Print 58 comment(s) - last by zsdersw.. on Mar 11 at 7:10 AM

BIOS issues and other concerns are addressed

Anandtech was able to score a second session with Intel's Conroe and an overclocked AMD FX-60. Some concerns and issues were raised concerning the first test, so Anand addresses them with this update.

The performance picture with regards to Conroe hasn’t really changed all that much - on average we’re still seeing a bit over a 20% increase in performance over an overclocked Athlon 64 FX-60.  While it’s worth noting that these results should be taken with a grain of salt, we really were not able to determine any cause for suspicion based on Intel’s setups.  The machines were as clean as they could get, with the BIOS oversight having no tangible impact on most performance.

After this second round of tests, there doesn't appear to be anything malicious or underhanded going on Intel's part. The BIOS had little to no effect on performance and the Conroe still comes out on top. Now all we have to do is wait for AM2 Athlons.

Comments     Threshold

This article is over a month old, voting and posting comments is disabled

Real Comparisons
By m3a2 on 3/9/2006 6:31:15 PM , Rating: 2
I wonder why Intel didn't just compare the new Conroe to one of its own currantly existing chips? Afraid to show how bad their old chips are or how little hey have improved on their new chip?

RE: Real Comparisons
By Brainonska511 on 3/9/2006 6:38:20 PM , Rating: 2
Why would they do that? It is better publicity to bash the competitor's product, since that is what Intel is facing in the market. They are competeting with AMD, not themselves.

RE: Real Comparisons
By clnee55 on 3/9/2006 6:47:06 PM , Rating: 2
I wonder why you wonder. Showing something better than your competitors versus showing something better than your own product. Do you need to ask that question? Actually, if they did as you suggest, the improvement should be larger assuming that Conroe > AMD > P4. They still can be proud of a large improvement. Your argument is so funny. It is like saying "you should be ashamed of your great aacchievement because you didn't do that in the past".

RE: Real Comparisons
By Regs on 3/9/2006 7:58:23 PM , Rating: 2
The K8 has been AMD's blockbuster for many years. Consider a company like mine, one I work for and one I wish not to disclose. Every CEO knows in today's business world that you can only market, sell, and then sit on a blockbuster for so long. Upper management will always stress the importance of research and development even when stocks are up. Sales and marketing is a major part of every successful company, no doubt about that. However on the other end you have the research and development effort. I work for research and development in my company, however I'm only a low level employee. On the first day of my employment I signed a confidentiality contract with the company which holds me responsible and liable for all information in R&D and within my department. Our R&D facilities and directors of research carefully manage all information within the company and more importantly within the individual departments themselves. This allows the containment of information. Even our own CEO, who addresses the shareholders, does not know more than half the information that our R&D department contains. He only known's what the share holders want to hear in reality. Everything else he knows is all numbers and cents. An all too familiar term used in business is number and cents. It's their common language. Our R&D department is basically in another world compared to the more business and legal end of our company. This is exactly the way AMD works. The executives know exactly what they are working on, however all they know right now is how much money and resources they are implementing on the research. Upper management plans the long term goals of the company by managing their resources, personal and capital. The truth is they try to know as little as possible. However when they ask a question to R&D, the expect answers. Conroe was Intel's R&D answer to Intel's executives, board of directors, and stock holders. Intel offered them the capital and resources (like the Israel Research and Development facilities) then R&D provided them with an answer when needed. A working model. One which Intel's business and marketing teams took over. And here it is, on Anandtech's very own site. A solution six months away.

AMD's executives are now asking AMD's R&D departments for an answer. So ask yourself this, did AMD offer capital and resources to their development facilities over the years just as Intel did? Maybe licensing to new technologies, contracts with other semiconductor companies, building or opening new production facilities, or even new machinery and engineers? They are asking questions. Will they deliver an answer? It all maters on one fundamental aspect of business -- did they have the resources in place to make their long term goals a reality?

RE: Real Comparisons
By m3a2 on 3/9/2006 10:13:41 PM , Rating: 2
What i'm saying is this, if they want to claim 20% over their previous chip let them benchmark against that chip and prove it. Benchmarking against your competators chip and claiming % over it and your other chips smells fishy. Oh well we will see in 6 months what happens.

