TLB bug still present in B3 rev Phenoms

This is yesterday news but as i saw it nowhere. This is in French and it says they got the story from the horse's mouth (AMD reps).

http://www.erenumerique.fr/ces_2008_...ews-11723.html

Quote:

---

On savait déjà que les versions cadencées à 2.4 et 2.6 GHz des Phenom étaient repoussées au second trimestre. Nous avons pu confirmer hier que les 9650 et 9550, version « débugguées » des premiers Phenom (voir notre article) sont également repoussées au second trimestre. AMD a en effet reçu ses premières puces utilisant le stepping B3 (early silicon), le bug n’est toujours pas corrigé. Un nouveau « spin » est donc nécessaire, ce qui peut repousser les Phenom au mois de mai.

---

Bolded text translation:

AMD got its first chips using the B3 rev, the bug is not addressed still, a new spin is needed which could postponed Phenoms till May.

OUCH... another blow... now they can just hope it aint true.

the TLB-Bug doesn't seem to occur very often. so what's the problem for us if B3 clocks better than B2?

let's just hope it's not true, though

AMD below 2$ anyone?

:owned: :(

not good news from amd.... again...
let's hope that this b3 can get a lot more clock, but i doubt it

I just heard the same thing from elsewhere and if really true, AMD is in deep trouble...

IF the bug is still present, then AMD should just ditch the L3 and increase the L2 cache sizes in the next revision. If it's that easy of course.

AMD is officially dead now, only low price might save them.

Well , this certainly makes you have a great deal of trust in their engineers. :rolleyes:

The B3 stepping supposed to fix B2 suffers from the same problem.Did they bothered to do a proper debug or some are sabotaging the company ?I'm pretty sure most EE aren't happy in Austin.

Btw , you get the beauty of this : Tier 1 OEMs won't touch the chip until this bug is fixed.This will do wonders for AMD's MS and credibility among Tier 1 vendors and enterprise customers.

Stick a fork in AMD, they're done.

B3 still having TLB bugs, and 45nm in 2009.

And yet their BOD still keeps Hector as CEO... incredible.

with this in mind what would NVIDIA need to buy AMD? i mean there shouldnt be VGA monopoly concerns right?

What does nvidia need to keep the x86 license going?

Quote:

---

Originally Posted by metro.cl

with this in mind what would NVIDIA need to buy AMD? i mean there shouldnt be VGA monopoly concerns right?

What does nvidia need to keep the x86 license going?

---

amd can be had for pocket change now... if anyone was serious about buying them i think it would have happened by now.

Quote:

---

Originally Posted by metro.cl

with this in mind what would NVIDIA need to buy AMD? i mean there shouldnt be VGA monopoly concerns right?

What does nvidia need to keep the x86 license going?

---

Why would they want to buy AMD? The CPU part is nothing but a hole to throw money into forever and obviously they would have to sell off the ATI part, and anyone that would buy ATI could only be a stronger and better funded company than AMD making Nvdia's life on the GPU front only more difficult.

Besides, if AMD dies and takes ATI with it then any attempt to move into the GPU business by Intel would be very likely seen as an attempt to leverage their CPU monopoly to generate another monopoly, something that would generate huge regulatory headaches. Nvidia would be better insulated from competition on the GPU front if AMD dies. Again, why would Nvidia want to buy AMD?

Anyway, AMD's x86 licence has a clause in it making it non-transferrable. There are some questions as to whether such clauses are enforceable or not, but at a minimum it would be a huge legal battle with an uncertain outcome. Once again, buying AMD makes no sense for Nvidia.

Quote:

---

Originally Posted by Periander6

Why would they want to buy AMD? The CPU part is nothing but a hole to throw money into forever and obviously they would have to sell off the ATI part, and anyone that would buy ATI could only be a stronger and better funded company than AMD making Nvdia's life on the GPU front only more difficult.

Besides, if AMD dies and takes ATI with it then any attempt to move into the GPU business by Intel would be very likely seen as an attempt to leverage their CPU monopoly to generate another monopoly, something that would generate huge regulatory headaches. Nvidia would be better insulated from competition on the GPU front if AMD dies. Again, why would Nvidia want to buy AMD?

Anyway, AMD's x86 licence has a clause in it making it non-transferrable. There are some questions as to whether such clauses are enforceable or not, but at a minimum it would be a huge legal battle with an uncertain outcome. Once again, buying AMD makes no sense for Nvidia.

---

For the x86 license, it is the cheapest way to get one and also to elminate competition on th GPU front.

NVIDIA is still evaluating to get a CPU working so this might be the best chance they can get. AMD is freaking cheap now.

Quote:

---

Originally Posted by naokaji

amd can be had for pocket change now... if anyone was serious about buying them i think it would have happened by now.

---

Anyone want to throw some money together and buy AMD?

Quote:

---

Originally Posted by metro.cl

For the x86 license, it is the cheapest way to get one and also to elminate competition on th GPU front.

NVIDIA is still evaluating to get a CPU working so this might be the best chance they can get. AMD is freaking cheap now.

---

Who's gonna pay those debts ?

Quote:

---

Originally Posted by metro.cl

For the x86 license, it is the cheapest way to get one and also to elminate competition on th GPU front.

---

No, the cheapest way to get an x86 licence by far would be to buy Via, not AMD. Remember that anyone that buys AMD has to take on their 5+ Billion in debt on top of buying up the stock.

There is no way in hell, zero, that Nvidia would receive anti-trust clearance to purchase ATI. To buy AMD they would have to sell off the ATI division to have any chance whatsoever for approval, and anyone other than AMD owning ATI would make ATI stronger.

The idea of Nvida buying AMD is a fantasy that won't happen, not before AMD goes into C11 at the very least.

I dont know if people dont understand it yet. Or maybe do not wish to understand it. Its been said 1000s of times.

Let me bend it in neon this time. You will NOT, NEVER, EVER get an x86 license by buying VIA or AMD. It will be LOST and GONE. No nobody can change that.

