Wednesday, May 23, 2007

AMD Q1 2007 Outlook

AMD's Q1 2007 results were certainly disappointing. This has caused the rumors about bankruptcy for AMD to start popping up like mushrooms. Beyond bankruptcy there is the question of AMD's Q1 sales and after that there is the question of outlook for the rest of 2007 and into 2008. Time to dig in.

The simplest way to handle the question of bankruptcy is to look at AMD's history and assume that AMD can survive whatever it has already survived. The most definitive ratio is the stockholders' equity to assets ratio. This essentially tells what percentage of the company is not in debt. AMD's ratio in Q4 06 was 44% but it has now dropped to 41%. In spite of the better than $600 Million loss in Q1 this really isn't too bad since AMD's lowest percentage was 35% back in 2003. This suggests that AMD could lose another $1.4 Billion to hit 35% and still survive (since it did in 2003). Therefore, bankruptcy does not seem to be an immediate problem.

Figuring out AMD's processor position is a bit tougher since they've now started combining processor, embedded, and chipset revenues. However, by any measure it is way down from Q4 06 (about 30% lower). As far as I can tell the dropoff is way above the usual seasonal decline, but the only other comparison is Intel. So, we have two possibilities. The first possibility is that Intel's numbers are indicative of the quarterly demand and demand for AMD has fallen sharply. The second possibility is that general demand has fallen sharply but Intel was bouyed by pent-up demand. It doesn't seem reasonable though to assume that general demand has fallen off as sharply as AMD's numbers so there must be another reason.

However, no other reason is obvious. AMD is now 100% 65nm on FAB36 and is producing more than half of its volume as 65nm. This is very confusing because AMD's inventory has not increased enough to swallow 30% of the production yet it is difficult to imagine that they've cut production that drastically after being maxed out just one quarter ago. The only other explanation I've seen is that AMD has unreported income due to payments next quarter from government sales. While Net 60 or 90 terms are not unusual, one would nevertheless think that if AMD had large revenues due next quarter that they would have mentioned this. Since AMD didn't mention this, I have to discount this line of reasoning. The only other possibility I can think of is that AMD's cpu sales fell off because of delays in delivering chipsets. About the only bright spot in this is that almost every possible reason for the sharp drop in both volume and revenue will get better in the next two quarters.

To see where AMD really stands it is necessary to dispense with the typical shell game that plagues most attempts to characterize AMD. For example, people compare AMD with nVidia for discreet graphics and then compare AMD with Intel for processors. People often dodge back and forth comparing AMD with either Intel or nVidia for integrated graphics and chipsets using whichever seems a tougher competitor. The truth is that Intel doesn't yet make discreet graphics and nVidia doesn't make processors. AMD is currently the only company that makes processors, chipsets, and discreet graphics yet I can't recall a single editorial or analysis that mentions this fact. And, the double standards for Intel and AMD are still very much alive. For example, I've seen comment after comment about AMD's R600 delay yet no mention of Intel's 965 chipset being two months late or the proper drivers being a full year late. Sadly for Intel, it looks like the finished drivers may arrive just in time to discourage upgrade to Intel's newer chipsets for Nehalem. I've now seen people proclaiming that R600 is a total failure when it has already been suggested in the trades that AMD's R600 orders will max out the capacity at foundry giant TSMC.

Without doubt though, the good news for AMD is mini-DTX. Again, the trades have suggested that mini-DTX will shove aside all other contenders to become the dominant small form factor standard on the desktop. Having compared AMD mini-DTX boards with Intel's latest, underpowered mini-ITX offerings, it is difficult for me to argue with this assessment. While Intel will offer cheap, Celeron class solutions on mini-ITX, AMD's only slightly larger mini-DTX boards will deliver real desktop power with double the cores, double the memory bandwidth and memory capacity. AMD can do this while delivering both PCI and PCI-e while Intel is only able to deliver PCI. There is little doubt that Intel is guilty of a major oversight in low cost desktop architecture but it still isn't clear what Intel can do to dig itself out. Mini-ITX is low cost but underpowered while pico-BTX has enough power but can't begin to compete in cost. This leaves AMD mini-DTX with no competitor and Intel with no sandbags to plug the gap in levee.

