Shop OBEX P1 Docs P2 Docs Learn Events
8-Bit 16-Bit 32-Bit Whats the difference? - Page 5 — Parallax Forums

8-Bit 16-Bit 32-Bit Whats the difference?

1235»

Comments

  • evanhevanh Posts: 15,126
    edited 2013-02-07 05:53
    Heater. wrote: »
    Legend has it that the 68xxx line was falling behind in performance. The power PC was pretty hot at the time so they had little choice.
    As luck would have it for them the PPC did not work out either so Intel it was.
    All of which was god practice I guess as they could easily go ARM.

    In effect, changing memory models is a similar step. Software has to be recompiled. AMD64 is a bigger step. All that bloatware does actually do something useful from time to time.

    As for the falling behind factor, it's a simple case of funds. Intel overflows with funds while everyone else starves. IBM have some financial strength behind the PPC but isn't going to go for broke, they know it's a no-win situation. Just one more chicken or egg situation. One born of establishment. You can't make an amazing design without the funds, and you can't get the funds without the amazing design.

    Take MRAM for example, in it's current state-of-the-art it's a real DRAM killer right now ... but the tech doesn't have the funds to ramp up to DRAM like production rates so thereby can't be produced as competitive prices. That, and DRAM is doing a great job as is so very few are looking for a replacement. So, MRAM stays in the shadows being used on corner products only. If MRAM can be up'd another notch and match SRAM for speed then things will likely change.
  • evanhevanh Posts: 15,126
    edited 2013-02-07 06:16
    potatohead wrote: »
    @Localroger: " It was only commercially successful because IBM. But there it was." Yeah, there it was... And I really do think "IBM" had way more to do with that whole thing than anybody really wants to recognize.
    Yep, IBM's big contribution was sitting back and letting the clones in.

    This whole mobile / ARM business is mostly about going backward now. "All day" type computers that can do lots of stuff, and or that really narrow down choices. It's kind of bizarre. All my earlier comments about that "Apple ARM" type machine sort of make sense, but not quite. Going backward, limiting choices makes some sense, but only so much. Are we really going to fragment people that way? Interesting times.
    Probably yes, and at the same time legislated stronger and stronger laws that prevent the user having hardware access. Where everything tends toward centralised continuously monitored control - big brother but where it's the corporations doing the control with USA government grabbing big chunks when they discover how much is being hoarded. The Web won't be the Web any longer - it becomes a mass of script-kiddie spying applets ... oh, wait, that's already here.
  • evanhevanh Posts: 15,126
    edited 2013-02-07 08:11
    On the subject of a "standardised" design. Just been rereading the postings ... What the XT and AT could be referred to as is a reference design. For the clone makers it served that purpose. Just the same as they do today.

    That was not a done thing back then. And in some ways has never happened again since IBM did it. The modern reference designs all come from chip makers, not computer makers.

    Although, there is now the newly forming open hardware groups too I guess. Again, though, not likely originating from computer makers.
  • evanhevanh Posts: 15,126
    edited 2013-02-07 08:36
    Intel, flush with untold billions in reserve, happily snatched away the providing of motherboard reference designs with the release of PCI and the Pentium 1. Even deploying the marketing monster of "over 50 MHz needs special design skills that most board makers don't have" type scare tactics just to ensure investors would strongly encourage designers go the Intel way ...

    Not that I have a problem with the results. The PC industry seriously needed to modernise at the time - This self important juggernaut that was seriously holding computing in the dark ages.
  • localrogerlocalroger Posts: 3,451
    edited 2013-02-07 12:24
    Tor wrote: »
    It sounds as a likely hypothesis, but there is no evidence at all that this was what happened. Remember also that the first Motorola Apple (the Lisa) came on the market in January 1983, much later than the 1981 IBM PC.

    The first Motorola Apple was the Apple I. The 6502 was a Motorola chip. Every single box Apple ever made was Motorola until they switched Mac to the PowerPC.
  • ctwardellctwardell Posts: 1,716
    edited 2013-02-07 12:36
    localroger wrote: »
    The first Motorola Apple was the Apple I. The 6502 was a Motorola chip. Every single box Apple ever made was Motorola until they switched Mac to the PowerPC.

