... Maybe the Smart Pins features can be simple. Get the P2 prototype chips on their way. While that's happening maybe continue on the Smart Pins in case there are bugs in the prototype, in which case the extras could be added too. This what you call a recipe to ensure there are no bugs found
The exercise with the FPGA is for exactly that. Rushing passed the FPGA testing/experimenting now would be a mistake.
More documentation is number one ... people are testing bits already; that will expand as the document fleshes out. The Prop123-A9 board would be nice but much can be tested already even on the DE0-Nano with it's two Cogs.
Cluso, is the RGB converter feature creep if it was already done and only took a couple of hours/a day to add back in? same with PIX blender, so it may take a couple of days to add back in, but what an addition! Chip and Andre' spent a long time developing that, it would be stupid to waste it, when it can be re-added in a matter of days, we've waited this long a couple of extra days for awesome returns is a no brainer in my book.
It seems that lots of valuable and in some cases rare testing time is being wasted having to trawl through forum threads that lead to nowhere. Let's get back to testing people. Hang in there Chip!
...is the RGB converter feature creep if it was already done and only took a couple of hours/a day to add back in?...
Isn't that the definition of feature creep? A constant trickle of small additions, each of which pushes completion further and further into the future.
I'm guessing that I am not unique, therefore there must be lots of people who lurk around in this forum, not saying much, but still very excited about the development. As much as I would love to participate, other demands on my time prevent me from keeping up to speed.
The changes and features, and changes to features I see are needed exploration of possibilities. I have VERY high expectations of Chip's ability to find the simplest implementation for special purpose features, and then make them generic enough that the possibilities allow use far beyond the stated purpose.
An example is the colorburst generator. As I understand the description, in light of my own interests, this basically equates to a module that does all the heavy lifting for a software defined radio transmitter!. (Maybe even one in each cog??!!)
I saw The CORDIC engine as a radio receiver from the beginning. I've shown some of the posts to amateur radio folks who don't know who or what Parallax is, and in minutes, they are excited about the Propeller 2 possibilities. These folks already cleaned me out of P1 protoboards! (Even the nasty used ones...)
So don't underestimate the excitement. I also have every expectation that the effort and time put into exploring just what the Prop 2 should be will find a proper balance and yield an amazing product when the time is right.
Heck, I'm considering jumping into this because of something you did by accident!
There's no turning back now, perhaps just a rationalization of features, of must-haves from a technical and marketing viewpoint vs those die if I don't get it desires because this is the real world and it's come to a do-or-die effort. The market and economy is changing way too fast and even though the marketplace is crowded and extremely competitive I feel that if a claim is staked in time then the P2 or whatever it will be called may indeed have a foothold to gain some traction. But that's a lot of ifs and a lot more hard work yet ahead.
The tools and documentation also make or break a lot of CPUs out there and I remember that the Propeller was so easy to get into with the tutorials in the PDF manual, the OBEX, and Spin. If P2 is made that easy to get into as well then I am sure it will take-off just like a catapulted carrier jet.
Perhaps the P2, as it is different from the Propeller, could be renamed the Plex, as in a complex structure. Not that it is complex from a software viewpoint, rather a behind the scenes hardware organisation and computing foundation.
I seem to recall the original name for data structures was plex, but that was many, many moons ago; way back in the early days.
Man as I read through this thread I just kept getting more and more depressed. "Did we bite off more than we can chew" is the last thing I wanted to see as a topic from Chip! I had to step away and come back and I was thinking why is this bothering me so much? I stopped planning on using the P2 for any of my future projects over a year ago. However, I continue to find new projects that are P1 based (working on one right now) and often come back to this forum for input and to see what is going on! So it shouldn't bother me.
As I read through this thread I saw so many names of people who have been a tremendous help to me on my prop projects. I can't help but think if this open P2 dev process wasn't going on they would have left the forum in boredom a long time ago. Heck if it takes two more years for the P2 to ship I'm fine with it as long as it keeps everyone interested and on the forum!
Nooo!!!! Not COMplex!!!! Makes it sound difficult!
Funplex isn't serious enough
Simplex - nope, don't want to be thought of as simple
Multiplex - no, people will think it is a movie theater
Googolplex - as long as you are careful with spelling....otherwise, Google will be all over it!
Metroplex - um, not sure what to think about this one...too '80's GQ for me
Flexiplex - just marketing gone bad......
Fundamentally, I'm NOT convinced a P2 is *THAT* different from a P1 until you light up the afterburners.
Don't really care what it is called as long as it is available at some point. P16X32, P2, or something else is fine by me. Personally I would prefer FLEX to PLEX considering how many applications the mix of new architecture and multiple cores could address.
