Comparing Apples and Amigas

[WolfToTheMoon]

As I understand it (but I can't find a corroborating source for this), the only reason that Apple didn't make that decision sooner was that Steve Jobs fell out with Intel because they wouldn't give him the good deal he was looking for.

Steve Jobs was not running Apple when they went PPC.

John Sculley did. And he says it was his biggest mistake.

When Jobs arrived back in Apple in 97', he pretty soon became dissatisfied with Motorola(and Motorola was pissed of when Jobs killed of the clones market)  and PPC and started making plans on switching to Intel.

[Duce]

Apple couldn't get the volume out of PPC that they needed, and when Jobs did come back he killed the clone market and made the Intel push - with a lot of niggling issues the entire way with Intel itself.

Jobs wanted X chips at X price that was far under what any other vendor was given deals wise for that sort of bulk, but they did manage to get to common ground and I suspect Apple and Intel are both glad of that.  More inportantly, I think Jobs likely made some threats in a paper tiger form that he'd flip to AMD or another x86 vendor and Intel simply called his bluff knowing Jobs would never have settled for anything but the "Cadillac" of the chip vendors, which was and still is, Intel.

Most of this is well documented in the various books and rags of the era.

[Sean Cunningham]

A big part of that mistake was believing Motorola and IBM could be partners.  Those two plus Apple getting together...you couldn't find three dissimilar tech companies with entirely different ideas or priorities.  They tried, with Somerset, but the writing was on the wall very quickly that things were likely going to go the way they did.  Scully had no choice but to go down that path though given the choice made before he was there to hitch Apple's wagon to Motorola in the first place.

Mistake or not the move to PPC was a logical one and the decision by Jobs to stick with it jumping to the G3 was also kind of a no-brainer at the time.  There was no other move to make that offered a transition to a real, modern OS.  He already had to "cut bait" and cancel the failed internal effort to update OS9 to something worth keeping around, sent the Newton team packing, there was simply no way to move to Intel when he took hold of the company again until some years later when it was the only decision that made any sense at all.  Once users were weened off OS9 they had options.  Not before.

That took as long as it did largely because big software developers are lazy (ie. Adobe being horribly, horribly lazy).

Motorola was/is an engineering-centric company that doesn't understand software, or anything much beyond the component level use of their tech and whose bread-and-butter was non-PC implementations of their technology.  Apple might have been the most high profile public PPC customer but they were just a fraction of the PPC market.  They were the tail trying to wag the dog.  That and, though they're not unique in this, they (Motorola) were and are a company run by Lumberghs.   Freakin' idiots.  And so they will continue to spin off parts of their company that have any value until Motorola will simply cease to exist.  Because they're idiots.

IBM could also not give a crap about much of anything but embedded applications which is why they didn't care about Altivec regardless of the fact that Apple's products were more than a little dependent here.   Embedded applications didn't need that kind of floating point acceleration.  The rub was IBM had the better manufacturing technology and could deliver a better version than Motorola, they just didn't care.

The switch to Intel was something he had in his back pocket the entire time because NeXT had already been running on "white hardware" for years and OpenStep was already powering a majority of the behind-the-scenes enterprise level web technology.  Apple was coy about this for years for the sake of their tenuous alignment until it was simply idiotic for them to stick with it given how utterly disappointing the G5 was and no indication that they could ever get back to being competitive in markets they had been dominating that were starting to slip.

[nicholas]

simply idiotic for them to stick with it given how utterly disappointing the G5

I hear this a lot but my 2008 Macbook Pro with 2.5GHz Core2 Duo 'Penryn' and 6GB RAM is not half as nippy as my 2005 PowerMac with 2.7GHz G5 and 4GB RAM when it comes to running Logic Pro.

Or are you referring to it's power consumption?  There the Core2 wipes the floor with the G5.

[amigakid]

It's simple, there really isn't any hardware/software forward momentum going honestly.  Yeah we have peg, morph, amigaone and so on but the cost of new hardware is so high and honestly Amiga is a hobby or passion for most users, and most users (such as myself) cannot afford an Amigaone.  Hardware is expensive compared to the PC side (and even Mac side) and there isn't much support compared to other platforms.  I agree if Commodore hadn't gone under and where still producing hw, then there is a possibility that we would also have an aPhone and so on.  Nice thread tho!

[Sean Cunningham]

Some of that could be the coding of Logic Pro.  I don't know that app specifically but I know the maker of ProTools was one of the "lazy" application developers I was talking about back in the OS transition periods.  Every single Adobe app, when you could dual-boot OS9 and OSX, was faster in OS9 because they hadn't actually written an OSX version, not really.  

It's been so long since I cared to keep track of the compatibility layers, it could be a lack of optimization in the app or it could also be a flaw in their earlier Core2 Macbook.  2008 marks a year where there's a performance difference in some applications between similarly spec'd Core2 models due to chipset implementation and memory.  I have one of the effected iMacs.  Just unlucky I guess.  It was what was available when I went to buy in 2008.

