Slashdot Mirror
A Transmeta Couplet
Godfather writes: "According to the heise-people the 600-MHz-TM5600 performs somewhere in between a Pentium III-400 and a Pentium III-600. It seems to be amazingly fast in memory access. The article is in German so you have to try the fish." A better translation would be appreciated, too, since Babelfish still leaves certain things murky. And if you've heard enough about the upcoming Picturebook, Timothy Brown writes: "Fujitsu is releasing (in early November) the Loox-T laptop, with a 500Mhz Crusoe chip. It's only available in Japan, but Dynamism, a company which sells Japan stuff to purchasers in the U.S., is accepting preorders." Here's that link.
http://www.spec.org/osg/web99/results/res2000q2/
Vendor: Dell
Model: PowerEdge 6400/700
Processor: 700MHz Pentium III Xeon
# Processors: 4
Memory: 8 GB
Disk Subsystem: 7 9GB 10KRPM drives
Operating System: Windows 2000 Advanced Server
File System: NTFS
HTTP Software
Vendor: Microsoft
HTTP Software: Internet Information Server 5.0
Conforming Simultaneous Connections (Median): 1598
Vendor: Dell
Model: PowerEdge 6400/700
Processor: 700MHz Pentium III Xeon
# Processors: 4
Memory: 8 GB
Disk Subsystem: 5 9GB 10KRPM drives
Operating System: Red Hat Linux 6.2 Threaded Web Server Add-On
File System: ext2
HTTP Software
Vendor: Red Hat
HTTP Software: TUX 1.0
Conforming Simultaneous Connections (Median): 4200
Of course, it all means nothing, I'm sure...
Right?
t_t_b
--
I think not; therefore I ain't®
I'm on PJ's "enemies" list! Are you?
THE CPUID INSTRUCTION DOES NOT VIOLATE PRIVACY.
repeat until you understand. most cpu's do provide cpuids. it's used mainly for licensing - and there are ways around that. obviously if a company can't depend on a cpuid for locally running apps, they'd be foolish to depend on one for non-local apps.
i am not an intel fan. but the cpuid fiasco was a tempest in a teapot. a marketing person at intel made a stupid comment - stupid for techincal as well as non-technical reasons. but marketing people are *supposed* to be stupid in any company! but the complete and utter stupidity of harping on it by "privacy activists" makes that marketing droid look downright brilliant in comparison.
so enough already with the cpuids. sun has had them since the 80's. macs have them (or they should). aix boxes have them, hp/ux, etc. big deal. it's not an issue. this windmill has been tilted at enough, go find something real to work on.
US Citizen living abroad? Register to vote!
True. Of course, the TM5400 also swallows part of the PCI bridge, and eliminates some of the other legacy devices that are normally out in the chipset by emulating them in software at the Code Morphing level. Both of these steps also help save power (fewer bus transactions, fewer off-chip peripherals) that wasn't being counted towards the CPU to begin with. (Incidentally, having part of the PCI bridge onchip may account for those impressive memcpy/memset scores.)
Still, this isn't quite the same as integrating the backlight onchip. ;-)
--Joe--
Program Intellivision!
When I saw the headline, I thought /. had sunk to a new low--I thought this was going to be a story about Transmeta poetry.
A bloody three-hour battery life.
That alone takes it out of the running.
The Psion 5mx gets a *FULL MONTH* off two AA batteries. And runs Office-compatible software.
Now to be sure, there are some deficiencies when looking at the Psion: relatively low-res, grey-scale LCD, and limited software selection. But it satisfies for about 90% of typical laptop use: wordprocessing, daytimer, calculator, web browser.
I'd like to see something that strikes in the middle: about a week on a full charge and a high resolution grey-scale screen, and more apps.
That'd make for a really great laptop.
--
--
Don't like it? Respond with words, not karma.
Good luck getting a week's battery life with a hard drive, colour TFT screen, and 500+ MHz Intel-compatible processor. Oh, the Psion doesn't have any of those, does it... Not surprising it has a completely different level of power usage, then.
:)
Try comparing apples to apples - it sounds like you would be better off using a WinCE notebook-format device, maybe with Linux on it since this is slashdot
still, double is over the top.
Preferential Voting: easy as 1-2-3
The CPUID instruction looks at a "function" number in a register. (EAX, if I recall). If it's one value, it returns what processor type it is. If it's another value, it returns a bitflag of known capabilities. If it's other values, it returns the range of legal values for this function index. New functions can be added by different processors.
