A .Net CPU

An anonymous reader writes "Windows for devices has an article about the .Net CPU. The chip is programmed with a subset of the CLR and runs the same software as the SPOT smart watches. Among other things, "[t]he computer module is implemented in the format of a 32-pin "DIP" (dual inline package) chip, allowing the module to conveniently plug into a standard 32-pin DIP socket. In addition, the ".netcpu CPU Module" integrates 4MB of nonvolatile Flash memory (interfaced via an SPI interface on the SoC). It also provides 24 general purpose digital I/O lines, which are multiplexed with other functions including 8 VTU ports, a USB port, two serial ports, and SPI and I2C interfaces." More information about the product can be found at the .netcpu company website."

Comments

Hmm, sounds familiar

[Anonymous Coward]

*cough* java chips *cough* *cough*

Re:Hmm, sounds familiar

[Anonymous Coward]

You should see a doctor about that cough.

Re:Hmm, sounds familiar

[Anonymous Coward]

JAVA IS DOGG SLOW YOU FCKING MORRON!

--
too many caps yelling
too many caps yadda yadda

choo-choo here comes the cum-train with a load of taco snot ... ok, this should be enough.

Re:Hmm, sounds familiar

[sosume]

Yeah riiiiight have you ever seriously looked at the spec of these 'java chips'? They are not as advanced as Sun may have you believe..

* No floating point 16-bit int instead of 32 bits.
* All types (byte, short, char, int and boolean) use 2 bytes,
though byte and short arrays use 1 byte per element.
* Only one-dimensional arrays (can use the index to simulate a 2-D array.)
* Single byte ASCII strings instead of two byte Unicode
* Only a single thread available, though a timer allows for
scheduling of multiple tasks. (Plus the VP objects run independently)
* No interfaces, though sub-classing of an abstract base class is allowed.
* A subset of the core libraries is available. (Remember also that
all linked classes must be downloaded with the program and fit into
the 32kb of memory.)
* No garbage collection. All objects created will last for the
duration of the program.

Compare that to this .NET chip....
* 384K of SRAM, single cycle access
* 27 MHz ARM7TDMI
* FBGA chip form
* ~450,000 instructions per second
* 4MB non volatile flash
* 1.8-volt core, 3.3-volt I/O
* 32768 Hz real-time clock
* 32-pin pinout, including 24 GPIO ports multiplexed with other functions (8 VTU ports, dual serial ports, SPI, and USB port)
* SPI and I2C interfaces

and its multithreaded, too

Re:Hmm, sounds familiar

[DasAlbatross]

Listing all the worst qualities of one thing and then listing all of the best points of another is not a valid comparison. Try comparying the variable sizes on one to the other, rather than comparing variable sizes to instruction speed.

Re:Hmm, sounds familiar

[DrXym]

But it depends what problem space the Java chips were trying to address. With those low specs it seems obvious they're meant to be embedded in simple devices, running simple apps who don't need unicode or other junk eating up the precious constraints or adding to the cost. Another clue was the name - picoJava.

Given that problem domain (and timeframe in which Java Chip was out), the cited restrictions seem perfectly reasonable.

Re:Hmm, sounds familiar

[maxwell demon]

Lisp machine.

Re:Hmm, sounds familiar

[Glock27]

The aj-100 doesn't seem to have most of the limitations you mention, in particular it has floating point and 32 bit ints.

Read about it and some other Java chips here.

Re:Hmm, sounds familiar

[escher]

27MHz? .NET in 27MHz?? Are they farking mad?

Well, lessee... 450,000 instructions/second? That would be 450KHz which is slower than my old PCjr.

Re:Hmm, sounds familiar

[sosume]

NOT really. I copied both specs from their vendors' websites

Re:Hmm, sounds familiar

[Anonymous Coward]

Well, lessee... 450,000 instructions/second? That would be 450KHz which is slower than my old PCjr.

Well... Those older processors took several clocks to perform each instruction. You could compare it to a 0.45MHz 486 (still very, very slow), but if you compare it to an older processor it'll be equivalent to 2 to 4MHz.

Re:Hmm, sounds familiar

[escher]

This is what I get for posting before the coffee has kicked in.

umm

[Anonymous Coward]

hi

yeah right..

[msh104]

so first you create a managed runtime...
to run on every cpu....
and then you write a cpu for .net?
doesn't really make sense.

Re:yeah right..

[borum]

I this case they made a small module with an ARM7 on it - running a .net runtime.
Did you even read the article :)

Anyway. There are some of this kind of modules around - some with java runtimes.

Atmel has a series of small footprint risc processors with flash, eeprom, sram and various pheripheral io. Easy to program, easy to use.
http://www.atmel.com/dyn/products/param_table.asp? family_id=607&OrderBy=part_no&Direction=AS C

There is also a 6 pad PIC processor http://www.microchip.com/stellent/idcplg?IdcServic e=SS_GET_PAGE&nodeId=2060

None of these runs .net or java as far as i know, but this would be a nice christmas project ;)

So

[Lady Griffin]

How long before every PC has one of these .Net Chips?

Seems scary

Re:So

[Tim C]

Why does it seem scary? What do you imagine the chip will do?

It's just a CPU for the .net CLR, that's all, in much the same way as Pentiums and Athlons, etc, are CPUs for x86 code. It's not going to prevent you from running Linux, or reach up and take control of your PC and/or spy on you for Bill.

Re:So

[Anonymous Coward]

I for one welcome our new .netcpu overlords.

Re:So

[Lady Griffin]

It will mean that the Java Runtime is obsoleted by the .Net Runtime: Microsoft wins, again.

Re:So

[Anonymous Coward]

go wash your mouth out with happywatersoap.

Re:So

[nietzsche_freak]

It will mean that the Java Runtime is obsoleted by the .Net Runtime: Microsoft wins, again.

I doubt it.

.NET will never 'obsolete' anything; it's just a monopolist's shoddy attempt to wipe out Java.

I can't believe anybody is 'afraid' of these chips. I laughed out loud when I read the article.

Re:So

[badfish99]

Probably just as long as it took for Sun to get their Java chip into every PC.

Re:So

[Sj0]

What of things I've read saying that .net will be the default api in windows longhorn?

As a former DOS programmer, I can tell you that when Microsoft wants to get rid of an API, they're quite good at it. If they want to do it, win32 will be dead before the end of the decade, just like dos.

Re:So

[Taladar]

It is an entirely different situation.

They can't just kill backwards compatibility now since it is the one big reason to stay with Windows. Most businesses are evaluating other OS now and if the change to a new Windows version requires rewriting all your programs (I know they will probably implement a compatibility layer but we know how well that worked in the past) then they might just as well rewrite them on Linux (or some other OS that 'lacks' MS Security Features (TM) ).

Re:So

[cherberos]

It's not like there isn't anything like this for Java. The first that comes to mind is the TINI-board, from Dallas. There was another one with a more arcane Java-implementation, but less resource-overhead (Can't remember the name right now..). And those are just the ones I worked with. There should be others. So nothing unique here, except maybe that this is the first of this kind of firmware for .Net

Re:So

[Tim C]

So?

Yes, that'll suck for anyone who's currently working with any API it replaces, but that's progress - technology moves on. Besides, the jobs won't disappear overnight, there are still openings for COBOL programmers, for example (there's even a COBOL binding for .net...)

I still don't see the big deal. One of the most frequent criticisms I hear on tech sites of Windows is the cruft that's accumulated due to always maintaining backward compatibility. Surely removing that cruft by removing the backward compatibility would be a good thing?

Not that it'll happen very quickly; there's simply far too much software available using the Win32 API to simply drop support. People would either not upgrade, or (worse) move to an alternative platform.

Re:So

[dan the person]

It's an ARM CPU, not a .NET CPU.

It loads ".NET Embedded" from firmware.

This is like saying an iPaq has a WindowsCE CPU.

Re:So

[Anonymous Coward]

This is like saying an iPaq has a WindowsCE CPU.

Or Slashdot has "editors". You'd think an editing staff would research the wording before posting the story to the site.

Re:So

[oberondarksoul]

Or, even more likely, skip re-writing altogether and keep using whatever they're currently on. Most schools, for example, can't keep upgrading to the latest and greatest version of Windows until it's been out for a while and they have the funds - if the price of upgrading involves an entire fleet of new, incompatible computers, chances are they'll just stick with what works.

Re:So

[aichpvee]

How is it worse to move to another platform except for micros~1's bank accounts? The only reason anyone even bothers with windows anymore is because of the wide range of software available. Take that away and people would be even bigger idiots than usual to stick around.

Re:So

[Glock27]

What of things I've read saying that .net will be the default api in windows longhorn?

Those things you've read are telling you that .Net will be the default API just as Microsoft begins losing serious marketshare. :-)

Re:So

[Joe U]

Windows doesn't really have a default API.

The Windows NT (2k/xp) kernel has the ability to use multiple APIs out of the box, with WIN32/Posix/OS2 1.x support built in. .NET will most likely wind up in this group with the longhorn release.

This will allow .NET apps to run about as fast as current Win32 apps. The average joe user won't know or care what API the software was designed for.

Re:So

[Anonymous Coward]

They aren't getting rid of win32.

The real question is whether win32 will become implemented upon .NET, but I've been told that won't happen by people who should know.

Re:So

[Spiked_Three]

Yeah, that's as dumb as saying I have a linux PC or a Windows PC - wait ..... n/m

Re:So

[Marxist Hacker 42]

A damn long time at $199/chip. My question is, if you're using this for an embedded solution, why wouldn't you just go down to Fry's and pick up an x86 archetecture AMD motherboard instead? With memory, it will cost you less.

Re:So

[aminorex]

that galileo, what an air head. the earth revolves around it's axis. it orbits around the sun.

Re:So

[Glock27]

that galileo, what an air head. the earth revolves around it's axis. it orbits around the sun.

I think you're confusing "revolves" and "rotates". I'll leave who's an airhead to the observer. :-P

From Wikipedia:

In 1600, astronomers were engaging in a great debate between the Copernican system (the planets revolved around the sun) and the geocentric system (the planets and sun revolved around Earth). In 1604, Galileo announced his support for the Copernican school of thought, but he lacked the means to reinforce the opinion.

I hope that cleared things up for you. :-)

Re:So

[Sj0]

the interesting thing about the responses I've reciveved is that none of them seem to notice the precident I mention -- there were lots of DOS programs too, and technically windows still supports it, but as new releases move on, we're going to see less and less in terms of actual support, just like DOS programs under XP will at best run horribly, and at worst be hit with crippling problems, like my buddy who can't use his keyboard in dos programs under XP.

I wonder...

[Anonymous Coward]

...If it would suffer the same fate as those "Java Chips" I see on the market...

Pretty Cool (Application lies herein)

[Anonymous Coward]

Maybe with this technology they can make a clunky cheap looking watch that imports my calendar and contact book by reading a flashing screen.

Re:Pretty Cool (Application lies herein)

[madaxe42]

I had one of those, years ago! It was great, but the fits it gave me completely negated the usefulness of the data on the watch - by the time I stopped frothing, hours had passed!

I for one...

[Anonymous Coward]

I for one welcome our new embedded cpu overlords!

Re:I for one...

[Anonymous Coward]

In the Old People's Soviet Republic of Korea the cpu embeds you.

Features in a nutshell

[amigoro]

From TFA:

• 384K of SRAM, single cycle access
• 27 MHz ARM7TDMI
• FBGA chip form
• ~450,000 instructions per second
• 4MB non volatile flash
• 1.8-volt core, 3.3-volt I/O
• 32768 Hz real-time clock
• 32-pin pinout, including 24 GPIO ports multiplexed with other functions (8 VTU ports, dual serial ports, SPI, and USB port)
• SPI and I2C interfaces

Moderate this comment
Negative: Offtopic Flamebait Troll Redundant
Positive: Insightful Interesting Informative Funny

Is this like a JavaChip

[Anonymous Coward]

Seems for a while, those were really hot too. I never realized just how much (and how far behind) MSFT was following Sun.

Re:Is this like a JavaChip

[Anonymous Coward]

It isn't a microsoft product. It is a .netcpu company product.

Re:Is this like a JavaChip

[Anonymous Coward]

We've been over this, jerk. The Java chip is practically useless because it supports such a small subset of Java.

Re:Is this like a JavaChip

[Glock27]

We've been over this, jerk. The Java chip is practically useless because it supports such a small subset of Java.

