• Announcements

    • Robin

      Welcome!   03/05/2016

      Welcome, everyone, to the new 910CMX Community Forums. I'm still working on getting them running, so things may change.  If you're a 910 Comic creator and need your forum recreated, let me know and I'll get on it right away.  I'll do my best to make this new place as fun as the last one!
Sign in to follow this  
Scotty

NP, Friday September 30, 2016

Recommended Posts

51 minutes ago, Vorlonagent said:

I didn't know that. 

I consider IBM's decision RE the PC to be a necessary stupidity, if such a thing can exist,

A 'necessary stupidity' is as good a definition I have ever heard of a category of precondition common in history. History is positively littered with bad decisions that nonetheless led to advantageous outcomes. For example, Columbus was not the only one to make the (correct) assumption that the Earth was round. However, the others considered his idea to sail around the Earth to reach India unworkable; they'd done the math and realised that the distance from Europe to Asia by sailing west actually exceeded the distance imposed by the trip around Cape Horn.

That Columbus just happened to bump into a new and entirely unknown land mass full of resources to exploit, loot and plunder was just one of these trivial facts that the simple math couldn't predict.

Share this post


Link to post
Share on other sites
8 minutes ago, The Old Hack said:

they'd done the math and realised that the distance from Europe to Asia by sailing west actually exceeded the distance imposed by the trip around Cape Horn.

I think we were talking about it .... yes, here (and around): The way around Cape Agulhas was not exactly short or safe ; but sailing west to reach China would mean traveling some 20 thousand kilometers across open ocean, which would mean as much as a year continuously out of sight of land while traveling in each direction ; and to quote the rest

On 05/11/2016 at 2:00 AM, ijuin said:

Carrying a year's worth of consumables was not an unmanageable burden in terms of bulk or weight if your ship was big enough to be sailing the high seas in the first place (i.e. hundreds of tons displacement--the Santa Maria was about 200 tons fully laden). The problem was that it would all go moldy/rancid (or get infested by bugs) after several months in the warm and humid air at sea. It wasn't until the invention of methods of sterilizing and airtight-sealing food (i.e. canning, whether in glass/clay jars or in metal tins) that food could be preserved for years on end in such a climate.

Share this post


Link to post
Share on other sites
6 minutes ago, hkmaly said:

True, but IBM was not only candidate. Some other, similar company could've done it. Like .... DEC. Ok, I must admit DEC is only other possible company I found ... Sun or SGI, for example, started only afterwards, although they would likely start even without the IBM's "mistake" ... and Cray Research seem to be in TOO big computers ...

DEC didn't seem interested in minicomputers. Dominance would have shifted to Apple, Atari, Commodore or possibly Radio Shack. 

What am I saying?  Not Apple.  Apple wins when it innovates and the premium it charges for its stuff is justified, loses when "good enough" overtakes Apple's focus on excellence.  The premium creates a natural niche for "good enough".  PCs filled that niche in the case of the desktop/laptop market, where Apple's share as been 5%.  Apple's failed lawsuit against Microsoft over the "look and feel" of Windows vs Mac OS was an attempt to protect Apple's market share.  When it failed, Apple was consigned to some niche applications which Macs were head and shoulders above PCs. 

I believe Steve Jobs could see the same writing on the wall when Google came out with Android.  I think that's why he is reported to want to destroy Google and willing to crash Apple in the process if he had to.

Share this post


Link to post
Share on other sites
8 minutes ago, Vorlonagent said:

DEC didn't seem interested in minicomputers.

Who knows what would they do in say 1983 if IBM PC wouldn't appear or would appear as yet another closed computer?

10 minutes ago, Vorlonagent said:

Apple wins when it innovates and the premium it charges for its stuff is justified, loses when "good enough" overtakes Apple's focus on excellence. 

"Excellence"? Apple is "one-size-fits-all". They are rarely ready to admit that there can be two ways to do something and neither is bad. They wouldn't be able to get major market share even if people would somehow be willing to pay the premium price.