    I suppose you could say that, but I would say had the team never left and formed MOS and created the 6502 the Apple may never have happened.
    Motorola had the 6800 priced way too high for it to have taken off, the 6501 and 6502 forced the prices down to a useful range.

    C.W.
  • TorTor Posts: 2,010
    edited 2013-02-07 12:55
    localroger wrote: »
    The first Motorola Apple was the Apple I. The 6502 was a Motorola chip. Every single box Apple ever made was Motorola until they switched Mac to the PowerPC.
    No, the 6502 was not a Motorola chip. It was a MOS Technology chip. The first Motorola Apple was the Lisa.

    -Tor
  • evanhevanh Posts: 15,126
    edited 2013-02-07 16:57
    ctwardell wrote: »
    Motorola had the 6800 priced way too high for it to have taken off, the 6501 and 6502 forced the prices down to a useful range.
    That's a likely reason for IBM choosing the x86 over the 68k. Motorola, like Digital Research, figured they had control. And they did while the market was a known.

    But no-one saw the clones coming the way they did.
  • evanhevanh Posts: 15,126
    edited 2013-02-07 17:06
    Just thinking a bit more about the reference designs thing. When Intel jumped in with PCI the 486 PCs were in full swing and volumes kept increasing. The games market was opening up for the PC. VESA LocalBus, tightly coupled to the 486, was a key factor in this. And something like that needs a reference design, making the jump from AT clones, for the clone makers to work from.

    I'm gonna guess the "Gang of nine" served this function, and presumably benefited directly from the clone makers. IBM wasn't in the game any longer, it had changed back to a royalty licensing model with the PS/2 and Microchannel.
  • localrogerlocalroger Posts: 3,451
    edited 2013-02-07 19:27
    Wow, MOS. I had completely forgot they were separate from Motorola. Well the tech was very similar and IBM wanted to be different, so the basic logic of IBM's decision still probably follows. They did not want to in any way be "like" any of the existing major competitors. I've programmed both 6800 and 6500 series assembly language and they are extremely similar, especially when you consider contemporary alternatives like the 1802, 8080, Z80, or TMS9901.
  • evanhevanh Posts: 15,126
    edited 2013-02-07 20:40
    Put a killer app on an open design and you get an unstoppable establishing process.

    And, because it establishes into culture, any further competing open designs are naturally stunted. Closed designs never have a chance outside of niches.

    I'm starting to see where the confidence of open software advocates is coming from now. And that also explains a lot about the amount of draconian laws being past to slow it down. There does appear to be a slightly secret war going on. Or at least the real reasons are not being aired when making the case for laws.
  • potatoheadpotatohead Posts: 10,253
    edited 2013-02-07 21:23
    You got that right evanh.

    (still wondering about that rebuttal, BTW)

    The body of open code is like a natural resource. If we are to have closed things, and I think we must for a lot of reasons, then those closed things need to be value adds, not rent seeking and the classic way we've resolved that is through basic competition.

    How much would Microsoft Office be if it were the only application to do basic things on computers? Truth is, that software did add value early on and it was worth the money people paid. Over time, our understanding of how to do things grows and now it's not so important that we buy software that enables basic computing tasks. OSS is always growing and it's always there to pressure those who claim to add value to actually do so.

    There are things that closed makes great sense for. Mechanical CAD is one of those close to my heart. There is still a lot of work to be done on mechanical CAD software and geometry kernels take insane huge man hours to develop, plus the audience for higher end capability is still small. OSS won't erode this for a very considerable length of time, though it can and currently is bringing some elements to ordinary people and that's good. When I started in CAD, there were a few open apps out there, very basic, difficult to use and limited in scope. There is a fair amount now, and the scope of stuff possible on open code is growing. Slowly, but it's growing. Hard niche. Not everything is that way though.

