If I were playing a FPS, I'd need a mouse. If I'm doing almost anything else, trackpoint so my hands never leave my keyboard (pointer just beteer 'g' and 'h'). Not much of a middle ground to me. Touchscreen only makes sense for me when it's the only input method. I don't want my hands moving back and fourth between input devices.
That's Ubuntu's thing. They have Gnome, KDE and they roll Unity. They have XBMC and *knowing* it is the base of endeavors like Boxee and Plex decide to roll it again.
I will say XBMC is certainly not perfect. For one, no sane mechanism to view iTunes, Amazon video, Netflix, or Hulu/Hulu plus in Linux. However, one would think if they could work the logistics for whatever they are up to, they could do it in an XBMC compatible way. FYI, I don't buy iTunes or Amazon, but I *would* if it were DRM free and therefore amenable to XBMC.
I will say XBMC could use some enhancenments. Namely: -Better Library scanning. First the default expressions seem significantly worse than say Plex. I always have to plop in a bunch of extra regex into advanced settings. For another thing, XBMC only scans the library at specific points in time. Should do filesystem monitoring and trigger a scan on update of contents. By extension, a headless XBMC just to do library management/refresh (a la Plex's PMS) -Better multi-system setup. If I work at it to make a mysql setup and advanced settings for pathsubstition to get shared art/thumbs it's serviceable enough but I do have to copy over the advencedsettings.xml to each new front end. If an XBMC instance (headless or not) could publish some network-wide defaults for new instances to detect and default to, then XBMC becomes more bullet-proof. -Better PVR integration. Allegedly in Eden, but when I tried it it failed and some discussion that the mythtv integration needed to basically be redone...
Not at this moment, but Ubuntu doesn't either. The difference between the two is XBMC is likely to start working, whereas Ubuntu has decided to ignore ARMv6 architectures going forward.
Doesn't their QNX-based platform now have an Android 'emulator'? I assumed that's what he was talking about.
What is this "nearly" you speak of?
Well, of the automakers that didn't actually go bankrupt, many of them came close. Going bankrupt kind of reinforces the point even more. Businesses functioning in a market that is not yet pure commodity have an *exceedingly* hard time as that market becomes more like a commodity market. Hyundai started from a position of relative weakness due to their poor quality in the 90s, but from all accounts now produces sufficient cars. I don't know to what extent it's because they've improved their processes or how much materials science, safety standards, and everything else has advanced to the point where it is harder to screw up making a car.
Of course, I don't think cars are quite commodity. I still see plenty of cars that are obviously *not* commodity priced on the road (pretty much any sports car, Audi, Lexus, infiniti, etc). In PC world, Apple holds the distinction of being the only status symbol brand, but many car companies enjoy that status suggesting a not-quite commodity industry yet.
The reason they avoided a cost uptick so long is because they spend a lot of money to stockpile replacement parts that *usually* will never be sold or shipped and thus had a huge stockpile. You are paying for that continuity being guaranteed throughout your usually lengthy warranty.
If you don't *need* that continuity then don't buy their components. Functionality wise you can get by without those vendors by using much cheaper alternatives and it may very well be more appropriate in your case (if you have enough skills, attention, and infrastructure to mitigate the risk in other ways, or the risk just isn't significant in your application). Some endeavors are exceptionally paranoid and once in a blue moon an innocuous difference in a new model somehow significantly breaks the system it is a part of (of the few times I can think of, usually poor in-house software hardcoded to awkward specific cases is to blame here). In any event I'd never buy from one of the enterprise vendors for my house, but for some things at work I do think an enterprise vendor is important even with their insane markup.
2) One of the various things that extra money goes toward in the enterprise vendor game is having *gobs* of extra inventory so that they can carry out 'business as usual' in the event of a supplier encountering a situation like this. If you call for warranty replace on a part you got cheap, direct from manufacturer, first the cheapo vendor tends to resist more and second when they do relent, in the cheapo case you may have to wait for it to ship internationally, whereas with many enterprise vendors they likely have a replacement part within 50 miles of your location, even for your obscure, 2 year old system.
When your supply chain choices dwindle down to about two companies (Seagate and Western Digital), there's only so much you can do. All the other vendors have been 'consoloidated' into those two.
On the first part, I would have no trouble believing that Nokia would sell themselves completely to MS. I wouldn't think selling their phones separately would ever happen, but Nokia is pretty well going to do anything that MS asks of them at this point unless their leadership changes away from the MS cronies in place now.
