Amiga.org
Amiga computer related discussion => General chat about Amiga topics => Topic started by: SIR-GERMAN on October 16, 2010, 06:39:05 AM
-
This is an interesting interview with the ex CEO of Apple , makes you wander about Amiga .
"I made two really dumb mistakes that I really regret because I think they would have made a difference to Apple. One was when we are at the end of the life of the Motorola processor… we took two of our best technologists and put them on a team to go look and recommend what we ought to do.
They came back and they said it doesn’t make any difference which RISC architecture you pick, just pick the one that you think you can get the best business deal with. But don’t use CISC. SISC is complex instructions set. RISC is reduced instruction set.
So Intel lobbied heavily to get us to stay with them… (but) we went with IBM and Motorola with the PowerPC. And that was a terrible decision in hindsight. If we could have worked with Intel, we would have gotten onto a more commoditized component platform for Apple, which would have made a huge difference for Apple during the 1990s. In the 1990s, the processors were getting powerful enough that you could run all of your technology and software, and that’s when Microsoft took off with their Windows 3.1.
Prior to that you had to do it in software and hardware, the way Apple did. When the processors became powerful enough, it just became a commodity and the software can handle all those subroutines we had to do in hardware.
So we totally missed the boat. Intel would spend 11 billion dollars and evolve the Intel processor to do graphics… and it was a terrible technical decision. I wasn’t technically qualified, unfortunately, so I went along with the recommendation."
http://www.cultofmac.com/john-sculley-on-steve-jobs-the-full-interview-transcript/63295
-
Ah, the beauty of hindsight. Nowhere near as challenging as its cousin foresight.Back then, when I'd try to look this far foward, what I'd see is very different to what is here today. Just as different as the things I confronted then, to the way I remember them, I'd say.
-
I wouldn`t regard it a mistake. Especially, back then PPC seemed to offer much more potential than the x86 platform. It was a clear cut RISC architecture.
Intel went to a lot of strains just to include some RISC functionality into it`s processors while maintaining compatibility. I believe that the demise of the PPC platform was more a matter of scale then anything else.
-
If he were techincally-orientated, then he would know why they went with PowerPC, though.
I don't know the facts, but I would guess the endianness of the PowerPC architecture had a lot to do with the choice to go that route. These days it's less of a problem, but in those days they needed all the power they could get, so byte-swapping would have slowed down legacy compatibility too much.
Hindsight is always 20/20 vision.
-
Without sounding like I am crowing, I tried to get Commodore then Gateway to x86 with Siamese and then PCI Amiga. Commodore would not listen and Gateway dumped the PCI Amiga project.
Trouble was, the technical software guys who talked about the beauty of the PPC missed the point, x86 had Intel behind it and was only going to get way better.
Still we are on x86 now, we just lack the Amiga badge for now.
-
A bad decission choosed PPC CPUs? if they choose at first the Intel family they neither have been the same level that Microsoft, pay more for one Apple with Intel? when a single and clone PC is more cheap that one MAC, the people there were done as currently buy a PC clone with Intel and install the MAC OS with the patched, the sucessful of MAC was the different with the PC clones and Windos, because its prices always have been and are very exapensives, very exclusives, having the MACs 68k, PPC or Intels.
-
The man is an idiot, and is just saying this now, after Intel has come out on top. You can bet that, if PPC had triumphed over Intel, this same man would be crowing about PPC was his own personal recommendation.
Back then, RISC, and PPC was by far the best choice for the future, was faster, more modern, based on he best technology, that is why the team of experts Apple put to investigate future recommended it.
Much of what he says doesn't even make sense
1990's, the processors were getting powerful enough that you could run all of your technology and software
What? So, before 1990 your processor couldn't run your software? Or they could run your software but not your technology? My Mac still doesn't run my technology even in 2010. Runs my software fine, though.
So we totally missed the boat. Intel would spend 11 billion dollars and evolve the Intel processor to do graphics…
What an absolute tool.