Share this post


Link to post
Share on other sites
1 minute ago, hkmaly said:

"Excellence"? Apple is "one-size-fits-all". They are rarely ready to admit that there can be two ways to do something and neither is bad. They wouldn't be able to get major market share even if people would somehow be willing to pay the premium price.

They generally do deliver a better product that is more accessible.  But Apple is also very controlling.  You have to do it their way whether it's how the OS works or how they source components.

The premium Apple puts on their products means they would always eventually be undercut.  The American consumer is more price conscious than anything else.  "Good enough" which is significantly cheaper wins over even "demonstrably better" if it's also "demonstrably more expensive".  With Steve Jobs gone, I consider Apple to be circling the drain. 

I personally think Apple's nifty new HQ to be something of a white elephant.  Too big a drain on resources when Apple really needs to focus on not just leading the market but blazing a trail at a speed where others can't catch up.  The alternative is death.  Either the death of Apple's premiums and the way Jobs did business or the death of the company.  Their business model isn't and was never sustainable.  Maybe it'll be a shopping mall in 20 years....

Share this post


Link to post
Share on other sites
1 hour ago, Vorlonagent said:
1 hour ago, hkmaly said:

"Excellence"? Apple is "one-size-fits-all". They are rarely ready to admit that there can be two ways to do something and neither is bad. They wouldn't be able to get major market share even if people would somehow be willing to pay the premium price.

They generally do deliver a better product that is more accessible

Accessibility? Yes. Better quality of components? Yes. But that's not enough for better product.

1 hour ago, Vorlonagent said:

But Apple is also very controlling.  You have to do it their way whether it's how the OS works or how they source components.

People who use tools in way they were designed to be used rarely make any sort of progress.

Also, it makes more sense to adapt tools for people than to adapt people to tools.

(Although, maybe Apple doesn't want their computers to be seen as tools. More like something to worship ... or to show to others.)

1 hour ago, Vorlonagent said:

The American consumer is more price conscious than anything else. 

Meanwhile, the non-american consumer can't AFFORD the premium at all (ok, not EVERY non-american consumer, but it was funny you were talking about american consumers as if there would be some category of consumers who wouldn't be concerned about getting twice as good product for ten times the price ; like, I guess people like Saudi family, but whole nation?).

1 hour ago, Vorlonagent said:

Too big a drain on resources when Apple really needs to focus on not just leading the market but blazing a trail at a speed where others can't catch up.

Apple doesn't need money to lead the market. It needs IDEAS.

1 hour ago, Vorlonagent said:

The alternative is death.  Either the death of Apple's premiums and the way Jobs did business or the death of the company. 

Agree.

1 hour ago, Vorlonagent said:

Their business model isn't and was never sustainable.

... to be fair, business models USUALLY don't value sustainability much, it's not specific to Apple.

Share this post


Link to post
Share on other sites
2 minutes ago, hkmaly said:

Meanwhile, the non-american consumer can't AFFORD the premium at all (ok, not EVERY non-american consumer, but it was funny you were talking about american consumers as if there would be some category of consumers who wouldn't be concerned about getting twice as good product for ten times the price ; like, I guess people like Saudi family, but whole nation?).

There are European consumers who behave differently, also Japanese and South Korean consumers and lately, Chinese consumers.  I can't speak to which if any would be more open to Apple's product line but Americans are distinct in their/our degree of price consciousness and Apple is located in the US, making the US its prime market.

5 minutes ago, hkmaly said:

Apple doesn't need money to lead the market. It needs IDEAS.

Apple needs ideas, yes.  Desperately.  That's what I meant by leading at a pace others can't catch up to. 

Apple also needs the money to develop and market those ideas.  Innovation is not done on the cheap.  Especially since some ideas are going to be failures.

They need money a lot more than they need a fancy new HQ.  I imagine it's considered something of a monument to Jobs since he put its construction in motion.  Certainly it's too late to back out now.  It's done or almost so.  It's just Jobs should have realized Apple needed the money more.

