EE Times had an article several years ago (40th anniversary maybe?) detailing the history of hard drives with a nice description of a lab with Dixie cups containing iron oxide paint for spin coating the platters by hand.
Seagate also made a couple of drives with 2 heads per platter surface and only 1 actuater. The head assembly was known as a C-Blk instead of an E-blk. The model number was ST-425(20 MB). Seagate also produced a half height version called the ST-212(10 MB) with 1 platter and 4 heads and the ST-206(5 MB)which only used 1 surface and 2 heads.
I was not able to find anything on the net about those but given the model numbers they would have been using stepper motors instead of voice coils so the track density would have been low enough to allow multiple heads to stay in alignment. This is an obvious way to improve transfer rates and seek times but it limits the track density if you want to use the heads in parallel. The dual head assembly Seagate drives I mentioned used voice coils and embedded servo tracking although the only reference I found to this was the Connor "Chinook" series so maybe I got the two manufacturers confused. Apparently the Seagate Barracuda ST12450W using one head assembly could read simultaneously from both heads on each platter but I do not remember it.
All of these techniques have been left behind because of track density and RAID which makes economic sense.
Personally, I'm waiting for them to cram 2 opposing sets of read/write arms (or even just a second set for reading) so that they can effectively halve the latency and seek times without having to go faster than the existing 15k screamers.
For a short time Seagate made a series of drives with dual head assemblies for transactional processing but they were not cost effective. I do not remember how the interfacing worked.
The fact of the matter is, a gun is far more likely to kill the owner of the gun, a casual bystander, or someone totally different (when the gun is stolen), than it is a criminal while protecting life or property.
Comparing the number of criminals killed with firearms to other firearm deaths leaves out the vast majority of cases where the presence of a firearm prevents a crime when not discharged or even brandished.
Lucky for us then this would be a civil war instead of a real war and unlike our earlier one which was more of a war between the States with defined boundaries.
As I am sure others will point out, ownership of everything up to machine guns is legal in most of the States although for the later considerable civilian licensing is involved. Given the targets, hunting weapons are more then adequate and probably preferred.
I would still prefer that U.S. versus Miller was defendable one way or the other since the lack of any defense opposition made the govenment's weak case possible and had the result of producing bad law which we continue to pay for.
In the movie, Aragorn randomly hands the four hobbits four short swords right before the Nazgul attack at Amon Sul. He doesn't explain where they come from nor how he came to have them. Later, Merry uses his to stab the Witch King in the back of the knee, which despite the admonition "no man can slay me," seems to be pretty effective at hurting him and rendering him vulnerable to Eowyn's coup de grace. But nobody knows why.
I always enjoyed this type of thing in books. Not only does the Witch King unknowingly face a woman rendering the prophecy in doubt but Merry happens to have a weapon specifically spelled to hurt him that was only gained through chance. Mysterious as he is, perhaps that was Tom's way of indirectly aiding the war.
What really bothered me about this in the movie was that they could very well have had Aragorn give the swords to the hobbits and in passing give mention to their origin (or someone else could have noticed the odd weapons and told them later if time was an issue) replacing the similar book scene with Tom.
Of course, I think Frodo's movie addition in Osgiliath among other unneeded additions should have been left out in favor of the Scouring of the Shire. I have never found a satisfactory explanation for that. I was rather looking forward to fierce hobbits and in the case of Merry and Pippin contrastingly tall ones. As it was, the only lasting impressions were on Frodo and Sam.
Once outed, what would they have left to loose through association. Conversely, outing them would make their association stronger and their voices louder. This would be the last thing the social conservatives would want. They're bullies. They don't want to actually fight them, they only want them to cower and submit to their will out of fear of alienation.
I do not particularly disagree with this however it requires only a minority of social conservatives to implement this which will have the effect of essentially committing the rest least they denounce their own which I find even less likely.
Since there is no upper bound on the complexity of existing law, create one by adding an ammendment specifying that all federal laws expire after a certain amount of time. Nothing would preclude congress from passing them again but this would create pressure to work on what they believe is truely important.