RE: Real Comparisons
By m3a2 on 3/9/2006 10:15:24 PM , Rating: 2
Also is it true, that they ran the AMD with video drivers optimized for the Intel chip?

RE: Real Comparisons
By JackPack on 3/10/2006 1:14:53 AM , Rating: 2

RE: Real Comparisons
By Maskarat on 3/10/2006 7:50:05 AM , Rating: 2
Are you high on dope?? Cause you certainly are blowin smoke out from your @*s

RE: Real Comparisons
By masher2 on 3/10/2006 8:55:22 AM , Rating: 3
> Benchmarking against your competators chip and claiming % over it and your other chips smells fishy..."

Lol, I actually had to read this twice to convince myself someone actually typed such nonsense. Comparing your product directly to your competitors is "fishy"?

A mind is a terrible thing to waste...

RE: Real Comparisons
By clnee55 on 3/10/2006 12:19:00 PM , Rating: 2

Since you don't believe what Anand did and said, I am 100% sure you won't believe what Intel would do and say. TRy another excuse to support your pain.

"What i'm saying is this, if they want to claim 20% over their previous chip let them benchmark against that chip and prove it. Benchmarking against your competators chip and claiming % over it and your other chips smells fishy. Oh well we will see in 6 months what happens."

By Sunbird on 3/9/2006 11:16:15 AM , Rating: 3
Is it me, or does there seem to be some underlying unhappiness that Intel has been found to be doing nothing wrong?

RE: Unhappy
By raskren on 3/9/2006 11:20:24 AM , Rating: 2
Not here.

Maybe if you're an AMDroid.

RE: Unhappy
By kattanna on 3/9/2006 11:36:37 AM , Rating: 2
LOL as a AMD fanboy seeing intel do this makes me VERY happy...

kick AMD out of its complancency

besides dual cores...the processor world has been frankly boring for the past 2 years

but hopefully that is changing

RE: Unhappy
By scabby on 3/9/2006 12:03:38 PM , Rating: 2
Agreed... two sleeping giants makes... well, for loud snoring. Wake up and give us some neat processors please!

A solid Intel product might get AMD off of their duffs. More competition = better products = lower prices in the long run.

RE: Unhappy
By bbomb on 3/9/2006 9:05:44 PM , Rating: 2
I too am hoping AMD gets their asses handed to them. They HAVE become complacent to the point where they aren't too much cheaper than Intel and thought they had overtaken them it seems. AMD has offered nothing really new aside from dual core, if you count that, and are still working with K8 architechture. Do they have any plans to finally abandon it and move on?

More trustworth
By Doormat on 3/9/2006 11:34:24 AM , Rating: 2
These benchmarks seem more trustworthy than their predecessors, however they still arent as good as independently run ones. It does look like Intel will regain the performance lead, and by more than 10% - which means AMD would have to release a FX64 (3Ghz dual core) on the AM2 style to get back close to even to a 2.66Ghz Conroe.

RE: More trustworth
By masher2 on 3/9/2006 11:45:23 AM , Rating: 2
> "Which means AMD would have to release a FX64 (3Ghz dual core) on the AM2 style to get back close to even to a 2.66Ghz Conroe"

And even then, Intel will still hold the performance-per-watt and performance-per-clock cycle crowns. Not that the latter one is in any way meaningful...but since AMD fans have been touting it the past several years, I wonder how they're going to react.

RE: More trustworth
By bupkus on 3/9/2006 1:29:24 PM , Rating: 2
...but since AMD fans have been touting it the past several years, I wonder how they're going to react.

Really... who cares?

RE: More trustworth
By bbomb on 3/9/2006 9:09:01 PM , Rating: 2
I will say that both AMD and Intel are doing pretty good jobs of getting people off the Ghz is better thinking. Quite a few less people consider that CPU speed anymore.

