't better if it don't work<---
March 30, 2003 9:29:39 PM

Bah, I tried many Celeron P4 systems at the local Radio Shack and such, and they ALWAYS had some kind of lag in WindowsXP. They simply are BAD with 128K L2. Menu fade effects lagged, highlighting with blue boxes was laggy, loading was sluggish as well.
The only Celeron I truly respect is the Tualatin 256K L2 one, now THAT made sense.


--
This post is brought to you by Eden, on a Via Eden, in the garden of Eden. :smile:
March 30, 2003 9:53:21 PM

Your question is an example of why AMD's PR Rating exists. People look at the speed of the processor (1.83Ghz compared to an Intel at 2.53 ghz, for example) and think the Intel processor is WAY faster. Fact is, that just isn't true. The AMD processor calculates more instructions per clock cycle and has lower memory latency. If you compared an Intel P4 clocked at 1.83 ghz, and an Athlon at 1.83 ghz, the AMD would totally annihalate the Intel. So AMD uses the PR system to try and counter the fallacy that "megahertz = performance." It just isn't true.

Now, for the most part, the Athlon processors are better than Intel at multitasking and in office applications. (Unless you throw down $535 for a 3.06 P4 with HT.) The two processors really are about equal for games--the Intel wins in games that have SSE2 support, but not many games do.

In the final analysis, you will very rarely NOTICE a difference between an AMD or an Intel when the PR rating / mhz are equal. You may see a difference in benchmarking scores, but a 2.53 gigahertz P4 and a 2500+ Barton are going to give you pretty much the same experience.

I personally think AMD processors are "peppier" in Windows XP. That's just my personal opinion, though, your mileage may vary.




<-----Insert witty sig line here.
March 30, 2003 11:01:54 PM

It's not that bad dude, I don't spend time at all studying the specific revisions and pr ratings and know them all. Not to sell amd on that basis is ridiculous. Also, they don't lie. I mean one could easily argue on the same basis that Intel lies in that there was some general conformity to ipc before P4, and now that's gone b/c them. And BTW, I don't know what kind of customers you get, but in certain cases Intel CPU's are just junk, esp in intense fpu work. So one would not be smart to get an Intel and pay for as much speed as they can afford in that case cause u could pay a little and get much better performance out of an AMD. Working with a lot of quantum simulations, I have experienced this first hand many many times.

Hilbert space is a big place.
March 30, 2003 11:08:16 PM

1.7Ghz Celeron- $53
XP1700+ Thoroughbred- $52
Guess which one won't lag in XP.
And the lag is a very real thing, I was fixing a friend's comp who had a 2.0Ghz Celeron, and OMFG! System pissed me off to no extent w/ how slow it was, and yes, it was optimized to the full.

Hilbert space is a big place.
March 31, 2003 12:33:05 AM

Put simply. Performance matters. Mhz doesnt.
A barton @ 1.83Ghz is more than enough to play games, especially since most of the work is done by the VIDEO CARD.

the Barton XP2500+ has the added bonus of being considerably overclockable, should you desire that.

<b>Damn War! I'm too young to watch other people die!</b>
<A HREF="http://members.iinet.net.au/~lhgpoobaa/images" target="_new">My Images!</A>
March 31, 2003 1:30:50 AM

They lag... compared to what?

If you are used to a 3ghz pentium or a top end AMD any 2ghz processor will seem slow and clunky. If, on the other hand, you are used to a 600mhz duron, the celeron 2ghz is a racehorse beside your snail.

Remember when we all marveled over the speed of 486s?

It's all relative.



--->It ain't better if it don't work<---
March 31, 2003 1:34:17 AM

I stopped selling AMD for 3 reasons:

1) They make up less than a quarter of my machines but occupy half of my service time.

2) I can sell 3 Intel systems in the time it takes to explain AMD's speed ratings and most often people opt for Intel anyway.

3) I got real tired of "heat emergency" calls.

As for the kind of customers I get. Well, I'm lucky. I get the kind of customers who trust me to make careful and sensible choices for them and are willing to pay high end prices to have custom built systems with 2 year warranties and full on-site service. Most of them are very well informed, a few to the point where I seek out their advice on things.



Bottom line... read my sig.

