The Barton Conspiracy?

slvr_phoenix

Splendid
Dec 31, 2007
6,223
1
25,780
I don't know. I've just been thinking about this whole Barton XP3000+ thing, and it comes to me that we really shouldn't be surprised. I mean honestly, when you think about it...

The last few AMD releases have been paper launches, which some think are ok to do and some don't. But anywho, they've also been launches where the performance leaves it very debatable if the rating system really means that your AMD's PR is going to give you more performance than Intel's clockspeed. The performance gap between the AMD/Intel bleeding-edge was minimal at best, where as AMD's PR rating <i>used</i> to mean that it would outperform a P4 of that speed. So anywho, we've been seeing the PR rating slowly mean less and less as time marches on.

On top of that though, AMD's bleeding-edge prices have been sliding higher and higher. Instead of being great price over Intel, they've been more or less matching Intel. (For the top end, not for the middle-to-low.)

So the trend has been that AMD's PR means slightly less and less, and AMD's prices are going up and up. When you really think about it, we should have seen the XP3000+ coming. The PR was bloated even more for Barton. AMD prices these over-inflated-PR CPUs the same (or more) than Intel counterparts. And so now, the price/performance just really sucks at AMD's top end.

Now here's the part where I just don't know...

A) Did AMD really mean to screw up their PR ratings on purpose with Barton <i>just</i> so that they could charge more for them?

B) Did AMD screw up their PR ratings with Barton <i>just</i> so that they could claim to be as fast as Intel once more?

C) A bit of A and B?

D) Did AMD <i>honestly</i> think that the Barton's extra cache warranted a +300 to the PR?

I suppose this post might kind of belong in the poll section, but I'm not really taking a poll. I'm just wondering what everyone else's thoughts are. Does anyone still try to put AMD up on any pedestal? Or has this been the last straw for you? Does anyone see another side I've missed to this?


PC Repair-Vol 1:Getting To Know Your PC.
PC Repair-Vol 2:Troubleshooting Your PC.
PC Repair-Vol 3:Having Trouble Troubleshooting Your PC?
PC Repair-Vol 4:Having Trouble Shooting Your PC?
 

RCPilot

Champion
I like you have noticed the same thing as well. The top end AMD's are way to expensive now days. Here's something else to consider, is the new agreement with IBM responsible for some of it as well? I don't know. But I do know that the times seem to be a changing for AMD.

I guess that price performance has meant something different to me. I choose my own level. My rule is $200 for a CPU & no more. I did violate my own rule with the XP2400 I got @ $204 in Nov. I felt the T-Bread B core was worth the extra $4. So $200 usually gets me to the latest core & I don't take such a burn. It's a good rule for me. I want a 333 FSB AMD & wanted it bad in Nov., but not until it's below $200. That way I'm just a tab off the top & still plenty fast.

If it ain't broke, take it apart & see why not!
 

nausicaa

Distinguished
Jan 13, 2003
45
0
18,530
barton will eventually replace the t-bred, and bring better performance to lower clock speeds...this is a good thing...just wish they didn't change the ratings...but this is not a failure or dissapointment, just a new chip. processors are getting better, faster, and more powerful. this may speed intel to release their 3.2 and cut prices...good also. take it in stride

<A HREF="http://www.anandtech.com/mysystemrig.html?id=22270" target="_new">My System</A>
 

tlaughrey

Distinguished
May 9, 2001
581
0
18,980
I don't really see it negatively the way you do. The AXP 3000+ offers performance close to the P4 3.06, so I think the PR is still justified. The PR is obviously not perfect, but then neither is a pure GHz to GHz comparison. I would say the Barton 3000+ should maybe have a slightly lower PR, but I don't think it's too far out of line.

As far as prices go, I think AMD is a for-profit organization, so they can try to make as much money as they want. CPU prices are always high on release and then steadily drop. You either pay more now or wait a while and pay less.

I guess I see AMD and Intel as huge corporations trying to make a lot of money. Their marketing departments are going to try to put the best spin on things possible. I don't think AMD is any more or less guilty of this than Intel, with its blue men and aliens and "clockspeed is everything" mantra. It's nice that AMD has seemed to care about the individual enthusiast/hobbyist up until now, but I don't expect them to base their whole strategy on what we want.

<i>Money talks. Mine always likes to say "goodbye." :smile: </i>
 

tlaughrey

Distinguished
May 9, 2001
581
0
18,980
Low yield
Not from what I've heard. Any data to back this up?

low performace
About the same as a P4 3.06, while running 900MHz slower. This is low "performace"? Get real.

over price
About the same price as a P4 3.06 for similar performance. What's wrong with that? The price will come down anyway.

bad overclocker
Actually, the reviews I've read say it should be a pretty good overclocker.

