Seems akin to the Atlantis Project, which hoped to build the city of Oceania from floating concrete-and-air hexagonal platforms. Sounded promising, but alas no artificial islands have come of it yet.
(volumes cannot be distinguished from random data)
Aye, there's the rub.
Most files CAN be distinguished from random data. If not outright human-readable (text, XML, etc.), they start with header data which can be visually recognized with a little experience. File sizes are predictably reflective of the directory context. Browsing the rest of a file's content usually reveals non-random components.
TrueCrypt claiming to be indistinguishable from "random data" is kinda like the hotel security guy who was checking out my activity when I was bored (playing with video camera menu settings, waiting for someone) in a hotel lounge. It was obvious he was hotel security because he didn't have any official-looking paraphanalia AND was dressed in "I'm trying to blend in but don't know how" attire. It was obvious he was checking out my activity because he wandered close, looked around like he was looking for someone, and left - when there was absolutely nobody else in the lounge. And from his "I'm not hotel security, no really" dress & demeanor, I knew something would come of it - manifest a few minutes later when the Federal Marshals showed up.
A TrueCrypt file (or partition) hits the "uncanny valley" realm: it tries so hard to blend in that we become keenly & deeply aware that it doesn't; the deep-seated human mechanism for sensing "something is wrong here" kicks in.
It stands out precisely because it so completely doesn't.
For the same price I can get an ultraportable (3lbs) Sony VAIO with ~10" screen, real keyboard (only slightly scrunched), 1280x768 screen, and real everything else including optical drive and WAN radio. Heck, I've had two models over 5 years, wishing only for a stronger case and boot-from-USB; I carry it everywhere.
I'm not sure where the author thinks this toy is usable for anything but an overblown cellphone without the phone.
Funny, everyone thinks their way is right, and expects others to comply... and when others don't, accusations of "insane balkanization" are flung, if not outright force, upon those who simply don't cooperate and have no interest in doing so. It's reasonable to cooperate, you see...
What consumers are increasingly going to want is a comprehensive telecom service: phone+TV+internet.
No, what they want is a telco that's going to deliver what they want, NOW. What they want will largely come from the "long tail" that a single provider won't.
Phone is just getting data from one specialized (audio i/o) device to another; if somebody can just map phone numbers to IP addresses and get an audio data stream from one to another, we don't need a "phone service".
TV? 300 channels and only 3 I want to watch? Get outta my way and let me get my video from iTunes, YouTube, and a thousand other niche providers. Comcast doesn't serve the content I want - but someone does, so just deliver that video data stream.
Internet? it's just moving data packets from A to B. Do that fast, efficient, and cheap, and I'll pay a bundle; throttle me and push lousy content & stupid "phone billing" in my face, and I'll find someone else who will just transfer my data fast, efficient & cheap.
Yes, I was thinking of GE as being an exception when I wrote that. (Viable posting sizes do not lend themselves to detailed analysis of every conceptual variation.)
GE came into being, and largely succeeded, by having the core competency "general electric": they did pretty much anything that had to do with electricity, and that at a time when a company _could_ (broadly speaking) do anything and everything having to do with electricity (kinda like IBM and computers for a long time). They stuck to their core competency, and it worked. As the company flourished, they were able to branch somewhat into other stuff - but kept that core alive, without which all would fail.
Eventually, the "electrical stuff" business got so vast and detailed and nuanced and competetive that General Electric had to largely get out of both the "general" and "electric" parts of the business. In came Jack Welch, who managed to do something _rarely_ done: change the core competency of a business, and survive. Since GE's massive growth had branched into so many subjects (not all electrical), and had gotten so successful at some of them (again, not all electrical), Mr. Welch re-wrote the core competency to "#1, #2, or not in the business". Everything GE (no longer an acronym, just a meaningless couple of letters) was not best, or second best, at was mercilessly pruned. "Neutron Jack" got his nick for vacating life from vast swaths of the company, but leaving the buildings standing. Plastics? Jet engines? Financing? not electrical, but darn good at it - so it stayed, adhering to the new core competency. Most consumer products (tape players, radios, TVs, etc.)? electrical, but losing out to Sony and other competetors, so cut the losses, don't fight where you won't win, dump the business. Train engines? actually giant electrical generators on wheels, and the department was really good at it, so that business stayed. Hydroponic farming? not electrical, they weren't good at it, and it was dropped - you probably didn't even know they tried it. #1, #2, or get out - that became the new core competency, and on a dime GE turned mercilessly to implementing it.