    I'm OK with closed software so long as they've actually added some value. Later on as things grow and change that may well end up open and I got the value when I needed it and that's all good. For those who make the investment, OSS can do a ton of stuff these days. The closed vendors hate it because they want to rent seek, thinking they are entitled to endless margins on things, always trying to stake out fiefdoms and to own parts of this and that when they should always be pushing the edge and earning their keep.

    Go OSS! It's good for everybody. Using it takes some work, but that work pays off over a very long period of time.

    There is a lot of confidence in the dynamics of it all. Anyone who experiences OSS has a good chance of getting how it all works. I share that confidence.

    However, the whole ecosystem is under threat due to law and the idea that laws can be purchased. It's not over yet. The state of things, particularly with software patents and the very aggressive US pressure for other parts of the world to adopt them (EU) is shaky at best. The amounts of money in play are huge. That's worth a sustained effort on their part, and that's exactly what we see out there today. I worry because an effort like that, well funded and compelling financially is a very strong motive. Way too many people don't get OSS and so there is work to be done. If enough people do that work, I think a longer term solid outlook is possible, but I really don't know whether or not enough people are doing that work.
  • evanhevanh Posts: 15,126
    edited 2013-02-07 22:53
    Rebuttal to what?
  • potatoheadpotatohead Posts: 10,253
    edited 2013-02-08 09:32
    I wrote "things are worth what people pay for them" and you wrote "that's how you want / think it, but that's not how it is" and then I asked you to expand on that. In my post above, I mentioned why, but the short story is I've challenged that assertion unsuccessfully myself over the years. If you were writing about something else, no worries. I've somebody I would love to bring a challenge to, that's all.
  • evanhevanh Posts: 15,126
    edited 2013-02-08 14:57
    Oh, that. I'd not wanted to start a second big topic. However, I've got very little to go on on that topic other than a recording of an Aussie economist being interviewed over a book he published in 2001 that describes the disconnect between typical economic thinking and the real world and how that causes much bigger bubbles than needed.


    EDIT: The phrase "It is what you want it to be" is closely related to another one - "We make our own beds". As in when a collective drives in a particular direction it can usually achieve it, but sometimes at great unintended cost that they become blind to or only belatedly shows up. And it can be painted as a success, even in hindsight, if enough of the mess is ignored. This is the "History is written by the victors" bit - most salesmen convince themselves they are total victors all the time.
  • potatoheadpotatohead Posts: 10,253
    edited 2013-02-08 15:03
    Well, I'm not sure that's the same thing. No worries then. That is just a topic of ongoing interest to me personally.
  • Heater.Heater. Posts: 21,230
    edited 2013-02-08 15:18
    potatohead,
    ...things are worth what people pay for them...
    Oh dear, room for a lot of confusion between the meanings of "cost", "price", "value", "worth", etc.

    Sure, when it comes to marketting most goods and products day to day what you say is true.

    If you can sell a paper clip and use the profits to aquire a house then that paper clip was worth a hundred or more thousand dollars to you. Sounds crazy but it happened, it's a famous internet story, sadly I can find the link now.

    Or there is the famous case of the "million dollar web page" where the guy sold each pixel of a 1024 by 1024 image on his web page for a dollar. Google it, it is still up.

    But, in order to take up your challenge:

    What is the Linux kernel and the GNU operating system worth? It has tremendos value and worth but it can be had for zero cost or price. Perhaps the former is because of the latter.

    What is your freedom worth, and how much do you pay for it?

    What if my government OK's smashing down Stone Henge to make way for a motorway, is that OK because nobody can afford enough money to stop them? Would such priceless human artifacts have no value because they are in the way of making money and no one can afford the cash to buy them out?

    Such ecconomic judgments without some degree of ethics are very dangerous.

    Have a listen to Richard Stallman http://www.youtube.com/watch?v=SNBMdDaYhZA for further views on this topic.
  • potatoheadpotatohead Posts: 10,253
    edited 2013-02-12 09:01
    Ok, that's a good confirmation. (On ordinary goods and services) Always on the lookout for a case where that's not true. So far, the only cases I've seen successfully argued are those involving artificial value due to lock in, legisation, monopoly position, collusion, etc... bad cases in other words.