Likewise, Intel's Geneseo has now shown itself to be not at all a competitor for AMD's Torrenza as was first claimed. First, while Torrenza products are already available, users will have to wait until 2009 to get any Geneseo products. And, these delayed Geneseo products will be little more than upgrades to existing PCI-e products. There is no doubt that this will hurt Intel both in terms of high powered servers and HPC. HPC accelerators benefit from Torrenza's cache coherency which is missing on Geneseo while high powered servers benefit from the much greater bandwidth of HTX for things like networking cards. In simple terms, while Torrenza is designed for maximum power with lowest latency and greatest bandwidth, Geneseo is designed for low cost and overlap with existing PCI-e. This is quite a handicap because Torrenza too is designed to leverage existing HyperTransport technology to minimize development costs. But since PCI-e was already far behind HyperTransport Intel's boost with Geneseo does nothing to close the gap with Torrenza. Essentially, Geneneseo is roughly competitive with HyperTransport 1.0 as AMD moves on to the considerably more powerful HyperTransport 3.0. Consequently, an HTX powered network will run rings around a Geneseo powered network. So, in spite of the new Nehalem architecture we see Intel giving ground in high end X86 server architecture and once again optimizing for low end single and dual socket systems. One certainly has to question Intel's direction given a renewed X86 attack from AMD with K10 and a very aggressive attack from IBM with much faster Power speeds.

So, while Geneseo's delay brings into question when an actual CSI implementation will be ready, we also have to wonder about mobile. Intel took most of the mobile share by delivering the specialized Pentium M processor with a good mobile chipset and wireless capability in the Centrino platform. However, the current situation is ironic indeed. While Intel had a separate architecture for mobile, AMD used the same K8 architecture for mobile, desktop, and server. Now, Intel is using the single C2D architecture for everything while AMD is splitting off a separate mobile architecture. Assuming that AMD can deliver a quality mobile chipset by mid 2008, Intel could find itself facing a much tougher mobile competitor with both a specialized mobile cpu and chipset. A reversal in battery life with AMD notebooks outlasting Intel notebooks could be a serious blow to Intel's mobile share. We will have to see if C2D will truly be able to cover all bases with just a chipset change or whether Intel will have to scramble to try to differentiate the mobile architecture once again.

We can also see that the battle for fair benchmark testing is still raging. Most people don't seem to be aware that with the dominance of the Intel Compiler that benchmark code is often poor quality on AMD processors. This poor quality code gives Intel processors an artificial boost in testing. Yet, there has been no call to have testing code compiled on the PGI compiler even when this code gives a boost to Intel as well.


Looking at this graph we can see that Core 2 Duo runs only 95% of its normal Intel Compiler speed when running PGI Opteron optimized code, however, it runs about 109% of its normal speed when using the PGI Unified Binary. One would think that getting a 9% boost in speed would be enough to make testers want to switch to PGI.
However, here we can see the main problem. While Intel Xeon only falls off slightly while running good Opteron code, Opterons take a much bigger hit while trying to chew through the poorly optimized Xeon code generated by the Intel Compiler. So, anyone who wishes to give Intel an advantage would not want to trade Intel's current artificial 19% advantage for a genuine 9% advantage. There doesn't seem to be any valid reason to continue to use the Intel Compiler to compile benchmarks unless the reason is indeed to tip the scales in Intel's favor. PGI code is 9% faster for Intel than Intel Compiler generated code and fully 19% faster for AMD than Intel Compiler generated code. It's time to stop the sham and make PGI the standard comparison compiler for all benchmark code. There is no doubt that Intel would oppose this since it would be losing not only its current artificial advantage but revenue from its compiler sales as well. However, one does wonder what would happen if all the popular review sites grew backbones and insisted on PGI benchmark code. Until this happens I'm not sure how we will ever know how Intel and AMD processors actually compare.

In conclusion, AMD's current position is terrible. There is no doubt that it would be better if 65nm had been released sooner or the new chipset had been out sooner or K10 had been out sooner. So, AMD is just going to have to bite the bullet until Q3 when things should improve. The chipset and graphic sales should be up by then and AMD should be fully anchored on the desktop with mini-DTX and DTX. These two should prevent any further erosion of the desktop although Intel will still control the top range. Barcelona should begin to take back at least the top end server sales (which are now going to Clovertown) and re-establish AMD's server chip ASP's. However, it will take until Q4 for AMD to be able to start hitting back at the upper desktop range and to re-establish its desktop ASP's. It doesn't look like AMD will drown before Q4 but it won't be much fun getting there. I've also seen suggestions from the Intel hopeful that Penryn will prevent an AMD comeback. I'm sorry but suggestions that Penryn is 30% faster are not exactly accurate. Penryn can be 30% faster than Clovertown in some circumstances because Clovertown bogs down badly with four cores. Penryn however is not 30% faster than Conroe (possibly 10%). And, in spite of Penryn's improvements, native quad is still more efficient than MCM at managing bus access. Unless Penryn gets a substantial boost in clock speed (which is possible) AMD will take back the lead on both desktop and server. Recent tests certainly seem to suggest that AMD will bump its K10 clocks by 100Mhz. However, other demonstrations suggest that AMD could hit 2.97Ghz in Q4 while Intel could pull 3.33Ghz Penryn into Q4. So, it looks like the actual leader in Q4 is going to depend on deliverable clock speed and that is a huge question at the moment since AMD has only admitted to 2.5Ghz while Intel has only admitted to 3.0Ghz.

