Slashdot Mirror
ARM — Heretic In the Church of Intel, Moore's Law
ericatcw writes "For 30+ years, the PC industry has been as obsessed with under-the-hood performance: MIPs, MHz, transistors per chip. Blame Moore's Law, which effectively laid down the Gospel of marketing PCs like sports cars. But with mobile PCs and green computing coming to the fore, enter ARM, which is challenging the Gospel according to Moore with chips that are low-powered in both senses of the word. Some of its most popular CPUs have 100,000 transistors, fewer than a 12 MHz Intel 286 CPU from 1982 (download PDF). But they also consume as little as a quarter of a watt, which is why netbook makers are embracing them. It's 'megahertz per milli-watt,' that counts, according to ARM exec Ian Drew, who predicts that 6-10 ARM-based netbooks running Linux and costing just around $200 should arrive this year starting in July."
i think those have been around for some time http://kurobox.com/
I don't mean to Dis-ARM, ARM or Armless...
But it will do exactly the same thing, 0.5 Watts now, 100K transistors now, 300 MHz now... it wont stay that way though, it's just a slimmer base to build upon, like using aluminum instead of steal. People will still keep reaching for the sky, and with a lighter structure, means they can reach even higher, even more MHz, more transistors, etc...
ARM chips are nice, but they are not as fast as Atoms and their low power usage does not guarantee long battery life. It needs to perform at least on the level of a Dothan 600MHz before I'm interested - web surfing is already a pain at that level of performance.
that some /.ers seem to need to create an enemy of conventional wisdom, even when conventional wisdom is conventional for a reason?
Yes, efficiency is good. But do you really need to smear the idea of higher processing power at the same time you're pointing out the good in low electricity consumption?
I mean... really?
If you can read this... 01110101 01110010 00100000 01100001 00100000 01100111 01100101 01100101 01101011
What happens when those 6-10 netbooks get sold? What about the rest of us? Seems like it's hardly worth it to build so few. They should be building them by the thousands!
"Tell me doctor, with all of your defenses, are there any provisions for an attack by killer bees?"
Web browsers are interpreters, which are going to be slower than machines that run pre-compiled code. Could web servers pre-parse the html for target platforms, to speed things up? I'm sure Microsoft would be willing to lead the way forward :s
Maybe we should make computers out of them. In fact, they did...
http://en.wikipedia.org/wiki/Roman_abacus
This is my sig.
The marketing term (not the architecture) MIPS == Million Instructions Per Second. It's not the plural form of some other TLA. ;)
Colorless green Cthulhu waits dreaming furiously.
And a 1 GHz Cortex-A8 core is probably in that ballpark.
And, there's always the Cortex-A9 MPCore, which should help even more.
A quarter of a watt is a percentage of the static I gather walking. A processor like that is powerful enough to run a tiny GPS, an insert in my shoe. Add a little foot-pad to power a HUD and attached map and I always know where I am. This is one of many, many uses. Anyone still thinking "cell phone" is missing the point.
If video games influenced behavior the Pac Man generation would be eating pills and running away from their problems.
opera for low-end phones uses a proxy that converts the html to a compressed image.
Do you even lift?
These aren't the 'roids you're looking for.
And as a regular user of Opera Mini 4 (not to be confused with Opera Mobile), their proxy based system can work impressively well on mobile phones that are clearly underpowered for web browsing (like the cheap pieces of crap I have used it on).
1 (short ton / firkin) = 89.1432354 slugs / keg
They are the only chips that you can program and keep your sanity.
The ARM code is just beautyful design, one weeps with joy after struggling through x86 hell.
And computing/electric power ratio is fantastic.
Web browsers are interpreters, which are going to be slower than machines that run pre-compiled code
It's worse than that: In addition to HTML, a web browser must parse/interpret JavaScript, Java, CSS, XHTML, Flash (if Adobe ever gets onboard), and regular XML just to display the modern, JavaScript-heavy web application. This gets resource intensive if, say, using an app such as Google Docs on a netbook with little memory, since the browser keeps the DOM structure in memory, and it gets exponential if the user has multiple tabs open with an app/page in each.
/. a while back, but how well they would maintain this for both x86 AND ARM remains another story, in addition to all of the other problems that could ensue, especially at the security level (a bug in the JS parser leading to direct remote code execution, etc.).
A server pre-parsing HTML would mean a browser/server handshake, something IE and IIS could easily do moreso than Apache(2)/Lighttpd and Firefox/Safari/Chrome. Opera does this with their mobile platform, but it is still far from perfecting JavaScript precompilation or even delegating this to the lower-resource device at the client end.
Google was contemplating compiling JavaScript to pure native code in a story I read here on
It's problems like these that keep 300Mhz netbooks with little RAM from being very efficient with full-scale web apps. Just my firefox I'm running now, I have about 20 tabs (mostly regular HTML) open and it runs up my dual-core CPU so high that my fan is running (not much in the background), and it eats memory like crazy. But as far as MS breaking the Wintel relationship to pursue ARM-based netbooks, I don't see it happening unless something drastic happens.
And a 1 GHz Cortex-A8 core is probably in that ballpark.
Perhaps. A 530MHz Dothan was about twice as fast as a 600MHz Cortex A8 in a benchmark I saw. That does not mean the A8 is slower for browsing, as a browser is so complex that a simple CPU bench isn't enough. One has to sit down and use the system.
Correct me if I'm wrong, but aren't ARM chips also used within the Nintendo DS(i) and the GameBoy (Advance)?
From TFA:
For 30 years, the PC industry has treated Moore's Law with religious reverence. Its immutable commandment -- thou shalt double the transistors on circuits every 18 months -- created an enviable business model with consumers spurred to buy new, more powerful PCs every few years.
