APassionForCriminalJustic wrote:I am hopeful that the release of Komodo 9 is only just around the corner. Please Mark and Larry, give us a new Komodo.
Is there anybody else out there eagerly dying for the release of Komodo 9?
I very much hope Komodo 9 will not be released until they reach at least 50 elo improvement over Komodo 8.
40 elo could also qualify as lowest bound.
The point is, a human simply can not make a distinction between an engine and another engine playing 20 elo stronger. You simply can not see what is improved, can not feel it. 50 or 40 elo though make sense.
Then just wait for the next version. There's no harm releasing something for those that want to buy it.
There is big harm, as this would be unethical.
How much are you going to pay for Komodo 9 that is 25 elo stronger than Komodo 8?
I guess half of what you would be ready to pay, if the new version was 50 elo stronger than older version.
But that is not the main point. The main point is that you simply will not notice any observable differences in Komodo play, the same way it is difficult to tell apart 2 GMs of roughly equal strength, but easy to do so with GMs that are 50-100-200 elo apart.
Not so.
First your point about ethics: if the team don't promise a certain figure, there is nothing unethical about it. If they declare openly what the figure is, they could ethically release K9 with just 1 elo improvement.
Secondly, your point about the difference being immeasurable : I can think of at least several situations where this doesn't apply. Someone could be upgrading from K7, where the difference will be obvious.
As for me I'm upgrading from K8 where even a small improvement of 30 elo is very measureable. A 30 elo improvement is about the same as a +40% hardware speed increase. To achieve a 40% speed increase on my IDeA network would cost about $20,000. Certainly I can measure the difference.
APassionForCriminalJustic wrote:I am hopeful that the release of Komodo 9 is only just around the corner. Please Mark and Larry, give us a new Komodo.
Is there anybody else out there eagerly dying for the release of Komodo 9?
I very much hope Komodo 9 will not be released until they reach at least 50 elo improvement over Komodo 8.
40 elo could also qualify as lowest bound.
The point is, a human simply can not make a distinction between an engine and another engine playing 20 elo stronger. You simply can not see what is improved, can not feel it. 50 or 40 elo though make sense.
Then just wait for the next version. There's no harm releasing something for those that want to buy it.
There is big harm, as this would be unethical.
How much are you going to pay for Komodo 9 that is 25 elo stronger than Komodo 8?
I guess half of what you would be ready to pay, if the new version was 50 elo stronger than older version.
But that is not the main point. The main point is that you simply will not notice any observable differences in Komodo play, the same way it is difficult to tell apart 2 GMs of roughly equal strength, but easy to do so with GMs that are 50-100-200 elo apart.
Not so.
First your point about ethics: if the team don't promise a certain figure, there is nothing unethical about it. If they declare openly what the figure is, they could ethically release K9 with just 1 elo improvement.
Secondly, your point about the difference being immeasurable : I can think of at least several situations where this doesn't apply. Someone could be upgrading from K7, where the difference will be obvious.
As for me I'm upgrading from K8 where even a small improvement of 30 elo is very measureable. A 30 elo improvement is about the same as a +40% hardware speed increase. To achieve a 40% speed increase on my IDeA network would cost about $20,000. Certainly I can measure the difference.
I remember hearing Mark say that they aim for at least 25 Elo before a new release of Komodo. Therefore, there would be absolutely nothing unethical about Komodo 9 with 30 Elo worth of improvement. There are also SMP improvements more than likely that will be good for most of us.
Where do you get your calculation that 30 Elo worth of improvement is approximately the equivalent to +40 percent hardware-speed increase?
Anyways, all good things come in good time. Komodo 9 is definitely something worth looking forward to.
APassionForCriminalJustic wrote:I am hopeful that the release of Komodo 9 is only just around the corner. Please Mark and Larry, give us a new Komodo.
Is there anybody else out there eagerly dying for the release of Komodo 9?
I very much hope Komodo 9 will not be released until they reach at least 50 elo improvement over Komodo 8.
40 elo could also qualify as lowest bound.
The point is, a human simply can not make a distinction between an engine and another engine playing 20 elo stronger. You simply can not see what is improved, can not feel it. 50 or 40 elo though make sense.
This is a very interesting perspective surrounding Elo improvement. Do you really think that this is true? Remember, Komodo development right now is somewhere around 30 Elo. Mark and Larry would never release an update with only 20 Elo worth of improvement.
lkaufman wrote:I voted for "Sometime in March" because I think we will reach +40 over Komodo 8 sometime this month. Not a promise, just the way I would bet.
I voted "Sometime in March", since it does not specify which year!