Bad enough when readers don't RTFA, but the submitters/editors???
Where in the world could anyone have gotten "world's first" from here? The article makes it pretty clear that this is an upgrade to an existing design and that plenty of similar tokamak reactors already exist.
Even at full capacity, it would've taken 10-20 years of taking all of Iraq's oil profits (or it may even have been total net sales...) to pay for the initial cost of the war. Iraq's oil fields aren't running anywhere close to full capacity due to initial damage from the war and constant ongoing damage from insurgent activity.
Note that by "initial cost", I mean the initial 80-100 billion that Bush requested for the war. What's the price tag up to now? 200b? 300b? It's a hell of a lot more. Plus there's the cost of upgrading/rebuilding Iraq's oil production infrastructure.
If this was about oil, it was a damned stupid financial decision.
"All that proves is that the screen and the piece of paper say the same thing. How do either of those relate to the actual value recorded as the vote?"
It doesn't. But the original posters' point was that if there is any suspicion of discrepancies/errors/hacking, the "system" (meaning the whole election process) can fall back on a more traditional/reliable method (paper votes).
Paper ballots have their own problems, but in general it's a different set of problems than the ones in electronic systems.
To get into a console, you can't win if you fall into categories 1 or 2. If the Cell were that expensive or underperforming, people wouldn't be putting it into consoles.
Now, there is one thing about the Cell you missed - It's a special-purpose processor designed for raw floating point performance. 8 of the cores can only do basic streaming floating point (although they do it EXTREMELY fast), the remaining CPU is a VERY stripped down PPC.
So for a gaming system or DSP, the Cell will kick ass. For general purpose computing, it's going to suck.
Do I need to say more? OK, I guess I can - EMT64? StrongARM? (Yes, they bought that from DEC, but rather than kill it and go with one of their own designs they stuck with the better design. The StrongARM lives on in the PXA series.)
As another person pointed out, SOI may simply be too expensive for Intel to license from IBM. Does AMD use SOI? I don't believe so.
Not all companies go for showing the "absolute best case" benchmark. AMD is generally rather conservative with their performance ratings, and it's pretty rare that third-party benchmarks show an Athlon with a given performance rating having significantly different average performance than the "baseline comparison" CPU (One of the older P4s) running at the same clock rate as that performance rating.
A great example of just how conservative AMD is - The Venice core Athlon 64 3200+ has a 2.0 GHz core clock and 512k of L2 cache, using a 90nm process. Its closest dual-core variant (the Manchester core X2 3800+) has the same core clock, L2 cache per core, and manufacturing process. (They also have the same FSB speed, 1 GHz HyperTransport) Yes, that's right, the dual-core variant is only rated 18% higher than its closest single-core counterpart. (This is because currently, on average, a second core usually doesn't net you much benefit because so many CPU-intensive tasks do all the work in a single thread.)
Apple, on the other hand, is notorious for being overly optimistic in their speed comparisons - They always pick the benchmark which will make the competition look as bad as possible, to the point of even failing to use important performance features of the competition's CPU. (For example, back in the P2/P3 era, Apple constantly marketed their systems as being faster than a P2 or P3 with twice the clock speed - While the PPC did in general perform somewhat better per clock cycle than Intel's CPUs, the difference was not anywhere close to what Apple claimed it to be. The benchmark in question used Altivec on the PPC but failed to optimize for Intel whatsoever - No MMX or SSE was used, despite being available.)
To compare it to my previous example, Apple would have called the Athlon 64 X2 3800+ a 6400+ because it had two cores equivalent to the 3200+.
When it comes to inflated/BS benchmarks, Apple is one of the kings.
Cable companies are different in different parts of NJ, as another user said.
I'm in a Cablevision service area that was formerly served by a small company called TKR Cable. (Cablevision bought them out.) Let's just say that TKR sucked, to say the least. CV has been a massive improvement, but there's some legacy TKR cruft in Cablevision's pricing and tier organization in our area.