So in short, buying VIA would give you...nothing. Buying AMD would give you a struggling GPU business.

The CPU parts of VIA and AMD would be trash and gone!

nVidia buying AMD would be like buying sand in Sahara. Pretty darn useless and dumb.

Quote:

---

Originally Posted by Shintai

I dont know if people dont understand it yet. Or maybe do not wish to understand it. Its been said 1000s of times.

Let me bend it in neon this time. You will NOT, NEVER, EVER get an x86 license by buying VIA or AMD. It will be LOST and GONE. No nobody can change that.

So in short, buying VIA would give you...nothing. Buying AMD would give you a struggling GPU business.

The CPU parts of VIA and AMD would be trash and gone!

nVidia buying AMD would be like buying sand in Sahara. Pretty darn useless and dumb.

---

You can buy AMD and keep the brand as "independent" company and have the license

Quote:

---

Originally Posted by Shintai

I dont know if people dont understand it yet. Or maybe do not wish to understand it. Its been said 1000s of times.

Let me bend it in neon this time. You will NOT, NEVER, EVER get an x86 license by buying VIA or AMD. It will be LOST and GONE. No nobody can change that.

So in short, buying VIA would give you...nothing. Buying AMD would give you a struggling GPU business.

The CPU parts of VIA and AMD would be trash and gone!

nVidia buying AMD would be like buying sand in Sahara. Pretty darn useless and dumb.

---

How many times do you need to be told that the U.S. government, not to mention the EU is going to allow Intel to have a monopoly on the X86 license? It's just not going to happen. They'll be forced to renew the agreement with a probable suitor or come to some type of financial agreement.

We need AMD or Nehalem will cost 2k$ per CPU until late 2009..

Really bad news for AMD, that's a shame...Looks like that bug is a really tough one to kill :D
Looks like the K8 will still remain quite much alive...at least until that error is fixed...

Quote:

---

Originally Posted by Shintai

I dont know if people dont understand it yet. Or maybe do not wish to understand it. Its been said 1000s of times.

Let me bend it in neon this time. You will NOT, NEVER, EVER get an x86 license by buying VIA or AMD. It will be LOST and GONE. No nobody can change that.

So in short, buying VIA would give you...nothing. Buying AMD would give you a struggling GPU business.

The CPU parts of VIA and AMD would be trash and gone!

nVidia buying AMD would be like buying sand in Sahara. Pretty darn useless and dumb.

---

not so sure if you checked GPU lately but they are doing just fine....

670 is going well
635 is on track to reduce costs again
lots of mobile designs won for 2008, not like 2007 where all designs went to nvidia
and not the least: they are now the best price/performance you can buy at every segment in the workstation market wit the firegl series


for the rest still a rumor....lets see when they are there end of february

Quote:

---

Originally Posted by R101

We need AMD or Nehalem will cost 2k$ per CPU until late 2009..

---

Why do people keep saying crap like this? :confused:

Quote:

---

Originally Posted by Faktor

Why do people keep saying crap like this? :confused:

---

If AMD goes under, they can charge whatever they want for their CPUs. Not saying they'd charge 2K for a top of the line CPU, but $1500 wouldn't be out of the question. They charge $1000 now for top of the line as they have no immediate competition from AMD, can you imagine what they'd charge if there was no AMD, period?

Quote:

---

Originally Posted by Faktor

Why do people keep saying crap like this? :confused:

---

Because of history, without competition , intel can charge (and has) whatever they want, for very little performance improvement.

Competition is what makes computers better and better, without competition innovation might not be invested it.

But on a serious note, I don't think the consumers would allow themselves to be taken advantage of this way again, if low end PC is costing $400 now, no one will buy a low end for $1000 in 2 years.

The prices of high end cpus, I don't think would go up, but low end cpus will.

It is the engineers playing stock games. They bought when the company was new. They made the 939, then sold their stock, then made K10. Now they are going to buy buy buy. Now they will fix the chip and make tons of money. lol

Quote:

---

Originally Posted by 2long4u

It is the engineers playing stock games. They bought when the company was new. They made the 939, then sold their stock, then made K10. Now they are going to buy buy buy. Now they will fix the chip and make tons of money. lol

---

Thats the coolest "conspiracy theory" Iv heard for ages :D

TBH.....Its kinda tempting to join in!

While this is never good, I guess my question is how much it effects sales. What are the level of Opteron sales currently, and how much does the bug effect them really? Sure, they would likely get business from new processors, but is the level of Opteron sales enough to sustain them in the meantime? I'm not sure I'm ready to call AMD done yet. They have issues, but this isn't the first time they've had to struggle either...

I can't really understand why are ppl give so much fus about this TLB errata thing. It is very uncommon to happen and as some resources mentioned can occur only in virtualization environment. Correct me if I'm wrong, but isn't virtualization only used to make many working places from fewer machines? And what tasks would be performed there? Office applications most probably. Now does any one gives a damn if word can accept 10k symbols per second or only 8k if patch is applied?

Though I still think that first respin of B3 was not concerned about TLB errata fix and there would be a proper one.

As far as I am concerned, I didn't know that the errata doesn't occur always. But hey, I'd rather choose a CPU that has no bugs, at least we don't know if it (or them) has bugs. For example the K8 Athlons and current Intel's lineup. Why would I spend my money on merchandise with an factory error? Sure, I don't know will my CPU have that error, or if it has it, I don't know will I'll ever have problems.
But, that leaves a pretty bitter taste if you know what I mean and makes people avoid it ;)
I don't know, maybe the sale results would be better if the current Phenom line up is cheaper. The 9500 has almost the same price as the Q6600, and the Q6600 overclocks better and is bug free. So people pick the Q6600 (or sth. else) instead...

Quote:

---

Originally Posted by Swarm dude

But hey, I'd rather choose a CPU that has no bugs, at least we don't know if it (or them) has bugs. For example the K8 Athlons and current Intel's lineup.

