Use gentoo. No I'm serious. I run the latest ~amd64 software on my 2P opteron box and aside from the occasional portage hiccup (like moving to glibc-2.4 or xorg-11 v7) it works just fine.
I get the latest software, latest bug fixed vanilla non-vendor-locked-in kernel and things are smooth. I run a 3xSATAI RAID-5 in this box and I max out the drives in terms of performance using software raid.
Suse is good, like redhat, for one thing and one thing only. Reproducibility.
I install SLES9 and I get SLES9. I install gentoo today and gentoo tommorow and even with the same USE flags I could get different installs (unless you tarball the FS and use that to install).
What about loadable module support? What if you want to add a probe detection callback, etc, etc.
There are many reasons to change the module interface format. Many of which include the ability to do things we take for granted today.
What the kernel really lacks is a good standard for coding practices, like say adding comments and indenting at least somewhat sensibly [yeah I know for some of you "elites" you can take reading a complete lack of consistent indentation but for the rest of us...]
2.6.16 has had a lot of bug fixes [from 2.6.16.5 to the current version is pretty much fixes].
I've been running 2.6 since 2.6.10 [or so] without any significant problems or stability issues. x86_64 support was better initially with 2.6 than 2.4 as well.
Perhaps they should spend a few months fixing bugs but I wouldn't favour 2.4 over 2.6 any day.
Part of the problem is experience. Projects like GCC and the Kernel are split along two problems.
1. The underlying technology is non-trivial
2. The implementation often is dirty, quick and without consistent method.
In the case of #1 it's the case that the technology is not trivial. How many people understand paging?
In the case of #2 the code in many cases lacks comments, uses cryptic variables and the documentation [even doxygen style comments] are just not there.
Those two issues fight against anyone willing to throw in a weekend to help out.
Clocks are not a high percentage of the power. They're not trivial but mostly the problems with clocks is the length of the line. The bus between the register file and ALU is probably 1/20th that of the clock traces.
Compared to all the other logic in a cpu from the decoders to the schedulers to the ALUs, load-store, and then all the support pipeline registers, control logic, etc not to mention the cache...
The problem with "doing away with the clock" is being able to co-ordinate things in some usable amount of time. Each pipeline stage would need bidirectional signalling to co-ordinate the state transition, etc.
There already is some of that in the modern processors. When you do
MOV EAX,[EBX]
Provided there is no dependencies the scheduler will assume it takes 3 cycles [on the AMD side] to complete. It will then stall the ALU for two additional cycles before attempting to feed something in. But the MOV may not be finished so there is some need for feedback.
That in mind, if they are truly 1-cycle ops the scheduler will pump out instructions without waiting. The write to the register file is on a clock edge, etc, etc, etc... A lot of things "just assume" data will be ready. Similarly talking with other things like memory will have to get into a clock domain making things a bit more complicated.
The fact of the matter is asynchronous circuits are not new. They're just not space efficient. You get the efficiency of running as fast as possible (e.g. if your ADD takes 0.25ns and your ADC takes 0.35ns then executing ADD will be faster as the ALU will signal it's finished 0.10ns sooner) but waste a lot of space with the syncronization steps.
If you look at the ARM case they maxed out at ~80MIPS or so. The typical Athlon gets ~1MIPS per Mhz at a minimum and upto 2MIPS per Mhz on more efficient code. So a 2Ghz processor is netting upwards of 2000MIPS. Sure it takes more power but if you install the requisit 25 ARM cores the "power efficiency" drops quickly.
That said, there are good uses for that core. It leaks less RF energy as it's not pulsing at a fixed frequency. That is, it does leak RF, just it's spread over more of the spectrum. Also if the cpu is idle it effectively is not switching which reduces power.
Intentionally unhelpful... I'd say that's par for the course for those stores...
Them: "Can I help you?"
Me: Yeah, actually stock cabling!!!
I don't know how many times I've gone there looking for some RCA or S-Video cable to find out all they have are the 6" long cables or the MONSTER diamond studded $300 an inch cables.
