Sign in with
Sign up | Sign in
Your question

Where AMD went wrong by enthusiasts....

Tags:
Last response: in CPUs
Share
January 18, 2007 1:27:26 AM

Notice: This is only opinion. But you're free to say what you will.

It is of my humble, and possibly misguided opinion that AMD "screwed up" long before the release of the Core 2 phenomenon or even the current debacle that is QuadFX (or 4x4, QuadFather or whatever nom de guerre you ascribe to it). Where exactly did they go wrong by the enthusiast and the overclocker do you ask? The answer to me is quite simple. Socket AM2. Now I know I'll be asked "Why AM2? AMD was kicking Intel ass at that time. Obviously, they lost ground to the Core 2". This is somewhat correct in my opinion, but lets look at the situation a bit closer.

AMD basically killed Socket AM2 by releasing it when they did. 939 fit the enthusiast market like a fine tailored glove. But no, AMD killed it for us. AM2 got basically the same performance with little variance as the older 939 socket, unless of course, you used super low latency, ultra tight timed DDR2. Only then did the user see an appreciable gain in performance. Now combine the fact that you got virtually the same performance with 939 on cheaper,looser, more available DDR, duplicate chips (3800 X2 anyone?) and the upcoming release of K8L and 65nm, AMD would have done well to wait to do a socket change. Pair that with the release of Core 2, and you have a one, two, punch to the enthusiast.

Now they've totally unbalanced the market by doing such, and look to do so even more. This "Fusion", the integration of the CPU and the GPU on one die looks to benefit the "mainstream" market and OEMs far more than it will overclockers and gamers. It seems AMD wants to leave us behind and move to innovating in the server and OEM market (mainstream), while Intel finally looks at us with offerings such as the XBX series, the Core 2 family, 45nm, and the future 45nm + IMC. Not only that, but the prices Intel is selling these new wonder chips at doesn't help for market stability either.

So to AMD, who I know reads enthusiast boards, please, don't forget about us.

/end rant.

More about : amd wrong enthusiasts

January 18, 2007 1:44:11 AM

Am2 wasnt just for gamers, it was simply an update for all users, although I do agree a rather poorly timed one. It would have been wiser to time it with the new arch, or even 65nm.
January 18, 2007 1:47:34 AM

Yeh, I agree. None of their products, spare the FX series CPU's are meant "Only for Gamers" but AMD always give the impression of being the enthusiasts company, and to make such a bad move....
Related resources
January 18, 2007 2:03:38 AM

Quote:
Notice: This is only opinion.

It is of my humble, and possibly misguided opinion that AMD "screwed up" long before the release of the Core 2 phenomenon or even the current debacle that is QuadFX (or 4x4, QuadFather or whatever nom de guerre you ascribe to it). Where exactly did they go wrong by the enthusiast and the overclocker do you ask? The answer to me is quite simple. Socket AM2. Now I know I'll be asked "Why AM2? AMD was kicking Intel ass at that time. Obviously, they lost ground to the Core 2". This is somewhat correct in my opinion, but lets look at the situation a bit closer.

AMD basically killed Socket AM2 by releasing it when they did. 939 fit the enthusiast market like a fine tailored glove. But no, AMD killed it for us. AM2 got basically the same performance with little variance as the older 939 socket, unless of course, you used super low latency, ultra tight timed DDR2. Only then did the user see an appreciable gain in performance. Now combine the fact that you got virtually the same performance with 939 on cheaper,looser, more available DDR, duplicate chips (3800 X2 anyone?) and the upcoming release of K8L and 65nm, AMD would have done well to wait to do a socket change. Pair that with the release of Core 2, and you have a one, two, punch to the enthusiast.

Now they've totally unbalanced the market by doing such, and look to do so even more. This "Fusion", the integration of the CPU and the GPU on one die looks to benefit the "mainstream" market and OEMs far more than it will overclockers and gamers. It seems AMD wants to leave us behind and move to innovating in the server and OEM market (mainstream), while Intel finally looks at us with offerings such as the XBX series, the Core 2 family, 45nm, and the future 45nm + IMC. Not only that, but the prices Intel is selling these new wonder chips at doesn't help for market stability either.

So to AMD, who I know reads enthusiast boards, please, don't forget about us.

/end rant.



You are thinking in terms of unlimited resources which AMD doesn't have. How could they wait until they release a new architecture to optimize for DDR2/3?