Which Java chip? Fittingly there are several, and two support the full J2ME spec, including floating point.

Hope that helped...

.Not a .NET CPU

[Gopal.V]

It is really just a CPU on which CLR runs , not a real .NET CPU in hardware. (or so the TFA seems to indicate from the diagram). Also of the more convenient peices of the ECMA 335 spec.

It's an embedded chip which has a CLR on top of it. Nice idea, sorry that Sun thought of it earlier ( The Green Project) - Sun seems to be consistently missing the BUS here. They came up with "Network is the computer" and now MS is selling ".NET " :)

I've seen a couple of stack based engines but by its polymorphic nature .NET bytecode is not suitable for a direct CPU (you could do something like dynamic translation like the Crusoe chip had). But then it's still a JIT , right ? :)

Re:.Not a .NET CPU

[Anonymous Coward]

the Jini's fall is solely based on cost. Why spend almost $100.00 for a single jini chipset for your devicewhile a regular embedded CPU costs $7.00 has more speed and can use established programming languages like C.

the Java on a Chip Jini is a really cool device but it is horribly overpriced for what it is, when the Dev kit costs almost $300.00 and the Jini board it's self is $100.00 in single quantities nobody will touch it, and that is exactly what happened.

if Microsoft wants this visual Basic chip to even try to make a dent in the embedded PCI world their pricing had better be on par with Microchips and Atmel's offerings. at $7.00 to $20.00 per chip single quantities for something equiliviant in that processing speed and power and storage.

the 4Meg of flash is insanely large for an embedded processor, are they looking to the future or is this typical Microsoft and that is how huge your executible+libs is going to end up?

Re:.Not a .NET CPU

[Anonymous Coward]

.NET is not JIT, it makes proper executables, just the runtime engine is huge.

Re:.Not a .NET CPU

[winfx]

Polymorphic translates : easier to write compilers, harder to JIT it.

Direct CPU mapping has the same difficulty as JVM bytecode, polymorphic instruction set is not a problem compared with the dynamic loading types, inlining, virtual calls, GC etc that the CPU architecture must solve

Re:.Not a .NET CPU

[Anonymous Coward]

Yes, I do hope that they can illegally leverage their monopoly in software to crank out a loss-leader for embedded devices. That would be such a nice win-win situation: devs get Microsoft lockin for cheap, and Microsoft gets the lockin.

Re:.Not a .NET CPU

[Anonymous Coward]

I wonder if it could run NetBSD if it had more memory/flash?

Re:.Not a .NET CPU

[JanneM]

Sun wasn't first either. There's been a number of "LISP Machines", designed to run LISP efficiently, and there's been several microcontrollers designed with some of FORTH:s lower-level words as opcodes.

Re:.Not a .NET CPU

[MoebiusStreet]

And don't forget the machines designed to run Pascal p-code.

Re:.Not a .NET CPU

[LWATCDR]

And the Lilith that ran M-code for modula 2

Re:.Not a .NET CPU

[squiggleslash]

And the 6809 that ran 6809 machine code.

Whatdayamean that doesn't count? Programming the 6809 was like programming in a high level language... well, it felt like it, back when the alternative was BASIC. ;-)

Re:.Not a .NET CPU

[farble1670]

Sun seems to be consistently missing the BUS here. They came up with "Network is the computer" and now MS is selling ".NET " :)

??? so, sun thought of it first, therefore they missed the boat? which is it? you knew that a pointless jab at sun would get you mod'd up, or that you could not resist the urge to make a little play on words?

Re:.Not a .NET CPU

[LWATCDR]

never used the 6809 I did program the 6800 and it was in interesting chip. The 68k was truly a thing of beauty to program :)

Re:.Not a .NET CPU

[RevAaron]

It's an embedded chip which has a CLR on top of it. Nice idea, sorry that Sun thought of it earlier ( The Green Project) - Sun seems to be consistently missing the BUS here. They came up with "Network is the computer" and now MS is selling ".NET " :)

First, this Green Project doesn't seem to have any sort of Oak/Java co-processor/cpu-layer, at least not referenced in that page- it's just SPARC. Even so, Sun has backed the idea of a Java-chip. Java-chip or not, Sun wasn't the first to have such an idea. There was the Western Digital Microengine, though, which implemented Pascal p-code on hardware. Far before the abortion known as Java ever came to be.

Re:.Not a .NET CPU

[RevAaron]

First, at least according to that article, the Star7's CPU was just SPARC- no Java/Oak co-processer/CPU layer. Perhaps there's a reference to that somewhere else, or just a mistake on your part. However, Sun has backed Java-chips in the past, until they were found not to make much sense. Even so, Sun wasn't the first to have the idea- as far as I know, the Western Digital Microengine was the first to implement a bytecode instruction set, the Pascal p-code, as real hardware. Another idea that Sun certainly didn't think up, though people tend to give them more credit than is due to such "innovations."

Re:.Not a .NET CPU

[squiggleslash]

I certainly agree about the 68000 (and its successors), never used the 6800.

Motorola seemed to have an intuitive feel for how a processor should be programmed, unlike the Intel processors and their offspring which were feature laden but... and as for the 6502, I'll never understand why people loved that chip. 8 bit addressing? Ouch!

Ahhhhhhh though, the 6809. I need to dig out the programs I wrote for the Dragon 32 (a TRS80 CoCo clone) at some point.

Ob6809Trivia - it had a SEX instruction, causing much chuckling to sad 15 year olds as I was when I programmed it. (Sign Extend, IIRC, made an eight bit signed value into a 16 bit signed value.)

Re:.Not a .NET CPU

[LWATCDR]

People loved the 6502 because there Atari, Commodore, or apple ran it. They where cheap and faster than just about anything else per clock tick.
A 1 mhz 6502 was just about as fast as a 4mhz Z-80.
In many ways the 6502 was the first risc chip. Just think of zero page as 256 registers.

Yea I often wonder how much better off we all would have been if IBM had picked the 68k over the nasty but cheap 8088.

Re:.Not a .NET CPU

[spiko-carpediem]

I believe 4MB is more an average computer user will ever need

Re:.Not a .NET CPU

[blankslate]

Yes, but does it ... oh, right.

ctrl-alt-del keys?

[rleyton]

They must be very small, but I think I can see them if I look really closely and squint a bit.

Re:ctrl-alt-del keys?

[nmg196]

As it doesn't run Windows, and .NET is widely accepted to be very stable, your joke really isn't very funny.

Re:ctrl-alt-del keys?

[Anonymous Coward]

I think I can see your humour if I look really closely and squint... just a bit...

wait no sorry I can't.

Re:ctrl-alt-del keys?

[Anonymous Coward]

Hmmm. Seeing ctrl-alt-del keys everywhere? Get that man some linux, stat!

Re:ctrl-alt-del keys?

[tyndyll]

BSOD on a chip - i'd buy that for a dollar

Re:ctrl-alt-del keys?

[DrSkwid]

it was funnier than *any* of your posts modded funny

Re:ctrl-alt-del keys?

[Anonymous Coward]

As it doesn't run Windows, and .NET is widely accepted to be very stable, your joke really isn't very funny.

Well, it is true that it is accepted that Ximian and Miguel de Icaza are widely accepted to write very stable code - so I agree it's an overgeneralzation to say .NET isn't stable.

But I think the guy was alluding to the fact that that other group who writes unstable buggy crap code that is less stable and less secure also has an implementation of C# and the .NET CLI. That will almost certainly have as many security holes (whether intentional back-doors or not) as their other products.

Re:ctrl-alt-del keys?

[theolein]

A Sense of humour would do you some good, you know?

Re:ctrl-alt-del keys?

[nmg196]

I have a sense of humour - a good one. But jokes about Windows being unstable are really boring now. Windows hasn't been unstable since Windows 98 and that's 7-8 years old. Nobody who uses Win2k or XP has to keep pressing the beloved Ctrl-Alt-Delete-Vulcan-Nerve-Death-Grip key combination...

OK - maybe to some people it was funny - and I've seen one (hundred) too may Windows is unstable jokes on here...

Re:ctrl-alt-del keys?

[gibson_81]

Nobody who uses Win2k or XP has to keep pressing the beloved Ctrl-Alt-Delete-Vulcan-Nerve-Death-Grip key combination...

That's true, cause once it bluescreens you can't use CAD - you have to press the reset button on your case to restart the box ... I'll agree that Win2K/XP is a great improvement, but it still isn't 100% ... (not that Linux is either, although the only times I've gotten OOPSes myself were when I was writing my own drivers ...)

Re:ctrl-alt-del keys?

[Ilgaz]

"OK - maybe to some people it was funny - and I've seen one (hundred) too may Windows is unstable jokes on here..."

I also saw dozens of people having some sort of MSCE license and fanatically defending windows here.

Its the hardware and software combined...

G5 mac user until some MSCE licensees figure how to code in objective C with Altivec extensions.

That time I am moving to "dead" SGI.

Need quality code you know ;)

Re:ctrl-alt-del keys?

[Ilgaz]

I don't know if there are any chip making/pro geeks here but I sure saw at Discovery that they embed their signatures, cartoon characters to chip plans.

Anyone with more info?

Re:ctrl-alt-del keys?

[nmg196]

> That's true, cause once it bluescreens you can't use CAD

I've never seen Windows 2000/XP bluescreen. Neither have any of my colleagues in the software development business. If you have decent branded hardware with decent drivers you simply do not get bluescreens.

In fact in a previous slashdot debate, a few people were actually arguing that XP didn't even have the code in it to render a bluescreen - and it looked like they were winning the argument until someone managed to dig out a screenshot from somewhere. So don't go telling me that an XP bluescreen is any more common than a linux kernel panic...

Re:ctrl-alt-del keys?

[Haeleth]

Windows hasn't been unstable since Windows 98 and that's 7-8 years old.

You must have missed Windows Me, which makes Windows 98 look rock-solid.

Nobody who uses Win2k or XP has to keep pressing the beloved Ctrl-Alt-Delete-Vulcan-Nerve-Death-Grip key combination...

...except to deal with badly-behaved applications. But those are hardly Windows' fault.

Re:ctrl-alt-del keys?

[dcam]

I've had 2000 blue screen on me a number of times. From memory most of them were due to hardware issues.

Re:ctrl-alt-del keys?

[gibson_81]

I'm not saying that XP BSODs more often than Linux does ... I think we can all agree that for both OSes it's seldom enough that it is hard to get any statistical certainty(sp?).

As for decent branded hardware, I consider Promise to be a 'decent' brand when it comes to IDE controllers, but since my bluescreens started to appear after I installed the drivers for my Ultra 100TX2 card (from Promise's homepage, but appareantly _not_ signed by Microsoft), I guess they're not ... On the other hand, if I use XP's driver for that card, my 180gig drive only registers as a 128gig, so unrolling those drivers is not really an option ...

Scary (saracasm)

[wcitechnologies]

This is about as scary as the current .NETs threat to take over the internet. Oh, wait, there is no threat, because nobody bought into .NET.

Sorry, Microsoft-- just because you want something to be true doesn't mean that it is.

Re:Scary (saracasm)

[Anonymous Coward]

Plenty of people develop for .NET. We've got ten plus here doing it today, and the rest of the company works in .NET from time to time. Our enterprise web app runs on .NET. We've got plenty of customers.

Sorry, wcitechnologies-- just because you want something to be true doesn't mean that it is.

Re:Scary (saracasm)

[Anonymous Coward]

Sorry AC just because you want something to be true doesn't mean that it is.

Re:Scary (saracasm)

[kahei]

I'd say that I see more .NET projects starting than any other single platform in my industry, although the lead over Java is slim and obviously there is far more Java already running. I wouldn't say that .NET has reached the 'will be with us forever' point that Java and C have, but it's certainly been very popular with devs and had a number of successful early projects. In the end it will probably stand or fall on the success of Longhorn (which everyone is quite skeptical about). But buy-in has been good.

Re:Scary (saracasm)

[Anonymous Coward]

I know. I wish I had mod points today so I could smack wcitechnologies (you?) down. His frankly-quite-immature tone really grates, doesn't it?

Re:Scary (saracasm)

[the angry liberal]

Sorry, Microsoft-- just because you want something to be true doesn't mean that it is.

Perhaps if you put your troll's club down long enough to take a look at sourceforge, you would notice most of the newer open source applications for Windows are being developed in .NET.