Which makes this story even less plausible, why would MS buy the cow while they get the milk for free? In Nokia they have a partner that is pretty well willing to bet their whole business on MS and do exactly what MS would have them do if they owned them, without the complications of an acquisition, particularly in fairly MS-hostile territory of EU.
In terms of other manufacturers being 'happy' with MS, I think the handset makers are likely not particularly pleased with how the WP7 ecosystem is set up anyway. By design, the hardware manufacturers are relegated pretty much to producing the exact same equipment with the same exact software, chips and screens as their competitors. There is pretty much zero room in the WP7 ecosystem for any differentiation, making it pretty much a pure commodity business with race-to-the-bottom margins. I think MS has most of them scared enough to at least participate by making a few handsets to hedge their bets in case of an Android collapse, but there seems to be pretty much no enthusiasm among manufacturers or carriers with MS and Nokia the only ones actively really *pushing* the platform.
In both that and your other issue frankly it really doesn't matter as the E-350 supports hybrid crossfire
Well, if I can't GPU accelerate the video decode at all, more GPU won't really help, unless you imply that Northern Islands or a follow-on will have offload for 10-bit encodes, but everyone has been awful quiet...
I'll agree that it pretty well beats out Atom even with Ion, and I really want to use something that low power, but I fear I might have to step up to 35W Core i3 to get a beefy enough main cpu. AMD's sufficiently beefy stuff bumps the power budget up a bit (at least in TDP numbers, but those are oversimplifications and flawed to compare, but it's a fair amount of work to get a good comparison).
1) It matters to me, and that's the only one who counts;)
2) That's a valid point.
3) Also a valid point, though I think BD players are more likely to have less emphasis on extensibility than either baked-in 'internet tv' solutions or streaming-dedicated boxes.
I will say that new platforms coming out today may be obsoleted sooner than you would guess. It seems to be that up to very recently, netflix and hulu+ covered most realistic mass-market canned setups (hulu *non* plus will be forever relegated to home-crafted things). It seems that various providers to netflix seem to be growing their own services, and things could start being hit or miss on the various integrated platforms. If some streaming service decides these people doing 10-bit encodes are on to something for better quality per bitrate, that could wipe out all the current popular embedded style hardware.
E-350 works great, *except* for: -Netflix (surprised you didn't mention this), high-def streams under silverlight don't get decode accelerated (or did they fix this? I'm not following windows too closely but that was my recollection) -New or 'obscure' hi-def codecs. The 10-bit profile of h264 is starting to see more and more use but there is no CUDA, OpenCL, DXVA, XvBA, or VA-API offload for it and E-350's CPU half lacks the horsepower to decode without dropping frames. Maybe this will change, but no one knowledgeable enough has provided any hint as to whether a hardware change is simply absolutely required before this can be supported.
Well, apart from *price*, I don't particularly feel like shelling out a ton of money for an OS when I prefer the alternative and I'm not wanting for anything more. Besides, I tried WMC and frankly prefer XBMC even on Windows for non-DVR content and I'm hoping improved Myth support in Eden will make XBMC complete for me...
Radeon has been fairly good. In Linux, VDPAU is way ahead of XvBA in popular support and across the board there is more CUDA work than OpenCL, but other than those details Radeon is serviceable. For $220 you can get a whole new box (e.g. Zotac AD10) with an E-350. ION-equipped Atoms might be an alternative for a die-hard nVidia fan, but it seems to me Intel and nVidia aren't playing together as well and currently the platform is pretty well inferior to e-350. My main caution would be that you are *heavily* reliant in both cases on hardware decode, and things change in ways that may break hardware decode (hi10p being the current headache).
If you are interested in *encoding* content and you bring up CoreAVC implying Windows anyway, you'd probably be best off pining for a Sandy Bridge processor. It's 'quick sync video' stuff is very impressive though it is sadly windows only. By most every account it blows all the CUDA/OpenCL based alternatives out of the water.
I think the 'thumb stick' form factor is a tad more than necessary and in general am disappointed with Roku as a limited platform, but TV integrated stuff is less desireable for two reasons:
1) It is generally even *more* limited than Roku.
2) Some people avoid all-in-one systems where a perfectly fine monitor has to be replaced to reasonably upgrade the system. A large display is significantly more cost than the ~20" desktop monitors. If the TV still is sufficient 5 years from now, but a new codec requires more horsepower than your set can cope with, you would have to add a set-top box anyway. For example, there is an increasing amount of h264 hi10p content out there, and all the current low-power/heat type boxes will likely never be able to play it.