It's nice that you sell 15 flavours of DVD players, but if you don't sell cables to hook them up what's the fucking point? You can reapply this with various computer gear.
The problem is they're not an electronics store. An electronics store sells devices and their accessories. Best Buy just sells the big ticket items and the media (e.g. movies and shit).
iirc NTP took what 1/3 or 1/4 of the cash reserve.
RIM going all psycho is not just to get revenge [well directed at someone else] but because they can't afford to lose again.
Frankly this makes me happy. I hate crackberry syndrome. It does change people, not always for the better.
More than once I would be interrupted during an interview to have some little dork read his email. Sure I'm not a rockstar so probably not 100% entertaining but during a Job interview when you ask a question, listen for at least 30 seconds for an answer....
It also happend to my brother. Used to be able to talk to him while driving or at a restaurant. Now he's reading SMS and e-mail all the time.
Meh... maybe I'm just not that interesting. I'll invest in a wig or something...
Fundamentally a good product [aside from those that feed addictions] are what people need not just want they want.
When it comes down to it, what in Vista do users actually "need"? I look at Vista and see a larger, shinier, fundamentally no better OS that is segregated into markets artificially to create the sense of premiums.
At some point MSFT has to come to the realization that the market is driven by the customers true needs and not just what some press agent drones up in advertisement as their needs.
It probably has a lot of meta data in it like changes, styles, fonts, etc.
I'm not defending word. I just made a blank document, e.g. CTRL+N CTRL+S test.doc [enter]. It's 23KB on disk. On the NTFS drive it's compressed to less than 4KB. So it's probably a whole bunch of reudndant crap. That said it's not super bad.
Like I said my text is less than 400KB for nearly 60 pages.
Of course there is a boat load of things Word does poorly [typesetting, formulas, etc] which it tries to make up with by having a half decent grammar checker and perdy GUI...
I'm writing a book in Word [yeah I know, shudder] and my 57 page 2nd chapter is about 340KB on disk. It sports 10 figures, lots of styles (from normal paragraphs, to emphasis to source code etc...)
Maybe you put high res graphics and are using tracked changes?
Yes. When I click on html files on my desktop I can't use PGUP/PGDN to scroll, if I open a new tab [say maps.google] I can't type without it going in the find field.
I use mozilla. I just don't has flash installed. Truth be told though did they actually use Flash for something highly scripted or interactive? Or was it just their really really neato super cool uber sweet way of doing a simple button menu?
People like that were probably hypercard junkies in the 80s and are getting their fix today.
That's the problem though. Have a degree nowadays really means nothing. I went through college. I know how a lot of students act. Cheating on lab assignments, cramming for exams the day of the exam, etc, etc.
You can't judge employees ability based on their education. You have to see what they've done before.
That's why OSS is a great thing. You can make up your own project or get attached to a project without really the huss and fuss of a "hiring process". It also shows initiative and the ability to perceive a niche market [e.g. a tool people need].
Over there or over here, a college degree basically means the same thing.
Slashdot Mirror
Use gentoo. No I'm serious. I run the latest ~amd64 software on my 2P opteron box and aside from the occasional portage hiccup (like moving to glibc-2.4 or xorg-11 v7) it works just fine.
I get the latest software, latest bug fixed vanilla non-vendor-locked-in kernel and things are smooth. I run a 3xSATAI RAID-5 in this box and I max out the drives in terms of performance using software raid.
Suse is good, like redhat, for one thing and one thing only. Reproducibility.
I install SLES9 and I get SLES9. I install gentoo today and gentoo tommorow and even with the same USE flags I could get different installs (unless you tarball the FS and use that to install).
Tom
What about loadable module support? What if you want to add a probe detection callback, etc, etc.
...]
There are many reasons to change the module interface format. Many of which include the ability to do things we take for granted today.
What the kernel really lacks is a good standard for coding practices, like say adding comments and indenting at least somewhat sensibly [yeah I know for some of you "elites" you can take reading a complete lack of consistent indentation but for the rest of us
Tom
2.6.16 has had a lot of bug fixes [from 2.6.16.5 to the current version is pretty much fixes].