It won't take over the Internet, but it has been well accepted and is easy to use.

I wonder though, with all this FUD, if anyone can produce real numbers showing which is in more demand in the workplace: Linux developers vs .NET developers. I'm not talking about which is more 31337, I am talking about which one will find more steady income and have less trouble when they need to change jobs.

Re:Scary (saracasm)

[shufler]

As you mentioned, newer Windows applications are being written in .NET, and you go on to ponder what the demand is. I don't know of any numbers, but I'd imagine .NET developers would be in all sorts of demand with respect to developing on any Windows platform, after all, .NET is the new API which replaces Win32.

I agree that there is all sorts of FUD flying around about .NET, and it's pretty sad that it is. I'm not a Microsoft fanboy, but anyone who cannot recognise the Official API of future Windows development is in serious trouble (if they intend on developing future Windows applications, that is). As you said, .NET isn't going to take over the Internet (who said it would in the first place?), but it will take over ALL Windows development.

All that said, I seem to remember reading about how Microsoft was dropping .NET, however I highly suspect I dreamt it.

Re:Scary (saracasm)

[benjymous]

Personally as someone who's long developed Windows Win32/MFC code (in C++) and is now moving to C# .NET stuff, I'm finding .NET an absolute doddle.

let's face it - MFC and Win32 are old and have been cobbled together, seemingly at random over the lifespan of the whole Windows family, meaning nothing feels like it's ever really been designed

One function returns a colour, another function needs a colour. Oh dear, one uses some kind of int, the other a struct (oh and another some kind of class) - lets bog down our code with lots of conversion functions - Most of the time the sensible obvious approach to a task is the wrong one.

So far in .NET, whenever I've wanted to do something, I've looked at the classes, thought "How would it be sensible to do this", and 9 times out of 10 it works perfectly

Re:Scary (saracasm)

[bloodredsun]

Which industry is that?

Both media and banking (my last 2 contracts) have been dominated by Java/J2EE, although this may well be due to the fact that I am primarily a server developer, with my only client programming being the web presentation layer. .NET looks good for serious windows GUI apps due to the shortfalls of Swing/SWT but anything distributed has gone to Java, especially at the enterprise level.

I know it's horses for courses but I'd like to know personal feedback for different industries.

Re:Scary (saracasm)

[Taladar]

You have a strange view of the Industry if you think Java has reached "will be with us forever" stage just 10 years after the initial release (and that didn't resemble Java as we know it today in any form). Java has too many problems and not enough advantages to stay as long as C has. Don't get me wrong, the idea of the virtual machine will probably stay for a long time but not Java as a language or as a Runtime Environment.

Re:Scary (saracasm)

[Anonymous Coward]

Any pointers to interesting projects?

Re:Scary (saracasm)

[Mant]

I can't imagine .NET is going to take over the Internet, but ASP.NET is a very nice platform for writing web apps. OK they will probably run on IIS, unless you use Mono, but it is a big step up from the scripting languages approach of basically just printing out the web page.

It gives some nice abstraction to writing web pages, you don't have to worry about hand crafting every bit of HTML that is going out to the browser (although you can if you want or need to), and can deal with the concepts, objects and events.

.NET does little that is new, Java was doing much of it first, but for writing web apps it is pretty simple and powerful and has good development software. We are moving to it at work because it makes us more productive.

Re:Scary (saracasm)

[tchernobog]

Although I agree with you that it isn't the case to troll everything that has "microsoft" into it, I think that an high income isn't the first requirement for someone that foreseek freedom of choice and information (why develop Free Software, else?).

The fact that 85% of the computer world use MS systems doesn't mean that it's the best thing to do. Still, things are (really) slowly changing. Maybe I'll live the day when the market share between MS and *nixes 'll be 50%-50%... and that would mean real competition, not just "smithe the infidel with teh big hammer" as almost everyone on both sides tries to do (often don't understanding really what's right to "fight" for).

Re:Scary (saracasm)

[Sj0]

Not really a fair comparison; Linux is an OS, .net is an API.

To be honest though, I'm not looking forward to the future. It looks like Microsoft is going to be using brute force to get developers to use their .net platform for just about everything,

I have a feeling that this may be the last generation of computers for a while that feels truly speedy. :/

Re:Scary (saracasm)

[Anonymous Coward]

Yes,
InterestingProject *mInterestingProjects = stackalloc InterestingProject[10];

Re:Scary (saracasm)

[ceeam]

.NET is the new API which replaces Win32

Say it when you have winword.exe - the dotnet executable. (Or quake4.exe for that matter).

Re:Scary (saracasm)

[dmayle]

All that said, I seem to remember reading about how Microsoft was dropping .NET

You're probably thinking of when Microsoft dropped the .NET branding from all of their server software. Originally, there was going to be a Windows Server.NET, and a SQL Server.NET, etc. Combined with the .NET branding on the languages, no one really knew what it meant to be ".NET".

Because of this, Microsoft cancelled all of their plans for .NET servers. This left Visual Studio.NET, which is used for developing the .NET languages. Programs written in .NET languages run on top of the .NET framework. (The Common Language Runtime.)

Antoher factor in the dropping of the .NET branding for the servers, I'm sure, is that it would have meant more stringent requirements for release dates, and it would have been more difficult to move to the next version (forced upgrades) without some branding confusion. (Think "Windows Server.NET Two!" Wait, is there a new version of .NET? Will my applications run on it? etc.)

Re:Scary (saracasm)

[doctormetal]

The fact that 85% of the computer world use MS systems doesn't mean that it's the best thing to do.

It depends what computer world you are looking at. In my line of work, dedicated embedded devices, the numbers are reversed and microsoft has a small marketshare.

Re:Scary (saracasm)

[tarunthegreat2]

Say what you want about .NET (I disagree with u completely BTW), but the IDE KICKS Hardcore BUTT! I have yet to see something better the VS .NET 2003 for development. Quite a few people have bought into .NET and if I have a choice between C++ and C#, I pick C# thanks....but then I was born and raised on C and Java anyway....

Re:Scary (saracasm)

[Tarwn]

I've seen .Net moving in quite heavily in the manufacturing world. This is one sector that MS has a strong hold on simply because there are so few people that want to sit down and write the hundreds of communications drivers, etc needed to create manufacturing data systems. Or maybe because once you buy a manufacturing system you don't want to switch brands until you get back out of the hole with it :P
A lot of the products I have seen (both data collection and warehouse-type) are moving to .Net SDK's, and a lot of internal application programming seems to be moving that way also.
The last major project I did was:
1 part config client and 1 part server
"please maximize uptime"
"please maximize scanning capabilities"
"please correct our last 9 months of errors and get it on the shelf in 2 months or less" .Net was the available choice (unless you consider VB6 as a valid choice...*shudder*) and despite the fact that the only other competing product was written in C++ (we think) we also managed to turn out a more efficient server (not that I don't think i could have made it even faster in C++, I just expected the other company's to suck that badly :P).

A lot of internal app's get written in .Net, as the valid choices are generally VB6 vs VC++ vs some flavor of .Net vs Excel VBA (you think I'm kidding). The few other languages the make it onto the plate are generally as bad as VB6, so I prefer to leave them unmentioned. About the only time I have seen it go beyond this is the few times I introduced the power of bash scripting something on my laptop's preferred partition (ie, not windows) ;).

Ok, enough rambling :)
-T

Re:Scary (saracasm)

[Tim C]

Jobserve seems to disagree with you:

search for ".net", any job type - 1629

search for "c", any job type - 1499

search for "java", any job type - 3009

search for "c++", any job type - 2300

Your comment has too few characters per line (currently 15.6). Guess I'd better explain then that jobserve.com is a major UK-based jobs web site, catering to a large number of industries. While the jobs advertised are mainly in the UK, they also cover parts of Europe, and have a site dedicated to jobs in Australia. How's that slashcode, better?

Re:Scary (saracasm)

[tarunthegreat2]

As you said, .NET isn't going to take over the Internet (who said it would in the first place?)

Well not take over, but give everything else a run for its money, because at least it's semi-standards compliant with SOAP and XML and stuff. I LOVE the fact that all DB objects are stored in an XML format by default. Now that is truly revolutionary thinking (which probably get me marked flamebait...)

Re:Scary (saracasm)

[ceeam]

Yes, new things tend to be slick and shiny. Applies to almost everything. When they grow up they become burdened with all kinds of supporting sticks, duct tape and patches. Take PalmOS, take the directories structure on your HDD, take Java, and even when the brightest guys handle it (mod me to hell) - take Python or Unix... Not as easy as it used to be "back in the day".

.Net (WTF - extremely "ungooglable" name, BTW) is young and peppy... Wait 5-10 years for it to mature though.

Re:Scary (saracasm)

[Anonymous Coward]

You forget one thing, Java is being used in Banks, that pretty much guarantees that it is gonna be around for a loooong long time.

Re:Scary (saracasm)

[Tim C]

it would have been more difficult to move to the next version (forced upgrades)

Forced upgrades? Releasing a new version of a language/API is a "forced" upgrade now? Who's forcing you to upgrade when a new Linux kernel or gcc comes out?

Re:Scary (saracasm)

[Tim C]

I can't do you Quake4, but how about Quake 2?

(Eactly how much Win32 do you think Quake 4 is going to use anyway?)

Re:Scary (saracasm)

[Anonymous Coward]

we've got 500 people programming in C.

and I can spend less on the hardware and get a better product than your beloved .net could ever think of.

so what was your point again??

the delay in product launches are NEVER software but hardware.

Re:Scary (saracasm)

[droolfool]

My experience with .NET:
- Build pages using Web Forms
- Watch the result in IE: just fine.
- In Mozilla: Horrible! Several CSS properties missing. (yes, missing, they were present in IE).

So, my so-called cross-platform web application became an IE-only application. It was not a web site, it was an interface to a power plant monitoring system (built with no Java/ActiveX/.NET, only HTML, Javascript, CSS... It even had zoomable charts. Yes, the chart was a static image :P

Re:Scary (saracasm)

[asliarun]

Sorry to pick nits in your comment, but .NET (or ADO.NET) does NOT natively store db objects in XML. Native database objects, such as DataSets, DataTables, etc. are stored in binary format, just like all .NET objects.

It only seems so because almost all the data objects in ADO.NET allow the user to readily serialize and deserialize the objects into and from XML. So yes, XML is the native serialization format for datasets, but not the format in which datasets are stored themselves (because of performance and size penalties).

Re:Scary (saracasm)

[Ours]

Maybe you are confusing the fact that Microsoft dropped the Hailstorm project (officially My.Net or something like that). It was supposed to regroup Passeport.Net, Calendar.Net, Hotmail and .Net itself. Fortunelly Microsoft realised that people weren't going to developp applications where most of the information would be stored on Microsoft's servers. A silly idea that gave a bad image of .Net. Good thing it's now well forgotten. It could have killed .Net.

Re:Scary (saracasm)

[GnuVince]

I'd rather use Seaside for web applications or Ruby on Rails for a web-based interface to a database than use ASP.NET. Seaside is probably the most advanced thing I've seen to date, you should check out the liveWeb project that is built with it, it's gonna blow you away.

Re:Scary (saracasm)

[GnuVince]

You are obviously not a Smalltalk person are you? The numbers of useful, productive features that a Smalltalk IDE has that VS.NET doesn't is not even funny. Check out the non-commercial version of VisualWorks Smalltalk (by Cincom) and you'll see that VS.NET is suddenly not so hot.

Re:Scary (saracasm)

[Anonymous Coward]

droolfool:

The machine.config configuration file on the server contains a section where the capabilities of each browser is defined by parsing the user agent via regex. As such these behaviors are entirely customizable. I would bet that since Mozilla and FireFox are newer than .NET 1.1 that they were never included. Either that or perhaps they're being matched as an earlier version of Netscape. Give it a look-see:

%SystemRoot%\Microsoft.NET\Framework\%FrameworkV er sion%\config\machine.config

Re:Scary (saracasm)

[SilentChris]

The plan for .NET is not to take over the internet (despite the name). The real goal is to simplify the development process. There's a lot of draws to it:

* Robust memory-protection and cleanup: Visual C developers can no longer screw up and create buffer overruns when using managed extensions.
* The ability to write in any language and mix them: Unlike other VMs, this isn't tacked on. It's a fundamental design of the platform. This is done exceptionally well.
* A better, simpler windowing scheme: Although Avalon looks to complicate this.

