Slashdot Mirror

Hardware involves real and significant material costs. Big difference.

Really? Silly me, all I thought I needed was a testing board and some hardware descriptions.

Thanks to FPGAs, complete hardware designs can be written (in source code no less!), downloaded directly to the chip, tested, and then sold for a profit without ever speaking to a chip fab or hardware factory. And places like Pad2Pad allow for custom test boards to be built for a VERY low cost.

If the Open Graphics Hardware project needs a million smackers, then they better damn well have a shippable product on their hands.

See Rover FPGAs and RSC.
Future NASA space computers may not look like what most expect.

The FPGAs on Spirit and Opportunity seem to be overlooked. NASA's new Reconfigurable Scalable Computer (RSC) Project for space applications is exploring using FPGAs (instead of CPUs) which offer increased performance and radiation tolerance at a fraction of the power consumption.

Xilinx offer EasyPath option by testing for a customer-specific application. Customers use EasyPath customer specific FPGAs to achieve lower unit costs for volume production once they know their design is fixed and no longer requires the full programmability of an FPGA.

Another reason that people don't use FPGAs that much in consumer applications is the security of the IP on the FPGA. They are loaded on power-on with a PROM chip and it is a somewhat trivial task to read the entire contents of the FPGA at power-on. This would be a nightmare for companies like Nvidia and ATI, who value their custom hardware.

Fortunately, there are some companies that are incorporating flash memory on to their FPGAs instead of using the standard SRAM. The problem is that flash-based FPGAs are usually a few generations behind SRAM-based FPGAs in terms of die size (and henceforth storage space and speed).

I think that as flash-based securable FPGAs become more popular, cheaper, and less power consuming, we'll start to see cards for the computer that come with completely configurable hardware.

-Montag

Check out Lava at Xilinx, Lava at Chalmers, Hawk, the Hardware Design and Synthesis section of Haskell Application Papers on readscheme.org.
The links above lead to programs that are used by companies like Xilinx and Intel to help designers build better chips with existing technology. There are more interesting hardware approaches being investigated with Haskell. Two that come to mind immediately are quantum computing and dataflow-based simulations more related to the Lustre and Lucid languages. Though I do know of some unfinished research in the dataflow/hardware design area, I can't find any published papers at the moment.

One day I'll get around to buying a PCI card with a FPGA and use Haskell to turn it into a reprogrammable coprocessor. So many cool things to learn, so little time...

An interesting (and excellent) link on VHDL coding standards in a working environment is also available off that page: the European Space Agency's VHDL coding standard (available in PostScript format here http://tech-www.informatik.uni-hamburg.de/vhdl/doc /style_guide/ModelGuide.ps.gz).

The Xilinx WebPack is a great place to start - you get everything you need to take you from text-edited files to a binary image on one CD (or download from here: http://support.xilinx.com/support/download.htm). It even comes with a (very) cut-down version of Mentor Graphics' ModelSim and Xilinx's own synthesis tool, XST.

If you use Altera chips, they have a similar offering, called Quartus II Web Edition (http://www.altera.com/products/software/products/ quartus2web/sof-quarwebmain.html)

Speaking of text editors, (X)Emacs has a great VHDL mode that can beautify your code, create makefiles and manage your projects, available here:http://opensource.ethz.ch/emacs/vhdl-mode.htm l.

And now strange, I also would not mind putting together something like that, including the PCB, it's just I don't have any free time for that! Also, all laptops that I have are in working condition :-)

I am lazy and don't want to look this all up myself if someone else already knows, so here is a question: Left over from my undergrad project, I have a FPGA (XC4010XL from Xilinx) plus prototype board (XS40 from XESS), and have used the Xilinx foundation tools to code up some nice VHDL designs under NT. How can I do similar design work under Linux? Will gEDA suffice? Or, will I need a slew of other tools? Any VHDL environments? Or maybe Verilog? Something else? Are any of you super-smart /.-ers doing this purely under linux?

