1MB 65XE Not Booting Consistently

[flashjazzcat]

I haven't had much time to actually use the 65XE I recently upgraded to 1MB using the kit currently available on Lotharek's website, but it doesn't always boot properly. It passes all the RAM tests I threw at it when I first installed the SIMM, but I've noticed it won't do a rapid powerdown/reboot. Unless I leave it for a good minute or so, it just drops straight back into BASIC when I power it on again, bypassing device polling, the boot sequence, etc. As I say, it seems to be OK if I leave it for a while, but it makes testing other stuff kind of difficult. I'm slightly confused, too: I thought you lost internal BASIC with the 1MB upgrade, since bit 1 of PORTB is now used for bank switching???

 

 

 

[Rybags]

Not sure... there's some arrangements where bits can change role dependant on the setting of other bits.

 

Supposedly some newer RAM modules can take some time to lose their contents. On the Atari, you need the magic numbers up in Page 3 to be wiped for a coldstart to be guaranteed, or for the COLDST ($244) flag to be nonzero.

I usually notice with my XE machines they'll often not coldstart unless powered down for >5 seconds or so (none have RAM expansions).

 

All that aside, I think your solution might well be one of those "freezer" mods that changes the TRIG3 (cartridge inserted) status. Then you can coldstart without having to power off.

 

Other benefits of course would be less wear/tear on the machine and guaranteed preservation of extended RAM contents.

[+Larry]

Wow - a minute's a long time for re-boot. I've got some XE's and XL's that take 15-20 seconds, but nothing in that 60-sec. range.

 

Can you post (or PM) a link to Lotharek's web site?

 

Thanks,

Larry

 

I haven't had much time to actually use the 65XE I recently upgraded to 1MB using the kit currently available on Lotharek's website, but it doesn't always boot properly. It passes all the RAM tests I threw at it when I first installed the SIMM, but I've noticed it won't do a rapid powerdown/reboot. Unless I leave it for a good minute or so, it just drops straight back into BASIC when I power it on again, bypassing device polling, the boot sequence, etc. As I say, it seems to be OK if I leave it for a while, but it makes testing other stuff kind of difficult. I'm slightly confused, too: I thought you lost internal BASIC with the 1MB upgrade, since bit 1 of PORTB is now used for bank switching???

[flashjazzcat]

http://www.mmj.pl/~lotharek/atari/ Edited December 14, 2009 by flashjazzcat

[tjlazer]

Cool. So does he sell a complete kit to upgrade to 1MB? What all does it entail? I have 1MB SIMMs (30-pin).

[atariksi]

Wow - a minute's a long time for re-boot. I've got some XE's and XL's that take 15-20 seconds, but nothing in that 60-sec. range.

 

Can you post (or PM) a link to Lotharek's web site?

 

Thanks,

Larry

 

I haven't had much time to actually use the 65XE I recently upgraded to 1MB using the kit currently available on Lotharek's website, but it doesn't always boot properly. It passes all the RAM tests I threw at it when I first installed the SIMM, but I've noticed it won't do a rapid powerdown/reboot. Unless I leave it for a good minute or so, it just drops straight back into BASIC when I power it on again, bypassing device polling, the boot sequence, etc. As I say, it seems to be OK if I leave it for a while, but it makes testing other stuff kind of difficult. I'm slightly confused, too: I thought you lost internal BASIC with the 1MB upgrade, since bit 1 of PORTB is now used for bank switching???

 

Yeah, I have a couple of XLs that take at least 5-10 seconds wait before I can turn on, but I also have some XLs that boot instantly. I have been using these latter for testing since I don't want to decrease my lifespan waiting for stuff to boot up. And for debugging stuff, I am rebooting like 100+ times/day.

 

I was thinking it's something to do with the capacitors in the machine causing the delay. Anyone researched the capacitors and whether their death or ratings affect boot up time?