Yes, companies can survive changing their core competencies. To do so, they must make the change wholesale - and _stick_to_it_. Most try but fail because they didn't really change, they just branched, got lopsided, and fell over. "Do or do not, there is no try."
To the thread's point: Telling a telco to get into the destination website business is lunacy. They're not in that business, they didn't develop competency in that business as facilitating their core, and the suggestion they try it comes directly from failing to succeed in their core competency - switching won't help because frankly they suck at both. GE succeeded in switching from making electrics to, well, making money because they were GOOD at the original core competency, and when they had to switch they had a good tangent to switch _to_, and they _made_ the switch _totally_. If telcos want to "win", they need to get GOOD at their core competency of bandwidth delivery; if they want to switch, it must be _to_ something they're already good at, developed as a tangent to the prior competency - and they have to switch completely, without mercy.
Considering their, ahem, lack of certain constraints, why not? Wi-Fi coverage is so broad and overlapping that suitable reprogramming of certain models of routers could easily implement an ad-hoc wide-area uncontrolled network grid serving their major markets - rapidly creating a vast "backbone" mesh almost completely independent of major telcos.
Yahoo, Google, etc. are going into the telco business because the telcos are not doing their job. Instead of facilitating customers' needs and making usage easy, pleasant and efficient, they are trying to squeeze every penny out of customer pockets with screwy billing plans, bandwidth & destination throttling, etc. - practices which hinder the services which customers want and which Google, Yahoo et al want to provide.
As long touted, the Internet is designed to work around breakdowns and bottlenecks. Current telcos ARE breaking links and implementing bottlenecks... so the businesses that suffer are taking advantage of the Internet's core purpose: distribute data efficiently around problems.
Funny thing is, if the telcos would just focus on getting packets from point X to Y quickly and cheaply, and pass that speed and savings on to the customer, they would make more money and not have to consider going into businesses they're not suited to.
Every company has an essence that it must stick to. If it gets too far outside that core product/service, it almost invariably suffers and often dies.
Retailers do not build major roads to facilitate reaching their stores. Road-building contractors do not go into the retail business.
For a _few_ businesses, expanding into infrastructure construction may be required - but only to jump-start the market, at which point they need to get out of the infrastructure business... and at which point they often get overrun. Compuserve, AOL, etc. needed to build infrastructure to serve their content business... but when the infrastructure was there, customers went elsewhere and both are now largely also-rans.
Electricity, natural gas, etc. providers have largely given up their infrastructure business.
Internet backbone service providers simply do not have what it takes to go into the content/destination business. It's simply not what they do, and others do it far better so long as there is sufficient infrastructure to support them. Google may be getting into the infrastructure business, but only to boost infrastructure capacity to match where they want to go in their core business; when Google gets the infrastructure to where they need it, they will have to let go of the infrastructure business because, simply, it's not what they do.
NiN may be giving away the raw audiomixes for Year Zero, but that's in a format that 99% of customers don't want to dork around with. It generates some buzz from those who think it's neat, and is "free" to those very few who actually use it, but most just shell out the $18.99 for the CD anyway. Those who go the literal "free" route with Year Zero "pay" by creating buzz often, to a non-trivial degree, by using the raw tracks to remix into other material that raises more awareness; NiN is buying advertising by giving away samples instead of $$$.
Goes to show that "free" hey-here's-the-whole-shebang-and-more doesn't work the way people claim they want it to. Most want FUN, NOW and are willing to pay for it.
While most commentaries on brilliant analysis are not brilliant, a few are.
Edward Tufte's analysis of Dr. Feynman's brilliant analysis is brilliant, warranting a full chapter in Visual Explanations. What makes it special is that it is not "hey, yeah, that's a good idea, I'm smart too" but instead a study of why Dr. Feynman's analysis is brilliant.
What is particularly stunning about these is that a non-trivial team of people actually were commissioned to: - explore possible characters - depict them - write detailed commentary - outsource actual manufacturing, including mold design, painting, packaging - publish & advertise the promotion - reserve & fill warehouse space - ship them etc.
It's what I've been calling the "SpongeBob Effect": someone actually had to convince deep-pocket high-level managers to fund a team to do this - and got a 'yes'!
no one is playing monopoly or life because they think those are the greatest games in the world.
