I've thought about this, too, but lots of nets between all cogs, would take a lot of power. Between adjacent cogs, maybe not so much - if they actually wound up adjacent in the layout.
Shame, it's just a gnats whisker too big to fit on a module with a pin out like so :
Let just call it the Circus chip. The hub can be renamed "The Big Top" and the cogs can be "Trailers" occupied by short, fat programs called "Carnies". When multitasking is on, they become "Fun Houses". All the new pipelined math circuits can be collectively called "Midway". The 512-long program memory can be featured in the "Freak Show". Wait, wait, that's all politically incorrect, nowadays. We need a bold, new name that commands reverence: "Sustainable Diversity Cluster". It thrives in an "Eco System" of mobile apps and exceptionally neutral people who were born about about five minutes ago.
Let just call it the Circus chip. The hub can be renamed "The Big Top" and the cogs can be "Trailers" occupied by short, fat programs called "Carnies". When multitasking is on, they become "Fun Houses". All the new pipelined math circuits can be collectively called "Midway". The 512-long program memory can be featured in the "Freak Show". Wait, wait, that's all politically incorrect, nowadays. We need a bold, new name that commands reverence: "Sustainable Diversity Cluster". It thrives in an "Eco System" of mobile apps and exceptionally neutral people who were born about about five minutes ago.
I absolutely agree, Core(s) it should be!
It will then come up in search entries and be compared with other multi-core chips. We are seeing numerous ARM multi-cores appearing. Microcontrollers will no doubt follow, and then Parallax is clearly in the lead. And you will get a lot of free press because they understand what cores is.
Next, I think the cog memory should be Core RAM Registers/Memory and described as Private Core RAM Memory 2KB (496 x 32bit longs with 128bit wide access to Common Memory).
And lastly, the hub memory should be described as Common RAM Memory 512KB (byte/word/long/quadlong), shared between all Cores, and usable as expanded instruction and/or memory space.
We have to ditch our favourite words to embrace the market that should be able to understand these simpler definitions.
Cluso is right on the MONEY here.
You NEED to use common terms. A big reason for that is SEARCHING. Your custom descriptors cause you to fail to come up, or rank low, in a lot of common search terms.
Like others here I have found it frustrating to find the Prop on retail sites when I KNEW they were there. All because you don't show up in the search functions.
Example: Go to mouser and enter multi-core in the search. You get 33 results, not one Prop reference. Yet they sell them! You have to search for 'propeller' to find them.
Why? Because searching is automated based on database field and pattern matching.
Go to digikey and search 'core'. Select Integrated Circuits (ICs) > Embedded - Microcontrollers. Then select '32-Bit 8-Core' in the Core Size filter. You get 31 XMOS chips, no propeller! Yet they too sell them!
You are losing out on potential sales because your terms don't match industry standards and you are getting lost in searches.
You NEED to use common terms. A big reason for that is SEARCHING. Your custom descriptors cause you to fail to come up, or rank low, in a lot of common search terms.
Like others here I have found it frustrating to find the Prop on retail sites when I KNEW they were there. All because you don't show up in the search functions.
Example: Go to mouser and enter multi-core in the search. You get 33 results, not one Prop reference. Yet they sell them! You have to search for 'propeller' to find them.
Why? Because searching is automated based on database field and pattern matching.
Go to digikey and search 'core'. Select Integrated Circuits (ICs) > Embedded - Microcontrollers. Then select '32-Bit 8-Core' in the Core Size filter. You get 31 XMOS chips, no propeller! Yet they too sell them!
You are losing out on potential sales because your terms don't match industry standards and you are getting lost in searches.
I'm not sure about that. Is "Snow Leopard" more than "Maverick" ??
People love to compare for whatever reason. The following may not be a fair analogy but even BMW has its "3 series", "5 series", an so on which imply superior performance. Without a clear orientation you need additional support information to compare or people will just look at the price and guess.
I have been saying this for a long time now, see my recurring story about the ELFA distributor a few posts back in this thread today.
Kerry S has put it much better than I did.
I believe this is keeping the Propeller off the radar for many engineers.
I think you are right, and I read your post about ELFA. I've noticed the same thing when I've bothered to search on Digi-Key or Mouser. It always makes me wince when I see stuff like that. But, you know, I always imagined there wasn't much we could do about it, because Parallax doesn't "pay to play" all the industry marketing games, and we are not favored by the system, so to speak.
There was some multi-core processing conference that we were polled to speak at, recently. They asked for submissions on ideas for talks, so we put something together and sent it to them, figuring it wouldn't hurt to speak about the Propeller at some conference. Well, they got back to us and said that the thing was filling up very quickly with 700 registered attendees and speaking spots were filling up fast, too, but there was some "good news": they still had some talks open for 'very reasonable rates'. Huh??? Ken got the scoop from them. They wanted $2500 from us for a 20 minute talk. Whoever paid them $10k got to give the keynote address. Woopie! What a waste of time that whole thing was to even consider. I wonder if many of those attendees realized they were listening to speakers who BOUGHT their way in.
Had a quick look
COGRUN,COGINIT?
Maybe a TASKID instruction
Looks good!
Brian
Edit: PASSCNT?
Oh, I was worried for a second, there. I was wondering how I could forget COGRUN, COGNEW, etc. The answer is, they all fall under the SYSOP umbrella. That will be used for the math circuits, too. Thanks for noticing, though.