That timing was as good as they could get it. I really admire their focus. I know what it's like to have a bigger, meaner competitor and how you have to be nimble to survive. Had AMD waited until now, the prices would still be the same while there would still be a need for continued DDR research and production.

Enthusiasts were already used to high prices. I paid $229 in Oct05 for 2GB 3-3-3-8 DDR400. I paid $179 for the same Corsair set around 5 months later. Somewhat a drop but that wasn't for 2-2-2-3 DDR400, where AMD did best (with DDR).
Most high-end DDR2 (800+) only needs 4-4-4 to match DDR. You could almost blame the memory companies since they are just getting CAS3 DDR2 even close to affordable at large sizes and fast timings.

The first 2GB chips have only just been released. 4GB is now around $500 and even that is wasted because drivers aren't supposedly mature enough for X64.

Enthusiasts can still use water and phase change cooling to get to 3.5GHz and with the speed RAM runs then how can you go wrong? Even without that, the 5200+ is only $289, which leaves you a nice chunk of change to get the better RAM.


I think if anything they should have went to 1066 RAM instead of 800, but again available resources and ROI are going to play a huge role in any corporate endeavor. If things like that were so easy, I'd have a successful SW business now.

All in all, they have still managed to keep a toe in the door for the perf crown though I wonder if you would think this if Core 2 were just coming out now.
January 18, 2007 2:12:14 AM

Quote:

I know what it's like to have a bigger, meaner competitor and how you have to be nimble to survive.


Are you referring to Jumping Jack? :twisted:
January 18, 2007 2:14:28 AM

I'll admit that the hind sight provides more insight on to the matter than just looking at in from the present view, and that we are used to the higher prices, but for one, that doesn't mean with have to roll over and accept those prices, and also, aren't most financial analysis made after things have happened and not while said events are occurring?
My entire point is that this really unbalanced the market, and will lead to higher prices down the road, if things don't turn out right.
January 18, 2007 4:19:50 AM

Quote:
Interesting take --- I recall the initial release of 939 and the 940 sockets, the reception was luke warm, but AMD did one hellva job on the IMC which proved to be their key to success in my opinion.

AM2, though I would argue, was not poorly timed at least not from a business stand point --- arguably the performance took a hit with less than DDR2-800 and loose timings; however, it was small. With the best RAM performance went up, again by a negligible amount but it wasn't necessarily a step backwards.

However, specifically to the enthusiast market. AMD pretty much used the underground word of mouth marketing by keeping the 'geeks' happy, now that they are a name in every venue --- this is less important so I am not surprised that the enthusiast is not getting the best attention --- 4x4 was an attempt but fell short, they could have done better.

But you could have seen this coming ---

AMD plans to spend the next three to five years carving out a name for itself among this lucrative market, Ruiz said. If that means taking the company's focus off of the consumer space for awhile, that's fine. "We don't feel we need to make huge strides in the consumer [space] in the near future," Ruiz told the audience.

http://www.tgdaily.com/2006/03/06/amd_ceo_capacity_coul...

Hector is chasing the money and he basically looked at you, me, Joe Sixpack, and Baron and said -- 'I want commercial, you guys can chill for a while' :)  Hector knows where the money is, he knows....

What AMD has tried but failed to do is emulate Intel's approach to commercial space (not servers, but laptops, DTs for business) -- offer up a stable, managable, platform that simply works. The SIPP (AMD's counter the CSIP) provides a stable environement by which corporations can lower costs and maintain hundreds or thousands of systems and roll out updates on proven platforms. Commerical markets have never trusted AMD to deliver a unwaivering platform package --- hence, ATI acquaition (at least one reason), and they (AMD) will focus on this before satisfying the enthusiast.

Jack

I'd be really careful about what you're saying about AMD's attempts at a platform being a failure. With ATI they are going to pose a very real threat in the near future. Besides, they own the supercomputer space and with Barcelona will probably continue to gain command of 4S+ server space, with probable further gains in the 2S market, regardless of how well Penryn is and how far Intel can push their clock speed. AMD is poised for further gains in commercial applications. I know the fanboyish "just wait until Barcelona comes" argument is trite as all hell, but from what I've been reading and the trends I'm seeing, AMD will be in good shape to keep chipping away at Intel's market percentages once their revised architecture hits the market.