I'm not as fond of ASP.NET (it complicates things by bringing a lot of old ASP constructs over) but developing in .NET is a pleasure compared to the old scheme. It's not going to take over Java, but will likely work side-by-side with it (.NET on Windows boxes, Java on everything else), which is fine by me.

Re:Scary (saracasm)

[Anonymous Coward]

You're right, fortunately the developer convention (for very nice googlability) is "dotnet"

Re:Scary (saracasm)

[Anonymous Coward]

You're a moron. Notice that the sister to this post schools you. Get the facts before you spout off!

Re:Scary (saracasm)

[droolfool]

Fuck you.
I got the facts, because I actually CODED IN .NET. Yes, I tried to edit the .config file, but it didn't work very well. I had serious problems with GC , like pieces of memory never freed, and guess what, not my variables, but .NET specific stuff I couldn't control. I had web sessions misteriously disappearing.
Such things actually HAPPENED to me. Stop whining, calling me a moron. I never said .NET is bad. But guess what? It's not perfect, it's not the right tool for every single job. Sometimes you're better off with J2EE (hint: clustering).

Re:Scary (saracasm)

[gcaseye6677]

At the risk of being trolled, I must ask, how much does the salaries of 500 programmers cost? Would you need this many programmers with a more modern language? There is more to project cost than hardware and software. The biggest expense is usually labor. Also, every project delay I've ever encountered has been because of software problems or combined hardware/software problems.

Re:Scary (saracasm)

[bushidocoder]

All that said, I seem to remember reading about how Microsoft was dropping .NET, however I highly suspect I dreamt it.

I'm pretty sure what you were reading about is their attempt to drop the name .NET, not the technology. Microsoft is not backing off .NET as platform - if anything they're more excited about it now than they ever were.

From a marketing perspective, everything that came out in 2001 was labeled X.NET. Office.NET. MS Money.NET. They dilluted the phrase .NET so much that people didn't ever figure out what it is, and from the looks of most people on slashdot's comments, people still don't know what it is. .NET is not an attempt to take over the internet (in fact, if you look at the technologies love of web services and XML, you might even think .NET is trying to play nice on the internet). MSN Passport.NET which shares the name was an attempt to take over the internet that failed miserably, but its unrelated to .NET.

Re:Scary (saracasm)

[Oblio]

VS2003 works really damn well with C#, but I still prefer VC6 for C++ devel. The IDE has really become much harder to navigate, and the only thing that mitigates that cost is that when intellisense works, it is truly a thing of beauty.

And intellisense always works with C# (or as nearly always as I can tell).

But I'm dissappointed with C++.NET (though it looks like this will be mitigated with VS2005), and I'm dissappointed at how keyboard navigation became harder. OH, and I'm REALLY dissapointed about the drop of support for Win9x. Remote debugging is a pain.

Re:Scary (saracasm)

[Digital11]

Now THAT is geek humor. Bravo!

Re:Scary (saracasm)

[Teckla]

You have a strange view of the Industry if you think Java has reached "will be with us forever" stage just 10 years after the initial release (and that didn't resemble Java as we know it today in any form).

I won't comment on the "will be with us forever" part, but I will say this: Java is showing every sign of being a long-term player. I'm not sure how you try to estimate the staying power of languages and platforms, but by every reasonable measure, Java will be with us a long time.

Java has too many problems and not enough advantages to stay as long as C has.

This is just a meaningless troll unless you explain what show-stopper problems you think Java has that'll limit its future potential. Java has some problems, yes, but none of them are show-stoppers, and the language continues to evolve nicely (see Java 5).

In addition, the areas where Java and C compete with each other are pretty small already. You typically won't write low level systems code in Java, and writing general purpose applications in C is making less and less sense every day.

Don't get me wrong, the idea of the virtual machine will probably stay for a long time but not Java as a language or as a Runtime Environment.

And you base this conclusion on...what? With all due respect, your post is all words and no substance. As I said before, by every reasonable measure, Java will be with us a long time.

Re:Scary (saracasm)

[voodoo1man]

If by "will be with us forever," you mean that in a decade .NET will leave millions of lines of bug-ridden, festering and unsupportable C# and Visual Basic .Net code, then yes, I agree with you. COBOL has touched every one of us in ways we shouldn't have been touched in, and everyone still remembers PL/1. Visual Basics 5 and less are pretty much in the graveyard already (and I know of some geniuses who decided to use it because they thought it would be the way of the future - so much so for trusting Microsoft to build a lasting product). Java may seem popular and widely used, but this is largely Sun spin; if it happened to just disappear spontaneously, I wouldn't notice a thing, and chances are neither would you. I think you can pretty much put it in the "living dead" category right now. Now C, there's a widely used language with a thriving and active programmer community. C++ even more so. You can't predict what will happen to a programming language in 10 years (much less "forever" - remember that computers as we know them today are barely 60 years old!), although it's tempting to try.

Re:Scary (saracasm)

[spiko-carpediem]

I don't get it :| stackalloc is a classname ?

Re:Scary (saracasm)

[abpimentel]

Seeing is believing...
Can you provide a video similar to what this guy did using .NET framework.
http://www.rubyonrails.org/show/HomePa ge

It is not a real CPU , from what I read.

[Gopal.V]

a CPU for the .net CLR, that's all, in much the same way as Pentiums and Athlons, etc, are CPUs for x86

No, it is a CPU for .NET CLR as much as a Gumstix is a CPU for Linux kernel. It's just a VM embedded on firmware, NOT a REAL CPU.

Btw, the JVM FPGA is a real example of a VM less execution (or more correctly , a native JVM + support libs).

Re:It is not a real CPU , from what I read.

[the angry liberal]

No, it is a CPU for .NET CLR as much as a Gumstix is a CPU for Linux kernel. It's just a VM embedded on firmware, NOT a REAL CPU.

I can only begin to guess what your definition of a CPU is. Anyway, it still isn't going to eat your mother or pull your cats tail. It is just a chip from a vendor you don't like. Move on.

Re:It is not a real CPU , from what I read.

[DrMrLordX]

I submit to you the notion that the .netcpu would be far superior product if it COULD eat someone's mother or pull a cat's tail.

Oh, the possibilities . . .!

Re:It is not a real CPU , from what I read.

[SenseiLeNoir]

Not quite correct, ALL modern CPU's are based on a type of firmware (read Microcode). Modern Athlons/Pentiums, have multiple RISC units onchip, with a translator intepreting the x86 code.

Re:It is not a real CPU , from what I read.

[Anonymous Coward]

Not quite correct, ALL modern CPU's are based on a type of firmware (read Microcode).

Shouldn't you include a disclaimer with that, like "ALL modern CPU's does not include all modern CPU's"?

Re:It is not a real CPU , from what I read.

[zero_offset]

It's just a VM embedded on firmware, NOT a REAL CPU.

The CLR is not a VM.

In Korea.....

[ucdoughboy]

In Korea only old people use .Net natively on their cpu's.

Re:In Korea.....

[Anonymous Coward]

Mods where is your sense of humour - this is farkin funny

Re:In Korea.....

[Anonymous Coward]

what a loser --- 2 in korea jokes in 5 days. give up already.

Erlang CPU

[Anonymous Coward]

There have been an Erlang CPU around for a long time:

http://www.erlang.se/euc/00/

Look for ecomp on the page.

--
Mickaël Rémond
http://www.erlang-projects.org/

Parrot

[hey]

I'm waiting for a Parrot chip.
Now that would be exciting.

Re:Parrot

[martin-boundary]

Just make sure it's not NAILED to its perch. Or PINING.

Re:Parrot

[Steve Cox]

The Finnish company Koncernen makes these already.

Steve.

Re:Parrot

[joshuaobrien]

Not as exciting as a Pirate Ship!

*Squawk! Pieces of Eight!*

Re:Parrot

[Anonymous Coward]

Mod this as (Score:-1,loser)

Big deal

[oakad]

I bet, every low cost Samsung printer (they like ARM7 microcontrollers) has this ".netcpu" embedded inside - though without dependence on stupid microsoft dev tools.
Anyway, it'll take few hours at most to get ucLinux running on this module.

Re:Big deal

[KlomDark]

At risk of getting burnt to a crisp - what do you mean by "stupid microsoft dev tools"?

I'd say that's one thing MS has done right. I haven't seen anything for Linux that approaches the level of usage of Visual Studio .NET 2003.

KDevelop is frustrating, since you have to hop back and forth between it and QT Designer just to build an interface and wire it up.

As far as anything else along a Linux IDE, I haven't seen anything impressive. Maybe Eclipse, but I haven't tried it out since it was first released (And it stunk bad then).

What's out there for good GUI-based development for Linux?

Re:Big deal

[Anonymous Coward]

What's out there for good GUI-based development for Linux?

If you think VS is good, I don't think you will find anything you like on Linux. That kind of crap is not tolerated in an environment where the users actually get to chose the software they like.

(Yes, I do use Visual Studio .NET 2003 everyday at work, because I don't get to chose. I can complain all I want).

Re:Big deal

[MasterDirk]

Then try out Eclipse and stop whining. If you don't like it, come back and whine some more.

Re:Big deal

[Anonymous Coward]

Hate to break the news to you Eclipse is not that good. Perhaps instead of using the knee jerk reaction of everything M$ does is evil and bad, perhaps the Open Source community could actually see some of the things M$ has done right, and make it better.

But, I guess it is always easier to criticize and bring your opponent down, then to make a better product.

Re:Big deal

[Bitchslap_69]

I'll agree with the statement that, "If you think VS is good, I don't think you will find anything you like on Linux." The second part of your statement is stupid and I don't care if you do use VS.Net every day at work (I tend to think you're a liar, but if not then you're just ideologically resistant to using the better tool). The closest thing on Linux you could use to VS.Net is IntelliJ, which is quite awesome. Depending on what you're doing, Eclipse is passable, but if you think it's remotely competitive with VS.Net in anything other than price, you're high. As a development tool, VS.Net is lightyears beyond anything available on Linux desktops. If you prefer emacs and command-line compilers, then that's cool, but without a lot of work and customization that's NOT a feature-rich environment and, whether you share that philosophy or not, that's what a lot of developers want. The main drawback with VS.Net is that you're locked into deploying on Windows. I work on both Windows and Linux, both for work and my own development. I prefer Linux as a deployment environment and Windows as a development environment and I'm not the only person to have that philosophy.

Re:Big deal

[Bitchslap_69]

Amen. It's worth noting that the lead on VS.Net, Anders Heljsberg, was also responsible for Delphi at Borland. So everyone that laments the passing of Borland's dev tools (as I do; I worked on Borland C++ for years) should realize that the same development philosophy from there is what's behind VS.Net.

Re:Big deal

[si618]

-=> Hate to break the news to you Eclipse is not that good.

Stop trolling knee jerker, give us some details as to why Eclipse blows.

I think Eclipse is great, it has allowed me to move from Windows to Linux whilst still developing. I use Eclipse for PHP, Actionscript, HTML & CSS, Ant tasks (build and deploy), Docbook markup (using OxygenXML plugin). Plus the integration of CVS and SVN version control is good and 3.0.1 crashes very rarely in both Windows and Linux environments.

But, I guess it is always easier to criticise without backing up your claims.

Also included

[Dachannien]

While supplies last, you'll also get a free acronym dictionary, so that laymen can understand what the hell this thing does.

Re:Also included

[Anonymous Coward]

CLR = Common Language Runtime (The CLR is to .NET what the Java Virtual Machine is to Java)

SPOT = Smart Personal Object Technology (MS marketspeak for gadgets with embedded computers)

DIP = Dual Inline Package (The old amateur friendly chip package with the through-hole pins on two sides of a rather big carrier)

CPU Module = Central Processing Unit Module (A carrier module with the processor and some other things on it, like memory, interface chips and glue logic)

I2C = Inter IC Bus (I squared C - IIC - I2C...)

SPI = Serial Peripheral Interface (like I2C, a data connection between chips)

SoC = System on a chip (combination of most components of a complete computer system on one chip)

I/O = Input/Output

USB = Universal Serial Bus (repeater based serial point to point communications interface forming a logical bus)

VTU ... no idea

Re:Also included

[frozen_kangaroo]

Yes, its saying you can now display a BSOD (Blue Screen of Death) on a small LCD (Liquid Crystal Display)

Re:Also included

[p0]

VTU = Very Terrifying Unit

Re:Also included

[Anonymous Coward]

GSOD or Grey Screen Of Death?