Hundreds of millions are spent on BMW's to establish their position in the marketplace. BMW is ubiquitous, non drivers know the difference. Highly paid research teams market the "3 series" to stimulate your emotions. The name is a number, and there are exceptions. Apple has spent tons also to market their software revisions" The only number that is used is generic "X". The flavor of the new revision is never about the number, it always includes a new graphic, new features but recent history always includes an animal association. Companies in marketing spend billions on research in trials, side by side tests, group studies. So, "people love to compare" for very good reason. It is not arbitrary or wasteful.
But I'm sure that's not the case with distributors. They want to sell stuff as much as you do, there is no gain for them in making things hard to find. I imagine they don't know what a Propeller is either and it just gets lost somewhere "on the shelf at the back".
Heater, the number alone does nothing much, the marketing behind 007 over decades does lot. Thousands of years of 666 references does a lot. It is NOT the number behind those references you make, but the history of associations.
But I'm sure that's not the case with distributors. They want to sell stuff as much as you do, there is no gain for them in making things hard to find. I imagine they don't know what a Propeller is either and it just gets lost somewhere "on the shelf at the back".
I think they have some grounded interests in selling our products, but as it gets into their system, it gets lost for the reasons cited.
Oh, I was worried for a second, there. I was wondering how I could forget COGRUN, COGNEW, etc. The answer is, they all fall under the SYSOP umbrella. That will be used for the math circuits, too. Thanks for noticing, though.
Wow! Look at how few times LOCKSET and LOCKCLR were used: 3 and 7, only! I wonder if they were even needed, or used just because their existence suggested they were needed.
I've never used these, myself, though I put them in the design because I could imagine they might be necessary, as some others now adamantly suppose. They've hardly been used on the OBEX, though. I would like to just get rid of them. There are other ways to do such things and, ideally, you wouldn't be writing code that required them. I think they might just be faking people out into thinking they are useful. In practice, I've never found them even remotely necessary. I SAY JETISSON THOSE THINGS! Anyone care, who's actually used them? Please speak up!
There was some multi-core processing conference that we were polled to speat at, recently. They asked for submissions on ideas for talks, so we put something together and sent it to them, figuring it wouldn't hurt to speak about the Propeller at some conference. Well, they got back to us and said that the thing was filling up very quickly with 700 registered attendees and speaking spots were filling up fast, too, but there was some "good news": they still had some talks open for 'very reasonable rates'. Huh??? Ken got the scoop from them. They wanted $2500 from us for a 20 minute talk. Whoever paid them $10k got to give the keynote address. Woopie! What a waste of time that whole thing was to even consider. I wonder if many of those attendees realized they were listening to speakers who BOUGHT their way in.
Pay to play is lame BUT sometimes you have to look at it as part of the marketing budget IF you can leverage it.
For example if that conference was one that was well known then: Were they doing live streaming? Were they posting the speakers on-line (video)? Would they let you film the talk and post on your web site? IF they had a name people searched for and IF they would either post the video or let you post THEN that $2500.00 might have been a good investment. You could have put pages up on Parallax.com linking to it (with key words in the page to hit the search engines) and perhaps got others with Prop related sites to do the same (to raise your ranking). NOW your talk has the potential to reach a lot of people (via keyword search) AND you start to get your name out there with people who might not otherwise have heard of, or really thought about, Parallax and the Propeller as an option. You HAVE to have an almost evangelical visionary message that really grips people. Needs to tie in to what they do while gently opening their eyes to new concepts and possibilities. I have no doubt that the Propeller could be the source of such a speech/white paper! I see guys, and gals, walking out with the eyes wide going 'wow! I never thought about THAT before!'
I had the pleasure of working with a small automation manufacturer in the way back. They were a no name UNTIL their president started hitting all of the trade conferences, started writing compelling articles for the trade magazines, and created a name for both the company and himself in the industry. That was all before the internet and <GOOGLE>!
Oh yeah, I'm already aware of this and I've tried to change it. There are two reasons for not being categorized correctly, and only one of them relates to our terminology being a bit unique. In these big corporations - and they're mostly the same - buyers handle similar product lines. Prior to making our first Propeller, we were a supplier of embedded tools - for the SX, the BASIC Stamp, and even the PIC. This is a product category in a distributor's view, assigned to a buyer who handles the product category. Once you're an embedded tools provider, and you introduce your first processor to 70 distributors, a bunch of them are going to trip up the listing no matter how much you manage them. While we now need to be represented by the internal semiconductor buyer, we've always been handled by "Joe, who does the Parallax stuff - the embedded tools". This is partly why ELFA has this goofy listing.
From a distance it's situations like this that make engineers think marketing people are just plain stupid.
And I've tried to fix this problem with ELFA. I've concluded that it's not in your benefit to have me spend my time on that particular case. I can only afford to give it another inquiry by Skype to our buyer. I'll do that, but I'm not putting 20 hours into resolving this problem or flying to Sweden anytime soon.
To address the distribution and listing issue, Parallax's strategy moving forward with P2 is to focus on three distributors exclusively (Mouser, Digi-Key and RS Components). All of these distributors have capable buyers and we have available bandwidth to communicate effectively with them. While you'll be able to buy all of our products at any distributor, it's these three who will be responsible for the introduction and communication of our new product to the market place, with us. They come to our office regularly and they understand what's going on with our product plan, even though it's changing. They lurk here, too, but they don't post anything.