The actual law is about reduction of cost, not increase of performance. Other formulation says:
The transistor cost shall halve every 2 years.
ARM is not breaking any "law".
"ARM chips" is far to general a term...there is a vast difference in the performance between an ARM7,ARM9,ARM11 and say a quad core ARM cortex-a9...
as for performance it really depends on the work load, the cortex is clockable at 1ghz and contains multimedia units that may well make it faster than the atom for activities like media playback.
a lot will come down to optimization for the core, ARM has the advantage in that it has to have custom compiled binaries for this architecture. the atom general deals with a stand x86 binary target.
the release of ubuntu 9.04 which will have a ARM version should allow much more concrete and real world performance comparisons
As I read through the article (I know, I've already violated Slashdot's law, but anyway), I couldn't help but go back to this whole idea of 'under-the-hood performance.' Cars built today don't necessarily have to have the 400 cubic inch plants and 500 horsepower that they sometimes had in the 60's. Engines are half that size and half the horsepower, but because they're designed better, it doesn't matter. (Although I'd love a 500 hp engine anyway.)
As well, continuing the car analog, just because there are still some cars with 500 horsepower engines made today, it doesn't mean everyone needs one. There are plenty of tiny cars doing just fine thankyou
This article suggests that because we're not using giant oversized processors in our iPods and cellphones, that somehow we've violated Moore's law. All it really means is that putting a Ferrari engine in golfcart is pointless.
What's your system overhead like? I run a 400 MHz Pentium 3 with 256Mb RAM (I know, its ancient). I've got no problems with web surfing, multiple tabs and all. Admittedly, the WiFi connection usually gags first. But I'm running Eclipse as well and MySQL is grinding away on a search in the background. No problems.
The only time my fan ever kicks on is when I'm rebuilding the kernel.
Have gnu, will travel.
and it gets exponential if the user has multiple tabs open with an app/page in each.
You keep using that word. I do not think it means what you think it means.
Megahertz per milli-watt doesn't make sense either. Some chips can accomplish more per clock cycle than others.
(And in this case, the statistic is going to make ARM look good.)
They may up the megahertz, but not at the expense of a more costly product or more power usage. Instead, the ARM chip vendors take a look at what needs the MHZ, such as video/audio decoding, and include special co-processors for those functions on the same silicon. Therefore they don't need to increase MHz for increased functionality.
It is a similar philosophy to using a script written in a slow interpreted language to drive a more complex system composed of high-speed modules written in C.
and it gets exponential if the user has multiple tabs open with an app/page in each.
You keep using that word. I do not think it means what you think it means.
He's right if you think about it, because the JavaScript/XML engine in most browsers is within the same process or thread and more pages open means bogging down the system's memory and processing power even more. Multiplying would have been a better term for it, but keeping in mind the memory leaks in firefox and the process vs. thread model for tabs IE8 and Chrome seem to use, it is actually accurate.
I think the ARM netbooks are going to have a monster market, like eventually over 100 million a year.
That may sound crazy, but you have to look at the demographics. There are about 6 1/2 billion people in the world. About 1 1/2 billion are in the developed world or the richer parts of the developed world. They all have computers. At the other end are about a billion who are are desperately poor.
That leaves around 3 billion who are in-between. These are the people who have enough money to buy things like bicycles, motor bikes, televisions, and cell phones. A great many would love to own a computer, and indeed many of them spend a lot of time at cybercafes. But they can't afford the price. And there is another problem, namely that half of these people live in areas with no electricity, and for most of the rest the electric service is very eratic.
The first generation of netbooks was too expensive for this gigantic potential market, and besides they used too much electricity. But the new ARM netbooks will be enough cheaper for perhaps 500 million more people, and they will use far less electricity, too. Furthermore prices are just going to keep going down. Pixel Qi is planing on designing $75 models in a few years. Every time prices drop another huge group will join the market.
This all is a huge problem for Microsoft. On the one hand, it would hate to charge the very low license fees it would need to get anywhere in this new market, on the other hand it can hardly afford to ignore it.
Look if i'm going to get a laptop that uses ARM i'm not going to be able to run it off a small solar panel. I'm going to have to have a battery and charge it regularly, just not quite as regularly as an x86. If i'm going to be doing that i may as well just get an x86.
I'd like to see a laptop maker go to the extreme. eg. Try taking an MSP430 CPU and put it into a small laptop with a big passive LCD and a nice keyboard.
http://focus.ti.com/docs/prod/folders/print/msp430f5437.html
Stats
Ultralow Power Consumption
* Active Mode (AM): 165 ÂA/MHz at 8 MHz
* Standby Mode (LPM3 RTC Mode): 2.60 ÂA
* Off Mode (LPM4 RAM Retention): 1.69 ÂA
* Shutdown Mode (LPM5): 0.1 ÂA
Yes that's microamps (@1.8-3.6V). Basically it could run off a small solar panel like your calculator does. The CPU runs at up to 18Mhz. This is more than enough for the word processors and tools we used years ago.
In my opinion this kind of thing above is exactly what the OLPC should have been. It would cost 10's of dollars and is exactly what we in the Western world has to learn on years ago.
The OS I was referring to in my comment is just stock-kernel Ubuntu 8.10. The laptop is a dual core Pentium, 2GB ram, no swap partition used. I run compiz as a WM, for eye candy's sake, and this may contribute to the fan issue, but I just checked system monitor and Firefox alone takes up 384MB of ram and averages around 15% CPU. I found some Flash running in the ads of some pages, contrary to what I said above, as well as some JavaScript in a few tabs I forgot about.
If anything, the hard drive is the biggest bottleneck of the system, but I'm thinking about writing a piece on a news website I run (thecoffeedesk.com) about which browser would be best suited for a Linux installation on ARM. I'm going to compare memory usage, average CPU usage, and anything else that comes to mind (I'm open to suggestions). I want to compare Opera, Firefox 3, Seamonkey, and Epiphany in this respect to see which is better suited resource-wise even though I'm not running ARM.
