Hi Joe, Yan and all, This issue is resolved via copying WSJT-X.ini file from other computer.
Unfortunately I was too quick in the decision to accept overwrite prompt and have lost broken .ini file ... The issue is solved and you have saved my computer for further WSJT-X usage. Thanks a lot for your assistance. 73 Igor UA3DJY > >Message: 4 >Date: Thu, 12 Nov 2015 08:35:37 +0700 >From: "Yannick Devos (XV4Y)" < yannick.de...@online.fr > >Subject: Re: [wsjt-devel] JT9 decoder performance with AMD processor >To: WSJT software development < wsjt-devel@lists.sourceforge.net > >Message-ID: < 6a2f2d93-e1be-49a8-a62c-e5bf519cd...@online.fr > >Content-Type: text/plain; charset="utf-8" > >Hi Igor, > >I have not followed all this thread, but just a few questions: >- What is your receiver ? >- What is your sound card ? >- Is your computer over clocked ? > >>From my experience this is more related to a software problem (drivers, >>settings) or hardware (sound card, USB) but not related to the CPU itself. > >Here my experience: >The computer I used for SDR and things like WSPR is an "old" Pentium Dual Core >E5300. >Not a monster, but largely good enough for this work. >The old MSI motherboard died and I had hard time to find a replacement one. >What I found is a Gigabyte with a lot of over-clocking capabilities. >The problem is that the integrated soundcard I much much worse than the one in >the MSI motherboard. I can see the noise floor much higher, hiding all the >weak signals. I/Q is also much harder to balance and good image rejection is >hard to achieve (increasing again the noise). >Also, when I over clock the computer there seem to be some timing issue on the >audio streams that make the SDR and WSPR software crash a lot. >It is not related to the CPU or memory as no other software crash even when >doing CPU or memory intensive work. > >73, >Yan. >--- >Yannick DEVOS - XV4Y >http://www.qscope.org/ >http://xv4y.radioclub.asia/ > >Le 12 nov. 2015 ? 03:28, Joe Taylor < j...@princeton.edu > a ?crit : > >> Hi Igor, >> >> Here's a brief note responding to your many messages about different >> decoding results on different computers. >> >> On 11/11/2015 1:09 PM, ????? ? wrote: >>> Just got 96 decodes from 'AGC ON' files and 84 decodes from 'AGC OFF' files >>> while running r6058 on Win7 / AMD Phenom II X4 945 Processor. >>> >>> Hence this issue is related to some narrow branch of AMD processors like my >>> AMD Athlon II X2 255 Processor and has low priority. >> >> I have not had time to explore any of your reported issues in detail, >> but I am not inclined to take seriously your claim that identical data >> processed by identical, identically configured versions of WSJT-X can >> produce different decoding results. >> >> At first you were blaming differences between the way Win/XP and Win7 >> execute WSJT-X. Now apparently you are suggesting the differences have >> something to do with "AMD Phenom II X4 945" vs. "AMD Athlon II X2 255" >> processors. I don't buy either of these conclusions. >> >> I don't see how any such distinction can cause differences in decoder >> output. If configured in the same way and presented with the same data, >> identical software should produce identical output -- at least, up to >> differences caused by non-deterministic differences in the way time >> slices have been allocated to different threads. In the JT9/JT65 >> decoder, multi-threading is used only in the combined JT9+JT65 operating >> mode, so such a cause could not be present in single-mode JT9 or JT65 >> operation. >> >> What have you done to ensure that your software configuration is exactly >> the same on both test machines? Do you install WSJT-X on both test >> machines from a single package file? Have you tried exchanging the >> WSJT-X.ini files? >> >> -- 73, Joe, K1JT >>
------------------------------------------------------------------------------
_______________________________________________ wsjt-devel mailing list wsjt-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wsjt-devel