No, they're playing Monopoly because it's sufficiently fun, 1 to >8 can play, most people already have it, and most people already know how to play it. Starting is simply a matter of "anyone up for Monopoly?", dump the contents on the table, and look up the starting $$$ distribution. The goal is FUN, NOW.
That in contrast with "the greatest game in the world", which probably requires conneseurship to appreciate, has player restrictions, few have it (from both cost & awareness), and takes 30 minutes just trying to read & explain the rules. Starting is a matter of a sales pitch on why anyone would want to play, figuring out what the heck is going on, and playing several rounds just to get the idea. The cost is complexity and education time... failing to pull off FUN, NOW.
(Finding myself in the odd position of justifying that side...)
The difference is that the FBA/TSA/BATFE/etc. lab researchers tried doing it without terminating themselves. If you're not particularly concerned about surviving the mixture part, and know that things can go sufficiently wrong to achieve your dastardly plan anyway, and don't realize that the funds spent would get better results at the Bunny Ranch, what is utterly unacceptable to most becomes entirely desirable. Notice that in the same page where the lab researchers had trouble getting it to work, they also note that some other folks did get it to work, and the results would have been satisfactory for them if only they were in a plane and not a scraggy apartment.
That reasoning amounts to expecting every doorway from every room to open onto a major automotive freeway.
LANs will survive indefinitely precisely because sometimes your data is just feet or yards away... and because even Internet backbones can't handle the load of routing data for everyone's personal networked printers, storage servers, and media terminals.
Seriously: with the new HDMI-based configuration, there is no way to "tape a movie" anymore. DVR gives you time-shifting and temporary storage, but you can't keep it anymore.
That's what I'm working toward. I don't want my video options dictated by a single local cable monopoly, and while Apple is heading in the right direction I don't want them limiting my options either.
Give me a single high-bandwidth data pipe to my TV, and source material & providers geared toward the TV-style viewing experience.
Cable/satellite/broadcast had their chance to provide what customers wanted: a variety of good material, without commercials, on demand or in a casual drop-in format. Like so many practical monopolies, they forgot who their customers are. Now that broadband exists, others can provide what customers want. Let's get a move on, people!
it represents a specific implementation of well known techniques developed nearly 25 years ago
There are many classic/old techniques which are only now being used - and very successfully - precisely because the hardware simply wasn't there. A recent/. post told of ray-tracing being soon used for real-time 3D gaming, and how it beats the socks off "rasterized" methods when a critical mass of polygons is involved; the techniques were well known and developed nearly 25 years ago, but only now do we have the CPU horsepower and vast fast memory capacities available for those "old" techniques to really shine. Likewise "old" "brute force" database techniques: they may not be clever and efficient like what we've been using for highly stable processing of relatively small-to-medium databases, but they work marvelously well when involving big unreliable networks of processors working on vast somewhat-incoherent databases - systems where modern shiny techniques just crumble and can't handle the scaling.
Sometimes the "old" methods are best - you just need the horsepower to pull it off. Clever improvements only scale so long.
the fact remains that we don't actually need this for games.
Your post is heavily dependent on availability of suitable hardware. Software can be ported and recompiled to new platforms, but hardware-dependent software has a short lifespan precisely because getting usable hardware doesn't last particularly long. There's a lot of otherwise good enjoyable games which are unplayable now because they depended on the early Voodoo cards or other unrepeated graphics hardware. Now with CPU power ramping back up (relatively speaking), we can pull away from lifespan-limiting dependence on dedicated graphics hardware, and move the full rendering process back to generic CPU hardware.
Torques me off when I want to play something but can't - not because I don't have the horsepower, but because it requires obsolete cards.
BTW: The article notes that RT vastly outperforms raster on very high poly count scenes - that alone is reason to switch.
Slashdot Mirror
Seems akin to the Atlantis Project, which hoped to build the city of Oceania from floating concrete-and-air hexagonal platforms. Sounded promising, but alas no artificial islands have come of it yet.
(volumes cannot be distinguished from random data)
Aye, there's the rub.
Most files CAN be distinguished from random data. If not outright human-readable (text, XML, etc.), they start with header data which can be visually recognized with a little experience. File sizes are predictably reflective of the directory context. Browsing the rest of a file's content usually reveals non-random components.
