Where would CPUs be today without AMD in the late 90s to mid 2006

Page 4 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Ketchup

Elite Member
Sep 1, 2002
14,482
32
106
#76
I still remember that 2006 article by AT. Talk about an amazing showing! Look at the jump over AMD, let alone Intel's own $1,000 Extreme chips!

http://www.anandtech.com/show/2045

I also remember Intel telling us how great a chip the P4 was due to scaling. Gee, I wonder what changed their minds?

AMD did that. No doubt. Intel would still be sitting there, rolling in the dough if it wasn't for them. In fact, they have already started it again. When was the last time Intel released a jump anywhere close to the only that occurred almost TEN YEARS ago? Oh yeah, they haven't.
 

.vodka

Golden Member
Dec 5, 2014
1,086
196
136
#77
When was the last time Intel released a jump anywhere close to the only that occurred almost TEN YEARS ago? Oh yeah, they haven't.
With Netburst having set the bar so low, in hindsight it wasn't too difficult for them to take P6 in its Pentium M form and bring it up to date, ending up with Conroe.

Today, well, I don't think they have another "secret" uarch in the works to utterly wreck the Nehalem lineage up to today's Skylake in the way P6 did to Netburst, with all the focus on mobile and no more low hanging fruit to be picked up after so many years... But then, what do I know? Intel has become quite tight lipped about their architectures lately.
 

Blitzvogel

Platinum Member
Oct 17, 2010
2,012
0
81
#78
Technically, AMD released the first native multi-core x86 process; Intel were the first overall by a couple of weeks.

In retrospect though, you have to wonder whether or not AMD might actually have been able to get away with making the Athlon 64 X2 an MCM chip instead of native dual-core. Based on what we later saw with Phenom, I think the X2's advantage was less its native dually status, and more the fact that the Pentium D's core just really, REALLY sucked.
I know the Pentium D was out right before the Athlon 64 x2, but I thought AMD released their server dual cores before Intel did for either desktop or enterprise?
 

BigDaveX

Senior member
Jun 12, 2014
384
111
116
#79
I know the Pentium D was out right before the Athlon 64 x2, but I thought AMD released their server dual cores before Intel did for either desktop or enterprise?
From a quick bit of research, it appears that Intel launched (or at least paper-launched) their first dual-core, the Pentium XE 840 in mid-April 2005, while the first dual-core Opterons didn't show up until the end of the following month. What you're likely thinking of is the fact that Intel didn't release any dual-core Xeons until near the end of that year... and when they did they were, to put it lightly, complete garbage.
 

BarkingGhostar

Diamond Member
Nov 20, 2009
5,827
98
126
#80
I do not understand this thread. It is almost as bad of a pity topic as this.

This topic seems like an AMD fan trying to justify AMD existence and a need to 'be recognized' based on the success of their competition. I could equally ask where would the world be with AMD if Intel wasn't riding their backs, but that would be looking at the other side of the coin.
 

zir_blazer

Senior member
Jun 6, 2013
886
23
91
#81
I know the Pentium D was out right before the Athlon 64 x2, but I thought AMD released their server dual cores before Intel did for either desktop or enterprise?
AMD launched A64X2 with inmediate availability, and they were often out-of-stock Ci7 6700K style. Intel paperlaunched the Pentium D like two days after the A64X2 launch, but you had to wait 3 or 4 months before being actually able to purchase one.
The only Pentium D worth considering was the 805, since it was cheaper than the cheapest A64X2 and had a decent overclocking headroom. However, Pentium D required new Chipsets/Motherboards, A64X2 only a BIOS upgrade, and lets not talk about power consumption... Basically, it crushed it on every metric.
 

BigDaveX

Senior member
Jun 12, 2014
384
111
116
#82
AMD launched A64X2 with inmediate availability, and they were often out-of-stock Ci7 6700K style. Intel paperlaunched the Pentium D like two days after the A64X2 launch, but you had to wait 3 or 4 months before being actually able to purchase one.
The rest of the Pentium D line didn't technically launch until after the A64 X2, but Intel were the first overall to the dual-core game with the Pentium XE 840. I remember the flame wars that ensued over the next year, with AMD fans claiming that the Pentium D wasn't a "true" dual-core, and Intel fans lording over their first-to-the-punch status, even though everyone and their grandma could see that the X2 was better overall.

