AMD INTERVIEW POSTED

[9-inch]

Damon Muzny: Intel said Netburst would take them to 10GHz. You know, we can both fab some pretty nifty one-off products that may never see mass production. The proof is in the pudding, as they say. When can you bring a product to market competitively? Making off-the-wall claims 5 years out isn't leadership.

Netburst was ultimately too inefficient and so the architecture was finally scrapped. Intel began scrambling to bring up a "plan-B," but that takes time. So meanwhile one way Intel attempted to compensate for not being competitive with Direct Connect was by increasing their cache sizes. We know that didn't get them to performance parity. What they got was an increase in die size with performance that still notched them down the value ladder when lined up to next to AMD chips. So while both AMD and Intel are both headed to 65nm, 45nm and beyond... Intel's complex and inefficient architecture with large "compensation caches" made 65nm necessary to bring die sizes back down on Pentium. And this is just one aspect of the big picture. So when you hear the competition crowing about starting a new, smaller design process sooner, understanding the big picture gives a more clear view of what is really going on.

Here's another good example. Intel has been trying to take the focus of their current product portfolio with Conroe demonstration stunts. So while I can't really comment about a competitor's product probably which won't be available for 6 or 9 months, I will say this much about their tactics: it's not like their typical "AMD who?" approach. Clearly we've forced our competitor to respond us. Every time they stress future products they are simply admitting their current products aren't competitive so customers shouldn't buy them. That's just an irrational and desperate move for them. Their claims of future architectural competitiveness largely assume AMD is standing still, and we're not. In fact, as they work to catch up to us, we'll move ahead with our own innovations. So the big picture is Intel in crisis-mode, reacting to AMD's leadership. It shouldn't come as a surprise.

www.amdzone.com/modules.php?op=modload&name=Sections&file=index&req=viewarticle&artid=251&page=1

 

[Action_Man]

 

[Action_Man]

So while both AMD and Intel are both headed to 65nm, 45nm and beyond... Intel's complex and inefficient architecture with large "compensation caches" made 65nm necessary to bring die sizes back down on Pentium 4.

 

[SidVicious]

So while both AMD and Intel are both headed to 65nm, 45nm and beyond... Intel's complex and inefficient architecture with large "compensation caches" made 65nm necessary to bring die sizes back down on Pentium 4.

Heheh ! Love that one !

 

[spud]

So while both AMD and Intel are both headed to 65nm, 45nm and beyond... Intel's complex and inefficient architecture with large "compensation caches" made 65nm necessary to bring die sizes back down on Pentium 4.

Word.

 

[gr8mikey]

9-Inch, my phallic friend, I was reading over the interview then it hit me -- I noticed while trying to study up to match you're dizzying depth of knowledge of processor making, I was hoping you could help me answer a question....

I noticed that some people call silicon Si(100), what the heck does that (100) mean and why is it important?

I eagerly await your answer.....

Jack

Jack you need to quit! Yer killin me! I can't breathe!

 

[Legenic]

So while both AMD and Intel are both headed to 65nm, 45nm and beyond... Intel's complex and inefficient architecture with large "compensation caches" made 65nm necessary to bring die sizes back down on Pentium

AMD is just as bad as their diehard fans. that's the one thing that really gets me about them, they're so arrogant and constantly call down their competition. they've never heard the saying "don't talk bad about your competition"? intel seems much more collected. I've only heard them talk bad about AMD once (correct me if there's more, there may be), and that was to do with the anti-trust suit.

