Don Valentine has died • Hacker News Board

Re: Don Valentine has died

#21

Post by ChuckMcM » Sat, Oct 26, 2019, 5:35 AM UTC

servrite wrote: post #19

tyingq wrote: post #10

Earlier quoted context omitted.

Great share. I remember the inflection point that Opteron was. Essentially, it killed Sparc, PA-RISC, Alpha, Itanium, etc. That was a unique moment in time. Hearing a personal anectode between someone relatively junior at the time, and a legend, is very interesting to me. It was, really, the moment Linux/x86 "won". That 64 bit hurdle defined where we are today. At the time, it inspired me to make a pitch that "RISC w…

Opteron never killed SPARC. The changing market dynamics of Cloud did.

I was at Sun when SPARC was invented. And I think Sun killed it.

Every time someone started trying to take advantage of the "open" nature of SPARC, Sun would out compete them, fairly or not. It was a pretty open secret that Sun had hitched its wagon to the "best" processor for running Unix servers. As I told the leadership after I left that Sun was becoming DEC. They were retreating to the enterprise data center, nobody was buying workstations any more, and Solaris was not getting any more open. I suggested they build servers with the Opteron too :-).

Re: Don Valentine has died

#22

Post by jorgenveisdal » Sat, Oct 26, 2019, 5:46 AM UTC

Rest in peace Don!

Re: Don Valentine has died

#23

Post by gumby » Sat, Oct 26, 2019, 6:23 AM UTC

monocasa wrote: post #20

gumby wrote: post #14

Earlier quoted context omitted.

> At the time, it inspired me to make a pitch that "RISC was dead", despite it's technical superiority. Is it? The microarchitectures of the big MPUs is essentially RISC -- as it always has been, but microcode isn't written 100% by scratch any more. At the compiler level it's kind of true, and for what turned out to be generally good reasons such as compilers aren't as smart as Radin & co thought they'd be, or concom…

The CISC just became RISC on the inside thing is greatly overstated. CISC always decoded to simple more or less single cycle ops internally, that's how microcode works. The RISC shtick was to get rid of that decoding into simple ops in the first place. Originally that didn't make sense because those ops' fetch bandwidth would be competing with data bandwidth. But notice how RISC popped up the same time as ubiquitous…

> CISC always decoded to simple more or less single cycle ops internally, that's how microcode works. The RISC shtick was to get rid of that decoding into simple ops in the first place

Having written microcode myself as a wee lad I would call that a gross oversimplification. And the micro machines inside a modern CPU are themselves fiendishly complex; my point is that microcode is no longer completely hand-crafted which is the only level to which the “RISC survived” argument might (IMHO) hold.

For the reasons above I also don’t agree with your final point.

I have to say I am not familiar with the microarchitechture of any of the early large-scale single-chip CISC CPUs (my microcode forays were for much larger machines) so we may be speaking to some degree at cross purposes. But again I think you mischaraterize the 801 and it’s descendants.

Re: Don Valentine has died

#24

Post by melling » Sat, Oct 26, 2019, 8:25 AM UTC

“ his high school friend, Steve Wozniak.”

This can’t be right.

Re: Don Valentine has died

#25

Post by tyingq » Sat, Oct 26, 2019, 8:27 AM UTC

servrite wrote: post #19

tyingq wrote: post #10

Earlier quoted context omitted.

Great share. I remember the inflection point that Opteron was. Essentially, it killed Sparc, PA-RISC, Alpha, Itanium, etc. That was a unique moment in time. Hearing a personal anectode between someone relatively junior at the time, and a legend, is very interesting to me. It was, really, the moment Linux/x86 "won". That 64 bit hurdle defined where we are today. At the time, it inspired me to make a pitch that "RISC w…

Opteron never killed SPARC. The changing market dynamics of Cloud did.

The changing market dynamics of cloud were pretty tied to cheap 64 bit servers. Opteron forced that on everyone. Intel would have gladly dragged it on x64 and sold high margin, "not cloud friendly" Itaniums.

Re: Don Valentine has died

#26

Post by ahartmetz » Sat, Oct 26, 2019, 10:12 AM UTC

ChuckMcM wrote: post #18

jacquesm wrote: post #11

Earlier quoted context omitted.