RE: More trustworth
By knowyourenemy on 3/9/2006 3:20:08 PM , Rating: 2
More likely AMD will ditch Toledo and just get a better design entirely. By the way, isn't Conroe 65nm?

sad day for computer enthusiasts
By cgrecu77 on 3/9/2006 12:21:48 PM , Rating: 1
I can't understand how some people can be happy with this ... If Intel regains the performance crown by a significant margin it could easily spell the end for AMD. With an agressive marketing strategy and a superior product they could bury AMD and I'm pretty sure that AMD won't have the ability to fight back once more ...
And if AMD goes down we all know what the future holds, 500 dollars for a medium range CPU, $2000 for a top-of-the-line and so on. We will see a 200 Mhz increase every other year if we're lucky ... And since a CPU is so amazingly complex and the technology and resources required to build a CPU are so difficult ti acquire you can be sure that there will NEVER be a competitor for Intel. I just can't understand how so many people hate Microsoft (who delivers decent products at decent prices) while they love Intel (who usually delivers poor products at astronomical prices) ...

By Sensai on 3/9/2006 12:30:40 PM , Rating: 2
hehe.. you're giving your hopes up for nothing... AMD won't go down that easily.... the current 939 and very soon AM2 is what they have now. 90nm am2 is not expected to do any significant jumps in power and performance as we perceive conroe will do. you can hope that the next significant architecture improvement or 65 nm (for wattage) move covers even more ground.

By masher2 on 3/9/2006 12:35:55 PM , Rating: 2
> "With an agressive marketing strategy and a superior product they could bury AMD"

Intel could have "buried" AMD any time it wanted in the past 10 years. Strategically, it would be a very poor move for them however. They'd gain only 20% market share (and not the most profitable 20% at that) at the cost of gaining a huge amount of antitrust attention.

Far better to keep an easily-controlled minor competitor in the marketplace. The past year or two, though, they slipped up, and the 'minor' competitor started giving them a little more heat than they wanted.

By trivik12 on 3/9/2006 1:03:23 PM , Rating: 2
I dont think AMD is going down at all. Only their revenue will be hit if they have to decrease prices once ICM releases. I think they will delay the decrease as much as they can. But by May-June we would know exact release date of ICM and discerning buyers will delay their purchase then if they know the prices will be going down.

So actually it will be good for computer enthusiasts (Intel and AMD) if they could get FX-60 for $500 like Conroe 2.67Ghz.

By DallasTexas on 3/9/2006 2:42:42 PM , Rating: 1
"..I can't understand how some people can be happy with this.."

I applaud your bleeding heart attempt to support AMD, however, at the end of the day, "computer enthusiasts" care about performance and PC adaptability. I don't neccessarily see that category of buyers subscribe to your views that AMD is some PC saviour. You are either drinking too much AMD koolaid on the outside or AMD is serving too much of it in their cafeteria. Clearly, AMD will surely dust off the underdog marketing campaign to cover their now lackluster performance compared to Intel.

The bottom line is 'enthusiasts' and 'gamers', in general, don't care. In fact, what AMD should be concerned about in the bleeding heart department is 90% of the manufacturing is done off shore. At Intel 80% is done in the good old USA. Not that it matters all that terribly, but AMD should be a tad concerned that Germany is "over there" and Idaho, massachussetts, Oregon, New mexico, Colorado, etc etc is "Over here".
AMD should not play that card too strongly and best to revertto crank up the clock at the expense of more heat in or der to compete.

By shabodah on 3/9/2006 12:34:45 PM , Rating: 2
I certainly understand that their is a significant improvement in performance with these new Intel designs, but I have yet to hear a good explanation as to why. It seems the most significant change is how SSE is processed, and seeminly, this would account for the breakthrough performance of the design, but nothing else seems to be updated in a meaningful way. Still FSB bottlenecks are going to be a bandwidth issue with more and more cores, right? Anyone have a in-depth explaination?

RE: Understanding.
By Questar on 3/9/2006 12:48:45 PM , Rating: 2
Serveral things have changed:

1. An additional ALU.
2. A new scheduler to take advantage of the ALU.
3. Macro-ops fusion (a new decoder).
4. Greatly massaged ALU's and FPU.
5. Bigger, better cache.

Basically, it's a new design.

RE: Understanding.
By PT2006 on 3/9/2006 9:34:45 PM , Rating: 2
6. Single cycle SSE1-3, plus new SSE instructions (also single stage)

RE: Understanding.
By JumpingJack on 3/11/2006 4:43:48 AM , Rating: 2
What the dramatic jump...what is different...

