AMD's response

[Artanis]

Yes, rendering is another story, which can be paralelised fine. But, for most CAD modelling work wich consist for example in computation (in complex 2D sketches), updating features and associative drawings, or managing large 3D assemblies with their constraints, a multi-core CPU hardly helps...

Now back ontopic and wait for AMD's response. Cause I do

 

[Duvie]

Originally posted by: Artanis
Yes, rendering is another story, which can be paralelised fine. But, for most CAD modelling work wich consist for example in computation (in complex 2D sketches), updating features and associative drawings, or managing large 3D assemblies with their constraints, a multi-core CPU hardly helps...

Now back ontopic and wait for AMD's response. Cause I do

CAD modeling is like I said more of a product of the video card then cpu anyways....most things in 2d sketching hardly stress the cpu.....


I agree back on topic....

Second part should arrive tomorrow.... You got to admit quad core would be huge to the market they already lead in....plus it has been their MO of recent...supply the server/business/worktation market first...tat is where a majority of the moeny is anyways...Just look at the cost of a 800's opteron....

 

[Artanis]

Yes, but for the average user it would be even more important if the software developers will build better and more multithreaded apps. Much more important...so your future dual or quad core CPU won't keep idle most of the cores most of the time

 

[Madwand1]

It's really great to have a response from AMD's chief of sales and marketing. This way we know that we're getting the in-depth technical accuracy without a preponderance of spin about the competition. Isn't it?

Edit: Typo.

 

[kknd1967]

emmm, get technical data from marketing people?

Originally posted by: Madwand1
It's really great to have a response from AMD's chief of sales and marketing. This way we know that we're getting the in-depth technical accuracy without a preponderance of spin about the competition. Isn't it?

Edit: Typo.

 

[SlitheryDee]

"Then their new-generation micro-architecture (NGMA), is, quite frankly, a quick fix on the front-side bus. I don't think that's the future of the Intel architecture. I think it's another quick fix until 2008 or later, when they're going to come out with a genuinely new architecture."

I dunno about you guys but this line rather worried me. He thinks that intel will only rely on Core for a couple of years? Judging from the benchmarks it may take AMD almost that long to catch up with Conroe in performance, then they have to square off against a "genuinely new architecture"? This is coming from AMD's chief of sales and marketing?

 

[dmens]

It's really great to have a response from AMD's chief of sales and marketing. This way we know that we're getting the in-depth technical accuracy without a preponderance of spin about the competition.

LOL that *really* cracked me up. In fact, the general attitude on this thread towards that interview is hilarious. That guy's only job is to create spin. He sounded like he was reading off simplified flipcards created by his gang of "technical liasons", i.e. minions. Tech marketing guys... they're all the same in every company.

I cannot imagine the same reaction on AT if Eric Kim gave a similar interview... which leads to an obvious conclusion.

 

[3chordcharlie]

Originally posted by: dmens

It's really great to have a response from AMD's chief of sales and marketing. This way we know that we're getting the in-depth technical accuracy without a preponderance of spin about the competition.

LOL that *really* cracked me up. In fact, the general attitude on this thread towards that interview is hilarious. That guy's only job is to create spin. He sounded like he was reading off simplified flipcards created by his gang of "technical liasons", i.e. minions. Tech marketing guys... they're all the same in every company.

I cannot imagine the same reaction on AT if Eric Kim gave a similar interview... which leads to an obvious conclusion.

Yeah he really said nothing of interest to mainstream consumers. The fact the AMD is well positioned to offer many-cored systems at a less than enourmous price is of very little interest to me.

The fact that I never buy top-of-the-line anything and AMD may be forced to sell me a $150 X2 come september sounds great though.

 

[IdaGno]

Originally posted by: chilled
I'm just relieved AMD are doing something...

I just hope it's too little & too late. Some high-end FX series cpu's at less than half todays prices would make me happy.