Did he yield?

Effectively yes. He did let me know that if it didn't work out that I would bear the consequences :-). By that time in my career I knew that you're only as good as your previous decision, good or bad. It became NetApp's best selling product at the time. I don't deserve any credit for that of course, the folks to did the work to get it out the door and sell it get the credit. My job was to do what the Army engineers d…

For all the times "nobody ever got fired..." is used when discussing technology, this might be the first time that I read about a manager directly warning you why you might get fired if you don't buy, errr, IBM :)

Re: Don Valentine has died

#27

Post by jacquesm » Sat, Oct 26, 2019, 10:38 AM UTC

ChuckMcM wrote: post #18

jacquesm wrote: post #11

Earlier quoted context omitted.

Did he yield?

Effectively yes. He did let me know that if it didn't work out that I would bear the consequences :-). By that time in my career I knew that you're only as good as your previous decision, good or bad. It became NetApp's best selling product at the time. I don't deserve any credit for that of course, the folks to did the work to get it out the door and sell it get the credit. My job was to do what the Army engineers d…

That's really great. The reason I asked is because I've seen that tactic used just to get people to let off steam and then the manager would push through their own vision regardless of whatever was said. So I'm really happy to see that he not only took input from you but was willing to commit to it as well over his own opinion.

Agreed on the management interviews, any kind of power should always come with accountability in case it is mis-applied.

Re: Don Valentine has died

#28

Post by sokoloff » Sat, Oct 26, 2019, 11:49 AM UTC

melling wrote: post #24

“ his high school friend, Steve Wozniak.” This can’t be right.

That line also gave me pause, but I think it was meant to say that Woz was Jobs’ high school friend (which is also wrong, but much less wrong).

Re: Don Valentine has died

#29

Post by monocasa » Sat, Oct 26, 2019, 5:03 PM UTC

gumby wrote: post #23

monocasa wrote: post #20

Earlier quoted context omitted.

The CISC just became RISC on the inside thing is greatly overstated. CISC always decoded to simple more or less single cycle ops internally, that's how microcode works. The RISC shtick was to get rid of that decoding into simple ops in the first place. Originally that didn't make sense because those ops' fetch bandwidth would be competing with data bandwidth. But notice how RISC popped up the same time as ubiquitous…

> CISC always decoded to simple more or less single cycle ops internally, that's how microcode works. The RISC shtick was to get rid of that decoding into simple ops in the first place Having written microcode myself as a wee lad I would call that a gross oversimplification. And the micro machines inside a modern CPU are themselves fiendishly complex; my point is that microcode is no longer completely hand-crafted wh…

I've written microcode too, and there's two type of micorcode. Vertical and horizontal.

Horizontal is your wide microcode like seen on what I programmed, the KB11A CPU inside a PDP-11/45. It was somewhere around a hundred or so bits wide, and you could pretty clearly see "ok, these five bits just latch into this mux over here, these over here", etc. in the micro-architecture. I've seen between 96-bit adn 256-bit wide singel instructions here.

Vertical microcode is what you see in designs that the 801 was trying to get away from having a full CISC decoder for. Much smaller fixed length instructions that represent higher level ops, and are what RISC was trying to get rid of mainly.

The non ascetic CISC machines would normally have at least two microcode ROMs: one in the decoder, and at least one in the backend, maybe more depending on how the separated out their execution units.

So for instance 68K had: * Decoder microcode of 544 17bit instructions * Execution unit "nanocode" of 366 68bit instructions

An ARM1 had: * No decoder microcode (but 32bit wide, fixed width, aligned ISA instructions with a I$) * Execution unit microcode of 42 36bit instructions

Re: Don Valentine has died

#30

Post by chriselles » Sat, Oct 26, 2019, 6:46 PM UTC

LeonM wrote: post #2

If you want to learn more about Don and how influential his work has been, I recommend listening the recent Acquired episode about Don and Sequoia [0]. [0] https://www.acquired.fm/episodes/sequoia-capital-part-1

Great episode. Will listen to it again on a long walk today.

I’ve always sought commentary and quotes from Don Valentine to help me learn more about being an angel investor the same way I have have always Warren Buffett’s commentary and shareholder letters to learn how to be a better long-term investor.