Honestly, everything but the desktop space is where they are the strongest, I really don't blame them, from a business point of view, for aiming there.
January 18, 2007 5:14:11 AM

I started wondering what was up with AMD, when it was annouced that price-cuts were gonna be coming, but only after a price-jump. I can't remember exact when this was....05 i think. :?
January 18, 2007 5:40:11 AM

Quote:

I know what it's like to have a bigger, meaner competitor and how you have to be nimble to survive.


Are you referring to Jumping Jack? :twisted:

true that
January 18, 2007 5:40:32 AM

Well I just hope AMD can give us enthusiast a chip that would finally challenge the Core 2 Duo. Seems right now the Core 2 Duo is the only choice for hard core gamers and overclockers. It's been 2 years now since I last use AMD chip. I hope my next upgrade would be the Core 2 Duo or AMD, if they can offer one by fall of this year.
January 18, 2007 6:30:49 AM

What AMD desparately needs is a K7S5A type motherboard, one that uses both DDR and DDR2 memory, and PCIe and AGP. There are tons of socket A upgraders like myself that went to Core2duo because of Asrock 775dual-vsta. I got to keep my memory and AGP card until I could dig up more money for an upgrade. An AMD solution was either too costly, or a upgrade dead end. the 754 pin fiasco was no help to anyone either.
January 18, 2007 6:54:50 AM

Quote:
What AMD desparately needs is a K7S5A type motherboard, one that uses both DDR and DDR2 memory, and PCIe and AGP.

That would be harder as AMD's IMC for DDR2 is not backwards compatible with DDR memory ...
January 18, 2007 8:28:43 AM

Not exactly Ninja; They had to use DDR2 and since the K8 IMC is on die, you also neded a new mombo if you wanted to use DDR2 because all 939 motherboards used DDR.
Even if they had released a S939 chip with DDR2 controller it would have been the same; you couldn't have used the new models on old motherboard or vice versa. the IMC is the real problem here.
January 18, 2007 8:41:09 AM

AMD will and can't forget about the enthusiasts, since it was us which kept them going when K7/K8 came out initially. Unfortunately, AMD doesn't have the resources that Intel has, and can't play the waiting game Intel can concerning with the releasing of their products. AMD has to act quickly in order to compete with Intel, whereas Intel can take a couple of hits, wait for a bit, then do a sucker-punch when AMD just happens to look away for a second or two. Its this difference in between these companies that shows how much AMD has to grow before it can really take on Intel on level ground. I'm sure AMD will provide the enthusiasts with the chips we crave, but unfortunately, it isn't us who keeps AMD going at the moment. The OEM and server markets are now the big bucks for AMD, and they need that money in order to bounce back again like they did before with K7/K8. They did it before, I'm sure they can do it again, otherwise us enthusiasts will still be paying craploads for our Intel quad-cores...
January 18, 2007 9:10:30 AM

Quote:
AMD will and can't forget about the enthusiasts, since it was us which kept them going when K7/K8 came out initially. Unfortunately, AMD doesn't have the resources that Intel has, and can't play the waiting game Intel can concerning with the releasing of their products. AMD has to act quickly in order to compete with Intel, whereas Intel can take a couple of hits, wait for a bit, then do a sucker-punch when AMD just happens to look away for a second or two. Its this difference in between these companies that shows how much AMD has to grow before it can really take on Intel on level ground. I'm sure AMD will provide the enthusiasts with the chips we crave, but unfortunately, it isn't us who keeps AMD going at the moment. The OEM and server markets are now the big bucks for AMD, and they need that money in order to bounce back again like they did before with K7/K8. They did it before, I'm sure they can do it again, otherwise us enthusiasts will still be paying craploads for our Intel quad-cores...

It was just their mistake; they had to jump to AM2, 65nm and barcelona much earlier, to prevent this drop.
January 18, 2007 9:14:17 AM

True... because they can't take those hits as well as Intel can. I just hope they do have something up their sleeves to bring out at the right time, instead of bringing out rushed products which could have done with better tweaking/testing.
a b à CPUs
January 18, 2007 9:39:37 AM

Quote:
Notice: This is only opinion.

It is of my humble, and possibly misguided opinion that AMD "screwed up" long before the release of the Core 2 phenomenon or even the current debacle that is QuadFX (or 4x4, QuadFather or whatever nom de guerre you ascribe to it). Where exactly did they go wrong by the enthusiast and the overclocker do you ask? The answer to me is quite simple. Socket AM2. Now I know I'll be asked "Why AM2? AMD was kicking Intel ass at that time. Obviously, they lost ground to the Core 2". This is somewhat correct in my opinion, but lets look at the situation a bit closer.