 

[Zebo]

what a con man. I cant stand PR peeps. Assuming this conroe/meron stuff was 100% legit...Bottom line is unless AMD makes a new core with Higher IPC they are going to get pwned. Quad core can't do it since Intel can quad easier since they are cooler and smaller (eg cheaper) and the lead gets exponetial when you add cores so it would actually work against AMD to start adding cores. (remeber how extra crappy prescott looked in smithfeild when X2's rolled, now switch places)

In fact, Intel themselves recognize that at some point they are going to have to move to an integrated memory controller. One of their scientists has just acknowledged that publicly.

Really? have to? I saw ~20% faster clock for clock w/o one...
AMD guy better hope intel does'nt add a mem controller either.

"AMD's response" = $50 chips again.. I for one can't wait

 

[potato28]

Originally posted by: Zebo

Really? have to? I saw ~20% faster clock for clock w/o one...
AMD guy better hope intel does'nt add a mem controller either.

"AMD's response" = $50 chips again.. I for one can't wait

Intel aint adding a memory controller for at least 2 years... too expensive for them atm.

 

[SlitheryDee]

Looks like it's better than 20% clock for clock. Remember, that FX60 was clocked 140mhz over the 2.66 Ghz Conroe and anandtech recorded an average of ~24% advantage. I think it would have been better to compare 2 equally clocked processors (slightly overclock the FX60 to Conroe's speed). The results would have been more impressive and it would give us a much clearer (or at least more immediate) picture of where the A64 architecture stands on a clock for clock basis.

I don't find the fact that Conroe outperforms current processors to be surprising in the least, it's supposed to. What I do find disturbing is that AMD seems to have nothing revolutionary in the works to answer this...other than the elusive K8L which is in all likelihood a year or more away.

 

[Arachnotronic]

I really hope AMD can close this clock for clock performance gap. They've held the lead for so long, I don't know why they can't continue it!

 

[Shortass]

Originally posted by: Intel17
I really hope AMD can close this clock for clock performance gap. They've held the lead for so long, I don't know why they can't continue it!

Strange, coming from a member with the name of Intel17...

 

[AnandThenMan]

Originally posted by: Zebo
Really? have to? I saw ~20% faster clock for clock w/o one...
AMD guy better hope intel does'nt add a mem controller either.

"AMD's response" = $50 chips again.. I for one can't wait

Give me a break. If you seriously are going to take Intel sanctioned benches as the gospel truth, and decide because of the benches that AMD is going to be soon selling "$50 chips" then you are a much, much bigger "con man" than you claim the AMD rep is.

Dunno what your deal is lately, but you used to be a rational voice around here, now anymore you seem to be jumping on the Conroe bandwagon and dissing AMD. What gives?

Here's a thought. Let's actually WAIT until AM2 shows up, and wait some more until Conroe is a real product you can actually buy before we all predict the demise of AMD.

 

[DrMrLordX]

I, for one, welcome my next $50-chip overlords.

On a serious note, AnandThenMan, I think Zebo predicted a smackdown on AMD months ago so long as AMD rested on their laurels. This marketting-drone AMD sent out to hype future products isn't really saying much that makes me think Zebo to be wrong. Obviously, the reasonable approach is to wait until these future products are actually available, but AM2 just isn't looking that hot. Hell, people were wondering if AM2 would ever significantly outperform s939 not too long ago . . . for overclockers, it may wind up being worse.

 

[Zebo]

AMD2 wont do anything for AMD - as far as fakeing benchmarks I'm with you.. notice I said "assuming these benchmarks are true" before begining? Do I need to spell it out for you? Ok - I don't like these lame reviewers were not allowed to even look inside box, gives me idea someones hiding something... It's also trivial to fake these CPUID's and properties on screen, theres long posts at xtreme about it and with very talented people working the intel they could do it to perfection especially on a major tools like CPUZ as of yet unidetifiable CPU like conroe. Hell they could have had dual opterons or a 3.6 GHZ conroe in the intel box for all know....But again assuming they are true and intel has'nt continued thier dispicable/deperate behavior along the lines alleged in their anti-trust lawsuit - AMD's in big trouble. Happy now?

edit: OH BTW $50 was very good times.. don't hate on it..You could buy a tbred 1700 for $48 and make into a very fast desktop cheap.. call it wishful thinking. Or you like paying $300 for a $20 peice of silicon?