Why would the libertarians and anarchists object? It sounds like one of the things they would try although the libertarians consider contract enforcement one of the proper functions of government.
Is it out of the question that after some future date it may not be feasable to travel while carrying portable storage do to increasingly strict national reactions to terrorism? This could not only be because of the percieved threat of smuggling a weapon but also because law enforcement could take the opportunity to duplicate and inspect user data as happened recently in a US border incident.
If our sole purpose was to reproduce a few times and die "young" (before 35), then why do our cells have so many proteins dedicated to detecting and repairing chromosome damage? Shouldn't they deactivate after 35 years?
It is not so much that they should deactivate after 35 years but more that they are left running.
Why would nature keep old people around? How does nature select for old age genes if you reproduce when young?
Nature would only select for long life genes if they gave preferential advantage to the caregiver's children who also have a significant chance of having the same genes.
Some theories are that older people pass their life's knowledge to the next generation, without the next generation having to experience it themselves. Older people act as secondary caregivers, freeing the younger generation to do "useful stuff".
Notice here that if the benefit of the older caregiver is shared by those who are not related to them, the genes involved will not be selected for as strongly unless group selection is significant.
There's no reason to believe our bodies were made to wear out at 60 or 70. Eat less calories, more fruit and veggies high in anti-oxidant compounds, exercise (physical labor and mental), and there's no reason that our bodies couldn't last... longer. How much longer? One study says maybe 120 years.
It is not so much that we were designed to wear out at a certain point but instead that we are running on inertia so to speak. Preventative maintenance certainly has an effect but it is not the whole story. An even better question is where are the genes that affect behavior such as to lead a healthy life style? They are in the same position as our other genes that affect lifetime.
Experiments have been done on short lived species like possums that live on the east coast. Their normal life span is about 2 years at which point if they have not been run over or eaten they start dying of cancer and other things that we would consider old age. It is not a change in their environment that has caused this but an accumulation of genes that cause or allow cancer and other problems because their life is so short anyway do to environmental hazards that they have to go all out just to reproduce before some calamity takes them. In the past, any gene that gave them a reproductive advantage was selected for despite any problems it caused a short time later. Even when kept in a benign environment, they only live to maybe 4 years and then succumb to old age. The situation is reminiscent of how salmon spawn though not as extreme.
When these possums were bred for extended life by delaying when they have children, the genes causing and allowing old age were quickly removed from the pool. Onset of cancer happened later and their lifetime was extended several times over. I am reminded of Heinlein's Lazarus Long and the Ira Howard Foundation.
Atmospheres are part of a planet, aren't they? If their atmospheres (or oceans) are covering the center of mass, then it's within the planet/planets.
That had not occurred to me. Except in the case of a gas giant where the outer fluid is a significant fraction of the planet's mass, I would not have considered the atmosphere when determining where the barycenter is because you could end up with a situation where with two otherwise identical double planets one meets the definition and another does not just because of the difference in having an atmosphere which is contributing almost nothing. At what depth does Jupiter's atmospere become liquid? That might be a more appropriate surface to compare the barycenter to.
The more I think about it, the more I think they might well be considered one planet, anyway. Given some time, they certainly will be!
I am not so sure about that. Since the planets are already tidally locked to each other they only have to deal with tides caused by their primary which are much smaller. They will radiate gravitational energy but because of their relatively low mass, I doubt it will be significant. If they were not tidally locked, then they would move apart until they were or they became seperated like the Earth's moon currently is doing.
You're right, a Rocheworld would be an extreme case, but that's half the fun of arguing definitions.
Which was why I argued it.:) Since none of the planets using their definition are true spheres in the geometric sense, their use of sphere must mean something else.
I was thinking more about whether they're two planets, as the point they orbit about is inside one of them!
It is? I remember them being tidally locked, very close in mass, and physically close enough that you could travel from one to the other through their combined atmosphere. I though the point they orbited was almost exactly between them.
