Back in the day, when Galileo was making his telescope, the best glass in the world was made on some island in Greece. The guys that worked there were not allowed to leave the island. In case the secret of the glass making process escaped.
If I recall my high school history correctly the preeminent glass making island was the island of Murano, in Venice, Italy around the time of Galileo, and indeed, the glass makers were not allowed to leave. It even seems there were "bounty hunters" that went chasing after the few ingrates who actually ran off.
Back in the day, when Galileo was making his telescope, the best glass in the world was made on some island in Greece. The guys that worked there were not allowed to leave the island. In case the secret of the glass making process escaped.
If I recall my high school history correctly the preeminent glass making island was the island of Murano, in Venice, Italy around the time of Galileo, and indeed, the glass makers were not allowed to leave. It even seems there were "bounty hunters" that went chasing after the few ingrates who actually ran off.
Seems worth it to me. Spin 2 can complie down really small, and can perform like Spin 1 did. On P2, that could be close to P1 PASM speeds.
It will have inline PASM, and ersmith has an early, suggestive Spin 2 compiling to HUBEXEC.
Frankly, if we see all that work well, it is an awesome set of choices! People may want SPIN as a framework type thing, or to perform a lot of complex logic, things that do not need speed, but may take space. Great. Let them do that. Maybe free up big buffers.
Compiling Spin2 to PASM2 while possible will not be as compact, and may not even perform better. Remember Chip isn't going to write an optimizing compiler (at least I doubt it). So a compile to hubexec would likely just stitch together snippets of PASM2 together that just perform the operations in order as translated, and is unlikely to be doing any of the things that would make it much (if at all) faster than the bytecode version.
Also, keep in mind that he's done some stuff to greatly improve the performance of the bytecode interpreter.
I assume that eventually, someone will make a Spin2 to PASM/hubexec compiler that does nice optimizing to make it much faster, but I doubt that will be out at launch.
Back in the day, when Galileo was making his telescope, the best glass in the world was made on some island in Greece. The guys that worked there were not allowed to leave the island. In case the secret of the glass making process escaped.
If I recall my high school history correctly the preeminent glass making island was the island of Murano, in Venice, Italy around the time of Galileo, and indeed, the glass makers were not allowed to leave. It even seems there were "bounty hunters" that went chasing after the few ingrates who actually ran off.
Ingrates? or Inmates? ;-)
Ingrates, in the sarcastic sense of the word. Of course the ones who remained were inmates. This of course raises the question of whether the inmates who ran off were inmate ingrates or just ingrates ;-)
I assume that eventually, someone will make a Spin2 to PASM/hubexec compiler that does nice optimizing to make it much faster, but I doubt that will be out at launch.
fastspin already does compile some of Spin2 to PASM/hubexec, and it is an optimizing compiler, so in some sense it's out before launch . Of course we'll have to see what else gets added to Spin2 to see what more needs to be added to fastspin. I don't think it'll be hard to support new features.
Are you really saying that your employer controls what you can think, say and do? And the code code you write in your own free time?
The day I showed up at IBM for my first intern position I did indeed have to sign a document wherein I asssigned to IBM anything that I may create or invent, whether it was done on my time or theirs, and whether it had anything to do with my IBM employment or not. I balked for a moment and signed. The need for this document spoke volumes about IBM's past. Clearly they'd been burned many times.
In a practical sense, I'm confident they would never go after an employee for his custom candle dipping business. The legal department simply wanted to claim all the ground they could. And the cusp of lucrative employment seemed as good a time as any to go for it.
In the long run I was the winner. What I got from IBM was far more valuable than anything they got from me. What I learned about advanced technology I could never have gleaned from the outside. Some things they were doing still have not come to light, private pun intended.
All sounds a bit murky to me. Obviously companies what to keep a lid on what they are doing and how they are doing. The mere fact that they present you with a problem to solve that new never knew existed before leans to wards their owning your solution, even if you did it at home over the weekend.
