“VIA’s newly launched processor architecture, known for the last three years by its codename, “Isaiah,” will keep the company’s focus on cost and power intact while taking things in a substantially different direction. In short, this year will see something truly odd happen on the low end of the x86 market: VIA and Intel will, architecturally speaking, switch places. Intel will take a giant step down the power/performance ladder with the debut of Silverthorne/Diamondville, its first in-order x86 processor design since the original Pentium, while VIA will attempt to move up into Intel’s territory with its first-ever out-of-order, fully buzzword-compliant processor, codenamed Isaiah. In this brief article, I’ll give an overview of Isaiah and of what VIA hopes to accomplish with this new design. Most of the high-level details of Isaiah have been known since at least 2004, when VIA began publicizing the forthcoming processor’s general feature list (i.e., 64-bit support, out-of-order execution, vector processing, memory disambiguation, and others). So I’ll focus here on a recap of those features and on a broader look at the market that VIA is headed into.”
We wont know until there are some real benchmarks available, but the fundamental design appears to indicate that this chip could come close to clock-for-clock parity with say a Pentium-M (hopefully at least that) or even a Core-2 Celeron (unlikely but that would be awesome).
I know alot of people look at that and say “so what?”. Well VIA’s Epia stuff (nano-ITX) is a _very_ nice platform for small embedded systems (firewalls, routers, etc.) A chip like this on a nano-itx board would make it possible to build _really_ small desktop machines, UMPCs, and even home theater PCs (the current C7 based stuff is just too slow for most uses). It _should_ have enough horsepower to at least do 720P decoding (1080P may be too much for it, but its certainly within the realm of possibility). A MythTV frontend using one of these would be sweet!
You talk about struggling performance, then you suggest that MythTV would be a good viewing interface??
“You talk about struggling performance, then you suggest that MythTV would be a good viewing interface??”
The keyword here is frontend
Edited 2008-01-25 16:33 UTC
If you’re talking about the mythtvfrontend package. That is the one with the ridiculous performance issues. If you’re talking about a seperate machine acting as a frontend, then having a small-form-factor is somewhat pointless, unless you live in a very small flat.
Im not sure I am parsing your reply correctly, but Ill try. MythTV IS a good viewing interface (imo). The question is whether this chip is suitable for it. In a nutshell I’m saying that VIA’s new chip mounted on a nano-ITX epia board _could_ be a wonderful platform for a MythTV frontend. The _could_ meaning there is basically a performance sweetspot for doing 720P decoding – and all this chip has to do is reach it with a bit of breathing room. I dont think it will be able to do 1080P, but 720P is definitely possible.
Not everyone cares about the same things. Many (if not most) people run Myth monolithically. Nothing wrong with that, but I prefer having my backend/tuners in a wiring closet and putting a frontend on each TV. A frontend the size of a small paperback book that can do 720P decoding for less than $400 would be pretty nice (for me anyway – to each his own).
You’re right, lucidity is not always one of my strong points, and I was a little distracted this morning.
My only point is that for many people, the performance/speed/latency of the mythtv frontend/user interface is not great at all. So much so that my PPC mac-mini [running ubuntu] can’t even properly cope with navigating the user interface, let alone playing back video/audio without major frame dropping. I know I am not alone in experiencing this, but maybe we are just a vocal minority of users.
Yeah… I cant argue with that. The UI isnt exactly glass smooth – even on a high-end system it still feels a bit sluggish. There are quite alot of tweaks that can be made to improve it, but its a pita…
That’s what I’m interested in with these. I love the ASUS Eee but I’ve been holding off to see what the competition brings in now that ASUS has shown there is a market. I’m also hoping to see ASUS bring in a Eee model that doesn’t come with the crippled 900mhz processor that’s killing their battery life. Now at last with the impending release of this it looks like that may happen sooner than later.
–bornagainpenguin
Combine a 22.5cm or 25cm (9 or 10 inches for behind-the-times Americans) low-power LCD 1024×640 widescreen (possibly one from Pixel Qi http://pixelqi.com/ ), a fast flash filesystem, some additional storage capacity and a full and popular GNU/Linux OS with a large application repository (perhaps Xubuntu or similar, or maybe even KDE4 would be quick enough), wireless mesh networking and one of these new Via processors could make a next-gen ASUS EEPC competitior machine a real game-changer.
Ultra-portable UMPC++ on a budget! Nice.
I was going to suggest the OLPC screen, but in fact the Pixel Qi is the ‘OLPC spinoff’ so it’s the same thing.
But yes, I agree, a 10″ OLPC-like screen, this new CPU, 512MB of RAM (upgradable), an Ethernet port, a ‘second generation’ (faster) SSD (upgradable), wireless meshing and the result is a killer small laptop.
