Some Preliminary Results of AMD Thuban @ 2.8GHz (OC'd too)

[MarkG]

Assuming were both talking about the 8008, thats part of it.

4004, surely?

 

[doive1231]

All the overall performance bemchmarks say that more cores are better. Nuff said.

 

[xaira]

All the overall performance bemchmarks say that more cores are better. Nuff said.

http://www.tomshardware.com/reviews/spider-weaves-web,1728-30.html

you were saying????????

architecture makes a bbbiggg diff

 

[FALC0N]

http://www.tomshardware.com/reviews/spider-weaves-web,1728-30.html

you were saying????????

architecture makes a bbbiggg diff

You just pointed to benchmarks that are almost three years old and are not optimized for multiple cores. Notice how the C2D at similar clock as the C2Q perform about the same as do the AMD X2 vs the X4.

Not saying your point is completely invalid, but you have to come up with something better than that.

 

[fazers_on_stun]

4004, surely?

According to Wiki yes it was the 4004:

The 4004 was released on November 15, 1971.[2] Packaged in a 16-pin ceramic dual in-line package, the 4004 was the first commercially available computer processor designed and manufactured by chip maker Intel, which had previously made semiconductor memory chips. The chief designers of the chip were Federico Faggin and Ted Hoff of Intel, and Masatoshi Shima of Busicom (later of ZiLOG, founded by Faggin).

Faggin, the sole chip designer among the engineers on the MCS-4 project, was the only one with experience in MOS random logic and circuit design. He also had the crucial knowledge of the new silicon gate process technology with self-aligned gates, which he had created at Fairchild in 1968. At Fairchild, in 1968, Faggin also designed and manufactured the world's first commercial IC using SGT - the Fairchild 3708. As soon as he joined the Intel MOS Department he created a new random design methodology based on silicon gate, and contributed many technology and circuit design inventions that enabled a single chip microprocessor to become a reality for the first time. His methodology set the design style for all the early Intel microprocessors and later for the Zilog’s Z80. He also led the MCS-4 project and was responsible for its successful outcome (1970-1971). Ted Hoff, head of the Application Research Department, contributed only the architectural proposal for Busicom working with Stan Mazor in 1969, then he moved on to other projects. Shima designed the Busicom calculator firmware and assisted Faggin during the first six months of the implementation. The manager of Intel's MOS Design Department was Leslie L. Vadász.[3] At the time of the MCS-4 development Vadasz's attention was completely focused on the mainstream business of semiconductor memories and he left the leadership and the management of the MCS-4 project to Faggin.

The Japanese company Busicom had designed their own special purpose LSI chipset for use in their Busicom 141-PF calculator with integrated printer and commissioned Intel to develop it for production. However, Intel determined it was too complex and would use non-standard packaging and so it was proposed that a new design produced with standard 16-pin DIP packaging and reduced instruction set be developed.[4] This resulted in the 4004, which was part of a family chips, including ROM, DRAM and serial to parallel shift register chips. The 4004 was built of approximately 2,300 transistors and was followed the next year by the first ever 8-bit microprocessor, the 2,500 transistor 8008 (and the 4040, a revised 4004). It was not until the development of the 40-pin 8080 in 1974 that the address and data buses would be separated, giving faster and simpler access to memory.

The 4004 was the world's first commercially available microprocessor - a complete CPU (central processing unit) integrated in a single chip. Before the 4004, CPUs comprised multiple SSI or MSI chips. The 4004 was part of the MCS-4 family of LSI chips that could be used to build digital computers with varying amounts of memory. The other members of the MCS-4 family were memories and input/output circuits, which while not part of a CPU are necessary to implement a complete computer. Specifically:

the 4001 was a ROM (read-only memory) with 4 lines of output
the 4002 was a RAM (random access memory) with 4 lines of input/output
the 4003 was a static shift register to be used for expanding the I/O lines, for example, for keyboard scanning or for controlling a printer
The 4004 included control functions for memory and I/O, which are not normally handled by the microprocessor.

The first commercial product to use a microprocessor was the Busicom calculator 141-PF.