Power and heat was a problem though, yeah.  It pointed to a cap they were going to hit on the performance they could expect in the future.  It didn't inspire confidence in their ability to keep scaling.  It had been similarly an issue with the dual-G4 desktops as well.  We got lucky when I built our Cinewave and I got the dual 1K G4 that I did and not the next generation.  It ran flawlessly for our configuration while folks with newer, slightly faster systems were having more issues and then more issues still with the G5.

[agami]

Ahh, revisionist history at its finest.
Y'all are off your rockers, 'crept for @nicholas.
The only thing intel was good at in the late '90s and early 2000s is pumping out low cost 32bit CPUs. The Pentium III and especially the Pentium 4 were relatively poor performers. And it's not so much just about the CPU, the Northbridge and Southbridge architecture from intel wasn't up to scratch either (Nvidia and AMD had better chipsets). They were floundering, AMD beat them to the 1GHz mark, and AMD beat them to the 64bit extensions mark, and AMD beat them to the multi-core mark, and the Opteron multi-pipeline design was mopping the floor with more expensive Xeon counterparts in the physical and virtual server space up until about 2009.

Fact was, the G3 (750), the G4 (74xx), and especially the G5 (970) and their specific Apple integrations outperformed anything intel was selling at the time.
2000 - My PowerBook G3 @ 333Mhz ran Adobe Premiere smoother than a dual PIII 733MHz.
2003 - I did a comparison in ripping a CD to mp3 between my PowerBook G4 @ 800MHz and a Toshiba Portege with Pentium M @ 1.6 GHz and the PowerBook beat it by a few seconds.
2003 - When the PowerMac G5 and later the Xserve G5 were released they were the foundation of several of the Top 10 supercomputing clusters of that time.
It wasn't until only a couple of years ago that a 4-core intel CPU outperformed my PowerMac G5 Quad @ 2.5GHz in converting video. And the PowerMac G5 Quad was released in late 2005.

But yes, as @nicholas stated, in terms of TDU, performance per watt, and small dies suited to mobile computing, intel's Core architecture is pretty impressive. But it still isn't the undisputed king, AMD is better at integrating GPU cores on the same die, ARM is better at sub-10w computing, and Nvidia is doing things with Tesla that intel is only dreaming about with the Xeon Phi. Cadillac my ass

[KimmoK]

Perhaps it would not have worked out, but if Apple had let PA Semi continue PPC, we might have had very cool 16core desktops now...

[Sean Cunningham]

You're not above revisionism either it seems, agami.  Adobe Premiere, in the G3 era, wasn't even frame accurate.  It ran very poorly, even though DV wasn't a terribly taxing format and even the slowest G3 offered realtime playback of the media.  In 2001 I co-authored one of the first 24P HD narrative features and we did the roughcut with DV proxy material in Premiere on my G3 while raising money to eventually build our dual-G4 Cinewave for online HD editing not quite a year later and switched to FCP because of how not up to the task Premiere was at the time(*).

Part of that was very niggling issues with Quicktime during the 1999-2003 period and a lot of third party vendors having issues but Premiere for Mac had languished at Adobe.  Rather than answer the assault from Final Cut Pro they killed the app for a number of years and I held my breath waiting for them to do the same with After Effects because not only did it work better under OS9 than OSX but the Windows version on Intel and AMD hardware killed After Effects performance on the fastest dual-G4 you could get at the time.  Photoshop users were likewise jumping to ship in droves and not because all of a sudden folks just up and decided to give Windows a try.  The hardware was much faster and a helluva lot cheaper, with AMD offering slightly better floating point performance per dollar.

This is why longtime Mac fanatics like Stu Maschwitz grudgingly switched The Orphanage over to NT because After Effects was their bread and butter and it was just too slow on the best Mac hardware available at the time (compared to what was available to run it under Windows).  I stuck with Mac as my main interactive machine but built a couple dual-AMD boxes to offload rendering to because they were so much faster.  I didn't have clients sitting on my shoulder so I just continued using it in the environment I preferred until it more or less caught up again and Adobe re-invested in making their apps on the platform.  

I was really happy when they re-introduced Premiere to Mac because I've always preferred Premiere to FCP.


(*) to put Premiere's poor performance on the G3 into perspective, it actually worked better and smoother a couple years prior on my DEC Alpha through FX32, which allowed the DEC RISC chip to run Intel Windows software through continuously optimizing and re-optimizing emulation, controlling a DPS Perception Video.  So that box was running Premiere through emulation talking to specialized hardware doing realtime A/V (AVI based) and it worked quite well and never introduced trimming errors like were rampant on the G3 + Premiere combo (Quicktime based).

[agami]

@Sean Cunningham
I don't doubt or negate anything you say. My example for the PowerBook G3 was subjective; My brother had the dual PIII Windows 2000 workstation on which he was doing various 2D/3D multimedia work. I come home one day and curiously find him using my PowerBook G3 and he says that the video stutters in Premiere on his dual CPU Windows box.

More raw CPU power? Definitely.
Let down by Windows et al? Perhaps.
Smooth end-to-end video editing experience? Nope.

Just sayin' is all.

[gertsy]

Comparing Apples with Amigas?
Come on guys, the only reason Jobs, Wozniac and Wayne chose the brand name Apple is because they knew you can only compare Apples with Apples.