Calling Last Release

[Kostask]

........
For those who remember, IBM was late to the market with 386-based systems because it had trouble managing the heat generated by the 386. This helped to end IBM's domination of the IBM-compatible personal computer space.
......

IBM lost its dominance in the PC space because it got greedy. Specifically, when the PC industry started going to higher speed expansion busses (EISA), IBM decided that it was going to do its own thing and try to lock in its customers with Micro-Channel (MCA). In one stroke, it cut itself off from the vast majority of expansion cards, and expansion card manufacturers declined to manufacture MCA expansion cards in favour of EISA expansion cards. IBM's MCA systems also no longer had ISA cards (the "standard" 8 and 16 bit cards used by the majority of PCs (XT and AT) until the EISA and MCA busses showed up), additionally cutting off users with any investment in ISA cards (which could be used in EISA systems). IBM aldo patented most of critical technologies used in MCA, and wanted to charge add in card makers licensing fees. After a while, IBM stopped making desktop PCs all together, building only the Thinkpad line of laptops, which it ended up selling off to Lenovo.

 

[MisterMe]

I have been a Mac user since I got a hand me down Macintosh Plus way back in the day, I was around for the 68k to PowerPC transition. I know the benefits that PPC brought to the Mac and I know that in it’s day it was fantastic hardware. I’m not disagreeing with you at all with that, I have a MDD PowerMac G4 set up on my desk right now. I’m just saying that at the time, PowerPC was no longer a viable platform and the switch to Intel was much needed. It was more than just power consumption and heat, it was about embarrassing low spec numbers that Apple was stuck defending. Steve even comments about his 3 GHz promise in the keynote announcement.

Reading spec sheets can be terribly misleading. I judge systems by their computational power and stability.

IBM lost its dominance in the PC space because it got greedy. Specifically, when the PC industry started going to higher speed expansion busses (EISA), IBM decided that it was going to do its own thing and try to lock in its customers with Micro-Channel (MCA). In one stroke, it cut itself off from the vast majority of expansion cards, and expansion card manufacturers declined to manufacture MCA expansion cards in favour of EISA expansion cards. IBM's MCA systems also no longer had ISA cards (the "standard" 8 and 16 bit cards used by the majority of PCs (XT and AT) until the EISA and MCA busses showed up), additionally cutting off users with any investment in ISA cards (which could be used in EISA systems). IBM aldo patented most of critical technologies used in MCA, and wanted to charge add in card makers licensing fees. After a while, IBM stopped making desktop PCs all together, building only the Thinkpad line of laptops, which it ended up selling off to Lenovo.

You have it backwards. IBM had already lost the PC-compatible battle when it introduced MCA. MCA was its futile attempt regain control. IBM lingered in the PC space for a few years until it sold its operations to Lenovo. More than just the Thinkpad went to Lenovo.

 

[Boil]

The Apple ][ started it all for me...!

 

[Spock]

The Apple ][ started it all for me...!

Apple II FOREVER!!

 

[Boyd01]

Got my Apple ][ on June 22, 1978 - $1225 for the 16K model. Disk drive hadn't even been introduced yet!

 

[Kostask]

My first Mac was a MacII that the company I was working for had bought for the competitive analysis lab. I was given an opportunity to buy it at $CAN20 in 1988, because nobody else wanted it.

 

[MisterMe]

My first Mac was a Mac SE/30 back in 1989. It belonged to my department and had been damaged by the departmental secretary's children. I was given permission to have it repaired. Before the year was out, I purchased a Mac Iicx as the first computer that I owned. I still have it although it has not booted in decades.

 

[the8thark]

The Apple ][ started it all for me...!

Got my Apple ][ on June 22, 1978

My first Mac was a MacII

My first Mac was a Mac SE/30 back in 1989.

Looks like this topic has degenerated into braging about how old we all are. I'll join in.

My first Mac was in the mid to late 80's. Mac 128k (that was upgraded to 512k)

 

[dmccloud]

Steve Job's stated reason switching to Intel is that Intel had developed a process for producing cool chips while IBM was not interested in producing processors that were suitable for consumer products. This is the reason that Apple was never able to produce a PowerBook G5. Intel's cool chip manufacturing process was a surprise based on its history going back to the 386. The 386 ran hot. For those who remember, IBM was late to the market with 386-based systems because it had trouble managing the heat generated by the 386. This helped to end IBM's domination of the IBM-compatible personal computer space.