ISE Webpack is free and can program up to a Spartan 3 1500. This is probably why the target device is a Spartan 3. It can also program all other chips up to their respective 300,000 gate versions except Virtex 4. It comes with a limited Modelsim license that allows unlimited simulation but slows down after 1000 lines of HDL code.
I've have used these tools myself to design a VLC decoder compliant with MPEG-4. My 5000 lines of VHDL took about 3 minutes to simulate 2 microseconds on a P4-2.53. If the work was partitioned correctly, doing a 1.2-1.3 million gate deisgn using these tools is very practical.
I would pay roughly $200 for it simply because FPGAs with PCI interfaces are usually very expensive and none of them have video DACS. I would imagine any working with visualization could have great use for this. Say I wanted to look at data coming from a wireless receiver in realtime, I could send the data in bulk to the FPGA card and have it do all the processing and display a histogram, FFT, any parameter I wanted while placing very little burden on the host CPU.

Boards such as the Multimedia Board http://www.xilinx.com/products/boards/multimedia/ contain everything you would need. Not cheap though...

They have not put the whole thing on a PCI card, probably because it's even more fun to integrate a CPU core and build the whole system-on-chip on the FPGA while at it.

Cheers!

...classic toys are better! That's why I asked Santa for an FPGA System-On-a-Chip for Christmas! :-D

Now Open Cores would be great! But as long as we don't have a home chip manufacturing unit

Actually, given that this is FPGA based, perchance a more generic piece of hardware would be appropriate.

I've looked at FPGA development cards (usually several thousand dollars, but they're meant for DSP work generally). What we really need to go with the Verilog core they're designing is a modular add on card so you can either get a compatible reference design from someplace like Opencores.

We'd need several base designs with, say 1 to 4 onboard FPGAs. I am not a hardware designer, but I've done some research and at least from my perspective, I'd go with Xilinx since they're supporting all their Spartan3 with their freely downloadable WebPAK software which will support some rather large Spartan3 chips. I personally would like to have one card-edge connector version for PCI and another for ISA systems, but for the video package we'd need an AGP or PCI-X implementation first. A card edge socket for a backplane connection suite (VGA or audio or SCSI connectors for instance), possibly a memory daughtercard (either premounted memory for video or DIMM for using commercial main memory), maybe even a PC/104+ setup onboard would be nice.

Now, with the above, you could mix and match internal cards, output connectors, and daughtercards to create your audio card, or a video card, or just about anything you can think of and find or create the reference design for. Now, they DO make cards like this but they don't use the Spartan3 low cost chips, and they cost several hundred to several thousand dollars, and those boards that do meet a good chunk of even the basic requirements, like the development kit at Nuhorizons but there's no way to plug that into a PCI or AGP slot. Others are available, but use older tech, like the X84 ISA card. So, for the general hobbyist or designer, the company might be able to spin off the hardware design to general DSP or other type of work and help recoup more of the hardware cost that way.

Actually, for the (even more) adventerous, grabbing an available PCI backplane from Ebay or something would let you build your own computer from the ground up with the higher FPGA count units.

Now, mind you, this may not work well for a video card based reference design due to the higher bandwidth requirements.

Actually, given that this is FPGA based, perchance a more generic piece of hardware would be appropriate.

I've looked at FPGA development cards (usually several thousand dollars, but they're meant for DSP work generally). What we really need to go with the Verilog core they're designing is a modular add on card so you can either get a compatible reference design from someplace like Opencores.

We'd need several base designs with, say 1 to 4 onboard FPGAs. I am not a hardware designer, but I've done some research and at least from my perspective, I'd go with Xilinx since they're supporting all their Spartan3 with their freely downloadable WebPAK software which will support some rather large Spartan3 chips. I personally would like to have one card-edge connector version for PCI and another for ISA systems, but for the video package we'd need an AGP or PCI-X implementation first. A card edge socket for a backplane connection suite (VGA or audio or SCSI connectors for instance), possibly a memory daughtercard (either premounted memory for video or DIMM for using commercial main memory), maybe even a PC/104+ setup onboard would be nice.

Now, with the above, you could mix and match internal cards, output connectors, and daughtercards to create your audio card, or a video card, or just about anything you can think of and find or create the reference design for. Now, they DO make cards like this but they don't use the Spartan3 low cost chips, and they cost several hundred to several thousand dollars, and those boards that do meet a good chunk of even the basic requirements, like the development kit at Nuhorizons but there's no way to plug that into a PCI or AGP slot. Others are available, but use older tech, like the X84 ISA card. So, for the general hobbyist or designer, the company might be able to spin off the hardware design to general DSP or other type of work and help recoup more of the hardware cost that way.