--->It ain't better if it don't work<---
March 31, 2003 1:36:25 AM

What he means is that past a given point, when u click on start menu or right click on an object/ any instance where the fade effect in xp is used, everything is smooth and instantaneous. My 1.2Ghz looks the same in that respect to my friend's XP1900+. However, celeron doesn't quite get the fade effect very smoothly/ instantaneously, and that's what's bothersome, especially about a processor running at 2.0Ghz.

Hilbert space is a big place.
March 31, 2003 1:50:23 AM

What type of customers do you get? They shouldn't be getting heat emergencies in any case if average user, b/c he/she should not be removing the heat sink and all. Another thing is P4's also generate heat, and when hsf fail, they simply overclock, but that still is a heat emergency. AMD's procs r just as reliable.

Hilbert space is a big place.
March 31, 2003 1:54:20 AM

Menu fades? That's your criterion for CPU performance?

Go into the registry and disable the paging executive and watch what happens... better still, just turn that s**t off... XP's eyecandy is the biggest waste of CPU horsepower ever invented.

There are exactly 2 critera that matter:

1) Does it do your essential tasks well enough?
2) Does it keep breaking down?

Beyond that, it's all bells and whistles.

Celerons are great little chips for writers, students and net surfers. They provide a low cost, high reliability system that's easily upgraded and doesn't require 307 case fans and a cooler that sounds like an F16 to keep everything sane.

Think about it... when's the last time you saw an Intel user talking about serious system instabilities or writing about heatsink problems? Yes, you do see it on the various boards occasionally but look how these discussions are almost totally dominated by AMD... there's a hint in there.

As I've told you in other threads, I have about 200 systems under my care. About 50 of those are AMD but they occupy fully half of my service time; seems like something is always going wrong with them. The Celeron and P4 systems I've sold, for the most part, disappear and I never hear about them again. In fact, I spend more time on Monitors and Printers than on Intel computers.

That's gotta tell you something.




--->It ain't better if it don't work<---
March 31, 2003 2:20:54 AM

We're just using the fade example as a an instance of real life performance where celeron is trash. I've seen intel users have heat issues, in fact there's an ongoing discussion in the cpu forums about high temps. Thing is that most people here r amd users who build the systems themselves, and someone will always have incorrectly installed an hsf. P4's generate a lot of heat too. YOU may never hear of ur customers' problems, but that doesn't mean they don't exist. Friends of mine ordered a system from a single guy who made comps, and he fixed it several times for em, but I live right next to them. I help them out rather than that guy. Also, people will be bound to be skeptical with their amd system, which is perhaps part of the problem with people calling u- creating a fault by looking for one. Also, there's the issue that yes, there r certain chipsets/ mobo which aren't too good, and then there are ones that are fine. It's up to the builder really there, so obviously a poorly chosen mobo will result in problems. If amd's were so bad, why would so many oems sell them? You have failed to mention specific cases, rather given general statement (for all we know, they may have windows problems), nor have you answered my questions as to what type of customers do you have. It honestly seems like bias on your part. This summer I will be integrating several amd systems for a hedgefund. I already configed them a test one and it absolutely shattered their intel systems' performances, so we shall see how everything goes from there. Given a good, sturdy setup, and AMD system runs fine.

Hilbert space is a big place.
March 31, 2003 2:35:12 AM

Teq:

I think what you're failing to acknowledge is that the "bells and whistles" are more important than 100% stability to many enthusiasts. A lot of us would rather be able to turn everything on and enjoy the eye-candy. I mean, if that stuff was totally unimportant, why not keep running Windows 3.11 for Workgroups level graphics? It may not be practical, but I think it's desireable.

After all, a Dodge Viper is not as reliable as a Toyota Camry. But the Viper sure is fast and pretty. :) 



<-----Insert witty sig line here.


<P ID="edit"><FONT SIZE=-1><EM>Edited by Twitch on 03/30/03 10:36 PM.</EM></FONT></P>
March 31, 2003 3:23:56 AM

Mmm, just like that Aston Martin DB7 or modded Opel Speedster I'm eyeing and may buy in a few years. It'd have a lot of bells:) 

Hilbert space is a big place.
March 31, 2003 3:38:55 AM

I'm not failing to acknowledge anything.

I have several friends and customers who really like the eye-candy of XP and even a couple running Windowblinds for some real wild visuals... nothing wrong with that so long as you understand you take a performance hit for doing it.