While digital cable may not be required for most channels, Sci-Fi is now only transmitted in encrypted digital in my aree. Well, my former area. I now live in upstate NY, and Time Warner kicks Cablevision's ass right and left.
For customers of Cablevision in New Jersey, getting Sci-Fi costs you $80-90/month, as for some reason it's in Cablevision's "silver" premium digital tier.
I'm not talking a package deal ($80-90 for cable modem plus cable service), I'm talking $80-90 for the TV service ALONE. Cable modem is another $40ish (plus/minus 5-10 dollars, I don't remember the exact amount.)
Cablevision sucks. A channel with commercials should NEVER be put into a top premium tier. Sci-Fi is the ONLY channel in the silver/gold premium tiers from CV that has commercials.
From what I recall, people in California basically pioneered the recycling movement.
Guess what - apparently almost no one in California recycles. (I could be wrong about this, I'm going on anecdotal evidence from relatives/friends.)
Meanwhile, New Jersey has quite a few recycling programs (although they still don't have the bottle deposit system - wtf?). For the NJ government, recycling IS cheaper than dumping in a landfill, because there is very limited landfill space in NJ and much of NJ's non-recycled trash gets carted to PA.
New York State varies as far as recycling, in some cases there are major variances between towns. Ithaca has hefty surcharges on normal trash pickup, forcing people to recycle or pay hefty fees. NYS has bottle deposits statewide, and it is EXTREMELY easy to get your deposits back - almost all supermarkets have automatic bottle/can recycling machines that take your cans and give you a voucher for store credit.
The "baseline comparison" CPU was the Pentium 4, not the Athlon Thunderbird.
The XP was slightly more efficient clock-for-clock than the Tbird, but not by very much. It WAS significantly more clock-for-clock efficient than the Pentium 4.
Like the XPs, if the Tbirds had used the same rating system, they would have been rated on average around 1.4 to 1.5 times their clock speed.
Other than that, your post was accurate.
One person bashed AMD's performance rating system for being confusing (for example, multiple CPUs rated at 3200+) - It may be somewhat confusing, but it's less confusing than trying to figure out which AMD CPU with a base clock rate of 2.0 GHz is going to perform better. (The Athlon XP 2800+ in my system is clocked at 2.0 GHz. The 512kb cache Venice-core Athlon 64s clocked at the same rate garner a 3200+ performance rating due to a much better memory subsystem than the XP. Not sure what the 1M cache version rates.)
Intel has gone to a performance rating system that is even more confusing. They seem to have a different rating system for every CPU, rather than a common baseline comparison unit.
"So if you're a gold farmer, hanging around with your gold farming buddies at the gold farming office, wouldn't you just team up with them instead of trying to solicit groups with American players, who are likely to just slow you down?"
Not if you can't read enough English to read various player guides and forums to point you to the really good farming spots/tactics.
A lot of these farmers group with more skilled American players to leech off of them. Said American players (like myself) quickly get tired of dying every 2-3 minutes to someone who can't even follow basic instructions, and subsequently refuse to ever group with someone who isn't able to communicate again.
When playing an FPS, even a team one, you can make a decent contribution to your team without any actual communication at all. When playing straight deathmatch or any other non-team based mode, communication doesn't matter whatsoever.
Nearly all MMOGs, on the other hand, are about teamwork. All it takes is one person who doesn't listen or communicate well to get an entire group killed, or even wipe an entire multi-group raid. I've seen numerous cases where the actions of one single person who didn't listen to the raid leader caused the entire raid to wipe. As a result, it's simply not acceptable to join groups/raids in an MMOG if you are unable to communicate with the other people in your group/on the raid.