Actually, for the (even more) adventerous, grabbing an available PCI backplane from Ebay or something would let you build your own computer from the ground up with the higher FPGA count units.

Now, mind you, this may not work well for a video card based reference design due to the higher bandwidth requirements.

Actually, given that this is FPGA based, perchance a more generic piece of hardware would be appropriate.

I've looked at FPGA development cards (usually several thousand dollars, but they're meant for DSP work generally). What we really need to go with the Verilog core they're designing is a modular add on card so you can either get a compatible reference design from someplace like Opencores.

We'd need several base designs with, say 1 to 4 onboard FPGAs. I am not a hardware designer, but I've done some research and at least from my perspective, I'd go with Xilinx since they're supporting all their Spartan3 with their freely downloadable WebPAK software which will support some rather large Spartan3 chips. I personally would like to have one card-edge connector version for PCI and another for ISA systems, but for the video package we'd need an AGP or PCI-X implementation first. A card edge socket for a backplane connection suite (VGA or audio or SCSI connectors for instance), possibly a memory daughtercard (either premounted memory for video or DIMM for using commercial main memory), maybe even a PC/104+ setup onboard would be nice.

Now, with the above, you could mix and match internal cards, output connectors, and daughtercards to create your audio card, or a video card, or just about anything you can think of and find or create the reference design for. Now, they DO make cards like this but they don't use the Spartan3 low cost chips, and they cost several hundred to several thousand dollars, and those boards that do meet a good chunk of even the basic requirements, like the development kit at Nuhorizons but there's no way to plug that into a PCI or AGP slot. Others are available, but use older tech, like the X84 ISA card. So, for the general hobbyist or designer, the company might be able to spin off the hardware design to general DSP or other type of work and help recoup more of the hardware cost that way.

Actually, for the (even more) adventerous, grabbing an available PCI backplane from Ebay or something would let you build your own computer from the ground up with the higher FPGA count units.

Now, mind you, this may not work well for a video card based reference design due to the higher bandwidth requirements.

Actually, given that this is FPGA based, perchance a more generic piece of hardware would be appropriate.

I've looked at FPGA development cards (usually several thousand dollars, but they're meant for DSP work generally). What we really need to go with the Verilog core they're designing is a modular add on card so you can either get a compatible reference design from someplace like Opencores.

We'd need several base designs with, say 1 to 4 onboard FPGAs. I am not a hardware designer, but I've done some research and at least from my perspective, I'd go with Xilinx since they're supporting all their Spartan3 with their freely downloadable WebPAK software which will support some rather large Spartan3 chips. I personally would like to have one card-edge connector version for PCI and another for ISA systems, but for the video package we'd need an AGP or PCI-X implementation first. A card edge socket for a backplane connection suite (VGA or audio or SCSI connectors for instance), possibly a memory daughtercard (either premounted memory for video or DIMM for using commercial main memory), maybe even a PC/104+ setup onboard would be nice.

Now, with the above, you could mix and match internal cards, output connectors, and daughtercards to create your audio card, or a video card, or just about anything you can think of and find or create the reference design for. Now, they DO make cards like this but they don't use the Spartan3 low cost chips, and they cost several hundred to several thousand dollars, and those boards that do meet a good chunk of even the basic requirements, like the development kit at Nuhorizons but there's no way to plug that into a PCI or AGP slot. Others are available, but use older tech, like the X84 ISA card. So, for the general hobbyist or designer, the company might be able to spin off the hardware design to general DSP or other type of work and help recoup more of the hardware cost that way.

Actually, for the (even more) adventerous, grabbing an available PCI backplane from Ebay or something would let you build your own computer from the ground up with the higher FPGA count units.

Now, mind you, this may not work well for a video card based reference design due to the higher bandwidth requirements.

Actually, given that this is FPGA based, perchance a more generic piece of hardware would be appropriate.

I've looked at FPGA development cards (usually several thousand dollars, but they're meant for DSP work generally). What we really need to go with the Verilog core they're designing is a modular add on card so you can either get a compatible reference design from someplace like Opencores.

We'd need several base designs with, say 1 to 4 onboard FPGAs. I am not a hardware designer, but I've done some research and at least from my perspective, I'd go with Xilinx since they're supporting all their Spartan3 with their freely downloadable WebPAK software which will support some rather large Spartan3 chips. I personally would like to have one card-edge connector version for PCI and another for ISA systems, but for the video package we'd need an AGP or PCI-X implementation first. A card edge socket for a backplane connection suite (VGA or audio or SCSI connectors for instance), possibly a memory daughtercard (either premounted memory for video or DIMM for using commercial main memory), maybe even a PC/104+ setup onboard would be nice.