TrueCrypt claiming to be indistinguishable from "random data" is kinda like the hotel security guy who was checking out my activity when I was bored (playing with video camera menu settings, waiting for someone) in a hotel lounge. It was obvious he was hotel security because he didn't have any official-looking paraphanalia AND was dressed in "I'm trying to blend in but don't know how" attire. It was obvious he was checking out my activity because he wandered close, looked around like he was looking for someone, and left - when there was absolutely nobody else in the lounge. And from his "I'm not hotel security, no really" dress & demeanor, I knew something would come of it - manifest a few minutes later when the Federal Marshals showed up.
A TrueCrypt file (or partition) hits the "uncanny valley" realm: it tries so hard to blend in that we become keenly & deeply aware that it doesn't; the deep-seated human mechanism for sensing "something is wrong here" kicks in.
It stands out precisely because it so completely doesn't.
Why waste time building it?
Because we LEARN. How many great scientific/engineering breakthroughs were preceeded by someone saying "don't waste your time on that"?
"Strange how much human accomplishment and progress comes from contemplation of the irrelevant."
- Scott Kim
Turing Machine!
has a small 800 x 480 pixel 7" touchscreen
For the same price I can get an ultraportable (3lbs) Sony VAIO with ~10" screen, real keyboard (only slightly scrunched), 1280x768 screen, and real everything else including optical drive and WAN radio. Heck, I've had two models over 5 years, wishing only for a stronger case and boot-from-USB; I carry it everywhere.
I'm not sure where the author thinks this toy is usable for anything but an overblown cellphone without the phone.
Next...
A second-tier priest illustrating a point as part of an interview does NOT translate to the Vatican pronouncing "7 new deadly sins".
The Constitution's Bill of Rights doesn't stop legislators from infringing on rights, so what's to think a new one would do any better?
Strange how much human accomplishment and progress comes from contemplation of the irrelevant. - Scott Kim.
That the us vs. them rhetoric
... and when others don't, accusations of "insane balkanization" are flung, if not outright force, upon those who simply don't cooperate and have no interest in doing so.
Funny, everyone thinks their way is right, and expects others to comply
It's reasonable to cooperate, you see...
Belkin tried that. Backfired bigtime.
For a while they sold a router that would, occasionally, take you to a Belkin ad page instead of the website you wanted.
Years later I still won't buy any Belkin products. I'm not the only one. That stunt cost them far more than they made.
What consumers are increasingly going to want is a comprehensive telecom service: phone+TV+internet.
No, what they want is a telco that's going to deliver what they want, NOW. What they want will largely come from the "long tail" that a single provider won't.
Phone is just getting data from one specialized (audio i/o) device to another; if somebody can just map phone numbers to IP addresses and get an audio data stream from one to another, we don't need a "phone service".
TV? 300 channels and only 3 I want to watch? Get outta my way and let me get my video from iTunes, YouTube, and a thousand other niche providers. Comcast doesn't serve the content I want - but someone does, so just deliver that video data stream.
Internet? it's just moving data packets from A to B. Do that fast, efficient, and cheap, and I'll pay a bundle; throttle me and push lousy content & stupid "phone billing" in my face, and I'll find someone else who will just transfer my data fast, efficient & cheap.
Yes, I was thinking of GE as being an exception when I wrote that. (Viable posting sizes do not lend themselves to detailed analysis of every conceptual variation.)
GE came into being, and largely succeeded, by having the core competency "general electric": they did pretty much anything that had to do with electricity, and that at a time when a company _could_ (broadly speaking) do anything and everything having to do with electricity (kinda like IBM and computers for a long time). They stuck to their core competency, and it worked. As the company flourished, they were able to branch somewhat into other stuff - but kept that core alive, without which all would fail.