 

[XBoxLPU]

Conroe isn't the end of AMD

 

[dexvx]

Originally posted by: Zebo
edit: OH BTW $50 was very good times.. don't hate on it..You could buy a tbred 1700 for $48 and make into a very fast desktop cheap.. call it wishful thinking. Or you like paying $300 for a $20 peice of silicon?

The silicon may cost $20, but the R&D Q&A to get to where it is adds much more to the price.

As for the integrated memory controller, its usually used by people who don't understand hardware basics as jingo to make them sound superior. Its been shown and proven time and time again that you do not need an integrated memory controller to gain performance. Alpha and Sun, AFAIK, abandoned their integrated memory controller years ago.

 

[PentiumIV]

Originally posted by: Link

Originally posted by: BrownTown
The problem is that they are jsut updating the old core next year whereas Intel has a compleyelt new core, so the improvements AMD gets won't be nearly as signifigant as Prescott to Conroe will be.

Well, I don't think Conroe isn't exactly a new core, either. Correct me if I'm wrong, but didn't Conroe/Yonah evolve (updated) from P-M, and P-M is an offspring of the tweaked P3 core?

More-or-less true, but there are VERY significant changes in Merom(Conroe-desktop version of it) compared to Yonah.

 

[darkdemyze]

Originally posted by: XBoxLPU
Conroe isn't the end of AMD

Let's assume Conroe is better than anything AMD has atm:
The failure of Prescott didn't ruin Intel, I doubt the success of Conroe will ruin AMD. Now granted Intel is much larger and has been around much longer, but AMD will bounce back. Desktop market isn't the only thing pulling in AMD's revenue.

 

[Keysplayr]

Of course AMD will bounce back. They will most likely copy Intel with a 4 issue wide CPU themselves. Or, AMD could make an "infamous right turn" of their own, repeating Intels Netburst mistake. You never know.

You mention that the desktop market is not the only thing pulling in AMD's revenue. You are very correct, but, Intel is releasing these Core MPU's platform wide. Mobile (Merom), Desktop (Conroe) and Server platforms (WoodCrest and ULV Woodcrest I guess for blade applications).

Anyone who says AMD is going under is just being foolish. However, if Core turns out to be what it seems to be, it is going to hurt AMD's pockets tremendously. If they can't compete in speed, thermals and price, and as of today, is does not look good for them, AMD will have to heavily, and I mean need to HEAVILY cut prices on the order of about 50 to 70%.

We will see in about 4 months. I am anxious to know the facts.

 

[Stoneburner]

You know, when people look at the r480 board vs whatever intel board was using crossfire, nobody seems to discuss the fact crossfire has not been very efficient so far, or has not shown the same percentage gains as SLI. Part of the gaming performance advantage might just stem from the fact whatever chipset and board intel was using implemented crossfire much better. I think in the end INtel's performance lead even in games will be closer to the media tests the various websites ran.

 

[darkdemyze]

Edited to first post.

 

[dexvx]

Originally posted by: Stoneburner
You know, when people look at the r480 board vs whatever intel board was using crossfire, nobody seems to discuss the fact crossfire has not been very efficient so far, or has not shown the same percentage gains as SLI. Part of the gaming performance advantage might just stem from the fact whatever chipset and board intel was using implemented crossfire much better. I think in the end INtel's performance lead even in games will be closer to the media tests the various websites ran.

Why? Because Rahul said so?

Go search any RD480 vs RD580 benchmarks. With mature drivers, the RD480 is a whopping 1-3% behind the RD580.