---

There are NO CPUs w/o erratas. Since 8086

Quote:

---

Originally Posted by Cooper

I can't really understand why are ppl give so much fus about this TLB errata thing. It is very uncommon to happen and as some resources mentioned can occur only in virtualization environment. Correct me if I'm wrong, but isn't virtualization only used to make many working places from fewer machines? And what tasks would be performed there? Office applications most probably. Now does any one gives a damn if word can accept 10k symbols per second or only 8k if patch is applied?

Though I still think that first respin of B3 was not concerned about TLB errata fix and there would be a proper one.

---

Maybe people fuss about it (from an investment point of view) because no Tier 1 OEM will ship K10 servers until AMD fixes it. They are selling the bugged Phenoms, but that is damaging the reputation of everyone involved, fair or not.

B3 was supposed to fix the TLB bug. It didn't. Say hello to Q3 for first OEM server K10 shipments. If AMD hasn't gone bankrupt first.

To answer someone else's question, no, the K8 DC Opterons do not cut it in the server space any longer. Intel has had QC parts out for over a year now. I'm guessing AMD's server share will fall to 5-7% by Q208, and that will just crush their ASPs and margins, and explode their losses.

Quote:

---

Originally Posted by Cooper

There are NO CPUs w/o erratas. Since 8086

---

But there are plenty of CPUs w/o errata so serious that OEMs will not ship them.

The "every CPU has errata" argument is silly. Of course they do. The question is, how bad is any given erratum, and the answer is determined by looking at OEM (read: CPU customer) behavior.

BTW, it's "erratum" (singular), "errata" (plural)

Quote:

---

Originally Posted by terrace215

B3 was supposed to fix the TLB bug. It didn't.

---

You say that like B3 already is in local PCWorld. This news is just a rumour until B3 will be actually released.

Quote:

---

Originally Posted by Cooper

You say that like B3 already is in local PCWorld. This news is just a rumour until B3 will be actually released.

---

Like you said its just a rumor as of now but we can all agree that this isn't the kind of news and/or rumors we want to hearing from AMD at this point in time. Let's hope the GPU development goes a lot smoother.

Quote:

---

Originally Posted by Cooper

You say that like B3 already is in local PCWorld. This news is just a rumour until B3 will be actually released.

---

With something this dire for the company, AMD would be denying it to the press if it were false. And note that the source says they got the information FROM AMD employees at CES:

google trans:

Here, in bulk, some information that we manage to steal employees from AMD.

We were able to confirm yesterday that the 9650 and 9550, "débugguées" version of the first Phenom (see our article) are also repelled the second quarter. AAMD has indeed received its first chips using the stepping B3 (early silicon), the bug is still not fixed. A new "spin" is needed, which could push the Phenom in May.

The manufacturer had published a roadmap in which the next generation of graphics cards were listed by the manufacturer in 2009 in the evolution of Spider platforms. Even under torture, our interlocutor refuses to confess if the R700 did emerge before the end of 2008.

Side integrated graphics, 700 series chipset AMD should see its life extended until 2009 pending the arrival of processors "Fusion" that incorporate graphics and heart-core processors traditional.

The manufacturing process of 45 nm AMD (who will not use high-k technology unlike that of Intel) is still scheduled for the second half of the year. Initial results were "encouraging". Hopefully ...

Cooper, There are two types of computer enthusiasts.

Theres guys like me and you who I like to call "Christmas" users.
We simply wait patiently for cool new things to come out, and its like christmas for us YAY.

Then theres guys like those that posted those comments in this thread that I call "Horse race" users.
They sit, eyes glued open, watching every moment, every move, ever scrutinizing, and betting. They bet early and call their bet outloud and wait

These are the two types of enthusiasts.

Quote:

---

Originally Posted by gdogg

Because of history, without competition , intel can charge (and has) whatever they want, for very little performance improvement.

Competition is what makes computers better and better, without competition innovation might not be invested it.

But on a serious note, I don't think the consumers would allow themselves to be taken advantage of this way again, if low end PC is costing $400 now, no one will buy a low end for $1000 in 2 years.

The prices of high end cpus, I don't think would go up, but low end cpus will.

---

Because of what history? When in history has Intel charged $1000 for a low end cpu and $2000 for mainstream? I don't recall that happening; even in the early pentium pro days (the time when intels prices were highest).

In the pre-k8 days, Intel's prices weren't as high as people think they remember them to be. In the pentium 1 through pentium 3 days Intel always priced thier processors the same. $200-300 for the low-end and $800 or $900 for the high-end.

Even if Intel was a true monopoly, they could not sell cpu's at very high prices because not enough people would buy them to cover the cost to make the cpu's. The market is what determines prices. If AMD were to die right now I doubt Intel would raise their prices much at all. The problem we would face with Intel being a monopoly is lack of inovation not pricing.

That is why when someone says that Intel would price thier low-end processors at $1000 if they were a monopoly or when someone says that nehalem cpu's would cost $2000 if they were a monopoly, they are either stupid fanboys or are just plain ignorant. (too many people)

Quote:

---

Originally Posted by terrace215

With something this dire for the company, AMD would be denying it to the press if it were false. And note that the source says they got the information FROM AMD employees at CES:

google trans:

Here, in bulk, some information that we manage to steal employees from AMD.

We were able to confirm yesterday that the 9650 and 9550, "débugguées" version of the first Phenom (see our article) are also repelled the second quarter. AAMD has indeed received its first chips using the stepping B3 (early silicon), the bug is still not fixed. A new "spin" is needed, which could push the Phenom in May.

R700 is alright, thank you. That's what the AMD's graphic department's reps told us. The manufacturer had published a roadmap in which the next generation of graphics cards were listed by the manufacturer in 2009 in the evolution of Spider platforms. Even under torture, our interlocutor refuses to confess if the R700 did emerge before the end of 2008.

Side integrated graphics, 700 series chipset AMD should see its life extended until 2009 pending the arrival of processors "Fusion" that incorporate graphics and heart-core processors traditional.

The manufacturing process of 45 nm AMD (who will not use high-k technology unlike that of Intel) is still scheduled for the second half of the year. Initial results were "encouraging". Hopefully ...

