Guest post by VIA is working on a new x86 compatible CPU codenamed Isaiah II, the first in years from the company. Its low power, highly efficient design compares favorably to offerings from AMD and Intel in the same market. It was tested on a VIA branded motherboard with a VIA chipset, giving hope to Free Software users who currently struggle with locked down or unsupported boards from the major manufacturers.
Sounds good, thanks for the heads up.
Was expecting my next board like this to be ARM rather than x86 but this is very interesting.
VIA has at times sold apparently low end boards at above par prices so I am not sure how much the free software will enjoy these.
Meh. There’s nothing interesting happening in the x86-scene. Sure, VIA is an unexpected company to be delivering a whole new CPU to the market, but they’re not really doing anything new or even improving much on anything. I suppose this is good news for those who need low-power x86, but I just can’t muster up enough excitement to care on a personal level.
ARM, on the hand? I dunno, I just find the ARM-scene so much more interesting, there’s always something going on, one or another company trying to find a new niche for themselves and fierce competition on every level, both in the terms of the various full-blown boards and just the plain SoCs themselves. I’m having a birthday in early December and I’m gonna treat myself to a higher-end ARM-board as a present.
Humming board caught my eye
Linux on ARM is an much larger mess, driver wise, than linux on x86
If we’re talking about video drivers, most have binary blobs for any remotely advanced functionality.
I really hope VIA doesn’t give up, Atom needs some competition. I’m talking about low power servers here, not low power desktops.
Edited 2014-09-05 09:20 UTC
I’d agree with the current ARM Linux situation. However, I’m optomistic that the work of Jon Masters at Redhat to standardize ARM 64 for servers will trickle down to desktops/tablets. I’d love to be able to just install plasma active on *any* tablet I chose off of the shelf. In fact, I’m really tempted to do look into the cheap windows x86 tablets that are coming out to see if they can be made to work with Linux.
That’s what I think VIA is trying to do here. Whether they are successful is another story, but it wouldn’t be the first time a company has reinvented themselves and released a great product.
As others have said, I’m not holding my breath, but I’m trying to be as optimistic as I can.
It would be nice to have a Tegra powered notebook with a proper OS and not the ChromeOS toy.
You mean like this:
http://www.cnet.com/products/lenovo-ideapad-yoga-11/
Yes, although Yoga leaves a lot to be desired.
How does Arndale Octa sound? I could part from it with a very reasonable price
Odroid-XU3 is better and cheaper.
http://www.hardkernel.com/main/main.php
UDOO FTW :
http://www.udoo.org/
http://www.udoo.org/features/
http://www.udoo.org/downloads/
http://shop.udoo.org/usa/?___from_store=eu&popup=no
http://shop.udoo.org/eu/?___from_store=eu&popup=no
Kochise
I was actually just checking that out yesterday. Planning for a DIY music streamer/player.
I’ve had my eye on that for a while too. I think it’s well suited for longer term serious projects that the Raspberry Pi is just not powerful enough for.
Why would Open Source people cheer a VIA based system? Unless the situation changed and I wasn’t paying attention (and I hope it has) VIA haven’t exactly been particularly open or supportive when it comes to their video drivers.
That’s why I said might be. I’m hoping at this stage VIA understands the need for a good Open Source friendly x86 chipset that doesn’t tie users to Windows via SecureBoot. Given its performance is right in the middle of current low-power offerings from Intel and AMD, and it is probably cheaper than either, a VIA-based mini PC running an open OS would be a good thing.
Hopefully VIA sees it that way.
What? Is there any processor that only supports Windows? And is there even a motherboard/notebook computer that doesn’t support disabling of secure boot?
I was strictly speaking of the motherboard possibly (but not confirmed) being free of SecureBoot, making it less F/OSS hostile. Most implementations of SecureBoot can be disabled, but not all. Try installing an alternate OS on most Windows 8.1 devices released this year, your luck will range from “possible with a ton of work” to “don’t even bother trying”.
Well I`m repairing desktops and notebooks, even the new ones, shipped last month from many manufactures (Dell, Samsung, HP, Lenovo, etc) and I never had any problems with disabling secure boot and running Linux.
