Conroe debunked

[cyks]

I just love it when people argue over stuff that is not out yet and that their future, bigger brother will beat up the future big brother of the other side.

Yea, and I love it how it is the same ones doing it every time.

 

[Jason711]

why didnt i see this thead TWO DAYS ago?? .. damnit.

 

[DejaWiz]

Donnie, you're talking overclocking comparisons.....

In the real world, overclocking is never a guarantee for any system. It's just a bonus for those of us that can accomplish it with a great degree of success. It's an unfair comparison, because every single processor will overclock differently. Just because I can overclock my RevE 3000+ to 2.4+ GHz with almost no increase in Vcore and with the stock hsf doesn't mean everyone out there with a RevE 3000+ will accomplish that. Many will get more, many will get less. And just because a reviewer says with confidence that all 920D's should reach 2.9GHz on air doesn't mean they all will by any means.

I'm talking about strictly non-overclocked comparisons....The way they perform out of the box with no tweaking. A majority of users out there aren't overclocking their Dells and HPs (the bulk of cpu sales and computers in use), and that sets the real benchmark for cost/performance in real-world testing...non-overclocked.

 

[coldpower27]

Well, let's look at this in current tense, since Conroe isn't available yet.

AMD X2 3800+ is about $295 avg price today. In all the results from tests performed by independent reviewers, this particular AMD chip can even beat out intels currently available top-of-the-line EE dual core chip (not to mention runs cooler and draws a lot less power). So from looking at the other side of the economic coin, spending $300 on an X2 3800+ is better than spending $1000+ on the intel EE for similar performance.

However, like you've repeatedly said and I repeatedly agree, this will surely be the inverse when Conroe is released. Spending $300 on a cpu that can outdo AMD's $1000+ top-of-the-line would be a very wise choice indeed!.

Then again, when Feb-Apr of next year arrives, I will get to compare and choose between either intel's or AMD's 65nm offerings. We don't know how future prices, availability, and comparable performance will be from either company yet. But like I said earlier, I may only have to upgrade my GPU(s) to see a decent performance gain in games.

Donnie, you're talking overclocking comparisons.....

In the real world, overclocking is never a guarantee for any system. It's just a bonus for those of us that can accomplish it with a great degree of success. It's an unfair comparison, because every single processor will overclock differently. Just because I can overclock my RevE 3000+ to 2.4+ GHz with almost no increase in Vcore and with the stock hsf doesn't mean everyone out there with a RevE 3000+ will accomplish that. Many will get more, many will get less. And just because a reviewer says with confidence that all 920D's should reach 2.9GHz on air doesn't mean they all will by any means.

I'm talking about strictly non-overclocked comparisons....The way they perform out of the box with no tweaking. A majority of users out there aren't overclocking their Dells and HPs (the bulk of cpu sales and computers in use), and that sets the real benchmark for cost/performance in real-world testing...non-overclocked.

Care to link comparisons that have the Athlon 64x2 3800+ beating the Pentium Extreme Edition 955 in all tests??? Stock of course, but I haven't seen where this is the case..

To me it strictly depends on the benchmark being run...

 

[DejaWiz]

Care to link comparisons that have the Athlon 64x2 3800+ beating the Pentium Extreme Edition 955 in all tests??? Stock of course, but I haven't seen where this is the case..

To me it strictly depends on the benchmark being run...

It is strictly dependant on the benchmark being run. I was recalling an article at TechReport about the X2's vs the 840XE.

But, I did find an article there about the 955XE vs the FX60. Using the POV-Ray Rendering test, the X2 3800+ scores slightly better than the 955XE.

X2 3800+ w/6800 Ultra and 1GB Mem
955XE w/7800GTX and 2GB Mem

All the other benchmarks seem to have severe variations, so I did not include those.

Man, I can't wait to see the Conroe benchmarks from these guys! Should be very interesting indeed.


[EDIT]
Found a comparison between the X2 3800+ and 965XE. Very surprising.

 

[coldpower27]

It is strictly dependant on the benchmark being run. I was recalling an article at TechReport about the X2's vs the 840XE.

But, I did find an article there about the 955XE vs the FX60. Using the POV-Ray Rendering test, the X2 3800+ scores slightly better than the 955XE.

X2 3800+ w/6800 Ultra and 1GB Mem
955XE w/7800GTX and 2GB Mem

All the other benchmarks seem to have severe variations, so I did not include those.

Man, I can't wait to see the Conroe benchmarks from these guys! Should be very interesting indeed.


[EDIT]
Found a comparison between the X2 3800+ and 965XE. Very surprising.

Yes which invalidates your statement, as the 955XE does win some tests against the X2 3800+. I think that was the issue that Donnie had with you before... Though it is major feat that the X2 3800+ is as competitive as it is with the Presler EE 955/965 but wins all test against it would be a distortion of reality.