Wow!

[DecayCell]

Imagine a beowu... oh, wait.
Nothing to see here, move along!

Re:Wow!

[maxwell demon]

Given that nowadays Linux gets ported to about everything, at some time we'll certainly see Linux for CLR, too.

BTW, has anyone already ported Linux to Emacs Lisp? :-)

Never will be a direct chip for Parrot

[Gopal.V]

Parrot is not a very good design to put on a chip, for one single reason.

Too Many opcodes (1500 at my current count and growing).

Morover parrot has opcodes which do very complicated things like "print_nc" which prints a FLOATVAL constant. Compared to that IL opcodes are simpler and JVM is still more simpler (CVM is even simpler - which is what I'm working on now).

Parrot is too complex, period.

Re:Never will be a direct chip for Parrot

[Anonymous Coward]

Either you have a complex (ie capable) chip or lots of memory chips to hold the bloated programs.

Re:Never will be a direct chip for Parrot

[RAMMS EIN]

At least the Parrot designers realize that the more sophisticated the operations your VM provides, the less emulated code their will be. Knowing this, I can't help but wonder what the real benefit of an instruction-level VM is. Why not have the VM interpret the parse trees of a high-level language instead?

Re:Never will be a direct chip for Parrot

[Anonymous Coward]

You do not have to support all instructions in the CPU. The MIPS CPU I'm working with does not have support for float point instructions. Such instructions cause an interrupt and the operation is performed in the software (interrupt handler). This is obviously slow but works ok for code that does not use unsupported instructions very often.

Stupid ramblings

[Neo-Rio-101]

But does it run Linux?

Oh who cares.... it's Microsoft hardware.
Hmm... bit of a condrum that one. If it's Microsoft hardware, shouldn't it be Microhard software? Or since soft+hard = firm, shouldn't it be Microfirm jigglyware...

(Note to self: STFU)

LINUX: Linux Is Not UniX

Re:Stupid ramblings

[Pants75]

Microfirm jigglyware Dude, stop. You're pushing all the right buttons!

Re:Stupid ramblings

[zr-rifle]

> But does it run Linux?

The real question is "does it run Mono?"

Re:Stupid ramblings

[Anonymous Coward]

>> But does it run Linux?

> The real question is "does it run Mono?"

Yes, of course, you are right. "Does it run Linux?" was a completely fake question.

YUO FAIL IT

[Anonymous Coward]

are having trouble worthwhile. It's surprise to the is mired in an OF AMERICA) is the a losing battle; the BSD 7icense, will not work. And bought the farm... empire in decline, so that their so on, FreeBSD went would mar BSD's deeper into the who sell another shout the loudest the official GAY is EFNet, and you Romeo and Juliet NetBSD posts on short of a miracle DURING PLAY, THIS to you by Penisbird come Here but now For it. I don't 'Yes' to any been sitting here Are you a NIGGER WASTE OF BITS AND There are somew so there are people

Maybe its just me but....

[ezelkow1]

this thing seems like an overpriced piece of junk just trying to hawk its .NET and VS support. Most of the microcontrollers out there i have seen can in some way or another be programmed in C and its various forms. 200 dollars just for the cpu seems to be asking a lot when the only advantage i see is that is 4mb of flash, and other MC's can always be expanded to that anyway. Besides the fact that other MC's out there that are cheaper also contain a whole lot more peripherals and features than this one. But maybe thats just me

Re:Maybe its just me but....

[mvdw]

Don't think of it as a product part, think of it more as a BASIC Stamp for people who want something more than a BASIC Stamp can manage.

BASIC Stamps are good for when you only want to do one, and don't want to lay out a board with crystal, peripherals, etc. Although I have a tendency to do my own boards, I can see that BASIC Stamps are good for some projects.

Re:Maybe its just me but....

[Fallen_Knight]

it looked like 200$ for JUST the modual, not the kit....

STAMP may be nice but its so much cheaper to buy a PIC for 5$ and use some veroboard.

Re:Maybe its just me but....

[Lumpy]

that is because it's trying to become a "super" basic stamp.

It's way overpriced and has no advantages over the microchip or Atmel offerings.

io and ports are multiplexed. what moron engineer though that was a good idea? put it in a 40 pin DIP if you want to follow that legacy form factor or a QFP for regular use (please do NOT use BGA, BGA is from the devil!)

I see this as a no starter. the .NET chip has a really hard time trying to get into a market that is dominated by HUGE companies with decades of history in embedded processing.

Hell if I really wanted a high level language chip I'd use the Java Offering or one of the plentiful "basic" interpeter embedded processors.

Re:Maybe its just me but....

[Fallen_Knight]

overpriced is right, i guess it needs that 4mb of memory for the huge .net binary?

Re:Maybe its just me but....

[hedge_death_shootout]

i guess it needs that 4mb of memory for the huge .net binary?

RTFA: the CLR that runs on the chip is 132 kilobytes.

[NB: The word 'dumbass' was deleted from this post to increase politeness quotient]

Re:Maybe its just me but....

[Anonymous Coward]

There's something kind of sick anyway about designing hardware for Microsoft bloatware.

Re:Maybe its just me but....

[Suidae]

The price would be tolerable if it included an ethernet port and TCP stack.

Re:Maybe its just me but....

[Fallen_Knight]

i was refering to the compiled .net programs...

Re:Maybe its just me but....

[mvdw]

For you and me, yes, but for many companies with only one guy who isn't a hardware guy, this is a viable option. Don't forget it would take someone a couple of days to get up to speed with the "raw hardware", costing much more than the $200 retail of this puppy.

Re:Maybe its just me but....

[hedge_death_shootout]

Which are normally a lot smaller than their win32 counterparts, in my experience.
But anyway, admit it, you were just having a whinge (Neville).

Merry Christmas.

Balmer claimed earlier :

[Anonymous Coward]

".net has nothingto do with Java"

This guy is realy brilliant ... or dumb ;-)

In your oppinion, which one is the worst nightmare for Gates, Linux or Java ?

IMHO, it is Java on Linux :)

Vive GNU's Classpath project !

It sounds SO good!

[OwlWhacker]

The chip... runs the same software as the SPOT smart watches.

I wonder if it's going to be as popular as the SPOT smart watches?

Security ?

[Anonymous Coward]

What happens if someone discover a flaw in the CLR ?
Do we have to buy another processor ? or flash another CLR ?

Placing anything on a processor is a *pretty* stupid idea.

Re:Security ?

[MoonFog]

It's basically just firmware, not hardware in any way. How do you upgrade the firmware for your DVD rom ?

Re:Security ?

[Anonymous Coward]

My DVD firmware does not run any code. The CLR...

Re:Security ?

[FrYGuY101]

Question. What happened when the f00f flaw was discovered for the Pentium?

Yep, that's right, you had to buy another processor.

The X86 instruction set isn't somehow immune to flaws.

Re:Security ?

[Anonymous Coward]

wrong! f00f could be fixed with software (look at the linux F00F bug detection and workaround ...)

Re:Security ?

[Anonymous Coward]

Uh, then couldn't any potential flaw with this chip also be fixed with software, further proving the point?

Re:Security ?

[Anonymous Coward]

No. THIS bug (the f00f bug) could be fixed via software. Any random bug could not.

It doesn't prove anything either way.

Re:Security ?

[lithron]

Question. What happened when the f00f flaw was discovered for the Pentium?

Yep, that's right, you had to buy another processor.

Most vendors offered patches to the operating system that rendered the f00f bug moot. Feel free to grep your Linux source tree if you don't believe me. Purchasing a new CPU was also a fix, albiet a costly one compared to applying a small patch.

Re:Security ?

[Anonymous Coward]

Most vendors offered patches to the operating system that rendered the f00f bug moot. Feel free to grep your Linux source tree if you don't believe me. Purchasing a new CPU was also a fix, albiet a costly one compared to applying a small patch.


Ignoring the whole thing worked amazingly well also.

Re:Security ?

[Anonymous Coward]

Given the history of Intel, IBM and AMD (pretty much the big names in desktop processor manufacturers) and the history of Microsoft, which one do you think I trust less? Bugs happen, but some company (read: Microsoft) is much more prone to creating bugs.

Re:Security ?

[Anonymous Coward]

From the previously linked page:

The fact that this fix works is only caused by the fact that the instruction fault (and nothing performance-critical) happens to be before the page fault in the IDT. If it hadn't been, Intel would be in some trouble.

A copy of...?

[DenDave]

Gee I though Gumsticks were already mainstream... oh.. but of course http://www.oreillynet.com/pub/wlg/4552

these thingys aren't from Redmond...

dang it.. too late...

Too many opcodes?

[hummassa]

Like the x86?

Re:Too many opcodes?

[lintux]

Yeah. Don't x86 CPU's convert those opcodes to simpler ones before executing them these days?

Re:Too many opcodes?

[Anonymous Coward]

Yes, and we all know that x86 is such an efficient design... NOT
There is always a owerhead in translating instructions, therefore a pure RISC processor will always be faster than a corresponding CISC->RISC processor.

Re:Too many opcodes?

[Short Circuit]

Yes. And x86 isn't the only architecture that does that. POWER and PowerPC. Actually, I think it's standard practice.

Re:Too many opcodes?

[runderwo]

You're absolutely wrong. First of all, a pure RISC processor needs a much larger instruction cache to avoid misses, because the instructions are usually fixed width to make the decoder simple, compared to variable width instructions which make the most common instructions very short. You also need more memory bandwidth to fetch the larger instructions in the first place.

Furthermore, the Pentium 4 has what they refer to as a "trace cache" which caches the already-decoded instructions - only if this cache is missed does the instruction decoder have to become involved.

You'd be hard pressed to say that instruction decoding has any significant overhead anymore in Intel processors.

Mod Parent Up

[Anonymous Coward]

How can this be over-rated??

Re:Mod Parent Up

[Anonymous Coward]

Oh come on!

Stop promoting your own posts and get rid of that stupid moderation shit.

Re:Mod Parent Up

[Anonymous Coward]

It's overrated because someone wanted to take a point away without being metamoderated.

I say fuck all the moderators. I metamod everything as unfair.

Blue device of death

[Anonymous Coward]

Clippy turns you off.
A drm hardware dream.

Re:Blue device of death

[antoy]

Wow, you have managed to compress all Microsoft stereotypes in a 12 word joke! Of course, none of it is relevant, but don't let that stop you!

Try for a haiku next time.

Re:Blue device of death

[x-caiver]

At least Clippy isn't turning you on!
That is an entirely different type of dream...

Actually, it's an ARM7

[david.given]

According to the products page on their website:

• 384K of SRAM, single cycle access
• 27 MHz ARM7TDMI
• FBGA chip form
• ~450,000 instructions per second
• 4MB non volatile flash
• 1.8-volt core, 3.3-volt I/O
• 32768 Hz real-time clock
• 32-pin pinout, including 24 GPIO ports multiplexed with other functions (8 VTU ports, dual serial ports, SPI, and USB port)
• SPI and I2C interfaces

I assume FBGA is a typo for FPGA. This thing sounds suspiciously similar to one of those standard FPGAs with a built-in ARM7 core.

It actually sounds like quite a nice little embedded system, a kind of grown-up Basic STAMP. I expect that the .net VM is in ROM; on start-up the FPGA is probably bootstrapped from it. I wonder if it would be possible to replace it with a real operating system?

Re:Actually, it's an ARM7

[NoseBag]

...or just a MicroChip PIC device.

Re:Actually, it's an ARM7

[Pemdas]

I assume FBGA is a typo for FPGA.

When referring to packaging, FBGA is usually Fine Ball Grid Array. I really doubt it's a typo. From the programmers point of view, the package virtually never significant.

Overall, this sounds remarkably similar to picoJava, which, last I checked, was going nowhere, and for good reason.

Designing bytecode formats for VMs is not really the same as designing opcodes for microprocessors -- shoehorning hardware that way is painful and generally results in less elegant, more expensive designs.

OTOH, the bytecodes in question aren't really significantly worse than, say, x86, and look where that is today...

Re:Actually, it's an ARM7

[Anonymous Coward]

it is FBGA packaging

Re:Actually, it's an ARM7

[ceeam]

27 MHz ARM7TDMI, 450,000 instructions per second

Wait a minute, does it spend _60_ clocks per instruction on average???

