Slashdot Mirror
Intel's Atom — First Benchmarks and a Full PC Review
Barence writes "PC Pro has received, benchmarked and discussed the first Intel Atom processor to be seen in the wild. A full analysis of the Atom processor itself is accompanied by a full review of the first PC — yes it's a PC, not a laptop — to use one. The benchmark results are pretty much as expected, but it's the power savings that really excite. And as a rep from the PC maker, Tranquil, joked — they could have left the Atom CPU uncooled if they'd really wanted to prove a point, as it's the old graphics chip that produces 70% of the heat coming from the motherboard. Exciting times ahead for the upcoming Atom-based Eee and friends."
MojoKid was one of several readers, too, to mention the upcoming
Eee Box mini-desktop from Asus (also Atom-based), which is supposed to start from $299, writing "although the actual dimensions are listed,
the image from ASUS' booth really gives a sense of scale. In the picture,
the Eee Box is standing next to a paperback book."
More info and benchmarks at http://anandtech.com/systems/showdoc.aspx?i=3321
those were the DEC boxes I think they were 166 Mhz Alpha's and they shipped with NT Alpha which was pretty much worthless unless you really liked to re-compile all programs. and didn't need support past SP4. so most boxes ended up with Linux pretty soon. Didn't Slashdot run on one in the early years?
I think you're thinking of the NetWinder. They were ARM based. I have a couple in my store.
The Atom is closest to the Pentium MMX than any other Intel CPU. It is in-order, for one thing, while every other Intel chip since the Pentium Pro has been out-of-order. It supports SMT, making it fairly unique among Intel chips (only the P4 did this before, and it has almost nothing else in common with the Atom), which helps avoid pipeline stalls caused by the lack of instruction re-ordering.
I am TheRaven on Soylent News
While the Atom certainly delivers impressive power statistics compared to our typical laptop processors, they are still far from the level of the ARM family. A recent article on Ars Technica will explain why. ARM processors are by far the most common processor on the low power frontier and the reason seems apparent; even at 1GHz they claim to reach operational power consumption around 300mW. Now, granted, it is on a RISC instruction set, but their upcoming Cortex-A9 will support multicore and starts to sound like a very interesting alternative for a notebook processor.
Could someone drop me a message as soon as those things start entering the market?
Snarky answers aside, MS is selling XP for miniature devices at a very, very low price, far lower than XP normally goes for. This allows OEMs to hit the low prices they want, as otherwise Windows would be a very big piece of the price. But Microsoft also had to keep the OEMs from installing this version of XP in place of a full version, so they set up fairly arbitrary limitations that ensure that it's only installed in such miniature (read: underpowered) devices. It's basically the same chain of logic as to why XP/Vista Starter Editions are so cheap; cheap Windows is for cheap devices, and hardware restrictions are a way to enforce that.
Also keep in mind that normal XP is also being retired (sales are ending) at the end of this month, MS doesn't want XP selling for so long that it's still in use in 2014 when long-term support ends, which might happen if it could be slapped on new high-powered computers after their cut-off date. This also spirals off in to the point that MS wants to retire XP sooner than later for API and security reasons.
You're mixing up two similar (in form factor) machines, that were otherwise quite different in architecture and time of availability.
One is the Alpha-based DEC Multia/UDB, from way back in the mid '90s. LITTLE-KNOWN FACT: Slashdot was originally run on one of these.
The other is the StrongARM-based Netwinder, which appeared around the year 2000.
They did have one thing in common other than their size - they both tended to overheat if they weren't stood up vertically.
OK, So I checked and to get a closer comparison of two new chips (the C7 is several years old now), Intel Atom (45nm) vs Via Nano (65nm).
Atom = 4 W.
Nano = 17W.
Keep in mind that the C7 has been shown to be faster than the Atom, and the Nano is twice as fast as the C7. On a performance/watt basis that puts Nano much closer to the Atom than even I thought.
I'd like to note that aside from the Core series of chips actually developing in terms of the microarchitecture, rather than just process shrinking an old design, the Atom is an all-new core from the ground up. It's a very different microarchitecture from the Core 2.