Comments below are a nutshell response:

Here is a good link --

The flash demo is oustanding visual description. The revolutionary items I believe are the Macro/Micro ops fusion, think of this as Hyperthreading along a single thread (serial hyperthreading) through each core. Just watch the flash, you will see what I mean.

AMD fanboys
By clairvoyant129 on 3/9/2006 1:21:37 PM , Rating: 2
Now that the BIOS issue has been settled, I wonder what AMD fanoys will think of next?

RE: AMD fanboys
By Burning Bridges on 3/9/2006 1:33:14 PM , Rating: 2
Well, there is always the fact that you are comparing next-gen Intel chips to current gen AMD ones... ;]

RE: AMD fanboys
By JackPack on 3/10/2006 5:25:25 PM , Rating: 2
Generation? AM2 is a refresh.

Yeah, Intel is supposed to get their hands on a heavily NDA'd AM2 and show it to the world.

They overclocked the FX-60 to represent the increase in performance from AM2.

RE: AMD fanboys
By bupkus on 3/9/2006 1:43:18 PM , Rating: 2
Now that the BIOS issue has been settled, I wonder what AMD fanoys will think of next?

"...fanoys"?? OMG, the fanbois have been neutered!

Interesting observations
By zsdersw on 3/10/2006 7:27:08 AM , Rating: 2
I've been reading the comment sections of all the DailyTech articles on Conroe lately and what struck me as being the most prevalent were two things:

- People saying that we shouldn't count AMD out.

- People saying that AMD's offerings are going to be better when Conroe is available.

The funny thing, though, is that if the situations were reversed.. and it was AMD who had demonstrated a very strong future new product.. I highly doubt there'd be as many people saying that we shouldn't count Intel out, nor would there be as many people saying that Intel's offerings are going to be better when AMD's new stuff is released. Double standards? Table for 100+.

The fact of the matter is that while AMD's offerings will be better when Conroe ships (AM2, DDR2, etc.) so will Conroe. That's something I've seen very little mention of in the comment sections of these stories here.

As for Conroe itself, I'm not particularly moved one way or another. Just like I expect with AMD, I expect a new architecture to outperform existing architectures. To me, this is just Intel meeting expectations. Have they exceeded expectations? We'll see when it ships.

RE: Interesting observations
By JackPack on 3/10/2006 5:15:40 PM , Rating: 2
What else can you expect from AMD apologists?

I especially love the one where they claim when 65nm AMD products ship in 1H'07, Intel will take the backseat again. Of course, they fail to mention Intel's 2H'07 45nm schedule.

Sometimes, they even claim 4 GHz K8 will be possible on 65nm, even though thus far 130nm->90nm has only brought AMD 400 MHz (FX-55->Opteron 256).

RE: Interesting observations
By JumpingJack on 3/11/2006 4:38:07 AM , Rating: 2
The funny thing, though, is that if the situations were reversed.. and it was AMD who had demonstrated a very strong future new product.. I highly doubt there'd be as many people saying that we shouldn't count Intel out, nor would there be as many people saying that Intel's offerings are going to be better when AMD's new stuff is released. Double standards? Table for 100+.

No not really, I liken it to "King of the Hill" effect, or the "Underdog Syndrome". It is natural to work hard to bring down number one. You see this everywhere, people live by MacOs why, well mostly because it is a better OS but also alot of Mac devotion is a despise of MS. How about more globally, USA is number one in just about everything, economy especially, we are bitterly despise, why? Well GWB is a very good part of that, but we were never really liked even during WHC. Look at Walmart, etc. etc. etc. the list is endless The human psyche loves an underdog, in some finite percentages people will latch on and even die with that underdog :) ....

RE: Interesting observations
By zsdersw on 3/11/2006 7:10:35 AM , Rating: 2
Indeed. Nothing I talked about was beyond my comprehension.. I just thought it was interesting.. and particularly telling of just how ridiculous many of the posters on here are.

