PFGW 4.0.4 (with gwnum v30.10) Released

[storm5510]

Originally Posted by rogue

I have tracked down and fixed the "40 bit" problem with primality testing. This code was changed over a decade ago and only impact terms less than 40 bits in size. I just had to revert part of those changes. In other words this bug has impacted all releases since 2011, which is when that change was made.

I will release a new build of pfgw once George releases an updated gwnum because it has a bug that causes pfgw to crash.

I found 4.0.7 Beta on Sourceforge. Is this what you are referring to?

[rogue]

Originally Posted by storm5510

I found 4.0.7 Beta on Sourceforge. Is this what you are referring to?

The beta has a change that someone had me implement. They are testing the changes. It is not ready for release.

[storm5510]

Originally Posted by rogue

The beta has a change that someone had me implement. They are testing the changes. It is not ready for release.

I gave it a rigorous task to do yesterday evening. It is still running now.

[gd_barnes]

Is there an update on the fix for the 40-bit problem?

[rogue]

Originally Posted by gd_barnes

Is there an update on the fix for the 40-bit problem?

I have not released it it because I was waiting for a fix on the GEC code. If I don't get a fix this week for it then I will release with the GEC code.

[Prime95]

Originally Posted by rogue

I have not released it it because I was waiting for a fix on the GEC code. If I don't get a fix this week for it then I will release with the GEC code.

30.13 sources have been released. Are you waiting on anything else from me?

[rogue]

Originally Posted by Prime95

30.13 sources have been released. Are you waiting on anything else from me?

I forgot about that. I am not waiting for anything from you now. Thank you!

[rebirther]

pfgw 4.0.6 is bugged,

150*295^292+1 is prime --->pfgw4.0.4, llr, llr2, with pfgw4.0.6 is not prime

tested a smaller range of S85 up to k=1000, n=2500, pfgw4.0.6 missed 3 primes and add them to remaining, pfgw4.0.4 found all

Its better not using this version on new bases

@rogue: Can you pls release pfgw4.0.8 with gwnum30.13 for testing? Otherwise we need to use older versions and maybe rerun all new bases while a lot of primes are missing

Note: Winpfgw 4.0.7 is stucking on testing S295 to max k=1000 (stucking at 150)

[rogue]

Originally Posted by rebirther

pfgw 4.0.6 is bugged,

150*295^292+1 is prime --->pfgw4.0.4, llr, llr2, with pfgw4.0.6 is not prime

tested a smaller range of S85 up to k=1000, n=2500, pfgw4.0.6 missed 3 primes and add them to remaining, pfgw4.0.4 found all

Its better not using this version on new bases

@rogue: Can you pls release pfgw4.0.8 with gwnum30.13 for testing? Otherwise we need to use older versions and maybe rerun all new bases while a lot of primes are missing

Note: Winpfgw 4.0.7 is stucking on testing S295 to max k=1000 (stucking at 150)

I'm waiting for a fix to the GEC code. If that is not ready by tomorrow I will disable that code and release.

[storm5510]

Originally Posted by rebirther

150*295^292+1 is prime --->pfgw4.0.4, llr, llr2, with pfgw4.0.6 is not prime...

I found something similar with PFGW v4.0.3. In a short run, using the -t switch causes it not to find much. Run it without, it finds much more. I cross checked it with cllr64. The results match PFGW's results without the switch.

I am not a fan of PRP. That is just me.

[rogue]

Originally Posted by storm5510

I found something similar with PFGW v4.0.3. In a short run, using the -t switch causes it not to find much. Run it without, it finds much more. I cross checked it with cllr64. The results match PFGW's results without the switch.

I am not a fan of PRP. That is just me.

The difference is that a primality test can take longer than a PRP test and in some cases much longer. It is better to do a PRP test first.