(Yes, I understand that supposedly they are a bit higher level than even x86, but still....)

Re:Actually, it's an ARM7

[mlock]

On the website is also written

The .netcpu CPU Module does not include a development board, a power regulator, or a serial interface. For a full development kit, check out the .netcpu development kits.

Preorders for the .netcpu CPU Module will begin shipping on 12/31/2004.

$199.99

two hundred DOLLARS! two HUNDRED dollars! TWO HUNDRED DOLLARS!
I know cheaper CPU's ...
(eg http://www.atmel.com/dyn/products/product_card.asp ?part_id=2983 - 200MHz, USB, Ethernet, ...)

(Yes, "Music with Rocks in" :-)

Re:Actually, it's an ARM7

[Vulcann]

I expect that the .net VM is in ROM...

Squeezing a strapped down portion of a platform (or OS) into ROM isnt particularly new. If fact, Linux has a version of this : http://www.linuxbios.org/

Re:Actually, it's an ARM7

[hattig]

I expect that is .NET instructions per second, not ARM instructions per second.

So yes, 60 clocks to perform one .NET instruction.

Part of me thinks "Sod .NET, write in native ARM assembly and get something that is up to 60 times faster".

0.45 MIPS is about as fast as a 4MHz 68000, or a 16MHz Z80 variant. And there are a lot more embedded programmers who knows these, and ARM, than .NET. And they are perfectly happy not using Visual Studio. All this hardware will do is bring Visual Studio weenie programming to the embedded space. Hurrah for your .NET washing machine boiling your delicate wool wash, and your .NET freezer defrosting every 2 days.

Re:Actually, it's an ARM7

[arkanes]

Maybe it's 450,000 .NET opcodes per second. 60 instructions per opcode would be pretty good, actually.

Re:Actually, it's an ARM7

[jrumney]

27 MHz ARM7TDMI

The .NET Compact Framework crawls on a 400MHz ARM9 chip. What sort of applications are they intending this thing to be used for?

Re:Actually, it's an ARM7

[Anonymous Coward]

FBGA = Fine (pitch) Ball Grid Array

Who cares what packages you are using in that product anyway?

Re:Actually, it's an ARM7

[XMyth]

That's strange...my .NET apps run perfectly fine on my 200mhz iPaq (ARM).

Re:Actually, it's an ARM7

[Anonymous Coward]

You can also get a similar sized ARM7 (the UNC20 http://www.unc20.net/) with Linux and 16 Mbytes of SDRAM and 2 Mbytes of Flash for a fraction of that price. It also includes Ethernet as an interface and runs Mika, a nice Javabased Interpreter.

Re:Actually, it's an ARM7

[tjw]

It actually sounds like quite a nice little embedded system, a kind of grown-up Basic STAMP.

Parallax already has a STAMP that you program in Java alled the Javelin. I don't know if it makes it more grown up or not though.

I've had a couple of these stamps over the years and they're great for hobbiests like me, but they'll never be widely used because they cost too damn much for mass production. I think this .netcpu thing will be in the same boat.

It seems like the only industrial use of these parts would be to make a functional prototype really quickly. Then rewrite everything in C for an Amtel part for production.

Virtual Java Virtual Machine

[L3WKW4RM]

Where is the hardware-implemented JVM we've been promised for years and years? Not like this gloified BASIC stamp, running an implementation on the .NET runtime in software, but a real hardware implementation that runs bytecode natively.

Re:Virtual Java Virtual Machine

[mukund]

This has been available for a long time with open access to the design from Sun as the picoJava CPU core. It was not an economically viable CPU and I think this's one of the reasons why Sun released it.

Re:Virtual Java Virtual Machine

[Anonymous Coward]

Lots of people have JVMs on FPGA chips. Try some of these:

http://www.jopdesign.com/index.jsp

http://www.csse.uwa.edu.au/~jamesp/thesis.pdf

http://crpit.com/confpapers/CRPITV26Parnis.pdf

http://www.iiit.net/research/vlsi/pub/JVM_on_FPG A. pdf

I think it's about 3rd year electronics engineering level to create some kind of FPGA based virtual machine so there are lots and lots of FPGA projects out there on the net. The day of open-source hardware is already here.

And people are putting other VMs on FPGA like Squeak Smalltalk. Again, virtual machine technology is also open source. Look at citeseer to find examples of people designing their own vms.

Re:Virtual Java Virtual Machine

[tigeba]

I think you want this: http://www.jstik.com/

Re:Virtual Java Virtual Machine

[upsidedown_duck]

It was not an economically viable CPU and I think this's one of the reasons why Sun released it.

I have seen advertisements for other embedded JVM chips, but I've never seen one in person. This was several years ago, so who knows if they are still around.

Regardless, Sun has tried and shelved two new chip architectures in the last decade: picoJava (JVM) and MAJC (VLIW). I suppose it shouldn't be much of a surpise when/if this .NET CPU (JVM) and the Itanium (VLIW) go the way of the dodo. And the MAJC wasn't a dud, either (something like several gflops at 500MHz).

Another "Innovation" from Microsoft?

[nathanh]

Isn't this exactly like the Java CPU that Sun was selling a few years back? And it was simply a close relative of the Lisp processors from the 80s.

C#, Java. .Net, J2EE. CLR, JVM. .NET CPU, Java CPU. So should we expect Microsoft to simply repeat everything that Sun did with Java? If so, wake me up when they declare they're going to release CLR under an open source license.

Re:Another "Innovation" from Microsoft?

[sosume]

If so, wake me up when they declare they're going to release CLR under an open source license.

*riiiiiiing* wake up call ... its called Rotor, released by Microsoft a few years ago and it runs on FreeBSD.....
Well, maybe not your definition of open source (no GPL or BSD license but Shared Source) but remember open != free as in beer

Re:Another "Innovation" from Microsoft?

[oxygene2k2]

shared source would be equivalent to SCSL.. SUN already upgraded to the "java research license" with jdk1.5 (still not OSI, but definitely nicer than SCSL)

so, the race for the first official VM and DK for one of the languages under an OSI approved license is still undecided

Re:Another "Innovation" from Microsoft?

[WJMoore]

Here's a link if you're interested: Rotor

Re:Another "Innovation" from Microsoft?

[YU Nicks NE Way]

You do realize that Microsoft isn't behind this particular chip, don't you?

Oh, wait, silly me -- that would have required actually reading the article summary on the front page. Never mind.

Re:Another "Innovation" from Microsoft?

[Homburg]

Well, given that Open Source is a trade mark, it doesn't matter what your or my definition of it is - Microsoft's shared source isn't open, end of story.

Re:Another "Innovation" from Microsoft?

[Anonymous Coward]

open != free as in beer

Wrong.

Re:Another "Innovation" from Microsoft?

[Anonymous Coward]

Open Source is a trade mark

Wrong.

Boring compared to... the Brainf*ck CPU!

[quigonn]

http://www.clifford.at/bfcpu/bfcpu.html This piece of hardware is tres cool, as it implements the _complete_ set of Brainf*ck instructions as native instruction set.

Re:Boring compared to... the Brainf*ck CPU!

[zxv]

ALL 8 OF THEM!? :O

Re:Boring compared to... the Brainf*ck CPU!

[Anonymous Coward]

That is not the original brainfuck - this compiler processes more than six tokens !

Re:Boring compared to... the Brainf*ck CPU!

[Vulcann]

hmm.... MICROSOFT .Net based CPU ....wouldnt that qualify sufficiently to be a BrainF**K CPU ? :-D

Re:Boring compared to... the Brainf*ck CPU!

[quigonn]

What compiler? I was talking about the CPU.

C&D

[News for nerds]

Is it OK for them to use the '.NET' in the CPU name instead of ILCPU or ECMAwhatCPU?

In addition, their 4-color windows symbol resembles the Windows symbol just too much.

Re:C&D

[a20vertigo]

Well, if they're payin' the license fee to MS, sure thing. Or maybe MS is giving them license to use those images in exchange for developing stuff that supports their cause. Either way, you can bet there's some lawyering going on.

Off this specific topic, but I can (indirectly) run Java code on Lego Mindstorms bricks. And they're what, $200 with some attachment devices? And hey, that makes me smile 'cause I've got one :D

Ahnetkpuh?

[martin-boundary]

Ahnetkpu? Is this an Elder God?

Re:Ahnetkpuh?

[saldek]

Let's see...

Ahnetkpu! Ahnetkpu! Ahnetkpu!

Nah, just a crappy marketi[NO CARRIER]

Re:Ahnetkpuh?

[Anonymous Coward]

Sounds more like an Egyptian pharaoh to me. Maybe there's someone of Egyptian descent in the company?

Re:Actually, it's an ARM7 in FBGA

[jcdr]

No. This is not a typo.

FBGA = fine BALL grid array
FPGA = fine PIN grid array

FBGA is very common for small processors and chipset. FPGA is not so much used but for old CPU as now Intel uses LGA and AMD uses uOPGA

Use Google to get more informations about FBGA.

Remember ROM Basic...

[Gopal.V]

Yeah, this is very much like ROM Basic.

Looks like this idea's been around for god knows how long ... So much for innovation, we seem to be going backwards here ?.

This is a plug , but I've been working on a .NET CLR which can be trimmed down to around 400k (for a full opcode set, no less !!) for the last 3 years.

Re:Remember ROM Basic...

[Anonymous Coward]

When MS asked the question "wWhere do you want to go today?" the answer has always been "Yesterday!"

Re:Remember ROM Basic...

[antoy]

That CLR of yours sounds very interesting. It will surely make packaging .net desktop applications easier and more convenient than including the the gigantic dotnetfx.exe with everything. How's it going? Anything we can see? Will it be a commercial product?

Re:Remember ROM Basic...

[stuffduff]

How about the Z80 Basic Interpreter?

Hey, is there any one going to make me a PERL CPU?

[afa]

If so, I'm gonna buy one. Why the big people are always thinking their damnned 'JVM' or 'C#' etc. Who can pay a little attention to our common persons who just want to boost their CGIs.

That's funny

[le_defaut_tragique]

Check out the company website, and Google them. I just did and it turns up that this company was founded on Oct10.2k4ce by Mark Phillips. A Google turned up... the company website, the original submission, and a couple other press releases. this is their only product, and they made it in two months.

Microsoft's only connection with them is that Mark Phillips guy, who, when googled investigatively, appears to have founded A Dot Corporation in Apr.2k3ce and they were involved in... SPOT Watch technology and claim microsoft to be a business partner (spotcorporation.com).

So is Mark Phillips using his work with microsoft's SPOT developer team to create something to market under a different name? Both companies list only Mark Phillips as founder and, in fact, confirmed employee, although one site listed A Dot as having 24 employees.

Yeah, so that's funny...

Re:That's funny

[Anonymous Coward]

Dude, while I know you're trying to be cool looking and all, I think we all already assumed you were talking about the common-frickin-era!

Re:That's funny

[Anonymous Coward]

friggin weirdo mac user... wtf is this MMM.[dumb] date-format supposed to be?

--Robert

Re:That's funny

[DAldredge]

It shows he is so afraid of offending people that he has to point out he is using Common Era instead of BC/AD.

Re:That's funny

[RevAaron]

This isn't much different than a SPOT CPU. The same setup is running in the SPOT watches. This is a spin-off of that technology, allowing hobbyists and others that would want to play with a .NET CPU to do so. I'm not sure waht kind of relationship this company has with microsoft, but you bet it's something close to have been able to have a product two months after it's founding. But that is what is going on, from what I've read.

Re:Actually, it's an ARM7 in FBGA

[mvdw]

The OP probably meant FPGA as Field Programmable Gate Array.

Hardware independence?

[linebackn]

It boggles my mind, every time someone comes out with a bytecoded language in order to attempt to achieve hardware independence, someone comes along and creates hardware to execute it! Thus defeating the original purpose.

Of course people see the need for hardware acceleration because interpreted or even JIT compiled bytecode languages are always going to be slower than precompiled native binaries.

Re:Hardware independence?

[hal2814]

Running bytecode will always be somewhat slower than native binary, but Sun has done a good job of getting most of the overhead out of the running code and into the VM startup. Most overhead people experience now with Java isn't from the VM at all but from the constraints Sun puts on the Java language specification (exs. ALL arrays must be bounds-checked, dynamically allocated memory must be garbage collected). C,C++,BASIC,etc. do not have these requirements built into their language specification and therefore their compiled code has a leg up if the user decides not to implementthese features.

