For instance what does she do when she needs to ask about/buy something in a store or whatever?
For my travels in Europe, including Germany, I made sure to at least know how to say hello, thank you, please, water closet and beer (the essentials) in the local language. I never really needed any more than hello. Whenever I walked into a shop or restaurant and said hello in the local language the other person smiled and started speaking to me in English. Exceptions were rare, although admittedly I was generally in the larger cities. And for the exception, a very small clothing shop near my hotel in Paris where I needed a belt (forgot to pack one) hand gestures worked just fine.
My understanding is that many Europeans speak English to each other when traveling. And that there is a bit of a generational component to it, the "younger" generations being more likely to speak English to some practical degree. I suspect that an American traveling in Europe needs only to show the some courtesy to the locals -- ex saying hello, please and thank you in the local language -- and they will find plenty of English speaking employees willing to offer their products and services.
Personal observation: if a bar is not terribly busy the bartender may be a valuable resource in getting the pronunciation of key words and phrases correct. A bartender in Prague was of great help to me with Czech.
The problem is that MBAs are hired to manage teams they have no experience with. Having a general knowledge of the business is not going to help you manage the tech support team, it's not going to help you manage the product development team that requires in-depth knowledge of the product, and it's not going to help you manage a group of web coders who deal with technical details.
And business school teaches that a manager must be familiar with the product, familiar with the production line, familiar with the development process, etc... in other words familiar with what they are managing. Do I really need to mention for a third time that 1/3 of my class were scientists and engineers? Working scientists and working engineers with many years of experience are getting MBAs too.
In business school you read lots of case studies. Case studies of successful companies and projects, and case studies of unsuccessful one. These generally include failed companies/projects that were simply managed by the wrong person. Sometimes a person without relevant experience, sometimes a person with relevant experience but who was too egotistical to listen to his team, sometimes a person with relevant experience who could not effectively communicate with others in the company or with customers,...
I've met ONE MBA who was a good manager, and he was a good manager before he got his MBA. Nor did he change his management style after completing his MBA. From a management perspective, his MBA was a complete and utter waste of time.
As I said, what you were when you entered business school is what you leave as, a scientist, an engineer or in this case engineering manager (assuming I guessed your field correctly). Most tech people who finds themselves promoted into management will benefit from a modern business school program. Your evaluation of your manager may be incomplete. His improvements may have been more outward facing with respect to the team, better representing your team and its interests to others in the company, others without an engineering background. Also being a good manager before doesn't mean that inward facing skills had not improved. The improvements may have merely not been things you were aware of or noticed.
Every other MBA manager I've dealt with in my life was cut from the same cloth: "respect me because I have an MBA."
In 30+ years I've seen fools and outright crooks too. It fed my erroneous image of MBAs, I don't actually know how many of the later actually had MBAs. However early in my career I also knew a senior programmer and team leader who was going to business school at night. I assumed it was all bean counting stuff, I eventually learned I was wrong. I also had a CEO at a medium sized company who I knew had an MBA, he sometimes came by to talk to us lowly software engineers. Part of the conversations usually involved asking us if we were getting everything we needed to do our jobs right. His background was management but he understood the functionality offered by our products and could use them in an operational sense. He did not understand the details of hardware or software design, he hired lower level managers who did. And when it came time to hire a new manager for my team he had us lowly software engineers participate in the interview process to make sure our future boss in fact did have the required knowledge. Later I'm my career I've seen some pretty talented coworkers go to business school. In business school I met new people who were pretty capable scientists and engineers and tech people.
If you think an MBA is anything like a CS degree, you're a complete retard. Oh, that's right, I should have known... you're an MBA. MBA's don't have a fraction of the intelligence it takes to finish a CS degree. Here's a litmus test: Anyone from any field can float in and earn an MBA; the same cannot be said of a CS degree.
Actually I used CS as an example because that is my field. I have both Bachelors and Masters degrees in CS. I also ranked fairly high on the CS specific GRE to get into grad school. And for two decades or so I held the severely misinformed opinions you and the GP held. Then I went to business school and learned the truth.
