ARM was no option because it was mainly headless, nowhere to buy or only super low profile overpriced boards with no real storage. VIA had everything in that position, including a media chipset to take over graphic stuff if the CPU was not fast enough for e.g. MPEG2 decoding.
But now intel has established Atom there, AMD is coming with Bobcats and all the ARM people are coming to the home market in various forms. VIA could have built a fan community with a good fellowship of users for their stuff til now.
But betray your friends once too often and they go.
And, yes, angry users (and it is not only the Linux users) ARE going to hurt on the long term.
Well, as current VIA CPU/GPU user i am disappointed too, for their promises. As i said some time before - their GPU technology is not so innovative as they think (even if so - both AMD and Intel outperforms it by another engineering solutions - by placing GPU in CPU and gaining much powerful performance), all the technology components are licenced in US, China and Taiwan not to let competitors to use some of VIA/S3 technologies - why they can't publish documentation? Well if previous GPU drivers code are written by third part, choose for next GPU drivers another programmers team, which allows to publish code as GPL. Now is coming new VN1000 and VX11 chipsets, and nothing wouldn't changed.
For those who want to boycott VIA for their promises (and do not want support Intel or AMD), there always are such low-power CPU players as RDC Semiconductor, XCore and DM&P Electronics. All these CPU are SoC type with integrated 2D (not 3D) Core. They all are early Pentium level CPUs with 1GHz speed and PCI interface. And all these companies are in touch with Linux kernel community in order to make for their users better Linux expierence.
VIA dropped the ball in the summer of 2003, when Alan Cox worked with them and made the first codedrop available. This is the code that i started the unichrome project off from.
VIA then often claimed to pick up the ball and to "do it right this time round!!!!!", but every time, they failed.
That VIA failed is one thing. In this whole story, I am mostly disgusted by those who a) believed VIA without getting anything solid or b) made big promises to VIA officials and the community that they could not live up to either.
Every time, big claims from many people ended up in a lot of noise, but little to no code being written. People were holding their breath, waiting for something to happen, instead of dealing with reality, and possibly getting active themselves. And so such events ended up being more damaging than anything else, and there were tons of those.
The last hickup in this story, the big claims openchrome made a few years ago when VIA handed out some docs (in response to the AMD docs of course), was the most exemplary of the lot. Secretly, the openchrome people (on an invite only irc channel and in private email) were coming up with plans and waiting for VIAs actions. Their plans of course never materialised, and the irc channel was littered with "I'm sorry, i didn't have much time in the last month or so, maybe in one month time i will be able to do something".
So, you end up with users waiting, and being told to wait, and then of course none of them stepping up and getting things done.
VIA officials also ended up doing something similar: they waited on openchrome people to deliver, only to see them come up with the above line. VIA now thinks that it is impossible to work with the open source community, and it knows that it doesn't have the resources to do anything much themselves.
Openchrome, this is your legacy.
It is better to only talk big and nice when you actually have something constructive to talk about, or otherwise talk big and nasty when some things needs to change. Stick to reality, tell the true story, and actually write the code.
All three of them?For those that were hoping that VIA Technologies...
I still don't get what market VIA is aiming for. Why would anyone buy a mini-ITX system from them when they can buy an equivalent equipped system with Atom that probably has better performance, lower power consumption and costs less?
If the GPU was capable of decoding video in linux it could be used to build a low-power, cheap and silent HTPC, but if on top of the already higher cost for the hardware you have to buy an OS (Windows), it looses any interest it could possibly have. The only use I see in a VIA mini-ITX system is in a POS terminal or something like that. And even in that role, Atom would be cheaper and not limited to Windows to take full advantage of its (limietd) capabilities.
VIA, you had your chance, but you pretty much blew it. Only way to fix that mess is to quit integrating S3 graphics or if you continue to do so, offering an MXM module or PCI-E 16 lane slot on your designs.
It's not even really useful in POS or Kiosk role right now. I can do better even if the Nano might be faster by going Atom (Or...better yet...ARM...same performance, even lower power...). They've bet the farm on Windows and they're about to find out that that wasn't a good bet. Windows might not be going away anytime soon- but neither is Linux and it's ascendant in the market segments they're selling to right now.If the GPU was capable of decoding video in linux it could be used to build a low-power, cheap and silent HTPC, but if on top of the already higher cost for the hardware you have to buy an OS (Windows), it looses any interest it could possibly have. The only use I see in a VIA mini-ITX system is in a POS terminal or something like that. And even in that role, Atom would be cheaper and not limited to Windows to take full advantage of its (limietd) capabilities.