It's only now, with the Core i7 series.. that Intel processors have began having built-in GPU, and, like every Intel GPU before it, is abysmal in performance. This has no relevance whatsoever to a decision 20 years ago.
-
But now games (at least) are developed for Power PC console, but snub the x86 PC.
-
Let's not forget what killed the PowerPC in the first place :)
-
The guy's an idiot. Parents are probably idiots too.
-
The guy's an idiot. Parents are probably idiots too.
I confirm that his parents are idiot too :biglaugh: (perhaps also his grandparents!:hat:)
-
From my perspective, x86 chips didn't actually get "good" until 4 or 5 years ago. Before that, they were power-sucking, hot, and inefficient across the entire range. AMD was a bit better than Intel, but PPC seemed to offer better performance per clock cycle while using the least electricity.
Now, however, x86 has strong multicore architecture and a good product range that encompasses the powerful and the power-efficient.
Whether this is due to a business relationship with Apple is impossible to say, but I think Apple's decision to switch has been fully vindicated.
(Which does make me a tiny bit sad because I do have a soft spot for PPC)
-
Ah the frantic shrieking of the cult of PPC
-
Yup, hinesight is 20/20, and his is spot on. Intel was the big desktop CPU maker and they should have hitched their wagons to it and not to IBM/Mot with their history of delayed CPU releases. Five years later when AMD hit the market with the K-5s would have been the best time to switched to Intels and use the threat of using AMD as a way to beat down Intel's pricing. Might have been a different OS landscape had Apple gone Intel. Piracy is what gave M$ the upper hand in the desktop market, it might have been kind to Apple as well had there been a chance to bring Apple into more offices around the world as the computer revolution took root.
-
Sculley doesn't know shit about the computer industry, and didn't back then, either; he was responsible for the deal that gave Microsoft a perpetual license to the Macintosh UI, (http://folklore.org/StoryView.py?project=Macintosh&story=MacBasic.txt&sortOrder=Sort%20by%20Date&detail=medium) one of the principle reasons that Windows even got to stay on the market, and was a driving force behind Apple's policy of absurdly-priced computers for the elite when the PC clonemakers were busy saturating the market with comparatively affordable machines, leading them into the rut they spent half of the '90s in. As for this nugget of "wisdom," while I'm not really a PPC fan, it was definitely the better choice at the time, even if it didn't work out so well in the long run. Moving to the 386 and 486 Intel chips of the time would have meant even worse emulation compatibility than the first PPC Macs offered, and given that the Mac's userbase was all about software, that would have been complete suicide.
-
and was a driving force behind Apple's policy of absurdly-priced computers for the elite when the PC clonemakers were busy saturating the market with comparatively affordable machines,
How is this different from the way Apple is now?
"I'd like a tower, please."
"That'll be $3000."
"WTF?!"
One home PC build and a trip to insanelymac.com later, and you're $2100 richer and still have a mac.
-
Oh, no argument there. The only difference these days is that Apple has a large enough cult of True Believers to make it a sustainable business model. But back in the early to mid-'90s it nearly killed them. Had they priced more affordably, they might have avoided that problem and become more than just "the computer everybody who doesn't want a PC buys."
-
One of the things he's missing is how incredibly difficult it would have been to switch processors back then..
It was a huge job for OSX, but the processors at that time were fast enough to handle it...
But trying the same thing, maintaining some level of compatibility, with the CPUs available then???
It would have been possible to write an Intel version of MacOS that was totally incompatible. Then they could have required all the app vendors to port all their apps to the new platform, and with some conversion programs to handle the resource fork bit, they could have had an OS that probably would have been better than Windows at the time...
But I'm still not sure what that would have gained them....
Yes, Intel chips cost less and were getting faster by leaps and bounds.
But that wasn't, and still isn't, the market Apple wanted to compete in. And it would have alienated quite a few of their customers.
It's not like waiting, as they did, for the CPUs (and RAM and other issues) to make it easier has really hurt them. They're doing OK..
desiv
-
If Apple wanted to stay CISC they could have in-housed the 68060 and continued from there. No transition to PPC or x86 was necessary. I guess the '060 was just late to production.