A while ago, there was a rash of Chinese gold farmers in the game. At least what I heard was that they were farmers. What I DO know is that grouping with them was a guaranteed waste of time. I don't know how many times they would do things that any even remotely competent player would be smart enough to do, and in other cases would not do things that they should have learned some time between 1 and 50. (I play DAoC, max level there is 50.) To put it simply - they sucked. They may have been intelligent, but they had no concept of group strategy, and most importantly, WOULD NOT LISTEN. My friend and I would routinely ask them to do something so that we would stop dying every 2-3 minutes, but they would never do it or even respond.
As a result, I refuse to group with anyone who can't speak decent English, and also have 2-3 predominantly Chinese guilds blacklisted.
It's a very, VERY stripped down PPC. Same basic instruction set, but missing a lot of features. (For example, I don't believe it has any instruction reordering capabilities, just like the Xbox 360's CPU is missing many features needed for good "general purpose" performance but not needed for specialized systems such as a gaming console. Those missing features might cause unmodified PPC binaries to fail miserably despite having an instruction set that is mostly identical.
Either way, there's no way we'll be seeing a general-purpose Apple desktop relying solely on a Cell CPU as its processor. Possibly a dedicated Apple media center device, or a coprocessor, but not as a desktop main CPU.
How, pray tell, did AMD invent the "Myth of the Megahertz"?
Consistently, since even the early 486 and Pentium days, AMD (and in fact also Cyrix) CPUs routinely beat Intel CPUs running at somewhat higher clock rates. With a small exception near the end of the Pentium III's lifetime and prior to the introduction of the Pentium 4, AMD CPUs have almost never been available with equivalent or higher raw clock rates than Intel's finest. They HAVE been available with performance matching or beating Intel's finest in many cases, but AMD has rarely ever been the winner in terms of raw clock rate. The one time they were, Intel's Netburst (Pentium 4) architecture was already well along in the development phase, Netburst was NOT a direct response to raw clock rates from AMD.
Intel CPUs have consistently been near the bottom of the barrel in terms of performance per clock cycle, with the exception of specialized low-cost/low-power CPUs.
AMD set about dispelling the "myth of the megahertz", but they did it in a reasonable fashion. Let's face it - clock rate isn't everything (the Pentium 4 proves that without a doubt), and the public needs to stop thinking that the clockrate of the CPU is important. Yes, the AMDs are clocked lower. Despite that, they routinely kill P4s clocked at over 1.5x their clock rate in nearly all applications. Cases where Intel wins are the rare exception and not the norm. Hell, even Intel has had to move away from publishing actual clock rate in preparation for the Netburst architecture's imminent demise.
Average performance on a wide variety of applications is an excellent performance indicator. Raw clock rate and peak performance on a single app (the former being a favorite of Intel and the latter being classic Apple) are both crappy methods of measuring performance.
They have keysets for only about ten games, if they even have ten.
If you don't play one of those games, it's just a really expensive (and in some ways inferior) QWERTY.
Logitech has the right idea with the G15 - A high-quality QWERTY keyboard with nifty extra features (namely a bunch of keys that can be assigned to macros, and an LCD display.) Same with Saitek's gaming keyboard - basically a standard keyboard + lots of macro functionality.
It has insane floating point throughput capability which will help on some apps, but for most desktop apps the Cell is extremely slow. It was designed for a very specific set of tasks.
Existing PPC binaries won't run fast on the Cell. In fact, they most likely won't run at all.
There is no way we'll see a general purpose desktop system based on the Cell - it's just not designed for that kind of purpose. We might see some sort of Cell coprocessor board become available though.
Slashdot Mirror
Bad enough when readers don't RTFA, but the submitters/editors???
Where in the world could anyone have gotten "world's first" from here? The article makes it pretty clear that this is an upgrade to an existing design and that plenty of similar tokamak reactors already exist.
Even at full capacity, it would've taken 10-20 years of taking all of Iraq's oil profits (or it may even have been total net sales...) to pay for the initial cost of the war. Iraq's oil fields aren't running anywhere close to full capacity due to initial damage from the war and constant ongoing damage from insurgent activity.