--->It ain't better if it don't work<---
March 31, 2003 3:51:47 AM

Ok, this is getting a little ridiculous.

I am not going to spend a lot of time justifying my decisions to you.

As I've already explained I made a decision based on the disproportionate amount of time I've been spending on AMD systems. If this troubles you, I'm sorry... but I have no intention of going bankrupt selling stuff that just doesn't work right.





--->It ain't better if it don't work<---
March 31, 2003 4:25:20 AM

In the end what gets to me is that YOU seem to mess up selling these amd systems. You say you have many issues with them, not worth time, confusing, etc. Yet there are so many others out there who say the exact opposite of you, and also have many more customers.

Hilbert space is a big place.
March 31, 2003 4:49:19 AM

Ok, this just got seriously ridiculous.

I presented a differing view to yours way over in the CRT Monitors discussion and here you are trouncing on my every word. Now you are accusing me of incompetence because I made a business decision...

Believe me, I've got far better ways of wasting my time than arguing with some fool who's nose is out of joint.

Bye Now!



--->It ain't better if it don't work<---
March 31, 2003 5:28:17 AM

dont listen to Teq

hes prolly 45 and grew up on Intel (well, i did too i guess) and is taking advantage of AMDs wierd PR to bring them down



but really.. a XP2600 (tbredB) is cheaper than a P42.53 and quite fasteer


really man. if you take a AthlonXP and put it with a Intel with the same Performance Rating, the AMD will own it 9 times outta 10. once you get into P4s with a 533 fsb, then its a little different.. but these cost a bit more also
March 31, 2003 5:38:01 AM

Well boys... AMD has trully lost it

www.canadacomputers.com

The top intel model 3.06HT is cheaper than the top AMD model!!! I never thought i'd see the day.. but here it is...



To err is human... to really screw things up you need a computer!
March 31, 2003 5:52:29 AM

Quote:
XP 2400 (1.93 GHz).

The XP2400 runs at 2ghz

Dichromatic for your viewing plesure...
March 31, 2003 6:23:23 AM

guys, i dont know who's post to reply to. but i always thot the Athlon XP rating is what u get when compared with the Athlon Thunderbird CPU. what i had thot is, a Athlon XP 1800+ performs on par with the equivalent of the Athlon Thunderbird 1800MHz though the former runs only at 1533MHz. But since Thunderbirds have always performed on par and sometimes even better than comparable P3s/P4s, the XP ratings apply to P4 equivalents too. Correct me if i am wrong.
March 31, 2003 7:17:20 AM

That's how it works--in theory. However, when Intel moved to the Northwood core and a 533 megahertz FSB, everything changed, and it became evident that AMD was really basing their PR rating to the equivalent P4 processor. And that's where many of the objections to the PR rating derive. And now, with Barton and its 512kb of level 2 cache, the whole PR rating scheme has become even more muddled.

In reality, if AMD had stayed with the "PR rating = equivalent T-bird" formula, there would probably be a lot fewer objections, since the PR rating would have remained consistent.





<-----Insert witty sig line here.
March 31, 2003 7:18:35 AM

you say you custom build systems for people and that 1/4 are amd and occupy half your time.. why? do you have some kind of problem wth building amd? you must be setting them up badly or using poor quality components elsewhere cos AMD are as reliable as intel when configured corectly. the new amds run cooler than intel. and what do you call a "heat emergency"? do you mean that some1 like you told them that amd run too hot? you dont need loads of fans in any system.. ive got 1 case fan in mine other than cpu hsf and guess what? my system is neither hot nor noisy. anyway no offence but it kinda sounds like you need to learn how to setup computers properly

I'm out of my mind, but feel free to leave a message.
March 31, 2003 3:57:31 PM

Quote:
In reality, if AMD had stayed with the "PR rating = equivalent T-bird" formula, there would probably be a lot fewer objections, since the PR rating would have remained consistent.

Are you kidding? Compare an AXP 1500+ to an T-Bird 1.4GHz and then try to tell me that the PR rating <i>ever</i> was based on the performance compared to a T-Bird.

Let's admit the truth, it was nonsense from the very beginning. The <i>only</i> reason that AMD came up with their rating system was so that they could charge prices that were closer to Intel's.