Even I signed away to stringent demands. I had to get security clearance and sign an oath under the British official secrets act before I could do contract work on military projects decades ago. Which basically meant I was not allowed to say anything about anything I worked on. Last I checked there was still a death penalty for violating that agreement!
Like you I think I got more out of that than I put in. It's not like I was ever about to invent anything significant.
On this topic of ownership of thought.. I've dealt with this a bit in the past.
Intellectual property is the real topic. You can work in a company and unless you signed an agreement ahead of time, you will own any patents that come out of things. A company does not provide intellect, its employees do. In states like Michigan, you are pretty much considered hired to invent. You don't have to sign over a patent to them, but they can sue you and win, which is why they should pay you for the patent even though you're considered hired to invent in Michigan (this status has come about from the automakers lobbying the state...) In Michigan, you don't have to sign an intellectual property agreement and the company will win the patent from you.
Now in general, because you were employed by the company, they have the right to make and sell what you develop if you were directed to come up with such a solution. Even if you own the patent, they have an intrinsic right to use the patent as they paid/directed you. If, you come up with something different along the way, related or not, or on your own time, your company does not own it unless you signed some far reaching agreement. If you did, you either sign it over and fight it in court and it will basically come down to if your salary is high enough to warrant it and if what you created on your own is related to your specific work. So they usually have you sign the "I'm hired to invent" documents to cover you.
Another example was two chemical engineers in china. Patents are pretty much worthless unless they prove useful in court and are upheld. In this case, the product the two engineers created was patented and a 40 BILLION dollar industry was protected. This allowed the two engineers to sue the company saying that their salaries were not considerable valuation for the 40 BILLION dollar industry in which the company now holds. The courts found for the engineers and they are now very wealthy.
I personally had a patent assignment placed before me that said "I herby assign ANY invention arising out of the subject matter"... Normally, you can only assign an existing patent and is says "i herby assign said invention...." Sneaky trick to put you into a permanent non compete. While it would not hold up in court, the fact that the document exists means you are going to court....
And that it the crux of all of these things. MANY of them are not valid contracts because they are one sided and do not provide valuable consideration. In terms of non-disclosures, you should be careful. Under regular laws, even here in michigan, you simply can not take actual company documents and use them. You CAN however, sit down with documents and information and write it down by hand on another sheet and use it. You simply can not keep someone from using something in their head. These non-disclosures are used to try and put in OTHER types of information and get you to agree it's covered under intellectual property laws, etc. This again means because the documents exist, you are going to court.
Moral of the story, if you are under these documents, do not use information in a way that will hurt your current/former employer. Because if you've signed one of these things, it means you'll go to court (lots of money) Now, you'll likely win, but be out a LOT of money. Otherwise, keep new ideas under your hat until you are not there anymore.
Back in the day, when Galileo was making his telescope, the best glass in the world was made on some island in Greece. The guys that worked there were not allowed to leave the island. In case the secret of the glass making process escaped.
If I recall my high school history correctly the preeminent glass making island was the island of Murano, in Venice, Italy around the time of Galileo, and indeed, the glass makers were not allowed to leave. It even seems there were "bounty hunters" that went chasing after the few ingrates who actually ran off.
Ingrates? or Inmates? ;-)
Ingrates, in the sarcastic sense of the word. Of course the ones who remained were inmates. This of course raises the question of whether the inmates who ran off were inmate ingrates or just ingrates ;-)
Glass = Silicon.... So what has changed in the last 5 or 6 centuries? Same slavery, different perks and compensation.
But back to the original tools part of the thread. What is to stop anyone from setting up on GITHUB or similar site a tools development project in the way other opensource was developed and controlled with a couple of people (with at least one Parallax person) controlling the final acceptance and attempting to referee the disagreements/flame sessions in a similar manner to what the Linux Kernal developers do? Aside from (un)willingness to share and play nicely when working towards a nicely integrated tool set rather than a widely scattered and fractured batch of similarly functioning tools that all do something the others don't or do, but not all. Admittedly, I am not at the level to contribute anything meaningful to such an effort, but there are a lot out there who can and do.