 

[DejaWiz]

Yes which invalidates your statement, as the 955XE does win some tests against the X2 3800+. I think that was the issue that Donnie had with you before... Though it is major feat that the X2 3800+ is as competitive as it is with the Presler EE 955/965 but wins all test against it would be a distortion of reality.

I can admit when I am wrong. I was wrong. It doesn't beat it in every single test. But damn, a 2GHz dual core within the same performance proximity of 3.73 GHz dual core is pretty damn impressive nonetheless.

Like I said, I can't wait for Conroe benchmarks from TechReport. If all that has been speculated is true, then we will all see one hell of an ass-whopping!

I hope you or Donnie don't regard me as an AMD fan-boi. To clear things up: I run systems with both brands. My system in sig for gaming, my dual PIII-1000 for remote access to work. The dually is still running strong and I haven't run into any instances of it requiring an upgrade. Oh yeah, and my laptop is a P4M 1.6GHz. Good, good stuff in those!

Which is why I am anxious to see which will be the better buy come Feb-Apr next year, not right when Conroe is released. I have faith in both companies (which many here seem to lack....but I could be wrong)

 

[coldpower27]

I can admit when I am wrong. I was wrong. It doesn't beat it in every single test. But damn, a 2GHz dual core within the same performance proximity of 3.73 GHz dual core is pretty damn impressive nonetheless.

Like I said, I can't wait for Conroe benchmarks from TechReport. If all that has been speculated is true, then we will all see one hell of an ass-whopping!

I hope you or Donnie don't regard me as an AMD fan-boi. To clear things up: I run systems with both brands. My system in sig for gaming, my dual PIII-1000 for remote access to work. The dually is still running strong and I haven't run into any instances of it requiring an upgrade. Oh yeah, and my laptop is a P4M 1.6GHz. Good, good stuff in those!

Which is why I am anxious to see which will be the better buy come Feb-Apr next year, not right when Conroe is released. I have faith in both companies (which many here seem to lack....but I could be wrong)

To me the fact that a K8 Dual Core 2GHZ processor that was designed with multi core in mind and HIGH IPC rather then pure clock scaling being in the same neigborhood as 3.73GHZ of a NetBurst Architecture that was designed for high clock frequency at the expense of actual performance as well as never intended to be in a Dual Core package hardly that impressive at all.

The architectures had clearly different goals in mind and it shows very clearly.

I can't wait for benchmarks of Conroe either from all kinds of sites such as TechReport, Anandtech, FiringSquad... etc etc.. The more the better to correlate results.

It's good that your using products from both vendors, though to me owning products from both camps doesn't stand as evidence of being unbiased, I evaluate that aspect by what a person says and how a person acts when presented with different information. I have witnessed a person owning both products showing bias for one particular group before so this statement hold little water with me.

Oh I have faith in AMD to survive, it's just that they will have to live without the performance crown on the desktop for sometime, I don't expect them to catchup that soon, this isn't the GPU sector when new generations of products arrive in relatively quick succession behind one another.

Of course but keep in mind while for you, your upgrade cycle and some others, doesn't arrive till the period you have stated, for just as many others their upgrade cycle will come when Conroe arrives and it would be those peoples best interest to go with Intel if the performance indications we have seen hold to be true. And there are also the Intel loyalists that have been holding out till Intel was competitive once again that will be upgrading when Conroe arrives as well.

 

[DejaWiz]

I can't wait for benchmarks of Conroe either from all kinds of sites such as TechReport, Anandtech, FiringSquad... etc etc.. The more the better to correlate results.

And this is exactly the information we all should be waiting for before creating threads to either put Conroe on a pedestal or piss on it.

It's good that your using products from both vendors, though to me owning products from both camps doesn't stand as evidence of being unbiased, I evaluate that aspect by what a person says and how a person acts when presented with different information. I have witnessed a person owning both products showing bias for one particular group before so this statement hold little water with me.

I hope you've been paying attention to my previous posts, then. "Can't wait for Conroe benchmarks, can't wait for Conroe benchmarks, should be interesting, will probably be an ass-whooping, etc..."
I go with what is best for the intended purposes of my system at the time when I purchase. Mobile: P4M, Gaming: AMD, Dually: PIII-1000.

Oh I have faith in AMD to survive, it's just that they will have to live without the performance crown on the desktop for sometime, I don't expect them to catchup that soon, this isn't the GPU sector when new generations of products arrive in relatively quick succession behind one another.

But my argument here is "how long" will they have to live without it? Could be a few months, could be a few years. No one knows until both companies have 65nm offerings available to the public and they are fairly compared. Many seem to be stating that Conroe will kick the crap out of anything and everything for some time to come. It might be true, it might not be. I'm willing to wait until both companies have released comparable technolgy before I make assumptions and post FUD.