By Sensai on 3/9/2006 12:23:54 PM , Rating: 2
seems like the AMD 64 is now becoming the latter Pentium 4... needs increased clocks to match or close in on performance and the FX-60 burns i think over 110+ watts to run over the 95 envelope 90nm they intend for the future.

RE: ..
By mxzrider2 on 3/10/2006 4:40:18 AM , Rating: 2
remember that amd and intel have a different way of measuring their thermal envolope. amds is an absolute max (theriotical) if i remember and intels is like average max ( real world), that is if i remember correctly. thing is my whole sytem takes less than 200 at idle, and thats with an otpy 144 oced to 2.4 and a 7800gt slightly oced. and less than 275 at full load. ya

RE: ..
By Questar on 3/10/2006 10:09:12 AM , Rating: 2
This is a myth. TDP is measured the same by Intel and AMD.

By unparalleled intellect on 3/10/2006 2:30:39 AM , Rating: 2
What do you think AMD will do about their number schemes? I understand it was meant to be in comparison to their Thunderbird or whatever it was but let's be honest here; it was meant to respond to Intel's GHz lead.

If Intel's chip really performs as advertised, then AMD would be misleading the consumer with a 5000+ or whatever they'll have.

RE: Question
By theprodigalrebel on 3/10/2006 3:34:42 AM , Rating: 2
Well, their next 2.8Ghz dual-core processor would be about 20% slower than Conroe 2.66...(Of course, this is assuming DDR2 does close to nothing to improve performance and the CPU architecture doesn't change much, like most people are saying around the Web)

How about AMD Athlon64 2128+? Maybe if it makes a difference, they can even up the rating to 2200+?;) Then, my A64 2800+ wouldn't look so bad!

RE: Question
By JumpingJack on 3/11/2006 4:32:09 AM , Rating: 2
If Intel's chip really performs as advertised, then AMD would be misleading the consumer with a 5000+ or whatever they'll have.

Good question, but if you really analyze the data out there, you will see Intel 3.0, 3.2, etc...3.6 GHz parts out performing 3800+ and 4200+ processings in many benchmarks, AMD is already "mis-leading" if they tell you "equivalent Intel GHz performance"....
Actually, they have changed their rationale on the numbers I read somewhere (if I find the link I will edit it in), that AMD was quizzed about this, and their response was they saw no reason to change because customers now understand that a 4400+ AMD part is better performing than a 3800+ AMD part. This is quite logical, and actually quite smart, as long as they don't imply "Intel GHz equivalence" any more.

By MrKaz on 3/10/2006 9:36:02 AM , Rating: 2
So you are loosing time looking at what have failed in the overclocked AMD system, instead of:

-Why dont they compare the new product with current offer?
-Where is the P4, the one that will be replaced by conroe?
-Why they changed the drivers of the Ati cards to recognize the processor?
-What would happen if the processor didn’t get recognized?
-Was the results achived by drivers optimizations?
-We are all looking at a huge 4MB cache processor, is the cache that is giving that (big) lead?

By Questar on 3/10/2006 10:15:47 AM , Rating: 2
What is loosing time?

I've lost time, but I've never loosed time.

By MrKaz on 3/10/2006 12:08:45 PM , Rating: 2
Losing time is trying to put the FX60 as a recognized processor, when saying unknown processors doesn’t affect the performance of it, or trying to achieve a little performance gain out of an already OVERCLOCKED processor, instead of seek out the conroe performance, how it was achived, and why!

Is it really that GOOD? I have little doubts, but where are the synthetic benchmarks, the Geforce 7 SLI, … even bother to test it with Ati REAL Catalyst 6.2 or 6.3 drivers and not the Intel supplied ones?

By BillyBatson on 3/9/2006 11:24:07 AM , Rating: 2
"The BIOS had little to know affect on performance"


RE: ...
By Xchus on 3/9/2006 12:02:45 PM , Rating: 2
I only can add that thank god Intel is up in its game again otherwise we would be running sucky hardware until the day we die, now at least we can see some competition and perhaps see more significant technological improvements come out faster, I mean who doesnt want to run a Quad-Core DDR3 based computer. LOL

More Understanding
By shabodah on 3/9/2006 1:02:13 PM , Rating: 2
So, if these structural changes would have also included a memory controller on-die and usage of a 1500 mhz HT, what type of performance would we now be seeing?