[+bob1200xl]

It's the capacitors in the DRAM chips that hold the data. You normally refresh (re-charge the capacitors) every 16ms while you're powered on. Think of them as little batteries whose charge leaks off into channels in the chip. Many of the newer, better built chips don't leak their charge on each bit in anything near 16ms - more like 16 seconds. Some machines can be turned off for a few seconds and have all their data intact when powered on. Look at flash memory or EEPROMs. They drive a charge into an area of the chip that doesn't leak off for 10 years or more. The only solution is to force a cold start with a switch. Or, use SRAM.

 

Bob

 

 

 

Wow - a minute's a long time for re-boot. I've got some XE's and XL's that take 15-20 seconds, but nothing in that 60-sec. range.

 

Can you post (or PM) a link to Lotharek's web site?

 

Thanks,

Larry

 

I haven't had much time to actually use the 65XE I recently upgraded to 1MB using the kit currently available on Lotharek's website, but it doesn't always boot properly. It passes all the RAM tests I threw at it when I first installed the SIMM, but I've noticed it won't do a rapid powerdown/reboot. Unless I leave it for a good minute or so, it just drops straight back into BASIC when I power it on again, bypassing device polling, the boot sequence, etc. As I say, it seems to be OK if I leave it for a while, but it makes testing other stuff kind of difficult. I'm slightly confused, too: I thought you lost internal BASIC with the 1MB upgrade, since bit 1 of PORTB is now used for bank switching???

 

Yeah, I have a couple of XLs that take at least 5-10 seconds wait before I can turn on, but I also have some XLs that boot instantly. I have been using these latter for testing since I don't want to decrease my lifespan waiting for stuff to boot up. And for debugging stuff, I am rebooting like 100+ times/day.

 

I was thinking it's something to do with the capacitors in the machine causing the delay. Anyone researched the capacitors and whether their death or ratings affect boot up time?

[flashjazzcat]

Oddly enough, with the SDX cart plugged in, the system cold starts repeatedly without problems.

[+bf2k+]

I was thinking it's something to do with the capacitors in the machine causing the delay. Anyone researched the capacitors and whether their death or ratings affect boot up time?

 

I noticed years ago that when I added Rambo mem upgrades to 2 of my 800xl's, both of them required longer off times (5 secs or so) than they did before adding the ram. I always assumed it had something to do with the additional memory.

[Marius]

Switching off and on "all the time" is a bad thing anyway to do.

I understand you want a solution for this, but you'd better concentrate on a better alternative.

 

I never switch off and on my atari, because SHIFT + RESET results in a coold boot, thanks for using different OS-es, which give this option.

 

There is even a work-around for a hardware cold-reset switch, if you do not want to use another OS.

 

Bob1200XL writes about that hardware cold-reset switch, he might know how to add that.

 

Greetz

Marius

 

p.s. using another OS in your Atari gives you also plenty of other benefits.

[+Larry]

I first read about this issue about the time that the XE's came out. The 4164's were definitely longer than 4116 for re-boot, and then when the 41256's started showing up, they were worse than the 4164's. (That would seem to fit the pattern mentioned by Bob1200xl.)

 

I've settled on using a Newell-modified 800XL as my main machine, so that is why I was lobbying Warerat to add the simulated cold start feature in the MIO. I am sincerely grateful to him every time I reboot!

 

-Larry

 

I was thinking it's something to do with the capacitors in the machine causing the delay. Anyone researched the capacitors and whether their death or ratings affect boot up time?

 

I noticed years ago that when I added Rambo mem upgrades to 2 of my 800xl's, both of them required longer off times (5 secs or so) than they did before adding the ram. I always assumed it had something to do with the additional memory.

[+Larry]

Hi Marius-

Could you recommend some (XL/XE-compatible) that have this feature? I know about MyIDE, but that only works with the MyIDE adapter.

-Larry

 

Switching off and on "all the time" is a bad thing anyway to do.

I understand you want a solution for this, but you'd better concentrate on a better alternative.

 

I never switch off and on my atari, because SHIFT + RESET results in a coold boot, thanks for using different OS-es, which give this option.