Of course but keep in mind while for you, your upgrade cycle and some others, doesn't arrive till the period you have stated, for just as many others their upgrade cycle will come when Conroe arrives and it would be those peoples best interest to go with Intel if the performance indications we have seen hold to be true. And there are also the Intel loyalists that have been holding out till Intel was competitive once again that will be upgrading when Conroe arrives as well.

I absolutely agree that when Conroe first arrives, it will be the better choice from a price/performance standpoint derived from what we have seen so far (I never said it wasn't?). I just so happen to be setting my upgrade cycle at a time that is logical and convenient for myself: when I get a bit of money thrown into my bank account about the same time that the newest 65nm tech will be available from both companies.

 

[Donnie27]

Donnie, you're talking overclocking comparisons.....

In the real world, overclocking is never a guarantee for any system. It's just a bonus for those of us that can accomplish it with a great degree of success. It's an unfair comparison, because every single processor will overclock differently. Just because I can overclock my RevE 3000+ to 2.4+ GHz with almost no increase in Vcore and with the stock hsf doesn't mean everyone out there with a RevE 3000+ will accomplish that. Many will get more, many will get less. And just because a reviewer says with confidence that all 920D's should reach 2.9GHz on air doesn't mean they all will by any means.

I'm talking about strictly non-overclocked comparisons....The way they perform out of the box with no tweaking. A majority of users out there aren't overclocking their Dells and HPs (the bulk of cpu sales and computers in use), and that sets the real benchmark for cost/performance in real-world testing...non-overclocked.

P4 comes close to being a guarantee, something I can't say about A64. Yes mine and most of the ones I saw were Rev-E made no frackin' difference LOL! 1.6V made no difference to mine.

I'm not about to come to an Enthusiast site and talk about not overclocking. Unless it's like mine and can't be overclocked worth a shit! Conroe *will* not need to be overclocked. 9xx and 6xx 65nm already overclocks like a mutha! Give them any kind of half-assed good RAM and they're not so easy to beat! Now of course I believe the Conroe test results and you guys don't, that's fine!

As sparks said, it'd make no difference. Conroe could turn out to be 45% faster and AMD leaning folks would find something Be-yotch about! Meanwhile I'll continue to buy both. If K-10 comes out next year and kicks Mucho-Ass, then you'll be able to sub K10 for what I'm saying about Conroe now. If Intel did cheat, then I'd be here to tear them a new one!

To the others who don't like my Intel comments please return to the AMD side of the Forum where I don't post any of this! Pretty simple uh? It's always the same folks complainning about anything positive about Intel, funny uh?

 

[DejaWiz]

P4 comes close to being a guarantee, something I can't say about A64.

Agreed. For the most part. P4 does seem to oc like mad. Much better % yield than any A64.

I'm not about to come to an Enthusiast site and talk about not overclocking. Unless it's like mine and can't be overclocked worth a shit! Conroe *will* not need to be overclocked. 9xx and 6xx 65nm already overclocks like a mutha! Give them any kind of half-assed good RAM and they're not so easy to beat! Now of course I believe the Conroe test results and you guys don't, that's fine!

It's fine to talk about overclocking, but what good is oc'ing really if the baseline sucks ass?

As sparks said, it'd make no difference. Conroe could turn out to be 45% faster and AMD leaning folks would find something Be-yotch about! Meanwhile I'll continue to buy both. If K-10 comes out next year and kicks Mucho-Ass, then you'll be able to sub K10 for what I'm saying about Conroe now. If Intel did cheat, then I'd be here to tear them a new one!

Agreed again! It would make no difference and all the AMDrones would find something (albeit petty) to bitch about just to try and make themselves feel better due to inferiority complexes. LOL

To the others who don't like my Intel comments please return to the AMD side of the Forum where I don't post any of this! Pretty simple uh? It's always the same folks complainning about anything positive about Intel, funny uh?

Agreed yet again!

 

[Donnie27]

It is strictly dependant on the benchmark being run. I was recalling an article at TechReport about the X2's vs the 840XE.

But, I did find an article there about the 955XE vs the FX60. Using the POV-Ray Rendering test, the X2 3800+ scores slightly better than the 955XE.

X2 3800+ w/6800 Ultra and 1GB Mem
955XE w/7800GTX and 2GB Mem

All the other benchmarks seem to have severe variations, so I did not include those.

Man, I can't wait to see the Conroe benchmarks from these guys! Should be very interesting indeed.