Eventually, the "electrical stuff" business got so vast and detailed and nuanced and competetive that General Electric had to largely get out of both the "general" and "electric" parts of the business. In came Jack Welch, who managed to do something _rarely_ done: change the core competency of a business, and survive. Since GE's massive growth had branched into so many subjects (not all electrical), and had gotten so successful at some of them (again, not all electrical), Mr. Welch re-wrote the core competency to "#1, #2, or not in the business". Everything GE (no longer an acronym, just a meaningless couple of letters) was not best, or second best, at was mercilessly pruned. "Neutron Jack" got his nick for vacating life from vast swaths of the company, but leaving the buildings standing. Plastics? Jet engines? Financing? not electrical, but darn good at it - so it stayed, adhering to the new core competency. Most consumer products (tape players, radios, TVs, etc.)? electrical, but losing out to Sony and other competetors, so cut the losses, don't fight where you won't win, dump the business. Train engines? actually giant electrical generators on wheels, and the department was really good at it, so that business stayed. Hydroponic farming? not electrical, they weren't good at it, and it was dropped - you probably didn't even know they tried it. #1, #2, or get out - that became the new core competency, and on a dime GE turned mercilessly to implementing it.
Yes, companies can survive changing their core competencies. To do so, they must make the change wholesale - and _stick_to_it_. Most try but fail because they didn't really change, they just branched, got lopsided, and fell over. "Do or do not, there is no try."
To the thread's point:
Telling a telco to get into the destination website business is lunacy. They're not in that business, they didn't develop competency in that business as facilitating their core, and the suggestion they try it comes directly from failing to succeed in their core competency - switching won't help because frankly they suck at both. GE succeeded in switching from making electrics to, well, making money because they were GOOD at the original core competency, and when they had to switch they had a good tangent to switch _to_, and they _made_ the switch _totally_. If telcos want to "win", they need to get GOOD at their core competency of bandwidth delivery; if they want to switch, it must be _to_ something they're already good at, developed as a tangent to the prior competency - and they have to switch completely, without mercy.
Considering their, ahem, lack of certain constraints, why not?
Wi-Fi coverage is so broad and overlapping that suitable reprogramming of certain models of routers could easily implement an ad-hoc wide-area uncontrolled network grid serving their major markets - rapidly creating a vast "backbone" mesh almost completely independent of major telcos.
Yahoo, Google, etc. are going into the telco business because the telcos are not doing their job. Instead of facilitating customers' needs and making usage easy, pleasant and efficient, they are trying to squeeze every penny out of customer pockets with screwy billing plans, bandwidth & destination throttling, etc. - practices which hinder the services which customers want and which Google, Yahoo et al want to provide.
... so the businesses that suffer are taking advantage of the Internet's core purpose: distribute data efficiently around problems.
As long touted, the Internet is designed to work around breakdowns and bottlenecks. Current telcos ARE breaking links and implementing bottlenecks
Funny thing is, if the telcos would just focus on getting packets from point X to Y quickly and cheaply, and pass that speed and savings on to the customer, they would make more money and not have to consider going into businesses they're not suited to.
Every company has an essence that it must stick to. If it gets too far outside that core product/service, it almost invariably suffers and often dies.
... and at which point they often get overrun. Compuserve, AOL, etc. needed to build infrastructure to serve their content business ... but when the infrastructure was there, customers went elsewhere and both are now largely also-rans.
Retailers do not build major roads to facilitate reaching their stores.
Road-building contractors do not go into the retail business.
For a _few_ businesses, expanding into infrastructure construction may be required - but only to jump-start the market, at which point they need to get out of the infrastructure business
Electricity, natural gas, etc. providers have largely given up their infrastructure business.
Internet backbone service providers simply do not have what it takes to go into the content/destination business. It's simply not what they do, and others do it far better so long as there is sufficient infrastructure to support them. Google may be getting into the infrastructure business, but only to boost infrastructure capacity to match where they want to go in their core business; when Google gets the infrastructure to where they need it, they will have to let go of the infrastructure business because, simply, it's not what they do.
NiN may be giving away the raw audiomixes for Year Zero, but that's in a format that 99% of customers don't want to dork around with. It generates some buzz from those who think it's neat, and is "free" to those very few who actually use it, but most just shell out the $18.99 for the CD anyway. Those who go the literal "free" route with Year Zero "pay" by creating buzz often, to a non-trivial degree, by using the raw tracks to remix into other material that raises more awareness; NiN is buying advertising by giving away samples instead of $$$.
Goes to show that "free" hey-here's-the-whole-shebang-and-more doesn't work the way people claim they want it to. Most want FUN, NOW and are willing to pay for it.
While most commentaries on brilliant analysis are not brilliant, a few are.
Edward Tufte's analysis of Dr. Feynman's brilliant analysis is brilliant, warranting a full chapter in Visual Explanations. What makes it special is that it is not "hey, yeah, that's a good idea, I'm smart too" but instead a study of why Dr. Feynman's analysis is brilliant.