Now, with the above, you could mix and match internal cards, output connectors, and daughtercards to create your audio card, or a video card, or just about anything you can think of and find or create the reference design for. Now, they DO make cards like this but they don't use the Spartan3 low cost chips, and they cost several hundred to several thousand dollars, and those boards that do meet a good chunk of even the basic requirements, like the development kit at Nuhorizons but there's no way to plug that into a PCI or AGP slot. Others are available, but use older tech, like the X84 ISA card. So, for the general hobbyist or designer, the company might be able to spin off the hardware design to general DSP or other type of work and help recoup more of the hardware cost that way.

Actually, for the (even more) adventerous, grabbing an available PCI backplane from Ebay or something would let you build your own computer from the ground up with the higher FPGA count units.

Now, mind you, this may not work well for a video card based reference design due to the higher bandwidth requirements.

I don't know much about FPGAs -- what will $30 get you?

That together with some DDR SDRAM and suitable video DACs should be sufficient for a fairly decently performing 2D card.

I don't know much about FPGAs -- what will $30 get you?

That together with some DDR SDRAM and suitable video DACs should be sufficient for a fairly decently performing 2D card.

What's really cool these days is the amount of free software available for DIY electronics. I'm not talking about free as in GNU, but free as in beer. Want to design your own digital chips? Get your free WebPack FPGA design software. Want to do your own circuit boards? Get your free CAD software... Want to program your embedded product? Most chip companies have free development tools.

You might be interested in FPGA's Which are practiaclly 'universal chips' you simply have a chunk of RAM (EEPROM) next to it that sets up the logic gates within the chip. They're only ~£15 ($20) per chip, and you can load whatever you want into them to perform any function. More info at http://www.xilinx.com/ (One of the industry leaders)

I remember I took a computer systems architecture class in college where we took a programmable Xilinx board, two "customized" NES controllers, and a monitor and created our own game system. We emulated our own MIPS processor, created a compiler, wrote our own OS and filesystem and then wrote Pong for the system and a driver for the monitor and the controllers. Needless to say, it was a learning experience. Timing was a pain. We had about 9 weeks to do it all and I don't think we finished. Most of the parts were completed but we didn't have time to integrate everything. I wish we did, it would be nice to say I developed my own "console" on my resumè ;-)

Xilinx was already offering chips that combined a processor core with an FPGA.

-jcr

Though not the same as this, the Xilinx Vertex II Pro combines an FPGA and PowerPC risc core on the same chip.

The Altera Excalibur does something similar with an ARM processor core and programmable logic.

Both of these have been around for a while...

If you think putting together a processor and an FPGA is revolutionary I think you need to look at Xilinx (Virtex II Pro) and Altera. Both of them have CPU + FPGA combinations. Xilinx is a PowerPC 405 core, Altera is an ARM9 (I think).

Xilinx also has an FPGA that has enough space to implement a full PowerPC 405 processor.

This to me indicates more of a simple evolution then revolution...

I do wonder how they deal with heat dissipation. :-)

Hardware manufacturers that need special hardware operations (IE MPEG-2 decoding) use dedicated, custom hardware for large volume production. Dynamically configurable hardware is expensive for large scales production, and small scale production will likely use FPGA for similar effect. I may be sceptical, but I doubt it'll catch on.

eBlocks don't run Linux, but this really small electronic logic blocks device sure does. How about an interface eBlock, that connects these simple sensor/actuator networks to the awesome power of a fully operational logic array?

eBlocks don't run Linux, but this really small electronic logic blocks device sure does. How about an interface eBlock, that connects these simple sensor/actuator networks to the awesome power of a fully operational logic array?

I'll shut up after this, promise. :-)

Multisecond RTT doesn't happen on anything but GPRS

I've seen it far too often on congested wifi networks. you easily get into a congested state with a crowded AP that forces lots of client waits for the DCF (i.e DIFS + padding, each in turn) and also induces lots of retransmission at the physical level due to collision with so many clients trying to talk to the same AP. Low power clients associated at the 1 or 2 Mbps rates drive this contention over the DCF even higher, severely punishing everyone associated.