You need to re-read the above more carefully. My comparison to CS was an attempt to frame the problem in a manner coders could understand, basically that both CS and MBA students are in fact taught to do things the correct way. However many fail to do so once they enter the field. Producing buggy unmaintainable crap on the CS side and mismanaging on the MBA side. Its the same failures in both camps, ego and overly optimistic. Both camps getting slammed by the inevitable unforeseen consequences of their shortcuts. That is the sense that CS and MBA are alike.
By the way. You are completely ignorant of the academic demands of some classes in an MBA program. In one elective marketing class I got to use some of my second year calculus from undergraduate days. Again, it was an elective. The mathematical bent of some classes was one of the pleasant surprises of business school, and I'm talking marketing and strategy not accounting nor finance. Areas where I never expected it.
I'll also repeat that 1/3 of my classmates had a scientific or engineering background. By that I meant they personally had scientific and engineering degrees.
If you're an MBA, remember that to everyone who works for you, that stands for "Moronic, Bullshitting Asshole".
I've seen plenty of of managers without an MBA degree fit that description. Some self-taught, some who had finished college, and some who had gone to grad school. And frankly, your post is suggesting you may be heading in that direction. You are also tossing a bit of BS.
I have to confess my opinions were once probably very similar to yours. And I had a couple of cherry picked examples in mind that matched my biased and uninformed opinion. Eventually I attended business school and one of the things I thoroughly enjoyed was seeing just how misinformed I had been. To my great surprise I was being taught to do things the "right way". For example that not all people are the same. That some perform better in one environment or using one methodology while others may perform better in a different environment or using a different methodology. That it was management's job to recognize what works best for each employee rather than try to pound each employee into the same mold.
Now do all MBAs do as they were taught, do they do the "right thing"? Certainly not. Its very much like what I see with Computer Science grads. We were all taught how to create working, reliable and maintainable software but when we enter industry some of us tend to ignore our lessons and start taking shortcuts and crap software results. Just like some MBAs ignore their lessons and start taking shortcuts and mismanagement results. MBAs are very much like CS grads in this regard. Taught how to do things right, but ego and time pressures lead them astray.
What is an MBA? The reality is that what comes out of business school is pretty much the same as what went in, good or bad. An MBA is *not* like other graduate degrees where you delve deeper into a particular field of study. An MBA is *not* about bean counting. An MBA is an overview of an organization. One gets overviews of accounting, finance, marketing, operations, strategy, organizational behavior (psychology, how people behave in groups, i.e. organizations, companies), consumer behavior, project management, product development, etc. One does not become an expert in any of these. Your area of expertise is whatever you brought to business school. In my class was about 1/3 were scientists and engineers.
So if one does not become an expert what's the point? The point is that with this overview of all these different areas you can now understand their wants and needs, their concerns. So you are still the same engineer you were before but now you are an engineer that can effectively communicate with the marketing people, the accounting people, etc. And the payoff is that you become more effective at getting the product requirements out of non-engineers, more effective at communicating the technical issues to non-engineers, you become more effective in persuading non-engineers to your side of an issue. You know how to frame things for the audience be it senior management, accounting, marketing, etc. And sometimes, you will realize that they have a legitimate non-engineering issue that trumps what you want.
An MBA is just like a degree in Computer Science. You are taught to be more effective, you are taught to do things the right way. However grads of both programs sometimes ignore their lessons, take shortcuts and these shortcuts screw projects and companies up.
To me, Celcius makes enormous sense with its 0 being the freezing point of water. The few degrees around 0 make a huge difference in weather and food preparation, to give a few examples where laypeople most think about temperatures.
If a few degrees matter then one should be aware that Celsius is somewhat arbitrary. Freezing and boiling is also a function of elevation. 0 and 100 C are only freezing and boiling at sea level. Go to 3,000m and the numbers differ, especially for boiling which is a more practical number for cooking and water purification.
Way too many family plots and small farms are already determined by acres, or by length of road frontage. Having to redo all of that to metric for every property as it changes hands through sale or inheritance would be a nightmare.
I really doubt that. I've seen surveying documents using simple 3 digit precision decimal numbers of feet (1.234 feet) rather than feet plus inches and fractions of inches. Converting to meters would seem quite trivial and a one-time event.