AMD basically killed Socket AM2 by releasing it when they did. 939 fit the enthusiast market like a fine tailored glove. But no, AMD killed it for us. AM2 got basically the same performance with little variance as the older 939 socket, unless of course, you used super low latency, ultra tight timed DDR2. Only then did the user see an appreciable gain in performance. Now combine the fact that you got virtually the same performance with 939 on cheaper,looser, more available DDR, duplicate chips (3800 X2 anyone?) and the upcoming release of K8L and 65nm, AMD would have done well to wait to do a socket change. Pair that with the release of Core 2, and you have a one, two, punch to the enthusiast.

Now they've totally unbalanced the market by doing such, and look to do so even more. This "Fusion", the integration of the CPU and the GPU on one die looks to benefit the "mainstream" market and OEMs far more than it will overclockers and gamers. It seems AMD wants to leave us behind and move to innovating in the server and OEM market (mainstream), while Intel finally looks at us with offerings such as the XBX series, the Core 2 family, 45nm, and the future 45nm + IMC. Not only that, but the prices Intel is selling these new wonder chips at doesn't help for market stability either.

So to AMD, who I know reads enthusiast boards, please, don't forget about us.

/end rant.



You are thinking in terms of unlimited resources which AMD doesn't have. How could they wait until they release a new architecture to optimize for DDR2/3?

That timing was as good as they could get it. I really admire their focus. I know what it's like to have a bigger, meaner competitor and how you have to be nimble to survive. Had AMD waited until now, the prices would still be the same while there would still be a need for continued DDR research and production.

Enthusiasts were already used to high prices. I paid $229 in Oct05 for 2GB 3-3-3-8 DDR400. I paid $179 for the same Corsair set around 5 months later. Somewhat a drop but that wasn't for 2-2-2-3 DDR400, where AMD did best (with DDR).
Most high-end DDR2 (800+) only needs 4-4-4 to match DDR. You could almost blame the memory companies since they are just getting CAS3 DDR2 even close to affordable at large sizes and fast timings.

The first 2GB chips have only just been released. 4GB is now around $500 and even that is wasted because drivers aren't supposedly mature enough for X64.

Enthusiasts can still use water and phase change cooling to get to 3.5GHz and with the speed RAM runs then how can you go wrong? Even without that, the 5200+ is only $289, which leaves you a nice chunk of change to get the better RAM.


I think if anything they should have went to 1066 RAM instead of 800, but again available resources and ROI are going to play a huge role in any corporate endeavor. If things like that were so easy, I'd have a successful SW business now.

All in all, they have still managed to keep a toe in the door for the perf crown though I wonder if you would think this if Core 2 were just coming out now.

Any C2D on stock cooling can get better overclocks then that without the cost of water or phase cooling, and is also a stronger performer clock for clock.
a b à CPUs
January 18, 2007 9:49:59 AM

Quote:
i think the biggest failure for me was when they claimed IMC and crossbar was better than the FSB. I know for a fact that the FSB is not a bottle neck for any current intel arch. I was disapointed when amd claimed they were better because of IMC and crossbar and not because of a better arch. i think amd should have been more honest with the market and their customers and should not have let the IMC and stuff propigate out like it was the magic of k8. now k8 is getting waxed by an intel product that still uses FSB and that has to raise a ton of questions/concerns.


Intels solution to that little issue was the larger (and shared) L2 cache, IMC and HTT like solutions are actually a good thing, atleast for servers and multi processor based systems.
a b à CPUs
January 18, 2007 9:50:59 AM

Quote:
What AMD desparately needs is a K7S5A type motherboard, one that uses both DDR and DDR2 memory, and PCIe and AGP. There are tons of socket A upgraders like myself that went to Core2duo because of Asrock 775dual-vsta. I got to keep my memory and AGP card until I could dig up more money for an upgrade. An AMD solution was either too costly, or a upgrade dead end. the 754 pin fiasco was no help to anyone either.


I dont get why people bother with the dual solution boards - if you want the full performance of a C2D get a new system or save up for one :?
a b à CPUs
January 18, 2007 11:08:40 AM

By stating that AMD went wrong by enthusiasts implies that they failed to deliver a viable product and enthusiast solution; which is just wrong. Only in the context of the Core uArch can such a statement be made. If you really think about it, the same arguement can also be made for Intel. So, as relevant as this thread attempts to be, I'm not exactly sure what the point is other than to further the Intel vs AMD debate and split hairs about who can get more fps and encode a file faster.