Perhaps the P2, as it is different from the Propeller, could be renamed the Plex, as in a complex structure. Not that it is complex from a software viewpoint, rather a behind the scenes hardware organisation and computing foundation.
I seem to recall the original name for data structures was plex, but that was many, many moons ago; way back in the early days.
What comes after "Propeller" is "Turbine". The T16X32.
"Turbine" might be more compelling than "egg-beater", in reference to the cog/hub memory interface. Egg-beater sounds like old people in pajamas.
From my own point of view, the thing with prop2-hot was that too much effort was made to make things completely orthogonal.
I'm referring in particular to the push to make hw slicing threads work with hubexec, when IMHO that was not necessary at all.
Like having a matrix of features with same labels on rows and columns, and then trying to put a checkmark in every cell.
To be frank, I see some signs of the same mindset again sometimes:
LUT was born with specific purpose, then people asks to use it for everything. Ok, it might be actually good.
But heck, even the HW stack, which was supposed to only be there to support interrupts, was asked to be fully addresable for the sake of having a third in-COG generic storage area. Does that really make sense?
With the P2-hot, in the end I was pleasantly surprised when I saw what ozpropdev made with a single COG.
And, despite the apparent negativity of this post, I have high confidence that the same thing is going to happen again, with no power issues this time.
Just remember and repeat with me: decimal mode in 6502 was an huge waste of silicon, making the chip surpass the 3500 transistor barrier for no real purpose!
Man as I read through this thread I just kept getting more and more depressed. "Did we bite off more than we can chew" is the last thing I wanted to see as a topic from Chip! I had to step away and come back and I was thinking why is this bothering me so much? I stopped planning on using the P2 for any of my future projects over a year ago. However, I continue to find new projects that are P1 based (working on one right now) and often come back to this forum for input and to see what is going on! So it shouldn't bother me.
As I read through this thread I saw so many names of people who have been a tremendous help to me on my prop projects. I can't help but think if this open P2 dev process wasn't going on they would have left the forum in boredom a long time ago. Heck if it takes two more years for the P2 to ship I'm fine with it as long as it keeps everyone interested and on the forum!
I think maybe we need to recognize why we really do these things, and perhaps stop thinking that our goal is to compete with the juggernaut of ARM, and such. This is akin to looking at all the movies being made, which are frustrating to watch, and supposing that the movie we make needs to fit that established genre, because that's what the herd is expecting - mainly because the herd has run in a single direction for a long time, and those in the herd don't realize, as they breathe dust and wear themselves out, that there are a few other cows up in the hills, having a much more pleasant life, not ruled by an endless frenzy which will continue to ignore what cows really enjoy.
Man as I read through this thread I just kept getting more and more depressed. "Did we bite off more than we can chew" is the last thing I wanted to see as a topic from Chip! I had to step away and come back and I was thinking why is this bothering me so much? I stopped planning on using the P2 for any of my future projects over a year ago. However, I continue to find new projects that are P1 based (working on one right now) and often come back to this forum for input and to see what is going on! So it shouldn't bother me.
As I read through this thread I saw so many names of people who have been a tremendous help to me on my prop projects. I can't help but think if this open P2 dev process wasn't going on they would have left the forum in boredom a long time ago. Heck if it takes two more years for the P2 to ship I'm fine with it as long as it keeps everyone interested and on the forum!
I think maybe we need to recognize why we really do these things, and perhaps stop thinking that our goal is to compete with the juggernaut of ARM, and such. This is akin to looking at all the movies being made, which are frustrating to watch, and supposing that the movie we make needs to fit that established genre, because that's what the herd is expecting - mainly because the herd has run in a single direction for a long time, and those in the herd don't realize, as they breathe dust and wear themselves out, that there are a few other cows up in the hills, having a much more pleasant life, not ruled by an endless frenzy which will continue to ignore what cows really enjoy.
Man as I read through this thread I just kept getting more and more depressed. "Did we bite off more than we can chew" is the last thing I wanted to see as a topic from Chip! I had to step away and come back and I was thinking why is this bothering me so much? I stopped planning on using the P2 for any of my future projects over a year ago. However, I continue to find new projects that are P1 based (working on one right now) and often come back to this forum for input and to see what is going on! So it shouldn't bother me.
As I read through this thread I saw so many names of people who have been a tremendous help to me on my prop projects. I can't help but think if this open P2 dev process wasn't going on they would have left the forum in boredom a long time ago. Heck if it takes two more years for the P2 to ship I'm fine with it as long as it keeps everyone interested and on the forum!
