No, latency is one reason Google is developing a new mobile OS.

On Feb 22, 2018 6:27 PM, "Robert" <i...@avantwireless.com> wrote:

I believe this was a major part of the improvements in android


On 2/22/18 3:58 PM, Bill Prince wrote:

> Pretty sure you need  RTOS to accomplish this.That will get pretty close
> to bare metal.
>
> -bp
>
> --
> bp
> part15sbs{at}gmail{dot}com
>
> On Thu, Feb 22, 2018 at 3:36 PM, <ch...@wbmfg.com <mailto:ch...@wbmfg.com>>
> wrote:
>
>     Had the command syntax wrong.
>     But got nice to work.  Have to sudo if you use negative nice numbers.
>     It made zero difference in my jitter.  I went from 19 to –20 on nice
>     and no change.
>     *From:* ch...@wbmfg.com
>     *Sent:* Thursday, February 22, 2018 4:29 PM
>     *To:* af@afmug.com
>     *Subject:* Re: [AFMUG] OT Raspberry PI
>
>     The problem is there is a crap ton of stuff out there that needs
>     network sync.  And it all has a T1 as an input.
>     But most T1 trunking circuits are getting replaced with SIP.
>     So, I am building a cheap and dirty T1 signal generator that is GPS
>     and rhubidium referenced.  The hard part is easy.  The easy part
>     should be easy but all the T1 framing chips that used to exist no
>     longer exist.
>     The ones that are out there have massive CPU interfaces and tons of
>     registers that need to get set to get them fired up and running....
>     Where is Exar when you need them....
>     *From:* Adam Moffett
>     *Sent:* Thursday, February 22, 2018 4:21 PM
>     *To:* af@afmug.com
>     *Subject:* Re: [AFMUG] OT Raspberry PI
>
>     Tell whoever's got the T1 that 1967 is way behind us and get a new
>     interface.
>     Problem eliminated LOL
>     ------ Original Message ------
>     From: ch...@wbmfg.com
>     To: af@afmug.com
>     Sent: 2/22/2018 6:16:45 PM
>     Subject: Re: [AFMUG] OT Raspberry PI
>
>>     I have to generate an alternate mark inversion signal on 1.544 MHz
>>     with every 193rd bit following a t1 framing sequence.
>>     Sure wish a 555 could do that.
>>     *From:* Dave
>>     *Sent:* Thursday, February 22, 2018 4:10 PM
>>     *To:* af@afmug.com
>>     *Subject:* Re: [AFMUG] OT Raspberry PI
>>
>>     Find a 555 timer ... I used many in the olden day when radioshacks
>>     were king LOL!
>>
>>
>>     On 02/22/2018 05:05 PM, ch...@wbmfg.com wrote:
>>
>>>     I am thinking of using some shift registers instead of using the
>>>     PI output directly as the timing signal.
>>>     Use the PI to load them.
>>>     I love me some hardware design anyhow....
>>>     *From:* Colin Stanners
>>>     *Sent:* Thursday, February 22, 2018 3:59 PM
>>>     *To:* af@afmug.com
>>>     *Subject:* Re: [AFMUG] OT Raspberry PI
>>>
>>>     Other than setting the process priority, you may need a custom
>>>     kernel. See
>>>     https://medium.com/@metebalci/latency-of-raspberry-pi-3-on-s
>>> tandard-and-real-time-linux-4-9-kernel-2d9c20704495
>>>     <https://medium.com/@metebalci/latency-of-raspberry-pi-3-on-
>>> standard-and-real-time-linux-4-9-kernel-2d9c20704495>
>>>
>>>     On Feb 22, 2018 4:48 PM, <ch...@wbmfg.com> wrote:
>>>
>>>         Anyone know how to get my program to run on bare metal?
>>>         Or at the very least tell Linux that my program is the most
>>>         important thing in the world and service it above all other
>>>         things.
>>>         I am trying to create a timing signal with the Pi.  It is
>>>         doing it but the jitter is pretty bad.
>>>         I have researched trying to use an interrupt but there is a
>>>         pretty low limit on how many times per second you can fire a
>>>         hardware interrupt.
>>>         Too low for my application.
>>>
>>
>>     --
>>
>
>
>

Reply via email to