Now, with that said, here's my rant. One observation about AMD and Intel that I've made is that AMD certainly seems to be focused more on a long term strategy and implementation. There are several examples of this, such as; early adoption of an IMC, exploring different platform solutions, and merging with ATI. It's a known fact that AMD does not have the same resources as Intel, and as a result must be more strategic and thrifty with their decisions. What a company does now may not seem to make much sense, but as the long term plans unfold, the big picture becomes a little clearer. I am involved with strategic planning for my company, and spend a number of hours explaining to the Union employees that it "may not make much sense now, but give it a few months, and you'll see"; this is just the nature of strategic business operations. And, it seems to me that as a defense mechanism, AMD has had to adopt this "lose the battle to win the war" approach in order to stay competitive.

Now, that is not to say that Intel isn't focused on the long term, is obvious that they are. But Intel has the resources to not only plan for the long term but also implement a short term strategy focused on maximizing current stock and fabrication. There are many examples of this, such as; "glueing" two cores into one die, depleting the P4 stock before really promoting C2D, and leveraging their influence to dominate the market.

This is not say that Intel's business strategy is better or AMD's is better, but is to say that every company has a different way of going about achieving the same goals of creating products, marketing that product, gaining market share, turning a profit, and returning a dividend to the stockholder.

So, to say that AMD has somehow let the enthusiast market down isn't a fair assessment of what the company currently has available on the market; not only from a product standpoint, but also from performance to cost and performance to watt standpoint. Bottom line is, if you build a new machine right now, whether you choose Intel or AMD, you're gonna have one beasty of bad a$$ box that will rip snort chew a hole into pretty much anything that either an enthusiast or average pc user can throw at it. [/rant]

Note: Edited a few times for grammar and spelling
January 18, 2007 11:47:58 AM

I've been asking why we need AM2 since the day it was released.

The answer i got most the time was for K8L.

*Sigh*
January 18, 2007 11:52:55 AM

Yes, you get a performance hit (10%) with such a board if you use PCI-Express (only 4x), but not with AGP.
But that is not much of an issue, and for budget oriented people, the ASRock 775Dual-VSTA is a great board.
January 18, 2007 12:02:47 PM

Quote:

I know what it's like to have a bigger, meaner competitor and how you have to be nimble to survive.


Are you referring to Jumping Jack? :twisted:

I said bigger and meaner. Not weaker and more clinging.
January 18, 2007 12:12:45 PM

I admit these very thoughts have crossed my mind in past months so obviously I agree.

Poor timing with a very underwhelming result helped unseat AMD as the enthusiasts choice. Even Netburst came out of the gate looking pretty decent for the time, it was only after a while it became obvious it was to go the way of the dodo. AM2 sort of fell flat still inside the gate.....

I do hope they get it together as I'm sure they will though. Competition is better for us in the end.
January 18, 2007 12:19:24 PM

Quote:
It was just their mistake; they had to jump to AM2, 65nm and barcelona much earlier, to prevent this drop.


I agree also. AMD failed in its product strategy and i believe AMD didn't evaluate well in time what were the possible Intel solutions to end its crisis.

In terms of product, Intel just produced the product all people was expecting:
- Less power consumption and heat production (AMD superiority until C2D);
- Higher processing capability in AMD FX field;
- Affordable prices for superior processing capability (the FX wasn't affordable)
- Dual core
- Great OC capability (due to the new cpu architecture)
- Kept DDR2 memory as the way (many Intel P4 buyers just kept their RAM dimms when upgrading to C2D, while old AMD customers had to change from DDR to DDR2).

AMD just slept in their 2y success and (unfortunately) it still hasn't a competitive product to compete with the C2D in the market for the masses.

Let's not forget also that for the common user (the masses ) and most of companies, Intel has a more trustful and known name (brand) than AMD, mainly because it was alone in the market many years during the PC boom (pretty much like Microsoft).
Even in the 2y golden age of AMD, i heard many common users that bought AMD referring it to "it's just like a Pentium but from another company"...
January 18, 2007 12:42:51 PM

Quote:
Up to now it has been a failure, hence ATI. I was careful to denote that it is commercial DT/Laptops --- server, they have been successful but CSIP or SIPP are not for the server space.

