Itsy Specs Updated

Jim Gettys (rock!) sent us some good news from the DEC/Compaq camp. Itsy, the ARM/Linux/PDA that has been under development for some time now, and is chock full of nifty little features, has had its schematics and software updated for you industrious do-it-your-selfers wanting to contribute to pocket computing.

Date: Mon, 7 Feb 2000 16:53:48 -0800
From: Deborah Wallach [kerr@pa.dec.com]
To: sa1100-linux@pa.dec.com, linux-arm-kernel@lists.arm.linux.org.uk
Subject: Itsy schematics and updated Linux software available

In order to promote collaboration and to advance the state of pocket computing research, Compaq Computer Corporation's Research Laboratories are making available the schematics and manufacturing information for the Itsy V1.5 Pocket Computer. The Itsy is a flexible research platform, based on the StrongARM 1100, intended to enable hardware and software research in pocket computing. It is a small, low-power handheld device with a highly flexible interface, designed to encourage the development of innovative research projects, such as novel user interfaces, new applications, power management techniques, hardware extensions, etc. The information we are making available includes the hardware specifications, a low level monitor, and the Linux source code for the Itsy.

Information about the project

More information on the distribution

Deborah Wallach
Western Research Laboratory
Compaq Computer Corporation

Comments

The Right Link

[thrash_]

Here is the correct link, in case the link on the page doesn't work for you.

http://crl.rese arch.compaq.com/downloads/register.cgi?download=It sy

Good as far as it goes, but still useless

[killbill]

Kudos to them for opening up the hardware and software designs (open in the "now I know what they did" sense but not the GNU sense, although Stallman himself insists on creating an arbitrary (and in my mind artificial) distinction between hardware IP and software IP, but I digress)...

Anyway, good for them for sharing, but they are using an LCD and touch screen that is unavailable commercially. What good is that?

If they were serious, they should redesign it to use an available module, or offer modules for sale from their site for a reasonable charge.

Although realistically, given what I suspect is a lot of surface mount components and very small trace widths, and probably multi-layer PC boards, I doubt many people will be throwing together one of these in their basement...

It would be nice if they offered an "unsupported" Itsy kit with all necessary parts, but I guess I can't complain about people giving me information.

Bill

Re:Good as far as it goes, but still useless

[gargle]

Anyway, good for them for sharing, but they are using an LCD and touch screen that is unavailable commercially. What good is that?

It's a research project. It's unreasonable to expect all the parts to be commercially available.

Re:Good as far as it goes, but still useless

[torpor]

Well, why don't you download the specs, the schematics, etc. and design it to use commonly available LCD screens yourself?

After all, isn't this what you'd do if it were an Open Source software project?

On the one hand, you disagree with Stallman in making a distinction between hardware and software IP, but on the other hand you refuse to apply the same successful formula to an Open Hardware project as you would to an Open Source project.

I for one can't wait to get the download of these schematics finished, so I can see just what *is* available and compatible with their design...

Re:Good as far as it goes, but still useless

[killbill]

Well, why don't you download the specs, the schematics, etc. and design it to use commonly available LCD screens yourself?

That is exactly my point. The obstacles are currently too high. I'm not saying others can't contribute, just that there are a couple of pretty substantial obstacles that prevent ME from doing so.

Like I said, Kudo's to them for opening it up, but "openness" is not the only thing that determines the success of an open source project. In addition to being "open", a project also be usefull, interesting, and modifiable. I could write a complete and perfect replacement for Microsoft Office for Linux and opensource it, but if it is written as a single 1,000,000 line main program in "obsfucated C", I should not expect developer support to come pouring in.

My apologies if it sounded like I was criticizing the project for opening up. I was just trying to state that if their goal is to develop an open community of loosly related developers, basing the hardware on a non-commercially available part is silly.

On the one hand, you disagree with Stallman in making a distinction between hardware and software IP, but on the other hand you refuse to apply the same successful formula to an Open Hardware project as you would to an Open Source project.

I knew I should have kept my mouth shut about that... everytime I mention Stallman I have gotten into firefights... :)

I was trying to make exactly the opposite argument. I do not see a distinction between a large subset of hardware IP and software IP, and don't see how a distinction could be made (where Stallman appeared to do so in a talk I heard him give two weeks ago).

Carefull with your terminology... Stallman is not "Open Source", he is "Free Software". There are some very important distinctions.

I am applying exactly the same standards, and therein lies my critic of the project. If the goal of "Opening" the hardware and software is to generate a spontaneous and dynamic community of developers, testers, and contributors, then the fact that this device requires a number of components that are not commercially available is the first and foremost obstacle to it's success.

I hear and accept your criticisim of my comments that I am just complaining about the status quo without doing anything about it.