What is particularly stunning about these is that a non-trivial team of people actually were commissioned to:
- explore possible characters
- depict them
- write detailed commentary
- outsource actual manufacturing, including mold design, painting, packaging
- publish & advertise the promotion
- reserve & fill warehouse space
- ship them
etc.
It's what I've been calling the "SpongeBob Effect": someone actually had to convince deep-pocket high-level managers to fund a team to do this - and got a 'yes'!
no one is playing monopoly or life because they think those are the greatest games in the world.
... failing to pull off FUN, NOW.
No, they're playing Monopoly because it's sufficiently fun, 1 to >8 can play, most people already have it, and most people already know how to play it. Starting is simply a matter of "anyone up for Monopoly?", dump the contents on the table, and look up the starting $$$ distribution. The goal is FUN, NOW.
That in contrast with "the greatest game in the world", which probably requires conneseurship to appreciate, has player restrictions, few have it (from both cost & awareness), and takes 30 minutes just trying to read & explain the rules. Starting is a matter of a sales pitch on why anyone would want to play, figuring out what the heck is going on, and playing several rounds just to get the idea. The cost is complexity and education time
(Finding myself in the odd position of justifying that side...)
The difference is that the FBA/TSA/BATFE/etc. lab researchers tried doing it without terminating themselves.
If you're not particularly concerned about surviving the mixture part, and know that things can go sufficiently wrong to achieve your dastardly plan anyway, and don't realize that the funds spent would get better results at the Bunny Ranch, what is utterly unacceptable to most becomes entirely desirable. Notice that in the same page where the lab researchers had trouble getting it to work, they also note that some other folks did get it to work, and the results would have been satisfactory for them if only they were in a plane and not a scraggy apartment.
That reasoning amounts to expecting every doorway from every room to open onto a major automotive freeway.
... and because even Internet backbones can't handle the load of routing data for everyone's personal networked printers, storage servers, and media terminals.
LANs will survive indefinitely precisely because sometimes your data is just feet or yards away
Seriously: with the new HDMI-based configuration, there is no way to "tape a movie" anymore. DVR gives you time-shifting and temporary storage, but you can't keep it anymore.
That's what I'm working toward. I don't want my video options dictated by a single local cable monopoly, and while Apple is heading in the right direction I don't want them limiting my options either.
Give me a single high-bandwidth data pipe to my TV, and source material & providers geared toward the TV-style viewing experience.
Cable/satellite/broadcast had their chance to provide what customers wanted: a variety of good material, without commercials, on demand or in a casual drop-in format. Like so many practical monopolies, they forgot who their customers are. Now that broadband exists, others can provide what customers want. Let's get a move on, people!
it represents a specific implementation of well known techniques developed nearly 25 years ago
/. post told of ray-tracing being soon used for real-time 3D gaming, and how it beats the socks off "rasterized" methods when a critical mass of polygons is involved; the techniques were well known and developed nearly 25 years ago, but only now do we have the CPU horsepower and vast fast memory capacities available for those "old" techniques to really shine. Likewise "old" "brute force" database techniques: they may not be clever and efficient like what we've been using for highly stable processing of relatively small-to-medium databases, but they work marvelously well when involving big unreliable networks of processors working on vast somewhat-incoherent databases - systems where modern shiny techniques just crumble and can't handle the scaling.
There are many classic/old techniques which are only now being used - and very successfully - precisely because the hardware simply wasn't there. A recent
Sometimes the "old" methods are best - you just need the horsepower to pull it off. Clever improvements only scale so long.
the fact remains that we don't actually need this for games.
Your post is heavily dependent on availability of suitable hardware. Software can be ported and recompiled to new platforms, but hardware-dependent software has a short lifespan precisely because getting usable hardware doesn't last particularly long. There's a lot of otherwise good enjoyable games which are unplayable now because they depended on the early Voodoo cards or other unrepeated graphics hardware. Now with CPU power ramping back up (relatively speaking), we can pull away from lifespan-limiting dependence on dedicated graphics hardware, and move the full rendering process back to generic CPU hardware.
Torques me off when I want to play something but can't - not because I don't have the horsepower, but because it requires obsolete cards.
BTW: The article notes that RT vastly outperforms raster on very high poly count scenes - that alone is reason to switch.