Share this post


Link to post
Share on other sites
45 minutes ago, Vorlonagent said:
57 minutes ago, hkmaly said:

Meanwhile, the non-american consumer can't AFFORD the premium at all (ok, not EVERY non-american consumer, but it was funny you were talking about american consumers as if there would be some category of consumers who wouldn't be concerned about getting twice as good product for ten times the price ; like, I guess people like Saudi family, but whole nation?).

There are European consumers who behave differently, also Japanese and South Korean consumers and lately, Chinese consumers.  I can't speak to which if any would be more open to Apple's product line but Americans are distinct in their/our degree of price consciousness and Apple is located in the US, making the US its prime market.

I think my "can't afford" was pretty good characterization of big part of Europe. It definitely is for the part I'm in. True, it may work differently in Switzerland or Norway. Or, yes, not sure how it works in Japan or South Korea.

45 minutes ago, Vorlonagent said:
57 minutes ago, hkmaly said:

Apple doesn't need money to lead the market. It needs IDEAS.

Apple needs ideas, yes.  Desperately.  That's what I meant by leading at a pace others can't catch up to. 

Apple also needs the money to develop and market those ideas.  Innovation is not done on the cheap.  Especially since some ideas are going to be failures.

Well, if they don't get any ideas, they don't need the money.

45 minutes ago, Vorlonagent said:

They need money a lot more than they need a fancy new HQ.  I imagine it's considered something of a monument to Jobs since he put its construction in motion.  Certainly it's too late to back out now.  It's done or almost so.  It's just Jobs should have realized Apple needed the money more.

I imagine they might've build little less fancy monument.

But ... the HQ was planed somewhen in 2011. Jobs knew he has tumor since 2003 and died later in 2011. I guess he didn't cared about longterm future of company so much :)

Our resident Pharaoh might relate.

Share this post


Link to post
Share on other sites
1 hour ago, hkmaly said:

the HQ was planed somewhen in 2011. Jobs knew he has tumor since 2003 and died later in 2011. I guess he didn't cared about longterm future of company so much :)

Our resident Pharaoh might relate.

A "permanent" location gives your organization (business, family, government, religion) an appearance of strength and stability.  A palace should remain stable and appear impressive throughout the lifetime of the ruler.  A Temple should be stable and impressive through the duration of a dynasty.  And a tomb needs to be stable and impressive throughout all time.

If Jobs knew he was dying, was the AppleHQ built as the iTemple or the MacMausoleum?

Share this post


Link to post
Share on other sites
6 hours ago, Vorlonagent said:

I consider IBM's decision RE the PC to be a necessary stupidity, if such a thing can exist, for the PC to become what we now know today.  The crucial thing was them publishing the schematics.  Not only could someone make devices to plug into the PC, they could make their own PC without paying a dime to IBM.  I consider that opening up of the market to be crucial since all other microcomputers were company-centric ecosystems the way Apple still is today.  IBM still defined the PC specification.

It took a second stupidity to completely pull IBM's fingers off the PC.  When Intel developed with the 80386 chip, IBM's execs looked at it and turned a bit pale.  They declined to make a computer around it.  They were concerned a 386-based PC would cannibalize their minicomputer sales.  A third-party PC-maker named Compaq picked up the challenge.

 

I wasn't aware of these stories either.  I knew DOS was a CP/M clone but I thought Microsoft developed it in-house. 

I'm sure the 150+ companies selling mutually-compatible computers and hardware based on the S-100 (aka IEEE 696) bus, most of said computers running some version of CP/M, would have been amazed to learn that they were company-centric ecosystems. The S-100 bus began with 8-bit Intel 8080 processors, but did okay with 16-bit processors such as the 8086 and 68000, and there was even a 32-bit processor board using a 68020. (Note: internally the 68000 is a 32-bit processor and the 68020 is a 64-bit processor. They had half-width paths to RAM. Just like the 8088 that IBM put in the first PC is a 16-bit processor internally with an 8-bit path to RAM.)

The S-100 bus, by the way, had 24 address lines and 16 data lines. For the IBM PC it was 20 and 8.