<font color=blue><pre>If you don't give me accurate and complete system specs
then I can't give you an accurate and complete answer.</pre><p></font color=blue>
March 31, 2003 4:25:30 PM

Come on folks, cut Teq some slack. It was a sensible business decision. You have to admit that with Intel it's <i>really</i> easy to set up a rock-solid system. With AMD though you <i>really</i> have to do your research first to get even close to the same level of system protection.

Just a year or two ago most AMD heatsinks came with just the one crappy little clip that was fine if your PC never moved, but if your wife redecorated the house regularly would be very prone to breaking the socket's tab and causing the heat sink to fall right off. (Not to mention shipping problems.) P4's simply never had that kind of a problem with their heat sinks.

On top of that, just a couple of years ago AMD didn't even <i>have</i> decent protection against a heat sink failure. If you had anything, you'd be lucky if it shut the system down even if the fan just failed. Should your heat sink actually come off (or even just reseat itself uneasily after a tab bent or broke) you were usually screwed and ended up with a fried CPU. Meanwhile Intel's P3 handled that just fine by shutting itself off in time to prevent frying, and the P4 went one step further to throttle itself so that you could save your work and then shut down the system to fix it.

A year ago (and definately earlier than that) VIA made some <i>really</i> touch-and-go chipsets. And many of the motherboard vendors considered AMD to be 'budget' and so cut every corner possible while making AMD motherboards. You had to research the market well (or just depend upon luck) to get a rock-solid AMD motherboard. Meanwhile, just about any motherboard with an Intel chipset was trustable, and if that wasn't good enough, you could go with a motherboard from Intel themselves and know it would be as stable as could be.

The simple fact is that a year ago (and more) building a rock-stable Intel system was as easy as bobbing for apples. While building a rock-stable AMD was either a matter of luck, or a considerable investment of time. And if you didn't spend that time (or didn't have any luck) you were pretty unhappy with the results.

It's not that all AMD systems were/are unstable. It's that making a stable AMD system takes research and time. Making a stable Intel system doesn't.

It's much less true today thanks to heat sink vendors and motherboard vendors <i>finally</i> starting to consider AMD seriously instead of just as a 'budget' system. It's also much less true today thanks to VIA <i>finally</i> learning how to make their chipsets better <i>and</i> learning to actually write dependable firmware drivers. And even more so for nVidia getting into the AMD chipset business as well.

Yet even just a year ago (and certainly two or three years ago, which is no doubt where the systems that Teq spends service time on come from) AMD was a much more demanding mistress. You really had to research AMD systems well before building them back then. And it was incredibly easy to build an AMD system that would have faults show up later back then.

Facing that kind of a frustration, I really can't fault anyone for simply not wanting to deal with it.

And then there's just simply trying to explain to a customer the whole rating that isn't actually MHz concept. Sure, we get it easily. We're however not exactly a 'typical' customer. I've had to deal with plenty of people (usually family and friends, sadly) who even after an hour straight of explanation can't even understand that a Dell system, a Gateway system, and my home system are all based on the exact same standards and can have the exact same parts inside of them.

And as Teq said, even if you <i>do</i> manage to finally get the customer to understand, they'll still usually go with the Intel system anyway.

If I was running a business and I had a product that moved only 33% of the time (or less) compared to another product that moved the other 66% of the time (or more), I'd streamline my inventory and simply stop selling the product that the customers clearly don't care about buying. It's called smart business. Anything that's sitting in inventory and not getting sold is <i>wasted money</i>. You don't make money by buying even more inventory that you <i>know</i> you won't get a quick turnaround on.

So for goodness sake, cut Teq some slack. He made a perfectly understandable <i>business</i> decision, and he's got some pretty good reasons to back them up in my opinion. By this-second-today's standards some aren't <i>as</i> true, but at the time that he probably made them I'd bet that they were.

<font color=blue><pre>If you don't give me accurate and complete system specs
then I can't give you an accurate and complete answer.</pre><p></font color=blue>
March 31, 2003 4:30:16 PM

I think that is fair to say that you stopped selling AMD for a business reason. Spending over half your time supporting less than 1/4 of the sold systems would cause me to do the same thing. Your comments are appreciated.

However, I have only built AMD systems for people. So far, I may have just been lucky, but I have not had to service a single machine. Most of the ones I build are for family members and they call me all the time for computer problems. So far, all their problems are with running the new software they buy and not with hardware.