The whole mess carried on until the Core 2 showed up the following year... at which point the debate turned into one about whether or not Intel had somehow cheated by slapping a huge L2 cache on the Core 2. :p
 

SPBHM

Diamond Member
Sep 12, 2012
4,858
81
126
#83
It's interesting to think about RDRAM vs DDR, it was basically DDR = AMD, Intel = RDRAM at some point...
 

zir_blazer

Senior member
Jun 6, 2013
886
23
91
#84
The rest of the Pentium D line didn't technically launch until after the A64 X2, but Intel were the first overall to the dual-core game with the Pentium XE 840. I remember the flame wars that ensued over the next year, with AMD fans claiming that the Pentium D wasn't a "true" dual-core, and Intel fans lording over their first-to-the-punch status, even though everyone and their grandma could see that the X2 was better overall.

The whole mess carried on until the Core 2 showed up the following year... at which point the debate turned into one about whether or not Intel had somehow cheated by slapping a huge L2 cache on the Core 2. :p
Here you go:
Smithfield Pentium D 820/830/840 - May 26 2005
Smithfield Pentium Extreme Edition 840 XE - May 1 2005
Manchester/Toledo Athlon 64X2 4200+/4400+/4600+/4800+ - May 31 2005
I didn't recalled than the 840 XE launched BEFORE, but that they were all released at the same time. And Intel actually paperlaunched the Pentium D before AMD - but it was that, a paper launch with no real products. The PD 805 arrived half a year later.
Indeed, Smithfield wasn't a "true" Dual Core, they were merely two Prescott dies cut together with no mean of internal communication, while A64X2s had a more elegant Crossbar. Smithfield supporting Chipsets had two FSBs, and for one Core to talk to the other one, it had to go through the FSB to the Chipset then back the other FSB channel. It was functionally EXACTLY IDENTICAL to a Dual Xeon setup of that era - I even recall that someone benchmarked a Dual Xeon vs Pentium D and was near the margin of error considering Buffered DDR RAM penalty, while A64X2 was supposedly faster than a equivalent Dual Opteron.

And yes, I also recall that AnandTech did a Conroe preview 6 months before launch date that got everyone holding back Athlon 64 X2 purchases expecting Conroe to delivered the preview results. And it did... The rest is a very sad history.


It's interesting to think about RDRAM vs DDR, it was basically DDR = AMD, Intel = RDRAM at some point...
May want to read about how Intel and RAMBus tried to take over the world. So, whoever thinks that AMD wasn't important to keep Intel in check, really needs to dig through a lot of info from previous eras (Redhill could be considered a 90' blogging-style site, and does it very good):

Always before, when Intel found the market for a CPU getting crowded, the company would move on to the next performance grade as fast as possible: not so with the DX/2. For the first time ever, Intel cut prices on a mainstream part to price-match AMD, and in consequence the Intel DX/2s remained popular and common on the market even after the cheaper AMD and Cyrix parts arrived.
The original Intel DX/4 was the fastest thing since sliced bread and mega-expensive. It was spectacularly poor value for money for a good long while until AMD got its own DX/4 into production, at which time the price of the Intel DX/4 halved — it dropped by $400 in just three weeks. Never has there been a more graphic illustration of the difference between monopoly and free market pricing.

Is actually extremely sad to see them becoming irrelevant and near-bankrupcy. But one thing is true: Maybe if AMD didn't did all what it do, probabily another manufacturer like Cyrix would be there now, or even another non-x86 player could have tried to fill its niche. It would have been fun to see x86 fail, merely so we could replace it with a more modern ISA, which is something that we can't really think that could happen these days.
 
Last edited:
Mar 13, 2006
10,138
132
126
#85
I do not understand this thread. It is almost as bad of a pity topic as this.

This topic seems like an AMD fan trying to justify AMD existence and a need to 'be recognized' based on the success of their competition. I could equally ask where would the world be with AMD if Intel wasn't riding their backs, but that would be looking at the other side of the coin.
I was thinking the same thing. The op isn't really interested in the correct answer, he wants his bias confirmed.
 
Mar 13, 2006
10,138
132
126
#86
Nvm, wrong info
 
Last edited:

geoxile

Senior member
Sep 23, 2014
325
0
51
#87
It's very obvious that AMD is the company that has contributed the most to CPU and GPU progress given its size. It's actually quite astonishing how many projects they have headed.