Maximum clock speed was 740 kHz[1][7]
Instruction cycle: 92,6 kHz [8] (740 kHz /8 = 92,6 kHz, not a 108 kHz, but 10,8 us instruction period)
Separate program and data storage (i.e., a Harvard architecture). Contrary to most Harvard architecture designs, however, which use separate buses, the 4004, with its need to keep pin count down, used a single multiplexed 4-bit bus for transferring:
12-bit addresses
8-bit instructions
4-bit data words
Instruction set contained 46 instructions (of which 41 were 8 bits wide and 5 were 16 bits wide)
Register set contained 16 registers of 4 bits each
Internal subroutine stack 3 levels deep.

Dunno how much RAM it had, but I would bet under 1K. And it was on a 10,000 nm process too, with 92Kips. The 8088 only improved that to ~400Kips on average as I recall. Although for some odd reason instructions per second has fallen out of favor as a benchmark , I guess we could approximate a 3GHz modern CPU's score by multiplying the average # of instructions per clock cycle by 3x10^9 by the # cores. Should be a fairly impressive number compared to the 4004, no doubt...

IIRC it was the self-aligned gate process that made LSI & VLSI practical, since yields went way up.

 

[FALC0N]

According to Wiki yes it was the 4004:

No it wasn't. It was the 8008. The 8008 with its pre-x86 instruction set started intel on its meteoric rise. And it was a CTC design that was also given to Texas Instruments for parallel developement. If TI had finished its chip and intel killed the 8008 (the exact reversal of what happened) TI would likely be the dominant player in the cpu field today, not Intel.

4004 was the first commercially available complete cpu in one package. But overall, its functionality was not unique, and did not sell well. It was a dead end chip. Intel is Intel because of the CTC designed 8008. The 4004 is a historic novelty.

 

[jimmysmitty]

Screw all that. My grandfather worked on punch based PCs in the 1940s so 😛

 

[FALC0N]

Did he really? If so, is he still with us? He must have some fun stories to share.

 

[jimmysmitty]

Yep. He is still kicking. Hes a ex WWII Marine so hes pretty hard to kill. Came out and became an accountant and worked on the punch card computers. Said it sucked if you screwed up. Then there was the Vacuum tube computers. Then the transistor hit and bam, ICs and them the x86.

Hell when I first started to live with him, his PC he had was one with a 8086 and the old DIP RAM.

 

[sonoran]

Screw all that. My grandfather worked on punch based PCs in the 1940s so 😛

I'm pretty sure your grandfather never worked on PCs with punch cards (since PCs never used punch cards). That said - my first computer program in college was done on punch cards. Thank goodness I talked my professor into signing me up for a TSO (terminal) account after that. Punch cards were a major pain in the arse - one mistake and you had to retype the whole card.

Back then computer programming went like this:

1) Write program (probably on a form looking something like this Fortran coding form)
2) Punch it on cards
3) Feed cards through reader
4) Wait (never knowing how long) for job to be run, and for results to print out on the line printer you specified, telling you how you messed up 😉
5) Fix mistakes and go back to step 2

BTW, the card readers only worked for cards in pristine condition. If they were folded, spindled or mutilated (you other geezers know where I got that phrase) they would not go through the reader. At least the card punch machines did have a "duplicate" function to make a copy of cards that were only slightly messed up.

 

[jimmysmitty]

^Fine computers. You know what I meant. He has worked on almost every version.

 

[FALC0N]

My uncle also had the honor of working with punch cards in the 60's.

Its kind of funny to think in 50 years people like us, maybe even us if we are lucky, will look back at what we have now in much the same way we look at punch cards.

I almost there as it is. I remember tape drives and 5 1/4 floppys on Vic20/C64/C128, atari 400/800/XL/XE, Apple II, TRS 80, etc. When 320*200, 256 color WITH a sound card was just the coolest gaming rig around! \

Ahhhh......the good old days. Before you know it, I will be telling my grand kids about how I walked 8 miles to and from school, uphill both ways and in 5 feet of snow year round! Even though it doesn't snow where I grew up.

 

[BadTrip]

My uncle also had the honor of working with punch cards in the 60's.