So basically, you just spewed out negative comments without backing them up in any way. Nice job.

<i>Money talks. Mine always likes to say "goodbye." :smile: </i>
 

slvr_phoenix

Splendid
Dec 31, 2007
6,223
1
25,780
I don't really see it negatively the way you do. The AXP 3000+ offers performance close to the P4 3.06, so I think the PR is still justified.
Close only counts in horse shoes and hand grenades. The majority of the time, it's still under the P4 3.06GHz. This is especially true if you look at other reviews that use a Granite Bay motherboard.

It's compounded though by the fact that the T-Bred 2800+ is <i>also</i> close. In fact, it really seems like the Barton 3000+ and the T-Bred 2800+ are neck-and-neck most of the time. So if the T-Bred 2800+ is tying and even sometimes beating the Barton 3000+, and is almost always losing to a P4 3.06GHz, what then actually warrants a PR of 3000+?

From all that I can calculate, the extra cache from Barton should <i>at most</i> have counted for an additional 150PR, not the 300PR that it was given. <i>AND</i> had AMD given it 150PR, then we would easily be able to distinguish between Bartons and T-Breds by the PR rating.

<i>That's</i> why I see it so negatively. AMD is clearly obfuscating the truth. The real questions in my mind are why and was it intentional or accidental?

I would say the Barton 3000+ should maybe have a slightly lower PR, but I don't think it's too far out of line.
I say it's 150PR out of line, which makes a <b>huge</b> difference in pricing. If it had actually been labelled a 2850+ instead of a 3000+ the price would be a lot more reasonable to it's actual performance.

As far as prices go, I think AMD is a for-profit organization, so they can try to make as much money as they want.
I don't disagree here. I would like to make the distinction however that over-inflation of prices <i>will</i> result in less sales. Since most of the people who would actually buy AMD are the same people who are educated enough to actually tell the difference between a Barton 2800+ and a T-Bred 2800+, let me ask you: Do you think that AMD's inflated PR to raise their prices is actually making them more, or less money? If it was Intel, we all know it'd be more. This however is AMD, and does not have many morons for customers. (Trolls maybe, complete morons no.)

I don't think AMD is any more or less guilty of this than Intel, with its blue men and aliens and "clockspeed is everything" mantra.
So you're saying that the company who invented their whole PR scheme to make PC shopping more fair for customers and who keeps trumpeting that we should compare by performance, not MHz, is not in any way more 'guilty' for suddenly fudging up their entire PR schema with their latest three processors?

You say that Intel has a 'clockspeed is everything' mantra. When was the last time that you even saw an <i>Intel</i> advertisement that mentioned clockspeed? Did the Aliens say "Look at my 2.8GHz chip!", or did they say "Look at my Pentium 4"? Did the blue men advertise clockspeeds, or did they advertise the Pentium brand name?

Sorry, but Intel is not the one pushing that mantra. It's the resellers who are trying to convince people to upgrade that are pushing it, just as they push AMD's PR numbering as though it was MHz. There's a noticable difference even there though, for Intel didn't invent the MHz system. It'd been used since the very beginning of microprocessors. AMD however <i>did</i> invent the PR system.

So can you <i>honestly</i> say that you don't think that "AMD is any more or less guilty of this than Intel"? I know that I can't.

It's nice that AMD has seemed to care about the individual enthusiast/hobbyist up until now
'Up until now' being the key. AMD <i>used</i> to <i>seem</i> to care about the individual enthusiast/hobbyist. That now is all in the past. AMD's latest actions only care about AMD, and are in fact trying to trick enthusiasts into buying chips that perform worse than the ratings they were given.

When Intel did that years ago, they had the excuse of moving to a completely different architecture <i>and</i> they had never once actually said "Use MHz to compare our chips." Sure, many assumed, but you know what we get when we assume.

AMD on the other hand <i>hasn't</i> changed their architecture very much at all. And AMD <i>has</i> told people to use their PR rating to compare their own chips against their own chips.

It's very sad if you ask me. AMD tried to stand for truth, but in the end only mislead customers even more than Intel.

but I don't expect them to base their whole strategy on what we want.
Funny, I thought the purpose of a manufacturer was to make goods to sell to customers. Since they make no money without customers, basing their whole strategy on what we want is the <i>very</i> thing that they should be doing if they want to make any sales. Without customers who actually want their product, they can't make <i>any</i> money. And when you're trying to steal market share from an entrenched competitor, pleasing customers is doubly if not triply important.