---

I inserted and bolded the sentence that strangely got lost in your translation :rolleyes:

Quote:

---

Originally Posted by Faktor

Why do people keep saying crap like this? :confused:

---

If you payed for Pentium 90 the ammount of $$ I did back in the day, you would know. When there's no competition, good hardware is expensive. Intel would be stupid not to raise prices if it could.

I think people forgot how much a 286 costed. Or a 50MB HD back then.

Back then your value end entry PC costed 4000-5000$ and not 400-500$ or less. And back then money was more worth than they are today. So with todays money it would easily have costed 10000$.

Or perhaps what some of the early TVs costed.

People should elarn history, and not just what they wish. Price is lowering, why? Competition? Only a tiny part. But the main part is VOLUME.

With no competition Intels prices would not be much different than today. maybe 10-15% higher. And they would still innovate is almost the same pace.

If not, other technologies would catch up and make the PC into the gadget segment. And without big volume they would earn as much. It can sound hard for some people, but Intel earns more by selling CPUs at 200$ than they do at 2000$.

Another reason why CPU were that expensive back then was because they were much more expensive to produce! A 65nm die in a 300mm wafer isn't that expensive, but a Pentium I with a 800nm proces technology just was much more expensive to make. You had alot less CPU's in a wafer then they do now.
Also about $1000 Extreme Editions, even the P4EE costed $1000 and those got crushed by AMD's FX cpu's, still intel managed to sell those. I don't think intel will raise their prices that much IF AMD dies. We can now buy complete computer for less then $500. If intel would increase CPU prices again and computer would cost atleast $1000 or more, then computer sales would take a huge dive!
Innovation on the other hand will go much slower as there is no compition.
That beeing said, I do hope AMD stays arround, but this sure isn't good news...

I really wonder why this bug gets this huge amount of attention?
Current 45nm Intel CPUs suffer several bugs (which can't be fixed with microcode update) which may be even worse than this as you can read in the recent specification updates.
For example, Harpertown and Yorkfield has an equally serious TLB bug, some mixed load commands couse system locks, FSB anomalies with 4 layer mainboard PCBs, and so on...

This whole "Phenom TLB bug = the sky is falling" crap is simply hype spewed up by people who don't know what they are talking about.

Quote:

---

Originally Posted by savantu

Well , this certainly makes you have a great deal of trust in their engineers. :rolleyes:

The B3 stepping supposed to fix B2 suffers from the same problem.Did they bothered to do a proper debug or some are sabotaging the company ?I'm pretty sure most EE aren't happy in Austin.

Btw , you get the beauty of this : Tier 1 OEMs won't touch the chip until this bug is fixed.This will do wonders for AMD's MS and credibility among Tier 1 vendors and enterprise customers.

---

I'm amazed how passionate are some people on this forum with every fart concerning the "news" about AMD's fall from glory!

Suddenly some unknown French site about multimedia is reliable source for ol' Intel boyz at XS! Pfff

BTW

http://www.x86watch.com/news/hp-phenom-desktop-200.html

Quote:

---

Originally Posted by freeloader

How many times do you need to be told that the U.S. government, not to mention the EU is going to allow Intel to have a monopoly on the X86 license? It's just not going to happen. They'll be forced to renew the agreement with a probable suitor or come to some type of financial agreement.

---

if Intel had a monopoly on computers, then yea governments wouldn't allow it. but they'd only have a monopoly on x86. it'd be a long expensive legal process to decide whether that was enough of a monopoly to justify government intervention (eg splitting Intel up if AMD disappeared, or overruling the x86 contract so that if a company bought AMD they would get to keep the license). i don't think you can assume it would happen.

consider microsoft, for example. a near total monopoly on windows (like Intel would be a near total monopoly on x86 without AMD), but the government didn't have a good reason to break them up because there are other operating systems.

expecting the government to intervene if Intel was going to become a monopolist in x86 is just a bit too much like expecting the government to break up coca cola if pepsi went out of business (and of course the government wouldn't) for it to be a sure thing.

Quote:

---

Originally Posted by freeloader

IF the bug is still present, then AMD should just ditch the L3 and increase the L2 cache sizes in the next revision. If it's that easy of course.

---

I really doubt that this situation is as bad as "we" make it seem. There are countless companies that go thru the same thing, and many that still operate after chapter 11. Kmart is still around... There will always be some sucker investor to lend a hand.

Quote:

---

Originally Posted by LowRun

I inserted and bolded the sentence that strangely got lost in your translation :rolleyes:

---

Sorry, it was unintentional. When you "copy" text from the google translation, it leaves the French sentences (which do not show on the page) spliced into the translation. So I had to hand-cut them out.

I wouldnt say AMD is no more not yet, give it till the middle of the year and see what happens, they still have GPU sales which are doing pretty well and on a OEM level for mobile GPU (where i work) they are doing really well. But it is going to hurt them if it is true. Lets hope they get it sorted soon as i want to go back to amd for a while :D

I say all of XS pitches in and buys AMD when they hit $.02 a share

Quote:

---

Originally Posted by Shintai

I think people forgot how much a 286 costed. Or a 50MB HD back then.

Back then your value end entry PC costed 4000-5000$ and not 400-500$ or less. And back then money was more worth than they are today. So with todays money it would easily have costed 10000$.

Or perhaps what some of the early TVs costed.

People should elarn history, and not just what they wish. Price is lowering, why? Competition? Only a tiny part. But the main part is VOLUME.

With no competition Intels prices would not be much different than today. maybe 10-15% higher. And they would still innovate is almost the same pace.

If not, other technologies would catch up and make the PC into the gadget segment. And without big volume they would earn as much. It can sound hard for some people, but Intel earns more by selling CPUs at 200$ than they do at 2000$.

---

true.
intel's biggest competitor right now is themselves. They still have to price new chips so that they can compete with the old ones.
though chances are margins will still go way up...

Not good news for AMD.

