Timeseal

[Richard Allbert]

This could win stupid question of the year... but where do I get timeseal for the mac and for my raspberry pi??

The best I could turn up was here:

https://www.unix-ag.uni-kl.de/~chess/soft/timeseal/

But these are very old - and tbh I've no idea which is applicable.

Freechess.org say to use their ftp.freechess.org, but it doesn't seem to be active.

Where do people get it?

Thanks!

[bob]

This could win stupid question of the year... but where do I get timeseal for the mac and for my raspberry pi??

The best I could turn up was here:

https://www.unix-ag.uni-kl.de/~chess/soft/timeseal/

But these are very old - and tbh I've no idea which is applicable.

Freechess.org say to use their ftp.freechess.org, but it doesn't seem to be active.

Where do people get it?

Thanks!

Send me an email, I can send you one that will work on linux and one for OS X.

[Richard Allbert]

Email sent, to the one on the https://www.cis.uab.edu/hyatt/ page.

Thanks so much!

[brtzsnr]

I use this one to connect zurichess to fics http://www.bergo.eng.br/chess/openseal.c

[Richard Allbert]

I use this one to connect zurichess to fics http://www.bergo.eng.br/chess/openseal.c

Thanks to you also

Thank goodness or forums...

[Dann Corbit]

It's broken.

From the C-FAQ:

Code: Select all

3.3b:	Here's a slick expression:

		a ^= b ^= a ^= b

	It swaps a and b without using a temporary.

A:	Not portably, it doesn't.  It attempts to modify the variable a
	twice between sequence points, so its behavior is undefined.

	For example, it has been reported that when given the code

		int a = 123, b = 7654;
		a ^= b ^= a ^= b;

	the SCO Optimizing C compiler (icc) sets b to 123 and a to 0.

	See also questions 3.1, 3.8, 10.3, and 20.15c.

Notice:

Code: Select all

#define SC(A,B) s[B]^=s[A]^=s[B],s[A]^=s[B]

[petero2]

It's broken.

From the C-FAQ:

Code: Select all

3.3b:	Here's a slick expression:

		a ^= b ^= a ^= b

	It swaps a and b without using a temporary.

A:	Not portably, it doesn't.  It attempts to modify the variable a
	twice between sequence points, so its behavior is undefined.

	For example, it has been reported that when given the code

		int a = 123, b = 7654;
		a ^= b ^= a ^= b;

	the SCO Optimizing C compiler (icc) sets b to 123 and a to 0.

	See also questions 3.1, 3.8, 10.3, and 20.15c.

Notice:

Code: Select all

#define SC(A,B) s[B]^=s[A]^=s[B],s[A]^=s[B]

I don't think that is broken, as long as you don't invoke the macro with A==B. Notice the comma in the #define. The comma operator introduces a sequence point.

[mar]

Code: Select all

#define SC(A,B) s[B]^=s[A]^=s[B],s[A]^=s[B]

I don't see the need to do xor-swap either, compilers will use a register anyway, so the only cool thing about xor-swap is that it obfuscates code.
Not to mention that this trick won't work with floats.

[Dann Corbit]

Code: Select all

#define SC(A,B) s[B]^=s[A]^=s[B],s[A]^=s[B]

I don't see the need to do xor-swap either, compilers will use a register anyway, so the only cool thing about xor-swap is that it obfuscates code.
Not to mention that this trick won't work with floats.

As Peter points out, it is not (strictly speaking) illegal.

It is clear from the code that it is never called with A==B as the XOR inputs, so it probably also created correct results.

The fact that the macro is not typesafe also does not matter, since it is only used with integer arguments in the function.

But I am with you -- a simple swap using a temp would have been more clear and therefore better. It's not like we will produce wrong timing results if we lose 10 cycles.

[mar]

It is clear from the code that it is never called with A==B as the XOR inputs, so it probably also created correct results.

xor trick works even if A==B:

assume a=44, b=44:

Code: Select all

a ^= b (a = 0, b = 44)
b ^= a (a = 0, b = 44)
a ^= b (a = 44, b= 44)