AMD's position does seem fairly good going into 2008 since it still insists that 45nm is on track and will be ready six months after Intel's. Meanwhile, Intel has severely downgraded expectations of Geneseo while also announcing that desktop versions of Nehalem will not use an IMC. However, even with an IMC, Nehalem is not going to be able to match K10's DC 2.0 connectivity. There is also little doubt that until Nehalem does arrive, Intel is going to get hit hard in power consumption with its monstrous quad FSB northbridge for 4-way. Even with Penryn's lower power consumption Intel is going to have a tough time matching AMD. This is doubly true with Intel still using FBDIMM. I think it is safe to say that Intel is going to lose some of its server position in late 2007 and into 2008 and presumbably is then going to start pushing Nehalem with its IMC advantages. Likewise, Intel is going to lose some of its current desktop position and it remains to be seen what they will do about mini-DTX. Intel may have to create a new small form standard to compete. My guess is that Intel will fair best in the upper desktop and lower server ranges. AMD may have a tough time taking back single socket and dual socket server share although having a real quad core to compete will certainly help. I would say that AMD's strongest position comes from the ability of quad K10 to be a drop-in replacement for dual core on socket F and AM2. So, AMD could potentially improve its volume and revenue position substantially by end of 2008. However, Intel could block this if it can deliver much greater clock speeds with Penryn or possibly if it can get Nehalem hardware out sooner than 2009.

37 comments:

enumae said...

Scientia
AMD's position does seem fairly good going into 2008...

I hope to make a more detailed post later, but just to point this out, you also thought that AMD was in a better position leaving 2006.

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

Also, when you use comments like
"Intel hopeful", you have to admit that you yourself are an "AMD hopeful", right?

No benchmarks have been released from AMD or Intel that could/should have anyone claiming a victory for either K10 or Penryn.

Scientia from AMDZone said...

enumae

My thoughts about AMD in Q1 were inline with what AMD said (and presumably expected). If you know why AMD dropped off so sharply then please let me know because I still haven't figured it out.

No, I'm not an AMD hopeful and I really shouldn't have to explain this to you. Since you don't seem to understand the difference though I guess I'll have to.

An AMD or Intel hopeful (enthusiast, fan) would want their favored company to do well on all fronts and wouldn't want to see any loss to the other company. Therefore, an enthusiastic AMD prediction would be that AMD would gain on all fronts. You should be able to tell that that is not what I'm saying.

I see AMD moving very aggressively into the low range with mini-DTX after mid year. This is not actually my view; this is from the trades, however, I can't think of any reason it wouldn't be true. The midrange should also do well with DTX although probably not quite as strongly since the advantages are less pronounced. The trades have again suggested that R600 will sell very well; I don't have any specific knowledge if this could be true or not. However, I'm inclined to think that this will require 65nm chips.

I could see AMD's doing better with mobile, although I don't see a really strong position until mid 2008 so AMD might continue to nibble at Intel's mobile share.

In servers, I see AMD as having an advantage with 4-way and up because Intel's quad FSB chipset pretty much neutralizes and reverses the low power draw from Penryn. Since single and dual socket servers aren't saddled with this monstrous chipset I see AMD as having less advantage in single and dual socket. Likewise, Geneseo's lukewarm capabilities are less of a factor on the server low end. It is obviously better for AMD to have a quad core competitor however Penryn will improve some of the performance problems of Clovertown so this may not be as strong for AMD as it could be. However, drop-in capability might help too. So, I think AMD is going to have a tough time taking back share in single and dual socket server but it is possible.

The upper desktop range is not a clear advantage for AMD. ATX is not really lower cost than BTX and Penryn will at least partially make up for the problems with Kentsfield. Unless Intel does something that hasn't been yet announced it appears that AMD will take back the FX range. However, it is not clear AMD will have a lead at the desktop top end. This could allow Intel to maintain ASP pressure at the upper desktop and the volume of FX is too low to create much revenue for AMD.

If you disagree with these comments then feel free to post something substantial instead of silly ad hominem remarks. Finally, I don't understand your "victory" comment. What is that related to?

sharikouisallwaysright said...

Hello, i know my english isnt that good.

I will ask some question:
AMD was unable to deliver the wanted numbers of cpus in q4/06?

Demand ist strong in q4 and people buyed what was left to buy - Intel?

The CPUs AMD could not deliver as demanded in q4 are in the stores now and as demand is low in q1/07 AMD is not selling its CPU as Stores are full from the late q4 delivery?

Maybe AMD was not producing the CPUs the market had wanted, i dont know what is best selling - is it single core, dual core, sempron or athlon or opteron if AMD?

Is the market changing away from Desktop to Mobile as more and more people use laptops and notebooks instead of desktopcomputers even at home?

The HD2900 is not as good as expected from some people, but the 2600 or 2400 maybe quit good - but late and even later, and what about the rumor AMD wanted to produce the chips itself as heard from the inq?