After going to his site, it looks on the surface the only thing you pay for is to create more than one 'letter'. This seems like a terrible example of 'Freemium' because there really isn't a lot of room for many potential customers who would want just one to pay for anything. Maybe I'm missing something....
Freemium usually manifests as ad-supported and/or 'the first hit is free' with priced DLC/in-game items. I've heard varying degrees of success brought up, but on the whole a more positive perspective than this site owner has experienced.
Given the relatively nice, straigtforward, and clean site design (e.g. ad-free), I would have assumed this to be mostly a labor of love. If he wanted or needed to get more money, he'd probably need to go ad-supported.
Now there are tons of other statistics which can be torturous to wade through, but: Overall revenue for purchased content back to 2003 has been nearly flat, with digital music now larger than physical sales in the US by revenue, compared to no official digital media content in 2003. At the same time though, recording industries have increased revenue from things like ad-supported streaming which provides a more tailored experience that, for a *lot* of people I know, has nearly completely displaced their need for 'owned' music.
My issue with the article linked is it seems fairly myopic. It focuses exclusively on *album* sales, when in my experience an album rarely is selected in digital download world anyway. In purchases, singles absolutely dominate. The use of terminology I was on the fence about if they were truly calling out only albums but when I saw them claim CDs comprised 76% of the sales, it almost certainly is excluding sale of singles, unless the UK is just *that* drastically different from the US, where revenue ignoring the detail of album vs. singles shows digital purchases as just over a 50% share. It stinks of an article fueled by the music industry to prod the government by torturing statistics to paint a more dire picture than reality.
I think the fact you must rely upon the site to make the 'correct' choices would be the meat of the complaint. If you are constructing *your* server and client, sure you can apply the discipline and make the correct technology choices. In his example, he is wanting to do some sort automation against arbitrary sites he comes across. Because the technology is so open-ended, he has no idea of how one site will behave versus another and rarely get to reuse code. I've been there a few times, using the web developer features of a browser to kind of reverse engineer how a given web interface is designed and works, with no control and subject to the redesign whims of the server provider, who has no idea or does not care that I'm broken by their new architecture.
There are a few examples of providers with usable interfaces that do care about API compatibility (like Google Maps), but there is a lot more to 'the web' than those and even amongst those you have no consistent API (if Google Maps were to frustrate you or shut down, it's not like you can just change the map data server from google to yahoo and have the same code work without a drastic rework.
In the server case, MS is embracing CLI with more and more Powershell instrumentation. In the *nix world we've had it since the inception of the platform, but MS admins are getting very enthusiastic about a CLI now that they are given it.
For the desktop end-user, the traditional CLI may not usually apply, but in many ways all the search dialogs in various places end up serving the role of a CLI,
If it is over the shoulder, third person perspective in a 3D game, almost universally the console control scheme is better (e.g. Assassin's Creed). For RTS (e.g. Starcraft, C&C), the console version is unplayable. For FPS, I don't understand why so many people prefer console. It's slightly more serviceable than RTS, but still worlds different.
The console sales figures I was considering as a rough indicator of overall popularity. E.g. Crysis low PC sales are not due to increased piracy, but because it just wasn't popular anywhere.
That's way beyond unrealistic. I don't believe even half of the pirates would pay $10 for a game, let alone 80%. By that logic, video and music -- both almost universally available for download at pretty darned reasonable prices -- piracy wouldn't be a booming business (and it is a business for some pirate sites).
At this juncture, any hypothetical price point is pulling something out of our asses, the point being that I think the optimal revenue/profit price point is not 50 (or now commonly 60). $10 might or might not be the price point (I actually think it's more likely than you think), but we can only guess while the publishers have their heads firmly up their asses ant not gathering particularly good data and still doing pricing in a way that doesn't make sense in this sort of ecosystem. In terms of video and music, I said right in my message that file sharing persists even for the music industry where your assertion holds true, but at some point the publishers have to recognize some portion aren't lost sales opportunities, they will *never* buy the product. By getting to the optimal price point for revenue, my suspicion is that the price is low enough to sate all but the most die-hard downloaders/cheapskates that are zero value to your business. That doesn't mean you can't continue pursuing injunctions and other traditional legal channels, but the value of DRM toward the intended purpose is none and harms your users. I would say you are completely wrong about reasonable video downloads. Amazon, Sony, and iTunes all force DRM even for 'offline' play. The price may be reasonable, but the DRM is not reasonable.