RE: More Understanding
By Questar on 3/9/2006 4:18:32 PM , Rating: 2

Nobody knows what those features do for AMD.

By bupkus on 3/9/2006 1:57:02 PM , Rating: 2
When there are 2 top dogs fighting it out the best scenario is for them to leapfrog eachother again and again. Just like ATI and nVidia, it's the best case possible for the consumer.
If one were to dominate for too long it would relagate the 2 spot to discount status allowing #1 to hold back and better manage it's R&D/Price ratio.
With leapfrogging as in this case, Intel couldn't hold anything back. They had to beat AMD or continue the impression they were becoming 2nd tier for desktop and perhaps other markets.
Doesn't anyone remember how ATI and nVidia have gone at it and still do?
When #1 leads for too long they then try to manage their profit margin and milk their lead as they can. This invites #2 to try to leapfrog them and then do the same. No, this model is a fact of economics and probably nature as well. AMD will not disappear unless they implode internally. Then someone else will emerge to take their place.

RE: Leapfrog
By bamacre on 3/9/2006 2:17:37 PM , Rating: 2
Exactly. With the Dell contract, Intel could continue putting forth a sub-par cpu, and on the other hand, AMD only had to stay one step ahead of Intel to gain in the retail market and in the enthusiasts market. This just went on way too long, as far as consumers are concerned. It was just a whole different ballgame for Nvidia and ATI, and the tougher competition meant much better products and prices for consumers.

I am no big fan of either AMD nor Intel, I always go by what's best for me, performance/$. I actually run an Intel currently, because I gotta a helluva deal on a Dell. But we really have AMD to thank for this new line of Intel cpu's, because if they didn't do such a good job of competing with Intel, we'd never see this. It's about time we see capitalism and competition actually work in the cpu market. As others have already said, this is good news for all consumers.

I also think it's good news for Dell as well. One of the only things holding Dell back from being an aggresive option for true gaming systems was the lack of AMD. With Intel now holding the performance crowd, Dell's sales to gamers and some enthusiasts should increase significantly. However, no way am I saying that BYO'ers would buy a Dell, no, but sales from those seeking prebuilts with higher performance should increase noticably.

AMD Intel Deathmatch
By PrimarchLion on 3/9/2006 2:02:56 PM , Rating: 2
I wonder if Intel is going to follow up on that challenge AMD made a few months after they release Conroe. I think if these benchmarks are for real, at least Athlon is more competitive with Conroe than P4 was with Athlon.

RE: AMD Intel Deathmatch
By DallasTexas on 3/9/2006 2:22:39 PM , Rating: 1
"..I wonder if Intel is going to follow up on that challenge AMD made a few months after they release Conroe..."

I don't see Intel stooping to AMD's juvenile challenge where nobody showed up and they declared themselves the winner. LOL

The real question is will AMD reopen the challenge or is the reason they put an expiration date of December 31, 2005 now makes sense?

By Motley on 3/9/2006 11:22:49 AM , Rating: 2
Shouldn't it read "The BIOS had little to no effect?"

By latino666 on 3/9/2006 12:08:50 PM , Rating: 2
Lets hope AMD can respond. Competition rocks!!!

Great job for Intel
By solgae1784 on 3/9/2006 12:49:59 PM , Rating: 2
Seems like Intel has finally done it with their new processors. This deserves many "good job well done" to Intel, even from a AMD fan like me.

Do know that AMD won't be standing still, either. And this kind of competition benefits us customers.

It's about time.
By exdeath on 3/9/2006 1:36:04 PM , Rating: 2
Wow this may be my first Intel CPU since the original Athlon Slot A 700 Mhz.

AMD has held the performance crown since then, loosing it briefly to Intel when the P4 crossed 3 Ghz, until the K7s arrived.

"Spreading the rumors, it's very easy because the people who write about Apple want that story, and you can claim its credible because you spoke to someone at Apple." -- Investment guru Jim Cramer
Related Articles
Conroe Benchmark Criticisms
March 8, 2006, 4:45 PM

Copyright 2016 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki