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

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

NTMBK

Diamond Member
Nov 14, 2011
8,320
315
126
#51
If it weren't for AMD, we'd probably have ARM Macbooks by now.
 
Mar 13, 2006
10,129
130
126
#52
Lol without AMD we'd be stuck with Intel trying to break 7GHz on single core CPUs :p

AMD owned IPC in K6 and for freaking ages after that, my 500 was winning the "booting into half-life race" at lans over P3 550s. And there's a reason the XP series were marketed with an Intel equivalent number instead of clockspeed, my 1.6Ghz chip was sold as "1900+" because it literally had 15-20% better IPC.

AMD was also pushing dual cores and 64bit architecture when the Intel super friends were arguing there's no point "because software didn't support it" (sound familiar?).

I'd say Intel has learnt a lot from AMD (and vice versa of course). And we're all in a better world because of it.
AMD didn't create K6. They bought Nextgen for the design.
 

Blitzvogel

Platinum Member
Oct 17, 2010
2,012
0
81
#53
I will just say that AMD kept Intel on their toes. Intel having to make some underhanded dealings goes to show how big of a threat AMD was to their business.

There should be no denial that Intel made some major mistakes and really underestimated what AMD was capable of, like releasing the first 64 bit and multi-core x86 processors. It must've been pretty humbling for Intel to accept that the Pentium M architecture, not the Pentium 4, was the real way forward. Of course nothing has been the same since Conroe, which AMD took over 2 years to match, by which time Nehalem had come a knockin'.

Knowing what I know now, I wish I had really been into PCs when I was in high school, because the early 2000s were pretty exciting times. It wasn't until 2005 that I really started delving into all of it.
 

TeknoBug

Platinum Member
Oct 2, 2013
2,078
1
91
#54
We'd probably still be using 32bit and using Pentium 5 or 6 clocked to 2.6GHz max.

If it weren't for AMD, we'd probably have ARM Macbooks by now.
That's hilarious.
 

Fjodor2001

Diamond Member
Feb 6, 2010
3,395
0
76
#55
Who somehow manages to forget Intel came out with the first multi core x86 CPU.
For the mainstream desktop market of course. Same for 64-bit. The AMD CPUs are what revolutionized the desktop market in this regard.
 

Fjodor2001

Diamond Member
Feb 6, 2010
3,395
0
76
#56
AMD didn't create K6. They bought Nextgen for the design.
https://en.wikipedia.org/wiki/AMD_K6

Despite the name implying a design evolving from the K5, it is in fact a totally different design that was created by the NexGen team, including chief processor architect Greg Favor, and adapted after the AMD purchase
So if you say AMD should not be credited for it at all, then Apple should not be credited for their Ax CPUs either, since they are based on ARM cores.
 
Mar 10, 2006
11,719
122
126
#57
https://en.wikipedia.org/wiki/AMD_K6



So if you say AMD should not be credited for it at all, then Apple should not be credited for their Ax CPUs either, since they are based on ARM cores.
What the...no. Apple's Ax chips come with custom ARM cores that dramatically outperform anything that ARM or its other licensees are selling into the mobile space (and ARM's cores are already very good).
 
Oct 10, 1999
11,915
0
76
#57
What the...no. Apple's Ax chips come with custom ARM cores that dramatically outperform anything that ARM or its other licensees are selling into the mobile space (and ARM's cores are already very good).
That's by the PA Semiconductor team. At what point do you say something is made by the parent company after they absorb another company? I mean, AMD wouldn't have K6, K7, or K8 without buying NexGen. Similarly, Apple wouldn't have A7, A8, or A9 without buying PA Semiconductor.
 
Mar 10, 2006
11,719
122
126
#58
That's by the PA Semiconductor team. At what point do you say something is made by the parent company after they absorb another company? I mean, AMD wouldn't have K6, K7, or K8 without buying NexGen. Similarly, Apple wouldn't have A7, A8, or A9 without buying PA Semiconductor.
The difference between AMD and Apple is that Apple really ran with the foundations that the PA Semi teams brought it. Today you will see at Apple top engineers from other semicos like Intel (Apple raided the Atom teams & even managed to score Per Hammarlund -- Haswell chief architect), AMD, Qualcomm, ARM, IBM (they nabbed a lot of the folks working on POWER9, AFAIK).

Anyway, I was just pointing out the flaw in Fjodor's analogy :p
 

Fjodor2001

Diamond Member
Feb 6, 2010
3,395
0
76
#59
What the...no. Apple's Ax chips come with custom ARM cores that dramatically outperform anything that ARM or its other licensees are selling into the mobile space (and ARM's cores are already very good).
The Apple cores are still based on ARM cores. Same as that AMD's K6 was based on a NexGen x86 core.

Both Apple and AMD started from an external base design and improved/added to it.
 

BigDaveX

Senior member
Jun 12, 2014
368
92
116
#60
There should be no denial that Intel made some major mistakes and really underestimated what AMD was capable of, like releasing the first 64 bit and multi-core x86 processors.
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.
 

NTMBK

Diamond Member
Nov 14, 2011
8,320
315
126
#61
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 wouldn't read too much into the Phenom I- the "true quad core" approach is meant to improve cache performance, but then they screwed the pooch with a broken TLB and a miniscule L3.

Nehalem followed the same basic scheme as Phenom (private L1 and L2, synchronised through an L3, integrated memory controller) and kicked serious ass because of it. The concept was sound, but the execution was lacking.
 
Jul 1, 2001
21,047
138
126
#62
We would still have lots of multi core lower power consumption processors, because that's what the market demanded as we moved away from desktops and towards laptops and tablets.

They would probably cost a bit more if AMD wasn't around at the time, though.
 
Mar 10, 2006
11,719
122
126
#63
Both Apple and AMD started from an external base design and improved/added to it.
You seem to be very confused.

The Apple CPU cores beginning with Swift (A6) onwards are totally custom cores that have nothing to do with the Cortex A-series CPU designs. The only thing they share in common with the ARM off-the-shelf designs is ISA compatibility.

Actually Apple was shipping an ARMv8 compatible CPU core before ARM's own core licensees were, which should tell you something.
 

krumme

Diamond Member
Oct 9, 2009
5,755
143
136
#64
The pc market would have eroded faster than it did.
Intel would have been less profitable than it is. :)
 

Hugo Drax

Diamond Member
Nov 20, 2011
5,646
18
91
#65
cpus would still cost 800 dollars no multicore. We would still be using netburst today with rimm memory.
 

leper84

Senior member
Dec 29, 2011
989
0
86
#66
High end what?

Laptops? They've come a long way.

Servers? They've also come a long way.

I don't think Intel sells much outside of those two groups, and probably wouldn't be too concerned even if they lost some of the desktop CPU market.

And let's not forget about the least common denominator of computing - Intel integrated graphics. It's in 99% of PCs, and what you get today is far and above what it was just 5 years ago.

I mean, yeah, I'm a desktop PC enthusiast, but even I'm starting to lose interest and I doubt performance gains are what this market is really lacking.
The goal is performance/watt. Anyone stepping out of this have no chance to compete. If AMD ignores this again, then you can pretty much count Zen sales on your hand. Same reason why they lost so much on the GPU side.
You should ask Oracle and IBM about it. Intel's marginally better at best processors are wreaking havoc on their big iron businesses.



Just in the next decade. Regardless of how Zen performs Intel line up is set in stone until the end of the decade, this for IC design, I guess the foundry roadmap is already set for the next 6-7 years.
To reply to all three of you- what you're saying is exactly the point. Where there is competition, whether in the server market or against ARM and Ipads, Intel flexes some muscle and puts out some amazing stuff.

Where there isn't- home pc/ enthusiast market- they put out just enough to maximize profits. So without competition from AMD up until bulldozer, I don't think CPU's would be near the same level.
 
Apr 12, 2015
107
0
36
#67
Intel would have saved billions from paying companies not to buy AMD.
 