Multi-core x86 revolution, 64-bit x86 CPU revolution, HBM based GPUs (and soon APUs), ARM based server CPUs, lower-level GFX API:s (Mantle, indirectly driving DX12 too), APUs, console APUs, the list goes on.

Now compare that to e.g. Intel and nVidia, relative to the size of those companies.
The (proto) Mantle spec was initially designed by Johan Andersson of DICE back in 2008. Provided, AMD did help bring it to a platform, but they didn't head it. Mantle was developed by DICE basically.
 

zir_blazer

Senior member
Jun 6, 2013
886
23
91
#88
Your memory is failing you. Pentuim D shipped before X2. And the 805D shipped a year after the 840, not six months.
Nope, A64X2 had availability from day one. Pentium D were MIA.

What did was true looking at early Threads is that Pentium D 820 was much cheaper than the A64X2 4200+ when they were initially launched. AMD actually charged a price premium for the A64X2.
 
Last edited:

Markfw

CPU Moderator, VC&G Moderator, Elite Member
Super Moderator
May 16, 2002
18,182
2,133
136
#89
I was thinking the same thing. The op isn't really interested in the correct answer, he wants his bias confirmed.
Well, I don't have links, its all in my memory, but the bottom line is that around this time AMD was kicking Intels arse all over the place, and remained in control and the leader in every respect until the CORE CPU's came out. Its been all Intel since then.
 
Mar 13, 2006
10,138
132
126
#90
Nope, A64X2 had availability from day one. Pentium D were MIA.

What did was true looking at early Threads is that Pentium D 820 was much cheaper than the A64X2 4200+ when they were initially launched. AMD actually charged a price premium for the A64X2.
You may be right. I withdrew my post.
 
Mar 13, 2006
10,138
132
126
#91
Well, I don't have links, its all in my memory, but the bottom line is that around this time AMD was kicking Intels arse all over the place, and remained in control and the leader in every respect until the CORE CPU's came out. Its been all Intel since then.
Blah, and I'm posting in the wrong thread.

Ignore me, I'm going to bed.
 

Techhog

Platinum Member
Sep 11, 2013
2,834
0
26
#92
I do not understand this thread. It is almost as bad of a pity topic as this.

This topic seems like an AMD fan trying to justify AMD existence and a need to 'be recognized' based on the success of their competition. I could equally ask where would the world be with AMD if Intel wasn't riding their backs, but that would be looking at the other side of the coin.
No pro-AMD posts allowed!
 

Techhog

Platinum Member
Sep 11, 2013
2,834
0
26
#93
I was thinking the same thing. The op isn't really interested in the correct answer, he wants his bias confirmed.
Unless you're taking Shintai's side, the correct answer is the confirmation of his bias, more or less. He just wants to hear that we may not be where we are today, and that's true. There's a chance that IA64 would have eventually caught on, but in the best case it would have taken much longer and 32-bit would likely still be quite dominant.

But hey, if you believe that AMD has contributed nothing and only held things back... Well, that's on you.
 
Mar 13, 2006
10,138
132
126
#94
Where did I say that? Did you see my reply to mark that I was in the wrong thread?
 

superstition

Platinum Member
Feb 2, 2008
2,219
0
101
#95
I do not understand this thread. It is almost as bad of a pity topic as this.

This topic seems like an AMD fan trying to justify AMD existence and a need to 'be recognized' based on the success of their competition. I could equally ask where would the world be with AMD if Intel wasn't riding their backs, but that would be looking at the other side of the coin.
Asking that defeats the purpose of your post, I'd say. It gets rid of Intel bribing OEMs to blackball AMD, Intel selling at a loss to stop OEMs from using AMD, Intel's compiler trickery, AMD losing sales to Intel at all, etc.
 

Techhog

Platinum Member
Sep 11, 2013
2,834
0
26
#97
I don't think we need pro-AMD or pro-Intel posts. I think we need objective posts. But, that's my personal bias.
This is true, but isn't actually related to what I meant or, frankly, realistic. It's more like, "no posts saying positive things about AMD allowed, true or not!"
 

superstition

Platinum Member
Feb 2, 2008
2,219
0
101
#98
Unlikely. First, IBM could not make the PowerPC 970 be more power efficient.
speculation:

Anything can be made more efficient but IBM didn't have a strong enough incentive to do it. It's not like one of their primary business angles is laptops/low power consumption. Apple wasn't enough back then to warrant the investment. So they gave lip service to Apple but didn't deliver.

