I'd rather define what's 'stupid for me to do' than have the government do it...
Can I put you in jail for not exercising today? How about that brownie you had? I'm sure you did somthing I'd consider stupid today... I want to stop you from doing that again! I'm you're friend!
IMO the real problem with Itanium is the compilers...
x86 decides how to do things in parallel in hardware (more or less)...
Itanium, while simplistic and clean, an architecture many academics would praise,
requires a compiler that is uber smart (for many tasks)... Itanium shines on ops where it's a 'no brainer' for a compiler to tell the chip how to do things in parallel.
Itanium 3 may have decent price/performance, but only if Intel sells them at a huge loss.
You say Itanium's been 'somewhat of a flop'... I'd call ~10billion spent for little revenue the biggest disaster ever in the semiconductor industry... I think Intel assumed they would be able to push everyone to a new (flawed IMO) architecture with their near monopoly.
Intel's sunk billions into Itanium; I hope they continue to push it (because I think it'll be a waste of cash and I have AMD stock!)
I'm assuming these guys were just running hubs on the machines.... I don't know why that would be illegal; a DC hub doesn't take part in data duplication, it just indexes the data.
Or did INDUCE pass without me noticing?
Sounds to me like all these guys are guilty of is facilitating copyright infringment (not a crime AFAIK). If it is a crime why isn't Microsoft, etc. guilty?
Re:welcome to commoditisation
on
You've Got PC
·
· Score: 1
You may be right. but I'd think a AL distro that'll do browsing, email, and some office apps might be an attractive buy for computerless folks (or those who want an additional pc).
Dumb it down, hide the console, make the ui for the few apps consistant and intuitive...
For a really casual internet user simply 'keeping it runnung' is a pain with windows/IE. You need a 2 hr class every year to keep a box 'halfass secure'.
There are so many frustrated casual users these days... I bet someone marketing a 'it just works and keeps on working' pc would have a chance... I've seen quite a few folkd buy new pc's not because they needed more, but because the software on theirs was hosed (spyware, etc.).
Well I guess apple kinda does that, but they're pricey. Cheap hardware for non demanding apps + free distro might have a market; throw in free updates at no cost; bandwidth is getting cheap.
"2) You have to ensure you are using the regenerative breaks instead of coasting to a stop."
BS. In order to use the regenerative breaking instead of coasting (on a flat surface), you'd need to expend gasoline to keep at cruising speed and then use regenerative breaking to recover some portion of that used gasoline as energy for your batteries (less than 100% of the energy).
Even in a hybrid, every time you touch the breaks you are hurting your mileage (unless you take collisions into account).
After rolling out ~35 new PC's at work (with user rights to the registry and c:\windows so our most used app will work) I was freaking amazed at how good some of our clueless users are at finding viri/spyware. If I put my mind to it I couldn't screw up a pc worse. Every time IE started (with the new xxx toolbar) around 30 popup windows with all sorts've educational pics came up.
In 24 hours, one machine had over 60 viri quaranteened and several pages of crap that spybot picked up.
After enabling immunize, their infection rate went to almost 0.
The reason for this multithreading per core is to reduce performance penalties while you're waiting for input. I think Sun's gone this route based on the assumptions:
1) Memory latency will be a bigger and bigger bottleneck in systems as cpu frequencies scale
2) Technology will not allow memory latency to keep pace with cpu frequency.
Chris Rijk [Ace's Hardware]: Stalled on waiting for data, basically.
Dr. Marc Tremblay: Yes. In large multiprocessor servers, waiting for data can easily take 500 cycles, especially in multi-GHz processors. So there are two strategies to be able to tolerate this latency or be able to do useful work. One of them is switching threads, so go to another thread and try to do useful work, and when that thread stalls and waits for memory then switch to another thread and so on. So that's kind of the traditional vertical multithreading approach.
The other approach is if you truly want to speed up that one thread and want to achieve high single thread performance, well what we do is that we actually, under the hood, launch a hardware thread that while the processor is stalled and therefore not doing any useful work, that hardware thread keeps going down the code as fast as it can and tries to prefetch along the program path. Along the predicted execution path [it] will prefetch all the interesting data and by going along the predicted path [it] will prefetch all the interesting instructions as well.