With the vast improvements being made over the VM's design in Java's case, I wonder if a chip that natively runs the code would really be "hardware acceleration" vs. a stock chip with a good VM.

Re:Hardware independence?

[Lisandro]

A CPU that can run bytecode directly could be very interesting for embeeded developement - if the price and power consumptions are low enough, otherwise writting assembler for PICs/CPUs is still the way to go.

Re:Hardware independence?

[spongman]

Thus defeating the original purpose

How, exactly, does having a CPU able to natively run a VM's bytecode/IL reduce that VM's hardware independance?

Re:Hardware independence?

[Anonymous Coward]

you are all idiots

Virus/trojan/spyware/malware

[Anonymous Coward]

Thank you MICROSOF~1
Now I will really be able to write my viree once in .NET assembly and spread them anywhere.

Glorious news indeed. Right now I've installed 'silent virouses' on about 17k windows boxens around the net. These viree silently wait wihtout doing anything besides spread themselves in binaries. Every now and then, they check a specific URL on a free WWW site, which may contain commands for them to exe. Until now the only cmd I've told them to run is report infected machines. Apart from that, I'm waiting for the right time to command them to do something big.

btw, if you're interested in paying me to put these viree do something for you, search goodle for "red bearded pirate".

Re:Virus/trojan/spyware/malware

[maxwell demon]

Thank you MICROSOF~1

Sorry, you got that wrong. It's MICROS~1. Only 8 characters including the ~1.

developers x3

[Anonymous Coward]

Not to nitpick, but shouldn't all .Net articles be on

http://developers.developers.developers.slashdot.o rg/???

...and from the zone file:

developers.developers.developers
CNAME bigsteves.anti.perspirant.net.

Mono is a heavyweight VM

[Anonymous Coward]

Dude, they can't even port it properly to PPC (exception handling is b0rked).

But that doesn't stop Novell from pushing out Cocoa# screenshots with the intepreter and benchmarks from the broken JIT together as if it's all done. (after all show me a benchmark which triggers an exception).

They're out to sell stuff - they'll "bend" the truth the way it suits them. Can't blame them either, 90% of all companies do it :)

Expensive and slow

[Anonymous Coward]

This is a bizarre piece of technology

- It's too expensive for volume marketing
- It's too slow for anything useful.

When you thing of what you can do in C on an AVR micro with GCC-AVR for a fraction of the cost (ie 10%) this thing is insane.

It looks like someone is trying to use .NET to gain a profile in the market - or they are using some kind of grant from Microsoft and this is one of the key milestones they had to meet.

"We are .not copying java"

[Anonymous Coward]

http://www.arm.com/products/solutions/Jazelle.html

http://www.netsilicon.com/products/netarmprocess or s/ns9775.jsp

http://www.developer.com/java/other/article.php/ 61 0041

http://www.ptsc.com/products/images/mpu.pdf

http://www.jopdesign.com/ (GPL'ed FPGA java cpu)

http://www.kiffer.be/k/products.html (?)

So will .net follow java everywhere without any own original ideas?

Re:"We are .not copying java"

[Anonymous Coward]

Wow, you just new to the industry? These "insightful" facts clearly demonstrate that. It's funny how Microsoft is "copying" Java, when the same lame argument could be placed on Linux for copying Windows GUI, installer, office suite, etc. It just doesn't work that way. In my heart of hearts I wish Java had anything as remotely powerful as Visual Studio .Net for C#. The best I've ever come to it using Java is WebSphere Studio 5.1 and frankly it's still just second class. Oh, and C# is a better language by several factors.

Re:"We are .not copying java"

[Anonymous Coward]

Is that because with Visual Studio, you don't actually have to remember any of the methods and M$ tries to do it all for you?

Re:"We are .not copying java"

[Anonymous Coward]

For why C# is "better" see Hot Topics and the tests at http://www.geocities.com/petilon/csharp/index.html - lovely!

Re:"We are .not copying java"

[Triskele]

Add in the JINI family - and the lovely little TINI from DalSemi which I'm writing an embedded web server on (can't tell you what for, though sysadmins will love it when it ships). This thing runs a reasonable JVM with several serial, ethernet, etc on an 8051 (the thing they put in keyboards as microcontrollers)!

Re:"We are .not copying java"

[RevAaron]

Though Java didn't do it first. This Java chip this wasn't first done with Java- there have been Pascal p-code chips, Forth chips and even BASIC chips for longer than Java existed. The rest of Java is just Smalltalk crammed into an inferior syntax so that LCD (lowest common denominator) programmer converts could grok it, coming from C++.

So will Java follow Smalltalk + Forth everywhere without any of it's own original ideas?

Microsoft has plans

[tobybuk]

What do you do when you have a monolopy on a Market and you shareholders want ever increasing earnings? You either charge people more for your monolopy products or you take other peoples business.

I fully expect Microsoft to enter the CPU Market in the next 10 years with a .NET style chip. Same old story - Call it the Windows CPU, spurt on about how much better it is than Legacy x86/AMD64. Throw in some BS about security being better and then make sure you need a .NET chip to have the 'best' Windows Experience.

They'll clean up.

Seems like beefed up GameBoy Advanced Specs

[Anonymous Coward]

The CPU is a 16.78 MHz ARM7tdmi RISC processor. It is a 32-bit processor but can be switched to....

http://www.cs.rit.edu/~tjh8300/CowBite/CowBiteSpec Frame.htm

32 Pin DIL

[Anonymous Coward]

What you say is true.

I think whoever typed the original article confused the issue... I thought it was 32 pin DIL (Dual InLine sockets)

32-pin pinout, including 24 GPIO ports multiplexed with other functions (8 VTU ports, dual serial ports, SPI, and USB port)

Only the insane and the anti-hobbist would BGA up a 32 pin device.

Re:32 Pin DIL

[Suidae]

The CPU is an FBGA package which is connected to some memory and other support components, all packaged on a 32 pin DIP.

Duh ! ... It's an exe + bootstrap for JIT

[Gopal.V]

.NET is a JIT engine , it's designed explicitly for JIT'ing ... It does produce a .exe file which has a main which calls the Mscoree.dll with the current file and starts up the VM using the bytecode data. The EXE part is just bootstrap code , the rest is JIT'd .

Read this paper about how many hoops you have to go through to get a decent interpreter for .NET. And it blatantly ignores the _Main() x86 native code that's in the .exe files.

Re:Duh ! ... It's an exe + bootstrap for JIT

[winfx]

1) the .exe bootstraper is never executed in Windows server 2003 which recognizes .net apps natively

2)What the paper demonstrates is a small switch/case with few if (.net)
vs a big switch/case (jvm), how can that be hoops it's beyond me

Re:Duh ! ... It's an exe + bootstrap for JIT

[Anonymous Coward]

> 2)What the paper demonstrates is a small switch/case with few if (.net) vs a big switch/case (jvm), how can that be hoops it's beyond me

You mean having to invent an entire new instruction set because .NET's default format sucked for execution is not going through hoops ?. (at least that's the first part of the paper - ie "Why ?", you've hit on the "How" part).

Re:Duh ! ... It's an exe + bootstrap for JIT

[M.C. Hampster]

1) the .exe bootstraper is never executed in Windows server 2003 which recognizes .net apps natively

By that, I hope you mean that Windows Server 2003 recognizes the native .NET managed executable format, and so it properly loads it using the .NET runtime, which still must JIT the code, right?

Your statement kind of sounded like you were saying .NET code doesn't need to be JIT'ed to run in Windows 2003 which ain't true.

Re:Duh ! ... It's an exe + bootstrap for JIT

[winfx]

Yes, that's correct with a minor detail

.NET can be pre-JITed (converted to machine code once and use that image at runtime)
That will eliminate the JIT stage, but is a practice that is rarely applied, because you will lose all optimizations that JITer does at runtime.

Re:Duh ! ... It's an exe + bootstrap for JIT

[winfx]

I never say that JITer suck, just is difficult to write one. MS and Mono they have pretty good JITers.

The "Why?" part is obvious

1) Makes compilers for .NET easier to write
2) Makes dynamic code (code on the fly) easier to write

That means that the difficult things are concentrated in one service (or component) and not spread to all tools and applications

Re:Hey, is there any one going to make me a PERL C

[m50d]

If it supports CLR, it supports python via the IronPython project. If that works there's no reason you couldn't do the same for perl.

Re:Hey, is there any one going to make me a PERL C

[Anonymous Coward]

It's a ***Ing subset !!

IronPython needs remoting and Reflection.Emit magic to work ...

This ships with the engine minus the Basic Class Libraries ...

Re:Hey, is there any one going to make me a PERL C

[m50d]

Sorry to reply twice, but according to wikipedia there is a Perl .net compiler, so you can run your perl directly on this cpu. Can't find a link though.

Try research before guessing

[fasura]

I assume FBGA is a typo for FPGA. This thing sounds suspiciously similar to one of those standard FPGAs with a built-in ARM7 core.

Fine-Pitch Ball Grid Array (FBGA).

Re:Hey, is there any one going to make me a PERL C

[m50d]

Gah, replying for the third time, but: YES, .NET INCLUDES PERL! If you have the activestate perl installed when you install .net, it will set things up so you can use perl with .net. Then you can compile the perl to CLR and use it on this CPU. So go and buy it.

Mod parent troll

[Anonymous Coward]

check out the posting history
another immature slashdot poster

Hmm

[Anonymous Coward]

I thought java is always a valid choice. It does run on windows, doesn't it?

--Coder

Re:Hmm

[Tarwn]

Yes, but the SDK's, API's, etc. that are bundled into most of the industrial apps I have worked with are all tied very heavily to MS products (if they are available at all).

The biggest break I have seen are some of the HMIs that still run on Solaris boxes and the data interfaces that hook into them to send the data to historians (on Windows boces). In some cases there areexcel plugins that are sold with products that are given tie-ins to VBA. Or some trending apps I have seen actually have a VBA coding section similar to Office products. This doesn't give you a lot of room to use Java. Granted you could use it if you really wanted to go to the extra work (writing VBA/VB/.Net apps that were basically interfaces for the Java app into the VBA/VB/.Net interfaces) but in most cases you would end up having to rewrite large portions of the SDK's/etc and still have most of the work being done from that middle interface, with bare tie-in classes in Java that just hooked into the objects in the interface.

So yes it could be done, just as I could write an ERP system all by myself from the ground up in any given language. Whether it is practical or not is the real issue.

-T

New bugs?

[priestx]

They'll patch it before it gets out. "Shipped with new errors from Microsoft!"

Fast enough for XP?

[Anonymous Coward]

Will THIS be fast enough to run XP at a useable speed?

This 2.6 GHz P4 can't even run it fast enough to keep up with my typing speed. It seems that 3 times as fast would be minimum to make XP bearable.

Re:Fast enough for XP?

[Anonymous Coward]

I know I really shouldn't reply to this, but I've heard too many people complain that their system is too slow lately. If you were to build your system properly and keep the software clean there is no slowdown. I run an Athlon 1800+ and get great framerates in both Doom3 and HL2, just an example. My machine hasn't slowed down in months, it is because I don't do stupid things and I have the hardware matched to get the most out of every part.

Re:Fast enough for XP?

[Anonymous Coward]

You can stop spreading lies now.

If "XP" can't keep up with your typing then you ned to start looking for viruses or buggy programs taking up 99.999% of your CPU.

Pure marketing hype?

[clevershark]

You could sell a brick to many CTOs these days if you told them that it was ".NET enabled"... to update a cliche for our times, no one ever got fired for buying Microsoft, although many should.

Instruction set

[Anonymous Coward]

Surely if microsoft or sun team up with intel or amd they could provide similar and add extensions at the cpu level? e.g. MMX or something similar but for .net or java.

YOU BASTARD!

[Anonymous Coward]

How dare you show up Sun's chip and demonstrate favoritism towards anything Microsoft on Slashdot! We don't deal with "facts" here!

Re:YOU BASTARD!

[SILIZIUMM]

Did you mean Sun Chips?

Re:YOU BASTARD!

[aichpvee]

Has anyone here ever actually seen a sun fanboi on /.? Because I sure see enough micros~1 ones.

Re:YOU BASTARD!

[grahamlee]

I was a Sun fanboy until I noticed that Apple were doing better UNIX systems for cheaper. I still like the W1100z and W2100z though :-)

Re:YOU BASTARD!