In 20+ years, prices have gone up, and distribution costs have gone down. In the case of digital downloads, there is *no* warehousing and logitstics concerns on the scale of physical distribution, no concern about over or under producing. Even in the case of new video games that cost $60 each for a stamped optical disc in a cheap mass-produced case the price is lower than the NES days where each game came on a circuitboard in a beefy, custom plastic shell.
You can reduce the price of software to barely above cost, but the torrented alternative would still be more attractive.
I challenge this assertion. Assuming the illegal and legal copies both are DRM unencumbered and the game is priced along the lines of 10 dollars or less, I'd wager that many fewer would go through the hassle of torrenting (you *still* have the legal threat/risk hanging over it without technical enforcement, distribution channels change, there is greater security risk associated with unauthorized channels, etc). I think the RIAA (after being forced by Apple) demonstrates this to be true. Before embracing digital downloads, file sharing was killing music sales. Now music file sharing certainly isn't gone, but the music industry is healthier than ever with low distribution costs and most casual users either listening to ad-supported streams or buying from amazon or itunes because its easy and not particularly expensive and much more convenient than CDs.
Even going less far, Crysis 2 can *obviously* be copied (3.6 million copies), but 500,000 people *still* chose to pay for their copy with the majority of those actually buying it shortly after release at presumably full retail. The free torrented alternative is *obviously* not universally more attractive, even when faced with a 50 dollar difference.
I think the answer is for the publishers to get less paranoid and implementing draconian DRM and use pricing to both alleviate the threat from illegal downloads and, perhaps more relevant to their business interests, the used market revenue. If they sold 500,000 at ~$50 , but could have hypothetically moved 3 million at ~$10, they might have come out 5 million dollars ahead (negligible per-copy cost incurred, so volume can pretty well be adjusted at will without repercussion). The biggest error I see made with content creators is this hubris where "someone doesn't *deserve* my content if they aren't willing to shell out 50 bucks", holding on to that to the point of ignoring that hubris may actually be getting in the way of higher profits for them. Software developers, movie industry, and ebook writers all still suffer from this witht he muisic industry largely embracing the optimal economic model.
Approximately 0%. The 3.6 million (guesstimated) downloaded copies were DRM-free. DRM did *nothing* to impeded the upload and mass distribution of these titles, did nothing to constrain the capabilities of those who opted to download it, but did constrain the capabilities of people who gave their cash to the software vendor and endeavor *not* to download their software via unauthorized channels. People who download a pirated version need not have a working internet connection to play single player, need not have a disc arbitrarily in the drive to play, and never have to worry about a system upgrade or reinstall invalidating their ability to install/play a game. A legitimate copy with DRM can inflict any combination of the above and more, making the people who pay have an inferior experience compared to those who get it for free.
MW3 appears to have sold about a million copies on PC, Crysis 2 has sold about 500,000.
Incidentally, Crysis 2 sold 1 million copies on xbox360, and 800,000 on the ps3. MW3 did 11.5 million on 360, and 9.2 million on PS3.
It's still hard to derive significant meaning. MW3 has a much bigger marketing push behind it and, frankly, Crysis 2 wasn't a particularly good game. It's initially interesting that Crysis 2 had such a higher rate of illegal downloading, *but* the leak ahead of launch explains that. It's impossible to tell if the month of availability ahead of 'launch' had a chilling effect on sales (my opinion is the sales look about in line with relative popularity with MW3, with the PC perhaps being kinder to Crysis than the console platforms in *relative* terms), and it's impossible to tell how many of those downloads coincided with a legitimate purchase (obviously less than 500k, but some do buy retail and then pirate for no-cd behavior or otherwise being free from DRM) and it's impossible to tell of the rest, how many would have *possibly* bothered to pay if they couldn't have gotten it for free.
Of course the one fact to take away: DRM does *nothing* except inconvenience legitimate users. Both titles were DRM encumbered and both were copied more than they were purchased. DRM does not impair those seeking it to copy in a *significant* way, but it does cause pain to your paying customers.
Even if you combine the estimated number of illegal downloads with the official sales figures, the number of copies on PC including people who didn't even pay is still less than half of either xbox 360 or ps3. It isn't the efficacy of DRM but rather the market in general.
As pointed out, Crysis 2 had a big leak a month ahead of launch. For a month the *only* way to get your hands on it was through illicit download. People are impatient and took whatever means necessary to get their game (the game companies love this impatience, a lot of people drop full retail early on knowing the price will fall like a stone in just a few months).