When looking to buy something, I always look at reviews first. Then I always select to just see the negative reviews. I think that the negative reviews often have more valuable information. (No product is perfect without any drawbacks) Obviously, there are ways you can build a system that would make an AMD choice less stable. That is attributed mostly to the much wider selection of motherboards. Based on price/performance and pure numbers of people that have perfectly stable AMD systems I wouldn't throw that choice out though.

I would recommend getting an AMD Athlon XP 1700+ Thoroughbred B CPU. They are very cheap, run very cool, and are very overclockable. You might not quite get it up to 2700+ performance, but you should get at least 2400+ performance and you will have paid much less (maybe $75.00 after a good heatsink/fan). If you really don't want to overclock, then go ahead and get a 2700+. Or go for a P4 2.8 GHz for only about $45.00 more.
March 31, 2003 4:43:25 PM

You know, you guys really must stop defending AMD every time you can... I completely understand Teq, you didn´t respect the guy at all. Look at this:
Quote:
AMD are as reliable as intel when configured correctly

It sounds stupid to me. OK, let´s assume, for the sake of argument and nothing else, that this guy is really incompetent (sorry, teq, I´m just trying to make a point here). Then <i>all</i> systems he configures are configured inadequately, right? Now if it takes so much more effort to "correctly configure" an AMD computer and you have to consider carefully their components, then I´d say (1) AMD´s processors have compatibility issues and (2) Very unlike Intel, AMD processors need highly specific conditions to operate. His Intel systems didn´t have any trouble! Why, if he´s so incompetent as you say? There is <i>no way</i> you can possibly argue logically that this statistic is not to Intel´s credit in any way (maybe Intel´s systems are easier to configure correctly! It´s much to their credit!), unless he´s being MORE incompetent with AMD than with Intel - and I don´t see that happening. Get reasonable, folks. If you´re still up for an argument, I suggest you use reason instead of love for AMD and start respecting the people you´re posting and don´t flame every word they say.

Anyway, slvr_phoenix is right in saying that the PR rating was bound to give AMD a headache. I think its problematic nature is just starting to surface, and we´ll see more trouble ahead, as processors don´t only scale in speed, but also get architectural differences which alter their performance in various areas in different ways... AMD better come up with something more solid really fast.
March 31, 2003 5:54:20 PM

I have to disagree on the stability issue and taking time or luck to get a stable amd system a year ago. I built two AMD based systems last year and both took almost no effort to configure and get up and running smoothly. I am not a tech and I do not have the time to fool with an unstable system. One of these systems has not been turned off for 6 months (with the exception of power failures)and it does not have any stability issues. The other system was for home and I finally turned it off after 5 months of continuous use running Suse 8.0 because I did not use it enough to justify leaving it on all the time.

If a person building boxes for a living can not build a stable system based on any CPU I would highly question their ability as a tech. I am no tech and I can do so even with differing OS's.
March 31, 2003 6:01:40 PM

Quote:
AMD better come up with something more solid really fast.

Or at least something <i>supportable</i>. Barton just completely killed whatever shred of supportability was left to their schema.

I'm not entirely against some sort of a rating system. AMD's, while consistantly questionable, was at least better than nothing most of the time and made a certain business sense ... for them anyway. But it should have been based on something tangible. Now it's more like a tally sheet. Okay, we got X MHz, a super-size on the FSB, and a double-order of cache. That comes to 3000 calories. ... I mean PR. That'll be $600. Please pick up your order at the drive-through window.

(Okay, so maybe it's not quite like eating at McDonalds, but it's certainly bad enough and getting darn close.)

Personally, I think that retail CPUs and <i>especially</i> PCs should come with basic facts from government-standardized benchmarks. (Much like the nutritional information that is forced to be on the packaging of pretty much every food product.) Not that government has to run the benchmarks. They just declair which ones are 'standard' and how they <i>have</i> to be run. It's up to each manufacturer to run the right ones and run them correctly, or face fines and whatnot.

Sure, it's still not a perfect solution, but at least that way you have <i>some</i> sort of <i>fair</i> measurement for the 'less technical folk' to compare systems instead of just trusting the salesperson to be telling the whole naked truth. So people about to buy a P4 2.8GHz system with an 845 mobo, CAS3 PC100 RAM and onboard graphics can compare actual benchmark results to a P4 2.8GHz system with an 850E mobo, 32ns PC1066 and an ATI R9700Pro, or an AXP 2800+ on an nForce2 with CAS2 PC2700 and a GF4Ti4200. If <i>every</i> reseller is required to provide the information upon request (or even better, on the packaging as well) then it would be a LOT more fair to the average PC buyer who can't spend hours a day reading THG to make sure that they make the 'best' choice.