-
Oh, no argument there. The only difference these days is that Apple has a large enough cult of True Believers to make it a sustainable business model. But back in the early to mid-'90s it nearly killed them. Had they priced more affordably, they might have avoided that problem and become more than just "the computer everybody who doesn't want a PC buys."
I dont know if 1 in 20 of all computers is a "true believer"
-
T
It's only now, with the Core i7 series.. that Intel processors have began having built-in GPU,
He was most probably thinking of MMX. But Apples problem in the 90s wasn´t the PowerPC. They were selling more than a dozen different models at the same time, so choosing the right one wasn´t easy and more importantly: they sold a lot of 68x00 boxes too and even those were more than twice as expensive as an similary specd Pentium.
Oh and maybe PCs even had a better OS: OS/2.
-
From my perspective, x86 chips didn't actually get "good" until 4 or 5 years ago. Before that, they were power-sucking, hot, and inefficient across the entire range. AMD was a bit better than Intel, but PPC seemed to offer better performance per clock cycle while using the least electricity.
Now, however, x86 has strong multicore architecture and a good product range that encompasses the powerful and the power-efficient.
Whether this is due to a business relationship with Apple is impossible to say, but I think Apple's decision to switch has been fully vindicated.
(Which does make me a tiny bit sad because I do have a soft spot for PPC)
AMD K8 Athlon 64 Mobile(1) in 2004 says Hi. IBM CPC 925's(PowerPC 970's NB) alone has a TDP of 30 watts. AMD K8 Athlon 64 Mobile has a 35 watt version i.e. "Odessa" (CG, 130 nm, 35W TDP(2)) up to 2Ghz.
1. At that time, I used Mitec 8355 laptop with Athlon 64 Mobile and ATI Mobility Radeon 9600 Pro.
2. Max power not typical power.
A 2GHz 970FX(90nm) consumes 39 watts maximum.
http://www-306.ibm.com/chips/techlib/techlib.nsf/techdocs/9DBF300EB19A60D287256E4B005E43EC/$file/970fx_thermal_an_7_20_05.pdf
Dated 2005.
By the time of 90nm for AMD, Turion MT-40 at 2.2Ghz has 25 watt version. Turion MT was first release around March 10, 2005.
-
The man is an idiot, and is just saying this now, after Intel has come out on top. You can bet that, if PPC had triumphed over Intel, this same man would be crowing about PPC was his own personal recommendation.
Back then, RISC, and PPC was by far the best choice for the future, was faster, more modern, based on he best technology, that is why the team of experts Apple put to investigate future recommended it.
Much of what he says doesn't even make sense
What? So, before 1990 your processor couldn't run your software? Or they could run your software but not your technology? My Mac still doesn't run my technology even in 2010. Runs my software fine, though.
What an absolute tool.
It's only now, with the Core i7 series.. that Intel processors have began having built-in GPU, and, like every Intel GPU before it, is abysmal in performance. This has no relevance whatsoever to a decision 20 years ago.
In the late 80s and early 90s, Intel has i860/i960 i.e. A VLIW CPU with 3D (fix function hardware) and SIMD(similar to MMX). This research was later recycled into Pentium's MMX. This pattern leads to the current position i.e.
1. Intel Itanium and it's VLIW/EPIC design.
2. Intel GMA IGPs.
3. Intel Larrabee.
Intel i860/i960 cGPU was used in SGI workatations as a graphic co-processor.
PS; Large parts of Microsoft Kinect's visual with depth recognition is done on XBox 360's AMD Xenos GpGPU. GpGPUs easily process images with depth.
-
Apple its just a small part of the worlds computers (overpriced and cuestionable harware quality at this times) then... who cares about apples words?
-
He was most probably thinking of MMX. But Apples problem in the 90s wasn´t the PowerPC. They were selling more than a dozen different models at the same time, so choosing the right one wasn´t easy and more importantly: they sold a lot of 68x00 boxes too and even those were more than twice as expensive as an similary specd Pentium.