So the delay of some products, the pricewar and maybe wrong expectations of the markets demand are hurting AMD?

Scientia from AMDZone said...

sharikouisalwaysright

"I will ask some question:
AMD was unable to deliver the wanted numbers of cpus in q4/06?"


True; they were behind on Q4 orders.

"Demand ist strong in q4 and people buyed what was left to buy - Intel?"

No. Intel's volume didn't go up that much in Q4. And, Intel's volume dropped in Q1.

"The CPUs AMD could not deliver as demanded in q4 are in the stores now and as demand is low in q1/07 AMD is not selling its CPU as Stores are full from the late q4 delivery?"

AMD dropped a lot more than Intel.

"Maybe AMD was not producing the CPUs the market had wanted, i dont know what is best selling - is it single core, dual core, sempron or athlon or opteron if AMD?"

It is possible that AMD isn't producing enough 65nm chips.

"Is the market changing away from Desktop to Mobile as more and more people use laptops and notebooks instead of desktopcomputers even at home?"

AMD shifted more to mobile in Q4.

"The HD2900 is not as good as expected from some people, but the 2600 or 2400 maybe quit good - but late and even later, and what about the rumor AMD wanted to produce the chips itself as heard from the inq?"

As far as I know these will be produced at TSMC.

"So the delay of some products, the pricewar and maybe wrong expectations of the markets demand are hurting AMD?"

There is no indication of falling prices for AMD systems. Perhaps graphics or chipset delays did hurt.

sharikouisallwaysright said...

I dont even know how to quote here, sorry ;)

What i want to say is, if there was a great demand for AMD-CPUs in q4 but they failed to deliver in time so there is a stockpile of this late delivered CPUs at the Vendors from q4 that is unsold.
If this happen and q1 has traditionally low demand there is no need for the business to buy that much AMD-CPUs in q1 and the drop for AMD is much bigger therefore than expected while Intel has nearly normal q1-business.

HD2xxx is produced at TSMC but the Inq wrote that AMD would have liked it to do it by itself and some of the delay is coming from this but this is only a rumor.

I only have this german link as i can not reach Inq UK yet:

http://de.theinquirer.net/2007/05/23/amdati_unsere_produktion_laeuf.html

At the bottom they say AMD must have planed to produce GPUs by itself.

Azmount Aryl said...

Let me get this straight - we still don't know the volume MS for 2007Q1?

Azmount Aryl said...

One more thing, what do you think about this review? I found it hard to believe that all so anticipated HD X2900 falls behind 8800GTS 640MB in almost every game tested by them! However i did notice that they used beta drivers for ATI card and that also they employed WinXP-32, that is an OS that i wouldn't care much for, I'd say most people looking to buy such video card would prefer that this testing be done in Vista-32. Still, makes me wonder, what if even with the newer driver that this X2900 wont be able to match Nvidia's offerings...

Roborat, Ph.D said...

AMD Q1 2007 Outlook?

It's impossible to have an "outlook" on something from the past nor the present, unless of course you wrote this article last year. It's either a "look back" or you change the title to a future date.

sharikouisallwaysright said...

Almost a Winner ;-)

The HD 2900 has some difficulties but can win against the 8800 GTS.

http://www.xbitlabs.com/articles/vi
deo/display/radeon-hd-2900-
games.html

Azmount Aryl said...

According to the link you gave that X2900 is at least twice slower than 8800GTX in NeverwinterNights-2, Supreme Commander and S.T.A.L.K.E.R.

This is simply unacceptable, ATI has never been to popular with gamers, and now it seems their new video card is only a 'cheaper, underperforming option'?
I am not very satisfied with this review by the way. They didn't mention the clocks for GPU/mem at all. For all we know they might be comparing some factory overclocked X2900 against normal 8800's. What do you think about that?

Unknown said...

http://hardocp.com/article.html?art=MTM0MSwsLGhlbnRodXNpYXN0

enumae said...

Scientia
If you disagree with these comments then feel free to post something substantial instead of silly ad hominem remarks.

I am actually out of town, but I do plan on commenting further.

In regards to the ad hominem comment, as it should be clear, I have no intent to be rude.

I was simply pointing out that your view of AMD leaving 2006 was that they were stronger than Intel, and it wasn't correct.

Until we have real information pertaining to K10, I think your being overly optomistic about the performance of AMD, while also under estimating initial clock speeds of Intel.

If they are running demoes at 3.33GHz, they could possibly release production higher. I say this based on there initial demos of Conroe, they were using the E6700 instead of the X6800.

Finally, I don't understand your "victory" comment. What is that related to?

You are already making predictions based on unknown performance and predicting adjustments to market share accordingly.

I will try and explain with more detail later, got go.

Scientia from AMDZone said...

azmount aryl

I don't have firm numbers for AMD's cpu volume but I would say my estimate of a 30% drop from Q4 is pretty good unless there have been much larger changes in chipset sales ratios.