I know benchmarks have been done before, but I'm out to see which browser could run with the smallest footprint while running full-blown web apps (I'll use Google Docs in my example, suggest others if you'd like). I don't have any idea how much memory comes on a standard ARM netbook, or what the default clock speed is, for that matter (and whether they use SSDs), so if anyone has more information than the Wikipedia article can offer, be my guest.
That last quote in the summary should read "...6-10 ARM-based netbooks running Linux and costing just around $200 should arrive this year starting in July and be done booting up sometime in early August"
Interesting that you say that about the Atom. The biggest fraud in things of energy usage, that I can remember.
Have you ever looked at an Atom mainboard? The big cooled thing is not the CPU. It is the freaking north bridge.
They just put as much of the CPU inside the NB, so it looks like it uses less power. In fact, if you add the NB, you still get a higher wattage than any low-power system on the market.
Any sufficiently advanced intelligence is indistinguishable from stupidity.
The problem with the PC industry is that it focuses on one or two numbers, without looking at how those specs actually creates a fast machine. It is like cars being sold by the number of cup holders. Computers are sold by the speed on the CPU and amount of memory. What is not talked about is whether it will function as a computer. Is the bus fast enough to keep the processor running. Is the memory fast enough. Is the memory available for CPU use, or is half of it going to the GPU. This si too much for many consumers to grasp but without it it can be hard to find a functioning computer. It is like cars with enough horse power and four wheel drive, but will tip over if and when the electronic suspension fails.
What we have is another useless arbitrary metric.
"She's a scientist and a lesbian. She's not going to let it slide." Orphan Black
...your average phone is more powerful than your average computer was in 1982.
Sonny, I was THERE in 1982 and I can tell you that my phone (an HTC Mogul) with its dual-core 400 Mhz ARM CPU knocks the socks off the 386 I had aound 10years later, around 1992! In fact, I can run DOSbox and run all the same games I used to play on my fire-breathing 386DX25 in emulation !!
If my phone today was released in 1982 it would probably have been considered a controlled military tool and banned from use by nonmilitary personnel!
Psssssttttt! Wanna guess what I'm typing this post with?
I have no problem with your religion until you decide it's reason to deprive others of the truth.
Moore's law isn't about what shall be done, or what should be done. It's about what can be done.
The size of a transistor on silicon has been steadily shrinking ever since the introduction of the first silicon chip. That's been driving down prices, and raising efficiencies. For example: the 6502 drew up to 160 mA at 7V. That's a little over one watt of power. The most power hungry Intel CPU on the market draws about 150 times that amount, but can do well over 150 times the work - getting hard facts is difficult, but I'd suggest four or five orders of magnitude, if not more.
So with the shrinking transistor, you can do three things. You can make the CPU more power efficient - able to do the same amount of work with less power. Or you can make the CPU more powerful - able to do more work, for the same amount of power. Or you can do both - able to do a bit more work, for a bit less power.
Intel has chosen to make their CPUs more powerful, at the cost of keeping the power usage high. ARM has chosen to make their CPUs a little more powerful, for a bit less power. Both are equally valid paths.
In the long run? Both choices will carve out their niche in the market place. There'll always be room for computing power at any price. But for the typical Joe Blow off the street, the ARM tradeoff - less performance, at a lower price - is more likely to be useful, assuming the software is there ... and with Linux, it pretty much is there.
ARM chips are famous for including special instructions and supporting silicon for things like MPEG4 encoding/decoding, MP3 encode/decode, etc. The "main" CPU core isn't involved in these "streaming" instructions, just the parameter setups for them. Given enough "heavy CPU" workloads implemented as custom silicon, the main CPU on an ARM chip can be relatively idle as all the heavy lifting is done by the stream coprocessors.
I do not fail; I succeed at finding out what does not work.
There are a couple of versions by Moore, but the two most common seem to be talking.
1) about the number of transistors that can cost-effectively be placed on a chip and,
2) about the density of transistors on a chip.
I don't see that the choice to produce a low-transistor-count chip has any more relation to this law than the fact that people still produce vacuum tubes does.
It would appear that you're really not aware of what's out there on the ARM department right now. Marvell is not the end-all, be-all for ARM processors, and the (relatively ancient) StrongARM CPUs are not even remotely comparable to what's on the market, in terms of performance.
Look at the Nvidia Tegra for a perfect example of ARM walking all over Atoms - per clock, per watt, and per actual performance.
There are a handful of other notable ARM chips out there right now which, while not comparable to the Tegra directly, offer considerable options above and beyond the Atom. Snapdragon and Tegra are just two examples; there are many others.
The performance is there, and has been there for quite a while. ARM chips do a LOT of things which an Atom couldn't come close to doing effectively (that fanless set-top box that does digital to analog conversion, or the DirectTV dvr, for instance).
~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers
OK, so WinCE/Mobile/whatever the hell isn't really Windows. It won't run all your apps. Linux won't either but is much more functional than Windows Mobile. Where will this leave MS with their strategy of forcing companies to bundle Windows instead of Linux on their Netbooks? What about the next OLPC which isn't supposed to have an Intel compatible processor either? Is this all a strategy to spoil MS's fun? I sure hope so!
"I have the attention span of a strobe lit goldfish, please get to the point quickly!"
No, it's not. The only way growth would be exponential is if each new tab caused each of the existing tabs to increase in size according to the number of open tabs. It is not accurate to describe the growth of memory utilization in a browser relative to the number of tabs as exponential.
You make a pretty good point. Couple that with that some websites are really poorly designed/implemented and there is a possible disaster. I mean, tech savy people will be fine, but I'm more worried about those who are not.