BTW, is AMD doing a K10 mobile chip?:confused:

Quote:

---

Originally Posted by happychappy

BTW, is AMD doing a K10 mobile chip?:confused:

---

no roadmap showed any k10 mobile parts for the near future....

Quote:

---

Originally Posted by happychappy

Not good news for AMD.

BTW, is AMD doing a K10 mobile chip?:confused:

---

They will have a K10 mobile processor, but that is pending 45 nm and much lower power consumption.

Quote:

---

Originally Posted by Shintai

I think people forgot how much a 286 costed. Or a 50MB HD back then.

Back then your value end entry PC costed 4000-5000$ and not 400-500$ or less. And back then money was more worth than they are today. So with todays money it would easily have costed 10000$.

Or perhaps what some of the early TVs costed.

People should elarn history, and not just what they wish. Price is lowering, why? Competition? Only a tiny part. But the main part is VOLUME.

With no competition Intels prices would not be much different than today. maybe 10-15% higher. And they would still innovate is almost the same pace.

If not, other technologies would catch up and make the PC into the gadget segment. And without big volume they would earn as much. It can sound hard for some people, but Intel earns more by selling CPUs at 200$ than they do at 2000$.

---

I remember my family bought a Pentium 1 when it first came out. Had 64MB of RAM.

It was $3,500 O_o

the tlb problem only becomes a problem on servers, there is little effect on desktops, there is a guy here figuring out how to make the tlb bug come to the surface, only way so far is 4 virtual programs running with cool and quiet on.
see if i can find a link.

edit: http://www.xtremesystems.org/forums/...d.php?t=172026

The TLB bug also effects OCing...

Quote:

---

Originally Posted by knightwolf654

the tlb problem only becomes a problem on servers, there is little effect on desktops, there is a guy here figuring out how to make the tlb bug come to the surface, only way so far is 4 virtual programs running with cool and quiet on.
see if i can find a link.

edit: http://www.xtremesystems.org/forums/...d.php?t=172026

---

This is true, the TLB bug is not a big deal for desktop users... a person is more likely to curse out Microsoft before they would suspect a TLB bug, BSODs and random OS lockups probably occur at probabilities 10x or 100x more often than this bug would...

However, server is where the margins are, AMD really needs to execute well there...

But honestly, this is really hearsay and rumor, so while fun to banter in a debate in back alley forums, it is a pretty mean rumor that would best be left alone .... if end of Q1 comes and volume shipments are still tardy... then we can point back and say... see... Barcelona B3 did not fix the bug... until then a pragmatic perspective is in order...

i.e. this is just a rumor, we cannot be certain.

TLB...TLB...TLB....:shrug:

I've had two Phenom 9500's since Nov.26
One crunching at stock speeds full load-> NO TLB CRASHES

The other one been overclocked
been running at 2772Mhz since Dec.27
TLB fix disabled
running games,benchies and crunching-> NO TLB CRASHES

If I had TLB crashing I don't my Phenom would have made it to 32nd spot on Riesel Sieve Boinc for daily RAC in less then a week
Computer ID 33998 owner Frank Encruncher
http://boinc.rieselsieve.com/orig/to...=expavg_credit

Does it matter if there is TLB bug on Phenom... Answer NO

Quote:

---

Originally Posted by terrace215

"...
The manufacturing process of 45 nm AMD (who will not use high-k technology unlike that of Intel) is still scheduled for the second half of the year. Initial results were "encouraging". Hopefully ...

---

This is strange, from what I've read, AMD will be using a high-k technology for 45nm.

Quote:

---

Originally Posted by Zytek_Fan

The TLB bug also effects OCing...

---

FUD also

Quote:

---

Originally Posted by ryboto

This is strange, from what I've read, AMD will be using a high-k technology for 45nm.

---

They were making fun of the hafnium based high-k process on their YouTube channel :rofl:

Quote:

---

Originally Posted by ryboto

This is strange, from what I've read, AMD will be using a high-k technology for 45nm.

---

If I recall, they will, but not initially. IBM and AMD had been working on a high-k solution I thought, but it wasn't able to be implemented at the outset.

Quote:

---

Originally Posted by ryboto

This is strange, from what I've read, AMD will be using a high-k technology for 45nm.

---

Nope will not happen.

They will retain the option, but they will not use it to begin with... this has also been stated in a few AMD analysts and techical day meetings:

Quote:

---

While ramping 45-nm production early next year, AMD said that it will retain the option to include high-k/ metal gates in that generation, and will include high-k/metal gate technology in its 32-nm production generation.

---

http://www.pcmag.com/article2/0,1759...049TX1K0100488

Frankly they are too far behind, even IBM is not using high-K at 45 but will do so at 32 apparently.

Quote:

---

Originally Posted by AMDDeathstar

FUD also

---

How do you konw ?? Perhaps the 2.8-3.0 Limitation for stable overclocks is hampered by the bug?? Without it, maybe 3.0 - 3.2 is doable.

Though I actually agre, I don't think the bug affects clocking at all, it is more desgin and, most importantly, process limited for getting to the 3.0 Ghz mark.

i hope AMD skips 45nm.

Quote:

---

Originally Posted by knightwolf654

i hope AMD skips 45nm.

---

They can't. You can't simply "skip" a manufacture process.

Quote:

---

Originally Posted by xlink

true.
intel's biggest competitor right now is themselves. They still have to price new chips so that they can compete with the old ones.
though chances are margins will still go way up...

---

True but back then there wern't as many people building their own computers, so the middleman got a higher %. There is almost no markup on pc's now, there can't be at $400 for a computer that includes a mouse, keyboard, monitor and printer.