Just such, but so many repositories, not quite sure which ones are current and which are not. Many appear to be flatlining. Much as I would like to do a bit more with C, not sure which tool is current/useful or still has a way to go before I commit to learning one. Between personal and Parallax versions of tools things seem a bit confusing. That was what I was trying to say.
Just such, but so many repositories, not quite sure which ones are current and which are not. Many appear to be flatlining. Much as I would like to do a bit more with C, not sure which tool is current/useful or still has a way to go before I commit to learning one. Between personal and Parallax versions of tools things seem a bit confusing. That was what I was trying to say.
The only Parallax tool is PNut. Chip has been posting updates to it on the Prop2 FPGA files!!!! thread. It has a GUI with an editor, assembler and loader.
The only C tool is p2gcc, which is a command-line tool. The thread is "Can't Wait for PropGCC on the P2?", the GitHub repository is https://github.com/davehein/p2gcc . It uses the GCC compiler, and contains a P2 assembler and loader. It's also integrated with the spinsim simulator, so you don't even need P2 hardware to test your code.
What did I do wrong.
What's going on here ?
I'm only interested in experimenting with the P2.
I think I can say that I published good reviews as beta tester for the GCC for P1.
The P2 is probably far too complex and powerful for my requirements.
However, even if I do not really need it, I will purchase a Parallax P2 activity board in recognition of the tremendous effort and commitment which has gone into the P2 product project.
Comments
https://forums.parallax.com/discussion/166416/spin2-operator-syntax/p1
Also, I know Chip spent a lot of time optimizing byte code but wouldn't it be better to compile to hubexec for spin2?
If I recall my high school history correctly the preeminent glass making island was the island of Murano, in Venice, Italy around the time of Galileo, and indeed, the glass makers were not allowed to leave. It even seems there were "bounty hunters" that went chasing after the few ingrates who actually ran off.
Ingrates? or Inmates? ;-)
Seems worth it to me. Spin 2 can complie down really small, and can perform like Spin 1 did. On P2, that could be close to P1 PASM speeds.
It will have inline PASM, and ersmith has an early, suggestive Spin 2 compiling to HUBEXEC.
Frankly, if we see all that work well, it is an awesome set of choices! People may want SPIN as a framework type thing, or to perform a lot of complex logic, things that do not need speed, but may take space. Great. Let them do that. Maybe free up big buffers.
Or, go the other extreme and run it HUBEXEC.
Or, something in between using in line PASM.
That kind of flexibility will prove valuable.
He has done some of the work for Spin2 already, but it got set aside to do other things, I assume he is going to pick it back up soon.
Also, keep in mind that he's done some stuff to greatly improve the performance of the bytecode interpreter.
I assume that eventually, someone will make a Spin2 to PASM/hubexec compiler that does nice optimizing to make it much faster, but I doubt that will be out at launch.
Ingrates, in the sarcastic sense of the word. Of course the ones who remained were inmates. This of course raises the question of whether the inmates who ran off were inmate ingrates or just ingrates ;-)
fastspin already does compile some of Spin2 to PASM/hubexec, and it is an optimizing compiler, so in some sense it's out before launch
In a practical sense, I'm confident they would never go after an employee for his custom candle dipping business. The legal department simply wanted to claim all the ground they could. And the cusp of lucrative employment seemed as good a time as any to go for it.
In the long run I was the winner. What I got from IBM was far more valuable than anything they got from me. What I learned about advanced technology I could never have gleaned from the outside. Some things they were doing still have not come to light, private pun intended.
Even I signed away to stringent demands. I had to get security clearance and sign an oath under the British official secrets act before I could do contract work on military projects decades ago. Which basically meant I was not allowed to say anything about anything I worked on. Last I checked there was still a death penalty for violating that agreement!
Like you I think I got more out of that than I put in. It's not like I was ever about to invent anything significant.
Intellectual property is the real topic. You can work in a company and unless you signed an agreement ahead of time, you will own any patents that come out of things. A company does not provide intellect, its employees do. In states like Michigan, you are pretty much considered hired to invent. You don't have to sign over a patent to them, but they can sue you and win, which is why they should pay you for the patent even though you're considered hired to invent in Michigan (this status has come about from the automakers lobbying the state...) In Michigan, you don't have to sign an intellectual property agreement and the company will win the patent from you.