Its kind of funny to think in 50 years people like us, maybe even us if we are lucky, will look back at what we have now in much the same way we look at punch cards.

I almost there as it is. I remember tape drives and 5 1/4 floppys on Vic20/C64/C128, atari 400/800/XL/XE, Apple II, TRS 80, etc. When 320*200, 256 color WITH a sound card was just the coolest gaming rig around! \

Ahhhh......the good old days. Before you know it, I will be telling my grand kids about how I walked 8 miles to and from school, uphill both ways and in 5 feet of snow year round! Even though it doesn't snow where I grew up.

TRS-80 Was my first computer. Got it when I was 5 or 6.

 

[Hellboy]

TRS-80 Was my first computer. Got it when I was 5 or 6.

Geez mine was a Commodore Pet.. the one with the bonnet catch so you could lift the lid up with the monitor on...

why cant we have them like the good old days...

Maybe ill mod a old broken commodore...


ooh looking on ebay right now...

 

[amdfangirl]

Apple II of some sort was my 1st computer.

I'm not that old, but I am certainly poor 😉

 

[jonahkirk]

Vic 20 was mine. I'm expecting a big clock-to-clock upgrade with Thuban.

 

[noob2222]

Adam was the first computer I had. http://en.wikipedia.org/wiki/Coleco_Adam It had some really cool games for it for the time and the coleco and atari cartridges would work on it as well.

 

[Hellboy]

oh the days when a tv screen wouldnt show 80 characters with out serious eye strain.


and people moan that 1080 p aint good enough..

pah.

 

[FALC0N]

Speaking of eye strain, remember those "screens" you could get for monitors to reduce the flicker from those poor quality 50 or 60hz video displays? :lol: A different world it seems!

 

[fazers_on_stun]

No it wasn't. It was the 8008. The 8008 with its pre-x86 instruction set started intel on its meteoric rise. And it was a CTC design that was also given to Texas Instruments for parallel developement. If TI had finished its chip and intel killed the 8008 (the exact reversal of what happened) TI would likely be the dominant player in the cpu field today, not Intel.

4004 was the first commercially available complete cpu in one package. But overall, its functionality was not unique, and did not sell well. It was a dead end chip. Intel is Intel because of the CTC designed 8008. The 4004 is a historic novelty.

OK, but that wasn't the topic - the "first" microprocessor, not the "first successful" microprocessor .

I gather that Busicom didn't sell a whole lot of those calculators then. I had an uncle who worked in a department store in the mid-70's, and he mentioned the first 4-function, LED-display calculators selling for something like $400. IIRC he sold exactly one during the year he worked there, mainly due to the "wow" factor 😛....

 

[fazers_on_stun]

Heh, my first computer was a used Osborne 1 suitcase computer, complete with a whopping 128K of memory, two 5-1/4" floppies, and a ridiculous 5" CRT screen that could only display something like 52 columns of characters - used scrolling to show the entire 80 columns. Paid something like $300 for it since I had just started my career and couldn't afford something decent like an IBM PC, let alone a Mac. I recall learning Supercalc, and some primitive word processing program whose name escapes me, plus the guy I bought it from threw in one freebie game - Ladders I think - where your 'character' was literally just that - a 'p' or 'q' depending which way he was facing - and had to climb ladders & jump stuff to avoid rolling boulders (the letter "o"). Used C/PM that only took 3K space on a boot floppy, so that if you wanted to copy some files you had to use a separate utility disc with small programs. The keyboard attached to the front of the case whenever you wanted to move the thing - weighed something like 25 lbs, so that wasn't too often 😛. I guess it was the world's first "laptop" - if you had a lap the size of Oregon 😀.

The computer's Z80 could only address 64K of memory, so the extra 64K was used as the video screen buffer for snappy screen updates 😛.

I sold it a year later to some old guy who collected various antique computers, for what I paid for it 😛. It's probably worth much more than that as a collector's item by now.

 

[Wisecracker]

I'm pretty sure your grandfather never worked on PCs with punch cards (since PCs never used punch cards). That said - my first computer program in college was done on punch cards. Thank goodness I talked my professor into signing me up for a TSO (terminal) account after that. Punch cards were a major pain in the arse - one mistake and you had to retype the whole card.