Slashdot Mirror
If I were playing a FPS, I'd need a mouse. If I'm doing almost anything else, trackpoint so my hands never leave my keyboard (pointer just beteer 'g' and 'h'). Not much of a middle ground to me. Touchscreen only makes sense for me when it's the only input method. I don't want my hands moving back and fourth between input devices.
That's Ubuntu's thing. They have Gnome, KDE and they roll Unity. They have XBMC and *knowing* it is the base of endeavors like Boxee and Plex decide to roll it again.
I will say XBMC is certainly not perfect. For one, no sane mechanism to view iTunes, Amazon video, Netflix, or Hulu/Hulu plus in Linux. However, one would think if they could work the logistics for whatever they are up to, they could do it in an XBMC compatible way. FYI, I don't buy iTunes or Amazon, but I *would* if it were DRM free and therefore amenable to XBMC.
I will say XBMC could use some enhancenments. Namely:
-Better Library scanning. First the default expressions seem significantly worse than say Plex. I always have to plop in a bunch of extra regex into advanced settings. For another thing, XBMC only scans the library at specific points in time. Should do filesystem monitoring and trigger a scan on update of contents. By extension, a headless XBMC just to do library management/refresh (a la Plex's PMS)
-Better multi-system setup. If I work at it to make a mysql setup and advanced settings for pathsubstition to get shared art/thumbs it's serviceable enough but I do have to copy over the advencedsettings.xml to each new front end. If an XBMC instance (headless or not) could publish some network-wide defaults for new instances to detect and default to, then XBMC becomes more bullet-proof.
-Better PVR integration. Allegedly in Eden, but when I tried it it failed and some discussion that the mythtv integration needed to basically be redone...
Not at this moment, but Ubuntu doesn't either. The difference between the two is XBMC is likely to start working, whereas Ubuntu has decided to ignore ARMv6 architectures going forward.
RIM doesn't do Android
Doesn't their QNX-based platform now have an Android 'emulator'? I assumed that's what he was talking about.
What is this "nearly" you speak of?
Well, of the automakers that didn't actually go bankrupt, many of them came close. Going bankrupt kind of reinforces the point even more. Businesses functioning in a market that is not yet pure commodity have an *exceedingly* hard time as that market becomes more like a commodity market. Hyundai started from a position of relative weakness due to their poor quality in the 90s, but from all accounts now produces sufficient cars. I don't know to what extent it's because they've improved their processes or how much materials science, safety standards, and everything else has advanced to the point where it is harder to screw up making a car.
Of course, I don't think cars are quite commodity. I still see plenty of cars that are obviously *not* commodity priced on the road (pretty much any sports car, Audi, Lexus, infiniti, etc). In PC world, Apple holds the distinction of being the only status symbol brand, but many car companies enjoy that status suggesting a not-quite commodity industry yet.
The reason they avoided a cost uptick so long is because they spend a lot of money to stockpile replacement parts that *usually* will never be sold or shipped and thus had a huge stockpile. You are paying for that continuity being guaranteed throughout your usually lengthy warranty.
If you don't *need* that continuity then don't buy their components. Functionality wise you can get by without those vendors by using much cheaper alternatives and it may very well be more appropriate in your case (if you have enough skills, attention, and infrastructure to mitigate the risk in other ways, or the risk just isn't significant in your application). Some endeavors are exceptionally paranoid and once in a blue moon an innocuous difference in a new model somehow significantly breaks the system it is a part of (of the few times I can think of, usually poor in-house software hardcoded to awkward specific cases is to blame here). In any event I'd never buy from one of the enterprise vendors for my house, but for some things at work I do think an enterprise vendor is important even with their insane markup.
1) Already marked up by a large margin.
2) One of the various things that extra money goes toward in the enterprise vendor game is having *gobs* of extra inventory so that they can carry out 'business as usual' in the event of a supplier encountering a situation like this. If you call for warranty replace on a part you got cheap, direct from manufacturer, first the cheapo vendor tends to resist more and second when they do relent, in the cheapo case you may have to wait for it to ship internationally, whereas with many enterprise vendors they likely have a replacement part within 50 miles of your location, even for your obscure, 2 year old system.
When your supply chain choices dwindle down to about two companies (Seagate and Western Digital), there's only so much you can do. All the other vendors have been 'consoloidated' into those two.
On the first part, I would have no trouble believing that Nokia would sell themselves completely to MS. I wouldn't think selling their phones separately would ever happen, but Nokia is pretty well going to do anything that MS asks of them at this point unless their leadership changes away from the MS cronies in place now.