    The kernel... You know, Eric Raymond actually fleshed out some great arguments aligned with this line of discussion.

    The worth of the kernel is directly related to the use value any one person sees. I like to compare Linux to natural resources. Oil is very valuable. In it's raw form where ordinary people can get to it, use value is intrinsic. Get the oil, burn it, done! Oil that is difficult to get at is still very valuable, but only in tandem with enabling technologies. Those same enabling technologies improve oil use value where that oil is transformed, refined, etc... into materials, fuels, etc...

    Notably, the easy access to oil early on fueled the enabling technology that brings us more use value from oil and the ability to get at more and more oil.

    Linux is like this in many ways. To somebody who has no skill with software, the worth of the kernel itself is dubious at best. This is use value. Stallman says it's not so important that people can use or modify or build with the software, only that it remains possible for them to do so. Stallman is right about it too. This is the basic environment that is necessary for use value to make sense and for the body of OSS to grow.

    Raymond expanded on this with the concept of use value. To somebody who has or wants to acquire the skill to use OSS, Linux, GNU and countless other OSS software components have a very high value! They can use the software as one would use any resource to save time and or enable things to happen. To somebody who does not have that desire, the worth is still there, but indirect. Others can employ the software, package it, combine it with hardware and deliver use value to somebody who would not otherwise see use value.

    I think the million dollar web page is brilliant. A buck a pixel! IMHO, that falls under goods and services. Clearly pixels are worth what people will pay for them. And what people will pay has as much to do with the seller showing value as it does the buyer seeing value for their own reasons. It may be obvious to both, and that's easy. There are few variances, the market in that case is stable and fairly predictable. Where it's not obvious to one or the other party, things are less stable and worth becomes far less clear. This has implications, some of which I've mentioned on this thread. One implication is new things are not always seen as worthy. Sometimes people need to be shown where the value is before they assign the worth necessary to form a market. Sales efforts add just as much value here as engineering efforts do, and the nexus of that is often industrial design, capable of adding very significant value to technology, despite the engineering being done and ready to go. Making technology accessible in that way is as necessary as innovating technology is in the first place. Of course, neither camp really wants to understand or cede ground to the other... Part of my professional work is bridging this gap and I deal with both sides of the matter regularly.

    I'll see a company or person develop some new innovation that's excellent, yet they see few returns! This is frustrating because the engineer in them sees the worth of it and they believe when the engineering is done right, said worth is intrinsic. Sadly, it's not. Mix in industrial design to make the innovation useful through packaging, user interaction design, human factors, etc... and more people see the worth because it's accessable to them. Add in the sales effort to educate people and position the innovation relative to others and worth improves due again to accessability. Costs of ownership, cost of purchase, etc... need to balance with returns or the innovation isn't seen as worth it. That circle goes on and on... The reverse happens too, and we all kind of hate that. Poor engineering or even no engineering packaged, designed, positioned, etc... ends up being "snake oil" and that circle goes on and on too.

    Coming back to Linux then. Stallman correctly asserts freedom as a core enabling principle. All of us benefit from it whether or not we choose to exercise said freedom. If it's there, it is enabling to those who would act. Raymond takes that and correctly asserts that use value is part of the process and that the cost of open is easily paid for with the use value one gets. Raymond also correctly asserts that software has a property that physical resources do not and that is the body is worth more than the sum of it's parts. People will contribute to Linux because it improves their use value, "scratch that itch" style. Everybody benefits from the contribution as well, growing the body of code that is basically a resource.

    "How much do you pay for it?" This isn't a dollars question, but a time one. The amounts of money in play surrounding open code are huge! I'm not talking shipped products, though there are a ton of those associated with a lot of dollars. I am talking about the competition. Without OSS, there is significantly less pressure for closed software to be honest about value added instead of rent seeking through lock in, patents, etc... The time argument? All of us need to invest the small amounts of time needed to preserve Stallman's assertions in the law and make sure that our society sees software in this way as normal, valuable, etc... or it may well end up litigated and legislated away.