Example, when my partner finishes some web browsing, I can hear the fan in my current machine going flatout ... watch your cpu usage when you visit this site. There are many other sites out there but this one sticks in mind.
.
Look at this PDF, page 8, top left picture
It's actually from here
http://www.snopes.com/inboxer/hoaxes/computer.asp
That said, I suspect whoever wrote it was aware of the Snopes article.
echo -e 'global _start\n _start:\n mov eax, 2\n int 80h\n jmp _start' > a.asm; nasm a.asm -f elf; ld a.o -o a;
No he's not, wtf? It's linear, not exponential. Not even polynomial, just linear. Geez.
All your base are belong to Wii.
Wow, and I thought I was using some old hardware! But I have to agree it is probably his OS, not the hardware. I am typing this on a 1.1GHz Celeron maxed at 512Mb of PC100 running Win2K pro, and even with Firewall(Outpost) Peerguardian, Windowsblinds 4, and active HDD monitor I'm still able to have multiple tabs in FF3, watch Youtube, etc.
According to process explorer I still have nearly 100Mb free and am using 227Mb for cache. FF3 is the biggest user at 94Mb, but it isn't sluggish or slow. And the only time the fan kicks in is when Perfectdisk does the auto defrag. So if you are slamming the fan I would look at the OS, because something is seriously wrong there. Because even with all that stuff I am only hitting around 23% on a 1.1GHz Celeron. So if you are kicking the fan on a dual core browsing then something isn't set right in your OS or you have a hardware problem. Because you really shouldn't be hitting the CPU that hard.
ACs don't waste your time replying, your posts are never seen by me.
Can you please post a link to this benchmark? Otherwise i do not think you know what your talking about.
then, who cares about ARM. 10 years battery life for ARM based processors, 1 year for x86-based. :)
Google was contemplating compiling JavaScript to pure native code in a story I read here on /. a while back, but how well they would maintain this for both x86 AND ARM remains another story,
Yeah, God knows that nobody who has a browser that runs on x86 and a variant that runs on ARM has ever thought about compiling JavaScript to native code, and that Google haven't done more than just contemplate making a JavaScript engine that translates to machine code, much less made versions for both x86 and ARM, and, of course, the Firefox people haven't done anything like that, either.
If you do try this be sure to use a virtual machine or a real low powered machine. Web browsers adjust their memory usage based on what system they run on and with other factors like cpu speed it is very difficult to extrapolate from a fast machine to a slow one.
Just a minor nitpick - do you really think that y=a0+a1*x is not a polynomial function?
Ezekiel 23:20
Some ARM chips these days go far beyond what one would expect from a RISC micro-controller.
Two great examples are the BeagleBoard, which also uses the same hardware as the OpenPandora gaming platform. Both of the aformentioned devices are running a fully open-source Linux-powered stack, complete with a fully-functional Desktop environment. For developers, you can program for these devices using the standar GNU tools (c,c++), and of course several JavaVM's are already ported to the OMAP (JamVM,OpenJDK,Kaffee,etc)
.
The TI OMAP 3530 chip that powers the BeagleBoard and OpenPandora has a 3D graphics acceleration unit, a 2D (video) graphics acceleration unit, and a built in DSP for audio and general-purpose number crunching.
Without question a large majority of users do little more with their mobile computers than browse the internet, write email, and occasionally stream video - all things that are fully achievable with today's ARM chips.
ARM chip manufacturers are also starting to put multi-core chips on the market, and while the transistor complexity is nothing like Intel's out-of-order logic, multi-core, still means serious multi-threading. With clock frequencies approaching and exceeding 1GHz, I would not be surprised to see ARM take over the mobile and netbook markets, especially considering how little power they actually consume.
Furthermore, even high-end x86 laptops could benefit from having an ARM co-processor for instant-on and mobile operation, while the x86 processor could be used for more compute-intensive applications such as CAD or video processing. Imagine battery power extended by a factor of ten for everyone!
YouTube Videos:
BeagleBoard
OpenPandora
A ARM Cortex A9 dual core clocked at 1GHz, that is a top of the line ARM, would run rings around any Atom while consuming a fraction of the power.
The trick is that the ARM instruction set is *WAY* more efficient than the x86. The fact that the current ARM's are basically in order units is less important due to the design of the instruction set.
You are right about the power though. The ARM needs to be coupled to a low power chipset, but guess what these also exist as well.
It also needs a low power display. Now if I could just get a netbook with an Cortex A9 a GB of RAM, with 8-16GB of flash and a LCD from a OLPC XO-1 for 200 USD I would be well chuffed. I would expect such a netbook to have around 10-12 hours battery life.
Actually the faster arms are pretty much up to par with Atom. Dont underestimate ARMs, they basically are everywhere, almost every cellphone sold runs on arm and a load of other devices...
Before you all slate the ARM for performance vs the Atom, have you looked at numbers other then clockspeed? Clockspeed is frankly a stupid way of comparing processors, especially ones so wildly different as ARM and x86.
Mips are no good either because the instructions sets are so different. For instance many ARM instruction can be conditional and you can also shifts and rotates into the data processing instructions. It takes different amount of instructions to do the same thing. Read : http://en.wikipedia.org/wiki/ARM_architecture
"This results in the typical ARM program being denser than expected with fewer memory accesses; thus the pipeline is used more efficiently. Even though the ARM runs at what many would consider to be low speeds, it nevertheless competes quite well with much more complex CPU designs."
I can't find a good way of comparing the performance. I've no doubt Atom is faster then it's contemporary ARM processes, but honestly, I don't have any idea of the gap. Really we need the machines in front of us running the same bench marks.
