nowhere in the article does he say that your pizzas will probably be delivered by autonomous cars within 5 years.
Can the submitter/editors be that bad at reading comprehension or is slashdot just making up/ approving clickbait?
Of course there will be experiments and publicity stunts but there is no way that most pizza deliveries will be done by autonomous vehicles within 5 years, current delivery drivers are just too cheap to compete with.
> You can purchase an FL/FD to ES adapter and continue to use those lens from 1964 if you wish.
But you lose infinity focus or there is some glass in the adapter which degrades the image quality.
They could have added electronics to the mount and kept it backwards compatible, but they chose not to. To me that says that they might do it again in the future, especially with full frame mirrorless around the corner. Also they usually stop producing parts for lenses after a few decades, so "lifetime" might really only be 20 years if the lens stops working, and with all the electronics in there today I think the rate of failure is much higher than on older mechanical only lenses.
I have canon cameras but have bought a few third party lenses. The canon lenses certainly hold their value much better but in a lot of cases I'm not willing to spend the premium - I don't think they will still be usable for my grandchildren.
The way to approach this is not explicitly eliminating/reducing left turns, you just need to factor in an average additional time to complete left turns. You would think this would but simple enough but my current Garmin GPS can't even account properly for stoplights. It grossly underestimates how much time will be lost waiting at them, preferring routes with high speed limits and long traffic lights every few miles instead of smaller roads or slightly out of the way freeways. Obviously traffic comes into play more when trying to account for stop lights and left turn delays, but many devices are traffic aware now.
and Rubin wasn't a huge fan of it either:
"If I could have my pick, I would like to learn that Newton's laws must be modified in order to correctly describe gravitational interactions at large distances. That's more appealing than a universe filled with a new kind of sub-nuclear particle."
I have high hopes for this new theory that can account for the galaxy rotation problem ( and the emDrive ):
http://physicsfromtheedge.blog...
from TFA: "At around $4 a subscriber, ESPN is the most expensive channel in the cable business"
I would pay $5 for just ESPN and the online ESPN content that you can't get now without spending $50+/month. You can deliver it online, there is no reason to have $20/mo "capital costs and maintenance" for one company's dedicated cable running to everyone's house just for TV when the internet is already there.
I'm sure they really won't let people choose just the channels they want without first buying a package of basic crap that is available online or OTA for free and that few people really want. I'm not going to pay $50/month now to get ESPN, and I'm not going to pay $25/month for ESPN in some pseudo-a-la-carte plan either.
Despite all the scare mongering, if you are reasonably healthy you can probably get a a high deductible plan for a few hundred dollars a month. I started my own business 6 years ago and have had my own insurance through two different providers. Things are different from state to state, but I used this site:
http://www.nationalhealthaccess.com/
to compare and get some basic quotes.
I have a $10000 deductible policy that costs just over $100/mo for me and my wife. We were able to get health insurance for our newborn through a state program and that has a very low deductible and that costs a little over $100/mo. In Wisconsin no matter how much money you make kids under age 18 can get some sort of care for reasonable cost. If we weren't in Wisconsin we'd probably be on a $300/mo plan for all three of us with a $2000 deductible, which is still pretty cheap in my opinion. If your health is one of the most valuable things in your life, it makes sense to me that you should be prepared to spend a corresponding amount of your money on it. I dread the prospect of it becoming "free" in the US.
If you don't have enough money in the bank to cover a few worst case years of $2k-10k medical bills, you probably shouldn't be leaving your corporate job. Make sure you get your new coverage started BEFORE you leave your current job, COBRA is very expensive and in my opinion is a last resort to avoid coverage gaps.
The health insurance situation in the US can be vastly improved with one simple change:
Make it ILLEGAL for any group (usually employers) to get preferential pricing from health insurance companies. Insurance should cost the same for any two people in the same situation, including taxes. I have my own business so I have to pay a lot more for health insurance. I know companies pay part of employees premiums, but the total of employer contribution plus employee contribution plus taxes is a lot less than it is for people who are self employed or work for a small business.