The big conference venues are notoriously bad about this, as you often end up with 10-20+ people associated with a single access point. That is just too many, and the 802.11 MAC was never meant to handle that kind of load efficiently. It is a pretty good solution for the general case that simply can't cover all the edge cases (long shots, high client loads, noisy RF environments).

This type of situation results in really weird ping times, for example. I've seen fluctuations myself that go from 80ms, 120ms to 3s!, 2s!, etc. then back down to a few score milliseconds. That is the 802.11 MAC trying to cope with scenario's it was never designed to encounter.

I mentioned software radios in the first post because having access to timing and congestion control in the MAC would allow mesh boxes, clients, and AP's to make very significant performance enhancements for situations where they were needed. Why be forced to use a static, inflexible, proprietary hardware layer when you can have the open flexibility associated with software radio? (It's coming, just not soon enough :-) There are also extensions to the ad-hoc routing protocols (like passive monitor of route info between other clients in DSR) that could be supported if only the hardware was open enough to do so.

I don't want to bitch too much; we have come a long way from sub-megabit data via FHSS over 900Mhz. I just want the really good stuff to hurry up and get here already so that things like mesh networks, low latency/loss voice over IP, and highly available multipath/redundant network configurations can be enjoyed to their full potential. (software radio + multiple input / multiple output + intelligent network stacks that can handle a diverse and volatile network environment). ... and a pony!

Gratuitous links:
congestion problems at TechEd conference

congestion melt down at CeBIT

GNU Radio's software defined radio (SDR)

software defined radio on $2,000 of 'roids [it's a dev kit, but would work very well for almost any kind of project]

Xilinx have silicon with embedded PowerPC processors, BlockRam (chunks of pre-generated SRAM) and huge swathes of FPGA cells and interconnect. The chips have other abilities too - built-in 18-bit multipliers and communications channges (10 Gbps/channel, 20 channels!). All very cool stuff. Very expensive too :-(

I'm sort of surprised there aren't more FPGA-hackers than there appears to be. It's not hard to learn verilog (very similar to pascal), and despite what most FPGA designers will tell you, as long as you keep your mind focused on 'everything happens in parallel', a decent programmer can produce good FPGA code too. The start kits (300,000 gates, about enough for a hardware JPEG core and maybe a network MAC) are cheap (100 or so), and designing a processor is a pretty simple operation, and immensely gratifying :-)

Just my thoughts,

Krik

Xilinx have silicon with embedded PowerPC processors, BlockRam (chunks of pre-generated SRAM) and huge swathes of FPGA cells and interconnect. The chips have other abilities too - built-in 18-bit multipliers and communications channges (10 Gbps/channel, 20 channels!). All very cool stuff. Very expensive too :-(

I'm sort of surprised there aren't more FPGA-hackers than there appears to be. It's not hard to learn verilog (very similar to C), and despite what most FPGA designers will tell you, as long as you keep your mind focused on 'everything happens in parallel', a decent programmer can produce good FPGA code too. The start kits (300,000 gates, about enough for a hardware JPEG core and maybe a network MAC) are cheap (100 or so), and designing a processor is a pretty simple operation, and immensely gratifying :-)

Just my thoughts,

Simon

In couple years, when DRM will be ubiquitous, there will be a booming black market with "coprocessor" devices of this kind. Fueled by the abundance of out-of-work engineers and developers, whose job went to the East.

Nature seeks balance, in medium-to-long-term ignoring the wishes of the money-hungry CEOs. [insert yin-yang sign here]

Xilinx radiation-tolerant Virtex(TM) FPGAs are being used in the "main brain" of the rover vehicle, controlling the motors for the wheels, steering, arms, cameras and various instrumentation, enabling the vehicle to travel about the planet.

[Disclaimer: I work for this great company.]

Someone may have already said this or you may have already seen it, but Xilinx has a press release about having their chips (FPGAs of course) in the rovers. They're touting the FPGA as being used as the "main brain" so you may be right on.

http://www.xilinx.com/prs_rls/design_win/0412_mars rover.htm

A funny side note, I think they posted the press release on the same day as the Spirit failure. That's some good pr timing ;)

Didn't Fortinet already do this???

It's a tad hard to tell because Fortinet's website doesn't seem to be too technical. However, they repeatedly use the phrase "ASIC-accelerated". That is, they have a specially designed, non-reprogrammable chip which is utilized by their software which implements their system.