I think maybe we need to recognize why we really do these things, and perhaps stop thinking that our goal is to compete with the juggernaut of ARM, and such. This is akin to looking at all the movies being made, which are frustrating to watch, and supposing that the movie we make needs to fit that established genre, because that's what the herd is expecting - mainly because the herd has run in a single direction for a long time, and those in the herd don't realize, as they breathe dust and wear themselves out, that there are a few other cows up in the hills, having a much more pleasant life, not ruled by an endless frenzy which will continue to ignore what cows really enjoy.
From what I observed growing up on a farm, cows mostly enjoy eating. Except for the bull who seemed to have other things on his mind.
BTW, how was the walnut harvest this year? I keep seeing this commercial on TV about California walnuts. It seems like the California Walnut Board is doing its part to keep demand up. They have a website at walnuts.org. There's also a mention of the California Walnut Commission which is funded by "mandatory" assessments of the growers. It sounds like you actually have to pay the commission to be able to grow walnuts in California. They have a webpage on how to prevent walnut theft. One of the items is "pay your dues to the Commission" (just kidding on that one).
Man as I read through this thread I just kept getting more and more depressed. "Did we bite off more than we can chew" is the last thing I wanted to see as a topic from Chip! I had to step away and come back and I was thinking why is this bothering me so much? I stopped planning on using the P2 for any of my future projects over a year ago. However, I continue to find new projects that are P1 based (working on one right now) and often come back to this forum for input and to see what is going on! So it shouldn't bother me.
As I read through this thread I saw so many names of people who have been a tremendous help to me on my prop projects. I can't help but think if this open P2 dev process wasn't going on they would have left the forum in boredom a long time ago. Heck if it takes two more years for the P2 to ship I'm fine with it as long as it keeps everyone interested and on the forum!
I think maybe we need to recognize why we really do these things, and perhaps stop thinking that our goal is to compete with the juggernaut of ARM, and such. This is akin to looking at all the movies being made, which are frustrating to watch, and supposing that the movie we make needs to fit that established genre, because that's what the herd is expecting - mainly because the herd has run in a single direction for a long time, and those in the herd don't realize, as they breathe dust and wear themselves out, that there are a few other cows up in the hills, having a much more pleasant life, not ruled by an endless frenzy which will continue to ignore what cows really enjoy.
From what I observed growing up on a farm, cows mostly enjoy eating. Except for the bull who seemed to have other things on his mind.
BTW, how was the walnut harvest this year? I keep seeing this commercial on TV about California walnuts. It seems like the California Walnut Board is doing its part to keep demand up. They have a website at walnuts.org. There's also a mention of the California Walnut Commission which is funded by "mandatory" assessments of the growers. It sounds like you actually have to pay the commission to be able to grow walnuts in California. They have a webpage on how to prevent walnut theft. One of the items is "pay your dues to the Commission" (just kidding on that one).
Yes, the walnut marketing board actually doubled its fees this year to something like $0.0385 cents per meat pound. If they can get people to buy walnuts, it's worth it, until people would buy them anyway, at which point it will not be worth it. Grower prices are expected to be down 40% from last year. Our Chandler block just did 7,000 lbs in-shell/dried per acre, which was really good. We've got another Chandler block started from seed in January and grafted, already. It's 5 years from production. Then we have a 60-year-old Hartley block which is dying quickly. We cleared another block earlier this year which is now awaiting Sudan grass to kill nematodes before fumigation and replanting. Meanwhile, the state is demanding that all counties come up with "sustainable groundwater management" plans, lest they take control, themselves, which nobody wants around here.
Regarding PIX and the Color Engine... Chip worked on those circuits for a long time. They are proven and really are just a drop in, with integration the thing we have to test.
I wasn't gonna ask for them, given hot, but I'm glad they are gonna make it. They are both great circuits that will add a lot of uses to this chip. I wrote here, "do it in software" and meant it, but man. It was not gonna be anything like what we get now. We will be just shy of pro grade signals, and that is gonna have a lot of uses besides pretty pictures. And it's easy! You don't want to see the ugly mess I was working on, sans these things. And I don't either. No fun at all.
Keep the trust people. The lessons learned are good ones, and we've learned 'em together too.
This has been a long haul, but this design is getting made. The right pieces are there and we are moving at a nice clip. And this whole effort is a novel thing. It's not like anything else, and the product of it will be like P1. It's gonna be fun, powerful, useful, etc... and having that exist amidst a sea of "me too" bog standard, often painful things out there, is totally worth the effort, if you ask me.