I want a ARM-netbook as a pocket linux-machine/glorified-mp3-player. I don't care about x86 compatibility or Windows, I don't care if it's not quite as fast as a x86 option if the battery life is so much longer, that's more important.
He's right, ARM chips are not that performant per Mhz. Their two main weaknesses are in order execution and their RISC instruction set not playing nicely with instruction cache. Although the ARM instruction set is huge for a RISC architecture, it is not nearly as powerful or compact as x86. Variable length x86 instructions end up fitting better in the instruction cache.
Of course it must also be noted that the Dothan had 2MB of Cache, the Cortex A8 Remembering that the Dothan had 2MB of cache Cache misses, the tiny tiny cache L1 and L2 cache, rather ancient CPU design make ARM processors at least 1/2 as fast per mhz as Intel x86 chips. (Speaking in huge generalizations here...)
Pentium M's had a damn nice microarchitecture though. :) My Pentium M laptop was noticably faster than my Pentium 4 desktop...
I can't find how much cache the TI SDP 3430 using the A8 has on it, but the A8 only goes up to 1MB of L2 cache max, and I am guessing most manufacturers opt for less than that for cost and power saving reasons.
Need help treating your acne? Come here!
Just use a properly written software...Opera, for example, seems to be one of the favourites among the makers of low-power devices. While it's a little sluggish on Nintendo DS, I've heard it's visibly better on (faster) DSi and quite good on the Wii. It also runs really good on my 266MHz dual P2 with 192 MiB of RAM that I still boot up sometimes.
Those ARM netbooks will be somewhat faster, I guess...
One that hath name thou can not otter
The trick is that the ARM instruction set is *WAY* more efficient than the x86. The fact that the current ARM's are basically in order units is less important due to the design of the instruction set.
A minor correction here: The ARM instruction set is simpler and is much less feature rich than x86. This, combined with self imposed limitations (mostly around in-order execution) ARM (the company) is able to design CPUs with a much lower transistor count than x86 chip designers are capable of managing.
Not having a huge instruction decoder, having to do instruction reordering, or basically doing any of the things that makes x86 so damn fast, and staying a generation or two behind on manufacturing techniques to avoid leakage issues, enables ARM CPUs to have their amazing power profile.
The instruction set is kind of a mess really. It has been hacked onto a number of times, with the latest additions really showing signs of having wedged into the existing instruction encoding space. With features like the original (crappy) thumb, and now the fixed "Thumb2" (which we are all supposed to be calling Thumb and ignore the old Thumb, or something like that), and the fark-up that is ARM's floating point support (They have 3 implementations, 2 of which are still in use, and those two versions respond dramatically different to some basic key floating point operations), the ARM instruction set isn't nice per say, but ARM did the right thing and by keeping their eye on power consumption always.
(For those who are getting linguistically confused: ARM the name of the company, the name of their CPU line AND the name of their instruction set. Oh it is also the name of their reference manual, the ARM-ARM.)
Need help treating your acne? Come here!
A ARM Cortex A9 dual core clocked at 1GHz, that is a top of the line ARM, would run rings around any Atom while consuming a fraction of the power.
The only benchmark I've seen for the A9 was a comment that it's about ~20% faster than the A8. A 1.2 GHz A8 does not run rings around a 1.6 GHz Atom... it's slower.
:-)
I'll gladly be proven wrong, but in the mean time I'm not getting my hopes up
Can you please post a link to this benchmark? Otherwise i do not think you know what your talking about.
My google-fu fails me but there are benchmarks around for that A8 equipped Pandora like motherboard. I considered getting a Pandora which is how I came over the benchmark. If the A8 performs as well as some claims the Pandora will be great for handheld emulation. However, it's months later now and there's still no Pandora :-(
I find surfing on an ipod touch quite pleasant, and that has a 400MHz ARM. Battery life is pretty decent too, in my experience.
Most don't *need* to follow it, as is shown by the reassurance of lower power machines. ( horsepower )
95% for the world doesn't need to be on the 'Moore train', and what was available years before is more then they need. Blindly following It just breeds inefficiencies and wasted resources. Just because you can fly at mach 2 doesn't mean it makes sense for everyone to do it.
---- Booth was a patriot ----
We have heard that before, and the 'super cheap' never quite pans out and ends up 2x.
---- Booth was a patriot ----
One motivation for putting an ARM in a netbook is to make a better product (overall coat-performance-battery life trade-off), not just a cheaper one, so why would a manufacturer not put a decent/large amount of RAM in one? Have you checked RAM prices recently - it's practically being given away.
As far as web compatibility, note that the iPhone is ARM based and has a decent browser (youtube compatible since youtube switched to H.264 video), and incidently Adobe is trying to get Flash working on the iPhone...
FYI Google's new JavaScript VM is here in the Google Chrome browser... The JVM is called V8 and does indeed compile to native code when it wants to (JIT), and runs rings around other JavaScript implementations (such as that in Firefox) in terms of speed. Considering how simple ARM machine code is (it's a totally orthogonal instruction set)it's hard to imagine porting the JIT compiler to ARM would be a big deal. As far as security, JIT makes no difference.
I find surfing on an ipod touch quite pleasant, and that has a 400MHz ARM. Battery life is pretty decent too, in my experience.
I've heard that it was 533 MHz. In any case I've not used one. The closed is a mips based PSP, which worked well enough on some web pages while badly on others (like Slashdot :-)
This is no wonder, the Intel architecture is an arcane CISC architecture an in its state of instructions it has been always the worst in existence, over the years it has become better but from an operating instruction standpoint it still is one of the worst in existence. Way too few general purpose registers, an arcane instruction set which still is dragged around as legacy and on top of that a heap of new instructions some of them probably better off not being in processors at all added for marketing reasons.