At the time IBM came out with the PC and "legitimized" the microcomputer industry - yes some clown said that - said industry was already as big (total revenues) as the pantyhose industry and coming up on rivalling the dog-food industry. Some big companies, including the US Air Force, had server farms clustering racks of S-100-bus computers.

Quote

I knew DOS was a CP/M clone but I thought Microsoft developed it in-house.

MS-DOS version 1 (aka PC-DOS, originally QDOS) was not nearly good enough to qualify as a CP/M clone. CP/M had years of service on a variety of hardware platforms behind it, and was much more robust and stable. CP/M-86 was further along in development than QDOS, until the guy who wrote QDOS decided to throw intended functionality out while the people doing CP/M-86 insisted on getting it to actually work. There were also multi-user and single-user multi-tasking versions of CP/M, and later of CP/M-86.

MS-DOS version 2 was almost entirely developed in-house by Microsoft. Version 1 was that bad, they basically threw it out - and replaced it with something that users would swear at for the next 17+ years.

Share this post


Link to post
Share on other sites
8 hours ago, Don Edwards said:

MS-DOS version 2 was almost entirely developed in-house by Microsoft. Version 1 was that bad, they basically threw it out - and replaced it with something that users would swear at for the next 17+ years.

At least it had staying swearing power.

Share this post


Link to post
Share on other sites
14 hours ago, Pharaoh RutinTutin said:

If Jobs knew he was dying, was the AppleHQ built as the iTemple or the MacMausoleum?

I would say iTemple.  Primarily because the iPhone and iPad were Jobs' crowning achievements.  But there are less-flattering reasons.

16 hours ago, hkmaly said:

I think my "can't afford" was pretty good characterization of big part of Europe. It definitely is for the part I'm in. True, it may work differently in Switzerland or Norway. Or, yes, not sure how it works in Japan or South Korea.

Really?  I'm surprised.  Heck I didn't know you lived in Europe....

16 hours ago, hkmaly said:

But ... the HQ was planed somewhen in 2011. Jobs knew he has tumor since 2003 and died later in 2011. I guess he didn't cared about longterm future of company so much :)

Jobs treated his tumor the way he ran his company...on his own terms with him thoroughly in control.  IIRC he ran down every blind ally and tried every quack-remedy looking for a cure.  Everything except chemotherapy.   And he died as he lived.  I don't know if he just didn't care near the end, was consumed by hatred of Google and wouldn't budge on anything, just wanted this one last thing for his ego, or honestly thought it would help his company survive. 

One of the guiding principles that went into Apple's HQ was ideas sparked by chance meetings in the hallway or in the breakroom.  The idea was to get everybody together and contour the place to passively pool creativity.  Like most such fad-ideas there's both something to it and confirmation bias that has over-sold it.  The new Nvidia HQ, a much less ambitious affair that's going up about 250 meters away from my desk, is being built with this same principle in mind.  My supervisor's supervisor is skeptical.

12 hours ago, Don Edwards said:

I'm sure the 150+ companies selling mutually-compatible computers and hardware based on the S-100 (aka IEEE 696) bus, most of said computers running some version of CP/M, would have been amazed to learn that they were company-centric ecosystems. The S-100 bus began with 8-bit Intel 8080 processors, but did okay with 16-bit processors such as the 8086 and 68000, and there was even a 32-bit processor board using a 68020. (Note: internally the 68000 is a 32-bit processor and the 68020 is a 64-bit processor. They had half-width paths to RAM. Just like the 8088 that IBM put in the first PC is a 16-bit processor internally with an 8-bit path to RAM.)

The S-100 bus, by the way, had 24 address lines and 16 data lines. For the IBM PC it was 20 and 8.

At the time IBM came out with the PC and "legitimized" the microcomputer industry - yes some clown said that - said industry was already as big (total revenues) as the pantyhose industry and coming up on rivalling the dog-food industry. Some big companies, including the US Air Force, had server farms clustering racks of S-100-bus computers

S100 was not a consumer industry and wasn't going to be.  It isn't relevant to what I was saying.  All the 6502-based consumer microcomputers were company-centric ecosystems as I stated.

