[Entertainment] When tuning goes wrong...

[ZirconiumX]

I've been tuning my PST array in Durandal (based on Sungorus 1.4) - and CLOP has given me some absurd values - and part of me says to ignore the values because the Sungorus eval was $*%# and needs replacing.

For your entertainment - find a copy of Sungorus 1.4 read through the PST init code and replace the values of line[] with this:

Code: Select all

{ 0, 46, 2076, 2595, 2890, 2312, 320, 0 }

Yes, those were the values that I was given by CLOP. I really think I should redo the search and eval to my liking - the Sungorus author has forgotten the rule of 'Premature optimization is the root of all evil'.

Matthew:out

[Richard Allbert]

Harsh criticism, and unjustified.

Sungorus is Strong and the eval really well balanced, especially when you consider the amount of parameters vs strength.

What gives you the right to publicly criticize it in this way?

[ZirconiumX]

Can we agree to disagree here? I really don't want to start an argument, Richard.

I personally think that Sungorus needs room to grow for want of a better phrase - Fruit is an example of having room to grow. Sungorus is optimized so much that you cannot make a few changes to the eval without a net Elo loss due to loss of speed.

With the Toga series of Fruit clones it is quite clearly Fruit.

The only Sungorus derivative I know of is Rodent - which from what I can tell rewrote most of the search, move ordering, etc to have more room to grow.

Matthew:out

[Richard Allbert]

I'm not arguing at all. I'm simply pointing out that you stated publicly that the author has 'forgotten the rule of premature optimisation'. Did you ask him? you also said the eval is *#*^. Did you ask the author about this?

My guess would be not.

Fits in well with the other endless drivel you post commenting on other peoples' engines.

[ZirconiumX]

Yup, that's me - a pile of drivel. Couldn't have described me better.

You do know I still have depression, and am still grieving for the loss of my mum? Or is that more drivel?

Matthew:out

[ZirconiumX]

I may be the resident dorpsgek around here, but occasionally I do useful things, like help people out, and cheer people up.

Mr Drivel.

[mar]

The reason why Pawel chose Sungorus must be design, read it is well designed (simple and clear), a few weights mean nothing.
I made a lot of subtle design mistakes myself, like en passant square being the same as last pawn push by two
instead of 1 square behind it and a poor choice of square numbering (I have bit 0 = A8 and bit 63 = H1).
The latter is problematic in that I have to number ranks in reverse order, i.e. rank1 is 7 and rank8 is 0 (to keep things simple).
No problem except that I can't compare ranks directly like if (a<b) ...
I find it a pity when authors of original open source engines are being criticized, that's cheap.
Even if you were right - one more reason for me to never go open source again - noone prevents you from starting
from scratch to show that you can do better

Cheers

Martin

[lucasart]

I've been tuning my PST array in Durandal (based on Sungorus 1.4) - and CLOP has given me some absurd values - and part of me says to ignore the values because the Sungorus eval was $*%# and needs replacing.

For your entertainment - find a copy of Sungorus 1.4 read through the PST init code and replace the values of line[] with this:

Code: Select all

{ 0, 46, 2076, 2595, 2890, 2312, 320, 0 }

Yes, those were the values that I was given by CLOP. I really think I should redo the search and eval to my liking - the Sungorus author has forgotten the rule of 'Premature optimization is the root of all evil'.

Matthew:out

Your gratuitious "comments" are not very intelligent and constructive:
- can you explain why Sungorus eval is "$*%# and needs replacing" ?
- can you explain how you used CLOP ? And when you stopped the CLOP experiment to use the QLR estimated values ?

Criticizing is easy, but what about writing your own engine ? What happenned to that project ?

[zullil]

I'm not arguing at all. I'm simply pointing out that you stated publicly that the author has 'forgotten the rule of premature optimisation'. Did you ask him? you also said the eval is *#*^. Did you ask the author about this?

My guess would be not.

Fits in well with the other endless drivel you post commenting on other peoples' engines.

Matthew is just 13 years old. He very recently lost his mother. Perhaps he shouldn't have made the post that he did, but let's please take it easy on him.

[Adam Hair]

I've been tuning my PST array in Durandal (based on Sungorus 1.4) - and CLOP has given me some absurd values - and part of me says to ignore the values because the Sungorus eval was $*%# and needs replacing.

For your entertainment - find a copy of Sungorus 1.4 read through the PST init code and replace the values of line[] with this:

Code: Select all

{ 0, 46, 2076, 2595, 2890, 2312, 320, 0 }

Yes, those were the values that I was given by CLOP. I really think I should redo the search and eval to my liking - the Sungorus author has forgotten the rule of 'Premature optimization is the root of all evil'.

Matthew:out

Can you describe exactly how you went about using CLOP to optimize those values? That is a result that I would expect to see if there was a problem in the test setup.