Intel currently in its 64 bit incarnation has 16 bit general purpose registers while the 68000 CPUs already had 24-32. Now this is one area which improves speed significantly and it took AMD to finally add some registers intel never did. Instead they worked on making deeper pieplines etc...
The funny thing is that the entire x86 instruction set runs nowadays in a semi hardware vm on top of a risc core so intel is not too far off anyway but that does not change the fact from a high level point of view!
It was always beyound me why Intel never saw it as necessary to improve the number of registers, while every compiler writer knows how important exactly this area is for code efficiency!
I dont think Intel has the knowledge, they write compilers themselves, but in the end probably marketing won over engineering. You cannot sell registers, but you can sell GHz (hence the shoddy Pentium 4 design)
Arm however had a different start. It started off as Risc with a load of registers and everything added afterwards was added in the aspect of keeping transistor count and memory consumption down. (After ARM was dead on the desktop because Windows and Intel took over, over the back then far superior ARM/RiscOS combination)
ARM found its nieche and Intel seems to be scared to death, because there is one processor vendor selling millions of those things indirectly with power consumption levels they never can reach with their lousy design they have to drag along and those beasts finally get up to speed levels comparable to the Intel processors.
Intel rightfully should be scared, the only thing which prevents Intel from being shot out of the market entirely, is the stubbornness of the Users of wanting to have Windows run on any computer like device they have, also Netbooks.
While an ARM machine can run 10 hours straight, and has hibernation without any problems Intel does not.
Problem is that being able to leverage Windows is almost as important for Intel as it is to Microsoft, without it they probably would have been reduced to a semi important processor designer of the size Transmeta once was. And in the end I doubt that ARM will make serious inroads in the Netbooks market in the long run, althoug ARM based netbooks are far superior to their shoddy Atom based ones in respect of power consumption! (The speed probably is the same especially if you can offload heavy graphics lifiting to a dedicated gpu)
Exponential, sir? I highly doubt it.
With chaching, I'd said the growth is less than linear.
entropy happens
If compression speeds things up for you, then it's your connections that's slowing you down, not the processing power of your device.
This sub-thread is pointless. The reply to the OP talked about pre-compiling so browsers wouldn't have to interpret every single time. The next reply suddenly mentions compressed HTML, which has got nothing to do with the topic of this thread.
Of course it runs NetBSD. BTC: 1NT7QvbetmANwaMzhpVL6
well considering that its reducible to y=aX, no I don't. Unless I am not understanding what you wrote and the 0 and 1 were supposed to be sub or superscripts.
Repeal the 17th Amendment TODAY! Also Please Read http://www.gnu.org/philosophy/right-to-read.html
Opera is snappy on YDL6.1 on the PS3 too, though I usually use firefox.
Since I'm running a PPC Linux I wish there was a way I could use the H.264 youtube version the iPhone uses. There's workarounds with greasemonkey and Totem, but that's not as elegant as being able to do say: http://h264.youtube.com/watch?v=videofoo
At least your fans only kick on under load. On my system, the fans run all of the time, and there is no way to control that behavior. I can can tell them to go faster or slower, but not on and off. Of course, this is on an old Abit VT7 based on VIA chips and running an Intel 2.80Ghz/512/1MB Prescott. Under load, my system reads 29-31 degrees C tops. I even unplugged the side fan and removed the case door....
@Mindless Drivel: 100% of Twitter posts ever Tweeted.
PSP (not the most relevant example) might be 300 MHz now, what about PSPII, still 300MHz? Doubtful.
The first PSP ran at 222 MHz. (The 3.50 firmware upgrade unlocked 333 MHz with the WLAN off in games that require firmware 3.50 or later.) The PSP-2000 added RAM, but not much else in the sense of processing power. The PSP-3000 didn't add processing power either.
The fault on that is the PSP's crappy version of netfront, which isn't as good as even the PS3's crappy netfront. Slashdot's code doesn't help sometimes. You've probably encountered the "slashdot locks up the browser" bug. When I want to do serious browsing on my PS3, I boot into Linux. Sure, Netfront's okay for basics, but it's no Firefox. SCEfoo seems to have a thing for Netfront, they used it in the Japan only BBN for the PS2.
TI has a line (DaVinci) of arm + super-duper-dsp + special hardware that can both decode AND encode H264. at HD resolution.
It needs to perform at least on the level of a Dothan 600MHz before I'm interested - web surfing is already a pain at that level of performance.
The iPhone CPU is clocked to run at 400MHz, and it's browsing performance, is perfectly acceptable, even with Javascript.
Higher Logics: where programming meets science.
In addition to HTML, a web browser must parse/interpret JavaScript, Java, CSS, XHTML, Flash (if Adobe ever gets onboard), and regular XML just to display the modern, JavaScript-heavy web application.
Then browse Web sites, not Flash sites or Java sites. The HTML recommendation states that the object or applet element that starts an SWF object or a Java applet is supposed to contain other elements that will replace the object or applet should SWF or Java be turned off.
A server pre-parsing HTML would mean a browser/server handshake, something IE and IIS could easily do moreso than Apache(2)/Lighttpd and Firefox/Safari/Chrome.
HTTP/1.1 has a header called Accept: designed for this and other purposes. How would IIS make negotiating a text/x-parsed-html media type any easier?
(a bug in the JS parser leading to direct remote code execution, etc.)
Isn't such a bug called ActiveX?
But as far as MS breaking the Wintel relationship to pursue ARM-based netbooks, I don't see it happening unless something drastic happens.
Like Microsoft finally getting CE-rious about Windows Mobile, perhaps?
Of course, ARM chips used to power full desktop systems back in 1987 so they weren't always designed for low-power, mobile or embedded systems. At that time they beat the pants off Intel/Motorola too.
http://www.old-computers.com/MUSEUM/computer.asp?c=75
I doubt the latest ARM processors you are all talking about (multi cores, reaching 1Ghz) only got 100,000 transistors... I would like to get information about this, but couldn't find any...