The ASUS EEE cpu is NOT crippled.
It’s a full blown Pentium3-M dotham processor with 512kb L2.
Very good processor and I believe blows away any via cpu produced clock per clock. The EEE uses at most 18W at the wall.
Really I’d like to see something more like a 1GHz Ti OMAP SOC with powerVR in the EEE which should push battery life to over 10+ hours. ARM might not be exactly right for a full use computer though.
Edited 2008-01-25 20:27 UTC
Ummm… did I blink again? Last time I checked the Eee used a 900 MHz Intel Celeron M ULV 353 which didn’t seem to have very much mobility support and had to be underclocked to prevent issues…
–bornagainpenguin
I’ve always loved VIA, ever since I’ve used their EPIA 5000 system (with the C3 CPU). Their focus on low-power design is great. As a programmer, I’ve found unless you’re writing a CPU-intensive program, you don’t need much to get the job done.
In fact, I’ve dropped my power usage about over 60W by downclocking my CPU from 3.2GHz to 1.2GHz. I don’t even notice the difference when programming.
When I upgrade I’m thinking of going VIA.
in Europe is low. I cannot find a gPC. I think this should be addressed first.
Much larger L2 cache than earlier, 64-bit extensions and so on. It looks promising. However, there is nothing said about their TDP. Still, it is possible that I will finally buy system with VIA processor.
Oh, the article mentions that TPD will be the same as current ones.
The name Isaiah strikes me as interesting. Many Intel chips of the last few years bear Biblical-era Hebrew codenames–Banias, Dothan, Yonah, Merom. These are all ancient Israeli cities and places, except Yonah. Yonah is an alternate transliteration of Jonah, who, of course, got swallowed by a Very Big Fish.
Now VIA comes up with a chip they’re calling Isaiah. Isaiah was a prophet who lived to see Israel conquered and subdued by the Assyrians. Isaiah lived under several Israeli kings who were subject to foreign rule.
Of course, Isaiah’s perspective was that of fighting back against the invaders, which completely spoils my poorly-researched analogy. But still, Isaiah was the most significant Messianic prophet, who predicted the entire string of events leading up to Jesus’s birth (you know, the Christmas story).
Even if the naming of “Isaiah” was not intended as a sly jab against Intel, the synchroneity of their codenames certainly rouses the inner conspiracy theorist.
*** Intel codenames comes from the Intel Design Center in Israel, they are ‘legitimate’ choices.
They produced both great design (PIII -> Core ) and weird ones (like the original 8087)
About VIA :
Imitation is the sincerest form of flattery
Interesting, in the past I thought I remembered their chips mostly being codenamed after rivers in the western US area (i.e. the Deschutes P3).
VIA is an interesting company. Although based out in Tapei County, Taiwan the boss there, Wenchi Chen, is a deeply Christian fellow (a majority of Taiwanese are roughly Buddhists). I had the pleasure of working as an on-site contractor doing devrel and some programming for them a few years back (one of the best free lunches I’ve ever had!).
There are Bible quotes on most doors and this is probably why their chips get biblical names. It’s not the sort of place where Christianity is forced down your throat though. Most of the quotes are the nice ones from the NT about being a good neighbour, treating people with respect, that sort of thing. Well, it made a pleasant change from those dreadful motivational posters!
Other names have been Eshter, Nehemiah, Joshua, Samuel, Ezra. There were also Mark, Luke and John which combined x86, Northbridge and GPU in a single chip.
They’re scared of intel. Intel sees the low power embedded market and have started for it, which VIA’s only tried to exploit it.
– Promise high performance, never delivered
– Price gouge on mini-ITX systems.
Here’s a mini-ITX board & cpu from intel: $75
http://www.mini-box.com/Intel-D201GLY2-Mini-ITX-Motherboard;jsessio…
Benchmarks shows this kills anything via has to offer. They need to get on the ball hopefully this new cpu isn’t just another bunch of vapor.
It seems to be a good system, especially for the price.
However, I still bought the VIA CN 10000 Board, as instead of 19 W (25 with HDD) it only consumes 14 W – and it is fanless.
The difference in power consumption amounts to 43.8 kWh per year or EUR 7 per year of saved electrical power if translated to my electricity supplier’s price of 0.16 EUR/kWh.
The VIA had a price of 118 EUR, the Intel had 65 EUR (at the same store, 3 weeks ago).
The price difference is therefore regained within 7.5 years. I hope, that as this board will be used as a router/small server it will last for this time, so the price advantage of the Intel is marginal for my case at best.
Other use scenarios will lead to a different conclusion of course.