The 2900 faired better in other reviews. For example in the LegitReviews it even managed to beat 8800GTX in one test. Obviously though, it draws a lot of power and with no top bin part and the two lower bins delayed it does look like someone fumbled a bit. This might look a bit better at 65nm and they could also get things in better shape by R700.

sharikouisalwaysright

I've heard nothing about R600 chips being produced in-house at AMD. Also, the decision to manufacturer in-house would have to be made early since AMD's 65nm process is quite different from TSMC's. This is another reason why I don't think it is likely. However, as AMD moves into Fusion I'm sure they will start making graphic components in their own FABs.

Scientia from AMDZone said...

roborat

You can only come up with a semantic criticism? The outlook is from Q1 rather than for Q1. You might recall that I talked about Q3 and Q4 07 and into 2008.

enumae

"I was simply pointing out that your view of AMD leaving 2006 was that they were stronger than Intel, and it wasn't correct."

What I actually said was that I expected AMD's gross margins to improve slightly in Q1 (which obviously was not correct). I don't recall saying that AMD would be stronger than Intel in Q1.

I did say that I expected AMD to do better (and gain share) later in the year after the new chipsets were released, production came up on FAB36, and K10 was out. However, what does that have to do with Q1?

"If they are running demoes at 3.33GHz, they could possibly release production higher."

I said the same thing in the article you just read:
However, Intel could block this if it can deliver much greater clock speeds with Penryn

I have to say that I'm baffled why it is that you mention the 3.33Ghz demos but then overlook reports that K10 has clocked to 2.97Ghz. I have no idea why when I say that Intel could deliver chips faster than 3.0Ghz in Q4 and that AMD could deliver chips faster than 2.5Ghz that that somehow makes me an AMD fan.

Apparently, by your definition, an AMD fan is anyone who doesn't strongly favor Intel.

"You are already making predictions based on unknown performance and predicting adjustments to market share accordingly."

AMD has said that K10 is 20% faster than Clovertown at the same clock; this is why I pointed out that Penryn improves on Clovertown. In other words, K10 wouldn't be 20% faster than Penryn at the same clock.

The only market change I've suggested as likely is the low range desktop. Again, the trades have suggested that mini-DTX will be the dominant small form standard. I've done my best to compare mini-DTX with the only existing competitor, mini-ITX and there is no comparison.

Now, I'm sorry if something that doesn't favor Intel is upsetting you but this really is a modest change since this is the lowest profit segement on the desktop. DTX is also ahead of Intel's BTX standards but it looks like Intel could drop the prices of its cpu's more to offset this advantage.

And, at the ATX and higher levels, I don't see an advantage. AMD's 4X4 with the new quad FX chips is going to take the lead again but this is primarily because Intel has not yet shown any interest in producing a dual socket X platform. If you know something about a dual socket X platform from Intel then feel free to post it. If Intel chooses to release its own dual socket competitor to QFX then Intel could very well hold the lead there.

Scientia from AMDZone said...

roborat

I'm still waiting to find out where you got your claim that Intel can make a quad core for $53 while it costs AMD $175.

sharikouisallwaysright said...

In some games the hd2900 can beat the 8800 GTX or compete with it, it is in most other games better or even with the GTS and in a few Games there are serious difficulties for the hd2900.

Personally, i will not buy a new GPU, my x1950xtx is good til 2008 or 2009, but if one wants to upgrade it is a choice when offered in the right pricerange.

And we dont know how hd2600/2400 are doing and if they beat Nvidias solution as i.E. the 8600 sounds not so good on paper, they will have success in terms of business.

Time will tell...

Scientia from AMDZone said...

enumae

You've mistated my position several times so, yes, I've been annoyed.

What that I have actually said do you disagree with?

Azmount Aryl said...

sharikouisallwaysright said...
In some games the hd2900 can beat the 8800 GTX or compete with it


By your own logic it seems that X2900 is not better than GeForce7900GT. Because In Some games it fell to the levels of 7900.

sharikouisallwaysright said...
and in a few Games there are serious difficulties for the hd2900


I'd like to point out that that is a problem. A very big problem. Nvidia's offerings do not suffer x2 drop in performance on selected games. And the fact that ATI's new card does, might.... no, better yet - Will turn away buyers.

sharikouisallwaysright said...
but if one wants to upgrade it is a choice when offered in the right pricerange.


...unless the person wants the best, in which case they only have one choice, and that is Nvidia.

Look, X2900 is a failure, lets just all admit that and move on. Personally I don't think its AMD's fault, in fact the only thing AMD did here is helping their ATI devision to move faster to the 65nm and that is good thing, the kind of a thing that will help X2900 get closer to the performance of top Nvidia's cards.

Running closer to the winner however, wont make you a winner. It will put you on 2nd place at best. And i was hopping for a winner...

enumae said...