Similarly in manufacturing I've seen simple 3 digit precision decimal numbers of inches. Fractions often being converted to a decimal 1/1,000th of an inch. Again, converting to millimeters would seem quite trivial and a one-time event for a legacy blueprint as its converted to a CAD document.
It really seems about a legacy personal preference, not technical issues.
"'E could 'a drawed me off a pint," grumbled the old man as he settled down behind a glass. "A 'alf litre ain't enough. It don't satisfy. And a 'ole litre's too much. It starts my bladder running. Let alone the price."
1984, George Orwell. [Doggone lousy Slashdot Unicode support.]
Wait, isn't a pint (16 oz) less than 1/2 liter in both US and Imperial? So a 'alf liter would be more than the accustomed amount.
Cans and bottles of beer are in millilitres; I don't know what they call the size of a glass at a pub because I don't drink.
A 10oz glass is a little less than 300ml, 16oz (pint) a little less than 500ml, 24oz a little more than 700ml, 32oz (quart) a little less than 1L. The differences are quite minor.
And no, people will not "use whatever they are used to". They'll use what's on the package and all of the road signs.
In the US our packaging has included both systems since the 1970s. In the 1970s many of our roadsigns listed both miles and kilometers as well as we were about to switch to metric. We all learned the metric system in school at the time. The US Government Board that was promoting the switch was eliminated during budget cuts in the 1980s. Replacement roadsigns went back to miles only but our speedometers in cars still indicate both systems. Packing remained imperial (well its US variant, not the same as UK and Canada) and metric.
Things are **slowly** switching in a voluntary manner, basically industry involved in foreign trade and the sciences. The US military is largely metric.
Its wasn't much bother to use metric. Calculators were rarely needed. Very rough approximations worked for most circumstances, ex: 1 meter was slightly over 3 feet, a kilometer a little more than half a mile, a liter was slightly more than a quart, a kilogram a little more than 2 pounds. It was only temperature measurements that required something approaching a "formula" to roughly approximate, F -> C then subtract 32 and divide by 2.
This will be a financial boon to the telecom industry. The black budgets are going to have to come up with money to pay for the storage and retrieval by the telcos. I expect this to be quite profitable for them. There are also going to be some nice contracts for redesigning the systems now that the stakeholders have changed.
No. There is nothing new here at all with respect to phone company infrastructure or practices. The metadata is basically what is on your phone bill. The phone companies have always hung on to this data for years. The only new thing that happened was granting the government direct unrestricted access to this data.
The metadata (both numbers, start date/time, duration) is basically what is on a phone bill. Yeah, phone companies have always hung on to that for a while.
...they just passed the cost of retaining all that metadata to the telcos. I pity the telcos.
The telcos have *always* done this. Phone number making the call, phone number being called, date/time call made, duration of call... sound familiar? That's the info on your phone bill. The phone companies have always hung on to this metadata for well over a year.
The *only* thing new was providing the government direct unrestricted access to this billing data so that the government could build an association graph of phone numbers.
Isn't cheating allowed? Write the original OS in C, then compile, and then work on the resultant assembly code, first on optimizing, and then on adding features?
Writing C code and playing with the generated assembly might be OK for a day or two's work in learning assembly on a particular architecture. However for any half serious project its probably a terrible way to get started.
Assembly programming is not about trying to out code generate a compiler. Its about thinking of implementations that are not restricted by the semantics of a particular high level language, of being able to leverage knowledge that can not be given to the compiler, of thinking in terms of the actual hardware architecture.
Well, you could optimize for current arches, too, if you wanted to squeeze all you can out of them.
Don't need to. Their performance was already good. Optimization of an old arch was only necessary to make its performance acceptable, to make it a minimum system requirement. To enlarge the potential market at the time.
To be honest support for the old arch (and WinXP) is being dropped from the next major release of the software. So the original reference C code is going back in to simplify maintenance. The old ASM is no longer needed. The segment of the market it served effectively no longer exists.
It would be fun to redo it for new archs, I enjoy such things, but its not justifiable. I will continue to occasionally compare it against the C code out of curiosity.
Single serving all the things! From flashlights to kittens, all protected with base64 encoded, dual rot13 encrypted DRM goodness!