Back then computer programming went like this:

1) Write program (probably on a form looking something like this Fortran coding form)
2) Punch it on cards
3) Feed cards through reader
4) Wait (never knowing how long) for job to be run, and for results to print out on the line printer you specified, telling you how you messed up 😉
5) Fix mistakes and go back to step 2

BTW, the card readers only worked for cards in pristine condition. If they were folded, spindled or mutilated (you other geezers know where I got that phrase) they would not go through the reader. At least the card punch machines did have a "duplicate" function to make a copy of cards that were only slightly messed up.

This. And don't forget the loose 'chads' 😀

My understanding is that certain 'mechanicals' in the Navy during WWII were set up for operation using punch cards without a computer - simply a card reader. This practice continued well into the 1960s before the ships were 'moth-balled'.

 

[bige420]

Gosh, I feel so young. My first pc was a Hewlett-Packard with an Athlon XP in it I believe, and 256mb of ram. I think it came with Windows 98 IIRC, I was so excited when we got it.

 

[keithlm]

Back then computer programming went like this:

1) Write program (probably on a form looking something like this Fortran coding form)
2) Punch it on cards
3) Feed cards through reader
4) Wait (never knowing how long) for job to be run, and for results to print out on the line printer you specified, telling you how you messed up 😉
5) Fix mistakes and go back to step 2

You forgot step #2a) Do not ever for ANY reason drop the cards on the floor and have them get unsorted. (Yes it happened to me one time.)

Anyway:

My first PC was an Altair 8800 with 4k of ram. It was later upgraded to 8k of ram and a paper tape to load the original Microsoft basic. I had trouble back then because I didn't know whether to spend my money on computer stuff or CB radio stuff. And making $1.35/hour as a busboy in a pancake restaurant didn't help.

In about 1978 or 1979 a friend bought a Motorola based SWTP and I was jealous because the Motorola chips were superior to Intel. (So I guess I started to prefer something non-Intel since the 70's.)

 

[FALC0N]

OK, but that wasn't the topic - the "first" microprocessor, not the "first successful" microprocessor .

Actually that particualr tributary of this thread had to do with Intel somehow deserving to be the dominant cpu maker for having developed the cpu. Not who developed the first microprocessor.

Your quote that started this was line of discussion was:

And for me Intel's "high crimes & misdemeanors" are just a silly thing to zero in on. As I said before, Intel invented the microprocessor, and AMD and VIA would still be selling ASICs and PGAs instead of cpus if not for Intel leading the way.

My counterpoint can be summarized as:

1) Yes, intel invented the first microprocessor, but it was just existing tech in a smaller package. It was so poorly implemented that almost everyone went with the old multi-stage design.

Calling this a seminal moment would be like calling the AMD phenom 9500 a seminal moment for being the first integrated quad core. Some things look better on paper.

2) The design that DID trigger their rise to dominance was entirely the work of CTC and OEM'd to Intel WITH R & D money from CTC. In fact, Intel's Bob Noyce, who negotiated the CTC deal was quoted as calling it "a dumb move". That what intel really thought of the microprocessor. Not exactly visionary.

The deal fell apart when Intel couldn't deliver in a timely fasion and CTC terminated the contract leaving intel to eat the R & D cost.

3) Faced with the prospect of eating the extensive R & D cost or finishing the last 10% or so on their own and TRYING to sell it, Intel chose the latter, hoping only to make up SOME portion of the thousands of dollars in losses they would surely take.

4) Once placed on the market, the chip unexpectedly took off and you know the rest...........


Take a close look. That is NOT the roadmap of the super deserving cpu pioneer "Leading the way" as you portray. In fact, had CTC NOT terminated the contract, the 8008 would have been CTC's intellectual property. And as I pointed out TI passed as well.

NOBODY wanted this chip. Intel got lucky in that unlike CTC they couldn't throw the losses at somebody else. Their ONLY choice was to send it to market and hope against hope it made more than it lost. Boy did it ever.

You know the old saying: Better to be lucky than good. Intel was both, but a lot more lucky.