This one simple change:
1. makes health insurance companies compete - people will shop around instead of just getting a provider chosen for you by your employer.
2. makes everyone realize what they are really paying for heath care. When things come out of a paycheck like taxes, people seem to accept it more than if they had to write out a check for it each month. If people realized how expensive it is they would be outraged and more likely to shop around or get involved politically. Most people don't need the low deductible, cover-everything plans some of employers offer at low costs - they would probably lower their coverage if paying for it out of pocket. Anytime people aren't paying with their own money costs skyrocket (just ask politicians). "Oh look, he has good insurance, let's run this test and this treatment too, your insurance will cover it"
The issues of pre-existing conditions and people who can't afford any form of health insurance are different. If you let anyone get medical treatment without paying, and cover pre-existing conditions, then you are no longer talking about INSURANCE which is a hedge against unforeseen problems. Then you are talking about government-run health CARE, which would likely be the end of the private health insurance industry as we know it. Whether or not that a good thing is an a different argument, but with all of their lobbyists I don't see it happening anytime soon, so let's start by fixing health insurance, and go from there.
You shouldn't be relying on the browser to clear all user data between users, and certainly not using IE with anything that sensitive.
Use the Groovix Public Access Computing solution, everything will be completely reset between users.
If the machines use too much electricity, are too slow, or are incompatible as thin clients, use Groovix SLIM http://groovix.com/slim.html software based on Ubuntu for the most cost effective approach. Get a $25 watt meter and see how much electricity those old boxes are using, it may not be worth it to keep them running.
Groovix software is free as in beer and freedom, as opposed to the proprietary and costly Discoverstation from Userful. The Userful software licenses will cost you more than new hardware and open source software combined.
Howard County library in Maryland is switching all of their public computers to Groovix software which is based on Ubuntu. They are having a product launch this Thursday, Sept. 21st, you should stop by.
"they were unable to max out the CPUs."
They've apparently never tried this benchmark:
i=0; while [ $i -lt 9 ]; do time pi 1000000 >/dev/null & let i=i+1; done
Despite multiple fixes to the nvidia drivers, I have never gotten my laptop to successfully resume from suspend to ram or from suspend to disk. This is a royal pain in the neck
Our ATI chipset based notebooks work perfectly with suspend to ram AND disk, as well as every other feature you'd expect on a notebook. All this right out of the box, no tinkering required. No need to buy a macbook, you just need to buy from a Linux vendor who has chosen, tested, and configured their hardware and software carefully. Then Linux IS ready for prime time, even on notebooks.
ATI has lagged behind nvidia for linux support in the past, but with the latest drivers that will be released with Dapper, the ATI mobile GPUs work perfectly. There is even a utility to scale back the clock on the GPU to save battery life at the expense of 3D performance. We found the MSI laptops with AMD 64-bit CPUS and ATI chipsets to be much better overall than the current Asus laptop offerings.
In desktops, we prefer nVidia cards, but the latest AMD-64 ATI mobile chipsets are clearly the best in our experience.
As for buying a Dell or Thinkpad, be very careful, or you'll end up with wifi or other aspects that have no hope of running under Linux. Dell/Lenovo can change their hardware without changing the model name, so you never know exactly what you're getting unless it is used.
I also wouldn't buy a Linux laptop from any place that does not specialize in Linux. We have ordered some laptops from stores with Redhat or Ubuntu preinstalled, but it is just a basic install and not all of the features are really working like 3D and full power management.
that is truly amazing, I have envisioned something like that for years. Compared to a live cd with usb flash solution it is going to be slower and more expensive. However, not having to reboot the host machine is priceless. It will also run on a windows machine with crappy windows-only hardware. I'd like to see what the real world performance is like.