On the other hand, this doctor has taken a wholly programmable device (FPGA = Field Programmable Gate Array -- see Xilinx for examples) and implemented complicated, useful stuff on top of it. So, even if you find his product sucks, hey, you still have an FPGA, right? More importantly, you have a device built from off the shelf parts. I don't think those ASICs that Fortinet uses are made by too many people. Again, I could be wrong, since their site is fairly non-specific in this area.

Also, his paper claims that this device, upon detecting a suspicious message can either block it outright or let it pass. Either way, though, it uses windows messaging to tell the targetted user that he has something bad sent his way and it was/was not blocked. Fortinet makes no such claims that I see. I wager they just drop 'bad' packets.

A hundred million PCs are getting sold this year without SATA support ....

Just because you don't have SATA doesn't mean you can't add it. You're going to spend $280(approx) for a SSD, spend $600-1000 to populate it with memory, and about $50-200 for battery packs and perhaps a UPS, but you wouldn't also spend just $32 to add a couple Serial ATA ports ???

Serial ATA makes a lot of sense for a many reasons, the main one being that it's simply faster than parallel ATA. Today, SATA is 150 Mbytes/sec... only slightly faster than the fastest parallel ATA at 133 Mbytes/sec. But in designing a high speed SSD, I'm certainly not going to skimp and I'll definately use a SATA PHY chip that supports 300 Mbytes/sec, or the planned 600 Mbytes/sec if a PHY chip is available, or a pin-compatible version is planned. I'm also planning on primarily designing around reconfigurable FPGA-based hardware, which reduces a lot of the risk and development costs, and might allow me to populate the boards with faster PHY chips as they become available.

With 300 Mbyte/sec SATA and the plans for 600 Mbyte/sec down the road, AND many gigabytes of DDR DRAM media that can actually have sustained I/O at those speeds with virtually zero latency, parallel ATA is looking like quite a dinosaur. Of course, future motherboards will need PCI-X or some other faster bus to transfer these amazing speeds... but all that is coming soon.

Anyway, the main point is that you can pretty painlessly add SATA ports to your existing PC with an inexpensive card. And lots of inexpensive adaptors are available to retrofit "legacy" parallel ATA drives to SATA (which will likely be needed if motherboards start phasing out parallel ATA connectors... which is expected soon since all the new semiconductor processes don't provide 5 volt tolerant I/O pins anymore).

Don't get out much, do you?

This is relatively simple to do, and most of the major FPGA vendors offer "PCI development kits" which allow you to develop your own PCI card using their FPGAs. They're quite expensive, though, as they're aimed at OEMs.

The biggest problem in this, is that the compilers are propetary and expensive.

That would be why FPGA Vendors like Altera, Xilinx and Lattice all offer free versions of their FPGA software that will place-and-route most of their lower-cost devices.

Learning to "program" an FPGA isn't all that hard, it's just a different paradigm to program a sequential language. Interfacing the PCI bus is probably the hardest part, but most of the vendors will help you with that, too. They're in the market to sell as many FPGA chips as they can, after all...

This prototype media appliance runs completely on FPGAs and downloads new hardware configurations from a webserver. Applications include VoIP phone (w/ TCP I/P stack), MP3 player and, yes, Space Invaders... complete with sound and a touch screen.

Part of Xilinx's whole Internet Reconfigurable Logic initiative... update hardware over the network.

How about 4 IBM PowerPCs onboard a large Xilinx FPGA chip. That allows significant flexibility. The tradeoff is just how parallel is your application. Some prefer more silicon for reconfigurable gates to the fixed CPUs on-board that may not maximize silicon use/cycle during an application.

"...running on hardware lacks the visibility needed to do debug..."

There's a neat tool called Chipscope Pro that ships with the ML300 that acts as an internal logic analyzer. You can trigger on a real signal and get back a sample of signals occuring in the FPGA. Very usefull for debugging designs that you don't want to put the work in to simulate.

Overall though, I'll have to agree that thoroughly simulating your design is the best way to go. The Swift model of the PPC is really usefull in that regard, you can have more confidence that your design will work with the real processor.

The bulk of the cost of the ML300 is not in the FPGA. The peripherals on the board and the accessories in the kit constitute a lot of the price.

If you're interested in a "standalone" development board those are also available.