Now in general, because you were employed by the company, they have the right to make and sell what you develop if you were directed to come up with such a solution. Even if you own the patent, they have an intrinsic right to use the patent as they paid/directed you. If, you come up with something different along the way, related or not, or on your own time, your company does not own it unless you signed some far reaching agreement. If you did, you either sign it over and fight it in court and it will basically come down to if your salary is high enough to warrant it and if what you created on your own is related to your specific work. So they usually have you sign the "I'm hired to invent" documents to cover you.
Another example was two chemical engineers in china. Patents are pretty much worthless unless they prove useful in court and are upheld. In this case, the product the two engineers created was patented and a 40 BILLION dollar industry was protected. This allowed the two engineers to sue the company saying that their salaries were not considerable valuation for the 40 BILLION dollar industry in which the company now holds. The courts found for the engineers and they are now very wealthy.
I personally had a patent assignment placed before me that said "I herby assign ANY invention arising out of the subject matter"... Normally, you can only assign an existing patent and is says "i herby assign said invention...." Sneaky trick to put you into a permanent non compete. While it would not hold up in court, the fact that the document exists means you are going to court....
And that it the crux of all of these things. MANY of them are not valid contracts because they are one sided and do not provide valuable consideration. In terms of non-disclosures, you should be careful. Under regular laws, even here in michigan, you simply can not take actual company documents and use them. You CAN however, sit down with documents and information and write it down by hand on another sheet and use it. You simply can not keep someone from using something in their head. These non-disclosures are used to try and put in OTHER types of information and get you to agree it's covered under intellectual property laws, etc. This again means because the documents exist, you are going to court.
Moral of the story, if you are under these documents, do not use information in a way that will hurt your current/former employer. Because if you've signed one of these things, it means you'll go to court (lots of money) Now, you'll likely win, but be out a LOT of money. Otherwise, keep new ideas under your hat until you are not there anymore.
'Me
Best not to get the monkey's attention.
Glass = Silicon.... So what has changed in the last 5 or 6 centuries? Same slavery, different perks and compensation.
But back to the original tools part of the thread. What is to stop anyone from setting up on GITHUB or similar site a tools development project in the way other opensource was developed and controlled with a couple of people (with at least one Parallax person) controlling the final acceptance and attempting to referee the disagreements/flame sessions in a similar manner to what the Linux Kernal developers do? Aside from (un)willingness to share and play nicely when working towards a nicely integrated tool set rather than a widely scattered and fractured batch of similarly functioning tools that all do something the others don't or do, but not all. Admittedly, I am not at the level to contribute anything meaningful to such an effort, but there are a lot out there who can and do.
Of course any of us can set up on github and meany here do so.
I have no idea how to resolve that. Yet another "community" repo could just confusing things even more.
Eric Smith has a Spin2 compiler and PASM2 assembler. It is posted at the fastspin compiler for P2 thread. It also contains a GUI called spin2gui for editing, compiling and loading. The GitHub repository is https://github.com/totalspectrum/spin2cpp .
The only C tool is p2gcc, which is a command-line tool. The thread is "Can't Wait for PropGCC on the P2?", the GitHub repository is https://github.com/davehein/p2gcc . It uses the GCC compiler, and contains a P2 assembler and loader. It's also integrated with the spinsim simulator, so you don't even need P2 hardware to test your code.
Hopefully there will soon be a development board in Germany.
It was some other weirdo posting junk that has now been deleted.
I got the correct suspect.
What's going on here ?
I'm only interested in experimenting with the P2.
I think I can say that I published good reviews as beta tester for the GCC for P1.
It was someone else we were complaining about.
Sorry for the confusion.
Soon I retire and I want to spend some of my time playing with the P2.
best regards
Reinhard
However, even if I do not really need it, I will purchase a Parallax P2 activity board in recognition of the tremendous effort and commitment which has gone into the P2 product project.