- - - -
Regarding terminology - and I see other threads have been started for it - sometimes being "unique" also means being "alone". Not only have we been standing in the corner of the playground while others played games together, we were also speaking our own language. Why make it difficult on ourselves? This is why I advocate we strongly use the word "core" and "shared RAM" and that we come to the playground speaking both our own language Spin and what everybody else wants to speak: C. I can prove the effectiveness of being bilingual with the recent Propeller C program. For the first time, educators are interested in looking at Spin because we support C! Prior to having C support they really didnt want to know about Spin. It we release Spin only again, we'll deserve every benefit of being unique.
Two tasks are good, but four are really plush. Actually, I've found that three tasks are usually optimal. Four is overkill, but two are too few. Three is just odd, so might as well go with four. In my usage, there is always a main task and a few helper tasks which run in loops and don't do any calls. They are like adding magic peripherals that never could have been anticipated in the chip hardware design. That's what I think of extra tasks: they are peripherals.
Multitasking is very simple. It's just a matter of mux'ing a few sets of Z/C/PC. That's all it needs to be.
OK well said. Yes I can imagine ELFA is a bit out of the way for you.
I'll set out to fix that Mouser and Digi-Key wonky listing today. That'll be easy to do by comparison to the ELFA problem, but I'll give it another shot.
I think you are right, and I read your post about ELFA. I've noticed the same thing when I've bothered to search on Digi-Key or Mouser. It always makes me wince when I see stuff like that. But, you know, I always imagined there wasn't much we could do about it, because Parallax doesn't "pay to play" all the industry marketing games, and we are not favored by the system, so to speak.
Quite true. I appreciate that you talk about this with complete freedom, as nobody inside the industry talks clearly about these "practices".
Everyone can easily check it: write "xilinx" at mouser's site. OMG, it turns out to be Al.... !
I usually check parts on several distributors. Its quite interesting to see that each distributor has its "friends".
(PS: Another day we can also talk about EDA tools and semiconductor industry, or maybe not. No, Its better not to talk about this if we want propeller to be done at smaller geometry than 180nm. I hope sharks not to be "smelling" around here)
Comments
Thanks Chip.
Let just call it the Circus chip. The hub can be renamed "The Big Top" and the cogs can be "Trailers" occupied by short, fat programs called "Carnies". When multitasking is on, they become "Fun Houses". All the new pipelined math circuits can be collectively called "Midway". The 512-long program memory can be featured in the "Freak Show". Wait, wait, that's all politically incorrect, nowadays. We need a bold, new name that commands reverence: "Sustainable Diversity Cluster". It thrives in an "Eco System" of mobile apps and exceptionally neutral people who were born about about five minutes ago.
...with Smart Pins(tm)!!
