I don't claim to be an Electrical Engineer but from what I was reading on the Titan they think it'll go to 2Ghz at 90nm.
Yeah, maybe you're right... though I think that's after some process tweaking. They seem to be using the Intrinsity technology, or something very much like it, on that CPU. They do a much of tweaking up just the bottlenecks, changing CMOS to NMOS circuity, and some other tricks to get the speed of things up. There are a bunch of CPU companies doing this... for example, Samsung used Intrinsity to get their ARM A8 core running at 1GHz (Samsung has the kind of ARM license that lets them the ARM... and keep in mind, they were the guys pushing the DEC Alpha to new levels, back in the day).
I have no idea, that's from a quote from the marketing on the X1000 page. However, I know the Titan is geared at the Telecommunications/Embedded markets,
Of course it is... that's "code" for "PowerPC" these days. At least some years back, Cisco was putting PowerPC in most every router and switch they made. Thus, PowerPCs grew all sorts of router and switch friendly hardware, like modern memory buses and RapidIO links.
Of course, every CPU that's not an "applications processor" (eg, desktop PC, PDA/phone, etc) is an embedded CPU... they're used in a box or board, hidden away somewhere. For example, I built an R/C controller system. which had two 32-bit ARM CPUs on the main controller, one 16-bit RI MSP430 CPU on the remote, even a tiny Zilog Z8 on an optional tachometer sensor. That's "embedded". So are the CPUs in your DVD player, TV, microwave oven, etc (in case anyone here's not familiar with the term).
The first TiVo used an embedded PPC... not super fast, though, so they had a bit of hardware to decode and encode. Second model used some MIPS CPU, fast enough to decode MPEG-2 in software, but they still needed the hardware encoder.
so you tell me does having an Xena chip with custom I/O socket on board give anyone the opportunity to develop/test Telecom/Embedded systems using this board before moving them onto a physical board or is it pointless with today's virtual circuit design and automatic PCB milling system?
I don't believe anyone building embedded telecommunications gear would bother with something like this. You can get a reference design board for any CPU you're after... all CPU companies either make the boards themselves, or partner with a board company. These are often small enough to fit your target device... in fact, many embedded designs just use off-the-shelf embedded CPU boards -- not everyone wants to design these things from scratch. But you wouldn't likely waste the space of a full-sized ATX motherboard, even for development.
The question would be, as you point out well, is there any use today for AmigaOS 4 running on what for the OS is a fast system with a Xena processor that would move this to other markets. I don't have a good answer but I for one would love to see AmigaOS get even a small foothold in the market so that it can continue to be developed.
That's the question. I'm not sure just what you'd do with that XMOS chip on its own. It's interesting, but I reject their example of MP3 player... that's well served by $3.00 DSPs these days, probably with a bunch of on-chip peripherals specifically designed for making MP3 players, ultra-low power, maybe even power managed to run off AA cells or a 3.6V Li-ion rechargeable. And they'd have reference designs, software, etc. There's a whole food chain for every microcontroller, which goes by application. Some companies web sites have hundreds of application notes, examples, code, hardware designs, etc.... for each CPU family they make. The fact I didn't really "get" the XMOS chip... ok, I wasn't motivated by "this is work", but you know, I do this for a living. I shouldn't have to guess :-)
I guess I'm still a dreamer. 8^) Those crazy Commodore engineers made me into one. 8^) Starting with the VIC20, progressing to the C64 onto the Amiga - they always had a way of pushing the envelope and creating something revolutionary not evolutionary.
None of those guys are working on this, far as I know. And it's a very different world. When we were doing this, personal computing was still very, very young. Right now, not so much.. it has matured. And yeah, it's a little sad, because who gets excited about a new computer release? Ok, maybe a few silly Macheads, but really, the PC you see this year is just a little better than last year. There are occasionally new CPU microarchitectures, but most of the time, it's just small improvements. Same with GPUs.
The reason isn't that no one's trying.. but rather, that many tried, and most ultimately failed. Those who are left are spending billions to incremental improvements, funded by the billions of chips they sell. Mature market.
Doesn't mean there can't be fun, or cool new things. I've been far more interested in the cool computing devices I can put in my pocket, or even my livingroom, than on my desktop. There's just more action in those places.
P.S. It was enough to get Dave to follow the thread so maybe it's at least a little intriguing. 8^) Good talking with you again after all these years. Last time would have been on a newsgroup in the early days of the Internet on my A4000D/030 over modem. 8^)
Ok, I'm hooked in via satellite modem... it's a little faster than dial-up. And many, many times more expensive. But all those dishes on the roof is great for my tech-cred :-)