PC Repair-Vol 1:Getting To Know Your PC.
PC Repair-Vol 2:Troubleshooting Your PC.
PC Repair-Vol 3:Having Trouble Troubleshooting Your PC?
PC Repair-Vol 4:Having Trouble Shooting Your PC?
 

slvr_phoenix

Splendid
Dec 31, 2007
6,223
1
25,780
Normally I really don't back up juin, because he's in a world of his own, but this time there really is support for a comprehendible statement from him.

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

Low yield



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

Not from what I've heard. Any data to back this up?
How about this snippet from AnandTech: "<font color=green>The added transistors increases the die size from 84 mm^2 to approximately 101 mm^2. Recently, we had the opportunity to ask AMD's CTO, Fred Weber about manufacturing, and one of the tidbits of information he left us with was that AMD's manufacturing sweet-spot exists between 50 mm^2 and 100 mm^2. Once dies get above 100 mm^2, they start to be significantly more expensive than those that can fit within that 50 - 100 mm^2 range; as you can tell, Barton is at the very edge of that scale, making it difficult for AMD to maintain as large of a price advantage over Intel.</font color=green>" Less processors per wafer = lower yield. AMD has left their 'sweet spot' to make Barton, and they're paying for it in yield per wafer.

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

low performace



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

About the same as a P4 3.06, while running 900MHz slower. This is low "performace"? Get real.
Compare apples to apples. The Bartons <i>do</i> have a <i>clearly</i> lower performance than their PR ratings indicate they should have when compared to T-Breds. Of all of the reviews that I've read, they each agree that the Barton's PR rating isn't quite what it should be. Get a Barton 2800+ and you'll definately be seeing 'low performance' when compared against a T-Bred 2800+ for a quite noticable number of benchmarks.

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

over price



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

About the same price as a P4 3.06 for similar performance. What's wrong with that? The price will come down anyway.
Again, apples to apples. When we consider the performance of a Barton with the same PRs as that of T-Bred, the prices <i>are</i> too high because the PRs are too high. It doesn't take much at all to make a huge difference in prices. The drop from 3000+ to 2800+ is about $200. From 2800+ to 2700+ is about $100. Were the Bartons 3000+, 2800+, and 2500+ listed as 2850+, 2650+, and 2350+ respectively (which is <i>much</i> closer to their actual performance) you would saving about $200, $100, and $50 respectively. It seems to me that constitutes considerable overpricing.

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

bad overclocker



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

Actually, the reviews I've read say it should be a pretty good overclocker.
I've read both. Some (like THG) can really get it up there. Others get noticably less of an OC. So it seems to me that the actual stability of Barton when OCed ranges greatly from chip to chip. It is too much so to trust those higher OCs that some are getting as an 'expected' OC. So at absolute best, I see Barton as an 'average' OCer. Certainly not bad as juin states, but certainly not good as you state either.


PC Repair-Vol 1:Getting To Know Your PC.
PC Repair-Vol 2:Troubleshooting Your PC.
PC Repair-Vol 3:Having Trouble Troubleshooting Your PC?
PC Repair-Vol 4:Having Trouble Shooting Your PC?
 

Crashman

Polypheme
Former Staff
This is the same direction Cyrix took just before they stopped making CPU's. And their planned Sammual chip was sold to VIA. I wonder what company will be making the Hammer?

<font color=blue>There are no stupid questions, only stupid people doling out faulty information based upon rumors, myths, and poor logic!</font color=blue>
 
I am just afraid that this is the last straw, AMD goes down, and there is no serious competition for Intel anymore......


I REALLY do like the price wars.....


<font color=blue> Ok, I took the HSF off, powered up, but it STILL won't do anything....what now? </font color=blue>
 

Flipphone

Distinguished
Jan 5, 2003
139
0
18,680
hmmmm seems like intel only fans invited here 1st didnt chip just come out seems like u jumping on band wagons. 2nd never buy a chip when its first unleashed every newb knowsthat so ignore price.3rd amd knows what thier doing have u ever expected a new deign chip from amd to overclock alot? no not at first. 4th barton is not some miracle chip its just an xp running at 333 mhz with more cach so for most of we will run asusa7n8x ddr400 and new geforce fx. 5th test that but dont be so ingnorant so soon.
 

Crashman

Polypheme
Former Staff
Cool, I'm going to have to get a 2800+ to replace my T-Bird 1400, because I encode video, and we know that video encoding is directly scaled to processor performance, so I should EASILY be able to get a video done in EXACTLY HALF the time, after all, AMD claimed the XP rating system is based on the actual MHz performance of the T-bird.