Oh and maybe PCs even had a better OS: OS/2.
QFT.
This guy is looking at the success under Jobs and simply says: its because of the move to intel. Its not.
Apple came off life support because of the ipod, their case designs (the transparent fruity imacs), the consolidation of two model classes (home and professional) and their marketing department. Jobs products were still using PPC, but prior to Jobs Apple were simply producing bland PC look-alikes with a gaziilion models. In fact Amelio wanted to head down the same PC path by licensing Apple clones, basically doing what the PC was doing and charging more, instead of proving a point of difference.
The CPU itself did not hold back wider acceptance of Apple products. Even today Apple differentiates its products from generic PC's primarily by their stylish case design. Sure OS X is nice but its not a world a way from Win 7 and isn't why people switch over. And as for the actual chips inside, they are identical to what you can get on the windows platform, but pay double for, which is why Apple will not advertise hardware specs to promote and identify its models, whereas in the PC world, the specs are what identifies a model and IS the point of difference between models. And ofcourse marketing. There is no-one that does it better
-
Even back in the day Apple were under fire from the humble C64... :)
[youtube]wyiUH9NmX0U[/youtube]
[youtube]796KD4SNzwE&feature=related[/youtube]
[youtube]BdfxfrTXyvc&feature=related[/youtube]
(showed back then, Commodore had a sense of humor, at least...:) )
-
No, I don't think he was referring to MMX either, when he talked about Intel processors doing graphics. I sincerely think that he simply has no clue about computing or technology.
This gem from the same interview
It was totally remarkable that you could deliver a machine when you think the first processor on the Mac was less than three MIPs (Million Instructions Per Second), which today would be — I can’t think of any device which has three MIPS, or equivalent. Even your digital watch is at least 200 or 300 times more powerful than the first Macintosh.
If the first Mac had 3 MIPS (actually it didn't, it had an 8Mhz 68000 at around 1 MIP, but you can't expect him to know that - he was only the CEO of the company that produced the thing) and an average digital watch today is 300 times faster? He seriously thinks that an average digital watch runs at "at least" 900 MIPS?
The Freescale MPC8272 running at 400Mhz only gives 760 MIPS!
The man is in cloud cuckoo land. He then goes on to say that an i3 is rated at about 40,000 MIPS, so in his World 45 digital watches have the same processing power as an Intel i3
And then this...
Microprocessors in those days were so weak compared to what we had today. In order to do graphics on a screen you had to consume all of the power of the processor. Then you had to glue chips all around it to enable you to offload other functions. Then you had to put what are called “calls to ROM.” There were 400 calls to ROM, which were all the little subroutines that had to be offloaded into the ROM because there was no way you could run these in real time.
He doesn't even seem to be able to grasp the most basic concepts of how computers work. ROM is just storage - the processor still has to run the code that's on them, and it still has to run them in real time, and it did just that. "calls to ROM" are calls in your program code which call these routines stored in the ROM and then the processor runs the code.
Putting graphics on a screen consumed all of the power of the processor? He has obiviously never used an Amiga... hell even a 1979 Atari 800 had graphics chips (GTIA) which handled graphics and did sprites in hardware so as not to use the processor.
On the other hand, Steve led the development of what was called AppleTalk and AppleLink. AppleTalk was the communications that enabled the Macintosh to communicate to the laser printer that enabled… desktop publishing.
Appletalk was a networking protocol to network Apple computers - yes, a part of that protocol meant that you could share printers (just as you can today with TCP/IP) but it was not what 'enabled desktop publishing' by any stretch of the imagination.
What is surprising is not that he got fired, but that he ever got to be CEO in the first place. No wonder the company was in such dire straits at that time with this guy in charge.
-
AMD K8 Athlon 64 Mobile(1) in 2004 says Hi. IBM CPC 925's(PowerPC 970's NB) alone has a TDP of 30 watts. AMD K8 Athlon 64 Mobile has a 35 watt version i.e. "Odessa" (CG, 130 nm, 35W TDP(2)) up to 2Ghz.