Interestingly the k-cup patent expired and many other companies began to produce single serve packs for Keurig coffee makers. This led Keurig to redesign and produce a v2.0 coffee maker and packs, with all new patent protection. The CEO just admitted that this 2.0 plan was a total failure and they lost a ton of money.
And you're assembly is probably easy to beat with even pretty crappy SSE2 code.
Apparently not by compilers.
You don't seem to understand the purpose of writing in assembly language. Its not to optimize for the current state of the art box. It is to get acceptable performance from old legacy boxes. Some assembly in the right spot(s) can make the difference between an old architecture making the cutoff in terms of acceptable performance, of being able to include that segment of the market in your minimum system requirements.
My point is that such optimizations for the sake of the old boxes doesn't necessarily do any harm to the new boxes. That worrying about future architectures is a red herring of sorts.
It's an object lesson in using pure assembly. By the time you get anything useful done, technology has moved on.
Not really. I have some computational code that I wrote in assembly 4 Pentium architectures ago. Every new architecture I run it against the C implementation, freshly recompiled with a current compiler. The assembly is still faster given all the hardware and compiler improvements. Now the performance improvement is getting much smaller but it is still a win.
I'm not reallyd sure that I understand that point. To me, thst would sound reasonable for educstionsl Ãr entertainment purposes, but are there any other meaningful reasons for writing an entire OS in assembler?
The entire OS would occupy about 1/3 of an Intel i7's cache. For ultra-high performance apps that might actually be useful.
Of course that includes user land apps and such so the footprint of the OS itself would probably be far smaller.
Slashdot Mirror
For instance what does she do when she needs to ask about/buy something in a store or whatever?
For my travels in Europe, including Germany, I made sure to at least know how to say hello, thank you, please, water closet and beer (the essentials) in the local language. I never really needed any more than hello. Whenever I walked into a shop or restaurant and said hello in the local language the other person smiled and started speaking to me in English. Exceptions were rare, although admittedly I was generally in the larger cities. And for the exception, a very small clothing shop near my hotel in Paris where I needed a belt (forgot to pack one) hand gestures worked just fine.
My understanding is that many Europeans speak English to each other when traveling. And that there is a bit of a generational component to it, the "younger" generations being more likely to speak English to some practical degree. I suspect that an American traveling in Europe needs only to show the some courtesy to the locals -- ex saying hello, please and thank you in the local language -- and they will find plenty of English speaking employees willing to offer their products and services.
Personal observation: if a bar is not terribly busy the bartender may be a valuable resource in getting the pronunciation of key words and phrases correct. A bartender in Prague was of great help to me with Czech.
The problem is that MBAs are hired to manage teams they have no experience with. Having a general knowledge of the business is not going to help you manage the tech support team, it's not going to help you manage the product development team that requires in-depth knowledge of the product, and it's not going to help you manage a group of web coders who deal with technical details.
And business school teaches that a manager must be familiar with the product, familiar with the production line, familiar with the development process, etc ... in other words familiar with what they are managing. Do I really need to mention for a third time that 1/3 of my class were scientists and engineers? Working scientists and working engineers with many years of experience are getting MBAs too.
...
In business school you read lots of case studies. Case studies of successful companies and projects, and case studies of unsuccessful one. These generally include failed companies/projects that were simply managed by the wrong person. Sometimes a person without relevant experience, sometimes a person with relevant experience but who was too egotistical to listen to his team, sometimes a person with relevant experience who could not effectively communicate with others in the company or with customers,
I've met ONE MBA who was a good manager, and he was a good manager before he got his MBA. Nor did he change his management style after completing his MBA. From a management perspective, his MBA was a complete and utter waste of time.
As I said, what you were when you entered business school is what you leave as, a scientist, an engineer or in this case engineering manager (assuming I guessed your field correctly). Most tech people who finds themselves promoted into management will benefit from a modern business school program. Your evaluation of your manager may be incomplete. His improvements may have been more outward facing with respect to the team, better representing your team and its interests to others in the company, others without an engineering background. Also being a good manager before doesn't mean that inward facing skills had not improved. The improvements may have merely not been things you were aware of or noticed.