<font color=blue>There are no stupid questions, only stupid people doling out faulty information based upon rumors, myths, and poor logic!</font color=blue>
 

slvr_phoenix

Splendid
Dec 31, 2007
6,223
1
25,780
This is the same direction Cyrix took just before they stopped making CPU's. And their planned Sammual chip was sold to VIA. I wonder what company will be making the Hammer?
Crashman, you make a good point. Luckily this time around AMD has at somewhat better market share and they have Hammers waiting in the wings. So perhaps it'll all be enough to counteract AMD's downward spiral into Cyrix-dom. It is however a rather scary similarity.


PC Repair-Vol 1:Getting To Know Your PC.
PC Repair-Vol 2:Troubleshooting Your PC.
PC Repair-Vol 3:Having Trouble Troubleshooting Your PC?
PC Repair-Vol 4:Having Trouble Shooting Your PC?
 

slvr_phoenix

Splendid
Dec 31, 2007
6,223
1
25,780
IBM and AMD maybe? :smile:
You know, that brings up a very interesting line of thinking. AMD's demise (or at least their CPU devision's demise) might not really be as bad as everyone makes it out to be. Suppose that someone like IBM bought out AMD's CPU resources. They would probably produce the same quality CPUs at the same FAB.

I mean AMD's ethics <i>are</i> dropping down into the 'questionable' category. (At least in my book.) So suppose that some company, like IBM, were to buy out AMD's processor devision. IBM is a much more stable company, with considerable resources. I've never really heard anyone question IBM's ethics. And IBM is certainly no slouch when it comes both computers and marketting.

We've all feared AMD's demise because it would mean going back to an Intel-only existence. Perhaps though, if a company like IBM bought out AMD, we would find that IBM then could really stick it to Intel far better than AMD did.

Granted, it's a lot of conjecture and hope, but I don't think it's all that crazy of an expectation should AMD become the next Cyrix. It's food for thought anyway.

Now if we only could get someone like Corsair or Samsung to buy out the Rambus patents on RDRAM and Yellowstone...


PC Repair-Vol 1:Getting To Know Your PC.
PC Repair-Vol 2:Troubleshooting Your PC.
PC Repair-Vol 3:Having Trouble Troubleshooting Your PC?
PC Repair-Vol 4:Having Trouble Shooting Your PC?
 

slvr_phoenix

Splendid
Dec 31, 2007
6,223
1
25,780
hmmmm seems like intel only fans invited here
Not at all. As it is, my next system is more than likely going to be an AMD system. (Specifically nForce2 mobo with an OCed AXP1700+.) Just because we can see that AMD is making mistakes doesn't mean we're Intel fans. You seem to be jumping to some pretty unbased conclusions.

1st didnt chip just come out seems like u jumping on band wagons.
So people who have similar observations are suddenly just jumping on bandwagons? Okay, I'll go explain that one to the scientific community. I'm sure they'll agree.

2nd never buy a chip when its first unleashed every newb knowsthat so ignore price.
If you'd read more, you'd have seen that price is hardly our primary concern. The real concern is the over-inflation of the PR rating just from the additional cache. <i>That</i> in turn affects price.

3rd amd knows what thier doing
Of course. That explains why they're constantly losing money. How silly of me. They know <i>exactly</i> what they're doing and they're running themselves into the ground on purpose. Thank you for clearing that up.

have u ever expected a new deign chip from amd to overclock alot?
1) Yes.
2) This is hardly a new design. It's actually the third revision to the Thoroughbred core. The first being of course the unhappy TBredA which was a combination of a process shrink and adjusting the layout of the chip in anticipation of adding an additional 256KB of L2 cache down the road. The second being the extra layer to fix the noise problems with the TBredA, called the TBredB. And now the third, which was simply adding that long-planned extra 256KB cache to the TBredB, now called the Barton.

4th barton is not some miracle chip its just an xp running at 333 mhz with more cach
1) The TBredB was <i>already</i> at a 2x166MHz FSB.
2) So you even admit that you knew it wasn't a new chip design.

so for most of we will run asusa7n8x ddr400 and new geforce fx.
1) What does this have to do with anything?
2) I highly doubt you're even right. No one seems to want to touch the GeForce FX, and a lot of people are disgruntled with Asus over the A7N8X. At least you got the PC3200 right.

5th test that but dont be so ingnorant so soon.
Could you please rephrase that into something comprehendible?