Quote:

---

Originally Posted by AMDDeathstar

TLB...TLB...TLB....:shrug:

I've had two Phenom 9500's since Nov.26
One crunching at stock speeds full load-> NO TLB CRASHES

The other one been overclocked
been running at 2772Mhz since Dec.27
TLB fix disabled
running games,benchies and crunching-> NO TLB CRASHES

If I had TLB crashing I don't my Phenom would have made it to 32nd spot on Riesel Sieve Boinc for daily RAC in less then a week
Computer ID 33998 owner Frank Encruncher
http://boinc.rieselsieve.com/orig/to...=expavg_credit

Does it matter if there is TLB bug on Phenom... Answer NO

---

The bug is very critical for servers.
Here is the problem:
http://www.x86-64.org/pipermail/disc...er/010281.html
Here is the answer:
http://www.x86-64.org/pipermail/disc...er/010282.html

Quote:

---

Originally Posted by JumpingJack

Nope will not happen.

They will retain the option, but they will not use it to begin with... this has also been stated in a few AMD analysts and techical day meetings:


http://www.pcmag.com/article2/0,1759...049TX1K0100488

Frankly they are too far behind, even IBM is not using high-K at 45 but will do so at 32 apparently.

---

Huh..I remember the news about that, but then I remember two more news articles that mentioned AMD reps speaking about High-k 45nm. Eh, I don't see how they expect to produce a competing product on their current process.

Quote:

---

Originally Posted by ryboto

Huh..I remember the news about that, but then I remember two more news articles that mentioned AMD reps speaking about High-k 45nm. Eh, I don't see how they expect to produce a competing product on their current process.

---

Yeah, the report was that either in the 2nd revision of 45 nm or in 32 nm.... I can dig up their presentation to technical analysts if you would like to see it.....

But no where has AMD officially announced high-K for any process technology on the horizon, only that the retain the option, what ever that means. But I agree, they will not be able to compete without it. I could be wrong, AMD could be 'smoke screening' in their earlier presentations for competitive reasons... i.e. ' make the other guy think we aren't going there', then surprise them.

Quote:

---

Originally Posted by knightwolf654

i hope AMD skips 45nm.

---

That can't happen for a few reason, but specifically because the industry typically 'moves' together more or less, going straight to 32 nm would not be possible becuse the equipment vendors do not have the appropriate equipment tuned up to support it.

Quote:

---

Originally Posted by R101

We need AMD or Nehalem will cost 2k$ per CPU until late 2009..

---

Might I suggest the main reason for continous price reductions has been the the dramatic improvement in process technology coupled with simple economies of scale rather than competative pressure.

With roughly equal die areas for each of the 80386 and Penryn the latter will be cheaper due to vastly more viable dies being delivered from the larger wafer being used in it's manufacture.

Frankly it seems Intel is already operating as a monoploy and has been since Core 2's introduction. They are releasing new product to their own schedule and setting compelling pricing without the slightest threat from AMD. They can't arbitrarily set prices and let the market stagnate as this would both impact on their revenue and allow other competitors in. Would most people really need a PC if a net enable, keyboard equipped games console bundle with Linux and an applications suite were available?

I'm not denying competative pressure plays a role in pricing, but I think it's current importance is overstated.

http://img529.imageshack.us/img529/7792/amdav9.jpg

Quote:

---

Originally Posted by AMDDeathstar

TLB...TLB...TLB....:shrug:

I've had two Phenom 9500's since Nov.26
One crunching at stock speeds full load-> NO TLB CRASHES

The other one been overclocked
been running at 2772Mhz since Dec.27
TLB fix disabled
running games,benchies and crunching-> NO TLB CRASHES

If I had TLB crashing I don't my Phenom would have made it to 32nd spot on Riesel Sieve Boinc for daily RAC in less then a week
Computer ID 33998 owner Frank Encruncher
http://boinc.rieselsieve.com/orig/to...=expavg_credit

Does it matter if there is TLB bug on Phenom... Answer NO

---

How do you know that a crash or data corruption problem was caused by the TLB bug ? Do you have advanced debugging tools ?

Coooper, what in da world is wrong with my post, except for the fact that it wasn't pro-AMD?
HP selling stuff at Wal-mart deserves a ridicule!

Quote:

---

Originally Posted by _Lone_Wolf_

Frankly it seems Intel is already operating as a monoploy and has been since Core 2's introduction. They are releasing new product to their own schedule and setting compelling pricing without the slightest threat from AMD. They can't arbitrarily set prices and let the market stagnate as this would both impact on their revenue and allow other competitors in. Would most people really need a PC if a net enable, keyboard equipped games console bundle with Linux and an applications suite were available?

I'm not denying competative pressure plays a role in pricing, but I think it's current importance is overstated.

---

Just keep telling yourself that.

AMD was eating away Intel's market share for years in a row, and you think the agressive pricing, marketing and development plan of Intel is just a coincidence? The tick-tock model didnt just fall out of the sky around 2005.

We can already see them taking some pressure off as AMD keeps failing.

Intel will cut down to what they think is an optimum between cost and production, as soon as they feel safe. Even someone with a moderate economical education can udnerstand that. And looking at the past, it's safe to say that what they have done with Core 2 and Quad is far above that optimum - especially in the light of the agressive pricing.

Quote:

---

Originally Posted by alfaunits

Coooper, what in da world is wrong with my post, except for the fact that it wasn't pro-AMD?
HP selling stuff at Wal-mart deserves a ridicule!

---

maybe in yr mind. I find this to be trolling

Quote:

---

Originally Posted by kl0012

The bug is very critical for servers.
Here is the problem:
http://www.x86-64.org/pipermail/disc...er/010281.html
Here is the answer:
http://www.x86-64.org/pipermail/disc...er/010282.html

---

Thank you for those links, this is the first time i see the bug in action.
The guy who answerd is working for AMD according to his sig, so it seems to be no BS this time.

I could not make the bug occure with xen and differen guest osses all running prime95 or other benchmarks.
It would be interesting what software he ran.

Quote:

---

Originally Posted by justapost

Thank you for those links, this is the first time i see the bug in action.
The guy who answerd is working for AMD according to his sig, so it seems to be no BS this time.

I could not make the bug occure with xen and differen guest osses all running prime95 or other benchmarks.
It would be interesting what software he ran.