Secureboot is not hostile to OSS, This laptop here has it on it, I just disabled. It now runs Xubuntu. Secureboot is a tempest in a teacup, not even worth worrying about.
Generally speaking, SecureBoot is a good idea. On a system that will only ever run Windows, I’m more than happy to have it protecting my system’s EFI.
The problem is when you try to disable it to install a different OS. Some OEMs make it easy, putting the options right there in plain view in the BIOS/UEFI setup. Other OEMs make it much more difficult, often hiding it in a manner that requires trial and error or extensive research. I’ve even seen systems that require a certain key combination to access the BIOS setup, but don’t allow keyboard input during boot. So, the OEM can say “we support disabling SecureBoot” but lock the customer out of the BIOS setup screen that allows it.
To me, anything that makes the normally trivial task of installing alternate OSes into a guessing game, or worse, impossible, is a roadblock and makes me want to seek out another, F/OSS friendly OEM.
I agree, that’s why I say Secureboot is not a threat to OSS, but some OEMS are hostile to it. Avoiding the technology altogether, instead of just those OEMs, is over kill.
When it comes to providing the documentation needed by OS developers and device driver writers (for CPUs and chipsets), VIA has always failed miserably.
– Brendan
I agree, that’s why I hope they have seen the light over the past few years since their last major chipset release. This is their moment to shine in the increasingly open “Internet of Things” movement, if they will seize the opportunity.
I would hardly say “failed miserably” – they usually provide all their docs to developers when asked as long as they sign an NDA. If you check the VIA forums on this topic (and I have), that’s the common response. So I’d say “makes it difficult with their restrictions” rather than “failed miserably” is closer to the truth.
Its the entire reason why they have failed. Crap documentation and crap restrictions on that documentation is why it’s completely stagnated.
VIA is just as bad as the ARM market.
Compare it to Intel and AMD, both companies put out docs and code and because of t have a community of devs helping out their drivers.
It doesn’ matter anyways, with AMD moving into the ARM sector with their Seattle series http://www.eweek.com/servers/amd-launches-arm-server-developer-kit…. of 64 bit Cortex-A57 based SoCs that support off the shelf ram dimms, SATA drives and PCIe cards the ARM market is going to be changing for the better.
We can hope that imgtech’s new mips processor with powervr gpu is going to be a good open source platform too.
Neither company has a good track record so far. There’s no reason to expect anything different from either.
Edited 2014-09-05 00:24 UTC
Not convinced either. Owner of a (working) C7 mini-itx server (backup file server) and I used to have a Via Nano based Samsung NC20 netbook,
VIA always announces products but does not deliver. It looks great on paper but I bet it will take months if not years before it it will really be produced and is hopelessly outdated by then.
Also their software support was pretty crap (graphics drivers)
Edited 2014-09-05 00:59 UTC
No, no we can’t hope that. It’s using a PowerVR GPU, aka the most open-source-hostile GPU ever to exist, and there have been company reps on Reddit that admitted they have no real chance to open source anything related to the GPU. Fuck PowerVR, and fuck people who use it in their products.
I saw some other stats for the VIA Is2 CPU and in those tests it performed even better.
I am very hopeful for the next release of the SoC
Who is strugging installing Linux on new motherboards?
We build about 20 computers a week for our clients and build systems based on Asus, Gigabyte, SuperMicro, Intel and then on the rare occasion, Asrock and MSI. We haven´t had any driver issues in a long, long time on Linux.
In fact, these are far more common on Windows.
For instance, just las night we discovered that an HP Microserver does not allow you to installl Windows 2012 R2 server unless you disable the built-in ethernet card. And what good is a server without a network card? The solution was to put in a new network card from Intel because otherwise it is impossible to complete the installation. This was confirmed by Microsoft and HP, who each blame the other for the situation.
The same hardware works perfectly on Linux. So, for the most part, the driver situation is really, really good these days, particularly for motherboards.
