Slashdot Mirror
Transmeta To Release Next Generation CPU
CodeShark writes: "According to this story at CNN, Transmeta is set to release their new TM6000 microprocessor this afternoon. The chip apparently incorporates some of the functions usually provided by high-performance (and high price!) chip sets. Transmeta is reporting a further reduction in power requirements by 44% and sees the laptop and sub-laptop markert as the primary markets for their new CPU. Intel and AMD claim to be catching up with the Transmeta chips in terms of power requirements, I'd be curious to find out what the real world comparisons might make of those claims ..." If anyone out there is at Microprocessor Forum, please say in comments any further details that are made clear there.
Ditzel said Transmeta will prove, despite Intel's claims to the contrary, that the TM5800 beats Intel's lowest power chip by a factor of 2 to 1. "And when we go to our highly integrated chip, we're going to take off another 44 percent," he said. "So we think we've got a substantial lead today, and we're going to keep that."
And yet when we look at these laptops with their lower power processors, there is VERY little added battery life, for the simple reason that the processor is not the major consumer of power in a notebook.
When you factor in that the processors are much slower than the equivalent Intel or AMD (by how much varies by who you ask and what you're doing), and there doesn't seem to be any price break, why would anyone want to use a Transmeta processor?
Transmeta needs to stop trying to sell me that they are "more l33t than Intel" and show me products that are SIGNIFICANTLY better. If they can give me, say, twice the battery life it might be worth switching to an off-brand processor that is much slower.
Sometimes it's best to just let stupid people be stupid.
The problem is that the Transmeta chip will optimize programs and make them faster when compared to the unoptimized first run of a program. It is possible to speed up subsequent runs by saving recent CPU paths instead of re-translating the instructions each time, essentially 'hardwiring' the processor to do a certain set of instructions with no translation step.
Compared to an actual Intel or AMD CPU that actually has these instructions hardwired, the Transmeta chip makes a pathetic showing.
From what I've heard, that 8 hour benchmark came from marketing. I don't believe the Transmeta chip can quite stack up to that. Good news is that it does well (4 or 5 hours?) against the power hungry Intel/AMD offerings. Results: less heat, longer battery life, smaller package. Sony has a nice subnotebook running a Crusoe chip, looks tempting but pricey. Transmeta will need a few more years to be taken seriously but given time they'll start showing up in cell phones, PDAs, car stereos, etc without enlarging the package.
There is no reasonable defense against an idiot with an agenda
:wq
I think a fairer comparison would be performance/watts rather than a synthetic bench that doesn't stress how much work you can do. It should also take into consideration the support chips that other traditional CPUs require (it looks like they're building in a bunch of other stuff that you'de need secondary chips for on Intel and AMD).
You poor, deluded, digestable protein.
... oh god i can't continue. This is like cleaning up someone else's vomit, and it's tripping my gag reflex.
Having just left Transmeta for gamier pastures, I assure you that the people writing press releases, designing websites, and manning show floors for VIA (Centaur), Intel, AMD, and Transmeta are after only one thing- money.
All of their "issues" and "features" are make-believe. They are fly vomit, meant to turn consumers into a common, runny soup of stupidity, that can be slurped without the need to chew on issues.
Speedstep is not a joke. It's a cheap, excellent hack, far easier to verify and debug than PowerNow or LongRun. Intel enjoys most of the power savings afforded by LongRun simply by implementing APM and getting the same job done faster than the p95 and therefore going to sleep sooner. Sensible, mundane, and vomit-free, but true nonetheless.
LongRun has problems with all kinds of applications featuring unpredictable loads. And so does APM. Each is good at a certain set of applications, but neither is clearly superior. And to overlook the critical importance of your choice of operating system, southbridge, video card,
Food, reconstitute thyself. Intel and Transmeta are in a deadly competetive battle. They are slitting their own wrists to give you 5% here and 3% there and need fly vomit because the numbers 3 and 5 don't sell product. Listen to your friends. Try a Transmeta notebook. Try an Intel notebook. You will like what you like. End of story. Every portable is completely different, no matter which CPU you use. Read reviews, friends, and personal experience, not corporate web sites.
about this whole Transmeta thing is the level of speculation and un-clearness.
/ 28365403.pdf. The datasheets says thermal power 10.1 Watt max. Well, we never _ever_ get that high. Also, the newer 500 Mhz ultra low power is 8 Watt max, 5 Watt under more normal conditions.
Talk all the shit you want about Intel, but I can tell you that I'm working on a board right now that uses a Mobile Celeron Mobile 400A: http://developer.intel.com/design/mobile/datashts
The thing is that TM _never_ published said figures (quickly: what's the MAX Watts a TM CPU can draw?), because supposedly all that we need to know is the power required to decode a DVD. Well, today that happens largely by the VGA controller now, doesn't it?
What suprises me even more is that Torvalds, if anyone, should know that using the simple HLT instruction in the idle thread, makes any Intel (or AMD) CPU draw a lot less power.
Even on paper I don't see the advantage of the TM CPU's. And I really hoped they would, believe me...