OpenChess

What happens when the freeware engines like IvanHoe, Fire, Houdini, Stockfish, etc. become stronger than the Rental Rybka 4 Clusters? I think Rental Rybka 4 Cluster will not reign for very long at all considering that the commercial release was expected to fend off the competition until the Rental System was fully designed and implemented. Seeing how that failed as Vas usually does, I think the freeware engines (and possibly freeware clusters) already have begun the process of overpowering "commercial" chess software.

xshat wrote:What happens when the freeware engines like IvanHoe, Fire, Houdini, Stockfish, etc. become stronger than the Rental Rybka 4 Clusters? I think Rental Rybka 4 Cluster will not reign for very long at all considering that the commercial release was expected to fend off the competition until the Rental System was fully designed and implemented. Seeing how that failed as Vas usually does, I think the freeware engines (and possibly freeware clusters) already have begun the process of overpowering "commercial" chess software.

How will you or any of the testers measure the strength of rental Rybka? It is in the clouds on unknown hardware.

Harvey Williamson wrote:

xshat wrote:What happens when the freeware engines like IvanHoe, Fire, Houdini, Stockfish, etc. become stronger than the Rental Rybka 4 Clusters? I think Rental Rybka 4 Cluster will not reign for very long at all considering that the commercial release was expected to fend off the competition until the Rental System was fully designed and implemented. Seeing how that failed as Vas usually does, I think the freeware engines (and possibly freeware clusters) already have begun the process of overpowering "commercial" chess software.

How will you or any of the testers measure the strength of rental Rybka? It is in the clouds on unknown hardware.

In the end the programming is more important than the hardware for chess engines.

I predicted this scenario before many moons,even before the appearance of Ippo/Robbo family....it's simply logical....
Dr.D

Even if not hosting the strongest engine available, cloud computing might appeal to certain users for other reasons.

Just to name a few,
1) better hardware than most of us have at home can compensate for a certain elo difference.

2) the cluster can be used on demand, eg. if you feel the need to prepare a certain opening line in a limited amount of time with tools like IDEA, it might be useful to rent another couple (n) of cores for a couple of hours (t), instead of spending n*t hours without using the service.

3) regular updates ?

4) reduced noise pollution in your living room / office

5) for less frequent users it might pay off to rent vs to buy a new piece of hardware, just because the rental-expense/year would be lower than the depreciation of the bought hardware

---

Of cause one can also mention a list of disadvantages, but after all the users have to decide for themselves where there priorities lie.

Cloud computing is a growing field. In the near future I wouldn't be surprised if there arose an independent organization that offered for any engine author to upload their engine, for any opening book author to upload their book, etc. And then the users were free to choose which to use. Once this comes I agree that the days of the Rybka cloud are counted.

Edmund wrote: 4) reduced noise pollution in your living room / office

But you would have to switch your heating back on in the winter The Rybka cloud concept also works on the assumption you do not want to run any other engines on a fast machine, if the reason to use it is to save money on hardware.

Edmund wrote:Even if not hosting the strongest engine available, cloud computing might appeal to certain users for other reasons.

Just to name a few,
1) better hardware than most of us have at home can compensate for a certain elo difference.

2) the cluster can be used on demand, eg. if you feel the need to prepare a certain opening line in a limited amount of time with tools like IDEA, it might be useful to rent another couple (n) of cores for a couple of hours (t), instead of spending n*t hours without using the service.

3) regular updates ?

4) reduced noise pollution in your living room / office

5) for less frequent users it might pay off to rent vs to buy a new piece of hardware, just because the rental-expense/year would be lower than the depreciation of the bought hardware

---

Of cause one can also mention a list of disadvantages, but after all the users have to decide for themselves where there priorities lie.

Cloud computing is a growing field. In the near future I wouldn't be surprised if there arose an independent organization that offered for any engine author to upload their engine, for any opening book author to upload their book, etc. And then the users were free to choose which to use. Once this comes I agree that the days of the Rybka cloud are counted.

My experience of being a publisher is that it always pays off to be first onto a new system/platform, not necessarily because the software release sells many units (or in this case gets rented) but because it gives your company a leg up into other developments for the platform, requests from other publishers to do other work for the platform, and offers from freelancers asking for their material to be published. Usually the money flows from these in time.

Most of the criticism of Vas for his cloud project concentrate on the chess only and miss the future arising opportunities.

Harvey Williamson wrote:

Edmund wrote: 4) reduced noise pollution in your living room / office

But you would have to switch your heating back on in the winter The Rybka cloud concept also works on the assumption you do not want to run any other engines on a fast machine, if the reason to use it is to save money on hardware.

Yes exactly, thats where I see a good potential for an independent organisation. Also I agree with Chris Whittington that timing is very critical.
So is there anyone with the right kind of entrepreneurial spirit interested?

Edmund wrote:

Harvey Williamson wrote:

Edmund wrote: 4) reduced noise pollution in your living room / office

But you would have to switch your heating back on in the winter The Rybka cloud concept also works on the assumption you do not want to run any other engines on a fast machine, if the reason to use it is to save money on hardware.

Yes exactly, thats where I see a good potential for an independent organisation. Also I agree with Chris Whittington that timing is very critical.
So is there anyone with the right kind of entrepreneurial spirit interested?

I see no entrepreneurs in computer chess. Maybe Vas is one but the rest are technicians who mostly give up their software for others to publish, or, if they publish themselves (very few) are incompetent at it.

Chris Whittington wrote:

Edmund wrote:... Yes exactly, thats where I see a good potential for an independent organisation. Also I agree with Chris Whittington that timing is very critical.
So is there anyone with the right kind of entrepreneurial spirit interested?

I see no entrepreneurs in computer chess. Maybe Vas is one but the rest are technicians who mostly give up their software for others to publish, or, if they publish themselves (very few) are incompetent at it.

Well, I for one am a student at a business university with the right ambitions. However for such a project I simply lack the technical know how. A hobby-project coming closest to cluster computing I ever did was writing a TCP/IP program capable of sending text messages between two computers of mine standing next to each other.