Well, that's for ahem, some people not supporting a move to "spud" which makes me want to be a little more irreverent, and besides... great ice breaker! You can drop that thing and people just sort of stop. Almost doesn't matter what it is. They just watch... and then whatever it was does change. Every time.
You watched it too. The whole thing. Almost everyone does. I did.
And yes, I like spud. Use it early and often.
BTW, while looking at VSYNC on the video driver, I happened to glance at Google Docs to see Chip knocking them out. It takes a while, and yes I decided to just watch. When somebody else is doing it, you can see the thought, pause, write a coupla words, pause, delete one, write three more, etc... in action! It's just tough.
I sat and watched Chip write into Google Docs one day (just for a little while, because otherwise it would be creepy) - it really does make you appreciate the documentation process and how long it takes.
It's not creepy at all. That is what suggestion functionality is for. I recently did a project with one of the Google Alphabet companies. We used the Google tools for most interactions, and it was a powerful experience. Really big docs get kind of slow, but for most things, it's great to have multiple people in there doing stuff, etc...
One item was a bill of activities. At one point, we had something like 10 people attacking a spreadsheet. It got done in like an hour. That same amount of intercommunication would have turned that into a couple days using email and file drops, or just slinging stuff back and forth.
Y.A.C.S. - Yet Another COW Story ... Sunday we took the girls to the horse jumping and roundup show at the Oklahoma City stock yard fair grounds. The roundup portion did not have any roping, but only herding techniques based on the rider and the horse. A few cows did not want to cooperate so the announcer of the event would periodically sound the buzzer and announce "New COW"... <-- I guess what was funny was the way in which the announcer said "New COW". Imagine some "billy-bob Texas tall-boot" with a big chew in a deep cowboy voice saying "New COW".
So the on-going joke at work here is that if a component isn't quite up to standard we have been saying ... "New COW"
... So along similar lines with regard to the question at hand in this thread, some of us might be tempted to say "New COW" if our expectations are less than what we had expected with the "Chip formerly called P2" , "Chip which shall not be named" , or whatever it is to be called.
Comments
The exercise with the FPGA is for exactly that. Rushing passed the FPGA testing/experimenting now would be a mistake.
More documentation is number one ... people are testing bits already; that will expand as the document fleshes out. The Prop123-A9 board would be nice but much can be tested already even on the DE0-Nano with it's two Cogs.
Isn't that the definition of feature creep? A constant trickle of small additions, each of which pushes completion further and further into the future.
The changes and features, and changes to features I see are needed exploration of possibilities. I have VERY high expectations of Chip's ability to find the simplest implementation for special purpose features, and then make them generic enough that the possibilities allow use far beyond the stated purpose.
An example is the colorburst generator. As I understand the description, in light of my own interests, this basically equates to a module that does all the heavy lifting for a software defined radio transmitter!. (Maybe even one in each cog??!!)
I saw The CORDIC engine as a radio receiver from the beginning. I've shown some of the posts to amateur radio folks who don't know who or what Parallax is, and in minutes, they are excited about the Propeller 2 possibilities. These folks already cleaned me out of P1 protoboards! (Even the nasty used ones...)
So don't underestimate the excitement. I also have every expectation that the effort and time put into exploring just what the Prop 2 should be will find a proper balance and yield an amazing product when the time is right.
Heck, I'm considering jumping into this because of something you did by accident!
The tools and documentation also make or break a lot of CPUs out there and I remember that the Propeller was so easy to get into with the tutorials in the PDF manual, the OBEX, and Spin. If P2 is made that easy to get into as well then I am sure it will take-off just like a catapulted carrier jet.
I seem to recall the original name for data structures was plex, but that was many, many moons ago; way back in the early days.
As I read through this thread I saw so many names of people who have been a tremendous help to me on my prop projects. I can't help but think if this open P2 dev process wasn't going on they would have left the forum in boredom a long time ago. Heck if it takes two more years for the P2 to ship I'm fine with it as long as it keeps everyone interested and on the forum!
Funplex isn't serious enough
Simplex - nope, don't want to be thought of as simple
Multiplex - no, people will think it is a movie theater
Googolplex - as long as you are careful with spelling....otherwise, Google will be all over it!
Metroplex - um, not sure what to think about this one...too '80's GQ for me
Flexiplex - just marketing gone bad......
Fundamentally, I'm NOT convinced a P2 is *THAT* different from a P1 until you light up the afterburners.
16th element
Ok, not really but it does have atomic weight of 16 (usually)
What comes after "Propeller" is "Turbine". The T16X32.
"Turbine" might be more compelling than "egg-beater", in reference to the cog/hub memory interface. Egg-beater sounds like old people in pajamas.