I've been running 2.6 since 2.6.10 [or so] without any significant problems or stability issues. x86_64 support was better initially with 2.6 than 2.4 as well.
Perhaps they should spend a few months fixing bugs but I wouldn't favour 2.4 over 2.6 any day.
Tom
Part of the problem is experience. Projects like GCC and the Kernel are split along two problems.
1. The underlying technology is non-trivial
2. The implementation often is dirty, quick and without consistent method.
In the case of #1 it's the case that the technology is not trivial. How many people understand paging?
In the case of #2 the code in many cases lacks comments, uses cryptic variables and the documentation [even doxygen style comments] are just not there.
Those two issues fight against anyone willing to throw in a weekend to help out.
Tom
Clocks are not a high percentage of the power. They're not trivial but mostly the problems with clocks is the length of the line. The bus between the register file and ALU is probably 1/20th that of the clock traces.
Compared to all the other logic in a cpu from the decoders to the schedulers to the ALUs, load-store, and then all the support pipeline registers, control logic, etc not to mention the cache...
The problem with "doing away with the clock" is being able to co-ordinate things in some usable amount of time. Each pipeline stage would need bidirectional signalling to co-ordinate the state transition, etc.
There already is some of that in the modern processors. When you do
MOV EAX,[EBX]
Provided there is no dependencies the scheduler will assume it takes 3 cycles [on the AMD side] to complete. It will then stall the ALU for two additional cycles before attempting to feed something in. But the MOV may not be finished so there is some need for feedback.
That in mind, if they are truly 1-cycle ops the scheduler will pump out instructions without waiting. The write to the register file is on a clock edge, etc, etc, etc... A lot of things "just assume" data will be ready. Similarly talking with other things like memory will have to get into a clock domain making things a bit more complicated.
The fact of the matter is asynchronous circuits are not new. They're just not space efficient. You get the efficiency of running as fast as possible (e.g. if your ADD takes 0.25ns and your ADC takes 0.35ns then executing ADD will be faster as the ALU will signal it's finished 0.10ns sooner) but waste a lot of space with the syncronization steps.
If you look at the ARM case they maxed out at ~80MIPS or so. The typical Athlon gets ~1MIPS per Mhz at a minimum and upto 2MIPS per Mhz on more efficient code. So a 2Ghz processor is netting upwards of 2000MIPS. Sure it takes more power but if you install the requisit 25 ARM cores the "power efficiency" drops quickly.
That said, there are good uses for that core. It leaks less RF energy as it's not pulsing at a fixed frequency. That is, it does leak RF, just it's spread over more of the spectrum. Also if the cpu is idle it effectively is not switching which reduces power.
Tom
Yeah, Athlon and Turion ... how unoriginal.
The "code" names can be boring but they're not meant to excite customers.
At least the product names are ok
"favor" would be an OR sound. Which is not how the word is pronounced.
Tom
No and for a good reason. Displays need in order of priority
1. Bandwidth
2. Realtime response
3. Low latency
They're also unidirectional.
PCI-E would give #1 but not #2 or #3.
So it makes sense that a digital bus specifically optimized for the nature of the traffic be designed.
Tom
American English in the press is really poor nowadays. I've heard "Proper noun and me are ..." many times as well as the "he did really good at VERB".
...
Then there is the whole "missing the u" bit
I think the problem is the media feeds off the people and the people off the media. It's a vicious cycle of stupidity and I want off!
Tom
Intentionally unhelpful... I'd say that's par for the course for those stores...
Them: "Can I help you?"
Me: Yeah, actually stock cabling!!!
I don't know how many times I've gone there looking for some RCA or S-Video cable to find out all they have are the 6" long cables or the MONSTER diamond studded $300 an inch cables.
It's nice that you sell 15 flavours of DVD players, but if you don't sell cables to hook them up what's the fucking point? You can reapply this with various computer gear.
The problem is they're not an electronics store. An electronics store sells devices and their accessories. Best Buy just sells the big ticket items and the media (e.g. movies and shit).