ZCDS (for D column: W=write, M=modify, R=read, L=read/immediate) ---------------------------------------------------------------------------------------------------------------------- ZCWS 0000000 ZC 1 CCCC DDDDDDDDD SSSSSSSSS RDBYTE D,PTRA/PTRB (waits for hub) ZCWS 0000001 ZC 1 CCCC DDDDDDDDD SSSSSSSSS RDBYTEC D,PTRA/PTRB (waits for hub if dcache miss) ZCWS 0000010 ZC 1 CCCC DDDDDDDDD SSSSSSSSS RDWORD D,PTRA/PTRB (waits for hub) ZCWS 0000011 ZC 1 CCCC DDDDDDDDD SSSSSSSSS RDWORDC D,PTRA/PTRB (waits for hub if dcache miss) ZCWS 0000100 ZC 1 CCCC DDDDDDDDD SSSSSSSSS RDLONG D,PTRA/PTRB (waits for hub) ZCWS 0000101 ZC 1 CCCC DDDDDDDDD SSSSSSSSS RDLONGC D,PTRA/PTRB (waits for hub if dcache miss) ZCWS 0000110 ZC 1 CCCC DDDDDDDDD SSSSSSSSS RDQUAD D,PTRA/PTRB (waits for hub) ZCWS 0000111 ZC 1 CCCC DDDDDDDDD SSSSSSSSS SYSOP D,# (waits for hub, # determines four write-long enables) ZCMS 0001000 ZC I CCCC DDDDDDDDD SSSSSSSSS ISOB D,S/# ZCMS 0001001 ZC I CCCC DDDDDDDDD SSSSSSSSS NOTB D,S/# ZCMS 0001010 ZC I CCCC DDDDDDDDD SSSSSSSSS CLRB D,S/# ZCMS 0001011 ZC I CCCC DDDDDDDDD SSSSSSSSS SETB D,S/# ZCMS 0001100 ZC I CCCC DDDDDDDDD SSSSSSSSS SETBC D,S/# ZCMS 0001101 ZC I CCCC DDDDDDDDD SSSSSSSSS SETBNC D,S/# ZCMS 0001110 ZC I CCCC DDDDDDDDD SSSSSSSSS SETBZ D,S/# ZCMS 0001111 ZC I CCCC DDDDDDDDD SSSSSSSSS SETBNZ D,S/# ZCMS 0010000 ZC I CCCC DDDDDDDDD SSSSSSSSS ROR D,S/# ZCMS 0010001 ZC I CCCC DDDDDDDDD SSSSSSSSS ROL D,S/# ZCMS 0010010 ZC I CCCC DDDDDDDDD SSSSSSSSS SHR D,S/# ZCMS 0010011 ZC I CCCC DDDDDDDDD SSSSSSSSS SHL D,S/# ZCMS 0010100 ZC I CCCC DDDDDDDDD SSSSSSSSS RCR D,S/# ZCMS 0010101 ZC I CCCC DDDDDDDDD SSSSSSSSS RCL D,S/# ZCMS 0010110 ZC I CCCC DDDDDDDDD SSSSSSSSS SAR D,S/# ZCMS 0010111 ZC I CCCC DDDDDDDDD SSSSSSSSS REV D,S/# ZCMS 0011000 ZC I CCCC DDDDDDDDD SSSSSSSSS ANDN D,S/# ZCMS 0011001 ZC I CCCC DDDDDDDDD SSSSSSSSS AND D,S/# ZCMS 0011010 ZC I CCCC DDDDDDDDD SSSSSSSSS OR D,S/# ZCMS 0011011 ZC I CCCC DDDDDDDDD SSSSSSSSS XOR D,S/# ZCMS 0011100 ZC I CCCC DDDDDDDDD SSSSSSSSS MUXC D,S/# ZCMS 0011101 ZC I CCCC DDDDDDDDD SSSSSSSSS MUXNC D,S/# ZCMS 0011110 ZC I CCCC DDDDDDDDD SSSSSSSSS MUXZ D,S/# ZCMS 0011111 ZC I CCCC DDDDDDDDD SSSSSSSSS MUXNZ D,S/# ZCMS 0100000 ZC I CCCC DDDDDDDDD SSSSSSSSS ADD D,S/# ZCMS 0100001 ZC I CCCC DDDDDDDDD SSSSSSSSS SUB D,S/# ZCMS 0100010 ZC I CCCC DDDDDDDDD SSSSSSSSS ADDX D,S/# ZCMS 0100011 ZC I CCCC DDDDDDDDD SSSSSSSSS SUBX D,S/# ZCMS 0100100 ZC I CCCC DDDDDDDDD SSSSSSSSS SUMC D,S/# ZCMS 0100101 ZC I CCCC DDDDDDDDD SSSSSSSSS SUMNC D,S/# ZCMS 0100110 ZC I CCCC DDDDDDDDD SSSSSSSSS SUMZ D,S/# ZCMS 0100111 ZC I CCCC DDDDDDDDD SSSSSSSSS SUMNZ D,S/# ZCWS 0101000 ZC I CCCC DDDDDDDDD SSSSSSSSS MOV D,S/# ZCWS 0101001 ZC I CCCC DDDDDDDDD SSSSSSSSS NOT D,S/# ZCWS 0101010 ZC I CCCC DDDDDDDDD SSSSSSSSS ABS D,S/# ZCWS 0101011 ZC I CCCC DDDDDDDDD SSSSSSSSS NEG D,S/# ZCWS 0101100 ZC I CCCC DDDDDDDDD SSSSSSSSS NEGC D,S/# ZCWS 0101101 ZC I CCCC DDDDDDDDD SSSSSSSSS NEGNC D,S/# ZCWS 0101110 ZC I CCCC DDDDDDDDD SSSSSSSSS NEGZ D,S/# ZCWS 0101111 ZC I CCCC DDDDDDDDD SSSSSSSSS NEGNZ D,S/# ZCMS 0110000 ZC I CCCC DDDDDDDDD SSSSSSSSS SETS D,S/# ZCMS 0110001 ZC I CCCC DDDDDDDDD SSSSSSSSS SETD D,S/# ZCMS 0110010 ZC I CCCC DDDDDDDDD SSSSSSSSS SETCOND D,S/# ZCMS 0110011 ZC I CCCC DDDDDDDDD SSSSSSSSS SETINST D,S/# ZCMS 0110100 ZC I CCCC DDDDDDDDD SSSSSSSSS MIN D,S/# ZCMS 0110101 ZC I CCCC DDDDDDDDD SSSSSSSSS MAX D,S/# ZCMS 0110110 ZC I CCCC DDDDDDDDD SSSSSSSSS MINS D,S/# ZCMS 0110111 ZC I CCCC DDDDDDDDD SSSSSSSSS MAXS D,S/# ZCMS 0111000 ZC I CCCC DDDDDDDDD SSSSSSSSS MUL D,S/# ZCMS 0111001 ZC I CCCC DDDDDDDDD SSSSSSSSS MULS D,S/# ZCMS 0111010 ZC I CCCC DDDDDDDDD SSSSSSSSS JMPSW D,S/@ (save Z/C/PC into D, load Z/C/PC from S/#) ZCWS 0111011 ZC I CCCC DDDDDDDDD SSSSSSSSS TOPBIT D,S/# ZCWS 0111100 ZC I CCCC DDDDDDDDD SSSSSSSSS DECOD5 D,S/# ZCWS 0111101 ZC I CCCC DDDDDDDDD SSSSSSSSS ESWAP8 D,S/# ZCWS 0111110 ZC I CCCC DDDDDDDDD SSSSSSSSS SPLITW D,S/# ZCWS 0111111 ZC I CCCC DDDDDDDDD SSSSSSSSS MERGEW D,S/# ZCRS 1000000 ZC I CCCC DDDDDDDDD SSSSSSSSS TESTB D,S/# ZCRS 1000001 ZC I CCCC DDDDDDDDD SSSSSSSSS TESTN