What I meant to communicate was "Great job. Too bad the obstacles to my contribution to your project are prohibitively high, even though I find the project compelling and interesting. By the way, here are the two things, in case you are looking for ways to increase your community... ".

I am encouraged that others do not find the obstacles prohibitively high. When you get the sourcing problem for the display/touchscreen module solved, as well as come up with a cost effective source for (what I suspect is a) multilayer surface mount PC board, please post the information so more of us can contribute to what appears to be a very interesting project.

Bill

Re:Good as far as it goes, but still useless

[torpor]

My apologies if it sounded like I was criticizing the project for opening up. I was just trying to state that if their goal is to develop an open community of loosly related developers, basing the hardware on a non-commercially available part is silly.

Well, I would think, given what I've been able to glean about this project as I've followed it for the last few years, that their primary goal wasn't so much to build an open hardware platform, but to prototype a bleeding-edge portable platform that just so happened to run Linux, one of the major open source products out there on the 'market' (!) today.

So it's not really fair to say that they're missing their goal of making a broadly useable open design product - I'd say that's a secondary goal for them, which just so happens to have become a reality recently. Otherwise, they would've been open from the beginning, and we wouldn't have this problem... :)

Note that I'm not 'attacking' you - certainly you are an intelligent enough person to see the value of politeness, and I respect that. I'd just like to point out to you that this project wasn't designed to be open from the beginning, and now that it is, yes there will be a few hurdles to overcome.

But, give it time. There are plenty of very skilled hardware hackers out there who are very probably dissecting the schematics right now, looking through reference manuals, and attempting to route around the problem.

I hope you benefit from this as much as I do, and if so, I hope we *both* find ways to contribute any way we can.

Download URL Broken

[Anomalous Canard]

Too bad I can't download the specs.

A problem occurred during the download process.

Download Itsyify=itsy-registration@pa.dec.com is unknown

If you do not understand what caused this, feel free to send us email describing what happened. This will help us to fix problems and improve the site.

Anomalous: inconsistent with or deviating from what is usual, normal, or expected

Re:Download URL Broken

[Anomalous Canard]

Viewing the source, I see the problem.

This link should work.

I suppose I should blame IE for maximum karma bonus.
Anomalous: inconsistent with or deviating from what is usual, normal, or expected

Re:ARM vs Crusoe

[tjwhaynes]

What would the advantages of putting an ARM processor over a Crusoe in a PDA? If the Crusoe can morph any instruction set, wouldn't it be able to run ARM instructions? How would a company benefit from using ARM chips in their PDAs, other than a slight decrease in power consumption?

Well, for a start, the SA-1100 uses less power at 200MHz (about 200mW) than the Crusoe at 400MHz (about 1W). Even given that power consumption tends to go as voltage squared, and that still means that the SA core gives you more MIPS/Watt. Secondly, unless the Crusoe RISC core matches well against the SA core in as much as having virtually equivalent instructions, emulating a RISC instruction set with a different RISC instruction set can lead to performance hang ups. For example, the ARM chips tend to have both conditional instructions and barrel shifters on each instruction, so while they are single issue they can end up doing several things for instruction. (Rusty ARM code warning alert!!)

e.g. ADDS r1,r1,r2 LSR#2; MOVNE r0, r1 LSL#4

Add r1 and r2DIV4 , place in r1 and set flags on result of sum. If negative flags set, move r1 shifted left 4 into r0.

As you can see, if for each of these instructions, you end up firing off two or three Crusoe RISC core instructions, performance can be degraded to a level at around a third of the base clock speed. Similar problems do not occur to the same extent with CISC, as you have more scope to interpret and pass on RISC instructions from CISC ones (as I believe AMD do in their intel compatible chips).

The other thing that springs to mind is that if Linux is your target OS, then either Crusoe running Linux under x86 emulation or a StrongARM running ARMLinux will fit your requirements well, so if battery life is your ultimate goal, then the StrongARM hardware may fit your requirements better.

Cheers,

Toby Haynes

Not possible

[Anomalous Canard]

The license dosn't allow it.

2. LICENSE GRANT
2.1 Subject to the terms and conditions of this Agreement, COMPAQ hereby grants CUSTOMER a non-exclusive, worldwide, non-transferable, royalty-free license for the term of this Agreement to duplicate and use HARDWARE INFORMATION solely for internal, non-commercial, research purposes.

2.2 CUSTOMER agrees not to distribute the HARDWARE INFORMATION in any form, other than for CUSTOMER's own internal, non-commercial, research purposes.

2.3 CUSTOMER grants to COMPAQ a non-exclusive, royalty-free, world-wide, unrestricted license to sell, duplicate, use, market, sub-license, distribute and create derivative works of DERIVATIVE HARDWARE.

You can't build and sell it and if you build one and modify it somehow (fruity colors?) they get the right to sell your product without compensation. It ain't very open.