This concept would be much better implemented by creating a special Ubuntu live cd to look at a usb flash disk for the/home partition or even to save all settings like http://puppylinux.org/
Not all old machines can even boot from usb, and the install on a flash disk approach doesn't make since Ubuntu would need to be resinstalled every time to get everything set up properly for the hardware that you booted with. For example, if you install Ubuntu on a machine with an nvidia video card, then try to boot that image on a machine with an ATI card, it will not work without changing xorg.conf. But a live cd configures xorg.conf dynamically each boot. If you want the speed of a mini hard drive instead of cd but still want maximum portability, put a bootable image of the the live CD on the mini hard drive but still have a data partition. Yes, it takes an extra minute to go through the debian bootstrap process with hardware detection at each boot, but if you want portability that is probably necessary.
I would love to develop a product like this, but puppy linux may already be good enough. If someone wants to work on this I'd love to help distribute the free software and sell the accompanying hardware and support, so contact me @ groovix.com.
Judging from the number of bad motherboards I've gotten from DFI and Gigabyte, I'm surprised motherboard makers test them at all. It seems they leave the testing up to the system builder. I've been searching to find the best motherboard maker in terms of reliability, but all of the reviews are mixed. It might be time to give ECS a try. Anyone have other suggestions for a reliable motherboard brand?
We've pitched KOHA to some libraries, but non-technical librarians are afraid that something will break and there will be no support.
Even if a closed source management system is used, make sure that clients can connect using a standard web browser. Then all of your client machines can be open source, running Firefox. Some people use LTSP thin client/server setups in large libraries, but our http://groovix.com/kiosk.html multi-user machines provide 4 workstations for $999 which is even more economical.
It just kills me to see library patrons referencing card catalogs and typing documents on $800 dell machines running XP. More money might be wasted on the client machines than the management system itself
My company wanted to try dual processors in our multi-user systems, but the motherboards and processors are just too expensive.
These dual-cores are just another reason we feel that multi-user computing is going to be the efficient choice of the future, instead of everyone having their own high-powered machine. Single user machines will probably not take full advantage of the dual-cores for some time
These are great for our local multi-user approach, but will also help networked thin client approaches like LTSP. Having one processor or core per user isn't even necessary, just having the second core will do a lot to lower latency when one cpu is busy with a thread.
We can't wait to start putting them in our http://groovix.com/ systems in June...
Could you please contact us and tell us exactly which parts you are using and where you get them from? You can send email to slashdot at groovix.com We would love to experiment with some systems like that if they really are that cheap. The local multi-user systems are just one aspect of our business, the goal of Open Sense Solutions is to provide cost-effective complete Linux solutions, and for setups > 10 users, LTSP is probably the most cost effective when you don't need high performance.
For 2-5 user setups, the incremental cost of another user for LTSP is $200, but for local multi-user it is just the cost of another video card and sound card, so local multi-user is guaranteed to be cheaper, and it is a whole lot faster!
Those cost $225 or $275 plus keyboards plus mice, plus a server, plus a little network infrastructure, nothing mentioned about sound, and that gets more expensive for a 3-4 user setup with not-so-great performance compared to a local multi-user setup.
Hi, Assuming your post was addressed at Open Sense Solutions, let me try to answer:
1. Business is slowly but steadily growing.
2. We take Debian sid, mirror our own copy, then adjust our repository to make sure that everything is tested and works well. We keep our repository stable, and add security updates. This gives our customers all the latest software, but without the churn and occasional breakage of Debian Sid. Our customers have liked the software repository, but some have wanted something more like linspire. We are actually duplicating a lot of the goals of Ubuntu, and we may switch to Ubuntu in the future.
3. If you could build us boxes for 50% to 60% of what we are charging, please contact us with more details, because we cannot aquire quality hardware for anywhere near those numbers. For example, we use Antec quiet cases and power supplies, which cost us over $60, not the usual $20-$30 junk that most online linux retailers use. Furthermore, we'll give you our software and support standalone for only $50, and you can put it on any box you build yourself! Look at our detailed specs, look up the prices, consider shipping, and remember we have to research and develop our systems, and there is almost no hardware markup. We encourage people building there own systems, but there is no way an individual can do it for less if there time is worth anything. If our prices were any lower, there would be no way we could stay in business.