D,S/# ZCRS 1000010 ZC I CCCC DDDDDDDDD SSSSSSSSS TEST D,S/# ZCRS 1000011 ZC I CCCC DDDDDDDDD SSSSSSSSS CMP D,S/# ZCRS 1000100 ZC I CCCC DDDDDDDDD SSSSSSSSS CMPX D,S/# ZCRS 1000101 ZC I CCCC DDDDDDDDD SSSSSSSSS CMPS D,S/# ZCRS 1000110 ZC I CCCC DDDDDDDDD SSSSSSSSS CMPSX D,S/# ZCRS 1000111 ZC I CCCC DDDDDDDDD SSSSSSSSS CMPR D,S/# ZCMS 1001000 ZC I CCCC DDDDDDDDD SSSSSSSSS INCMOD D,S/# ZCMS 1001001 ZC I CCCC DDDDDDDDD SSSSSSSSS DECMOD D,S/# ZCMS 1001010 ZC I CCCC DDDDDDDDD SSSSSSSSS CMPSUB D,S/# ZCMS 1001011 ZC I CCCC DDDDDDDDD SSSSSSSSS SUBR D,S/# ZCWS 1001100 ZC I CCCC DDDDDDDDD SSSSSSSSS MSGINA D,S/# (waits up to 64 clocks to receive %010_data32 message on INA, C=timeout) ZCWS 1001101 ZC I CCCC DDDDDDDDD SSSSSSSSS MSGINB D,S/# (waits up to 64 clocks to receive %010_data32 message on INB, C=timeout) ZCMS 1001110 ZC I CCCC DDDDDDDDD SSSSSSSSS ROLNIB D,S/# ZCMS 1001111 ZC I CCCC DDDDDDDDD SSSSSSSSS ROLBYTE D,S/# ZCMS 1001000 ZC I CCCC DDDDDDDDD SSSSSSSSS DJZ D,S/@ ZCMS 1001001 ZC I CCCC DDDDDDDDD SSSSSSSSS DJNZ D,S/@ ZCRS 1001010 ZC I CCCC DDDDDDDDD SSSSSSSSS TJZ D,S/@ ZCRS 1001011 ZC I CCCC DDDDDDDDD SSSSSSSSS TJNZ D,S/@ ZCMS 1001100 ZC I CCCC DDDDDDDDD SSSSSSSSS WAITCNT D,S/# --WS 1010xnn n0 I CCCC DDDDDDDDD SSSSSSSSS GETNIB D,S/#,#0..7 --MS 1010xnn n1 I CCCC DDDDDDDDD SSSSSSSSS SETNIB D,S/#,#0..7 --WS 1011xxn n0 I CCCC DDDDDDDDD SSSSSSSSS GETBYTE D,S/#,#0..3 --MS 1011xxn n1 I CCCC DDDDDDDDD SSSSSSSSS SETBYTE D,S/#,#0..3 --LS 1100000 0L 1 CCCC DDDDDDDDD SSSSSSSSS WRBYTE D/#,PTRA/PTRB (waits for hub) --LS 1100000 1L 1 CCCC DDDDDDDDD SSSSSSSSS WRWORD D/#,PTRA/PTRB (waits for hub) --LS 1100001 0L 1 CCCC DDDDDDDDD SSSSSSSSS WRLONG D/#,PTRA/PTRB (waits for hub) --LS 1100001 1L 1 CCCC DDDDDDDDD SSSSSSSSS WRQUAD D/#,PTRA/PTRB (waits for hub, zero-extends #) --LS 1100010 0L I CCCC DDDDDDDDD SSSSSSSSS WAITPAE D/#,S/# (waits for INA) --LS 1100010 1L I CCCC DDDDDDDDD SSSSSSSSS WAITPAN D/#,S/# (waits for INA) --LS 1100011 0L I CCCC DDDDDDDDD SSSSSSSSS WAITPBE D/#,S/# (waits for INB) --LS 1100011 1L I CCCC DDDDDDDDD SSSSSSSSS WAITPBN D/#,S/# (waits for INB) --LS 1100100 0L I CCCC DDDDDDDDD SSSSSSSSS MSGDIRA D/#,S/# --LS 1100100 1L I CCCC DDDDDDDDD SSSSSSSSS MSGDIRB D/#,S/# --LS 1100101 0L I CCCC DDDDDDDDD SSSSSSSSS MSGOUTA D/#,S/# --LS 1100101 1L I CCCC DDDDDDDDD SSSSSSSSS MSGOUTB D/#,S/# --LS 1100110 0L I CCCC DDDDDDDDD SSSSSSSSS JP D/#,S/@ --LS 1100110 1L I CCCC DDDDDDDDD SSSSSSSSS JPD D/#,S/@ --LS 1100111 0L I CCCC DDDDDDDDD SSSSSSSSS JNP D/#,S/@ --LS 1100111 1L I CCCC DDDDDDDDD SSSSSSSSS JNPD D/#,S/@ --LS 1101000 0L I CCCC DDDDDDDDD SSSSSSSSS SETINDS D/#,S/# --LS 1101000 1L I CCCC DDDDDDDDD SSSSSSSSS ADJINDS D/#,S/# --LS 1101001 0L I CCCC DDDDDDDDD SSSSSSSSS WAITVID D/#,S/# (waits for vid) --LS 1101001 1L I CCCC DDDDDDDDD SSSSSSSSS REP D/#,S/# ---- 11100xx x0 n nnnn nnnnnnnnn nnnnnnnnn AUGS #23bits (appends n to upper bits of next immediate S in same task) ---- 11100xx x1 n nnnn nnnnnnnnn nnnnnnnnn AUGD #23bits (appends n to upper bits of next immediate D in same task) ---- 11101xx 00 0 CCCC 0 nnnnnnnnnnnnnnnnn JMP #abs ---- 11101xx 00 0 CCCC 1 nnnnnnnnnnnnnnnnn JMP @rel ---- 11101xx 00 1 CCCC 0 nnnnnnnnnnnnnnnnn CALL #abs ---- 11101xx 00 1 CCCC 1 nnnnnnnnnnnnnnnnn CALL @rel ---- 11101xx 01 0 CCCC 0 nnnnnnnnnnnnnnnnn CALLA #abs ---- 11101xx 01 0 CCCC 1 nnnnnnnnnnnnnnnnn CALLA @rel ---- 11101xx 01 1 CCCC 0 nnnnnnnnnnnnnnnnn CALLB #abs ---- 11101xx 01 1 CCCC 1 nnnnnnnnnnnnnnnnn CALLB @rel ---- wr 11101xx 10 0 CCCC 0 nnnnnnnnnnnnnnnnn LINK #abs (jump and write {%0000000000000,Z,C,PC[16:0]} to $1EF) ---- wr 11101xx 10 0 CCCC 1 nnnnnnnnnnnnnnnnn LINK @rel (jump and write {%0000000000000,Z,C,PC[16:0]} to $1EF) ---- wr 11101xx 10 1 CCCC 0 nnnnnnnnnnnnnnnnn LOCINST #abs (write 17-bit instruction address to $1EF) ---- wr 11101xx 10 1 CCCC 1 nnnnnnnnnnnnnnnnn LOCINST @rel (write 17-bit instruction address to $1EF) ---- 11101xx 11 0 CCCC 0 nnnnnnnnnnnnnnnnn LOCPTRA #abs (write 17-bit instruction address << 2 to PTRA) ---- 11101xx 11 0 CCCC 1 nnnnnnnnnnnnnnnnn LOCPTRA @rel (write 17-bit instruction address << 2 to PTRA) ---- 11101xx 11 1 CCCC 0 nnnnnnnnnnnnnnnnn LOCPTRB #abs (write 17-bit instruction address << 2 to PTRB) ---- 11101xx 11 1 CCCC 1 nnnnnnnnnnnnnnnnn LOCPTRB @rel (write 17-bit instruction address << 2 to PTRB) --L- 1111xxx 00 L CCCC DDDDDDDDD xxx000000 OFFP D/# (pin operations that affect DIR and OUT simultaneously) --L- 1111xxx 00 L CCCC DDDDDDDDD xxx000001 NOTP D/# --L- 1111xxx 00 L CCCC DDDDDDDDD xxx000010 CLRP D/# --L- 1111xxx 00 L CCCC DDDDDDDDD xxx000011 SETP D/# --L- 1111xxx 00 L CCCC DDDDDDDDD xxx000100 SETPC D/# --L- 1111xxx 00 L CCCC DDDDDDDDD xxx000101 SETPNC D/# --L- 1111xxx 00 L CCCC DDDDDDDDD xxx000110 SETPZ D/# --L- 1111xxx 00 L CCCC DDDDDDDDD xxx000111 SETPNZ D/# ZCL- 1111xxx ZC L CCCC DDDDDDDDD xxx001000 GETP D/# (pin into !Z/C via WZ/WC) ZCL- 1111xxx ZC L CCCC DDDDDDDDD xxx001001 GETNP D/# (pin into Z/!C via WZ/WC) --L- 1111xxx 00 L CCCC DDDDDDDDD xxx001010 WAITPR D/# (waits for pos edge on pin) --L- 1111xxx 00 L CCCC DDDDDDDDD xxx001011 WAITPF D/# (waits for neg edge on pin) --L- 1111xxx 00 L CCCC DDDDDDDDD xxx001100 WAITPX D/# (waits for any edge on pin) --L- 1111xxx 00 L CCCC DDDDDDDDD xxx001101 SETTASK D/# --L- 1111xxx 00 L CCCC DDDDDDDDD xxx001110 SETPTRA D/# --L- 1111xxx 00 L CCCC DDDDDDDDD xxx001111 SETPTRB D/# --L- 1111xxx 00 L CCCC DDDDDDDDD xxx010000 SETINDA D/# (set INDA/INDB, INDA/INDB increment on each use) --L- 1111xxx 00 L CCCC DDDDDDDDD xxx010001 SETINDB D/# --L- 1111xxx 00 L CCCC DDDDDDDDD xxx010010 ADJINDA D/# (adjust INDA/INDB by adding value) --L- 1111xxx 00 L CCCC DDDDDDDDD xxx010011 ADJINDB D/# --L- 1111xxx 00 L CCCC DDDDDDDDD xxx010100 JMPT0 D/# --L- 1111xxx 00 L CCCC DDDDDDDDD xxx010101 JMPT1 D/# --L- 1111xxx 00 L CCCC DDDDDDDDD xxx010110 JMPT2 D/# --L- 1111xxx 00 L CCCC DDDDDDDDD xxx010111 JMPT3 D/# --L- 1111xxx 00 L CCCC DDDDDDDDD xxx011000 WAIT D/# (waits 1+ clocks, 0 same as 1) --L- 1111xxx 00 L CCCC DDDDDDDDD xxx011001 SETVID D/# --L- 1111xxx 00 L CCCC DDDDDDDDD xxx011010 PUSH D/# (pushes into 4-level stack) Z-W- 1111xxx Z0 0 CCCC DDDDDDDDD xxx100000 GETPTRA D Z-W- 1111xxx Z0 0 CCCC DDDDDDDDD xxx100001 GETPTRB D Z-W- 1111xxx Z0 0 CCCC DDDDDDDDD xxx100010 POP D (pops from 4-level stack) --R- 1111xxx 00 0 CCCC DDDDDDDDD xxx110000 LOCPTRA D (PTRA = {D[16:0],%00}) --R- 1111xxx 00 0 CCCC DDDDDDDDD xxx110001 LOCPTRB D (PTRB = {D[16:0],%00}) --R- 1111xxx ZC 0 CCCC DDDDDDDDD xxx110010 JMP D (D[18:17] into Z/C via WZ/WC for JMP..LINK D) --R- 1111xxx ZC 0 CCCC DDDDDDDDD xxx110011 CALL D (call using 4-level stack) --R- 1111xxx ZC 0 CCCC DDDDDDDDD xxx110100 CALLA D (call using PTRA) --R- 1111xxx ZC 0 CCCC DDDDDDDDD xxx110101 CALLB D (call using PTRB) --R- wr 1111xxx ZC 0 CCCC DDDDDDDDD xxx110110 LINK D (jump and write {%0000000000000,Z,C,PC[16:0]} to $1EF) ZC-- 1111xxx ZC 0 CCCC xxxxxxxxx xxx11100x RET (return using 4-level stack) ZC-- 1111xxx ZC 0 CCCC xxxxxxxxx xxx111010 RETA (return using PTRA) ZC-- 1111xxx ZC 0 CCCC xxxxxxxxx xxx111011 RETB (return using PTRB) ---- 1111xxx 00 0 CCCC xxxxxxxxx xxx111100 DCACHEX (invalidate data cache) ---- 1111xxx 00 0 CCCC xxxxxxxxx xxx111101 ICACHEX (invalidate instruction cache) ---- 0000000 00 0 0000 000000000 000000000 NOP PUSHA D/# (alias for WRLONG D/#,PTRA++) PUSHB D/# (alias for WRLONG D/#,PTRB++) POPA (alias for RDLONG D,--PTRA) POPB (alias for RDLONG D,--PTRB) x = don't care, use 0 ---------------------------------------------------------------------------------------------------------------------I think this will provide us with a nice programming platform, while keeping pretty skinny. This is like a non-fat version of Prop2.