---

It is not as easy as such. The guess is its due to hot spots. Thats also one reason it could be safe at under 2.4Ghz. So Its abit like OC and chips. Random picks. Maybe only 5% of all chips have this bug at stock speed. And with an increased chance with higher frequency.

if it was a fixed bug so to say. I am pretty sure AMD would have caught it in their QA process.

We have a few people here with 0 bugs, and we have an example of a person in WCG forum where it crashes.

Quote:

---

Originally Posted by Shintai

It is not as easy as such. The guess is its due to hot spots. Thats also one reason it could be safe at under 2.4Ghz. So Its abit like OC and chips. Random picks. Maybe only 5% of all chips have this bug at stock speed. And with an increased chance with higher frequency.

---

There is a detailed description of the TLB-thing at LostCircuit here.

Quote:

---

Originally Posted by Shintai

if it was a fixed bug so to say. I am pretty sure AMD would have caught it in their QA process.

---

I dont want to say the bug does not exist, but i experienced it is pretty hard to trigger it.

Quote:

---

Originally Posted by Shintai

We have a few people here with 0 bugs, and we have an example of a person in WCG forum where it crashes.

---

Thx, gona try to trigger the bug with WCG. Do you remeber when the crash was reported in the WCG forum.
According to LostCircuit the bug can only occure with hardware enabled virtualisation. What did the person at WCG use?

Quote:

---

Originally Posted by justapost

There is a detailed description of the TLB-thing at LostCircuit here.

I dont want to say the bug does not exist, but i experienced it is pretty hard to trigger it.

Thx, gona try to trigger the bug with WCG. Do you remeber when the crash was reported in the WCG forum.
According to LostCircuit the bug can only occure with hardware enabled virtualisation. What did the person at WCG use?

---

He just ran WCG and it crashed a few times.

Also i think you misunderstood my post. Its not hard to trigger as such. It just depends on your luck of CPU. If there is a hot thermal spot or not and if so, at what frequencies. I do not believe it s a virtualization only bug. Its a pretty lame excuse. Also you would just have the BIOS setting where you disable or enable virtualization to enable or disable the fix aswell. I call it a smoke screen ;)

I don't think WCG can trigger this bug. I've been running Phenom 9600 at stock on WCG for 4 days non stop - not a single crash. I beleive there are few more ppl that run WCG or any other F@H app on Phenom even on higher clock speeds. this is just not that kind of load. Though if you running something elsevery CPU intensive...don't know. This guy at x86-64...there is no explanation of what exactly he was doing to get this error

Quote:

---

Originally Posted by LowRun

[TLB bug still present in B3 rev Phenoms]
http://www.erenumerique.fr/ces_2008_...ews-11723.html
[...]
AMD got its first chips using the B3 rev, the bug is not addressed still, a new spin is needed which could postponed Phenoms till May.

---

I tried to translate that French site and i cannot see where it sais that the bug not fixed is the TLB bug (there are quite a few bugs can remain normally from revision to revision). Is it the reason why they pushed to 2nd quarter those 9650 and 9550?

What type of calculations set off the bug though? Surely something cache and CPU intensive...

Quote:

---

Originally Posted by Shintai

He just ran WCG and it crashed a few times.

Also i think you misunderstood my post. Its not hard to trigger as such. It just depends on your luck of CPU. If there is a hot thermal spot or not and if so, at what frequencies. I do not believe it s a virtualization only bug. Its a pretty lame excuse. Also you would just have the BIOS setting where you disable or enable virtualization to enable or disable the fix aswell. I call it a smoke screen ;)

---

NP I wrote a mail to x86-64 with a link to the description, waiting for feedback now. The reviewer seems to be an amd-fan so i do not trust that guy.

The hotspot thing is speculation atm. No phenom user in the AMD section could find a relation between stability and temps or between stability and the tlb-fix.

Maybe you are right and not all cpu's are affected. But that whould be something AMD-marketing whould have mentioned.

Here is an success report for the linux tlb-patch.

Quote:

---

Originally Posted by savantu

How do you know that a crash or data corruption problem was caused by the TLB bug ? Do you have advanced debugging tools ?

---

Best answer I can give you is it not crashing under full load at all. The stock one is running Boinc 24/7. The OC'ed one, the only time it's not running Boinc is when I'm gaming or benchin'

Quote:

---

Originally Posted by kl0012

The bug is very critical for servers.

---

True. Especially if your running virtualization on that Server

The thread title is about Phenoms not Barcelonas.
I don't think you'll see too many using a Phenom for virtualized mission critical servers.

Is TLB bug on Phenom going to crash Average Joes(sinlge OS Windows machine) multithreaded game...No

The TLB bug doesn't prevent you from overclocking but overclocking should increase your chances of tripping the bug

Can anybody show me that somebody actually tripped the TLB bug on Phenom(not Bacelona) running a single OS. Please, because I want to try too.

The crashing on WCG probably has more to do with ram compatibility.

Quote:

---

Originally Posted by justapost

Maybe you are right and not all cpu's are affected. But that whould be something AMD-marketing whould have mentioned.

---

You cant do that as a company. Just imagine it. "Only some of our CPUs are faulthy, we dont know which ones".

It would be a PR suicide.

Also patching agains 1 or a few possible bugs is not a fix really in what we usually see.

what is the argument for, or technical details behind, an overclocked phenom being more likely to trip the bug? or is that just people's experience?

Quote:

---

Originally Posted by hollo

what is the argument for, or technical details behind, an overclocked phenom being more likely to trip the bug? or is that just people's experience?

---

I think it's not the fact it's overclocked I think its tied that with the higher bus speed when you upp the fsb it gets more... common say if a BE I think.... just higher multi ther'd be normal chance :)?

please correct if im wrong

Quote:

---

Originally Posted by Jamesrt2004

I think it's not the fact it's overclocked I think its tied that with the higher bus speed when you upp the fsb it gets more... common say if a BE I think.... just higher multi ther'd be normal chance :)?

please correct if im wrong

---

The speed sould be the same. Its only the Core speed that change. Even the bugged L3 is at same speed.