Which makes this story even less plausible, why would MS buy the cow while they get the milk for free? In Nokia they have a partner that is pretty well willing to bet their whole business on MS and do exactly what MS would have them do if they owned them, without the complications of an acquisition, particularly in fairly MS-hostile territory of EU.
In terms of other manufacturers being 'happy' with MS, I think the handset makers are likely not particularly pleased with how the WP7 ecosystem is set up anyway. By design, the hardware manufacturers are relegated pretty much to producing the exact same equipment with the same exact software, chips and screens as their competitors. There is pretty much zero room in the WP7 ecosystem for any differentiation, making it pretty much a pure commodity business with race-to-the-bottom margins. I think MS has most of them scared enough to at least participate by making a few handsets to hedge their bets in case of an Android collapse, but there seems to be pretty much no enthusiasm among manufacturers or carriers with MS and Nokia the only ones actively really *pushing* the platform.
In both that and your other issue frankly it really doesn't matter as the E-350 supports hybrid crossfire
Well, if I can't GPU accelerate the video decode at all, more GPU won't really help, unless you imply that Northern Islands or a follow-on will have offload for 10-bit encodes, but everyone has been awful quiet...
I'll agree that it pretty well beats out Atom even with Ion, and I really want to use something that low power, but I fear I might have to step up to 35W Core i3 to get a beefy enough main cpu. AMD's sufficiently beefy stuff bumps the power budget up a bit (at least in TDP numbers, but those are oversimplifications and flawed to compare, but it's a fair amount of work to get a good comparison).
1) It matters to me, and that's the only one who counts ;)
2) That's a valid point.
3) Also a valid point, though I think BD players are more likely to have less emphasis on extensibility than either baked-in 'internet tv' solutions or streaming-dedicated boxes.
I will say that new platforms coming out today may be obsoleted sooner than you would guess. It seems to be that up to very recently, netflix and hulu+ covered most realistic mass-market canned setups (hulu *non* plus will be forever relegated to home-crafted things). It seems that various providers to netflix seem to be growing their own services, and things could start being hit or miss on the various integrated platforms. If some streaming service decides these people doing 10-bit encodes are on to something for better quality per bitrate, that could wipe out all the current popular embedded style hardware.
E-350 works great, *except* for:
-Netflix (surprised you didn't mention this), high-def streams under silverlight don't get decode accelerated (or did they fix this? I'm not following windows too closely but that was my recollection)
-New or 'obscure' hi-def codecs. The 10-bit profile of h264 is starting to see more and more use but there is no CUDA, OpenCL, DXVA, XvBA, or VA-API offload for it and E-350's CPU half lacks the horsepower to decode without dropping frames. Maybe this will change, but no one knowledgeable enough has provided any hint as to whether a hardware change is simply absolutely required before this can be supported.
Well, apart from *price*, I don't particularly feel like shelling out a ton of money for an OS when I prefer the alternative and I'm not wanting for anything more. Besides, I tried WMC and frankly prefer XBMC even on Windows for non-DVR content and I'm hoping improved Myth support in Eden will make XBMC complete for me...
Radeon has been fairly good. In Linux, VDPAU is way ahead of XvBA in popular support and across the board there is more CUDA work than OpenCL, but other than those details Radeon is serviceable. For $220 you can get a whole new box (e.g. Zotac AD10) with an E-350. ION-equipped Atoms might be an alternative for a die-hard nVidia fan, but it seems to me Intel and nVidia aren't playing together as well and currently the platform is pretty well inferior to e-350. My main caution would be that you are *heavily* reliant in both cases on hardware decode, and things change in ways that may break hardware decode (hi10p being the current headache).
If you are interested in *encoding* content and you bring up CoreAVC implying Windows anyway, you'd probably be best off pining for a Sandy Bridge processor. It's 'quick sync video' stuff is very impressive though it is sadly windows only. By most every account it blows all the CUDA/OpenCL based alternatives out of the water.
I think the 'thumb stick' form factor is a tad more than necessary and in general am disappointed with Roku as a limited platform, but TV integrated stuff is less desireable for two reasons:
1) It is generally even *more* limited than Roku.
2) Some people avoid all-in-one systems where a perfectly fine monitor has to be replaced to reasonably upgrade the system. A large display is significantly more cost than the ~20" desktop monitors. If the TV still is sufficient 5 years from now, but a new codec requires more horsepower than your set can cope with, you would have to add a set-top box anyway. For example, there is an increasing amount of h264 hi10p content out there, and all the current low-power/heat type boxes will likely never be able to play it.
