New Beta Client Out.......

[Jator]

....and this one doesn't have an expiration date. (Saw this on TA News Page)

8014.468

EDIT - Here's the list of improvements on the 8013 and 8014 cores:

2.8014.468 fix: all: bad results caused by build .466.
fix: all: packets were saved and reloaded after buffer update.
fix: all: current packet wasn't discarded on contest closure.
fix: all: exited without crunching with "-n 1" option.

2.8013.467 new: 68k: new, faster RC5 cores (previously only available in
the AmigaOS client)
new: x86: new, faster RC5 core for the P4.
new: ia64: new 4 pipeline ansi core, used by ia64.
imp: all: add support for a second argument to -test/-bench, the
number to use. example: 'dnetc -test rc5 8'
imp: some: OGR behaves more predictably on non-preemptive systems.
new: macos: dnetc can now find its support files in the active
systems "Application Support" or "Preferences" folder.

 

[Fardringle]

Nice! ~16% speed increase in OGR on Celeron2/PII systems over the 463 client! 🙂

 

[ottawanker]

just to add a note to that, on my duron and my k6-2, the 468 client is still slower than the 463b..

 

[optoman]

Has anyone upgraded from 463 and seen an increase in performance for RC5?

 

[Adul]

Man I got to find a P4 to test out new core.

 

[BGod]

It was a big letdown for me:
dnetc v2.8010-463-CTR-00071213 for Win32 (Windows 4.10).
[Mar 14 04:46:35 UTC] Benchmark for RC5 core #6 (RG/HB ath)
0.00:00:16.19 [4,661,524.32 keys/sec]

dnetc v2.8014-468-GTR-01031016 for Win32 (Windows 4.10).
[Mar 14 04:50:01 UTC] RC5: Benchmark for core #6 (RG/HB re-pair II)
0.00:00:17.28 [4,585,812 keys/sec]

I'll stick with what works for now. 🙁

 

[phatstyl]

man 463 just wont go away =(, the people at DNET gotta try harder =)

 

[Viztech]

That's why the 8011-464g was so popular with discriminating OGR counters. It was better than 10% faster on P2/P3/Celerons, but slower with K6s.

viz

 

[micron]

The beta is slower in RC5 on my T-Bird than 8010.463 🙁.

 

[Jator]

I benchmarked the 8013 on my fathers P4. There is nothing really to get excited about, no great speed improvement. 🙁

 

[Mad Pierre]

I think 8014 is a bit slower because there is debug code in the client. It should speed up when it's released

 

[JonB]

While the new 8014 client ran flawlessly, it did not run as fast as 436b, so I've switched back. Mad Pierre, I hope you're correct about the debug code. The executable certainly is larger.