Technically the 8088 had a sort-of 12-bit memory addressing scheme.  It had two registers associated with each memory access register. Each segment register accessed an 8-bit wide data path (64K),  but there was a second register that changed what 8 bit segment was being accessed.  The second register was overlaid on the first such that it added 4 bits to the effective range of accessible memory, though moving the segment around was a pretty big deal codewise...

The 8088 was a cut down version of the 8086, which IIRC was a true 16-bit processor, including memory access.  Intel has a history of making the "good" version you want to buy and the brain-damaged version you can afford (*cough*Celeron*cough*)

Share this post


Link to post
Share on other sites

For what it is worth, I live in Europe and I have an iPhone. And an iPad. I love the silly things. Admittedly I saved up a while to have the money to throw at them.

3 hours ago, Vorlonagent said:

My supervisor's supervisor is skeptical.

Good job. You have managed to send my mind into that eerie place where I am now wondering about supervisors whose job is to supervise other supervisors as they supervise lower ranking supervisors on duty observing still lower ranking supervisors, and whether those first supervisors then go report to their supervisors who then in turn have supervisors they have to keep satisfied, on and on, ad infinitum ad nauseam, worlds without end...

Share this post


Link to post
Share on other sites
Just now, The Old Hack said:

For what it is worth, I live in Europe and I have an iPhone. And an iPad. I love the silly things. Admittedly I saved up a while to have the money to throw at them.

Good job. You have managed to send my mind into that eerie place where I am now wondering about supervisors whose job is to supervise other supervisors as they supervise lower ranking supervisors on duty observing still lower ranking supervisors, and whether those first supervisors then go report to their supervisors who then in turn have supervisors they have to keep satisfied, on and on, ad infinitum ad nauseam, worlds without end...

Yup.  Two grades of "supervisor" then two grades of "Vice President" and then the CEO. 

It makes the most sense when you talk about concentric circles of responsibility.  My supervisor = responsible for my unit.  Her supervisor = responsible for all related units. The grades of VP work the same way.

Share this post


Link to post
Share on other sites
7 minutes ago, Vorlonagent said:

Yup.  Two grades of "supervisor" then two grades of "Vice President" and then the CEO. 

It makes the most sense when you talk about concentric circles of responsibility.  My supervisor = responsible for my unit.  Her supervisor = responsible for all related units. The grades of VP work the same way.

Ah. That does help. You could call this a simplified chain of command, really.

Share this post


Link to post
Share on other sites
3 minutes ago, The Old Hack said:

Ah. That does help. You could call this a simplified chain of command, really.

Yes.  My company is medium sized with maybe 10,000 people total, mostly in the US (Silicon Valley) and India, but with small offices everywhere it seems.   There's only so many links in the chain needed for that.

Share this post


Link to post
Share on other sites
4 hours ago, Vorlonagent said:

Really?  I'm surprised.  Heck I didn't know you lived in Europe....

You can't guess based on time of my posts. :) Sometimes I'm going to sleep when it's ALREADY light outside.

4 hours ago, Vorlonagent said:

Technically the 8088 had a sort-of 12-bit memory addressing scheme.  It had two registers associated with each memory access register. Each segment register accessed an 8-bit wide data path (64K),  but there was a second register that changed what 8 bit segment was being accessed.  The second register was overlaid on the first such that it added 4 bits to the effective range of accessible memory, though moving the segment around was a pretty big deal codewise...

The 8088 was a cut down version of the 8086, which IIRC was a true 16-bit processor, including memory access.  Intel has a history of making the "good" version you want to buy and the brain-damaged version you can afford (*cough*Celeron*cough*)

Ehmm ... the numbers don't add up. You need 16 bits to address 64KB. 8086 used segment register to address 1MB of RAM in something similar to trick you described. The 8088 looked the same internally, only difference was that the EXTERNAL bus was 8bit. Also I'm pretty sure it is 8 bit as in data width, not address.