PC Repair-Vol 1:Getting To Know Your PC.
PC Repair-Vol 2:Troubleshooting Your PC.
PC Repair-Vol 3:Having Trouble Troubleshooting Your PC?
PC Repair-Vol 4:Having Trouble Shooting Your PC?
 

lhgpoobaa

Illustrious
Dec 31, 2007
14,462
1
40,780
Well to be honest i really dont care what happens, so long as we NEVER go back to the days of just intel, only intel, lets all pay thousands of bucks for a cpu intel, all hail great intel.

Thats bad for innovation, bad for consumers.

<b>Anyone claiming they can see the difference
between 450 and 500 FPS in Quake3 deserves to
be severely beaten with a rock. :smile: </b>
 

juin

Distinguished
May 19, 2001
3,323
0
20,780
Thats bad for innovation, bad for consumers.


I not so sure for the inovation as windows still evolve very fast.

Just next to the lab and the bunker you will find the marketing departement.
 

eden

Champion
Demand from market causes it. Especially for the Windows server family, which at some point needed more than Windows 2000's already robust abilities.
As for the SOHO Windows market, it's because of OEMs mainly, and of course the thirst for money! (go figure)

Innovation-wise, yeah Intel is quite motivated to go alone if they have to, but they'll likely slow down. 0.09m technology, or SSE-3 would not come out so early without AMD, for sure.

--
This post is brought to you by Eden, on a Via Eden, in the garden of Eden. :smile:
 

juin

Distinguished
May 19, 2001
3,323
0
20,780
Intel Itanium project totaly unrelate to AMD or any others

Intel manufacturing process ir totaly unrelate to AMD as intel have allwayse try to meet or beat Moore law.

The only thing this have made is hyper boosted P4 Nw that run over there planned specification force a P3 release at 1.13 GHZ a early P4 release.The only difference is they push there limit higher cutting in the validation time cutting in the testing time.

SIS even now offer A3 stepping to improve there time to market.Early AGP 8X implentation i like to remind all the unstability of this platfrom.

You should now that monopole usely have better inovation as you live in Canada does Bell got a better network that in the US yes does Videotron were slow to introduce LAN conection no.

Just next to the lab and the bunker you will find the marketing departement.
 

dhlucke

Polypheme
I not so sure for the inovation as windows still evolve very fast.
Yeah, but it didn't work. Windows sucked before 2000/XP.

<font color=red>
<A HREF="http://kevan.org/brain.cgi?dhlucke" target="_new">If you were to have sex with your clone would that be considered incest or masturbation?</A></font color=red>
 

eden

Champion
Yeah you have a point about Bell and Vidétron.
Wish Bell would compete now, they have almost ceased any competition and increased all unfairness in the DSL service. 5GB limits, no new speeds, more expensive unless you got the rare contract deal. I wanted to switch to Vidétron lately as they increased from 2.8 to 3.1Mbps standard cable speed, and 10GB DL, as well as offer free modem, so you pay 39.99$ like Bell's DSL that I have now, though I am on the contract deal, so I'd pay a 100$ fine if I cancelled :frown: . Wish they would compete already.

Wasn't Itanium's main goal to get Intel in the enterprise/corporate server sector? If not for super computing or heavy graphical editing?

--
This post is brought to you by Eden, on a Via Eden, in the garden of Eden. :smile: <P ID="edit"><FONT SIZE=-1><EM>Edited by Eden on 02/13/03 04:06 PM.</EM></FONT></P>
 

Copenhagen

Distinguished
Oct 21, 2001
552
0
18,980
Slvr, I think the answer is plain and simple.

By chosing such a low frequency for Barton 3000+ (2.167MHz), AMD is leaving enough headroom in the Thoroughbred core, to be able to come up with a response when Intel releases P4 3.2GHz and maybe P4 3.4GHz.

They don't care whether the XP3000+ rating truely corresponds to the P4 3.06GHz. They assume that people by now has adopted the PR rating, and as long as they are able to best the P4 3.06 in at least one benchmark, they can claim some sort of equality. It's far more important to be able to come up with some sort of response, i.e releasing a new speed grade, when Intel makes the next move. At least this gives the impression that AMD is fighting back.

<i>/Copenhagen - Clockspeed will make the difference... in the end</i> :cool:
 

juin

Distinguished
May 19, 2001
3,323
0
20,780
Wasn't Itanium's main goal to get Intel in the enterprise/corporate server sector? If not for super computing or heavy graphical editing?

More becoming the GOD of the CPU.
Sole supplier
Best performance by far
Biggest volume
All market
All the profit

I am just again the 1 ligne as only transemeta have experience with Vliw.



Just next to the lab and the bunker you will find the marketing departement.