I should have brought it up in the summary, but my hope is that VIA would be willing to release any new motherboards carrying this CPU and chipset without SecureBoot, in effect targeting F/OSS users, tinkerers, IoT dabblers, etc. Basically, repair their damaged image going forward.
It’s getting more difficult every day finding motherboards that haven’t been infected with SecureBoot, though the workarounds when one is faced with it are getting easier. I just think that VIA has an opportunity here to shine in a space that deserves an inexpensive, low wattage, good quality setup for alternative OS use cases.
What’s wrong with Secure Boot, besides the usual utterly incorrect FUD?
By itself, there’s nothing wrong with it. When it’s used as a roadblock to alternate OSes by pressure from Microsoft on the OEMs to prevent it from being unlocked or disabled is where the issue lies.
I’ve even seen one instance where it was locked in a way that prevented other OSes from being installed, and it had nothing to do with Microsoft. Asus had released a very nice netbook with Ubuntu 12.04 installed (we have two of them at work) and it was impossible to install other versions of GNU/Linux without a complicated chroot jail type setup. The bootloader could not be unlocked. It was impossible to even install Windows on it due to this.
Which model netbook is that?
If I recall correctly, it’s in the 1015 series. I’m not at work so I can’t say for sure.
I’ve never looked for mainboards specifically without this feature, but I haven’t run into a problem like this. It seems that almost all mainboards have legacy BIOS support, which works fine.
Having VIA base these mainboards on coreboot would be the best solution of course.
Good quality from VIA?
I don’t know, I’m still jaded from the times when their chipsets were very sub-par; issues with PCI bus and such.
Overall, I call this inaccurate reporting / false advertising / the title seems a bit too optimistic…
I think it was abundantly clear the issues being discussed had to do with VIA kit, since VIA has a pretty poor record of FOSS support. So I fail to see how intel chipsets have anything to do with the article/issue at hand.
whats with the muslim code name?
Taiwanese…
Biblical CPU code names are nothing new for VIA: http://en.wikipedia.org/wiki/List_of_VIA_microprocessor_cores – you have Joshua, Samuel, Ezra, Nehemiah, Esther, and Isaiah with first Nano; lower in the table the Eden platform.
But BTW, the Centaur CPU design team which VIA acquired is US-based…
http://en.wikipedia.org/wiki/Centaur_Technology
http://www.via.com.tw/en/products/processors/
Edited 2014-09-07 22:51 UTC
I can’t tell if you are just horribly ignorant or a troll. Maybe a troll as you have a new account but anyway…
Isaiah is a Hebrew name, the book of Isaiah in the Old Testament is considered cannon for the Jewish and Christian faiths.
This is a tech site so please leave your xenophobia at the door…
I was always surprised that when the interest in low power x86 flourished in the netbook/nettop space, that VIA was effectively a non-player to Intel’s Atom. Power to performance was their forte, and they were left in the dust. The only people who have EVER seemed to care about them were the “build your own car audio” folks at the ITX and smaller form factor, and even there VIA has looked the past few years about as successful as the Patriots were this past weekend!
They really do need SOMETHING, and I’m really hoping this isn’t their swan song.
In a lot of ways it reminds me of AMC. IF AMC had held on another few years, they could have owned the SUV and crossover market; as evidenced by the BMW X6 being the wet dream of every AMC Eagle SX/4 fan. High clearance all wheel drive standard? Who was doing that in the 70’s and good chunk of the ’80’s?
Performance per watt x86? Who was doing that ten years ago?
We’re no more 10 years ago, no wonders why ARM go a great market share and not VIA.
BTW, I own a Jetway J2F7WE2G-LF and I’m very happy with it, a breeze under Windows 2000 and 28/38°C 32/48W idle/burn with a Cooler Master CM Blue Ice as CPU cooler and a Noctua NF-A4x10 FLX as fan replacement :
http://www.jetwaycomputer.com/J7F2.html
http://www.coolermaster.com/service/support/model/RT-UCL-L4U1/
http://www.noctua.at/main.php?show=productview&products_id=47&lng=e…
Kochise
Edited 2014-09-09 15:20 UTC