1. At that time, I used Mitec 8355 laptop with Athlon 64 Mobile and ATI Mobility Radeon 9600 Pro.
2. Max power not typical power.
A 2GHz 970FX(90nm) consumes 39 watts maximum.
http://www-306.ibm.com/chips/techlib/techlib.nsf/techdocs/9DBF300EB19A60D287256E4B005E43EC/$file/970fx_thermal_an_7_20_05.pdf
Dated 2005.
By the time of 90nm for AMD, Turion MT-40 at 2.2Ghz has 25 watt version. Turion MT was first release around March 10, 2005.
As I said, my perspective - that doesn't necessarily mean it was factually accurate :)
Of course there were good x86 chips back then, but they were the high-end market exception rather than the norm. Remember how many machines were being sold with those awful Celerons?
Now, pretty much any x86 chip you buy will be a damn good processor.
-
He doesn't even seem to be able to grasp the most basic concepts of how computers work. ROM is just storage - the processor still has to run the code that's on them, and it still has to run them in real time, and it did just that. "calls to ROM" are calls in your program code which call these routines stored in the ROM and then the processor runs the code.
Putting graphics on a screen consumed all of the power of the processor? He has obiviously never used an Amiga... hell even a 1979 Atari 800 had graphics chips (GTIA) which handled graphics and did sprites in hardware so as not to use the processor.
Ooh lordy, now I almost regret not reading the interview. This is comedy gold. (And as an aside, even on the Mac, which didn't have any video hardware besides a simple framebuffer, it didn't take the whole power of the processor to do graphics; the thing would've been completely unusable if it had.) Guy doesn't have a clue.
-
So how come Apple didn't fracture into two camps x86 and PPC? I suppose the Apple user is a different beast to an Amiga user.
On another topic I don't think Apple even makes and effort to escape from it's delicate, sanitary institution. I'm glad they are picking up market share, but it is probably from disenchanted wintel users.
If you look for some programming options on Apple you will find empty forums and clueless people on the other end of a Apple store phone.
Okay so it's not the case with the iphone at least.
Apple gaining market share is the classic saying "Out of the frying pan into the fire."
-
Ooh lordy, now I almost regret not reading the interview. This is comedy gold. (And as an aside, even on the Mac, which didn't have any video hardware besides a simple framebuffer, it didn't take the whole power of the processor to do graphics; the thing would've been completely unusable if it had.) Guy doesn't have a clue.
It just makes one wonder. Why would Jobs hire a sugared-water company CEO? Was there some hidden genius logic in that he must be brilliant in selling a 90-year old sugared-water brand around the world, so he must have the mad skillz to sell computers around the world?
I mean, I would think Sculley's contacts would be 50% bottling plants around the world, cola nut growers, 40% marketing/branding people and ad agencies, and perhaps 10% the other non-food industry contacts.
How did his skills/experience fit into Apple's plan of dominating the world of personal computers?
Sculley (to board): Sure, yeah, I know this guy who has a bottling plant in Argentina, I could tell him to mold the casings for us. I'm sure we could use bottle crate plastic for the new models. We could probably even tell him to use the same colors as for the crates. I hear in Argentina they're baby blue crates.
Years later Jobs brings out the baby blue iMac; thanks to Sculley :)
-
So how come Apple didn't fracture into two camps
They did twice: first NeXT, later BeOS. But only the main branch survived. Well ... you could argue that NeXT survived and BeOS isn´t entirely dead.
If you look for some programming options on Apple you will find empty forums and clueless people on the other end of a Apple store phone.
You asked a sales guy about programming? There are a lot of develloper resouces for Mac and iPhone. You can easily get advice on the mailing lists (even sometimes stuff that is still under NDA, but don´t count on that). Remember though that Apples platform isn´t about choice: Cocoa is king for them. Oh, and et´s hope they don´t sue MUI for patent infringement. ;)