I also disagree with everything but DT --- Intel holds 92% of mobile before Q3, AMD did a good job getting up to 12 or 13%, but it is still Intel dominated there.


You're right, I mentally included laptops with desktops - I should have just said "consumer space."
January 18, 2007 1:57:08 PM

Quote:
By stating that AMD went wrong by enthusiasts implies that they failed to deliver a viable product and enthusiast solution; which is just wrong. Only in the context of the Core uArch can such a statement be made. If you really think about it, the same arguement can also be made for Intel. So, as relevant as this thread attempts to be, I'm not exactly sure what the point is other than to further the Intel vs AMD debate and split hairs about who can get more fps and encode a file faster.

Now, with that said, here's my rant. One observation about AMD and Intel that I've made is that AMD certainly seems to be focused more on a long term strategy and implementation. There are several examples of this, such as; early adoption of an IMC, exploring different platform solutions, and merging with ATI. It's a known fact that AMD does not have the same resources as Intel, and as a result must be more strategic and thrifty with their decisions. What a company does now may not seem to make much sense, but as the long term plans unfold, the big picture becomes a little clearer. I am involved with strategic planning for my company, and spend a number of hours explaining to the Union employees that it "may not make much sense now, but give it a few months, and you'll see"; this is just the nature of strategic business operations. And, it seems to me that as a defense mechanism, AMD has had to adopt this "lose the battle to win the war" approach in order to stay competitive.

Now, that is not to say that Intel isn't focused on the long term, is obvious that they are. But Intel has the resources to not only plan for the long term but also implement a short term strategy focused on maximizing current stock and fabrication. There are many examples of this, such as; "glueing" two cores into one die, depleting the P4 stock before really promoting C2D, and leveraging their influence to dominate the market.

This is not say that Intel's business strategy is better or AMD's is better, but is to say that every company has a different way of going about achieving the same goals of creating products, marketing that product, gaining market share, turning a profit, and returning a dividend to the stockholder.

So, to say that AMD has somehow let the enthusiast market down isn't a fair assessment of what the company currently has available on the market; not only from a product standpoint, but also from performance to cost and performance to watt standpoint. Bottom line is, if you build a new machine right now, whether you choose Intel or AMD, you're gonna have one beasty of bad a$$ box that will rip snort chew a hole into pretty much anything that either an enthusiast or average pc user can throw at it. [/rant]

Note: Edited a few times for grammar and spelling
You can sit here and talk about strategies, future line-ups, who has more market-share, stock performance, etc.etc....but it doesn't mean anything in the "here and now"....which is what most people are interested in. Yes, Intel failed in that respect, as well, with Prescott..but they payed dearly for it. They lost many customers/potential customers to AMD's "better" K8. Anybody who jumped ship from Intel to AMD did it because, at the time, it was the superior architecture, and they didn't give a rats a$$ what future plans AMD had, or where their marketshare sat. Future "potential" only really matters if you're a fanboy, because if you're just a fan of the fastest processor you can get,or fastest/$...it doesn't matter if it's Intel or AMD. I don't care if Intel has 50 fabs, and AMD only has 2..if AMD has the better product(or more in my case..better product that i can afford), than i'll take the AMD. /my rant :wink:
January 18, 2007 4:59:25 PM

Quote:
Future "potential only really matters if you're a fanboy, because if you're just a fan of the fastest processor you can get,or fastest/$...it doesn't matter if it's Intel or AMD. I don't care if Intel has 50 fabs, and AMD only has 2..if AMD has the better product(or more in my case..better product that i can afford), than i'll take the AMD.


If someone asked me what % of buyers like you (ie, that buy more performance for less cost) would i imagine, i'd say about 50% (including companies). I imagine a strong base of permanent Intel buyers around another 40% and maybe only 10% for permanent AMD buyers.

IMO in a good Intel phase/year the market could be around 80% for Intel and 20% for AMD.
In a good AMD phase/year, the market could be around 55% for AMD and 45% for Intel.
This is my perception which obviously can be very precise, but maybe not too far from reality.

Does anyone knows where to check the current market segmentation between AMD and Intel?
January 18, 2007 7:38:35 PM

We can only make observations about the past from the future so I fail to see what your point is. Core or no Core, the factors that it had identical processors, required more expensive memory and only gave a marginal increase in speed still crippled AM2.
January 19, 2007 4:36:41 AM

I see your point Ninja, the am2 intro left a lot of people wondering WTF is that all about and hesitated to adopt it.