In the Piii, CPUID supported a function to return a unique per-chip number. They later recanted, returning identical garbage or zeros on all Piiis.
To be compatible, Athlons followed suit: they made a damaged CPUID function akin to Pentiums.
The article SEEMS to be saying that the Transmeta emulates the Athlon's implementation, but that a NEW function is available to get unique per-cpu numbers.
[
That note was created in the 1940's by US computer scientists working with the imported Nazi scientists. The US scientists decided to have a little fun one day... and the rest, as they say, is history!
---
GetSystemMetrics(SM_SECURE) == FALSE
Also, I found the translation in post #28 much easier to read due to better formatting and more natural-sounding English (sounded less like a literal translation than #15).
So maybe you could shut up? Thanks.
Latin is great for talking about religion, and Caesar's conquests in Gaul, but since it is a dead language, it would be kind of hard to use it to discuss the latest in laptop technology.
Crusoe T-Labor transforming tM5600 of Transmeta (in Sony Vaio) must at this moment prove-lhea T-Labor. First Benchmarkergebnisse of the movement 600-MHz-TM5600 in the sector enters Pentium III-400 (approximately it it t-Apfelmaennchen with badly 15 million iterations to as) and Pentium III-600. The operation of the memory is total clean for the small notebook, MByte/s starts above approximately stops of MemCopy with the 60 and hangelt with of the blow hidden in the translations of intermediary of the storage of the transforming one in 170 MByte/s. The letter in the memory (for Memset) cría the transforming one with the 280 MByte/v. good with the matching, Pentium III 500 with of the Chipsatz iI is delayed distant in this only disciplines with the 70 respectively 150 MBYTE. Pentium-III-Coppermine-800 in the board i815 is approximately immediately with 190 and 255 MByte/s. It is so interesting as crusoe of the one of hashing gives a number of CPUID work (not to abschaltbar with Vaio). This number of the work of logical software that morphing CMS () only produced not obstante obvious. Commando " the truth ", (in a compatible way in Athlon aMD) does not announce the number of the work for the transforming one. In the other hand, Transmeta did not define to exist commando $$MPXCHG8 that a transforming one of the Pentium of the classroom worked perfectly with " " the official features, since that Windows NT comes beyond with exception to a stool, if (this capacity offers to the families - identification 5). T the following week will publish other values of consumición electric the power and the test test standard (as/c ' T)
Boycott this translation! Babelfish is closed source! You should be using GPLTrans instead! We cannot let the corporate hegemony of Slashdot dictate our translation ways! [recommended moderation: funny] But eh seriously... why doesn't GPLTrans ever get mentioned? I find in most of the cases with these non-english articles, it produces much more readable text than MangleFish.
Has anyone actually used one of those things? I currently am using a Sony Viao F409. It has a 650mhz PIII w/15" display. Very nice, and with 2 batteries I get almost 4 hours out of the thing (Its floppy drive swaps out for an optional second battery).
Pro's of a big laptop: It's nice when it's actually on your desk, cables all undone, plugged in, re-set my network connection settings.. (I don't use DHCP in the office or at home lan). Booted the machine, plugged in my external mouse and keyboard. (No docking station for me!)
The Picturebook seems nice, with exception that you don't have the large screen. (Only a half-hight ??). This might be ok if you are plugging into a monitor at the home/office
I have upgraded my machine to 256 megs of ram which is really nice, and I have RedHat 6.2 working nicely with my USB mouse. The main problem is getting my External monitor to work correctly in XWindows, but maybe that is probably more due to my ignorance in XWin configuration than the laptop.
I guess, the whole point of my babeling.. I have purchased and invested in what I thought would be the 'Dream' laptop. It's fast and nice with a great display that runs linux very nice... As now I spend really no time programming and all my time doing network diagrams, technical specifications and network administration on it.
But despite all the great hardware on this laptop, the reality is
Anyway, this is from someone that has invested a very large amount of memory into a laptop. Frankly I am a bit disapointed and am considering a picturebook to replace it.
Has anyone got any real practical experience using a picturebook? It sounds to me that the battery life/size and ease of porting it around. (Just toss it in my rook-sack and take it, battery lasts the whole working day and bring it home and charge it!) No cables.. etc.
Anyway, don't want to run down a fools paradise.
;-)
--------------------
> The "real" CPUID command, (compatible to the AMD-Athlon) returns no Serial-Number back.
