What's New: New Features in Version 22.x of Prime95
NEW FEATURES IN VERSION 22.x of PRIME95.EXE
New Features in Version 22.13 of Prime95
A P-1 and ECM QA suite was implemented. A bug in ECM for exponents below 172,700
and near the limit of an FFT size and using SSE2 code was fixed.
New Features in Version 22.12 of Prime95
A bug was fixed that caused some factors to be missed in stage 2 of P-1 when
the available memory did not let the program allocate 12 temporary variables.
If testing a number in the 16 millions using an FFT size of 768K, then each temporary
takes 768K * 8 bytes or 6MB. If your memory setting was less than 72MB (6MB *
12 temporaries) then you were affected by the bug. Actual the program allocates
some fixed tables so anything less than about 75MB triggered the bug.
New features in Version 22.11 of Prime95
A bug that caused factors to be missed in the last stage of P-1 and ECM factoring
was fixed. The bug was introduced in executables built after Sept. 28, 2002.
New features in Version 22.10 of Prime95
When running a double-check assignment the preferred trial factoring depth
is reduced one bit. This is done because finding a factor will save one LL test
instead of the usual two LL tests.
On a Windows NT/2000/XP system, if prime95 is run by a user without administrator
privileges then prime95 cannot install itself as a service. Thus, the Start at
Bootup option will set prime95 to run at login time rather than bootup time.
A crash bug when selecting "Start at Bootup" on Windows NT3 and Windows NT4
systems was fixed.
Bug in starting more than one prime95 as NT services on multi-processor machines
was fixed.
New features in Version 22.9 of Prime95
Error rate for a clean run is now estimated at 1.8% raising your chances of
finding a Mersenne prime while double-checking.
You can now stop and continue testing from the system tray menu.
You can now pause prime95 when another program starts running. See the PauseWhileRunning
option in undoc.txt.
Fixed bug introduced in 22.8 where No Icon did not work if Start at Bootup
was also specified in Windows 98.
A bug in unscrambling the proxy server password in primenet.ini was fixed.
New features in Version 22.8 of Prime95
Soft FFT crossovers have been implemented. If you test an exponent that is
within 0.2% of the old hard FFT crossover point, then 1000 test iterations are
run to determine if the smaller or larger FFT is appropriate for the exponent.
Although not recommended, you can adjust prime95's behavior to be more aggressive
using the SoftCrossover and SoftCrossoverAdjust features discussed in undoc.txt.
To better stress main memory, the torture test will now use up to the amount
of memory specified in the Options/CPU dialog box.
Iterations with roundoff checking are a little faster for non-SSE2 CPUs.
New features in Version 22.7 of Prime95
Trial factoring above 2^64 on a P4 now uses the SSE2 instructions. This is
more than four times as fast as the previous version.
Old Advanced/Factoring save files are not compatible with this version. They
are named p0000000 and should be deleted. This should not be a problem as no one
should be using this obsolete and unsupported feature.
Four more x87 FFT crossover changes as suggested by Brian Beesley's research.
You can now change the NT/2000/XP service name and service display name. See
undoc.txt for details.
Bug where found factors were silently truncated to 200 digits was fixed.
New features in Version 22.6 of Prime95
The Start At Bootup option now runs prime95 as a Windows NT/2000/XP service!!
Normally Windows NT/2000/XP will not run GUI programs as services, but several
tricks have allowed us to get this working almost flawlessly.
To reduce spurious error reports on x86 LL tests near the crossover points,
the program now allows roundoff errors less than 0.40625 rather than less than
0.4.
New features in Version 22.5 of Prime95
A bug in P4 roundoff error recovery was fixed. Bug was present in version
22 executables dated between May 22 and June 18.
Prime95 now properly detects if the program is running by another logged in
Windows XP user.
When Tray Icon or Hide Icon is set, the "X" in the upper right corner no longer
closes prime95 rather it minimizes it. Also right clicking the tray icon now brings
up a popup menu.
Changed the way prime95 shuts down. Logoff in WinXP used to hang.
New features in Version 22.4 of Prime95
A little more tweaking of the FFT crossovers.
A GCD hang bug was fixed in case where there were a large number of entries
in the lowm.txt file. Another GCD bug affecting ECM on F18 was fixed.
New features in Version 22.3 of Prime95
Given more data, the roundoff error checking is now done on every iteration
for exponents within 0.5% of the maximum that can be tested by the current FFT
length. If a roundoff error above 0.4 occurs, then the iteration is now repeated
without any change to the shift count. It now takes a roundoff error greater than
0.6 to corrupt the results.
Many of the FFT ranges have changed. Version 21 was too aggressive in choosing
FFT sizes for the P4. The new handling of roundoff errors above 0.4 lets us be
more aggressive with the non-P4 code.
Result lines are now WYn rather than WXn.
A crash bug affecting P-1 and ECM using the 2^N+1 option for large N was fixed.
A rare memory corruption and possible crash bug in the GCD code was fixed.
The -t command line argument will run the torture test.
To reduce wild fluctuations in the RollingAverage, it will be updated roughly
twice per day.
New features in Version 22.2 of Prime95
Factoring assignments go to machines slower than a 233 MHz Pentium. Double-checks
are assigned to machines slower than a 500 MHz P-III.
The speed of the GCD has been improved.
Some of the P4 FFT ranges have changed. Version 21 was too aggressive in choosing
FFT sizes for the P4.
Roundoff error checking is now done on every iteration for exponents within
2% of the maximum that can be tested by the current FFT length. See undoc.txt
for ways to disable this new feature.
The previous version would get into an infinite loop if a repeatable roundoff
error larger than 0.48 was discovered. Now, if a repeatable roundoff error greater
than 0.45 is found, then the iteration is repeated after shifting the FFT data.
New features in Version 22.1 of Prime95
Some bugs in error handling when communicating with the server have been fixed.
Communicating with the server by RPC is no longer supported. The HTTPNET.DLL
and RPCNET.DLL have been deleted. Only "Basic" Proxy server authentication is
supported. Version 21 may have supported MS Proxy Server 2.0's NTLM (NT Lan Manager)
challenge/response authentication.
The program now uses a high resolution timer rather than the RDTSC instruction
to time events. This should help the program display accurate timings on laptops
with SpeedStep or desktops that can hibernate. You can force the program to use
the old RDTSC timing method with the RdtscTiming option described in undoc.txt.
The program uses a updated algorithm to automatically detect CPU type and
speed. The Options/CPU dialog box no longer let's you set this information. Instead,
the Options/CPU dialog displays the detected CPU type and speed. This new feature
helps prevent incorrect settings when users upgrade or try several overclocking
speeds. If the new algorithm fails, you can override the settings as described
in undoc.txt.
Benchmark now writes the program version, timing methodology, cpu type and
speed, L1 and L2 cache information to results.txt. It will refuse to benchmark
if rdtsctiming is 10.