[EDIT]
Found a comparison between the X2 3800+ and 965XE. Very surprising.

http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2725&p=10

Looking Ahead, the Conroe Comparison

Intel was well aware of what it was doing when it showed off Conroe at IDF, and it knew that there's no way that we could recommend the Pentium Extreme Edition 965 when Conroe was a little over a quarter away. If you need more of a reminder of why it will only help to wait, here are a few of the Conroe tests that we were able to run at IDF, which we duplicated on our systems here (same test configuration, test files and hardware):

Let's also not forget that the Conroe E6700 will list for $530, and there will eventually be an Extreme Edition based on it as well that will offer even better performance than what we're seeing here. In other words, as difficult as it usually is to recommend any of the FX or Extreme Edition series, it's virtually impossible to do so when you think of what's coming.

One of the few times I agree with Anand. My twist is that I'd buy NOTHING right now, Intel's or AMD's. That much performance for that PRICE!

 

[mwarps]

60% means more then you would like to believe, you can never simply use theoretical conversions and label them as correct. The 60% figure is more real world and represent smore on the lines of what is gained in actuality from a die shrink then theoretical caclulation will give you. You don't need all available information to accertain what a die shrink will cause what effects, using past optical transistion as a guideline is a good place to start and shouldn't leave you too far off from where the actual size will be.

I imediaitely noticed your math is off from when you stated the 440mm^2 figure as there is no way in hell, that Conroe would be that size on 90nm node. Double an X2 my ass lol. Assuming atrociously incorrect figures for Conroe die size on 65nm and then doing math on those would lead to incorrect figures on 90nm as well, and then comparing to the correct figure for X2, is simply atrocious and represents distorted reality. If you wanted to do a comparison to X2 figures then you should have used more accurate numbers for Conroe's 65nm size from the get go and this situation would not have occured.

Thank you for sidestepping the fact, yet again, that I am still right about your math. 60% means utterly nothing when speaking of the 90nm to 65nm node. Regardless of whether or not the die size figures are correct, your inability to understand simple mathematics and relate them to valid scientific arguements was the only point was I was trying to make. Engineering and science, unlike faith and insanity, dictate process changes and mathematics in semiconductor manufacturing. Your inability to properly recognize a simple calculation error (14mm vs 12.4mm for the linear, and 196mm^2 vs 155mm^2 for area) versus "distorted reality" (which you seem to throw around a lot in here) tells me you're someone who is suffering from a case of distorted reality, yourself.

Proper numbers for Conroe are 155mm^2, unlike the "14x mm^2" that you quoted in the thread earlier. Talk about distorted reality, it's a released statistic. Making it smaller to making something inside feel bigger? Also, the proper theoretical die size of Conroe is 299mm^2 at 90nm. Your "14x mm^2" is where my original 14mm on a side came from, I am ashamed to say. Your distorted notation caused me to use an incorrect number. Ironic?

Comparing the die sizes of two unrelated architectures at two different process sizes for anything other than computing the area (and maybe power dissipation) is scientifically baseless. It is, however, major e-penis-enlargement. I'm done, I've spent too much precious time on this.

 

[coldpower27]

Thank you for sidestepping the fact, yet again, that I am still right about your math. 60% means utterly nothing when speaking of the 90nm to 65nm node. Regardless of whether or not the die size figures are correct, your inability to understand simple mathematics and relate them to valid scientific arguements was the only point was I was trying to make. Engineering and science, unlike faith and insanity, dictate process changes and mathematics in semiconductor manufacturing. Your inability to properly recognize a simple calculation error (14mm vs 12.4mm for the linear, and 196mm^2 vs 155mm^2 for area) versus "distorted reality" (which you seem to throw around a lot in here) tells me you're someone who is suffering from a case of distorted reality, yourself.

Proper numbers for Conroe are 155mm^2, unlike the "14x mm^2" that you quoted in the thread earlier. Talk about distorted reality, it's a released statistic. Making it smaller to making something inside feel bigger? Also, the proper theoretical die size of Conroe is 299mm^2 at 90nm. Your "14x mm^2" is where my original 14mm on a side came from, I am ashamed to say. Your distorted notation caused me to use an incorrect number. Ironic?

Comparing the die sizes of two unrelated architectures at two different process sizes for anything other than computing the area (and maybe power dissipation) is scientifically baseless. It is, however, major e-penis-enlargement. I'm done, I've spent too much precious time on this.

I recognize quite well how your figures are incorrect, thank you very much, you are side stepping the issue that real world optical shrinks will not work out to how pure mathematical calculations would dictate you have to account for the information of past process transistions in order to extrapolate how an optical shrink would affect Conroe.

Pure scientific arguments are worthless without taking into account how things actually happen in reality, so they mean nothing to me.

The only who is living in distorted reality is yourself who is not able to accept that you have to take real world data as well into account to get a more accurate picture of what will happen. Instead of just completely hypothetical math calculations.