 

There is even a work-around for a hardware cold-reset switch, if you do not want to use another OS.

 

Bob1200XL writes about that hardware cold-reset switch, he might know how to add that.

 

Greetz

Marius

 

p.s. using another OS in your Atari gives you also plenty of other benefits.

[flashjazzcat]

Having killed an ATMEL chip on my SIO2SD when hot-swapping it in the past, I take care to do a complete power down now when plugging devices in and out of my five Ataris. It's not a huge PITA, but this was really why I questioned the longer reboot cycle.

 

 

 

[HiassofT]

Hi Larry!

 

Could you recommend some (XL/XE-compatible) that have this feature? I know about MyIDE, but that only works with the MyIDE adapter.

You could use my Highspeed SIO Patch and program the patched OS into an EPROM, it supports cold-boot with SHIFT+RESET and with SHIFT+CONTROL+DELETE (almost like on PCs :-)

 

so long,

 

Hias

[+Larry]

Thanks, Hias-

 

I do use HISIO.COM to provide Ultra Speed (and beyond!), but I totally forgot that it had the "three-fingered salute" built into it. A very, very nice piece of work!

 

-Larry

 

Hi Larry!

 

Could you recommend some (XL/XE-compatible) that have this feature? I know about MyIDE, but that only works with the MyIDE adapter.

You could use my Highspeed SIO Patch and program the patched OS into an EPROM, it supports cold-boot with SHIFT+RESET and with SHIFT+CONTROL+DELETE (almost like on PCs :-)

 

so long,

 

Hias

[flashjazzcat]

For closure, one of the +5v lines on the memory upgrade had come loose. That's why the computer wasn't booting properly...

Edited December 28, 2009 by flashjazzcat

[atariksi]

It's the capacitors in the DRAM chips that hold the data. You normally refresh (re-charge the capacitors) every 16ms while you're powered on. Think of them as little batteries whose charge leaks off into channels in the chip. Many of the newer, better built chips don't leak their charge on each bit in anything near 16ms - more like 16 seconds. Some machines can be turned off for a few seconds and have all their data intact when powered on. Look at flash memory or EEPROMs. They drive a charge into an area of the chip that doesn't leak off for 10 years or more. The only solution is to force a cold start with a switch. Or, use SRAM.

 

Bob

 

 

 

Wow - a minute's a long time for re-boot. I've got some XE's and XL's that take 15-20 seconds, but nothing in that 60-sec. range.

 

Can you post (or PM) a link to Lotharek's web site?

 

Thanks,

Larry

 

I haven't had much time to actually use the 65XE I recently upgraded to 1MB using the kit currently available on Lotharek's website, but it doesn't always boot properly. It passes all the RAM tests I threw at it when I first installed the SIMM, but I've noticed it won't do a rapid powerdown/reboot. Unless I leave it for a good minute or so, it just drops straight back into BASIC when I power it on again, bypassing device polling, the boot sequence, etc. As I say, it seems to be OK if I leave it for a while, but it makes testing other stuff kind of difficult. I'm slightly confused, too: I thought you lost internal BASIC with the 1MB upgrade, since bit 1 of PORTB is now used for bank switching???

 

Yeah, I have a couple of XLs that take at least 5-10 seconds wait before I can turn on, but I also have some XLs that boot instantly. I have been using these latter for testing since I don't want to decrease my lifespan waiting for stuff to boot up. And for debugging stuff, I am rebooting like 100+ times/day.

 

I was thinking it's something to do with the capacitors in the machine causing the delay. Anyone researched the capacitors and whether their death or ratings affect boot up time?

 

It was the RAM in my case causing the delay (10 seconds) on the 800XL. I just replaced the 8 64kbit MB8264A-15 with 8 MK4564N-15 and the system boots up in one second. And putting in TMS4256-10NL in same sockets delayed the boot time past 10 seconds.

[flashjazzcat]

Oddly, the machine has no problems downing a quick power-down cycle now that INTSDX is fitted. Just didn't seem as happy before that.