Of course then again I also believe that anyone who spends that much money on a purchase without doing their own research deserves what they get if they buy a lemon. Heh heh. I mean it's just common sense.

<font color=blue><pre>If you don't give me accurate and complete system specs
then I can't give you an accurate and complete answer.</pre><p></font color=blue>
March 31, 2003 6:30:20 PM

Thats why I'm going with Intel? AMD Barton XP 3200 2.2 Gig. 400 FSB. Is faster than Intel's 3.2 Gig 800 FSB Next AMD will tells us that their new Chips. Will be faster than Intel's Tejas with 1.2 Gig FSB. When AMD has 500 FSB an 2.5 gig CPU? Right!!!!
March 31, 2003 6:54:24 PM

Exactly, it was a business decision...

As I've been explaining in private messages, my personal machine is a 14 month old gigabyte 7zmmh, currently sporting an xp1800+ and 256mb of ram and I'm in no hurry to get rid of it. I like it just fine.

The problems I've been running into are financially annoying...

Mostly it's minor incompatibilities between some video cards, sound cards and Socket A motherboards. More than once I've had sound cards (for example) burble and flurp in an AMD machine but work spitting perfect on an Intel board.

The heat emergencies I was referring to come up mostly after a machine is moved. The clip and foam pads system on AMD, with such a small hard contact point, is not stable enough to survive being bumped or jarred while moving the system. Almost every time I get a heat emergency call, I end up doing nothing more than cleaning and re-seating the cooler but almost always I discover, in conversation or by observation, that the machine was moved right before the problem arose.

These aren't huge problems... but they soak up my time, taking me away from revenue increasing activities.

Every system configuration on my list is tested before it goes on the list. I also test all the upgrade parts on my list before I add them. Still, as I'm sure you can appreciate it's simply not practical to test all possible combinations, problems do arrise despite my every attempt to minimize them.

Those who buy a system from me pay a good price but they get a free 2 year service contract. During the first 2 years they will never pay labour on repairs or upgrades of their hardware and they pay for parts only when the individual part warrantees expire. During this 2 year period I deliberately visit each machine 4 times for cleaning, voltage checks and any necessary adjustments. I also have IM contact with my customers, during office hours, so they can ask questions or put in service calls interractively.

What kind of customers do I get? The kind who understand the value of service and are willing to pay for systems they can count on. Most are very savvy people who understand computers and know that a good deal doesn't always mean paying the lowest price.

For those who like AMD... good. I like it too. But lets be realistic about this... there are problems and those problems can and do play hell with the service promises I give my customers.




--->It ain't better if it don't work<---
March 31, 2003 7:16:46 PM

To build a stable system, you need trustworthy parts...

You say that a good tech can build a stable system with any combination of parts... The truth is that a <i>smart</i> tech knows which stuff is going to be a problem and doesn't inflict that upon their customers. They count on the reliability of my systems to do their jobs, I count on it to make a profit. Nobody comes out ahead when you sell crap.

Remember, this isn't a hobby and it extends beyond my circle of friends. I don't have infinate time to spend tinkering a balky system. Time is money and before I bankrupt myself taking care of a quarter of my systems I will swap them out and sell the problem children to someone who does want to tinker.



--->It ain't better if it don't work<---
March 31, 2003 7:54:31 PM

This has absolutely nothing to do with our discussion in the CRT forum. I do not wish to turn anything into a flame war/ fight, rather a debate. Obviously I find something to argue against, but you seems to take it specifically as an attack. BTW, don't call me fool. Somehow I've magically gotten AMD to work for a business where a downed comp can cost millions in several seconds (so it's my ass that gets kicked then, and it's not fun- I've seen it happen). I will be further implementing this over the summer. You sir are the incompetent one it seems. Like someone said, you probably have grown up with Intel and it's been ingrained as the only trustable solution. Like you, this AMD experience is something professional- it goes beyond a circle of friends, and into something very, very serious, certainly more so than a user's system going down so they can't look at their email.