It is also not my problem you had a reading problem and couldn't see that I had listed the die area as 14x mm^2 in the first place creating errors on your end. As well until I find information that correlates to the figure of 155mm^2 as you listed I am going to have to stick with 145 mm^2 as the correct figure. As it stands Conroe is known to be decently smaller then Presler, 14x mm^2 is as accurate as it gets given the amount of data we have now.

I recognized that you had calculations errors, but it isn't my job to correct your incorrect work. It's up to you to fix your own errors.

That is only the proper size in a purely hypothetical scenario which I don't support, you need to account for real world data. 52% for an optical shrink will never happen as never in the real world will you get a perfect optical shrink.

Comparing the die sizes of two different architectures at two different nodes has much basis in reality something you don't wish to accept, as they will be in direct competition with each other intially. A direct 65nm to 65nm comparison will not happen till sometime H1 2007. Since they will be direct competitors and the best each corporation has they are directly comparable. When Conroe launches all AMD will have is Windsor so that is the comparison that has to be made.

I am glad you are done you can go back to your purely hypothetical world now, where you don't account for how things actually happen in real life as opposed to nothing but theories. My you must be desperate to prove your argument with the need for constant references to your e-manhood.

 

[Donnie27]

My you must be desperate to prove your argument with the need for constant references to your e-manhood.

I was thinking the same thing LOL!

I thought I saw somewhere that they said it was 142mm^2, may have been EETimes.

 

[Bona Fide]

I don't think we'll be waiting till 2007 for a 65nm AMD chip. AMD just announced that Fab36 will be rolling out 65nm parts by the end of this year.

 

[Donnie27]

I don't think we'll be waiting till 2007 for a 65nm AMD chip. AMD just announced that Fab36 will be rolling out 65nm parts by the end of this year.

The problem is PRICES

The 2.67GHz Conroe CPU featured in the IDF benchmarks last month will have an official MSRP of $530 at launch. The 2.6GHz AMD Athlon 64 FX processor used in those benchmarks costs a little over $1000 today, though AMD roadmaps have revealed this product will not get a price cut until June at the earliest. AMD's top of the line DDR2 2.6GHz processor will ship for just under $700 at the end of Q2'06.

The slightly slower 2.4GHz version will come with a $316 price tag. Even slower versions with half of the L2 cache will start at $209. As expected, Pentium D processors will take a huge price cut. The Pentium D 950, available today for a mere $637, will drop to an incredible $241 at the time of the Conroe launch.

Meanwhile

http://dailytech.com/article.aspx?newsid=1601

Athlon 64 FX-62 AM2 $1,236
Athlon 64 X2 5000+ AM2 $696
Athlon 64 X2 4800+ AM2 $645
Athlon 64 X2 4600+ AM2 $558
Athlon 64 X2 4400+ AM2 $469
Athlon 64 X2 4200+ AM2 $365
Athlon 64 X2 4000+ AM2 $328
Athlon 64 X2 3800+ AM2 $303
Athlon 64 3800+ AM2 $290
Athlon 64 3500+ AM2 $208
Sempron 3600+ AM2 $123
Sempron 3500+ AM2 $109
Sempron 3400+ AM2 $97
Sempron 3000+ AM2 $77
Sempron 2800+ AM2 $67

Trying to squeeze their Fans before the bottom falls out!

 

[DejaWiz]

The problem is PRICES

The 2.67GHz Conroe CPU featured in the IDF benchmarks last month will have an official MSRP of $530 at launch. The 2.6GHz AMD Athlon 64 FX processor used in those benchmarks costs a little over $1000 today, though AMD roadmaps have revealed this product will not get a price cut until June at the earliest. AMD's top of the line DDR2 2.6GHz processor will ship for just under $700 at the end of Q2'06.

The slightly slower 2.4GHz version will come with a $316 price tag. Even slower versions with half of the L2 cache will start at $209. As expected, Pentium D processors will take a huge price cut. The Pentium D 950, available today for a mere $637, will drop to an incredible $241 at the time of the Conroe launch.

Meanwhile

http://dailytech.com/article.aspx?newsid=1601

Athlon 64 FX-62 AM2 $1,236
Athlon 64 X2 5000+ AM2 $696
Athlon 64 X2 4800+ AM2 $645
Athlon 64 X2 4600+ AM2 $558
Athlon 64 X2 4400+ AM2 $469
Athlon 64 X2 4200+ AM2 $365
Athlon 64 X2 4000+ AM2 $328
Athlon 64 X2 3800+ AM2 $303
Athlon 64 3800+ AM2 $290
Athlon 64 3500+ AM2 $208
Sempron 3600+ AM2 $123
Sempron 3500+ AM2 $109
Sempron 3400+ AM2 $97
Sempron 3000+ AM2 $77
Sempron 2800+ AM2 $67

Trying to squeeze their Fans before the bottom falls out!