20 hours ago, hkmaly said:

big part of Europe. It definitely is for the part I'm in. True, it may work differently in Switzerland or Norway

25 minutes ago, The Old Hack said:

For what it is worth, I live in Europe and I have an iPhone. And an iPad. I love the silly things. Admittedly I saved up a while to have the money to throw at them.

I already knew Denmark is close to Norway.

4 hours ago, Vorlonagent said:
18 hours ago, Pharaoh RutinTutin said:

If Jobs knew he was dying, was the AppleHQ built as the iTemple or the MacMausoleum?

I would say iTemple.  Primarily because the iPhone and iPad were Jobs' crowning achievements.  But there are less-flattering reasons.

Or iCathedral. Temples are supposed to be smaller than cathedrals, aren't they?

Share this post


Link to post
Share on other sites
20 minutes ago, hkmaly said:

You can't guess based on time of my posts. :) Sometimes I'm going to sleep when it's ALREADY light outside.

Nope.  My screen says you posted your reply "20 minutes ago". 

Your english usage was about my only possible cue.  You're overall very good but you don't always do plurals right for an American, but then, neither do the British.  :) 

21 minutes ago, hkmaly said:

Ehmm ... the numbers don't add up. You need 16 bits to address 64KB. 8086 used segment register to address 1MB of RAM in something similar to trick you described. The 8088 looked the same internally, only difference was that the EXTERNAL bus was 8bit. Also I'm pretty sure it is 8 bit as in data width, not address.

You're right about data width.  I misremembered and got my powers of 2 messed up.

Both 8086 and 8088 used the overlay scheme I mentioned to achieve 20-bit addressing, so both processors could address a full MB of memory if the programmers were willing to jump through the necessary code hoops.

Share this post


Link to post
Share on other sites
2 hours ago, Vorlonagent said:
2 hours ago, hkmaly said:

You can't guess based on time of my posts. :) Sometimes I'm going to sleep when it's ALREADY light outside.

Nope.  My screen says you posted your reply "20 minutes ago". 

I wasn't talking about NOW. Now it's still dark outside. Still, most people in my time zone are sleeping.

2 hours ago, Vorlonagent said:

Your english usage was about my only possible cue.  You're overall very good but you don't always do plurals right for an American, but then, neither do the British.

Whoa thanks. Although you would speak differently if you hear me speaking.

2 hours ago, Vorlonagent said:

Both 8086 and 8088 used the overlay scheme I mentioned to achieve 20-bit addressing, so both processors could address a full MB of memory if the programmers were willing to jump through the necessary code hoops.

Sure they were willing. They didn't liked it but it wasn't the worst PC (and/or DOS) feature they needed to adapt to. Compilers were giving you the choice of "tiny", "small" and "large" memory model, and in large pointers were 32bit. Obviously, you still could only use 640kiB, as the area above that was reserved for video RAM and BIOS.

With 80286, the processor was able to use 16MiB ... in protected mode, which was not compatible with DOS, so most people continued using 1MiB, with additional 64kiB-16B if you turned on A20 gate.

With 80386, the V86 mode came, which allowed to run DOS in virtualized environment under protected mode. You might remember device driver EMM386 .... well, THAT was technically the core of OS, running in protected mode with single V86 mode application, which was DOS. In that time, it was possible to make programs in protected mode and flat addressing, but only if you were willing to basically write your own operation system with drivers for all necessary devices (because the devices DOS had were not working in protected mode). Lot of games did. Some programs instead made use of XMS and/or EMS memory, and THAT was much more code hoops.

(Technically, EMS standard was originally developed for physical expansion cards, but most people only ever got it emulated by the EMM386 driver.)

 

Share this post


Link to post
Share on other sites

With 80286, the processor was able to use 16MiB ... in protected mode, which was not compatible with DOS, so most people continued using 1MiB, with additional 64kiB-16B if you turned on A20 gate.