Note that by "initial cost", I mean the initial 80-100 billion that Bush requested for the war. What's the price tag up to now? 200b? 300b? It's a hell of a lot more. Plus there's the cost of upgrading/rebuilding Iraq's oil production infrastructure.
If this was about oil, it was a damned stupid financial decision.
"All that proves is that the screen and the piece of paper say the same thing. How do either of those relate to the actual value recorded as the vote?"
It doesn't. But the original posters' point was that if there is any suspicion of discrepancies/errors/hacking, the "system" (meaning the whole election process) can fall back on a more traditional/reliable method (paper votes).
Paper ballots have their own problems, but in general it's a different set of problems than the ones in electronic systems.
To get into a console, you can't win if you fall into categories 1 or 2. If the Cell were that expensive or underperforming, people wouldn't be putting it into consoles.
Now, there is one thing about the Cell you missed - It's a special-purpose processor designed for raw floating point performance. 8 of the cores can only do basic streaming floating point (although they do it EXTREMELY fast), the remaining CPU is a VERY stripped down PPC.
So for a gaming system or DSP, the Cell will kick ass. For general purpose computing, it's going to suck.
Do I need to say more? OK, I guess I can - EMT64? StrongARM? (Yes, they bought that from DEC, but rather than kill it and go with one of their own designs they stuck with the better design. The StrongARM lives on in the PXA series.)
As another person pointed out, SOI may simply be too expensive for Intel to license from IBM. Does AMD use SOI? I don't believe so.
Not all companies go for showing the "absolute best case" benchmark. AMD is generally rather conservative with their performance ratings, and it's pretty rare that third-party benchmarks show an Athlon with a given performance rating having significantly different average performance than the "baseline comparison" CPU (One of the older P4s) running at the same clock rate as that performance rating.
A great example of just how conservative AMD is - The Venice core Athlon 64 3200+ has a 2.0 GHz core clock and 512k of L2 cache, using a 90nm process. Its closest dual-core variant (the Manchester core X2 3800+) has the same core clock, L2 cache per core, and manufacturing process. (They also have the same FSB speed, 1 GHz HyperTransport) Yes, that's right, the dual-core variant is only rated 18% higher than its closest single-core counterpart. (This is because currently, on average, a second core usually doesn't net you much benefit because so many CPU-intensive tasks do all the work in a single thread.)
Apple, on the other hand, is notorious for being overly optimistic in their speed comparisons - They always pick the benchmark which will make the competition look as bad as possible, to the point of even failing to use important performance features of the competition's CPU. (For example, back in the P2/P3 era, Apple constantly marketed their systems as being faster than a P2 or P3 with twice the clock speed - While the PPC did in general perform somewhat better per clock cycle than Intel's CPUs, the difference was not anywhere close to what Apple claimed it to be. The benchmark in question used Altivec on the PPC but failed to optimize for Intel whatsoever - No MMX or SSE was used, despite being available.)
To compare it to my previous example, Apple would have called the Athlon 64 X2 3800+ a 6400+ because it had two cores equivalent to the 3200+.
When it comes to inflated/BS benchmarks, Apple is one of the kings.
It blew away MMX back when it was first released, and was somewhat better than SSE.
It isn't really much better than SSE2 at all.
The issue here is that Apple had years to do hand optimization of key routines for Altivec, they haven't had as much time to optimize for SSE2.
The Itanium was known as Merced (for a river in Oregon or Washington, I believe), not Mercedes
Hmm, maybe they changed it due to (negative) customer feedback.
It was only in the "silver" digital tier 1-2 years ago. We didn't look since then and just stuck with OTA broadcast + BitTorrent.
Cable companies are different in different parts of NJ, as another user said.
I'm in a Cablevision service area that was formerly served by a small company called TKR Cable. (Cablevision bought them out.) Let's just say that TKR sucked, to say the least. CV has been a massive improvement, but there's some legacy TKR cruft in Cablevision's pricing and tier organization in our area.