Scientia

I would like to make one statement, and then tomorrow, with a clear head, and no feeling of being rushed (as I was traveling), create an entirely new response to your article while deleting the ones that may have been short and unclear/misinturpreted.

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

An AMD or Intel hopeful (enthusiast, fan)...

I take "Intel hopeful", or "AMD hopeful" as someone who has brand loyalty/preference, not a fanboi.

My interpretation of your comments was along those lines, and I had no intention of calling you a fanboi, or an "AMD fan".

I have no desire for AMD to go away, nor do I want a lack of competition.

I was trying to hurry, and it seems I should have just waited and made a more detailed post, for that I apologize.

Thank you for your patience.

sharikouisallwaysright said...

I dont expect much from a simple shrink, ATI had to do some redesign or evolution on the 2900 to meet the expectations from the people who want to see an all time winner.

Azmount Aryl said...

2900 is not a simple shrink.

From THIS slide you can see that 2900 is nothing like X19XX in its design, it just they fell short of what people demand this days. I do wonder, now that it is made clear that 2900 is not competition to the top 8800's, what will be the next move by AMD? Will they bump the clock when 65nm parts are out? Will they bump it high enough to make energy saving features of the process to disappear/be-sacrificed in order to gain performance crown ?

sharikouisallwaysright said...

K10 12% faster as K8 in Socket AM2 ?

http://xtreview.com/addcomment-id-
2508-view-k10-vs-k8.html

Azmount Aryl said...

What that is, is a 'blop' - pure speculations based on test results that don't exist just to make an article that explains nothing to nobody, because 'nobody' are the people who will take that article serious .

Neeext!

savantu said...

I have to say that I'm baffled why it is that you mention the 3.33Ghz demos but then overlook reports that K10 has clocked to 2.97Ghz. I have no idea why when I say that Intel could deliver chips faster than 3.0Ghz in Q4 and that AMD could deliver chips faster than 2.5Ghz that that somehow makes me an AMD fan.


Because Intel demoed 3.2 and 3.33GHz Quad Cores 2 months ago at
IDF while AMD demoed Task Managers.

In 6 months Intel can go at least 1 bin higher , possibly even 2 if the need arises.

Azmount Aryl said...

savantu said...
Because Intel demoed 3.2 and 3.33GHz Quad Cores 2 months ago at
IDF while AMD demoed Task Managers.


So what? AMD and intel don't do business the same way, history says so. Benchmarks for original Opteron were under NDA till the day Opterons were released be cause thats how AMD does things. And i don't even have to remind you about that incident where intel demoed all known processor 6 month before it actually arrived to the stores, we all know what that did to their sales (making it the worst Q for them ever).

savantu said...
In 6 months Intel can go at least 1 bin higher , possibly even 2 if the need arises.


Apple been selling 3.0GHz intel quads for some time now. Its a 160 Watt part. If intel has to sell Sun-hot chips just to keep up with AMD in performance then we have yet another 'Pentium(core2duo) vs. Athlon(phenom)' situation. Remember who won the Wattage race.

Heat said...

So what? AMD and intel don't do business the same way, history says so. Benchmarks for original Opteron were under NDA till the day Opterons were released be cause thats how AMD does things. And i don't even have to remind you about that incident where intel demoed all known processor 6 month before it actually arrived to the stores, we all know what that did to their sales (making it the worst Q for them ever).

Going by the statement you make it sound like AMD is making money these past months...releasing benchmarks will not hurt them either way if anything they will regain public confidence by revealing them and not showing slideshows which they have become really good at doing......

Apple been selling 3.0GHz intel quads for some time now. Its a 160 Watt part. If intel has to sell Sun-hot chips just to keep up with AMD in performance then we have yet another 'Pentium(core2duo) vs. Athlon(phenom)' situation. Remember who won the Wattage race.

First of all the 3.0ghz is a quad core part and need i remind you that the AMD dual core 6000+ STILL TAKES MORE POWER AT LOAD THAN THE QUAD CORE C2D.

In case you havent been keeping up the penryn is not only a die shrink but also includes the high k gate which potentially will make it run cooler as well as be able to clock higher. This is a whole new generation buddy get off the netburst and live in the present and presently AMD is getting a whooping both at all out performance, performance per watt and anything else AMD's marketing company can come up with.

On a side note i like how scientia compares a stock clocked Penryn to INQUIRER reports (no less) of an overclocked K10......werent you crying about the same thing when tomshardware used to overclock processors and compare it to stock......hypocrisy by an AMD hopeful .....naw cant be.

Scientia from AMDZone said...

heat

"On a side note i like how scientia compares a stock clocked Penryn to INQUIRER reports (no less) of an overclocked K10......"

Until these chips are actually released there is no such thing as "stock" or "overclocked". If Intel releases Penryn at 3.0Ghz then the demo would have been "overclocked" as well. Both of these are merely suggestions of what might be released.