After going to his site, it looks on the surface the only thing you pay for is to create more than one 'letter'. This seems like a terrible example of 'Freemium' because there really isn't a lot of room for many potential customers who would want just one to pay for anything. Maybe I'm missing something....
Freemium usually manifests as ad-supported and/or 'the first hit is free' with priced DLC/in-game items. I've heard varying degrees of success brought up, but on the whole a more positive perspective than this site owner has experienced.
Given the relatively nice, straigtforward, and clean site design (e.g. ad-free), I would have assumed this to be mostly a labor of love. If he wanted or needed to get more money, he'd probably need to go ad-supported.
Well, in North America:
http://www.grabstats.com/statmain.asp?StatID=74
Revenues have increased even in the face of the significant economic slowdown.
Now there are tons of other statistics which can be torturous to wade through, but:
Overall revenue for purchased content back to 2003 has been nearly flat, with digital music now larger than physical sales in the US by revenue, compared to no official digital media content in 2003. At the same time though, recording industries have increased revenue from things like ad-supported streaming which provides a more tailored experience that, for a *lot* of people I know, has nearly completely displaced their need for 'owned' music.
My issue with the article linked is it seems fairly myopic. It focuses exclusively on *album* sales, when in my experience an album rarely is selected in digital download world anyway. In purchases, singles absolutely dominate. The use of terminology I was on the fence about if they were truly calling out only albums but when I saw them claim CDs comprised 76% of the sales, it almost certainly is excluding sale of singles, unless the UK is just *that* drastically different from the US, where revenue ignoring the detail of album vs. singles shows digital purchases as just over a 50% share. It stinks of an article fueled by the music industry to prod the government by torturing statistics to paint a more dire picture than reality.
I think the fact you must rely upon the site to make the 'correct' choices would be the meat of the complaint. If you are constructing *your* server and client, sure you can apply the discipline and make the correct technology choices. In his example, he is wanting to do some sort automation against arbitrary sites he comes across. Because the technology is so open-ended, he has no idea of how one site will behave versus another and rarely get to reuse code. I've been there a few times, using the web developer features of a browser to kind of reverse engineer how a given web interface is designed and works, with no control and subject to the redesign whims of the server provider, who has no idea or does not care that I'm broken by their new architecture.
There are a few examples of providers with usable interfaces that do care about API compatibility (like Google Maps), but there is a lot more to 'the web' than those and even amongst those you have no consistent API (if Google Maps were to frustrate you or shut down, it's not like you can just change the map data server from google to yahoo and have the same code work without a drastic rework.
In the server case, MS is embracing CLI with more and more Powershell instrumentation. In the *nix world we've had it since the inception of the platform, but MS admins are getting very enthusiastic about a CLI now that they are given it.
For the desktop end-user, the traditional CLI may not usually apply, but in many ways all the search dialogs in various places end up serving the role of a CLI,
If it is over the shoulder, third person perspective in a 3D game, almost universally the console control scheme is better (e.g. Assassin's Creed). For RTS (e.g. Starcraft, C&C), the console version is unplayable. For FPS, I don't understand why so many people prefer console. It's slightly more serviceable than RTS, but still worlds different.
The console sales figures I was considering as a rough indicator of overall popularity. E.g. Crysis low PC sales are not due to increased piracy, but because it just wasn't popular anywhere.
That's way beyond unrealistic. I don't believe even half of the pirates would pay $10 for a game, let alone 80%. By that logic, video and music -- both almost universally available for download at pretty darned reasonable prices -- piracy wouldn't be a booming business (and it is a business for some pirate sites).
At this juncture, any hypothetical price point is pulling something out of our asses, the point being that I think the optimal revenue/profit price point is not 50 (or now commonly 60). $10 might or might not be the price point (I actually think it's more likely than you think), but we can only guess while the publishers have their heads firmly up their asses ant not gathering particularly good data and still doing pricing in a way that doesn't make sense in this sort of ecosystem. In terms of video and music, I said right in my message that file sharing persists even for the music industry where your assertion holds true, but at some point the publishers have to recognize some portion aren't lost sales opportunities, they will *never* buy the product. By getting to the optimal price point for revenue, my suspicion is that the price is low enough to sate all but the most die-hard downloaders/cheapskates that are zero value to your business. That doesn't mean you can't continue pursuing injunctions and other traditional legal channels, but the value of DRM toward the intended purpose is none and harms your users. I would say you are completely wrong about reasonable video downloads. Amazon, Sony, and iTunes all force DRM even for 'offline' play. The price may be reasonable, but the DRM is not reasonable.