Intel has been trying to take the focus of their current product portfolio with Conroe demonstration stunts. So while I can't really comment about a competitor's product probably which won't be available for 6 or 9 months, I will say this much about their tactics: it's not like their typical "AMD who?" approach. Clearly we've forced our competitor to respond us

I agree with this. AMD has forced things out of intel for sure. and I also agree it was somewhat foolish to talk up conroe so early. but at the same time, with the p4 being more and more outrun by the A64, intel had to convince investors and enthusiasts that there is light at the end of the tunnel.

on the one hand, it's not smart to talk so much about future tech when it's all about how much better it is than what's out now. but on the other hand, who reads their comments? investors, who will stick with them because they have a good product coming, and enthusiasts, who may very well avoid AMD now (since they're the best choice for performance currently), and wait to buy future intel chips.

also, they're not 6 to 9 months away (unless this article is old, I didn't click the link). intel is launching core in Q3, maximum 5 months away, minimum 3.

their claims of future architectural competitiveness largely assume AMD is standing still, and we're not. In fact, as they work to catch up to us, we'll move ahead with our own innovations.

fair enough. absolutely they're not standing still. but at the same time, intel isn't playing catchup. if conroe performs like it's been show so far (and there actually is a decently large amount of conroe performance info floating around), then intel is very much passing AMD. AMD's gains may play catchup to intel.

 

[RichPLS]

Actually, Intel was catching up with the 800 series and caught up with AMD with the 900 dual core series which run roughly equal, but have much more overclocking potential, and now Conroe will be swooshing past them in price, perfromance and thermal...

 

[chime]

Jack,

You should change your name to Professor4AMDfanboi. :lol: :lol:

Really enjoy your posts (really educational) vs. the inquirer posters.

Keep up the effort.

Thanks.

 

[butitoy]

So while both AMD and Intel are both headed to 65nm, 45nm and beyond... Intel's complex and inefficient architecture with large "compensation caches" made 65nm necessary to bring die sizes back down on Pentium

AMD is just as bad as their diehard fans. that's the one thing that really gets me about them, they're so arrogant and constantly call down their competition. they've never heard the saying "don't talk bad about your competition"? intel seems much more collected. I've only heard them talk bad about AMD once (correct me if there's more, there may be), and that was to do with the anti-trust suit.




TRUE!

these people think they are the only ones who had the best proccessor.

in real world situation..its not just games.
in graphics world. my INuTEL3.2ghz @3.52 beats my venice A643200 in proccessing power. photoshop and sorts.
but reign in 3d games.
so what?..

both had strength and weakness.

fanboyism sucks!! don't they know that?

or are they just defending they're ILL decided buy?

IN PANIC?
pity

 

[endyen]

Actually, Intel was catching up with the 800 series and caught up with AMD with the 900 dual core series which run roughly equal, but have much more overclocking potential,

Do you really believe the 965 is equal to the FX60?
I did a tally of the cpu chart. Of the 28 benchmarks (price/perf not included as no price for the 965), the 965 won 12, while the FX only won 16.
I guess that sounds close right? Then again, since 10 of the 965s wins were from the 10 synthetic benchmarks, that paints a different picture.
Let me put it this way. Of the real life benchmarks, the score is intel-2, Amd-16. Who wins?
As to OCing, even THG got the FX to 3 ghz. Sure, that's only 15%, but who is going to overclock their extremely expensive past 4.3ghz?

 

[hashv2f16]

Keep in mind the very small margin at which SOME of the AMD chips are "better" by. That's all they need to be "the best". It seems being only "almost as good" (1 or 2 FPS, 1-3 second less encoding time) just doesn't cut it at all..

And some of you AMD guys out there who call yourself "overclockers", you're not real overclockers - unless you buy intel. that is all

 

[Rustol3um]

And some of you AMD guys out there who call yourself "overclockers", you're not real overclockers - unless you buy intel. that is all

ya right !!!! whatever guy......glad to see you know what you are talking about..

 

[hashv2f16]

And some of you AMD guys out there who call yourself "overclockers", you're not real overclockers - unless you buy intel. that is all

ya right !!!! whatever guy......glad to see you know what you are talking about..

I may not know what I am talking about, but I know that the Intel 65nm stuff o/c's like a bat out of hell.

amd's so called "high-end" FX series only overclock what. 200-400mhz?? that's redicuous eh?

