yep
cool thanks and i can solder , but if its surface mount you cant solder it to normal prototypeing board can you?
Chips
Well this is why we have a small range of dev boards so the chips are pre mounted with their I/O exposed for you to do what you want. The chips themselves (depending which one you are referring to) comes in following formats:
LQFP
TQFP
BGA
Fair enough
OK, sorry I mis-understood the question.
This is why we are giving away dev boards, so you don’t need to, all the IO is exposed which you can solder to easily. Therefore you should just be able to connect whatever hardware to it (along with any electronics to drive those components - clearly you are not going to be running any motors from the pins of a chip!) and then write your program in C/XC on our processor and get data from the IO, do something with it, and put it back out again or whatever it is you want to do.
I believe with our new L1 chip it is possible to do a bit of hand soldering if you are amazingly 1337 with a soldering iron as they have pins that stick out on the sides. I have seen people in the office do it.
**oh **
oh what i meant is , if we like the dev kit then wanted to buy just some of the chips to play around with thenm it makes it hard for the general hobbyist to use ur products
The general hobbyist
The general hobbyist probably doesnt need the capabilities these chips offer, but if you really did need 4 cores and 32 threads then you probably also have the capability to make PCBs at home.</p><p>Even if you cant make PCBs, you can get break out boards from the likes of sparkfun or futurlec and solder the TQFP package chips on (although I have been searching for the right pin count boards and cant find them).
Indeed
Indeed, if you want to use the raw chips themselves and design your own boards for something highly specific, this is most certainly do-able, and there are plenty of companies that can do this for you quite cheaply (especially if you are ordering say 3 -4 boards, the price comes down dramatically) - we have had a few members of our own community do this already! If you want to know more about this, as this is not my specialist area, ask on our community site Xlinkers.org in the forums as this problem has been solved there already! Hope that helps!
/pokeLooking forward to
/poke
Looking forward to those new dev boards 
Well feel free to have a roam around the tutorials for programming in XC etc so you can utilise them to full potential when we have got them to give out! Stay tuned!
Sorry for the slight delay, but we want to make them really usable for these type of projects, so a few mods being made. I will make an announcement of some shape of form when they are ready and the projects chosen!
I might be
interested, but the cost of the brains turns me away. Sorry.
I think you missed the free part…
Not sure if you read all of my post, but I will be choosing the most suitable/best ideas and giving away some of our kits for FREE so you can have the chance to make them!
So please feel free to post your idea anyhow and it might become a reality.
Out of interest, what would you consider a good price for a basic dev board, with one of our chips on it, with the I/O exposed so you can connect your own hardware to it etc? I am always interested to find out about budgets for hobbiests etc and might be able to consider this in future designs. Thanks for your interest/feedback, it is all useful.
Robot Idea
Project Title: Follower
Prject Description: I would make a robot which would follow me around the house and could carry small loads for me; maybe obey voice commands.
Number of Dev kits: 2
Photo Diary and maybe a video here and there
**Hi **
point cloud ranging
Using a laser and a cmos camera mounted on a robot it will take various points in a room and build up a point cloud to map the room out and broadcast the coordinates back to a central server. I have been looking for a processor powerfull enough to handle the data and it sounds like this would be ideal.
I would need one kit and would be willing to keep a video diary and i will also post the progress on my blog.
Good Price for Xmos stuff
I must admit that I’m not a good one to ask. Mainly because I don’t really got that much of an income. Maybe $20 per month. Pretty poor. But I it were up to me, I would say free. But in reallity, maybe $20 - $30 bucks.
An update to my thoughts on an xmos project
My original post on my thoughts for uses for the xmos.
After giving it more thought…
I would use at least one possibly two xmos’s as the actuator controls on a robot that is windows xp based.
A mini itx board feeding commands to the xmos connect via usb to a hub.
Xmos catches commands, and in turn interrupts the commands into physical movement. Could also attach the robots sensors to the xmos and relay that info to the computer.
By using it in a setup like this i am able to use a wireless connection to the interent, and do all of my programming in vb or a multitude of other languages, and just send commands to an xmos or two that are doing all of the motor and sensor interaction.
Checking out the XC-5 dev
Checking out the XC-5 dev board, IO seems a little sparse. I can see only 7 IO pins available for use. From a 64 pin chip. There are a lot used by bicolor LEDs and some by 4 switches, but it would be helpful to have more available for robot functions.
In comparision, an Atmel Mega169 also has 64 pins. Their Butterfly demo/development board has 16 GPIO available, some limited ADC, a UART port and TWI pins available. And it runs an LCD, has a joystick with similar 4 contacts like above, drives a speaker, reads a temp sensor, reads a light sensor.
Another device is the Texas Instruments ez430-2013 with all 14 pins of the MSP430F2013 available on 0.01 spacing. These include timers, some analog and I2C too. USB tool and chip are $20, though this only has 1 LED for an indicator.
The Butterfly is also $20. They have determined that they are a chip company too, and so are willing to sell dev/demo tools for cheap to get them easily into the hands of anyone who might want to try them out.
For robotics, reading sensors and outputing drive controls and indicators are needed. Sensors can be simple digital on/off, or analog or timer based. Outputs include LED indicators, but also LCD or even serial screens, as well as speakers. H-bridges can require anywhere from 1 to 4 driving lines each. The 0.1" spacing is good for accessibility.
I know that each of these other micros do not have the processing capabilites of the XMOS devices, but they have been made more available by their pricing and more accessible/usable in their PCB layout. It’s like with these micros, they have been well matched to their vehicle, like a motorcycle engine on a motorcycle. The XC-5 almost seems like a Jaguar V-12 on a unicycle.
What I’ve seen for price points seems to be in the under $50 category, for demo tools of any sort.
Personally, I have a Butterfly board, and really enjoy it. I have considered the ez430 for possible low power apps, but not tried yet.
sorry
sorry i sound pushy dont i? it just i like what the chip has to offer ,other chips with these abilities cost hundreds .
Well I had a brainstorm
Well I had a brainstorm while chatting in the shoutbox and came up with another project. This one sounds really cool although probably more difficult to implement than the robotic planter.
Title: Undecided.
Description: Swarm AI. A central mother brain (based on the xmos dev kit). The mother is solar powered by panels on its back, but cannot move and has no sensors other than detecting its own battery level. Of course, the mother must find sunlight, so it can communicate wirelessly with several smaller worker bots (powered by AVRs and equipped with various sensors) which can move about the area seeking bright lights. Once a good place is found, the mother recalls all the workers who then push or pull the mother towards the light.</p><p>Im toying with the idea now of having the mother charge all her workers, that way the whole "hive" must work together to survive. Other requirements for the mother could be added later. ie must seek darkness when people threaten her.
Kits: 1
Log: Photos and videos.
Nice!
Thanks for the project proposal. That sounds pretty cool!
Hopefully XMOS would be a great match for you. If there is not enough power, you can simply chain XMOS processors together to get more.
For example we have a demo board that has 16 quadcore XMOS processors on it, of which each of the 16 processors is 1600 MIPS. Thats 25,600 MIPS in total!
I believe our new board is aimed in this range
It is interesting you should say around 20-30 bucks, as one of our new boards may be in this price point. Though it depends what bits and pieces end up on it, and the volume at which they are made (still being finalised). Your dream might come true! Watch this space!
Intresting approach
Interesting approach.
Indeed, I believe we have a UART tutorial already on how to talk to your dev kit via USB, so this should be pretty easy to set up, and then you just have the programming either end to do. Cool!