While I learned to touch type on a mechanical typewriter for many years I used various terminals and computer systems that had the control key to the left of the A and the caps lock key in the lower left corner. For a while AT style keyboard had a switch to swap the keys and after that I would take them apart and rewire them permanently. When they switched to using a membrane instead of a printed circuit board I resorted to software. When software no longer works I will probably build a keyboard dongle to do the substitution although that might take some effort for USB.
In short, it does not but only because Rocheworld is an extreme case.
They use the term sphere in the definition but that is the ideal where the planet is not influenced by outside gravitational force. Even the moon distorts the earth into a non spherical shape although it is cyclic since the earth is not tidally locked to the moon. The Rocheworld was tidally locked and the shape followed (please excuse me if I get the term wrong) an equalpotential curve of the graviational force which imo is what they really mean by sphere in the definition.
DNA reuse extends from large structures all the way down to the smallest chemical reactions. Any why not? There are many more ways to do something incorrectly than correctly so beneficial mutations are more likely to be based on existing DNA code and structures than entirely new ones.
What I always found amazing if not unexpected is the comparison to higher level languages and interpreted code. The HOX genes are used to define body structures in a universal way across species as diverse as mammals and insects in a way that looks a lot like a state machine driven interpreter. It is possible to swap these genes between species who last shared a common ancestor hundreds of millions of years ago and still have the state machine interpret them correctly.
As the magnetic domains written to the disk become smaller in volume the particles making up the magnetic media on the disk have to become smaller also. The super paramagnetic limit is reached when thermal energy alone is enough to alter them. Perpendicular recording makes better use of the available volume of material and lowers the surface area needed for each written domain.
Enterprise drives are optimized specifically for low access time which requires higher rotation speeds to lower latency and smaller platters to lower seek time. Since they are typically used in large arrays to further raise the number of operations per second, storage capacity is secondary.
The larger the population per server the more opportunity there will be for scaling problems on both the client side and the server side when too many players enter one area and interact. EQ faced this problem a lot when players organized very large raids in single zones. Players would increasingly go link dead during the raid or worst case the zone would crash and reset kicking everybody off often with disastrous (or humorous depending on your point of view) results if they logged back on individually and appeared in a dangerous area one by one. We actually did better by artificially limiting the size of our raids to specifically avoid these problems because the added control and reliability more then made up for the loss of firepower and healing. At the extreme, larger and better equipped guilds used 4 times as many players to fail at events we considered routine.
As the game ages and the character demographics on each server change, you have a compounding problem where there are not enough characters at some level of advancement to viably form groups. This gets even worse as expansions are added because you split the existing server population more finely. In EQ for example, the LDoN expansion which was mostly composed of instanced content became completely unusable 2 expansions later because at any one time on one server there were never enough players to form a group for it. Eventually players are lost to other games because they have no way to meet up with those at a similar level which only compounds the problem. They can start new characters and potentially face the same problem, pay(!) to transfer to another server with the risk of again facing the same problem, or switch to another game. I would expect game companies to at least make the options that retain the player more likely but have never seen them take steps that effectively did so.
I do not know how Blizzard is handling these issues but they would do well to learn from SOE's mistakes unless they wish to travel the same path. I was planning on at least trying Vanguard but after it was announced SOE would be involved I lost all interest for fear of repeating my experiences with them.
One odd thing that has bothered me about these types of games is that they always seem to make the same mistakes and not learn from even the documented past. I am not sure if it is because the designers are unaware or if it is because it is not worth the small effort needed.
The 68008 was not available until sometime after the 8088 was in production. The Wikipedia article on the subject is pretty close to what I remember of the time. Keep in mind that the Amiga and Atari ST came out some time after the IBM PC and the cost premium of a 16 bit data bus had lowered significantly nevermind the availability improvements in cheap 68000 compatable peripheral chips. I never did any 68000 bus designs but my understanding is that 6800 bus compatability was marginal.
I'm confused now: AFAIK Motorola didn't sell the Z8000
I was refering back to the reasons they did not use the 68000 instead of the 8088. The technical reasons are pretty well cut and dried so this is just speculation on my part.