Heat said...

Until these chips are actually released there is no such thing as "stock" or "overclocked". If Intel releases Penryn at 3.0Ghz then the demo would have been "overclocked" as well. Both of these are merely suggestions of what might be released.

That would be true but just going by official reports alone we have Intel demo a 3.33ghz quad core part.....AMD officially has stated that their highest part will be clocked at 2.5Ghz only the inquirer states otherwise.

Then why would you comment about overclockability factor and bring up the inquirer when someone is asking you about the K10s clock speed and on the other hand cry about the C2D overclockability prowess on different websites when it is proven fact on every website.

Didnt you post the same thing about C2D not a while back why Intel wont release a higher clocked C2D even though it is proven that they can easily be overclocked way higher than 2.93Ghz.......what would make AMD any different....if that isnt AMD hopeful i dont know what is..

savantu said...

Apple been selling 3.0GHz intel quads for some time now. Its a 160 Watt part. If intel has to sell Sun-hot chips just to keep up with AMD in performance then we have yet another 'Pentium(core2duo) vs. Athlon(phenom)' situation. Remember who won the Wattage race.

It's 130w , the classic Intel TDP range.

May I ask with who has Intel to keep up ?

I see no K8L in sight for the coming months.

By septemeber when it'll arrive Cloverton will see new chipsets with vastly improved memory performance including a low end 2S chipset with plain DDR2 .

Mark my words and mark them well : Barcelona/K10/K8L won't touch the high end Cloverton&new chipsets in performance except on some synthetic tests that heavily rely on memory performance.

That's the same way in which current K8 boast higher memory BW than C2D.

Scientia from AMDZone said...

heat

"That would be true but just going by official reports alone we have Intel demo a 3.33ghz quad core part.....AMD officially has stated that their highest part will be clocked at 2.5Ghz only the inquirer states otherwise"

Okay, what you've just done is use a double standard. You are going by an "official" demo for Intel and then using unnoficial roadmaps for AMD. This is nonsense.

The highest clock that Intel has unofficially listed for 2007 is 3.0Ghz while AMD's highest unofficial clock is 2.5Ghz. You try to spin this by using the term official but until either AMD or Intel actually announces it or releases it it is not official.

Once again, your use of the term "overclocking" is silly since without a released or offically announced speed from Intel there is no way to tell if the chip is stock or not.

Intel has demoed a 3.33Ghz chip but we don't know if it intends to release it in 2007. Intel has 3.0Ghz chips now and AMD's unofficial roadmaps only show 2.5Ghz. So, once again, AMD could release something faster than 2.5Ghz and Intel could release something faster than 3.0Ghz.

Scientia from AMDZone said...

savantu

"By septemeber when it'll arrive Cloverton will see new chipsets with vastly improved memory performance including a low end 2S chipset with plain DDR2 ."

There is nothing wrong with the memory speed of FBDIMM. I'm not sure why you assume that DDR2 will make it faster. The primary disadvantage of FBDIMM is higher power draw. In other words, your notion of "vastly improved memory performance" is a myth.

In reality, C2D will improve memory performance in proportion to the increase in FSB speed while changing to DDR2 will reduce power draw slightly.

"Mark my words and mark them well : Barcelona/K10/K8L won't touch the high end Cloverton&new chipsets in performance except on some synthetic tests that heavily rely on memory performance."

Your words are simply incorrect. K10 will do well on the same benchmarks that C2D currently does well on.

"That's the same way in which current K8 boast higher memory BW than C2D."

You obviously are not familiar with the K10 architecture. The difference in performance between K8 and K10 is quite large in terms of SSE and significant in terms of Integer. K10 will outperform Clovertown at the same clock regardless of what chipset it uses. On the other hand, Penryn has improved IPC and a higher clock which could keep it in the lead.

Heat said...

Okay, what you've just done is use a double standard. You are going by an "official" demo for Intel and then using unnoficial roadmaps for AMD. This is nonsense.

So its nonsense for me to compare an Official benchmark of Intel running a computer at 3.33ghz because they mite not release that part even though they have a chip running stable enough to let everyone demo. But it isnt nonsense for you to believe an inquirer report with not a shred of proof to back it up.

Once again, your use of the term "overclocking" is silly since without a released or offically announced speed from Intel there is no way to tell if the chip is stock or not.

What i am talking about is what you are doing in your blog. Taking information available and coming up with a speculation about an unreleased product granted my information is actually more reliable than your inquirer reports. If you want me to talk about something after it gets released i would suggest you to practice what you preach and talk about either the crap 6000+ or the garbage quadfather instead of the K10 as well...............

Intel has demoed a 3.33Ghz chip but we don't know if it intends to release it in 2007. Intel has 3.0Ghz chips now and AMD's unofficial roadmaps only show 2.5Ghz. So, once again, AMD could release something faster than 2.5Ghz and Intel could release something faster than 3.0Ghz.