Firefox alone takes up 384MB of ram and averages around 15% CPU. I found some Flash running in the ads of some pages, contrary to what I said above
Have you tried the Flashblock extension? It lets you whitelist YouTube, Newgrounds, and other allegedly worthwhile SWF sites while turning SWF objects on other sites into click-to-play buttons.
Minor nitpick - no such thing as a 400MHz P3; the P3 series started at 450MHz. Or you've got it underclocked :)
For the last time, PIN Number and ATM Machine are redundancies!
Welcome to the internet...
do you really think that y=a0+a1*x is not a polynomial function?
well considering that its reducible to y=aX, no I don't. Unless I am not understanding what you wrote and the 0 and 1 were supposed to be sub or superscripts.
It was fairly obvious to at least me that a0 represented the variable a with the subscript 0, and a1 represented the variable a with the subscript 1. Because Slashdot has been configured not to allow the sup or the sub elements in user comments, nor CSS that simulates them, Slashdot users have had to improvise different conventions to notate algebraic expressions. These conventions often draw from programming languages familiar to Slashdot users, which allow variable names to have letters or numbers after the first letter (e.g. "a0" and "a1) Numbers before a variable represent factors to multiply; numbers immediately after a variable (with no * in between) represent subscripts. Numbers after a ^ represent exponents. For example, 3a = a*3, but a3 = a with subscript 3, and a^3 represents a cubed.
Opera, for example, seems to be one of the favourites among the makers of low-power devices. While it's a little sluggish on Nintendo DS, I've heard it's visibly better on (faster) DSi and quite good on the Wii.
The Wii has a 729 MHz PowerPC G3 CPU, and Internet Channel still lags for nearly a minute to process the tags on stories when I load Slashdot's homepage. Is it a JavaScript problem or a reflow problem?
It's linear, not exponential. Not even polynomial, just linear.
It sure can feel polynomial once you open enough pages that the browser starts thrashing swap.
Although the ARM instruction set is huge for a RISC architecture, it is not nearly as powerful or compact as x86. Variable length x86 instructions end up fitting better in the instruction cache.
Are you talking with or without Thumb encoding? Thumb speeds up instruction fetches across slow buses because two instructions fit into each 32-bit fetch.
The battle between the Core and the Arm is about to begin! Finally I can put my stock of tins of baked beans to use!
The problem for Microsoft is they've already been there and done that. NT 4 had versions for non-Intel architectures but nobody was interested. Worse, if an ARM XP or Vista did appear nothing would run natively on it except .NET apps. Microsoft would probably have to implement x86 emulation because vendors wouldn't bother to target the device.
More likely is that Microsoft would be pushing some version of Windows CE. This would be a good fit except it would mean that anyone expecting a full desktop experience is completely screwed. All the apps would be the cut down sort seen in PDAs and phones. Maybe this wouldn't be so bad, but then again maybe it would.
Linux is in much better shape to deal with multiple architectures. Most of the source is already portable. It wouldn't escape entirely unscathed since some binary (and source) content like Flash player, Sun Java, WINE and some emulators mightn't work, but (memory / CPU speed permitting0 you could almost have a full Linux desktop experience no matter what the architecture was.
The iPhone CPU is clocked to run at 400MHz, and it's browsing performance, is perfectly acceptable, even with Javascript.
What one person finds acceptable is not true for everyone. I have never browsed on an iphone but I have used vastly faster computers. Though judging from people claiming that my cell phone has a responsive interface (when I think it's dog slow) my annoyance threshold must be lower than that of most people :-)
:-/
Miss my old celly. Everything instantly available within at most two clicks... could use it in my sleep, in fact that might be why it broke
Just to nitpick, V8 doesnt "compile to native code when it wants to". V8 is not an interpreter, but a pure JIT VM (unlike Java HotSpot which is both) and thus it cannot run code without compiling it first.
I don't remember the details from a group of computer architecture friends interested in forming a startup, but if you are interesting in licensing the ARM instruction set to develop your own ARM processor for sale, good luck with that. If I recall correctly, you get something like 12 months of exclusive rights, but then you have to give up your design to ARM. Someone correct me if I am way off here. Someone correct me if I am way off base.
Its going to be hard for ARM to kill intel. Thats roughly the same as saying x86 or itanium is going to kill intel.
Generally when people are buying a product your company makes, your company does better, not worse.
Guess no ones heard of XScale? What am I missing here?
Persistent Volume manager for Kubernetes - https://github.com/dwimsey/openshift-pvmanager
No, the proxy compresses it into something smaller and easier to process. It converts it to OBML (Opera Binary Markup Language) aka it pre-compiles it as mentioned higher up the chain. So a phone without the memory and CPU power to run a full rendering engine that supports HTML and CSS (and javascript) can probably still run the smaller more streamlined rendering engine of Opera Mini which doesn't have to deal with all the disparate formats out there on the web. It just has to deal with OBML.
On my phones, I can either use something that looks like this or use Opera Mini which looks like this.
Its a lot more helpful than mere compression.
1 (short ton / firkin) = 89.1432354 slugs / keg
Uhh, I was talking about orders of growth not functions.
All your base are belong to Wii.
I suppose it can sort of seem exponential when plugins like Flash or unruly javascript start to chew through a lot of memory, but in terms of order of growth, there's no way that opening a new tab somehow magically increases the memory usage of other tabs, or that the overhead for the second tab is exponentially higher than that of the first. If anything, the overhead per tab probably goes down. I think one main problem is how much Firefox caches in memory; eg. if I've opened 30ish tabs and a few youtube videos in HQ or HD, my memory usage can easily go past 300MB, and then closing them all doesn't do much for the memory usage. If it ever gets bad enough that it affects performance I usually just have to close it.