Both run x86 code pretty fast; Xeon can 'be for anything' too.
Try running a windows xp workstation on a dual Xeon system and you'll be very disappointed.
Only if I had to pay for it. Of course if I were in the market for a speedy dual workstation it'd most likely be an opteron due to price (and performance).
Got a Barton 2500 for $90 the other day. Changed one bios setting and it's run stable for ~1 month @ 3200. AMD heatsink.
On the other hand I have another Barton 2500 that doesn't like high fsb speeds... it's running about the same cpu clock as the one above at a lower FSB.
If you've got more time than $, overclocking still can make sense.
But it's a dirt dumb patent... a schoolchild could come up with the idea easily and then (if funded) pay some smart patent lawyers to patent it, and make it as brodly reaching as possible.
My conclusion:
(patents/copyright are now detrimental to innovation)
I'd be willing to be on the 'you can't pay? you die!' list if I was allowed to bypass the laws meant to protect me from myself.
Here and here
(btw filtering is off by default)
Can I put you in jail for not exercising today? How about that brownie you had? I'm sure you did somthing I'd consider stupid today... I want to stop you from doing that again! I'm you're friend!
Much better... :)
Sorry to hear about your accident; downtime sucks.
I don't, however, like the government forcing me to protect myself. Helmet laws stink!
IMO the real problem with Itanium is the compilers...
x86 decides how to do things in parallel in hardware (more or less)...
Itanium, while simplistic and clean, an architecture many academics would praise,
requires a compiler that is uber smart (for many tasks)... Itanium shines on ops where it's a 'no brainer' for a compiler to tell the chip how to do things in parallel.
Itanium 3 may have decent price/performance, but only if Intel sells them at a huge loss.
You say Itanium's been 'somewhat of a flop'... I'd call ~10billion spent for little revenue the biggest disaster ever in the semiconductor industry... I think Intel assumed they would be able to push everyone to a new (flawed IMO) architecture with their near monopoly.
Intel's sunk billions into Itanium; I hope they continue to push it (because I think it'll be a waste of cash and I have AMD stock!)
I'm assuming these guys were just running hubs on the machines.... I don't know why that would be illegal; a DC hub doesn't take part in data duplication, it just indexes the data.
Or did INDUCE pass without me noticing?
Sounds to me like all these guys are guilty of is facilitating copyright infringment (not a crime AFAIK). If it is a crime why isn't Microsoft, etc. guilty?
You may be right. but I'd think a AL distro that'll do browsing, email, and some office apps might be an attractive buy for computerless folks (or those who want an additional pc).
Dumb it down, hide the console, make the ui for the few apps consistant and intuitive...
For a really casual internet user simply 'keeping it runnung' is a pain with windows/IE. You need a 2 hr class every year to keep a box 'halfass secure'.
There are so many frustrated casual users these days... I bet someone marketing a 'it just works and keeps on working' pc would have a chance... I've seen quite a few folkd buy new pc's not because they needed more, but because the software on theirs was hosed (spyware, etc.).
Well I guess apple kinda does that, but they're pricey. Cheap hardware for non demanding apps + free distro might have a market; throw in free updates at no cost; bandwidth is getting cheap.
p.s. I hate AOL
Only need to backup 160GB (do not forsee that growing in 5 years). Gonna just buy two 160GB IDE HDD's & 2 firewire enclosures.
~$340 for both. Keep one plugged in for daily backup, keep the other in a safe place... swap them every month.
Pretty cheap, plenty fast, and won't take up much space!
Looks like $2k used.
I wonder how hard it would be to get MAME to use a force feedback wheel?
BS. In order to use the regenerative breaking instead of coasting (on a flat surface), you'd need to expend gasoline to keep at cruising speed and then use regenerative breaking to recover some portion of that used gasoline as energy for your batteries (less than 100% of the energy).
Even in a hybrid, every time you touch the breaks you are hurting your mileage (unless you take collisions into account).