I may have been thinking of the Z800 but the comparability via mode switching looks questionable and it would have been released a couple of years too late. I assumed Zilog had a 16 bit alternative to the 8080 available at that time but it does not look like it. I doubt IBM would have selected it even had it been available because of second source issues.
Besides the technical reasons I wonder if IBM had problems dealing with Motorola in other ways. While I never had trouble getting data sheets and support documentation from them actual availability was always an problem. I assumed it was because they and their distributors were not interested customers smaller then super colossal giant but maybe something else was going on.
Most of the later CP/M systems I used were Z80 based naturally but I switched over to the 8086 soon afterwards and never used any of the later Zilog processors. Internally we used 8085s because of the simplified interfacing and system requirements.
The 32 bit address space includes the PCI and peripheral address space. Hardware and software permitting you can use PAE mode to increase the physical address space but Windows XP is artificially limited to 4 GB of physical addresses. Adding a couple of 512 MB video cards and the rest of the memory mapped I/O will limit maximum usable physical RAM to not much more then 2 GB and perhapse not even that much.
Thank you IBM for picking the cheapest chip you could for your little test machine called the 5150. If you only knew that it was going to become the standard for the next 20+ years I bet you would have done things differently.
To be fair to IBM, the 16 bit 8088 had an 8 bit external bus that could use the existing 8 bit peripherals and allowed easy assembly language translation from the existing 8080 CP/M code base. Zilog probobably had a similar processor available based on the Z80 but if IBM had used it we would still have the same issues in the present.
Sigh. I miss PIP. Of course, I miss FID also if only for sentimental reasons. I did not have access to hardware capable of running Unix for a long time or I am sure I would have made that transition earlier.
One of the first utilities I wrote for CP/M involved pacing the output to our printers because we had no flow control by keeping track of both carriage and roller movement. If I had had the source to PIP, I would have modified it instead of writing a separate program.
Slashdot Mirror
EE Times had an article several years ago (40th anniversary maybe?) detailing the history of hard drives with a nice description of a lab with Dixie cups containing iron oxide paint for spin coating the platters by hand.
Seagate also made a couple of drives with 2 heads per platter surface and only 1 actuater. The head assembly was known as a C-Blk instead of an E-blk. The model number was ST-425(20 MB). Seagate also produced a half height version called the ST-212(10 MB) with 1 platter and 4 heads and the ST-206(5 MB)which only used 1 surface and 2 heads.
I was not able to find anything on the net about those but given the model numbers they would have been using stepper motors instead of voice coils so the track density would have been low enough to allow multiple heads to stay in alignment. This is an obvious way to improve transfer rates and seek times but it limits the track density if you want to use the heads in parallel. The dual head assembly Seagate drives I mentioned used voice coils and embedded servo tracking although the only reference I found to this was the Connor "Chinook" series so maybe I got the two manufacturers confused. Apparently the Seagate Barracuda ST12450W using one head assembly could read simultaneously from both heads on each platter but I do not remember it.
All of these techniques have been left behind because of track density and RAID which makes economic sense.
Personally, I'm waiting for them to cram 2 opposing sets of read/write arms (or even just a second set for reading) so that they can effectively halve the latency and seek times without having to go faster than the existing 15k screamers.
For a short time Seagate made a series of drives with dual head assemblies for transactional processing but they were not cost effective. I do not remember how the interfacing worked.
The fact of the matter is, a gun is far more likely to kill the owner of the gun, a casual bystander, or someone totally different (when the gun is stolen), than it is a criminal while protecting life or property.
Comparing the number of criminals killed with firearms to other firearm deaths leaves out the vast majority of cases where the presence of a firearm prevents a crime when not discharged or even brandished.
Lucky for us then this would be a civil war instead of a real war and unlike our earlier one which was more of a war between the States with defined boundaries.
As I am sure others will point out, ownership of everything up to machine guns is legal in most of the States although for the later considerable civilian licensing is involved. Given the targets, hunting weapons are more then adequate and probably preferred.