 

[gOJDO]

i will try to find a brain-washing machine picture, it will look just fine in the combo :lol: 8)

 

[pat]

Actually, Intel was catching up with the 800 series and caught up with AMD with the 900 dual core series which run roughly equal, but have much more overclocking potential, and now Conroe will be swooshing past them in price, perfromance and thermal...

Problem with Intel's 900 serieis, even if it OC well, and that OC may represent 99% of all poster on this forum, in real life, it apply to less than 1% of all computer sold out there...

So, they are still behind IMHO..

I don't meant that they are bad, but a bit outdated. Hopefully, new one will be more up to date, when they'll be available..

 

[spud]

Actually, Intel was catching up with the 800 series and caught up with AMD with the 900 dual core series which run roughly equal, but have much more overclocking potential,

Do you really believe the 965 is equal to the FX60?
I did a tally of the cpu chart. Of the 28 benchmarks (price/perf not included as no price for the 965), the 965 won 12, while the FX only won 16.
I guess that sounds close right? Then again, since 10 of the 965s wins were from the 10 synthetic benchmarks, that paints a different picture.
Let me put it this way. Of the real life benchmarks, the score is intel-2, Amd-16. Who wins?
As to OCing, even THG got the FX to 3 ghz. Sure, that's only 15%, but who is going to overclock their extremely expensive past 4.3ghz?

Look, you gotta go with Endyen here, when you encompass the entire data pool, K8 beats Netburst, pretty handidly for raw compute power for most items. Multimedia/video maybe 965, but most anything else??

Price/Performance well that may end up different in the near term, but at this level a hundred bucks or so is not that much different.

Word.

 

[GherkinPekul]

sphincter

 

[RichPLS]

Yonah is available and performs excellent on 65nm cool and overclockable extreme... and same technology conroe is based on... sort of a preview but shipping Conroe will be much faster still

 

[Caboose-1]

sphincter

Choad.

 

[turpit]

So while both AMD and Intel are both headed to 65nm, 45nm and beyond... Intel's complex and inefficient architecture with large "compensation caches" made 65nm necessary to bring die sizes back down on Pentium 4.

ROFLMAO. That was toooooo funny. Did you do that yourself, or find out in the ether?

 

[weskurtz81]

From my experience... if you aregoing to buy a dual core chip... do not buy intel.... yet. X2's and the Opty's are faster in almost everything, including encoding and other video tasks.... that is from personal use I say this....

 

[weskurtz81]

Jumpingjack.... you ever run SuperPi on a dual core cpu? There is a reason you can't compare a single core test reult of super pi to that of dual core and call it legit..... at least from the testing I have done. One instance SuperPi will not run both cores to 100%. When I run superpi it only runs each core to 50%. So trying to compare a single core yonah, to the most expensive consumer based amd cpu isn't isn't apples to apples. See what I am saying?

 

[weskurtz81]

seriously though.... if there is a way that you can make it run an utilize both core I would like to know... cause when I watch the cpu usage when I run it, it only uses 50%. Unless I run two instances..... any thoughts?

 

[weskurtz81]

come on Jack.... I understand the statement is funny. But I guess it is quite an accomplisment to be clock for clock as fast as a K8. I already knew why PI would not run on both cores, the same as prime.... but I was not sure why you were making that statment. So in games and single threaded apps it fast, how about encoding and other related apps. That is what I am truly interested in. And I am not making an argument for buying an FX-60, I would NEVER buy one even if it fit nicely into my budget. What they could have done was compare it to a 2gz A64 or opty... would have been just as easy.... but the whole point of the article was to slant it towards Intel. If they wanted to prove clock for clock it was as fast as K8, it could have been easily done. I can clock my opty to 2.8, granted not 3, but you can compare that. Just a rather pointless article, and people look at it and think it's soo fast. And it is in single threaded apps, but I on a day to day use basis notice the difference of using a dual core cpu, and would never switch back to single core if I don't have to..... that is pretty much it..... I am sure you understand.