4. You can play 3d games head-to-head at reasonable resolutions with reasonable frame rates, but these are not hard core gaming machines. You are free to use more powerful video cards, any nVidia cards will work.
5. These systems are extremely fast for the money, you would have to pay 50% more to get 10% better real world performance. Our generous amounts of RAM and fast hard drives are key. We optimize for cost/performance.
6. There is absolutely no network latency, that is the beauty of local multi-user systems.
Thin client setups cannot even begin to compare in terms of performance, and they aren't cheaper if you're buying new hardware.
Thank you for the feedback, and if you have any other questions drop up a line.
you're right, obviously Microsoft can and will compete in this arena, but from a technical and cost standpoint I'm confident that the open source solutions have a considerable advantage.
Whether or not big corporations choose those open source solutions is a different story.
Even low end thin clients can cost $300 each, not to mention the cost of the server, and the performance you get running 40 people off one box is abysmal compared to local multi-user systems, where you usually can't tell there are other people sharing the machine, and you can run 3D games and see full motion video without a problem.
If you wanted, you could even run a 4-user local multi-user system as 4 thin clients, and get the thin-client administration advantages combined with the local multi-user advantages like power, noise, space, and cost savings.
Unless you have a ton of old reliable boxen to run LTSP or other thin client solutions on, thin clients are way too expensive new for what you get. Local Multi-user systems are much more efficient. Especially when running 4 people on one box, open source (free) software is the only way to avoid killer software costs, so I don't think Microsoft can compete in this arena. You can get new hardware (and all the software you need) for 3 or 4 users for less than $1000 with an open source solution.
For more info on local multi-user systems, check out http://groovix.com/ (that's my company, so obviously I'm biased!)
Slashdot Mirror
nowhere in the article does he say that your pizzas will probably be delivered by autonomous cars within 5 years.
Can the submitter/editors be that bad at reading comprehension or is slashdot just making up/ approving clickbait?
Of course there will be experiments and publicity stunts but there is no way that most pizza deliveries will be done by autonomous vehicles within 5 years, current delivery drivers are just too cheap to compete with.
You can only get a grocery list app on a $1000 phone?
> You can purchase an FL/FD to ES adapter and continue to use those lens from 1964 if you wish.
But you lose infinity focus or there is some glass in the adapter which degrades the image quality.
They could have added electronics to the mount and kept it backwards compatible, but they chose not to. To me that says that they might do it again in the future, especially with full frame mirrorless around the corner. Also they usually stop producing parts for lenses after a few decades, so "lifetime" might really only be 20 years if the lens stops working, and with all the electronics in there today I think the rate of failure is much higher than on older mechanical only lenses.
I have canon cameras but have bought a few third party lenses. The canon lenses certainly hold their value much better but in a lot of cases I'm not willing to spend the premium - I don't think they will still be usable for my grandchildren.
According to the produce pages they also removed the gyroscope in the 2017 8" version.
The way to approach this is not explicitly eliminating/reducing left turns, you just need to factor in an average additional time to complete left turns. You would think this would but simple enough but my current Garmin GPS can't even account properly for stoplights. It grossly underestimates how much time will be lost waiting at them, preferring routes with high speed limits and long traffic lights every few miles instead of smaller roads or slightly out of the way freeways. Obviously traffic comes into play more when trying to account for stop lights and left turn delays, but many devices are traffic aware now.
and Rubin wasn't a huge fan of it either:
"If I could have my pick, I would like to learn that Newton's laws must be modified in order to correctly describe gravitational interactions at large distances. That's more appealing than a universe filled with a new kind of sub-nuclear particle."