After rolling out ~35 new PC's at work (with user rights to the registry and c:\windows so our most used app will work) I was freaking amazed at how good some of our clueless users are at finding viri/spyware. If I put my mind to it I couldn't screw up a pc worse. Every time IE started (with the new xxx toolbar) around 30 popup windows with all sorts've educational pics came up.
In 24 hours, one machine had over 60 viri quaranteened and several pages of crap that spybot picked up.
After enabling immunize, their infection rate went to almost 0.
It's not perfect, but it is a great help, IMO.
1) Memory latency will be a bigger and bigger bottleneck in systems as cpu frequencies scale
2) Technology will not allow memory latency to keep pace with cpu frequency.
See ace's previous interview
A snippet:
Chris Rijk [Ace's Hardware]: Stalled on waiting for data, basically.
Dr. Marc Tremblay: Yes. In large multiprocessor servers, waiting for data can easily take 500 cycles, especially in multi-GHz processors. So there are two strategies to be able to tolerate this latency or be able to do useful work. One of them is switching threads, so go to another thread and try to do useful work, and when that thread stalls and waits for memory then switch to another thread and so on. So that's kind of the traditional vertical multithreading approach. The other approach is if you truly want to speed up that one thread and want to achieve high single thread performance, well what we do is that we actually, under the hood, launch a hardware thread that while the processor is stalled and therefore not doing any useful work, that hardware thread keeps going down the code as fast as it can and tries to prefetch along the program path. Along the predicted execution path [it] will prefetch all the interesting data and by going along the predicted path [it] will prefetch all the interesting instructions as well.
So are Opterons at the moment.
wheras Opeteron's can be for anything.
Both run x86 code pretty fast; Xeon can 'be for anything' too.
Try running a windows xp workstation on a dual Xeon system and you'll be very disappointed.
Only if I had to pay for it. Of course if I were in the market for a speedy dual workstation it'd most likely be an opteron due to price (and performance).
Seems to me they'll go after any DVR producer who implements their fairly obvious time shifting features.
Will they drop the suit if Replay 'adds some thought of their own?'
From Yahoo,
:)
Shares Outstanding: 13.85M
Hmm so how much per share if SCO wins the IBM suit?
So 3bil/13.85mil = $216 per share...
So it's more like 10:1 odds.
Unless I made an error
I didn't... but I did it at ~$10.
Ack. Darl's puttin the squeeze on me. Unbelievable.
replacing outlook... Linux Journal has a good article on kgroupware.
Some high-resource digital filmers were using a MPEG-2 variant and big portable raid arrays for filming ~9 months ago.
Saw some setups for that at Comdex last year.
I do have some sympathy for your viewpoint, but it's duplicating, not taking.
I'm not knowledgeable w/crypto, but I haven't seen anyone question how much RSA encryption would:
:)
1) Increase amount of computing to save pic
2) Increase amount of time to save pic
3) Increase amount of power to save pic
Maybe they weren't so dumb... just naive
IMO it's still a good bang for buck with certain processors (usually the slowest on a given process)... i.e.
- Slowest P4
- Slowest Athlon XP (1700)
- Slowest Barton (2500)
Got a Barton 2500 for $90 the other day. Changed one bios setting and it's run stable for ~1 month @ 3200. AMD heatsink.
On the other hand I have another Barton 2500 that doesn't like high fsb speeds... it's running about the same cpu clock as the one above at a lower FSB.
If you've got more time than $, overclocking still can make sense.
I'm a happy overclocker.
You're not seeing the big picture.
Sure, this dork extorted $ from M$...
But it's a dirt dumb patent... a schoolchild could come up with the idea easily and then (if funded) pay some smart patent lawyers to patent it, and make it as brodly reaching as possible.
My conclusion:
(patents/copyright are now detrimental to innovation)
I'm stuck with Exchange 2k for a few more years... easy to put a linux box with this running in front of it?
Lots've local businesses here do a radio-station in the background for 'on hold' music.
If they're using VoIP (or their phone company is using TCP/IP somewhere between the callers) isn't this technically infringing?