IBM's big iron processor designs suit their business target. Redesigning Power chips for mobile was too big an ask, particularly with their foundry. I bet if Intel had the Power chip it could have done what Apple wanted with it. It's not like RISC is inherently power hungry. It was a very wide and relatively deep design, as far as I recall. Such designs, to my knowledge, tend to have more power consumption, although less than small and very deep like Netburst.

Just found the article:

arstechnica said:
If the P4 takes a "narrow and deep" approach to performance and the G4e takes a "wide and shallow" approach, the 970's approach could be characterized as "wide and deep." In other words, the 970 wants to have it both ways: an extremely wide execution core and a 16-stage (integer) pipeline that, while not as deep as the P4's, is nonetheless built for speed. Using a special technique, which we'll discuss shortly, the 970 can have a whopping 200 instructions on-chip in various stages of execution, a number that dwarfs not only the G4e's 16-instruction window but also the P4's 126-instruction one.

You can't have everything, though, and the 970 pays a price for its "more is better" design.

...one reason that the 970's pipeline is a little shorter than the P4's: the 970's pipeline lacks the "drive" stages, which the P4 inserts in order to allow signals to propagate across the chip. Inserting whole pipeline stages just to account for wire delay is necessary only if you plan to push a design to insanely high clock speeds. The PowerPC 970's GHz rating, then, will top out at a much lower number than will the P4's.

The PowerPC 970, on the other hand, is designed from the ground up with multiprocessing in mind--IBM intends to see the 970 used in 4-way or higher desktop SMP systems. So instead of increasing the performance of desktop and server systems by using a few narrow, very high clockspeed, high power CPUs IBM would rather see multiple, slower but wider, lower power CPUs ganged together via very high-bandwidth connections.

The 970's lengthy 16-stage pipeline is one of its greatest strengths and its greatest weaknesses. Keeping such a long pipeline full requires massive branch prediction resources, but in the end the 970's instruction window is just so large that it can afford to have fair number of pipeline bubbles before it starts to see serious performance degradation. The P4 can tolerate fewer bubbles, because there aren't as many execution slots available.
This can be part of the problem as well:

arstechnica said:
It has been suggested that the technical reason for the "tacked on" nature of the Altivec unit stems from the fact that IBM leans heavily on automated tools for chip layout, as opposed to doing a lot of custom layout by hand. Custom layouts, like hand-tuned assembler code, are painstaking and costly to do, but they can be tweaked by a skilled design team in order to make the processor faster and more efficient by fine-tuning circuit timings. Custom layouts were supposedly the secret behind the late Alpha processor's industry leading clock speeds. So the idea is that if IBM had gone the slower and more painful route of hand-tuning the 970's layout, the Altivec unit would be better integrated onto the die and wouldn't need these extra transfer stage.
Bulldozer's power consumption is rumored to be, in part, due to heavy reliance on automated tools, a consequence of AMD having less money at its disposal.

Dailytech said:
Apple Veterans Plotted to Kill PowerPC Long Before Jobs' Return

Tesler: "It was actually one of the reasons that the company decided to acquire Next… We had actually tried a few years before to port the MacOS to Intel, but there was so much machine code still there, that to make it be able to run both, it was just really really hard. And so a number of the senior engineers and I got together and we recommended that first we modernize the operating system, and then we try to get it to run on Intel... And when we realized that wouldn’t work we realized we had to acquire an operating system, either BeOS or Next, and one of the plusses was once we had that we could have the option of making an intel machine."

According to Rama Menon, a senior Intel engineer, the reasons PowerPC fell behind in the race was not because of the design itself, but rather the AIM alliance's manufacturing capability and the smaller installed base. The collapse of Apple computer sales meant there weren't many PowerPC computers on the market, which meant there wasn't much PowerPC software design going on -- at least compared to x86.

Intel killed the PowerPC backers in its ability to churn out close to 10 million CPUs a year.

PowerPC appeared far faster at processing and more power efficient than Intel, but the lack of a healthy quantity of unbiased benchmarks makes it hard to say if that advantage was real or but a mirage.
So, now that I've investigated it seems like my speculation is supported.
 
Last edited:


ASK THE COMMUNITY