Every other MBA manager I've dealt with in my life was cut from the same cloth: "respect me because I have an MBA."
In 30+ years I've seen fools and outright crooks too. It fed my erroneous image of MBAs, I don't actually know how many of the later actually had MBAs. However early in my career I also knew a senior programmer and team leader who was going to business school at night. I assumed it was all bean counting stuff, I eventually learned I was wrong. I also had a CEO at a medium sized company who I knew had an MBA, he sometimes came by to talk to us lowly software engineers. Part of the conversations usually involved asking us if we were getting everything we needed to do our jobs right. His background was management but he understood the functionality offered by our products and could use them in an operational sense. He did not understand the details of hardware or software design, he hired lower level managers who did. And when it came time to hire a new manager for my team he had us lowly software engineers participate in the interview process to make sure our future boss in fact did have the required knowledge. Later I'm my career I've seen some pretty talented coworkers go to business school. In business school I met new people who were pretty capable scientists and engineers and tech people.
If you think an MBA is anything like a CS degree, you're a complete retard. Oh, that's right, I should have known... you're an MBA. MBA's don't have a fraction of the intelligence it takes to finish a CS degree. Here's a litmus test: Anyone from any field can float in and earn an MBA; the same cannot be said of a CS degree.
Actually I used CS as an example because that is my field. I have both Bachelors and Masters degrees in CS. I also ranked fairly high on the CS specific GRE to get into grad school. And for two decades or so I held the severely misinformed opinions you and the GP held. Then I went to business school and learned the truth.
You need to re-read the above more carefully. My comparison to CS was an attempt to frame the problem in a manner coders could understand, basically that both CS and MBA students are in fact taught to do things the correct way. However many fail to do so once they enter the field. Producing buggy unmaintainable crap on the CS side and mismanaging on the MBA side. Its the same failures in both camps, ego and overly optimistic. Both camps getting slammed by the inevitable unforeseen consequences of their shortcuts. That is the sense that CS and MBA are alike.
By the way. You are completely ignorant of the academic demands of some classes in an MBA program. In one elective marketing class I got to use some of my second year calculus from undergraduate days. Again, it was an elective. The mathematical bent of some classes was one of the pleasant surprises of business school, and I'm talking marketing and strategy not accounting nor finance. Areas where I never expected it.
I'll also repeat that 1/3 of my classmates had a scientific or engineering background. By that I meant they personally had scientific and engineering degrees.
If you're an MBA, remember that to everyone who works for you, that stands for "Moronic, Bullshitting Asshole".
I've seen plenty of of managers without an MBA degree fit that description. Some self-taught, some who had finished college, and some who had gone to grad school. And frankly, your post is suggesting you may be heading in that direction. You are also tossing a bit of BS.
I have to confess my opinions were once probably very similar to yours. And I had a couple of cherry picked examples in mind that matched my biased and uninformed opinion. Eventually I attended business school and one of the things I thoroughly enjoyed was seeing just how misinformed I had been. To my great surprise I was being taught to do things the "right way". For example that not all people are the same. That some perform better in one environment or using one methodology while others may perform better in a different environment or using a different methodology. That it was management's job to recognize what works best for each employee rather than try to pound each employee into the same mold.
Now do all MBAs do as they were taught, do they do the "right thing"? Certainly not. Its very much like what I see with Computer Science grads. We were all taught how to create working, reliable and maintainable software but when we enter industry some of us tend to ignore our lessons and start taking shortcuts and crap software results. Just like some MBAs ignore their lessons and start taking shortcuts and mismanagement results. MBAs are very much like CS grads in this regard. Taught how to do things right, but ego and time pressures lead them astray.
What is an MBA? The reality is that what comes out of business school is pretty much the same as what went in, good or bad. An MBA is *not* like other graduate degrees where you delve deeper into a particular field of study. An MBA is *not* about bean counting. An MBA is an overview of an organization. One gets overviews of accounting, finance, marketing, operations, strategy, organizational behavior (psychology, how people behave in groups, i.e. organizations, companies), consumer behavior, project management, product development, etc. One does not become an expert in any of these. Your area of expertise is whatever you brought to business school. In my class was about 1/3 were scientists and engineers.