I would still prefer that U.S. versus Miller was defendable one way or the other since the lack of any defense opposition made the govenment's weak case possible and had the result of producing bad law which we continue to pay for.
In the movie, Aragorn randomly hands the four hobbits four short swords right before the Nazgul attack at Amon Sul. He doesn't explain where they come from nor how he came to have them. Later, Merry uses his to stab the Witch King in the back of the knee, which despite the admonition "no man can slay me," seems to be pretty effective at hurting him and rendering him vulnerable to Eowyn's coup de grace. But nobody knows why.
I always enjoyed this type of thing in books. Not only does the Witch King unknowingly face a woman rendering the prophecy in doubt but Merry happens to have a weapon specifically spelled to hurt him that was only gained through chance. Mysterious as he is, perhaps that was Tom's way of indirectly aiding the war.
What really bothered me about this in the movie was that they could very well have had Aragorn give the swords to the hobbits and in passing give mention to their origin (or someone else could have noticed the odd weapons and told them later if time was an issue) replacing the similar book scene with Tom.
Of course, I think Frodo's movie addition in Osgiliath among other unneeded additions should have been left out in favor of the Scouring of the Shire. I have never found a satisfactory explanation for that. I was rather looking forward to fierce hobbits and in the case of Merry and Pippin contrastingly tall ones. As it was, the only lasting impressions were on Frodo and Sam.
Once outed, what would they have left to loose through association. Conversely, outing them would make their association stronger and their voices louder. This would be the last thing the social conservatives would want. They're bullies. They don't want to actually fight them, they only want them to cower and submit to their will out of fear of alienation.
I do not particularly disagree with this however it requires only a minority of social conservatives to implement this which will have the effect of essentially committing the rest least they denounce their own which I find even less likely.
Since there is no upper bound on the complexity of existing law, create one by adding an ammendment specifying that all federal laws expire after a certain amount of time. Nothing would preclude congress from passing them again but this would create pressure to work on what they believe is truely important.
Something better like Stalin, Mao Tse-tung, Hitler, and Pol Pot? We hardly need the libertarians to create our own dystopias.
Why would the libertarians and anarchists object? It sounds like one of the things they would try although the libertarians consider contract enforcement one of the proper functions of government.
Is it out of the question that after some future date it may not be feasable to travel while carrying portable storage do to increasingly strict national reactions to terrorism? This could not only be because of the percieved threat of smuggling a weapon but also because law enforcement could take the opportunity to duplicate and inspect user data as happened recently in a US border incident.
If our sole purpose was to reproduce a few times and die "young" (before 35), then why do our cells have so many proteins dedicated to detecting and repairing chromosome damage? Shouldn't they deactivate after 35 years?
It is not so much that they should deactivate after 35 years but more that they are left running.
Why would nature keep old people around? How does nature select for old age genes if you reproduce when young?
Nature would only select for long life genes if they gave preferential advantage to the caregiver's children who also have a significant chance of having the same genes.
Some theories are that older people pass their life's knowledge to the next generation, without the next generation having to experience it themselves. Older people act as secondary caregivers, freeing the younger generation to do "useful stuff".
Notice here that if the benefit of the older caregiver is shared by those who are not related to them, the genes involved will not be selected for as strongly unless group selection is significant.
There's no reason to believe our bodies were made to wear out at 60 or 70. Eat less calories, more fruit and veggies high in anti-oxidant compounds, exercise (physical labor and mental), and there's no reason that our bodies couldn't last... longer. How much longer? One study says maybe 120 years.
It is not so much that we were designed to wear out at a certain point but instead that we are running on inertia so to speak. Preventative maintenance certainly has an effect but it is not the whole story. An even better question is where are the genes that affect behavior such as to lead a healthy life style? They are in the same position as our other genes that affect lifetime.