Then when the c2d benches started flooding in, there was a clearer upgrade option available to look at.

I agree the timing was not good.
a b à CPUs
January 23, 2007 7:06:02 AM

Quote:
Yes, you get a performance hit (10%) with such a board if you use PCI-Express (only 4x), but not with AGP.
But that is not much of an issue, and for budget oriented people, the ASRock 775Dual-VSTA is a great board.


im talking with the combo of alot of things - crappy manafacturer, crappy chipset (chipshit, via? comeon), crappy memory speeds etc
a b à CPUs
January 23, 2007 7:11:11 AM

Quote:
What AMD desparately needs is a K7S5A type motherboard, one that uses both DDR and DDR2 memory, and PCIe and AGP. There are tons of socket A upgraders like myself that went to Core2duo because of Asrock 775dual-vsta. I got to keep my memory and AGP card until I could dig up more money for an upgrade. An AMD solution was either too costly, or a upgrade dead end. the 754 pin fiasco was no help to anyone either.


I dont get why people bother with the dual solution boards - if you want the full performance of a C2D get a new system or save up for one :?

People bother with these types of boards because they can not afford a brand new system right away and want the performance increase now not months from now. At least with a dual solution board you can do that. Some people that bought 939 systems early last year may have expensive ddr1 that they do not want to get rid of yet for one instance. Not everyone is rich on here like jack and ninja. :lol: 

Quote:
Not exactly Ninja; They had to use DDR2 and since the K8 IMC is on die, you also neded a new mombo if you wanted to use DDR2 because all 939 motherboards used DDR.
Even if they had released a S939 chip with DDR2 controller it would have been the same; you couldn't have used the new models on old motherboard or vice versa. the IMC is the real problem here.

They could have mad a memory controller that supported both ddr1 and ddr2. The AM3's are supposed to support ddr2 and ddr3 so I do not see where it is not possible for them to have done something similar with 939. Plain and simple, AMD crapped on us 939 users when they went to AM2.

DDR2 required a diffrent pin out, why not fix things and change the socket while they were at it.

this it the first time AMD has done this to its customers (forgotten bout backward compatibility :(  )
January 23, 2007 7:15:33 PM

Quote:
Yes, you get a performance hit (10%) with such a board if you use PCI-Express (only 4x), but not with AGP.
But that is not much of an issue, and for budget oriented people, the ASRock 775Dual-VSTA is a great board.


im talking with the combo of alot of things - crappy manafacturer, crappy chipset (chipshit, via? comeon), crappy memory speeds etc
Bah, typical enthusiast's mentality..
The benchies of the ASRock 775Dual-VSTA are overall pretty respectable, and anyway, even if you stack up all those performance handicaps, you still get a pretty fast PC which can do pretty much any task, at a ridiculous price.
I have spare 512 MB of DDR1, i could sell them on ebay for maybe 20 bucks.. instead, i'm thinkin to buy the ASRock, a cheap Netburst dual core (Pentium D 805-820, but i'm waiting a couple of weeks to see if they get discontinued and a 65nm drops in that price range), an cheap video card with TV-out, a 400GB HDD, and an HTPC case... and hook it up to my TV.
a b à CPUs
January 23, 2007 8:10:53 PM

Quote:
Yes, you get a performance hit (10%) with such a board if you use PCI-Express (only 4x), but not with AGP.
But that is not much of an issue, and for budget oriented people, the ASRock 775Dual-VSTA is a great board.


im talking with the combo of alot of things - crappy manafacturer, crappy chipset (chipshit, via? comeon), crappy memory speeds etc
Bah, typical enthusiast's mentality..
The benchies of the ASRock 775Dual-VSTA are overall pretty respectable, and anyway, even if you stack up all those performance handicaps, you still get a pretty fast PC which can do pretty much any task, at a ridiculous price.
I have spare 512 MB of DDR1, i could sell them on ebay for maybe 20 bucks.. instead, i'm thinkin to buy the ASRock, a cheap Netburst dual core (Pentium D 805-820, but i'm waiting a couple of weeks to see if they get discontinued and a 65nm drops in that price range), an cheap video card with TV-out, a 400GB HDD, and an HTPC case... and hook it up to my TV.

i dunno, i prefer upgrading in hits not wasting my money here and there on parts i cant use later.
!