All your base are belong to Wii.
...and here I was thinking KDE4 (kmail, ktorrent, kopete), eclipse and firefox should be able to run w/o any swapping on a opteron 165 with 2gb of RAM, but for some reason they don't.
You're either lying or don't know what you have.
That said, I have yet to see any mobile browser which works acceptably with real world content. They mangle the layout and usually have half baked HTML, CSS, JS and DOM implementations which means web 2.0 sites (and even traditional ones) are just plain broken.
I agree, Eclipse with 400MHz and 256MB sounds impossible. Maybe he could launch it.. and swap out everything else on his system, but I doubt he could actually use it.
Yeah, but even your average phone is more powerful than your average PC was in 1982...
So "in the meantime" they will somewhat stick with the low+low, what happens when laptops, phones, GPS, etc all become the same device? People are content with the low power they have now, and with stuff like anti-jailbreaking etc, puts a limit on the push for better/faster/stronger because not many see it yet. People thought your body would fall apart at 50mph 100 years ago... "640kb ought to be enough for anybody"...
PSP (not the most relevant example) might be 300 MHz now, what about PSPII, still 300MHz? Doubtful.
I think the use of ARM is very promising and a really interesting approach. The ARM may not be quad Intel but if you look at the direction of cloud computing... it doesn't need to be. All it has to do is boot an operating system and a Browser and then connect to the more powerful resources available in the cloud. Look at what you can do now with just a browser and Amazon's EC2 AMIs?? use a browser, kick of 1 or 1000 virtual machines to work with. People have got to start thinking beyond the current Desktop paradim.
Then let da feet give you a hand there bud. As an old PC repair guy I have run into this problem. Hell I once even built a "super Frankenstein" for my boss with two mid tower case skeletons spot welded together(so he could hold a dozen SCSI drives) and didn't have anything but a CPU fan. Now this ain't pretty, but it is quiet and it works.
Since you already have the side off of it, all you have to do is this: go to Walmart(or Fred's or Family Dollar, etc) and buy yourself a $10 box fan. Place the box fan on its lowest setting and set it right beside your now open case. We called this "white trash supercooling" but it was the only way you could get rid of the noise with those old HP Pavilion "knuckle busters" mini towers that were popular between 98 and 2002. Now you will have to wipe the inside down once a month to keep the dust bunnies out, but you will find that your temps will drop to room temp and their won't be any need for anything but the CPU fan which will stay quiet.
Since I moved into a tiny apartment I used the "white trash supercooler" on an AMD Athlon 1.5GHz that was always running hot and noisy. Since using the white trash method my CPU is cool enough to touch even under load and running full blast my HDDs are reading 89 degrees f. But if your PC is noisy enough that you need to keep the side off(let me guess, mini tower? That design sucks for heat dissipation) this will get rid of the noise and extend the life of the hardware at the same time. At the lowest setting those cheapo box fans are whisper quiet. So give it a try. It ain't pretty, but if it could keep a pair of mid towers that were spot welded together cool with a dozen SCSI drives then I'm sure it won't have a problem with your P4.
ACs don't waste your time replying, your posts are never seen by me.
Agreed, the "other" architecture, MIPS actually leads ARM in certain areas, and is more "open" than the proprietary ARM. The 32MX series from Microchip kicks butt!
How about a 4-bit microcontroller like the National Semiconductor COP411? 512 Bytes of ROM, 32 nibbles (4 bits) of RAM, and 16usec instruction time baby!
Moore's Law is about the bleeding edge. Netbooks are about compromise. Different churches, no hereticism. C:\>
If all you have is a hammer, everything looks like a thumb.
Browsers already have a way to tell servers to preprocess things, thats what the accepts header is for.
While you could probably cut down a lot of poorly designed pages, one a site like google, they've already cut it down pretty good to keep their bandwidth costs down. When you function on the level they do, making your number of requests and content responses smaller is almost always cost effective, even if you have a team of 20 guys doing it. Since they are the ones most likely to push this sort of feature because it helps them most, and since its easier for them to just write good html, I don't really see anyone putting any effort into making preparsing a reality. The overwhelming response is likely going to be something along the lines of: make your pages with better/cleaner/smaller html. Which sucks when you are using someone elses website that you can't fix. :/
Persistent Volume manager for Kubernetes - https://github.com/dwimsey/openshift-pvmanager
Desktop and laptop ram is incredibly cheap, however it consumes more power then an ARM CPU. If you look at mobile DDR (what cell phone use) it is still almost $7-10 for 128MB.
in terms of order of growth, there's no way that opening a new tab somehow magically increases the memory usage of other tabs
In modern multiuser operating systems, RAM is in effect a cache for the swap file. Opening a page in a new tab enlarges the working set, and once this grows bigger than your RAM (e.g. 512 MB in a low-cost Linux subnotebook), you start seeing capacity misses in RAM. This increases the time to access memory associated with pages in other tabs once the machine starts having to hit the swap file more often.
Or it's a laptop, as there was a mobile P3 at 400Mhz (kind of rare though).
Actually, baths are bad. Specifically if you wash your skin, you remove the protective layer of oil - this is a fact.
Recently, a few years ago, they paid a bunch of students not to wash for a few months. They monitored them quite closely. After about 6 weeks or so, the skin stabilized in terms of pH etc..
In a "confrontation" about the time Malaysia was formed, British and New Zealand soldiers made many patrols. It was found that the NZers had a higher percentage of their soldiers reporting sick, but not because they wanted to avoid going on patrol (British officers had very high regards for the courage and endurance of the NZers). It was put down to New Zealanders generally being used to cleaner environments than the British soldiers.
-Gavin
ARM = "Acorn Risc Machine"