Those prices are definitely going to open some eyes. This is looking very good for intel. Cheap production + cheap msrp = market owned. I'm thinking AMD will need to cut prices in half of what they have listed to maintain sales to both consumers and oems.

 

[drizzt81]

Yea, and I love it how it is the same ones doing it every time.

agreed. Then again I am stupid enough to always come back and read them.. I guess I never learn.

 

[Donnie27]

Those prices are definitely going to open some eyes. This is looking very good for intel. Cheap production + cheap msrp = market owned. I'm thinking AMD will need to cut prices in half of what they have listed to maintain sales to both consumers and oems.

It might be greed on my part but I'm also not about to let my replaced-by-Conroe system go to waste. I'll wait for prices to fall then upgrade my 3500+ to something like 4400+ or etc..

 

[Bao01]

It might be greed on my part but I'm also not about to let my replaced-by-Conroe system go to waste. I'll wait for prices to fall then upgrade my 3500+ to something like 4400+ or etc..

I don't really need to upgrade. I can always use one opteron rig for DVDShrink and share the other for everything else. Faster times would be great but not completely necessary right now. If the prices come down real low around a hundred something and assuming they might for S939 which I doubt, AMD will be my first choice because I won't need to upgrade anything else. If it happens, it happens and if it doesn't, I might just hold over for Intel's 45nm chips.

 

[Killdozer]

So did I read that right?
The new Intel CPUs are only 32bit?
Holy shit. Sucks to be you if that's true.

 

[Warmonkey]

So did I read that right?
The new Intel CPUs are only 32bit?
Holy shit. Sucks to be you if that's true.

WRONG.

They are 64bit, just like existing models of the Pentium D / 4 lines.

 

[ChronicTrees]

http://www.xtremesystems.org/forums/showthread.php?t=95021

You guys have seen this right?

 

[Sovereign]

I've said this at MaxPC and am going to say it here: I do not care who makes the processer or videocard as long as it is affordable and runs games fast fast fast! PH@nb0iz SUCK! LOL

 

[Donnie27]

I don't really need to upgrade. I can always use one opteron rig for DVDShrink and share the other for everything else. Faster times would be great but not completely necessary right now. If the prices come down real low around a hundred something and assuming they might for S939 which I doubt, AMD will be my first choice because I won't need to upgrade anything else. If it happens, it happens and if it doesn't, I might just hold over for Intel's 45nm chips.

Mine would be just a swapp upgrade, replace a 3500+, that's all.

 

[Donnie27]

I don't really need to upgrade. I can always use one opteron rig for DVDShrink and share the other for everything else. Faster times would be great but not completely necessary right now. If the prices come down real low around a hundred something and assuming they might for S939 which I doubt, AMD will be my first choice because I won't need to upgrade anything else. If it happens, it happens and if it doesn't, I might just hold over for Intel's 45nm chips.

Yes and here's an example of the Company at least 3 posters here say they trust more.

[H]ardocp front page.

AMD’s 300mm Fab Off Track?

According to EETimes, AMD’s forward guidance in its quarterly results has raised some red flags among analyst. This comes just 10 days after announcing its first shipments from Fab 36 ramped up the “volume production in record time, hitting every major milestone on schedule”.

“We believe that this means AMD ramped Fab 36 slower than expected and that this will be viewed in one of two ways. Either the ramp has gone slowly as a result of poor execution ( this would be the bear point of view), or management was being conservative in previous guidance by giving a large depreciation number, and after the fab began producing mature yields at a quick pace, demand did not warrant further equipment start-up until later in the year,” he said.

Simply put, they lie and spun the facts LOL!

Sales of all Processors are DOWN=P Intel sales are down more because folks buying the most Intel processors are waiting on Cheaper and Next Gen Processors. Folks looking for Deals from AMD and or Intel are doing the same. Unlike AMD Fans who feel they need to Feed AMD extra money and buy now, the folks who really buy the Most AMD processors are waiting and will only buy absolutely what they can't put off. Even as AMD gains market share, it's VERY misleading. AMD and Intel Can RAMP all they like, there's not as many folks willing to spend alot right NOW.

After the 23rd of this month and again after Conroe, the real "Tale of the Tape" will be told. The Deadline is the Back to school buying season, this would be more of a nightmare for Intel since they will have already suffered through slow sales.

 

[DejaWiz]

It might be greed on my part but I'm also not about to let my replaced-by-Conroe system go to waste. I'll wait for prices to fall then upgrade my 3500+ to something like 4400+ or etc..

I certainly won't let mine go to waste, either; that is if I even upgrade to Conroe.

Heck, I've got everything already, so a dual core might be in my current setup's future. I guess it really depends on what kind of games/programs come out that I would actually use and actually take significant advantage of SMP to make the purchase worthwhile to me. If AMD's prices aren't very attractive for what I'll be getting compared to a Conroe cpu, mobo, and memory, then it's going to become either a backup system or a nice donation to upgrade a family members system.