Saying "PPC really fell behind" is a very broad statement. Where it fell behind was in the area of heat production. It did not so much fall behind as IBM has always thought of itself as the producer of Big Iron. It did not concentrate on designing processors suitable for the consumer market. After Jobs announced the switch to Intel, then IBM suddenly saw the error of its ways. By then, it was too late.

However, heat production is not the only system metric. Having used a number of Intel-based systems going back to the 1980s and more recent systems going back to 2007 or so to present and having used PPC systems going back to 1994, I can state without fear of contradiction that PPC-based systems are much easier to live with than Intel. By this, I mean that my PPC-based Macs allow me to get my work done without hassle. I find the difference is subtle, but Intel-based Macs do not age as well as PPC-based Macs.

You forgot the part where IBM was incapable of ramping up the clock speeds of their PPC parts, so if Apple had remained on that platform, the Mac would have fallen even further behind Intel when it came to real world performance. Heat production was a minor blip compared to the lack of performance.

 

[Waragainstsleep]

Didn't IBM break a speed record not long after this though? It got little to no attention at the time but I'm sure they managed to clock what would have been a "G6" at something mad like 5-6GHz. Maybe I'm just going nuts and my memory is shot.

 

[Joe The Dragon]

You forgot the part where IBM was incapable of ramping up the clock speeds of their PPC parts, so if Apple had remained on that platform, the Mac would have fallen even further behind Intel when it came to real world performance. Heat production was a minor blip compared to the lack of performance.

they also did not get the G5 to an point that would work in an laptop.

 

[Boil]

Was it the dual G5 Power Mac that had the water-cooling...?

 

[rezwits]

My Xcode Testing Beds

(Mac Pro 2008 & MacBook Pro 2011/17) end OS El Capitain 10.11.6
Xcode 6.4 (Yosemite)
Xcode 7.3.1 (El Capitain)
Xcode 8.2.1 (Sierra)

(Mac Pro 2010) end OS High Sierra 10.13.6
Xcode 8.3.3 (Sierra)
Xcode 9.4.1 (High Sierra)
Xcode 10.1 (Mojave)

(Mac Pro 2013) end OS Catalina 10.15.6
Xcode 10.3 (Mojave)
Xcode 11.6 (Catalina)
Xcode 12.y (Big Sur)

(MacBook Pro 2015/15 & MacBook Air 2015/11) end OS 10.17.z = 11.1.z
Xcode 12.z (Big Sur) 10.16.x = 11.0.x (2020)
Xcode 13.z (Future macOS 1) 10.17.x = 11.1.x (2021) (wanted Universal2)
Xcode 14.y (Future macOS 2) 10.18.x = 11.2.x (2022)

(MacBook Pro 2019/15 & MacBook Arm 11 or 12) end OS 10.19.z = 11.3.z
Xcode 14.z (Future macOS 2) 10.18.x = 11.2.x (2022)
Xcode 15.z (Future macOS 3) 10.19.x = 11.3.x (2023)
Xcode 16.y (Future macOS 4) no 10.20.x, will run Xcode (last for Intel)

(My Future Arm Macs 15 and, 11 or 12) (2022 Sept/Oct)
Xcode 16.z (Future macOS 5)
...

So I am thinking:
last Intel macOS will be 11.3.z aka 10.19.x (2023)
last Intel Xcode will be 16.x.y (2024 wwdc, could be 15.z)
16.x.y is saying a 2019 MacBook Pro, has an Xcode shelf-life of <6 versions?!? That's kinda low, so...

n.b. the above "end OS" are just where I ended most of them for development testing, I know some can go further...

Laters...

 

[thunng8]

Didn't IBM break a speed record not long after this though? It got little to no attention at the time but I'm sure they managed to clock what would have been a "G6" at something mad like 5-6GHz. Maybe I'm just going nuts and my memory is shot.

The IBM POWER6 was released in 2007 at 5ghz. It blew away Intel server chips at the time.

POWER6 - Wikipedia

en.wikipedia.org

It wasn't appropriate for laptops though because it was very power hungry. IBM concentrated on the enterprise and that where it excelled being more than double the speed of the equivalent Intel server processor at the time (in enterprise benchmarks like Database etc.).