As I recall, when the 80286 first came and and we learned about a number of people observed that the second most salient feature of protected mode (behind the fact that DOS wasn't compatible with it) was that nothing was protected.

Share this post


Link to post
Share on other sites
10 hours ago, Vorlonagent said:

Yup.  Two grades of "supervisor" then two grades of "Vice President" and then the CEO. 

It makes the most sense when you talk about concentric circles of responsibility.  My supervisor = responsible for my unit.  Her supervisor = responsible for all related units. The grades of VP work the same way.

I have worked, (other than the army, where it was 8 or 9, assuming you count the Secretary of Defense as CEO.  Add on if you count the President as the CEO) in 3 places where the gap between me and CEO was more than 2 people.

  1. Radio shack
  2. Wendy's
  3. EDS

Rat shack was a high school job, Wendy's was my college job, and EDS was my time of horror programming job, where I learned that Dilbert is in fact a documentary, and why I don't like the movie "Office Space".  Brings back too many bad memories.

Share this post


Link to post
Share on other sites
16 hours ago, hkmaly said:

I wasn't talking about NOW. Now it's still dark outside. Still, most people in my time zone are sleeping.

My forum tells me this post was made 16 hours ago.  Again, no local date or time.

16 hours ago, hkmaly said:

Whoa thanks. Although you would speak differently if you hear me speaking.

I don't talk as well as I write either.  In my case English is my first language, but I also deal with social anxiety so I have a harder time both thinking of things to say and developing my ideas as completely.

16 hours ago, hkmaly said:

Sure they were willing. They didn't liked it but it wasn't the worst PC (and/or DOS) feature they needed to adapt to. Compilers were giving you the choice of "tiny", "small" and "large" memory model, and in large pointers were 32bit. Obviously, you still could only use 640kiB, as the area above that was reserved for video RAM and BIOS.

With 80286, the processor was able to use 16MiB ... in protected mode, which was not compatible with DOS, so most people continued using 1MiB, with additional 64kiB-16B if you turned on A20 gate.

With 80386, the V86 mode came, which allowed to run DOS in virtualized environment under protected mode. You might remember device driver EMM386 .... well, THAT was technically the core of OS, running in protected mode with single V86 mode application, which was DOS. In that time, it was possible to make programs in protected mode and flat addressing, but only if you were willing to basically write your own operation system with drivers for all necessary devices (because the devices DOS had were not working in protected mode). Lot of games did. Some programs instead made use of XMS and/or EMS memory, and THAT was much more code hoops.

(Technically, EMS standard was originally developed for physical expansion cards, but most people only ever got it emulated by the EMM386 driver.)

I remember.  I was reading PC magazine at the time and writing assembly code on the forgotten x86: the 80186.

As a DOS user I've seen many EMM386 errors...

Share this post


Link to post
Share on other sites
19 hours ago, Don Edwards said:
Quote

With 80286, the processor was able to use 16MiB ... in protected mode, which was not compatible with DOS, so most people continued using 1MiB, with additional 64kiB-16B if you turned on A20 gate.

As I recall, when the 80286 first came and and we learned about a number of people observed that the second most salient feature of protected mode (behind the fact that DOS wasn't compatible with it) was that nothing was protected.

I think it could be done ... although it was hard. The page-based protection which came with 80386 was much easier to use than segmentation-based one.

4 hours ago, Vorlonagent said:
20 hours ago, hkmaly said:

I wasn't talking about NOW. Now it's still dark outside. Still, most people in my time zone are sleeping.

My forum tells me this post was made 16 hours ago.  Again, no local date or time.

Sigh. I was talking about the fact that if you look at when I'm posting, you could assume it's day where I am and guess where it is based on that. And that assumption would be incorrect.

4 hours ago, Vorlonagent said:

I remember.  I was reading PC magazine at the time and writing assembly code on the forgotten x86: the 80186.

Ok, that's experience I don't have. I started to care about assembly code and stuff like this only with 80386. So, if they would upgrade DOS properly, I wouldn't be affected ... but they didn't and I needed to deal with all the stuff "inherited" from 8086.