While digital cable may not be required for most channels, Sci-Fi is now only transmitted in encrypted digital in my aree. Well, my former area. I now live in upstate NY, and Time Warner kicks Cablevision's ass right and left.
For customers of Cablevision in New Jersey, getting Sci-Fi costs you $80-90/month, as for some reason it's in Cablevision's "silver" premium digital tier.
I'm not talking a package deal ($80-90 for cable modem plus cable service), I'm talking $80-90 for the TV service ALONE. Cable modem is another $40ish (plus/minus 5-10 dollars, I don't remember the exact amount.)
Cablevision sucks. A channel with commercials should NEVER be put into a top premium tier. Sci-Fi is the ONLY channel in the silver/gold premium tiers from CV that has commercials.
Yeah, although they make it extra easy in many of them with the automated machines.
From what I recall, people in California basically pioneered the recycling movement.
Guess what - apparently almost no one in California recycles. (I could be wrong about this, I'm going on anecdotal evidence from relatives/friends.)
Meanwhile, New Jersey has quite a few recycling programs (although they still don't have the bottle deposit system - wtf?). For the NJ government, recycling IS cheaper than dumping in a landfill, because there is very limited landfill space in NJ and much of NJ's non-recycled trash gets carted to PA.
New York State varies as far as recycling, in some cases there are major variances between towns. Ithaca has hefty surcharges on normal trash pickup, forcing people to recycle or pay hefty fees. NYS has bottle deposits statewide, and it is EXTREMELY easy to get your deposits back - almost all supermarkets have automatic bottle/can recycling machines that take your cans and give you a voucher for store credit.
From what I recall, the coprocessors on the Cell were optimized for floating point calculations.
Most video codecs use integer calculations. The Cell won't really be that much of a benefit.
It would kick ass for a software defined radio for HD reception, but not the actual video encoding/decoding.
The "baseline comparison" CPU was the Pentium 4, not the Athlon Thunderbird.
The XP was slightly more efficient clock-for-clock than the Tbird, but not by very much. It WAS significantly more clock-for-clock efficient than the Pentium 4.
Like the XPs, if the Tbirds had used the same rating system, they would have been rated on average around 1.4 to 1.5 times their clock speed.
Other than that, your post was accurate.
One person bashed AMD's performance rating system for being confusing (for example, multiple CPUs rated at 3200+) - It may be somewhat confusing, but it's less confusing than trying to figure out which AMD CPU with a base clock rate of 2.0 GHz is going to perform better. (The Athlon XP 2800+ in my system is clocked at 2.0 GHz. The 512kb cache Venice-core Athlon 64s clocked at the same rate garner a 3200+ performance rating due to a much better memory subsystem than the XP. Not sure what the 1M cache version rates.)
Intel has gone to a performance rating system that is even more confusing. They seem to have a different rating system for every CPU, rather than a common baseline comparison unit.
"So if you're a gold farmer, hanging around with your gold farming buddies at the gold farming office, wouldn't you just team up with them instead of trying to solicit groups with American players, who are likely to just slow you down?"
Not if you can't read enough English to read various player guides and forums to point you to the really good farming spots/tactics.
A lot of these farmers group with more skilled American players to leech off of them. Said American players (like myself) quickly get tired of dying every 2-3 minutes to someone who can't even follow basic instructions, and subsequently refuse to ever group with someone who isn't able to communicate again.
Gold farmers aren't always skilled. A few known farmers on the Dark Age of Camelot server I play are some of the most incompetent players around.
When playing an FPS, even a team one, you can make a decent contribution to your team without any actual communication at all. When playing straight deathmatch or any other non-team based mode, communication doesn't matter whatsoever.