Makes me glad Radioshack is still around...
Tom
iirc NTP took what 1/3 or 1/4 of the cash reserve.
... maybe I'm just not that interesting. I'll invest in a wig or something...
RIM going all psycho is not just to get revenge [well directed at someone else] but because they can't afford to lose again.
Frankly this makes me happy. I hate crackberry syndrome. It does change people, not always for the better.
More than once I would be interrupted during an interview to have some little dork read his email. Sure I'm not a rockstar so probably not 100% entertaining but during a Job interview when you ask a question, listen for at least 30 seconds for an answer....
It also happend to my brother. Used to be able to talk to him while driving or at a restaurant. Now he's reading SMS and e-mail all the time.
Meh
Tom
Fundamentally a good product [aside from those that feed addictions] are what people need not just want they want.
When it comes down to it, what in Vista do users actually "need"? I look at Vista and see a larger, shinier, fundamentally no better OS that is segregated into markets artificially to create the sense of premiums.
At some point MSFT has to come to the realization that the market is driven by the customers true needs and not just what some press agent drones up in advertisement as their needs.
Tom
It probably has a lot of meta data in it like changes, styles, fonts, etc.
I'm not defending word. I just made a blank document, e.g. CTRL+N CTRL+S test.doc [enter]. It's 23KB on disk. On the NTFS drive it's compressed to less than 4KB. So it's probably a whole bunch of reudndant crap. That said it's not super bad.
Like I said my text is less than 400KB for nearly 60 pages.
Of course there is a boat load of things Word does poorly [typesetting, formulas, etc] which it tries to make up with by having a half decent grammar checker and perdy GUI...
Meh...
viva la ODT!
While we are bragging about our Linux boxes...
tom@bigbox ~ $ uname -a
Linux bigbox 2.6.16.12 #1 SMP PREEMPT Tue May 2 14:09:55 EDT 2006 x86_64 AMD Opteron(tm) Processor 885 GNU/Linux
I'm writing a book in Word [yeah I know, shudder] and my 57 page 2nd chapter is about 340KB on disk. It sports 10 figures, lots of styles (from normal paragraphs, to emphasis to source code etc...)
Maybe you put high res graphics and are using tracked changes?
Tom
Solution is to use IE. /me ducks ...
I have no idea, I haven't filed a bug report because frankly I don't really care (that and bugzilla bothers me).
Tom
Yes. When I click on html files on my desktop I can't use PGUP/PGDN to scroll, if I open a new tab [say maps.google] I can't type without it going in the find field.
Tom
it's if you want all the addons.
A lot of the styles/classes are not part of the base install.
Tom
There is TeX for Windows. It's called MikTeX.
It takes forever to fully install but works decently fine.
Tom
No. Wrong. Using shared memory bad.
Tom
I use mozilla. I just don't has flash installed. Truth be told though did they actually use Flash for something highly scripted or interactive? Or was it just their really really neato super cool uber sweet way of doing a simple button menu?
People like that were probably hypercard junkies in the 80s and are getting their fix today.
Tom
Yeah...
I'd explore the website except their webdev team is stupid.
Flash is meant for things Flash is meant for. Not menus.
That's HTML.
Tom
That's the problem though. Have a degree nowadays really means nothing. I went through college. I know how a lot of students act. Cheating on lab assignments, cramming for exams the day of the exam, etc, etc.
You can't judge employees ability based on their education. You have to see what they've done before.
That's why OSS is a great thing. You can make up your own project or get attached to a project without really the huss and fuss of a "hiring process". It also shows initiative and the ability to perceive a niche market [e.g. a tool people need].
Over there or over here, a college degree basically means the same thing.
Tom
TurboCache makes sense if all you need is a 2D VGA device. It's a cheap design.
... might as well go with the 6600. Or given the price the 7600 is a good idea too.
If you plan on running even the most mild GL program on it though
Tom
Um not really. It's up to me to choose what I apply to my software not RMS.
RMS is a proponent of GPL. So am I. Doesn't mean I should get credit for all the GPL software out there.
Tom