locked Re: Decode lag


Andy_501 <andrew.webb.501.ve4per@...>
 

Just managed yesterday to get JTDX working reliably on Linux ubuntu 20.04LTS OS ; IC-7300 rig;  JTDX v 2.2.0rc.155

hardware list:

*-cache:0
          description: L1 cache
          physical id: 45
          slot: L1 Cache
          size: 32KiB
          capacity: 32KiB
          capabilities: synchronous internal write-back data
          configuration: level=1
     *-cache:1
          description: L1 cache
          physical id: 46
          slot: L1 Cache
          size: 32KiB
          capacity: 32KiB
          capabilities: synchronous internal write-back instruction
          configuration: level=1
     *-cache:2
          description: L2 cache
          physical id: 47
          slot: L2 Cache
          size: 256KiB
          capacity: 256KiB
          capabilities: synchronous internal write-back unified
          configuration: level=2
     *-cache:3
          description: L3 cache
          physical id: 48
          slot: L3 Cache
          size: 8MiB
          capacity: 8MiB
          capabilities: synchronous internal write-back unified
          configuration: level=3
     *-cpu
          description: CPU
          product: Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz
          vendor: Intel Corp.
          physical id: 49
          bus info: cpu@0
          version: Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz
          serial: To Be Filled By O.E.M.
          slot: LGA1151
          size: 1888MHz
          capacity: 4200MHz
          width: 64 bits
          clock: 100MHz
          configuration: cores=4 enabledcores=4 threads=8
     *-memory
          description: System Memory
          physical id: 4a
          slot: System board or motherboard
          size: 32GiB
          description: DIMM DDR4 Synchronous 2133 MHz (0.5 ns)

Resource monitor on Panel indicates max of up to 40% CPU when JTDXjt9 -s time slice is active all other workspace timeslices are in range of 11%- 20% CPU usage. Max Mem usage with all apps running using multi workspaces less than 4 GB and zero swap space is used.

Apps running, weewx wx stn software, firefox 9 tabs open, chromium 6 tabs open but not streaming (streaming sites),chrome (qrz.com db lookup, hamqth.net db lookup, Cdn Government ham call db search, PSKreporter) GridTracker, Thunderbird email, JTDX,

When I see the Band Activity window top line showing headers for text message columns and Avg.= and Lag= and NO signal coming in I get Lag -0.1 /0 as traffic shows up and increases the Lag indicator shows -0.04/40 with Avg=0.18.

The time, freq and DT I understand but I can't find how to interpret the Lag or Avg figures or understand what exactly they represent. If I click on Decode button Lag just goes to /54 and remains there until the next decode cycle.

Can anyone point me to a good wiki or floss type manual explanation respecting Lag and Avg measurements are arrived at as the ops manual and user manual do not seem to touch on the subject at all let alone even in sketchy detail.

Seems strange that the resource monitor/task manager taskbar panel indicators indicate max CPU resources peak out at 40% only when time slice devoted to JTDX and the rest of the time less than 20% and memory never exceeds 1/8 of what is available that the software would seem to be taxed so high as contact messages increase only to effect Lag and Avg. so widely.

 Thanks all pointers appreciated.

VE4PER  Andy



On 2021-01-01 9:17 a.m., Nc8q-mesh@... wrote:

Sounds to me like a huge operating system on a minimal CPU.
View system resources to see how busy the system/CPU is.
During a decode you may see one core/thread at 100%.
The decode elapsed time depends on CPU performance.
Response time is based on level of decode setting.
Deep consumes maximum CPU time.
Fast uses less.
Normal uses least.
I hope this helps,
Chuck


Join main@WSJTX.groups.io to automatically receive all group messages.