Experiments have been done on short lived species like possums that live on the east coast. Their normal life span is about 2 years at which point if they have not been run over or eaten they start dying of cancer and other things that we would consider old age. It is not a change in their environment that has caused this but an accumulation of genes that cause or allow cancer and other problems because their life is so short anyway do to environmental hazards that they have to go all out just to reproduce before some calamity takes them. In the past, any gene that gave them a reproductive advantage was selected for despite any problems it caused a short time later. Even when kept in a benign environment, they only live to maybe 4 years and then succumb to old age. The situation is reminiscent of how salmon spawn though not as extreme.
When these possums were bred for extended life by delaying when they have children, the genes causing and allowing old age were quickly removed from the pool. Onset of cancer happened later and their lifetime was extended several times over. I am reminded of Heinlein's Lazarus Long and the Ira Howard Foundation.
Atmospheres are part of a planet, aren't they? If their atmospheres (or oceans) are covering the center of mass, then it's within the planet/planets.
That had not occurred to me. Except in the case of a gas giant where the outer fluid is a significant fraction of the planet's mass, I would not have considered the atmosphere when determining where the barycenter is because you could end up with a situation where with two otherwise identical double planets one meets the definition and another does not just because of the difference in having an atmosphere which is contributing almost nothing. At what depth does Jupiter's atmospere become liquid? That might be a more appropriate surface to compare the barycenter to.
The more I think about it, the more I think they might well be considered one planet, anyway. Given some time, they certainly will be!
I am not so sure about that. Since the planets are already tidally locked to each other they only have to deal with tides caused by their primary which are much smaller. They will radiate gravitational energy but because of their relatively low mass, I doubt it will be significant. If they were not tidally locked, then they would move apart until they were or they became seperated like the Earth's moon currently is doing.
You're right, a Rocheworld would be an extreme case, but that's half the fun of arguing definitions.
:) Since none of the planets using their definition are true spheres in the geometric sense, their use of sphere must mean something else.
Which was why I argued it.
I was thinking more about whether they're two planets, as the point they orbit about is inside one of them!
It is? I remember them being tidally locked, very close in mass, and physically close enough that you could travel from one to the other through their combined atmosphere. I though the point they orbited was almost exactly between them.
While I learned to touch type on a mechanical typewriter for many years I used various terminals and computer systems that had the control key to the left of the A and the caps lock key in the lower left corner. For a while AT style keyboard had a switch to swap the keys and after that I would take them apart and rewire them permanently. When they switched to using a membrane instead of a printed circuit board I resorted to software. When software no longer works I will probably build a keyboard dongle to do the substitution although that might take some effort for USB.
In short, it does not but only because Rocheworld is an extreme case.
They use the term sphere in the definition but that is the ideal where the planet is not influenced by outside gravitational force. Even the moon distorts the earth into a non spherical shape although it is cyclic since the earth is not tidally locked to the moon. The Rocheworld was tidally locked and the shape followed (please excuse me if I get the term wrong) an equalpotential curve of the graviational force which imo is what they really mean by sphere in the definition.
DNA reuse extends from large structures all the way down to the smallest chemical reactions. Any why not? There are many more ways to do something incorrectly than correctly so beneficial mutations are more likely to be based on existing DNA code and structures than entirely new ones.
What I always found amazing if not unexpected is the comparison to higher level languages and interpreted code. The HOX genes are used to define body structures in a universal way across species as diverse as mammals and insects in a way that looks a lot like a state machine driven interpreter. It is possible to swap these genes between species who last shared a common ancestor hundreds of millions of years ago and still have the state machine interpret them correctly.
As the magnetic domains written to the disk become smaller in volume the particles making up the magnetic media on the disk have to become smaller also. The super paramagnetic limit is reached when thermal energy alone is enough to alter them. Perpendicular recording makes better use of the available volume of material and lowers the surface area needed for each written domain.
Enterprise drives are optimized specifically for low access time which requires higher rotation speeds to lower latency and smaller platters to lower seek time. Since they are typically used in large arrays to further raise the number of operations per second, storage capacity is secondary.