Nearly all MMOGs, on the other hand, are about teamwork. All it takes is one person who doesn't listen or communicate well to get an entire group killed, or even wipe an entire multi-group raid. I've seen numerous cases where the actions of one single person who didn't listen to the raid leader caused the entire raid to wipe. As a result, it's simply not acceptable to join groups/raids in an MMOG if you are unable to communicate with the other people in your group/on the raid.
A while ago, there was a rash of Chinese gold farmers in the game. At least what I heard was that they were farmers. What I DO know is that grouping with them was a guaranteed waste of time. I don't know how many times they would do things that any even remotely competent player would be smart enough to do, and in other cases would not do things that they should have learned some time between 1 and 50. (I play DAoC, max level there is 50.) To put it simply - they sucked. They may have been intelligent, but they had no concept of group strategy, and most importantly, WOULD NOT LISTEN. My friend and I would routinely ask them to do something so that we would stop dying every 2-3 minutes, but they would never do it or even respond.
As a result, I refuse to group with anyone who can't speak decent English, and also have 2-3 predominantly Chinese guilds blacklisted.
It's a very, VERY stripped down PPC. Same basic instruction set, but missing a lot of features. (For example, I don't believe it has any instruction reordering capabilities, just like the Xbox 360's CPU is missing many features needed for good "general purpose" performance but not needed for specialized systems such as a gaming console. Those missing features might cause unmodified PPC binaries to fail miserably despite having an instruction set that is mostly identical.
Either way, there's no way we'll be seeing a general-purpose Apple desktop relying solely on a Cell CPU as its processor. Possibly a dedicated Apple media center device, or a coprocessor, but not as a desktop main CPU.
How, pray tell, did AMD invent the "Myth of the Megahertz"?
Consistently, since even the early 486 and Pentium days, AMD (and in fact also Cyrix) CPUs routinely beat Intel CPUs running at somewhat higher clock rates. With a small exception near the end of the Pentium III's lifetime and prior to the introduction of the Pentium 4, AMD CPUs have almost never been available with equivalent or higher raw clock rates than Intel's finest. They HAVE been available with performance matching or beating Intel's finest in many cases, but AMD has rarely ever been the winner in terms of raw clock rate. The one time they were, Intel's Netburst (Pentium 4) architecture was already well along in the development phase, Netburst was NOT a direct response to raw clock rates from AMD.
Intel CPUs have consistently been near the bottom of the barrel in terms of performance per clock cycle, with the exception of specialized low-cost/low-power CPUs.
How is saying that the Cell is a totally unsuitable CPU for a desktop machine "Mac fanaticism"???
AMD set about dispelling the "myth of the megahertz", but they did it in a reasonable fashion. Let's face it - clock rate isn't everything (the Pentium 4 proves that without a doubt), and the public needs to stop thinking that the clockrate of the CPU is important. Yes, the AMDs are clocked lower. Despite that, they routinely kill P4s clocked at over 1.5x their clock rate in nearly all applications. Cases where Intel wins are the rare exception and not the norm. Hell, even Intel has had to move away from publishing actual clock rate in preparation for the Netburst architecture's imminent demise.
Average performance on a wide variety of applications is an excellent performance indicator. Raw clock rate and peak performance on a single app (the former being a favorite of Intel and the latter being classic Apple) are both crappy methods of measuring performance.
They have keysets for only about ten games, if they even have ten.
If you don't play one of those games, it's just a really expensive (and in some ways inferior) QWERTY.
Logitech has the right idea with the G15 - A high-quality QWERTY keyboard with nifty extra features (namely a bunch of keys that can be assigned to macros, and an LCD display.) Same with Saitek's gaming keyboard - basically a standard keyboard + lots of macro functionality.
It has insane floating point throughput capability which will help on some apps, but for most desktop apps the Cell is extremely slow. It was designed for a very specific set of tasks.
Existing PPC binaries won't run fast on the Cell. In fact, they most likely won't run at all.
There is no way we'll see a general purpose desktop system based on the Cell - it's just not designed for that kind of purpose. We might see some sort of Cell coprocessor board become available though.