So if one does not become an expert what's the point? The point is that with this overview of all these different areas you can now understand their wants and needs, their concerns. So you are still the same engineer you were before but now you are an engineer that can effectively communicate with the marketing people, the accounting people, etc. And the payoff is that you become more effective at getting the product requirements out of non-engineers, more effective at communicating the technical issues to non-engineers, you become more effective in persuading non-engineers to your side of an issue. You know how to frame things for the audience be it senior management, accounting, marketing, etc. And sometimes, you will realize that they have a legitimate non-engineering issue that trumps what you want.
An MBA is just like a degree in Computer Science. You are taught to be more effective, you are taught to do things the right way. However grads of both programs sometimes ignore their lessons, take shortcuts and these shortcuts screw projects and companies up.
To me, Celcius makes enormous sense with its 0 being the freezing point of water. The few degrees around 0 make a huge difference in weather and food preparation, to give a few examples where laypeople most think about temperatures.
If a few degrees matter then one should be aware that Celsius is somewhat arbitrary. Freezing and boiling is also a function of elevation. 0 and 100 C are only freezing and boiling at sea level. Go to 3,000m and the numbers differ, especially for boiling which is a more practical number for cooking and water purification.
Way too many family plots and small farms are already determined by acres, or by length of road frontage. Having to redo all of that to metric for every property as it changes hands through sale or inheritance would be a nightmare.
I really doubt that. I've seen surveying documents using simple 3 digit precision decimal numbers of feet (1.234 feet) rather than feet plus inches and fractions of inches. Converting to meters would seem quite trivial and a one-time event.
Similarly in manufacturing I've seen simple 3 digit precision decimal numbers of inches. Fractions often being converted to a decimal 1/1,000th of an inch. Again, converting to millimeters would seem quite trivial and a one-time event for a legacy blueprint as its converted to a CAD document.
It really seems about a legacy personal preference, not technical issues.
1984, George Orwell. [Doggone lousy Slashdot Unicode support.]
Wait, isn't a pint (16 oz) less than 1/2 liter in both US and Imperial? So a 'alf liter would be more than the accustomed amount.
No. We don't generally use ratios in metric.
Really, 1/1000 and 1/10 seem to be commonly used. :-)
Cans and bottles of beer are in millilitres; I don't know what they call the size of a glass at a pub because I don't drink.
A 10oz glass is a little less than 300ml, 16oz (pint) a little less than 500ml, 24oz a little more than 700ml, 32oz (quart) a little less than 1L. The differences are quite minor.
And no, people will not "use whatever they are used to". They'll use what's on the package and all of the road signs.
In the US our packaging has included both systems since the 1970s. In the 1970s many of our roadsigns listed both miles and kilometers as well as we were about to switch to metric. We all learned the metric system in school at the time. The US Government Board that was promoting the switch was eliminated during budget cuts in the 1980s. Replacement roadsigns went back to miles only but our speedometers in cars still indicate both systems. Packing remained imperial (well its US variant, not the same as UK and Canada) and metric.
Things are **slowly** switching in a voluntary manner, basically industry involved in foreign trade and the sciences. The US military is largely metric.
Its wasn't much bother to use metric. Calculators were rarely needed. Very rough approximations worked for most circumstances, ex: 1 meter was slightly over 3 feet, a kilometer a little more than half a mile, a liter was slightly more than a quart, a kilogram a little more than 2 pounds. It was only temperature measurements that required something approaching a "formula" to roughly approximate, F -> C then subtract 32 and divide by 2.
This will be a financial boon to the telecom industry. The black budgets are going to have to come up with money to pay for the storage and retrieval by the telcos. I expect this to be quite profitable for them. There are also going to be some nice contracts for redesigning the systems now that the stakeholders have changed.
No. There is nothing new here at all with respect to phone company infrastructure or practices. The metadata is basically what is on your phone bill. The phone companies have always hung on to this data for years. The only new thing that happened was granting the government direct unrestricted access to this data.
And much more I'm sure.
Yes, analyzing the phone network utilization and capacity. Again, nothing mysterious nor new.