Hilbert space is a big place.
March 31, 2003 8:37:41 PM

Why would anyone buy a Celeron when they could have an Athlon at a comparable price !?
March 31, 2003 9:02:57 PM

Oh my, now we are hurling the "incompetent" word around...

Well, ok, lets play "wigger wagging" for a minute...

I'm 52 years old, have just over 25 years of computer experience, more than 30 years in electronics and before that I was a hobbyist building experimental audio gear for fun and profit.

So, lets do the math... I've been working with computers since before there were PCs. The first machine I worked with was a DEC pdp8 with a KSR33 telletype and 8k of register memory. From there I moved to IMSAI 8080 "build your own" machines and then to Z-80s and 6809s. In my time I've had experience with almost every CPU out there... Zilog, Intel, Cyrix, Via, Motorola, AMD... you name it.

I was also on the design team of a computer based on the 8088 that was on the market a full year before IBM got into it. I've designed everything from simple interfaces to cash registers to industrial controllers and I've worked in just about all areas of electronics, ranging from microprocessors to tv repair. I also did a 5 year stint as the service manager for a national computer/cashregister company in Canada.

Although I do not have formal engineering papers, I did write and pass the electronics exams without taking the course.

Now... I work in a highly competative area providing rock stable machines to the SOHO market. Do you really think I would have gotten that far... that I would stay in business if I didn't know what I'm doing?

Get real!

Look... I don't know why you think you can dump on someone like this, but I can tell you this is really starting to piss me off. I have not impuned your reputation or even challenged your positions... yet here I stand accused of incompetence simply because you like AMD and I've decided it's not a good way to earn a living.

Tough. Not my problem, pal!



--->It ain't better if it don't work<---
March 31, 2003 11:17:29 PM

B/c they are biased and beleive amd is unstable.

Hilbert space is a big place.
March 31, 2003 11:21:35 PM

Earn your living however you want. I state tho that AMD is a clear choice compared to a celeron and in certain cases P4. It's simply a debate for others who r reading.

Hilbert space is a big place.
March 31, 2003 11:40:33 PM

If I were 52 yrs old I would act more mature then you.


*MadHatter's brain just got a blue screen of death*<P ID="edit"><FONT SIZE=-1><EM>Edited by TheMadHatter on 03/31/03 07:53 PM.</EM></FONT></P>
March 31, 2003 11:55:27 PM

listen (or rather read, sorry), Mr <i>flamethrower205</i>, how can you possibly expect people not to call you a fool when you´re blatantly saying
Quote:
You sir are the incompetent one it seems.

If you want a rational debate, stop calling other people incompetent! And if you <i>think</i> you are indeed talking to incompetent people, then do something better with your time and please get lost. Stop insulting people. If you are that much wiser than us (which I honestly doubt, but hey, that´s just me) then insulting us is not the way to spread your (probably nonexistant) wisdom around.

Oh, and nothing personal, by the way. Just try to respect people, and you <i>might</i> just get respect in return! :smile: Remember, Teq only called you a fool after you said these words:
Quote:
In the end what gets to me is that YOU seem to mess up

Doesn´t sound like you´re respecting the guy´s opinion.
April 1, 2003 12:02:52 AM

When you do get to my age you will understand that maturity also entails not taking crap from people who plainly are looking for a fight.

I have yet to see anything even vaguely resembling a reasonable rebuttal to my original position... just a bunch of AMD fans with their noses out of joint.

Want to handle this in a mature and adult way?
Lay off the insults and provide a reasoned point of view.


--->It ain't better if it don't work<---
April 1, 2003 12:09:22 AM

Quote:
If I were 52 yrs old I would act more mature then you.

That´s about the funniest thing I´ve ever heard... You guys are really starting to invent some creative insults, I´ve got to admit it...

How can you possibly foretell how mature you´ll be when you´re 52 when you´re not even mature enough to ascertain how pointless your remark is right now, today?

Let me try to show you...
<i>You know, be careful. When <b>I´m</b> 52, I´ll be so mature and know so many things that I´ll be SUPERMAN! :cool: And I´ll kick everyone´s sorry ass all the way to the moon!!! </i>

Did you get just how pointless it is?

If you didn´t, don´t worry, you´ll probably understand it when <i>you</i> get to 52. Or not, who knows :smile: .
!