I have high hopes for this new theory that can account for the galaxy rotation problem ( and the emDrive ): http://physicsfromtheedge.blog...
from TFA: "At around $4 a subscriber, ESPN is the most expensive channel in the cable business"
I would pay $5 for just ESPN and the online ESPN content that you can't get now without spending $50+/month. You can deliver it online, there is no reason to have $20/mo "capital costs and maintenance" for one company's dedicated cable running to everyone's house just for TV when the internet is already there.
I'm sure they really won't let people choose just the channels they want without first buying a package of basic crap that is available online or OTA for free and that few people really want. I'm not going to pay $50/month now to get ESPN, and I'm not going to pay $25/month for ESPN in some pseudo-a-la-carte plan either.
Despite all the scare mongering, if you are reasonably healthy you can probably get a a high deductible plan for a few hundred dollars a month. I started my own business 6 years ago and have had my own insurance through two different providers. Things are different from state to state, but I used this site: http://www.nationalhealthaccess.com/ to compare and get some basic quotes.
I have a $10000 deductible policy that costs just over $100/mo for me and my wife. We were able to get health insurance for our newborn through a state program and that has a very low deductible and that costs a little over $100/mo. In Wisconsin no matter how much money you make kids under age 18 can get some sort of care for reasonable cost. If we weren't in Wisconsin we'd probably be on a $300/mo plan for all three of us with a $2000 deductible, which is still pretty cheap in my opinion. If your health is one of the most valuable things in your life, it makes sense to me that you should be prepared to spend a corresponding amount of your money on it. I dread the prospect of it becoming "free" in the US.
If you don't have enough money in the bank to cover a few worst case years of $2k-10k medical bills, you probably shouldn't be leaving your corporate job. Make sure you get your new coverage started BEFORE you leave your current job, COBRA is very expensive and in my opinion is a last resort to avoid coverage gaps.
The health insurance situation in the US can be vastly improved with one simple change: Make it ILLEGAL for any group (usually employers) to get preferential pricing from health insurance companies. Insurance should cost the same for any two people in the same situation, including taxes. I have my own business so I have to pay a lot more for health insurance. I know companies pay part of employees premiums, but the total of employer contribution plus employee contribution plus taxes is a lot less than it is for people who are self employed or work for a small business.
This one simple change:
1. makes health insurance companies compete - people will shop around instead of just getting a provider chosen for you by your employer.
2. makes everyone realize what they are really paying for heath care. When things come out of a paycheck like taxes, people seem to accept it more than if they had to write out a check for it each month. If people realized how expensive it is they would be outraged and more likely to shop around or get involved politically. Most people don't need the low deductible, cover-everything plans some of employers offer at low costs - they would probably lower their coverage if paying for it out of pocket. Anytime people aren't paying with their own money costs skyrocket (just ask politicians). "Oh look, he has good insurance, let's run this test and this treatment too, your insurance will cover it"
The issues of pre-existing conditions and people who can't afford any form of health insurance are different. If you let anyone get medical treatment without paying, and cover pre-existing conditions, then you are no longer talking about INSURANCE which is a hedge against unforeseen problems. Then you are talking about government-run health CARE, which would likely be the end of the private health insurance industry as we know it. Whether or not that a good thing is an a different argument, but with all of their lobbyists I don't see it happening anytime soon, so let's start by fixing health insurance, and go from there.
My company developed the Groovix Public Access/Kiosk environment and it will do everything you need.
u ct_info&cPath=1&products_id=17
http://groovix.com/solutions_public_access.html
Just download the Groovix Public Access environment demo disk and boot it. You can install it to hard drive for faster boot up if necessary, and customize it yourself if you're a linux geek. We can customize it for you starting at $99.
http://groovix.com/store/index.php?main_page=prod
http://groovix.com/solutions_public_access.html
It is open source, and completely free if you do your own customizations.
Disclaimer: I'm the developer of that product.
Groovix software is free as in beer and freedom, as opposed to the proprietary and costly Discoverstation from Userful. The Userful software licenses will cost you more than new hardware and open source software combined.
Howard County library in Maryland is switching all of their public computers to Groovix software which is based on Ubuntu. They are having a product launch this Thursday, Sept. 21st, you should stop by.