Does anyone see any pitfalls, or something that needs to be in there, but isn't? Or something that IS in there that shouldn't be?
Great, I can get on with that 68000 emulator now and get it into the correct size!
Cluso is right on the MONEY here.
You NEED to use common terms. A big reason for that is SEARCHING. Your custom descriptors cause you to fail to come up, or rank low, in a lot of common search terms.
Like others here I have found it frustrating to find the Prop on retail sites when I KNEW they were there. All because you don't show up in the search functions.
Example: Go to mouser and enter multi-core in the search. You get 33 results, not one Prop reference. Yet they sell them! You have to search for 'propeller' to find them.
Why? Because searching is automated based on database field and pattern matching.
Go to digikey and search 'core'. Select Integrated Circuits (ICs) > Embedded - Microcontrollers. Then select '32-Bit 8-Core' in the Core Size filter. You get 31 XMOS chips, no propeller! Yet they too sell them!
You are losing out on potential sales because your terms don't match industry standards and you are getting lost in searches.
Ken needs to see this.
Ken really needs to see that.
I have been saying this for a long time now, see my recurring story about the ELFA distributor a few posts back in this thread today.
Kerry S has put it much better than I did.
I believe this is keeping the Propeller off the radar for many engineers.
Had a quick look
COGRUN,COGINIT?
Maybe a TASKID instruction
Looks good!
Brian
Edit: PASSCNT?
I'm not sure about that. Is "Snow Leopard" more than "Maverick" ??
People love to compare for whatever reason. The following may not be a fair analogy but even BMW has its "3 series", "5 series", an so on which imply superior performance. Without a clear orientation you need additional support information to compare or people will just look at the price and guess.
Alex
I think you are right, and I read your post about ELFA. I've noticed the same thing when I've bothered to search on Digi-Key or Mouser. It always makes me wince when I see stuff like that. But, you know, I always imagined there wasn't much we could do about it, because Parallax doesn't "pay to play" all the industry marketing games, and we are not favored by the system, so to speak.
There was some multi-core processing conference that we were polled to speak at, recently. They asked for submissions on ideas for talks, so we put something together and sent it to them, figuring it wouldn't hurt to speak about the Propeller at some conference. Well, they got back to us and said that the thing was filling up very quickly with 700 registered attendees and speaking spots were filling up fast, too, but there was some "good news": they still had some talks open for 'very reasonable rates'. Huh??? Ken got the scoop from them. They wanted $2500 from us for a 20 minute talk. Whoever paid them $10k got to give the keynote address. Woopie! What a waste of time that whole thing was to even consider. I wonder if many of those attendees realized they were listening to speakers who BOUGHT their way in.
Added: I emailed Ken the link the Kerry S' post.
7, 007, 42, 69, 666, 1729,
Those numbers don't move you?
Back in the day we had the Motorola 6800, their 32 bit chip became the 68000.