All that was new was granting the government direct unrestricted access so that the government could create an association graph.
The metadata (both numbers, start date/time, duration) is basically what is on a phone bill. Yeah, phone companies have always hung on to that for a while.
...they just passed the cost of retaining all that metadata to the telcos. I pity the telcos.
The telcos have *always* done this. Phone number making the call, phone number being called, date/time call made, duration of call ... sound familiar? That's the info on your phone bill. The phone companies have always hung on to this metadata for well over a year.
The *only* thing new was providing the government direct unrestricted access to this billing data so that the government could build an association graph of phone numbers.
Isn't cheating allowed? Write the original OS in C, then compile, and then work on the resultant assembly code, first on optimizing, and then on adding features?
Writing C code and playing with the generated assembly might be OK for a day or two's work in learning assembly on a particular architecture. However for any half serious project its probably a terrible way to get started.
Assembly programming is not about trying to out code generate a compiler. Its about thinking of implementations that are not restricted by the semantics of a particular high level language, of being able to leverage knowledge that can not be given to the compiler, of thinking in terms of the actual hardware architecture.
Well, you could optimize for current arches, too, if you wanted to squeeze all you can out of them.
Don't need to. Their performance was already good. Optimization of an old arch was only necessary to make its performance acceptable, to make it a minimum system requirement. To enlarge the potential market at the time.
To be honest support for the old arch (and WinXP) is being dropped from the next major release of the software. So the original reference C code is going back in to simplify maintenance. The old ASM is no longer needed. The segment of the market it served effectively no longer exists.
It would be fun to redo it for new archs, I enjoy such things, but its not justifiable. I will continue to occasionally compare it against the C code out of curiosity.
Apparently not by compilers.
Duh. Compilers are shit at things like vectorization. They can fail at generating decent code even with intrinsics.
Thank goodness we still have assembly language. :-)
Single serving all the things! From flashlights to kittens, all protected with base64 encoded, dual rot13 encrypted DRM goodness!
Interestingly the k-cup patent expired and many other companies began to produce single serve packs for Keurig coffee makers. This led Keurig to redesign and produce a v2.0 coffee maker and packs, with all new patent protection. The CEO just admitted that this 2.0 plan was a total failure and they lost a ton of money.
And you're assembly is probably easy to beat with even pretty crappy SSE2 code.
Apparently not by compilers.
You don't seem to understand the purpose of writing in assembly language. Its not to optimize for the current state of the art box. It is to get acceptable performance from old legacy boxes. Some assembly in the right spot(s) can make the difference between an old architecture making the cutoff in terms of acceptable performance, of being able to include that segment of the market in your minimum system requirements.
My point is that such optimizations for the sake of the old boxes doesn't necessarily do any harm to the new boxes. That worrying about future architectures is a red herring of sorts.
The 32-bit version is open (GPL), the 64-bit is currently not.
if the timing is that tight, it would allow using secondhand PCs instead of ladder logic controllers.
Think high performance computing too . An entire OS that takes a small fraction of the CPU Cache.
It's an object lesson in using pure assembly. By the time you get anything useful done, technology has moved on.
Not really. I have some computational code that I wrote in assembly 4 Pentium architectures ago. Every new architecture I run it against the C implementation, freshly recompiled with a current compiler. The assembly is still faster given all the hardware and compiler improvements. Now the performance improvement is getting much smaller but it is still a win.
I'm not reallyd sure that I understand that point. To me, thst would sound reasonable for educstionsl Ãr entertainment purposes, but are there any other meaningful reasons for writing an entire OS in assembler?
The entire OS would occupy about 1/3 of an Intel i7's cache. For ultra-high performance apps that might actually be useful.
Of course that includes user land apps and such so the footprint of the OS itself would probably be far smaller.
For a page so spartan, it shouldn't have needed 3 seconds to load on a 100mbit connection.
Well it is hosting its own web page and running on a Pentium MMX 200. :-)
It fits on a floppy disk? We are in 2015, right? What is a 'floppy disk'?
Its a unit of storage space measurement equivalent to about 30 seconds of music from iTunes.
and their website looks like it's from 1995 as well!
So its not bloated and it is fast too? Seems appropriate. :-)