[Anonymous Coward]

Is there any such thing as a Sun Fanboi? I've certainly never met one. Maybe Kowboi Kneel is one?

Visual ALL - M$ Programming Poster Child

[Anonymous Coward]

Why must M$ continue to push .NET like its some poster child of programming. The .NET attempt to take over the internet has failed.. Few to no companies use .NET for what M$ envisioned.

I'm sorry.. the minute you put VISUAL in front of any language, you just breed the next generation of lazy programmers.

Keep it real..

jP?

[Doc Ruby]

Where's my Java chip? They promised us cheap, safe little Java chips, with embedded DSPs and ethernet. I want one in my phone!

Re:jP?

[Alioth]

Dunno whether it's a Java chip (I suspect not) but my diminutive Nokia 6820 phone runs only Java applications.

Re:jP?

[NullProg]

TINI

Its been around for a while.
Enjoy,

Want even more performance?

[Anonymous Coward]

Take out the MS CLR chip from the design and program your app in C/C++, have your code executed DIRECTLY by the microprocessor.

Speed increase guaranteed!

You also get more room for hardware that's actually needed, like RAM.

Fate like the Java CPU

[max909]

Hmm, i feel like this system is going to have the similar fate like the Java CPU Announced some time ago.

Huh?!

[Anonymous Coward]

You are comparing characteristics in the logic domain (software) to hardware specs. How do you compare?!

comparing Apples and Margarine ?

[Gopal.V]

What you pasted for JVM was the engine specs and for this thing was the CPU/Embedded specs.

The guys haven't really given out WHAT the "embedded.net" runs - looks like it's about the same as what the embedded JVM runs (not the Java "chip"). It's not a ".NET" chip first off and secondly it's almost the same as those "jvm" embedded (ie 400k sdram for what I have) in features. Multi-threading is not really multi-threading either, it is a kind of co-operative environment.

It's really not the big badass ".NET" at all , despite the name and the endorsment.

Re:Hey, is there any one going to make me a PERL C

[keli]

Er.... nope....

It's still the plain Perl 5 written in C. The Active state Perl.Net just allows you to call .Net code from perl. (and maybe possibly to call perl code from .Net)... But the perl code is not being compiled to CLR.

What, the .NET chip doesn't have 640K of ram?

[Anonymous Coward]

Shouldn't that be enought for anyone?

CVM - where's the source code?

[Anonymous Coward]

I only see doc at the site.

(number OS projects) != (demand)

[Anonymous Coward]

Most OS projects are crap. Just like most closed-source, propietary projects are crap. It's just that we don't get to see the crappy closed-source projects.

So just because there are a lot of open-source projects for a certain platform or language being thrown out by their authors doesn't mean there's a demand for those projects.

Who was it, when asked why 90% of sci-fi was crap, replied, "Because 90% of everything is crap!"

Re:(number OS projects) != (demand)

[the angry liberal]

Who was it, when asked why 90% of sci-fi was crap, replied, "Because 90% of everything is crap!"

If 90% of everything is crap, that includes the collection of Linux open source project too, right? I would post AC too if my logic was as terribly flawed as your's appears to be.

Why is this a good thing?

[parvenu74]

One of the major *advantages* of .NET over Java is that a .NET is supposedly compiled to NATIVE CODE on the first run, not interpreted bytecode like java. That being the case, what good is the .NET runtime in hardware? Faster compilation? .NET's compilation doesn't take forever anyway -- it's not Gentoo for crying out loud...

Re:Why is this a good thing?

[Nautica]

Are you kidding me, Please do not post on ./ anymore. You have no idea on what you are talking about. .NET is a interpreted just like java! you idiot!

Re:Why is this a good thing?

[parvenu74]

Clearly you are not familiar with .NET, which is compliled to NATIVE CODE before it is run.

Re:Why is this a good thing?

[Nautica]

It is still interpreted, Java has the same using Jrocket from BEA..

CLR CPU

[nurb432]

This would be more accurate then calling it a .Net cpu...

Dont expect to see these in a general use pc anytime soon ( if at all )..

There are several 'dedicated language' chips out there, like for java and forth.. but none really catch on outside their little niche markets..

They may be neat, but not too practical.

Re:CLR CPU

[RevAaron]

Why would anyone expect this to be in a general use PC?

Also, what do you think .NET is? Sure, it's the CLR and some other stuff. But you don't call it a JVM Chip, you'd call it a Java Chip. Same thing.

The greatest technology hits of 1979, rerun today.

[human bean]

They weren't single chip (single board), but they did take bytecodes (and sometimes source code) over a bus and execute them. Sometimes it was BASIC, and sometimes it was PICK assembler, etc.

If this puppy ran bytecodes for a simple BASIC (Which I am sure could be arranged...) then it would be very close to the old MAI Basic-4 minis, which did this on a single(8x12) circuit card.

Makes me feel sort of, uhm, nostalgic...

Can you imagine...

[wjsteele]

you guessed it... a Beowolf Cluster of these?

Bill

Re:Hey, is there any one going to make me a PERL C

[Anonymous Coward]

Cool...a write-only cpu.

ALRIGHT!

[Jahf]

I can't wait to update my JavaStation to a .NETstation ...

yeah.

Re:Actually, it's an ARM7 in FBGA

[jcdr]

I don't think so. The specification say "FBGA chip form" and it's easy to 3 of them on the photos (The CPU and 2 flash memory). I see no relation to programmable chip.

More confusion

[poot_rootbeer]

Fantastic! Another market where Microsoft is slapping the ".NET" branding on everything, even though it has nothing to do with the top-level domain of the same name, nor even with networks in general!

This is the worst marketing idea MS has had since they renamed their eHome concept to the generic sounding "Windows Media Center", without changing the names of any of the binaries.

450,000 CLR instructions/sec?

[LordByronStyrofoam]

450,000 CLR instructions, I'd guess. Each of these would chew through many ARM assembly instructions.

Rotor what ?

[Anonymous Coward]

Well, of course doing a port of an C++ application to non windows is feasible ;-)

The problem is availability of the application on that platform !

Here the VM is an application for me.

The main issue with .net is not technical but strategical for MS. They have build a clone, and they don't know how far this clone will endanger their core business supremacy ...

Let's see how will MS handle both ".net" supposed "compatibility" and Windows & Office official Lock-down at teh same time.

IMHO, MS will never support any non MS platform for .net or any other technology that can endanger theyr busines.

As a longterm consequence, either windows is dead (.net has won) or windows is still the #1 (and .net has join OS/2 at MS museum).

Let's see ...

.net Java

[damicatz]

.net is still inferior to Java. JIT is really pointless IMHO if you aren't bothering to make it cross-platform. And Mono doesn't count because you have to alter your code in a lot of cases to get it to compile in Mono.

Re:.net Java

[lupus-slash]

And I'm sure you filed a bug report about that, right?

Shenanigans!

[Anonymous Coward]

Their bid to hijack java is now complete. I, for one, welcome our new Virtual Machine and process stealing overlords.

Micro$hit makes me want to puke with every .Not article I see.

l8,
AC

Python CPU

[edcrypt]

Hey, I want a Python CPU!!!

Re:.net Java

[nberardi]

You have to modify you code alot in Java to get it to work in both Windows and Linux. Ever heard of code once test everywhere? That came from Java. The only problem that Mono has is the lack of IO support for Windows, but luckily if they didn't use the Windows GTK# there isn't going to be a problem, and these companies are activily developing technology to take advantage of the System.Windows interface to map to QT# and GTK#. Give it another year and you will see code once, test once. Something Java still is not able to claim.

Re:.net Java

[Triskele]

You have to modify you code alot in Java to get it to work in both Windows and Linux.

You do? That's news to me and a lot of other Java developers. Quit the FUD, Java really does do what it says on the packet about write once run anywhere. I regularly write on Windows and run on Linux for server code. Or write on Windows and run on Mac for GUI code. Hell I've even had little apps run without translation on a Palm and on a desktop.

That's some genius research right there

[PCM2]

Well, given that it says right in the article summary that the supposed .Net CPU runs the same software as SPOT smart watches, making a link between the CPU and SPOT watches didn't take too much research, did it?

And get this ... the SPOT watches were designed by Microsoft and require you to subscribe to an MSN service to use them. There, see?! Clear proof of a conspiracy linking the .Net CPU to Microsoft.

If ADot outsources to India...

[Anonymous Coward]

will its contractors be ... 'ADot'Heads ?

Shouldn't the title be...

[SuperKendall]

Developers, Developers, Developers!!!!!: .Net CPU

Nothing new under the sun...

[kaaona]

Long ago Western Digital made a Microengine that executed UCSD Pascal P-code. And Intel's iAPX-32 was supposed to execute Ada directly.

As hardware implementations, neither could economically evolve and survive.

Re:ctrl-alt-del keys? - Blue Screens

[fbartho]

Ok, first off, how the hell can you take a screenshot of a bluescreen unless you have a digital camera??? lol

Anyways, I know for a fact that bluescreen for XP exists, because I routinely trip one in an hp pavilion laptop of mine. 2.5 yrs old, off the shelf for 1.5 years, it has started to die in 1 out of 5 hardboots. It gets near to letting me enter a password, and sometimes after, sometimes before, I get a bluescreen coredump of bad_pool_header or page_fault_in_a_non-paged_area now (among others), truth be told, this laptop was badly dropped once, but its a beast, and all that was visible that happened was the breaking of the protective flap on a corner floppy drive bay (fixed) these blue screens and problems surrounding them did not start appearing till 6months later... do to the bluescreens and the fact that I have access to much nicer computers, I use it mainly for the purpose of a remote terminal to the computers I use primarily for work, along with backup HD space, music storage/playback, and wifi connectivity...

The upshot of using it like this means that I rarely stress the cpu, and so it runs much colder than it did when i used to have it as my main platform

You anti-MS jokers are missing out...

[Anonymous Coward]

There are too many negative comments about this, and this only proves the inexperience of the posters.

I'm an extremely experienced and competent embedded developer (C, C++, RTOS and not), and I believe this is truely meaninful advancement in technology.

Sure, it isn't a hardware CLR. What it is is a CLR that fits on a SoC. If any of you are familiar with the state of the highly embedded markets, you'll realize that this is not only a major hardware advancement, but a giant leap forward in embedded development productivity. Afterall, embedded developers are traditionally far harder to come by than VB or C# programmers. Now you can hire a C# programmer to do embedded work.

Is this a wise thing to do? Not traditionally, but the CLR and C# are so well thought out (I also have extensive .NET and C# experience) that I would actually hire these guys into an embedded position.

Get with it, people. This is truely an innovation in embedded development.

Microcodable RISC?

[Joseph_Daniel_Zukige]

Thinking of FORTH CPUs of yore.

And some old IBM CPUs, too, come to think of it -- writeable microcode, was it?

And i really don't care much for Borland's tools.

[Joseph_Daniel_Zukige]

So there.

Dans says NOT ENOUGH opcodes!

[Anonymous Coward]

For God's sake - put more opcodes in Parrot! Python can't run without anything less than 60,000 opcodes. Perl6 needs at least 5.3 billion opcodes at last count. And can't we have a way to change Parrot's calling convention 60 times per second?

Re:And i really don't care much for Borland's tool

[Bitchslap_69]

Fair enough, but you're certainly in a minority. Just out of curiosity, what development tools rise above the level of the "crap" you're complaining about?

CLR?

[teklob]

a subset of CLR? Does that mean it clears part of the screen?

Re:.net Java

[RevAaron]

.net is still inferior to Java. JIT is really pointless IMHO if you aren't bothering to make it cross-platform. And Mono doesn't count because you have to alter your code in a lot of cases to get it to compile in Mono.

What does that statement mean? The first sentence makes sense, though I'd disagree with it. At least .NET gives me access to the language I want to code on- one that is completely open and free, rather than restricting me to one piece of rubbish. And no, pointing me to that Java languages page does not suffice- no languages on that list have the same capabilities as .NET languages do, especially as far as two-way communication.

What do you think a JIT is? A JIT has nothing to do with being cross-platform. Having a JIT'd language doesn't make the language non cross-platform. Sure, the VM isn't cross-platform, but guess what? VMs generally aren't. That's why most intelligent languages (e.g. Smalltalk, Lisp, etc) make the VM tiny- 500 KB of binary or so- and do the rest in the target language. I can't help it that Sun can't tell the diference between a class library and a VM.