IBM has continued with the lineup and just recently announced the POWER10. It is a monster - but this time they also concentrated on performance/watt, not just outright performance. Performance is still well over 2X faster than Intel or AMD per core (but it is using many more threads per core as each core can execute 8 threads).

Edit:
If you want a bit of history. The PowerPC G5 is based on the POWER4 chip. The POWER5 added SMT (multithreading) and on chip memory controller which made memory access much faster.

We do not know what IBM presented to Apple about future chips they were going to design for Apple - whether the G6 would be based on POWER5 or POWER6.

 

[Waragainstsleep]

The IBM POWER6 was released in 2007 at 5ghz. It blew away Intel server chips at the time.

POWER6 - Wikipedia

en.wikipedia.org

Thats the one. Nice to know I'm not losing the plot. Yet.

 

[MisterMe]

You forgot the part where IBM was incapable of ramping up the clock speeds of their PPC parts, so if Apple had remained on that platform, the Mac would have fallen even further behind Intel when it came to real world performance. Heat production was a minor blip compared to the lack of performance.

The clock speed problem is really another description of the heat generation problem. The PPC to Intel transition began about 15 years ago. The clock problem may have been an explicit issue, but I don't recall. I will say this: When the transition began in 2005, my computer at work was a 2.5 GHz Power Mac G5. As I write this, I am using a 2015 MacBook Pro sporting a 2.8 GHz i7. As of this writing, the top-of-the-line MacBook Pro is the 16" MacBook Pro based on the 2.3 GHz i9 TurboBoost to 4.8 GHz. The top-of-the-line Mac is the 2019 Mac Pro sporting a 2.5 GHz 28-core Xeon W Turbo Boost to 4.0 GHz. The fastest clock speed Mac Pro sports a 3.5 GHz Xeon W Turbo Boost to 4.0 GHz.

What you see is that Intel clock speeds are at best marginally faster in 2020 than the IBM PPC G5's were in 2005. It is interesting to note that the top-of-the-line original Mac Pro introduced in 2006 sported two 3.0 GHz Xeon 5100 Woodcrest processors--a faster clock than today's top of the line Mac Pro 14 years later. There is, of course, much more to processor performance than clock speed. Nobody knows this better than Apple.

As I said above, I don't recall clock speed being part of Apple's consideration for switching from PPC to Intel. Certainly Apple would have known had access to Intel's roadmap for clock speeds. As now with the transition to Apple Silicon, fan boys back in 2005/2006 said a lot of silly things related to the PPC to Intel transition. The clock speed discussion sounds like a fan site discussion, not an Apple discussion.

 

[dryjoy]

The clock speed problem is really another description of the heat generation problem. The PPC to Intel transition began about 15 years ago. The clock problem may have been an explicit issue, but I don't recall. I will say this: When the transition began in 2005, my computer at work was a 2.5 GHz Power Mac G5. As I write this, I am using a 2015 MacBook Pro sporting a 2.8 GHz i7. As of this writing, the top-of-the-line MacBook Pro is the 16" MacBook Pro based on the 2.3 GHz i9 TurboBoost to 4.8 GHz. The top-of-the-line Mac is the 2019 Mac Pro sporting a 2.5 GHz 28-core Xeon W Turbo Boost to 4.0 GHz. The fastest clock speed Mac Pro sports a 3.5 GHz Xeon W Turbo Boost to 4.0 GHz.

What you see is that Intel clock speeds are at best marginally faster in 2020 than the IBM PPC G5's were in 2005. It is interesting to note that the top-of-the-line original Mac Pro introduced in 2006 sported two 3.0 GHz Xeon 5100 Woodcrest processors--a faster clock than today's top of the line Mac Pro 14 years later. There is, of course, much more to processor performance than clock speed. Nobody knows this better than Apple.

As I said above, I don't recall clock speed being part of Apple's consideration for switching from PPC to Intel. Certainly Apple would have known had access to Intel's roadmap for clock speeds. As now with the transition to Apple Silicon, fan boys back in 2005/2006 said a lot of silly things related to the PPC to Intel transition. The clock speed discussion sounds like a fan site discussion, not an Apple discussion.

As I recall, it was really about heat, but the headline intel clock speeds were seen as an additional marketing advantage, i.e. no matter whether PPC systems might measurably outperform intel based systems at the same clock speed, the computer buying public would see higher clock speeds advertised for Intel systems, and assume they were more powerful than the Mac equivalent.