I'm referring in particular to the push to make hw slicing threads work with hubexec, when IMHO that was not necessary at all.
Like having a matrix of features with same labels on rows and columns, and then trying to put a checkmark in every cell.
To be frank, I see some signs of the same mindset again sometimes:
LUT was born with specific purpose, then people asks to use it for everything. Ok, it might be actually good.
But heck, even the HW stack, which was supposed to only be there to support interrupts, was asked to be fully addresable for the sake of having a third in-COG generic storage area. Does that really make sense?
With the P2-hot, in the end I was pleasantly surprised when I saw what ozpropdev made with a single COG.
And, despite the apparent negativity of this post, I have high confidence that the same thing is going to happen again, with no power issues this time.
Just remember and repeat with me: decimal mode in 6502 was an huge waste of silicon, making the chip surpass the 3500 transistor barrier for no real purpose!
I think maybe we need to recognize why we really do these things, and perhaps stop thinking that our goal is to compete with the juggernaut of ARM, and such. This is akin to looking at all the movies being made, which are frustrating to watch, and supposing that the movie we make needs to fit that established genre, because that's what the herd is expecting - mainly because the herd has run in a single direction for a long time, and those in the herd don't realize, as they breathe dust and wear themselves out, that there are a few other cows up in the hills, having a much more pleasant life, not ruled by an endless frenzy which will continue to ignore what cows really enjoy.
Herd direction, hmm, there we go, The Maverick!
BTW, how was the walnut harvest this year? I keep seeing this commercial on TV about California walnuts. It seems like the California Walnut Board is doing its part to keep demand up. They have a website at walnuts.org. There's also a mention of the California Walnut Commission which is funded by "mandatory" assessments of the growers. It sounds like you actually have to pay the commission to be able to grow walnuts in California. They have a webpage on how to prevent walnut theft. One of the items is "pay your dues to the Commission" (just kidding on that one).
Yes, the walnut marketing board actually doubled its fees this year to something like $0.0385 cents per meat pound. If they can get people to buy walnuts, it's worth it, until people would buy them anyway, at which point it will not be worth it. Grower prices are expected to be down 40% from last year. Our Chandler block just did 7,000 lbs in-shell/dried per acre, which was really good. We've got another Chandler block started from seed in January and grafted, already. It's 5 years from production. Then we have a 60-year-old Hartley block which is dying quickly. We cleared another block earlier this year which is now awaiting Sudan grass to kill nematodes before fumigation and replanting. Meanwhile, the state is demanding that all counties come up with "sustainable groundwater management" plans, lest they take control, themselves, which nobody wants around here.
I wasn't gonna ask for them, given hot, but I'm glad they are gonna make it. They are both great circuits that will add a lot of uses to this chip. I wrote here, "do it in software" and meant it, but man. It was not gonna be anything like what we get now. We will be just shy of pro grade signals, and that is gonna have a lot of uses besides pretty pictures. And it's easy! You don't want to see the ugly mess I was working on, sans these things. And I don't either. No fun at all.
Keep the trust people. The lessons learned are good ones, and we've learned 'em together too.
This has been a long haul, but this design is getting made. The right pieces are there and we are moving at a nice clip. And this whole effort is a novel thing. It's not like anything else, and the product of it will be like P1. It's gonna be fun, powerful, useful, etc... and having that exist amidst a sea of "me too" bog standard, often painful things out there, is totally worth the effort, if you ask me.
Go write some PASM...
You watched it too. The whole thing. Almost everyone does. I did.
And yes, I like spud. Use it early and often.
BTW, while looking at VSYNC on the video driver, I happened to glance at Google Docs to see Chip knocking them out. It takes a while, and yes I decided to just watch. When somebody else is doing it, you can see the thought, pause, write a coupla words, pause, delete one, write three more, etc... in action! It's just tough.
let's make sure we recognize and appreciate that.
(thanks chip!)
I like that!
I guess the hub ram is the wide green pasture
There'd be 16 sets of teeth, and 16 stomachs for processing.
Streaming? That'd have to be via 64 teats. Smart pins must be for modulated milk
And now we have LUTs for colors... colored milk!...
One item was a bill of activities. At one point, we had something like 10 people attacking a spreadsheet. It got done in like an hour. That same amount of intercommunication would have turned that into a couple days using email and file drops, or just slinging stuff back and forth.
So the on-going joke at work here is that if a component isn't quite up to standard we have been saying ... "New COW"
... So along similar lines with regard to the question at hand in this thread, some of us might be tempted to say "New COW" if our expectations are less than what we had expected with the "Chip formerly called P2" , "Chip which shall not be named" , or whatever it is to be called.