This kit comes with some really neat debugging features that hopefully help to ease the pain of debugging such a huge system. You can use ChipScope Pro to trigger on a real signal in the FPGA (like an interrupt) and send back sample data as waveforms. Basically a built in logic analyzer.

Also, the whole system can be simulated using a Smartmodel of the PPC, so you can see all the signals as simulation waveforms when running your software. And of course there's GDB w/ GUI for software.

The 3000 gate figure is just the nand gate equivalent count, in reality it only has 128 moderately configurable cells. By comparison the FPGAs that people use for anything slightly serious start at about 600 cells (~40k gates, the cells in FPGAs are more complex). In the stuff I do (image processing mostly) the smallest FPGA I use is a xcv1000e with 27,648 cells (~1.5M gates), and this is fairly out of date.

OTOH you really don't want an FPGA in a handheld, as they are very inefficient in terms of power/heat. They get way too hot to touch running in my computer (admittedly without active cooling, but that would be difficult to achieve in a handheld anyway), and suck a huge amount of power (some of our machines with weaker 300W PSUs can't provide enough current to boot with an FPGA card in). Admittedly the parts I use are more heavyweight than you'd think of putting in a mobile device, but even the smaller ones need alot of power.

Annapolis (a slightly odd FPGA platform vendor) do a PCMCIA card called the Wildcard which apparently works in an iPaq (plus adaptor), for example used in this project for wireless encryption project. But they cost $999 :)

There is no way you'll be reconfiguring your CPU any time soon,

An interesting exercise for undergrads. If he used a Virtex II Pro he'd have upto 4 power PC 405's on a single chip (not to mention all that reconfigurable logic).

Now some real nice research would be to see how to those processors could be combined with reconfigurable logic to give some real power.

point 1

Not only does MS screw its competitors, MS is causing a ruckus with its *partners*.

Kind of like killing the hen that lays golden eggs....

They have the muscle to be able to do this, but its baad bad business.

=====

point 2

Why wasn't Nvidia thinking ahead on this issue either?
They could use a CMOS-like chip containing the security key/algorhythm, that would be far less costly to produce, especially if its likely it will need to be changed in the future.

Maybe use a Complex Programmable Logic Device?

Yes MS kind of screwed Nvidia, but Nvidia appearing to be a "forward thinking" company... why did they get themselves in this position to begin with.

Is there a real PCB-layout tool available for Linux? No, I've looked and can't find one, not even a commercial tool.

Yes, there is a commercial VHDL/Verilog simulator (ModelSim), but they only have the high-end version ($$$$) available for Linux; the "Personal Edition" is only available for Windows (and it's still $5K).

If you do 8051 stuff, the Keil compiler (Windows-only) beats the pants off of SDCC, which hasn't been upated since last September.

And, as others have pointed out, there are no good audio apps available for Linux -- nothing like SawStudio, SoundForge, Cubase, ProTools, etc. (OK, I know, I should be doing my audio stuff on a Mac.)

I haven't found an equivalent for Apple's ColorSync for Linux. (OK, I know, I should be doing all of my photo editing stuff on a Mac.)

The solution to the problem, of course, is to get the various software vendors to target Linux. It's getting there: Xilinx supports their latest (non-free, non-cheap) tools under Wine, with native support expected next year. But, in parallel with urging vendors to support Linux, work needs to get done, so Windows it is.

--

"The Devil has many faces. Yours just fits." -- Swiz

anyone who has anything to do with a standards body nowadays know's that people try and hijack standards so their tech/patent gets into it

this way if you implement the standard you have to pay

you can't have an opensource MPEG 4 without paying 3million bucks when you distribute it and they call that a standard

ok real hardware and software

in terms of a kernel their is in My Humble Opinion

Linux

Open BSD

netbsd for every arch under the sun (joke included)

then we have the problem of hardware

Opencores provides some of the effort BUT my favorate is

LEON-1 VHDL model
- Functional SPARC compatible processor core integer unit. Runs on Altera, Mietec, Temic MG2, Xilinx. Developed for space missions. Implemented as a highly configurable, synthesisable GPL VHDL model.

Altera 10K200E FPGA or Xilinx XCV300 enable this you can also get a LCD and keyboard AMBA devices from www.gaisler.com

what I would like is a machine that you could say that the whole thing is opensource

regards

john jones