"they were unable to max out the CPUs." /dev/null & let i=i+1; done
They've apparently never tried this benchmark:
i=0; while [ $i -lt 9 ]; do time pi 1000000 >
Despite multiple fixes to the nvidia drivers, I have never gotten my laptop to successfully resume from suspend to ram or from suspend to disk. This is a royal pain in the neck
Our ATI chipset based notebooks work perfectly with suspend to ram AND disk, as well as every other feature you'd expect on a notebook. All this right out of the box, no tinkering required. No need to buy a macbook, you just need to buy from a Linux vendor who has chosen, tested, and configured their hardware and software carefully. Then Linux IS ready for prime time, even on notebooks.
ATI has lagged behind nvidia for linux support in the past, but with the latest drivers that will be released with Dapper, the ATI mobile GPUs work perfectly. There is even a utility to scale back the clock on the GPU to save battery life at the expense of 3D performance. We found the MSI laptops with AMD 64-bit CPUS and ATI chipsets to be much better overall than the current Asus laptop offerings.
In desktops, we prefer nVidia cards, but the latest AMD-64 ATI mobile chipsets are clearly the best in our experience.
As for buying a Dell or Thinkpad, be very careful, or you'll end up with wifi or other aspects that have no hope of running under Linux. Dell/Lenovo can change their hardware without changing the model name, so you never know exactly what you're getting unless it is used.
I also wouldn't buy a Linux laptop from any place that does not specialize in Linux. We have ordered some laptops from stores with Redhat or Ubuntu preinstalled, but it is just a basic install and not all of the features are really working like 3D and full power management.
that is truly amazing, I have envisioned something like that for years. Compared to a live cd with usb flash solution it is going to be slower and more expensive. However, not having to reboot the host machine is priceless. It will also run on a windows machine with crappy windows-only hardware. I'd like to see what the real world performance is like.
This concept would be much better implemented by creating a special Ubuntu live cd to look at a usb flash disk for the /home partition or even to save all settings like http://puppylinux.org/
Not all old machines can even boot from usb, and the install on a flash disk approach doesn't make since Ubuntu would need to be resinstalled every time to get everything set up properly for the hardware that you booted with. For example, if you install Ubuntu on a machine with an nvidia video card, then try to boot that image on a machine with an ATI card, it will not work without changing xorg.conf. But a live cd configures xorg.conf dynamically each boot. If you want the speed of a mini hard drive instead of cd but still want maximum portability, put a bootable image of the the live CD on the mini hard drive but still have a data partition. Yes, it takes an extra minute to go through the debian bootstrap process with hardware detection at each boot, but if you want portability that is probably necessary.
I would love to develop a product like this, but puppy linux may already be good enough. If someone wants to work on this I'd love to help distribute the free software and sell the accompanying hardware and support, so contact me @ groovix.com.
Judging from the number of bad motherboards I've gotten from DFI and Gigabyte, I'm surprised motherboard makers test them at all. It seems they leave the testing up to the system builder. I've been searching to find the best motherboard maker in terms of reliability, but all of the reviews are mixed. It might be time to give ECS a try. Anyone have other suggestions for a reliable motherboard brand?
We've pitched KOHA to some libraries, but non-technical librarians are afraid that something will break and there will be no support.
Even if a closed source management system is used, make sure that clients can connect using a standard web browser. Then all of your client machines can be open source, running Firefox. Some people use LTSP thin client/server setups in large libraries, but our http://groovix.com/kiosk.html multi-user machines provide 4 workstations for $999 which is even more economical.
It just kills me to see library patrons referencing card catalogs and typing documents on $800 dell machines running XP. More money might be wasted on the client machines than the management system itself
My company wanted to try dual processors in our multi-user systems, but the motherboards and processors are just too expensive.