Zilog had the Z80 and their 32 bit chip became the Z8000.
Something about all those zeros.
Oh, I was worried for a second, there. I was wondering how I could forget COGRUN, COGNEW, etc. The answer is, they all fall under the SYSOP umbrella. That will be used for the math circuits, too. Thanks for noticing, though.
Oh yeah, the conference circuit can be like that.
But I'm sure that's not the case with distributors. They want to sell stuff as much as you do, there is no gain for them in making things hard to find. I imagine they don't know what a Propeller is either and it just gets lost somewhere "on the shelf at the back".
I think they have some grounded interests in selling our products, but as it gets into their system, it gets lost for the reasons cited.
And what about PASSCNT?
Edit: and SETZC?
Chip,
Locks are absolutely necessary in some form to allow multiple tasks / threads to access shared resources safely.
Example: I allow multiple cogs to access the shared external memory on Morpheus, something I could not safely do without locks.
If locks are removed, many things will become impossible (or dangerous).
32 sounds great, if it does not take too much logic.
Pay to play is lame BUT sometimes you have to look at it as part of the marketing budget IF you can leverage it.
For example if that conference was one that was well known then: Were they doing live streaming? Were they posting the speakers on-line (video)? Would they let you film the talk and post on your web site? IF they had a name people searched for and IF they would either post the video or let you post THEN that $2500.00 might have been a good investment. You could have put pages up on Parallax.com linking to it (with key words in the page to hit the search engines) and perhaps got others with Prop related sites to do the same (to raise your ranking). NOW your talk has the potential to reach a lot of people (via keyword search) AND you start to get your name out there with people who might not otherwise have heard of, or really thought about, Parallax and the Propeller as an option. You HAVE to have an almost evangelical visionary message that really grips people. Needs to tie in to what they do while gently opening their eyes to new concepts and possibilities. I have no doubt that the Propeller could be the source of such a speech/white paper! I see guys, and gals, walking out with the eyes wide going 'wow! I never thought about THAT before!'
I had the pleasure of working with a small automation manufacturer in the way back. They were a no name UNTIL their president started hitting all of the trade conferences, started writing compelling articles for the trade magazines, and created a name for both the company and himself in the industry. That was all before the internet and <GOOGLE>!
Oh yeah, I'm already aware of this and I've tried to change it. There are two reasons for not being categorized correctly, and only one of them relates to our terminology being a bit unique. In these big corporations - and they're mostly the same - buyers handle similar product lines. Prior to making our first Propeller, we were a supplier of embedded tools - for the SX, the BASIC Stamp, and even the PIC. This is a product category in a distributor's view, assigned to a buyer who handles the product category. Once you're an embedded tools provider, and you introduce your first processor to 70 distributors, a bunch of them are going to trip up the listing no matter how much you manage them. While we now need to be represented by the internal semiconductor buyer, we've always been handled by "Joe, who does the Parallax stuff - the embedded tools". This is partly why ELFA has this goofy listing.
From a distance it's situations like this that make engineers think marketing people are just plain stupid.
And I've tried to fix this problem with ELFA. I've concluded that it's not in your benefit to have me spend my time on that particular case. I can only afford to give it another inquiry by Skype to our buyer. I'll do that, but I'm not putting 20 hours into resolving this problem or flying to Sweden anytime soon.
To address the distribution and listing issue, Parallax's strategy moving forward with P2 is to focus on three distributors exclusively (Mouser, Digi-Key and RS Components). All of these distributors have capable buyers and we have available bandwidth to communicate effectively with them. While you'll be able to buy all of our products at any distributor, it's these three who will be responsible for the introduction and communication of our new product to the market place, with us. They come to our office regularly and they understand what's going on with our product plan, even though it's changing. They lurk here, too, but they don't post anything.
- - - -
Regarding terminology - and I see other threads have been started for it - sometimes being "unique" also means being "alone". Not only have we been standing in the corner of the playground while others played games together, we were also speaking our own language. Why make it difficult on ourselves? This is why I advocate we strongly use the word "core" and "shared RAM" and that we come to the playground speaking both our own language Spin and what everybody else wants to speak: C. I can prove the effectiveness of being bilingual with the recent Propeller C program. For the first time, educators are interested in looking at Spin because we support C! Prior to having C support they really didnt want to know about Spin. It we release Spin only again, we'll deserve every benefit of being unique.
Ken Gracey
Motorola could have called their chip the 79360 or whatever random looking number but no it had to be 68000.
OK well said. Yes I can imagine ELFA is a bit out of the way for you.
Sounds like a winner to me.
One feature of the tasks we have on the P2 that I REALLY liked was the advanced debugging it made possible.
I'll set out to fix that Mouser and Digi-Key wonky listing today. That'll be easy to do by comparison to the ELFA problem, but I'll give it another shot.
Ken Gracey
Quite true. I appreciate that you talk about this with complete freedom, as nobody inside the industry talks clearly about these "practices".
Everyone can easily check it: write "xilinx" at mouser's site. OMG, it turns out to be Al.... !
I usually check parts on several distributors. Its quite interesting to see that each distributor has its "friends".
(PS: Another day we can also talk about EDA tools and semiconductor industry, or maybe not. No, Its better not to talk about this if we want propeller to be done at smaller geometry than 180nm. I hope sharks not to be "smelling" around here)
How about "i-Pin" (Intelligent Pin) ??
Alex