Jul 1, 2001
21,047
138
126
#68
OK, here is an interesting question... Would IBM and Motorola would still be making Power PC processors for Apple if AMD wasn't around? The lack of competition in the PC arena might have caused Intel not to invest as much in R&D, allowing the RISC guys to keep up.
 
Oct 10, 1999
11,915
0
76
#69
OK, here is an interesting question... Would IBM and Motorola would still be making Power PC processors for Apple if AMD wasn't around? The lack of competition in the PC arena might have caused Intel not to invest as much in R&D, allowing the RISC guys to keep up.
Unlikely. First, IBM could not make the PowerPC 970 be more power efficient. Secondly, at the same time, while Intel was dabbling with Pentium 4, they were hedging their bets with Pentium M. This would have happened regardless of AMD.
 

richaron

Golden Member
Mar 27, 2012
1,306
1
131
#70
You seem to be very confused.

The Apple CPU cores beginning with Swift (A6) onwards are totally custom cores that have nothing to do with the Cortex A-series CPU designs. The only thing they share in common with the ARM off-the-shelf designs is ISA compatibility.*1

Actually Apple was shipping an ARMv8 compatible CPU core before ARM's own core licensees were, which should tell you something.*2
1) Proof of designs please. Prove what you just claimed.

2) You're arguing against what it really a very good (if not entirely perfect) analogy Apple/ARM vs AMD/K6+. I'm not sure why you're so blind to it, or maybe you just can't let anyone say something positive about AMD. And you're last point is just... Pointless? Inane? Trolling? Or do you really think AMD was not producing x86 cores before K6?

3) Regardless of who designed K6, AMD marketed, sold, and developed upon them. And thus entirely legitimate for this thread/topic/conversation.
 

Fjodor2001

Diamond Member
Feb 6, 2010
3,395
0
76
#71
You seem to be very confused.

The Apple CPU cores beginning with Swift (A6) onwards are totally custom cores that have nothing to do with the Cortex A-series CPU designs. The only thing they share in common with the ARM off-the-shelf designs is ISA compatibility.

Actually Apple was shipping an ARMv8 compatible CPU core before ARM's own core licensees were, which should tell you something.
A custom core does not mean that it has been designed from scratch by the company producing it. Often in these cases they start from a base design (e.g. some standard ARM core), and then just make modifications to certain parts of the CPU core.
 
Last edited:

Dresdenboy

Golden Member
Jul 28, 2003
1,730
0
136
citavia.blog.de
#72
The goal is performance/watt. Anyone stepping out of this have no chance to compete. If AMD ignores this again, then you can pretty much count Zen sales on your hand. Same reason why they lost so much on the GPU side.
That's exactly the case.
 
Oct 10, 1999
11,915
0
76
#73
A custom core does not mean that it has been designed from scratch by the company producing it. Often in these cases they start from a base design (e.g. some standard ARM core), and then just make modifications to certain parts of the CPU core.
How do you know they weren't designed from scratch? And by what criteria would you consider it so? Bulldozer and Sandy Bridge are considered "designed from scratch". But the designers also started from standard logic blocks too. By this criteria, you could argue that those designs are also based on something and not "designed from scratch".
 

krumme

Diamond Member
Oct 9, 2009
5,755
143
136
#74
Michael wouldnt have had to call Otellini so many times asking for more money
 
Last edited:

Fjodor2001

Diamond Member
Feb 6, 2010
3,395
0
76
#75
How do you know they weren't designed from scratch? And by what criteria would you consider it so? Bulldozer and Sandy Bridge are considered "designed from scratch".
What source do you have that Apple A6 was designed from scratch by Apple without basing it on parts of an existing ARM core? A very common setup is to base a custom core on a standard ARM core, optimize some instructions or building blocks, and add some customized specialized instructions.

And how did you arrive at the conclusion that Sandy Bridge was designed from scratch?
But the designers also started from standard logic blocks too. By this criteria, you could argue that those designs are also based on something and not "designed from scratch".
So same as K6. Based on an existing core and existing building blocks, with some modifications and additions. Which was exactly my point to begin with.
 
Last edited:


ASK THE COMMUNITY

TRENDING THREADS