In 20+ years, prices have gone up, and distribution costs have gone down. In the case of digital downloads, there is *no* warehousing and logitstics concerns on the scale of physical distribution, no concern about over or under producing. Even in the case of new video games that cost $60 each for a stamped optical disc in a cheap mass-produced case the price is lower than the NES days where each game came on a circuitboard in a beefy, custom plastic shell.
You can reduce the price of software to barely above cost, but the torrented alternative would still be more attractive.
I challenge this assertion. Assuming the illegal and legal copies both are DRM unencumbered and the game is priced along the lines of 10 dollars or less, I'd wager that many fewer would go through the hassle of torrenting (you *still* have the legal threat/risk hanging over it without technical enforcement, distribution channels change, there is greater security risk associated with unauthorized channels, etc). I think the RIAA (after being forced by Apple) demonstrates this to be true. Before embracing digital downloads, file sharing was killing music sales. Now music file sharing certainly isn't gone, but the music industry is healthier than ever with low distribution costs and most casual users either listening to ad-supported streams or buying from amazon or itunes because its easy and not particularly expensive and much more convenient than CDs.
Even going less far, Crysis 2 can *obviously* be copied (3.6 million copies), but 500,000 people *still* chose to pay for their copy with the majority of those actually buying it shortly after release at presumably full retail. The free torrented alternative is *obviously* not universally more attractive, even when faced with a 50 dollar difference.
I think the answer is for the publishers to get less paranoid and implementing draconian DRM and use pricing to both alleviate the threat from illegal downloads and, perhaps more relevant to their business interests, the used market revenue. If they sold 500,000 at ~$50 , but could have hypothetically moved 3 million at ~$10, they might have come out 5 million dollars ahead (negligible per-copy cost incurred, so volume can pretty well be adjusted at will without repercussion). The biggest error I see made with content creators is this hubris where "someone doesn't *deserve* my content if they aren't willing to shell out 50 bucks", holding on to that to the point of ignoring that hubris may actually be getting in the way of higher profits for them. Software developers, movie industry, and ebook writers all still suffer from this witht he muisic industry largely embracing the optimal economic model.
Approximately 0%. The 3.6 million (guesstimated) downloaded copies were DRM-free. DRM did *nothing* to impeded the upload and mass distribution of these titles, did nothing to constrain the capabilities of those who opted to download it, but did constrain the capabilities of people who gave their cash to the software vendor and endeavor *not* to download their software via unauthorized channels. People who download a pirated version need not have a working internet connection to play single player, need not have a disc arbitrarily in the drive to play, and never have to worry about a system upgrade or reinstall invalidating their ability to install/play a game. A legitimate copy with DRM can inflict any combination of the above and more, making the people who pay have an inferior experience compared to those who get it for free.
MW3 appears to have sold about a million copies on PC, Crysis 2 has sold about 500,000.
Incidentally, Crysis 2 sold 1 million copies on xbox360, and 800,000 on the ps3. MW3 did 11.5 million on 360, and 9.2 million on PS3.
It's still hard to derive significant meaning. MW3 has a much bigger marketing push behind it and, frankly, Crysis 2 wasn't a particularly good game. It's initially interesting that Crysis 2 had such a higher rate of illegal downloading, *but* the leak ahead of launch explains that. It's impossible to tell if the month of availability ahead of 'launch' had a chilling effect on sales (my opinion is the sales look about in line with relative popularity with MW3, with the PC perhaps being kinder to Crysis than the console platforms in *relative* terms), and it's impossible to tell how many of those downloads coincided with a legitimate purchase (obviously less than 500k, but some do buy retail and then pirate for no-cd behavior or otherwise being free from DRM) and it's impossible to tell of the rest, how many would have *possibly* bothered to pay if they couldn't have gotten it for free.
Of course the one fact to take away: DRM does *nothing* except inconvenience legitimate users. Both titles were DRM encumbered and both were copied more than they were purchased. DRM does not impair those seeking it to copy in a *significant* way, but it does cause pain to your paying customers.
Even if you combine the estimated number of illegal downloads with the official sales figures, the number of copies on PC including people who didn't even pay is still less than half of either xbox 360 or ps3. It isn't the efficacy of DRM but rather the market in general.
As pointed out, Crysis 2 had a big leak a month ahead of launch. For a month the *only* way to get your hands on it was through illicit download. People are impatient and took whatever means necessary to get their game (the game companies love this impatience, a lot of people drop full retail early on knowing the price will fall like a stone in just a few months).