Interesting - I thought the AMD Athlon has no CPUID `feature' - there was old recent issues with Red Hat 6.2 and Athlons, where the kernel would detect a PIII compatible chip pthis was before Red Hat started doing Athlon Kernerls] and attempt to disable the CPUID feature. Following finding a PIII, thre kernel would try to turn of CPUID [yay kernel people!] and die horrible, because AMD Athlons apparently didn't have one.
Or do they?
So we have a study of performance on NT, ONE study. As anyone who has been following benchmarking issues (c.f. recent IIS vs. Apache on Dell) will know, we can't really guess anything from this test.
What would be VERY nice is if some enterprising lab got together (ArsTechnica, overclockers.com etc.?), played with this beasty with a number of configs, a number of OSs (NT, Linux, BeOS, BSD, etc), and ripped out a few stats.
Hey, even a comparitive IIS vs. Apache on THIS machine would be fun. How would PHP work with all of this code-morphing cache stuff?
Still, I have to wonder how much input Linus has in all - the focus at the press release time was VERY much on NT, except for the embarrassing Quake moment. Anyone know of a publicly available set of Linux tests on one of these?
Can't wait 'til the webpads though. That'll sort out the "debian on iPaq" trolls...
--
We may be human, but we're still animals.
Another interesting fact is that the Crusoe processor supplies a serial number via CPUID... The "true" CPUID command (compatible with the AMD Athlon) does not yield a serial number.
I don't get it. They want to appear "privacy friendly" by not implementing the Intel Pentium III CPUID serial number code, but they are still per-cpu identifiable.
Which will come first, the big Linux-world outcry that Linus Torvalds works for a privacy sellout, or a piece of software that sniffs the Crusoe's cpuid-stained butt to target ads more effectively?
[
Well, my english is not perfect, so please bear with me =).
I hope this translation is better than what the fish blurps.
<Translation>
The TM5600-Processor of Transmeta (used in the Sony Vaio) has to prove itself in the c't-Test-Facality right now.
First Benchmark results show that the performance of 600-MHz-TM5600 is between the
performance of a PIII 400 (while using the Appleman-testprogram, with about 15 Million iterations / sec)
and a PIII 600.
The Memory performance is pretty good for such a small Notebook,
it starts with 60 MByte/s (MemCopy) , and goes up with Cache-Hits in the Translation buffer (to 170 MByte/s).
Writing to the Memory (using Memset) is at about 280 MByte/s.
To compare:
Pentium III 500 (Via-Apollo-II-Chipsatz) writes with 70 Mbyte/s and 150 Mbyte/s.
Pentium-III-Coppermine-800 (Solano-i815-Board) is about as good, with 190 MByte/s and 255 MByte/s.
An interesting thing is, that the Crusoe-Processor responds to the CPUID command. (Cannot be disabled in the Vaio.
But the Serial-Number seems to be calculated by the Code-Morphing-Software.
The "real" CPUID command, (compatible to the AMD-Athlon) returns no Serial-Number back.
There is no correct working CMPXCH8 command in the official specs,
because Windows NT depends on the wierd behavior of a non correct CMPXCH8 if it detects a Pentium-Family CPU (Family -ID 5)
(my note: AFAIK Intel first implemented a non RCF CMPXCH8 command)
</Translation>
Before you email me, remember: "There is no god!"
Crusoe in the c't labs
Transmeta's TM5600 processor (built into a Sony Vaio) is currently undergoing c't labs' scrutiny. The first benchmark results put it into the range between a Pentium III-400 (e.g. the c't Mandelbrot fractal at about 15 million iterations per second) and a Pentium III-600. Memory performance is impressive for a small notebook, starting with about 60MB/s for MemCopy and increasing to about 170MB/s for cache hits within the processor's translation buffer. Write performance (measured via Memset) peaks at about 280MB/s. In comparison, a Pentium III-500 with Via-Apollo-II chipset won't reach more than 70MB/s and 150MB/s, respectively. A Pentium III-Coppermine-800 on a Solano i815 board comes in close at 190 and 255MB/s.
Another interesting fact is that the Crusoe processor supplies a serial number via CPUID, which cannot be disabled, at least in the Vaio configuration. This serial number seems however to be generated exclusively by the code-morphing software. The "true" CPUID command (compatible with the AMD Athlon) does not yield a serial number. Additionally, Transmeta declared the otherwise flawlessly functional CMPXCHG8 command as not available in the list of official features, since Windows NT can't cope with a Pentium-class processor (family ID 5) supplying this feature.
Further benchmarks and power consumption measures will be published by c't during the next week.