(On the other hand, I never made program in assembly which would actually require more than 64KB of code anyway. Moved to C. Yet I still miss the days where I was writing programs which didn't actually done much else than writing something on screen, but utilized 80% of instruction set and all modes the CPU had for it ...)

 

Share this post


Link to post
Share on other sites
13 minutes ago, hkmaly said:

Sigh. I was talking about the fact that if you look at when I'm posting, you could assume it's day where I am and guess where it is based on that. And that assumption would be incorrect.

I have no idea how I could tell what time of day it is where you are.  Maybe I could go to your account, poke around and try to find out how many hours off GMT you are compared to me, but I don't see anything like an absolute timestamp on anybody's posts, yours included. 

Maybe you have a forum setting enabled that I don't?  Otherwise where am I expected to go for this information?

17 minutes ago, hkmaly said:

Ok, that's experience I don't have. I started to care about assembly code and stuff like this only with 80386. So, if they would upgrade DOS properly, I wouldn't be affected ... but they didn't and I needed to deal with all the stuff "inherited" from 8086.

I was one of two programmers attempting to automate the compressor staging for an agricultural warehouse.  I wrote my code on a PC in 8086/8088 assembly.  It was then burned onto an EEPROM and then tested on an industrial computer which had next to no inputs or outputs.  Needless to say the process was slow and difficult to debug.

Share this post


Link to post
Share on other sites
2 hours ago, Vorlonagent said:
2 hours ago, hkmaly said:

Sigh. I was talking about the fact that if you look at when I'm posting, you could assume it's day where I am and guess where it is based on that. And that assumption would be incorrect.

I have no idea how I could tell what time of day it is where you are.  Maybe I could go to your account, poke around and try to find out how many hours off GMT you are compared to me, but I don't see anything like an absolute timestamp on anybody's posts, yours included. 

Maybe you have a forum setting enabled that I don't?  Otherwise where am I expected to go for this information?

No I don't. I also see everything in my timezone. Nevermind. Considering I was explaining how it would be useless ...

2 hours ago, Vorlonagent said:

I was one of two programmers attempting to automate the compressor staging for an agricultural warehouse.  I wrote my code on a PC in 8086/8088 assembly.  It was then burned onto an EEPROM and then tested on an industrial computer which had next to no inputs or outputs.  Needless to say the process was slow and difficult to debug.

Oooooh ... imbedded programing. Never done any. Must be fun.

Quote

The salient characteristic of an imbedded system is that it cannot be allowed to get into a state from which only direct intervention will suffice to remove it.  An imbedded system can't permanently trust anything it hears from the outside world.  It must sniff around, adapt, consider, sniff around, and adapt again.  I'm not talking about ordinary modular programming carefulness here.  No.  Programming an imbedded system calls for undiluted raging maniacal paranoia.  For example, our ethernet front ends need to know what network number they are on so that they can address and route PUPs properly.  How do you find out what your network number is?  Easy, you ask a gateway.  Gateways are required by definition to know their correct network numbers.  Once you've got your network number, you start using it and before you can blink you've got it wired into fifteen different sockets spread all over creation.  Now what happens when the panic-stricken operator realizes he was running the wrong version of the gateway which was giving out the wrong network number?  Never supposed to happen.  Tough.  Supposing that your software discovers that the gateway is now giving out a different network number than before, what's it supposed to do about it?  This is not discussed in the protocol document.  Never supposed to happen.  Tough.  I think you get my drift.

System programming is also fun, but on normal PC you at least have reset button, so if you make a bug and the computer is completely stuck because half of code which under normal OS is supposed to do error recovery was not yet installed and the second half, well, that's where the bug is in, you can simply press reset and put more debugging print inside. You can even usually rely on the debug print getting on screen, unless you are just in the process of switching videomode or you mapped video RAM incorrectly ...

(By "fun", I obviously mean "fun to talk about years later, from safety of position which no longer requires fixing any problems possibly caused by the code you wrote back then")

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this