Right now, I see the routes I can take when upgrade time arrives as this:
1. Keep everything but get D3D10 gpu(s).
2. Upgrade to a S939 DC, and maybe a D3D10 gpu.
3. Get a Conroe, mb, ddr2, and maybe a D3D10 gpu.
4. Get an AM2 mb, 65nm cpu, ddr2, and maybe a D3D10 gpu.

Of course, whichever route yields the most bang for the buck and can take advantage of current software will be the path I will follow.

Since I mostly game with my primary system, I'm still leaning towards getting just the D3D10 gpu(s), as long as my current cpu doesn't choke it out.

 

[savantu]

Yes and here's an example of the Company at least 3 posters here say they trust more.

[H]ardocp front page.

Simply put, they lie and spun the facts LOL!

.

How dare you say something like this of Saint AMD ?

 

[DejaWiz]

Simply put, they lie and spun the facts LOL!

There could be a lot more to it than what meets the eye. There might have been difficulties with construction or whatever. Heck, we all went through something similar when ATI was having difficulties with their X1800 production. But who knows. That's just one of the troubles with a publicly traded company...they often have to make up excuses that might be far from the truth to keep the stock prices at an acceptable level when projections aren't being met.

 

[savantu]

Yeah , AMD is forgiven all the time and has plenty of excuses while Intel is hammered down..nice double standards.

 

[Donnie27]

How dare you say something like this of Saint AMD ?

Hehehehe!

 

[DejaWiz]

Yeah , AMD is forgiven all the time and has plenty of excuses while Intel is hammered down..nice double standards.

It's not that alone...intel has a lot more demand than any other cpu manufacturer with the OEM makers. If they can't meet those demands, they get hammered down on. Who else is going to supply all those cpu's to the OEMs? intel is about the only company the OEMs can rely on for volume fulfillment.

 

[Donnie27]

I certainly won't let mine go to waste, either; that is if I even upgrade to Conroe.

Heck, I've got everything already, so a dual core might be in my current setup's future. I guess it really depends on what kind of games/programs come out that I would actually use and actually take significant advantage of SMP to make the purchase worthwhile to me. If AMD's prices aren't very attractive for what I'll be getting compared to a Conroe cpu, mobo, and memory, then it's going to become either a backup system or a nice donation to upgrade a family members system.

Right now, I see the routes I can take when upgrade time arrives as this:
1. Keep everything but get D3D10 gpu(s).
2. Upgrade to a S939 DC, and maybe a D3D10 gpu.
3. Get a Conroe, mb, ddr2, and maybe a D3D10 gpu.
4. Get an AM2 mb, 65nm cpu, ddr2, and maybe a D3D10 gpu.

Of course, whichever route yields the most bang for the buck and can take advantage of current software will be the path I will follow.

Since I mostly game with my primary system, I'm still leaning towards getting just the D3D10 gpu(s), as long as my current cpu doesn't choke it out.

As a Current Socket 939 user I see very little reason to upgrade to AM2 and DDR2 but more power to those who do. If this were K10 I'm ssure Id see it differently.

I can see myself going with Conroe before the IDF test came to light. Dothan and Yonah had already won me over. The only reason I hadn't already went with them? I wanted fully Desktop supported products and not a Notebook processor with half assed motherboard support. That even as I loved D & Y performance.

My Intel system is 3 years old in May. It's been due for a change for some time now. I never really liked Prescott and if not going with Conroe I would give Presler a closer look.

 

[Donnie27]

It's not that alone...intel has a lot more demand than any other cpu manufacturer with the OEM makers. If they can't meet those demands, they get hammered down on. Who else is going to supply all those cpu's to the OEMs? intel is about the only company the OEMs can rely on for volume fulfillment.

QFT!

 

[Donnie27]

There could be a lot more to it than what meets the eye. There might have been difficulties with construction or whatever. Heck, we all went through something similar when ATI was having difficulties with their X1800 production. But who knows. That's just one of the troubles with a publicly traded company...they often have to make up excuses that might be far from the truth to keep the stock prices at an acceptable level when projections aren't being met.

If you keep giving rosey projections, sooner or later you loose trust. IMHO, they could have been a little more honest without killing their Stock's Value. You can't keep saying you're on target when you're not. Then a bump in the road becomes and expected bump and the stock doesn't fall as fast.

 

[Bao01]

If you keep giving rosey projections, sooner or later you loose trust. IMHO, they could have been a little more honest without killing their Stock's Value. You can't keep saying you're on target when you're not. Then a bump in the becomes and expected bump and the stock doesn't fall as fast.

