RSTS boots very slow

[Bill B]

So I can now get to the "PiDP-11 simulator" menu on my old B+ by running Raspberry Pi OS Lite. I decided next to try and run RSTS/E without a PiDP-11 as described in the PiDP-11 manual. It took around 1 hr. for the startup to complete.

It takes around 10sec on my Ryzen desktop using simh.

So I assume my "too slow" message is the reason.

I also get a "Undervoltage detected!" message after starting the Pi.

I've been reading some of the posts in this group and it seems that power can be the cause of many problems. I'm using the power supply that came with my CanaKit...kit. It says it provides 5V. But I believe the specs recommend 5.1V. Can that small of a difference be a factor in my slowness?

[Marco]

An undervoltage warning on a Pi B+ won't slow it down much, but can lead to instability (unwanted reboots) and possible SD card corruption. It can either be caused by drawing too much power from the Pi's USB ports by peripherals (could even be  a mouse and/or keyboard), or a power supply delivering not enough Amps. I believe for a B+ the minimum is 1.8A. The 0.1 volts difference won't matter, in fact safe voltage values for a Pi are between 4.7 and 5.25 volts (but with 5.1 regarded as the optimum).

Pi's are picky with their power supply, and it is generally recommended NOT to use a phone charger or any other power supply not specifically intended for a Pi, because these are likely not to have enough Amperage, and/or do not actually supply the Amperage and Voltage as printed on the label.

But again, your Pi B+ is slow as it is, not much can be done to speed it up. Some overclocking maybe, with the risks that comes with that. For smooth operation you really need a Pi 3 or 4.

Op zaterdag 6 mei 2023 om 14:59:26 UTC+2 schreef bill...@gmail.com:

[Bill B]

Thank you so much for your response. I'll eventually buy a better Pi. If they ever return to reasonable prices. Which was one of their biggest selling points when they first came out.

[Eric Bruno]

The last projection I have seen is available and price are supposed to return to “normal” in Q3/Q4 2023

--
You received this message because you are subscribed to the Google Groups "[PiDP-11]" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-11+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-11/36c05ab0-5b32-408d-8226-2911e790947an%40googlegroups.com.

[Dave Gray]

Have a look at https://rpilocator.com/ - they keep an eye on what is available now. You'll have to move fast if you see what you want come available. Note that the official outlets (in the US, anyways) are contracted to sell at recommended list prices - which is pretty reasonable, but the supply isn't there yet.

To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-11/CANpxRj65aZKwurqVynBwz-X-UuP%3D1ACu8_KFWTH1WF3nN-6YvQ%40mail.gmail.com.

[isy...@gmail.com]

Dear All,

 RSTS does boot very slowly and I think this is as expected. It takes some time before the initial dialogue appears apart from the message DPxx does not interrupt. About 45 secs!

 The versions that are around have a lot of devices configured. I suspect that these are all probed to see if they exist. Later on in the boot process a long list of the disabled devices is printed.

 This can be up to 50;

 Increasing the Pi clock speed will not make any difference as the probe process is likely to have a timeout defined by the PDP11 system clock. As judged from boot times for a Pico at 200MHz and a PC 3GHz.

Regards, Ian.

[Johnny Billquist]

I suspect this isn't correct.

Probing for devices are not handled by some clock, but is based on bus
timeouts. On real hardware, there is a bit of timing in there, because
devices have a certain time allowed for responding on the bus, but with
emulation like simh, the non-existant address trap can happen
immediately, as it is well known if there is anything that is expected
to respond to an address.

RSX in simh boots within a few seconds, depending a bit on how fast
hardware simh is running on.

RSX do a little less probing than RSTS/E, so I could buy a little longer
time, but if it takes 45s to boot, it seems crazy to me.

Anyway, the autoconfiguration scheme for the PDP-11 is certainly not
timing based. It's all about accessing CSRs, and having the hardware
generate traps if there is nothing responding on the address.

Johnny

On 2023-05-08 16:03, isy...@gmail.com wrote:
> Dear All,
>
>  RSTS does boot very slowly and I think this is as expected. It takes
> some time before the initial dialogue appears apart from the message
> DPxx does not interrupt. About 45 secs!
>  The versions that are around have a lot of devices configured. I
> suspect that these are all probed to see if they exist. Later on in the
> boot process a long list of the disabled devices is printed.
>  This can be up to 50;
>  Increasing the Pi clock speed will not make any difference as the
> probe process is likely to have a timeout defined by the PDP11 system
> clock. As judged from boot times for a Pico at 200MHz and a PC 3GHz.
>
> Regards, Ian.
>
> On Monday, May 8, 2023 at 3:13:34 AM UTC+1 gray...@gmail.com wrote:
>

> Have a look at https://rpilocator.com/ <https://rpilocator.com/> -

> tooSlow.jpg
>
> I also get a "Undervoltage detected!" message after
> starting the Pi.
>

> Undervoltage.jpg
>
> I've been reading some of the posts in this group
> and it seems that power can be the cause of many
> problems. I'm using the power supply that came with
> my CanaKit...kit. It says it provides 5V. But I
> believe the specs recommend 5.1V. Can that small of
> a difference be a factor in my slowness?
>
> --
> You received this message because you are subscribed to the
> Google Groups "[PiDP-11]" group.
> To unsubscribe from this group and stop receiving emails
> from it, send an email to pidp-11+u...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/pidp-11/36c05ab0-5b32-408d-8226-2911e790947an%40googlegroups.com <https://groups.google.com/d/msgid/pidp-11/36c05ab0-5b32-408d-8226-2911e790947an%40googlegroups.com?utm_medium=email&utm_source=footer>.

>
> --
> You received this message because you are subscribed to the
> Google Groups "[PiDP-11]" group.
> To unsubscribe from this group and stop receiving emails from
> it, send an email to pidp-11+u...@googlegroups.com.
>
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/pidp-11/CANpxRj65aZKwurqVynBwz-X-UuP%3D1ACu8_KFWTH1WF3nN-6YvQ%40mail.gmail.com <https://groups.google.com/d/msgid/pidp-11/CANpxRj65aZKwurqVynBwz-X-UuP%3D1ACu8_KFWTH1WF3nN-6YvQ%40mail.gmail.com?utm_medium=email&utm_source=footer>.

>
> --
> You received this message because you are subscribed to the Google
> Groups "[PiDP-11]" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to pidp-11+u...@googlegroups.com

> <mailto:pidp-11+u...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/pidp-11/5c781086-b6e3-4e86-9d9c-18eced3dabbcn%40googlegroups.com <https://groups.google.com/d/msgid/pidp-11/5c781086-b6e3-4e86-9d9c-18eced3dabbcn%40googlegroups.com?utm_medium=email&utm_source=footer>.

--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: b...@softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol

[isy...@gmail.com]

Hi Johnny,

 I agree that a bus timeout is the way to detect non-existent devices. For my app, this interval is very short ie I don't have any delay waiting for an SSYN. Given this, I made the baseless assumption

that the system tries for a period and then gives up. However, a quick check on my PC app shows a lot of addresses are scanned prior to the DPxx message but after that ... silence for about 30sec.

During this time, there is a rather puzzling sequence of instructions which decrement a location until overflow which appears to repeat many times. I haven't bothered with a full debug here!

I am not sure what it is waiting for. However, you are absolutely right that this should not happen as simh does not show this delay. Oooops, emulation not quite right!!!!

Regards, Ian.

[isy...@gmail.com]

Dear Johnny,

 Thanks for this reminder .... RK emulation not quite right! Disable device ... no delay after DPxx message. So, back to the start above,  boot time is 7 sec same as simh.

Regards, Ian.