    The threat of that legislation runs high! The US is dubious in this area because we are not secure. Software and process patents threaten OSS. It's not that OSS can't exist. It can, but the patents and litigation in general can sharply limit the use value OSS has to everyone and that inhibits growth and constrains freedom. In other words, this would result in a "spoiled" resource that requires dollars to properly utilize, an analogy would be dirty water that people cannot consume without first paying for a service or device to make it useful. Very non optimal case.

    Other parts of the world, the EU in particular, are more secure in that they don't have the same law as threat, but that's not a sure thing! Again, the amounts of money in play are huge! Closed vendors see a very large opportunity cost in Linux and they also see it diluting their own software value (as it should), and the best, optimal case for them is to render that threat moot so that revenue is high, value and labor low. And again, the analogy is a "pure" resource like clean water that people can use to their benefit directly. Optimal case.

    So that's what freedom is worth. It's worth a little of your time, every year, and when it counts. There is no central advocate for OSS. There isn't some core law or right that has been established yet. The deal isn't done and OSS exists right now largely because enough of us value it and the freedom associated with it to warrant said existence. The stroke of a pen could render it moot, and there are a lot of people who would use that pen.

    Just FYI, I don't talk about it much here, but I do invest a percentage of my free time on advocacy issues and one of those issues is OSS. I got started doing that in the 90's when I learned about all this stuff, reading the greats kind of as it happened and Stallman, Raymond and others frankly too numerous to mention here educated me. I get it. Many of my peers get it. But a lot of people just don't get it, and that's the threat and it's not their fault either. This just isn't the kind of thing that ordinary people can get motivated on because it's not tangible to them, nor accessible. The State of Oregon had potential legislation to increase the amount of OSS used by the State for the public business. I joined that effort and followed some legislation from it's genesis through committee, etc... traveling to the State capital several times to lobby, testify and educate legislators on use value and freedom and costs and cost savings.

    That effort failed, though the education didn't and some good was done, but the key thing was to note the failure mode. Proprietary software companies dropped a *lot* of money onto the State government, and there were not enough of us to check that. And after it was all said and done, I came to understand that time commit as very important because the number of us and whether or not we are active is the primary means by which OSS will continue to survive as the potent resource it is, not marginalized to non-competing uses and the much less significant use value case that many closed software entities would rather it be. In this specific case, closed vendors did not want to compete with the use value inherent in OSS, and the product of it was the workings of State government cost significantly more than they could otherwise. Lots more to say there, but not right now, save to say that closed software adding a lot of value is a fine State purchase, and the whole effort was about making sure value was actually added to avoid rent seeking. Sadly, that didn't go well. Good lesson though.

    Maybe not the answer you were looking for Heater, but time is really what the freedom is worth. Interestingly, our personal wealth is measured in time as well. If we have lots of our time available for us to purpose, we are wealthy people! We get to do a lot of stuff we want to do. If most of our time is purposed for us, we are poor people. This is true regardless of the money we may or may not have. Those of us who are wealthy in this way should be making those investments in time needed to check those who have lots of dollars so that we continue to see the benefit of OSS in our lives and marketplaces.

    In a very real sense then, Linux too is worth what people will pay for it, but it's often time not dollars that is the payment.. Investing the time needed to use it, or advocate for it, etc... helps with the use value problem, which improves the value perception out there which continues the viability of OSS overall. If we slack on that, too many of us that is, then freedom wasn't worth it, and it may be lost. Funny too, when it's lost, suddenly it's very worth it! Optimal case is to not experience that loss in the first place, but sometimes that is just how it has to go. Hope not.

    One other artifact of that is Linux and it's software properties and how it acts like a resource means it's enabling. Wealthy people, and I mean in the basic sense having time, can invest in Linux and get seriously good returns for very low capital costs in terms of real units of money. If I were king, every single kid everywhere would be exposed to that so all of them who have the inclination to advance their skills know they can do so, will do so and get all the use value they can out of it. Unlike natural resources, Linux doesn't "run out" and on that basis presents a stunningly huge value proposition to the world in general, given the enabling work is done. I can think of few basic investments that would pay so darn well.
Sign In or Register to comment.