Quote:

---

Originally Posted by Shintai

You cant do that as a company. Just imagine it. "Only some of our CPUs are faulthy, we dont know which ones".

It would be a PR suicide.

Also patching agains 1 or a few possible bugs is not a fix really in what we usually see.

---

The whole launch was PR suicide.
They know in what circumstnaces the bug occures, there is a good decription from amd here.
I guess it should be possible to write a test case for the errata.
Such a test whould be no option for tier1's but for desktop users it whould be a nice stability test.
So if you are right that whould be actualy good news for phenom/b2 users. :rolleyes:

The NB and the L3 cache runs at ref HT (FSB) * NB multi (max 9 for locked phenoms). The multi can be modified with wprcedit.
The NB speed is independant from the memory speed (ref HT * mem multi/mem divider).
NB speed becomes an issue above ~2100MHz, ref HT at ~260MHz.

Paul Demone's view on the bug :

Quote:

---

It probably isn't a straight forward logic bug in synchronous
logic. The four cores and the L3/IMC can all run at different
frequencies and thus there is asynchronous interfaces between
them. AMD might have tried something a bit more aggressive
than straight out double sampling in both directions to reduce
latency crossing clock boundaries. And when two CPUs running
at different frequency and/or phase hit their L3 interfaces at
nearly the same time under some rare set of circumstances
some condition that should never happen, happens. You know,
like the brown stuff. It just happens and now AMD is up that
particular creek.

---

This is the reason why AMD locked the NB speed , they tried to get rid of at least one variable.No wonder that the bug is hard to fix ; it's non-deterministic.
In other words you can't really try to replicate it.Even with code known to cause the bug , it might not happen all the time.

Montecito experienced a similar problem with different clock domains for the cores and the L3 which was asynchronous.Foxton which modified frequency in real time depending on power consumption made this a real nightmare.The fix : Foxton disabled and L3 locked at 1600MHz.

It will be interesting to see how Nehalem will handle this problem since it also has independent clocks for each core , NB and IMC.Knowing Intel , the caches are also far more aggressive in terms of latency and BW.Add to that HT ; you have 8 threads per device , all targeting a shared l2.

Quote:

---

Originally Posted by .Tret

Because of what history? When in history has Intel charged $1000 for a low end cpu and $2000 for mainstream? I don't recall that happening; even in the early pentium pro days (the time when intels prices were highest).

In the pre-k8 days, Intel's prices weren't as high as people think they remember them to be. In the pentium 1 through pentium 3 days Intel always priced thier processors the same. $200-300 for the low-end and $800 or $900 for the high-end.

Even if Intel was a true monopoly, they could not sell cpu's at very high prices because not enough people would buy them to cover the cost to make the cpu's. The market is what determines prices. If AMD were to die right now I doubt Intel would raise their prices much at all. The problem we would face with Intel being a monopoly is lack of inovation not pricing.

That is why when someone says that Intel would price thier low-end processors at $1000 if they were a monopoly or when someone says that nehalem cpu's would cost $2000 if they were a monopoly, they are either stupid fanboys or are just plain ignorant. (too many people)

---

Never said low end cpu would cost $1000, said the PC itself.

Since amd is now both video and proc, its safe to say, if they go bottoms up, that the price of a low end pc could rise from $400 to more around $1000 again.

This is because, intel won't need to produce such a large line of processors to compete, and their can just be like 5, 2 low end , 2 mid range, and a EE.

Anyway, wasn't trying to imply anything in my post, was just saying what I have seen over the years since 1995 in computer prices, and how that could go in reverse w/o competition.

Not a fan boy or whatever, only really 1 company I don't like with good reason and its not amd or intel ,love them both, amd I still use for low end cheaper gaming machines, that can just focus on the video card, then rest of system is done for under $300

Just wondering if anyone noticed the early silicone in brackets, possible early samples of B3 still contain the bug and possible later ones wont. But still doesnt make any different as this bug hasnt really happended to any one has it..

Its like the chances perhaps of a CPU HSF falling off, its a possibility but never really happens to anyone.. well unless your god dam unlucky :ROTF:


If it doesnt effect you why worry...

Quote:

---

Originally Posted by justapost

The NB and the L3 cache runs at ref HT (FSB) * NB multi (max 9 for locked phenoms). The multi can be modified with wprcedit.
The NB speed is independant from the memory speed (ref HT * mem multi/mem divider).
NB speed becomes an issue above ~2100MHz, ref HT at ~260MHz.

---

Who cares above 2100Mhz right now? It doesnt even work at 1800 or 2000Mhz. The only thing that matters is the core speed for the bug.

Also they can not test it as such.

And if AMD had actually told the truth. It wouldnt even be an issue for desktops users. In short they lied about the impact as so many other companies.

The simple most straightforward fix for a hardware virtualization issue is DISABLE IT. Yet AMD choose to default have the TLB patch on with or without virtualization.

Quote:

---

Originally Posted by SEA

I tried to translate that French site and i cannot see where it sais that the bug not fixed is the TLB bug (there are quite a few bugs can remain normally from revision to revision). Is it the reason why they pushed to 2nd quarter those 9650 and 9550?

---

They don't mention the TLB bug specifically but this is the one they're talking about anyway.

Don't know about 9550 and 9650 being pushed back.

Now there is another French site saying they contacted AMD France to clarify the matter and have been officially told that there actually were no problem and that 9700 and 9900 had started to get in big OEMs hands, that they were ramping up the delivery and that everything was still on track for a worlwide launch at the end of this quarter.

http://www.presence-pc.com/actualite...-retard-27203/

.

inq confirmed NO TLB BUG ON B3! ;)

Quote:

---

Originally Posted by tictac

inq confirmed NO TLB BUG ON B3! ;)

---

Theinq also confirmed it to be 50% faster than Core 2 and doing 30K 3Dmark06. Not even to mention the "dancing in the aisle".

Quote:

---

Originally Posted by tictac

inq confirmed NO TLB BUG ON B3! ;)

---

How ironic, the King Rumor-Monger himself squelching a rumor.... how quaint.