These dual-cores are just another reason we feel that multi-user computing is going to be the efficient choice of the future, instead of everyone having their own high-powered machine. Single user machines will probably not take full advantage of the dual-cores for some time
These are great for our local multi-user approach, but will also help networked thin client approaches like LTSP. Having one processor or core per user isn't even necessary, just having the second core will do a lot to lower latency when one cpu is busy with a thread.
We can't wait to start putting them in our http://groovix.com/ systems in June...
Could you please contact us and tell us exactly which parts you are using and where you get them from? You can send email to slashdot at groovix.com We would love to experiment with some systems like that if they really are that cheap. The local multi-user systems are just one aspect of our business, the goal of Open Sense Solutions is to provide cost-effective complete Linux solutions, and for setups > 10 users, LTSP is probably the most cost effective when you don't need high performance.
For 2-5 user setups, the incremental cost of another user for LTSP is $200, but for local multi-user it is just the cost of another video card and sound card, so local multi-user is guaranteed to be cheaper, and it is a whole lot faster!
Those cost $225 or $275 plus keyboards plus mice, plus a server, plus a little network infrastructure, nothing mentioned about sound, and that gets more expensive for a 3-4 user setup with not-so-great performance compared to a local multi-user setup.
Hi, Assuming your post was addressed at Open Sense Solutions, let me try to answer:
1. Business is slowly but steadily growing.
2. We take Debian sid, mirror our own copy, then adjust our repository to make sure that everything is tested and works well. We keep our repository stable, and add security updates. This gives our customers all the latest software, but without the churn and occasional breakage of Debian Sid. Our customers have liked the software repository, but some have wanted something more like linspire. We are actually duplicating a lot of the goals of Ubuntu, and we may switch to Ubuntu in the future.
3. If you could build us boxes for 50% to 60% of what we are charging, please contact us with more details, because we cannot aquire quality hardware for anywhere near those numbers. For example, we use Antec quiet cases and power supplies, which cost us over $60, not the usual $20-$30 junk that most online linux retailers use. Furthermore, we'll give you our software and support standalone for only $50, and you can put it on any box you build yourself! Look at our detailed specs, look up the prices, consider shipping, and remember we have to research and develop our systems, and there is almost no hardware markup. We encourage people building there own systems, but there is no way an individual can do it for less if there time is worth anything. If our prices were any lower, there would be no way we could stay in business.
4. You can play 3d games head-to-head at reasonable resolutions with reasonable frame rates, but these are not hard core gaming machines. You are free to use more powerful video cards, any nVidia cards will work.
5. These systems are extremely fast for the money, you would have to pay 50% more to get 10% better real world performance. Our generous amounts of RAM and fast hard drives are key. We optimize for cost/performance.
6. There is absolutely no network latency, that is the beauty of local multi-user systems.
Thin client setups cannot even begin to compare in terms of performance, and they aren't cheaper if you're buying new hardware.
Thank you for the feedback, and if you have any other questions drop up a line.
you're right, obviously Microsoft can and will compete in this arena, but from a technical and cost standpoint I'm confident that the open source solutions have a considerable advantage.
Whether or not big corporations choose those open source solutions is a different story.
Even low end thin clients can cost $300 each, not to mention the cost of the server, and the performance you get running 40 people off one box is abysmal compared to local multi-user systems, where you usually can't tell there are other people sharing the machine, and you can run 3D games and see full motion video without a problem.
If you wanted, you could even run a 4-user local multi-user system as 4 thin clients, and get the thin-client administration advantages combined with the local multi-user advantages like power, noise, space, and cost savings.
Unless you have a ton of old reliable boxen to run LTSP or other thin client solutions on, thin clients are way too expensive new for what you get. Local Multi-user systems are much more efficient. Especially when running 4 people on one box, open source (free) software is the only way to avoid killer software costs, so I don't think Microsoft can compete in this arena. You can get new hardware (and all the software you need) for 3 or 4 users for less than $1000 with an open source solution. For more info on local multi-user systems, check out http://groovix.com/ (that's my company, so obviously I'm biased!)