The larger the population per server the more opportunity there will be for scaling problems on both the client side and the server side when too many players enter one area and interact. EQ faced this problem a lot when players organized very large raids in single zones. Players would increasingly go link dead during the raid or worst case the zone would crash and reset kicking everybody off often with disastrous (or humorous depending on your point of view) results if they logged back on individually and appeared in a dangerous area one by one. We actually did better by artificially limiting the size of our raids to specifically avoid these problems because the added control and reliability more then made up for the loss of firepower and healing. At the extreme, larger and better equipped guilds used 4 times as many players to fail at events we considered routine.
As the game ages and the character demographics on each server change, you have a compounding problem where there are not enough characters at some level of advancement to viably form groups. This gets even worse as expansions are added because you split the existing server population more finely. In EQ for example, the LDoN expansion which was mostly composed of instanced content became completely unusable 2 expansions later because at any one time on one server there were never enough players to form a group for it. Eventually players are lost to other games because they have no way to meet up with those at a similar level which only compounds the problem. They can start new characters and potentially face the same problem, pay(!) to transfer to another server with the risk of again facing the same problem, or switch to another game. I would expect game companies to at least make the options that retain the player more likely but have never seen them take steps that effectively did so.
I do not know how Blizzard is handling these issues but they would do well to learn from SOE's mistakes unless they wish to travel the same path. I was planning on at least trying Vanguard but after it was announced SOE would be involved I lost all interest for fear of repeating my experiences with them.
One odd thing that has bothered me about these types of games is that they always seem to make the same mistakes and not learn from even the documented past. I am not sure if it is because the designers are unaware or if it is because it is not worth the small effort needed.
The 68008 was not available until sometime after the 8088 was in production. The Wikipedia article on the subject is pretty close to what I remember of the time. Keep in mind that the Amiga and Atari ST came out some time after the IBM PC and the cost premium of a 16 bit data bus had lowered significantly nevermind the availability improvements in cheap 68000 compatable peripheral chips. I never did any 68000 bus designs but my understanding is that 6800 bus compatability was marginal.
I'm confused now: AFAIK Motorola didn't sell the Z8000
I was refering back to the reasons they did not use the 68000 instead of the 8088. The technical reasons are pretty well cut and dried so this is just speculation on my part.
I may have been thinking of the Z800 but the comparability via mode switching looks questionable and it would have been released a couple of years too late. I assumed Zilog had a 16 bit alternative to the 8080 available at that time but it does not look like it. I doubt IBM would have selected it even had it been available because of second source issues.
Besides the technical reasons I wonder if IBM had problems dealing with Motorola in other ways. While I never had trouble getting data sheets and support documentation from them actual availability was always an problem. I assumed it was because they and their distributors were not interested customers smaller then super colossal giant but maybe something else was going on.
Most of the later CP/M systems I used were Z80 based naturally but I switched over to the 8086 soon afterwards and never used any of the later Zilog processors. Internally we used 8085s because of the simplified interfacing and system requirements.
The 32 bit address space includes the PCI and peripheral address space. Hardware and software permitting you can use PAE mode to increase the physical address space but Windows XP is artificially limited to 4 GB of physical addresses. Adding a couple of 512 MB video cards and the rest of the memory mapped I/O will limit maximum usable physical RAM to not much more then 2 GB and perhapse not even that much.
Thank you IBM for picking the cheapest chip you could for your little test machine called the 5150. If you only knew that it was going to become the standard for the next 20+ years I bet you would have done things differently.
To be fair to IBM, the 16 bit 8088 had an 8 bit external bus that could use the existing 8 bit peripherals and allowed easy assembly language translation from the existing 8080 CP/M code base. Zilog probobably had a similar processor available based on the Z80 but if IBM had used it we would still have the same issues in the present.
Sigh. I miss PIP. Of course, I miss FID also if only for sentimental reasons. I did not have access to hardware capable of running Unix for a long time or I am sure I would have made that transition earlier.
One of the first utilities I wrote for CP/M involved pacing the output to our printers because we had no flow control by keeping track of both carriage and roller movement. If I had had the source to PIP, I would have modified it instead of writing a separate program.