Anomalous: inconsistent with or deviating from what is usual, normal, or expected

Re:Not possible

[technos]

Say I put together three Itsy prototypes, one for myself and two for friends. The components cost me a total of $1,300, and taking the sick day needed to assemble them cost me $200. If I charge each friend $500, and no more, I doubt Compaq would have a problem with it.. I am not engaging in a commercial venture, per se. I could almost write it off like this: I am founding a small think-tank. My two assistants are taking an equal share in the startup expenditures, and we are researching the plausibility of wearable PC's. Since I only have three people in my company, and I produce three Itsy prototypes for 'internal research purposes', it is natural that I give each partner one. Then, having a dispute over which of us drank the last beer, we disband the company. All in the course of a day or two.

No problema!

Re:Not possible

[ralphclark]

Yeah. Didn't do IBM much good, though.

Consciousness is not what it thinks it is
Thought exists only as an abstraction

ATTENTION COMPAQ!!!!

[kevin lyda]

SELL THEM TO US!!!!!!

WE WANT TO BUY THEM!!!! WE WANT TO SPEND MONEY!!!! YOUR ADVERTISING BUDGET WILL BE A MAIL MESSAGE TO LINUX-KERNEL!!!!!

ahem

thanks

Re:Limited run production.

[SEWilco]

Just ask if they'll give you permission to build your limited run. A few hundred wouldn't affect their company income much. They'd be letting you do a market test and there'd be several hundred more developers out here tinkering with the devices. This license forbids your doing this, so simply try to get a different license from them.

Limited run production.

[Fixer]

So, who would be willing to organize a limited run production of these devices for the community? I'd be willing to pay 700 or so for an Itsy.
They seem like near-perfect development platforms, great for working on new interfaces, or adapting them to wearable uses. I want one.. *whine*
-Fixer

Re:Limited run production.

[stienman]

We would need to form an organization (anyone got itsy.org?) and then declare that it's memebers are all researching itsy. We can then trade resources, code and info back and forth and still be within their agreement.

Or we can just go it alone, and start from scratch with a totally open solution...

-Adam

CPU time is cheaper than programmer time.

Re:Limited run production.

[Thomas Charron]

I wish I had the time and the resources for this. Heck, I wonder if they themselves would produce a one time run of 100+ units, for 'usability' testing, and/or software R&D.

Re:Limited run production.

[Thomas Charron]

I wonder if this would be at all possible. I'd DEFINATLY buy one, that's for sure..

OTOH

[Andy Dodd]

They're giving away the results of who knows how much $$$ worth of research. This could be useful to a LOT of people. I'm working on a videoconferencing project here, and one of our next goals is wireless handheld video - the Itsy (if it has sufficient performance) could save us a lot of time. Otherwise, we'll have to design a Crusoe-based device. :)

Some things we've been needing

[Captn Pepe]

There was a discussion here a while back about options regarding embedded linux solutions. Several gripes came up, including the lack of a filesystem or drivers for flash memory, support for varying CPU clock speed, and a few other details. Look at this. From Itsy's page,

• Although there are much newer versions of ARM Linux available today, the Itsy port contains several pieces of software not yet available in the newer versions, including the FTL flash file system, power management support, support to dynamically change clock speeds, etc. We expect that these features will eventually be integrated into later versions of Linux, by the Linux community.

Looks like they've gone to the trouble to write a bunch of these for us. Yay! Now anybody want to look into turning these into a 2.3/2.4 kernel mod?

Re:I would rather have one of...

[technos]

Vaporized hardware?

Blue smoke!

Re:I would rather have one of...

[ralphclark]

Nah, Magic Smoke. You know, the stuff in CPU's that makes them work...

Consciousness is not what it thinks it is
Thought exists only as an abstraction

Re:ARM rocks (Power consumption)

[commanderfoxtrot]

They certainly do: I have used Acorn computers (running on ARM chips) ever since they came out in 1987... very fast for their time.

I bought a Psion Series5 last year, which is very good for certain things, such as essays, diaries and the like. It's not underpowered, and it uses (IIRRC) an ARM7500 at 30MHz. The power usage I get from it is 45mA/h when idling, going up to around 110mAh/h for maximum processing. When the backlight is on, the maximum power throughput is around 200mAh.

Looking at the Itsy numbers, they are a _lot_ higher. How is this research device powered? Even the most highly powered AAs give out 1000-2000mAh. I get a more realistic 1000mAh out of each pair of NiMHs.

Great science, but that sort of processing power is unfeasible, in my view, and impractical for such a small device.

OT - Good Volley

[weston]

certainly you are an intelligent enough person to see the value of politeness

Wish I saw more of that on Slashdot; the above posts have dramatically increased my faith that perhaps I will see more of it...

Thanks, guys.