I dont care if the Intel chip is releasing 2007 or 2008 i dont remember seeeing a date in my post unless you read my mind or something. All i said is that the next generation intel chip will be clocked at atleast 3.33Ghz or better and according to AMD their highest barcelona part will be at 2.5Ghz or maybe 2.6....

The reason i pointed this out is because you were talking about how the inquirer reported that they can be overclocked to 3.0ghz (with no proof to show you just accepted this as true typical sharikou move) therefore AMD mite come up with a faster part which is totally contradictory to your previous statements about C2D when you posted about how user overclocking potential does not equate a chip being released by a manufacturer at that overclocked speed..........

If you choose to remain in your delusions that AMD will release a 3.0Ghz part to start with the release of barcelona so be it. Not one logical person has that much faith in AMD so good for you.

AMD has always had issues releasing high clocked parts and though this is a new arch you have to realize that not only is it a quad core part but AMD recently changed their process node to 65nm which is also not as efficient as Intel's.

Coupled with the fact that AMD is suspiciously secretive about clockspeed this late in the game should be an indicator and AMD running their mouth and delaying their products should be another.

savantu said...

Scientia said

There is nothing wrong with the memory speed of FBDIMM. I'm not sure why you assume that DDR2 will make it faster. The primary disadvantage of FBDIMM is higher power draw. In other words, your notion of "vastly improved memory performance" is a myth.

You didn't get what I've said : the new chipsets Intel is about to launch ( the Stoakley platform ) offer 25% better memory performance with the same FBDIMMs.Also the snoop filter is increased to 24MB.


Your words are simply incorrect. K10 will do well on the same benchmarks that C2D currently does well on.


Like PovRay ?


You obviously are not familiar with the K10 architecture. The difference in performance between K8 and K10 is quite large in terms of SSE and significant in terms of Integer. K10 will outperform Clovertown at the same clock regardless of what chipset it uses. On the other hand, Penryn has improved IPC and a higher clock which could keep it in the lead.


I think I'm more familiar than you. Please tell us why the difference is significant for Integer between K8 and K10.

As for free comments like "K10 will outperform C2D clock for clock" I think you're in for a big surprise.

Christian H. said...

My thoughts about AMD in Q1 were inline with what AMD said (and presumably expected). If you know why AMD dropped off so sharply then please let me know because I still haven't figured it out.

I think it's rather simple. Combining GPU,CPU and chipset margins caused a drop of probably 12-15%.

Then the X2 went down by half. Even with 4-6% growth, those two things caused the major problem.

Now that they have 65nm Turions, they can raise those ASPs a little with the higher clock and lower power.

As was recently reported by nVidia the channel is clearing so AMD can actually ship more.

Toshiba has signed on with 20% of their laptops going AMD which we should se the effect of this quarter.
A possible issue will be the 90nm vs 65nm chips. Since Fab36 is now 100% 65nm and at supposedly 20K WSPM, there will be a struggle for die space between Brisbane and K10.

Added to that is the fact that Fab30 will bechurning out 90nm for awhile to come, so they have to really be careful what chips stay 90nm.

I would say Opteron as it is the only one that doesn't have a shrink planned.

Then the Brisbane/Kuma change over has to be efficient in favor of Kuma/Rana. Brisbane should EOL in Q4/Q108. That space can be taken up by Griffin, which is where the majority of growth is now.

I don't think the future is terrible but even roses have thorns so a rosy outlook is the good with the bad.

Maybe next time they'll not drop prices so drastically. All it did was gave reviewers something to talk about (price/perf).

Aguia said...

I'm still waiting to find out where you got your claim that Intel can make a quad core for $53 while it costs AMD $175.

Not according to Intel own presentations. Maximum 12% savings I was expecting a much larger number. Maybe the double dual core packing is much more expensive than a single one. Besides they still have to manufacture 2 processors to get one out, right?

Benefits of a dual die quad core processor>
and
First quad core in 4Q 2006

I wonder why Intel releases this kind of data to the public.
I can’t find one situation where the general public wins except the faster time to market, but even that AMD as released their dual core part about the same time of Intel. But now, AMD is already 6 months late and still counting.

savantu said...

Not according to Intel own presentations. Maximum 12% savings I was expecting a much larger number. Maybe the double dual core packing is much more expensive than a single one. Besides they still have to manufacture 2 processors to get one out, right?

Imagine a DC costs $40 to make.2x of them and you have $80.

12% is $10 savings for a quad core ( 70 vs. 80 ).

Now make 5 million of these.You save $50m.Average QC price is 3x that of DC.

In the end you save $50m from manufacturing and your revenues are 1.5-2x as high.

george said...

GCC is probibly the best benchmark
compiler for linux systems because it's the standard and you got the source and.

On servers Opterons are unmatched. because of the bus arch.