Check out: Investors practice tough love with AMD

Nothing we haven't discussed before. Just interesting to see it actually being said by industry analysts. I love how they pointed out the AMDfangirl analyst, JoAnne Feeney, an analyst at Punk, Ziegel & Co., claiming an absurd $45 target for AMD shares.

Oh well.

In other news, AMD tries to sink it's own ship by dragging Microsoft into this scuffle.

 

[Bao01]

Yes and here's an example of the Company at least 3 posters here say they trust more.

[H]ardocp front page.



Simply put, they lie and spun the facts LOL!

Sales of all Processors are DOWN=P Intel sales are down more because folks buying the most Intel processors are waiting on Cheaper and Next Gen Processors. Folks looking for Deals from AMD and or Intel are doing the same. Unlike AMD Fans who feel they need to Feed AMD extra money and buy now, the folks who really buy the Most AMD processors are waiting and will only buy absolutely what they can't put off. Even as AMD gains market share, it's VERY misleading. AMD and Intel Can RAMP all they like, there's not as many folks willing to spend alot right NOW.

After the 23rd of this month and again after Conroe, the real "Tale of the Tape" will be told. The Deadline is the Back to school buying season, this would be more of a nightmare for Intel since they will have already suffered through slow sales.

Cool stuff. All the signs were there. Just gotta follow the peanut trail.

Regarding the 4/23 price cuts, Intel won't be the one with the nightmares. AMD is going to be on the losing side of the PD 950 vs X2 3800 battle.

 

[coldpower27]

And this is exactly the information we all should be waiting for before creating threads to either put Conroe on a pedestal or piss on it.

While the argument of waiting till the final product is in the reviewers hands has some merit, there are heavy points to consider, Intel wouldn't lie about Conroes performance considering how bad of situation they are in regarding that area, they will have to eat major backlash if Conroe is indeed less then promised, and I beleive all they promised is 20% faster on average which is a midly conservative estimate IMO. So they are safe from my perspective. Conroe was benched in an mainly Intel controlled environment and produced great results, however consider Victor Wang got a hold of a early revision of Conroe and benched it and also had great results, further reinforces what Intel said, and that they were right now with what they said. It isn't very likely anymore things are going to change in AMD's favor given how long it takes to get processors out the door. I myself am expecting better per clock performance then Athlon 64x2 Socket AM2, and it is very unlikely I will be disappointed at this point.

I hope you've been paying attention to my previous posts, then. "Can't wait for Conroe benchmarks, can't wait for Conroe benchmarks, should be interesting, will probably be an ass-whooping, etc..."
I go with what is best for the intended purposes of my system at the time when I purchase. Mobile: P4M, Gaming: AMD, Dually: PIII-1000.

Yes I have been paying attention to your posts, no need to link to the ones that are Intel Pro.

But my argument here is "how long" will they have to live without it? Could be a few months, could be a few years. No one knows until both companies have 65nm offerings available to the public and they are fairly compared. Many seem to be stating that Conroe will kick the crap out of anything and everything for some time to come. It might be true, it might not be. I'm willing to wait until both companies have released comparable technolgy before I make assumptions and post FUD.

This point I will have to a disagree with there is no such things as a "fair" comparison anymore, you compare whatever is out at a given time to whatever is out at a given time.

For example:
Q3 2006
Conroe vs Windsor because that will be what each has at the given moment in time. We don't wait for AMD to shift to the 65nm fabrication process. They are 9-12 months behind Intel, this is their problem not Intel's. It is the best technology they have at their disposal at the present time and is what will hae to be compared. If you are going to pull the 65nm vs 65nm I can also pull the argument that Intel hasn't moved to SOI technology yet and that it isn't fair on those grounds as well, and until Intel does move to that technolgoy they should not be compared.

H1 2007
Higher Clock Conroe vs Brisbane. This looks to be still K8 technology, but simply a die shrink of the older Windsor Core, I would still give the win, to Intel given the per clock performance advanatge Core Architecture maintain over K8 Dual Core technology.

You can wait till a certain time frame for a comparison but it doesn't change that to me comparisons are to be made constantly, as one company will always have some edge on the other at a given time frame however small it may be. TO me it just given the opportunity for bias as you can simply wait till the compairson favors one company over the other, constant comparison at each interval time frame ellminiates this as is fair.

I absolutely agree that when Conroe first arrives, it will be the better choice from a price/performance standpoint derived from what we have seen so far (I never said it wasn't?). I just so happen to be setting my upgrade cycle at a time that is logical and convenient for myself: when I get a bit of money thrown into my bank account about the same time that the newest 65nm tech will be available from both companies.

Yes, and that is fine, but I did want to make it clear to others, your situation doesn't apply to everyone.

 

[JetUsafMech]

In other news, AMD tries to sink it's own ship by dragging Microsoft into this scuffle.

That's the first thing I thought when I read that today. But you had to kinda see it coming.