Odd behaviour

[#1]

I have 2  desktops both with dual booting  windows 7 &  Linux mint.
One machine has SSD whilst the other is a standard HDD.
The odd behaviour occurs on the desktop with the SSD,  for example it boots automatically into Linux or if I select the windows option it also boots into that with no problem but if I have been using Linux then shut the machine down, power up and select the windows option it just freezes up, a second attempt to get into windows gets me to the windows menu with the option of booting in safe mode or normal mode, selecting normal mode it then boots up fine but now we come to the really odd bit, if the machine is left switched off for an hour or more, I can power up select windows and it boots up windows ok.
Is it possible that SSD could be retaining additional data for a short period after the desktop is switched off or is this not possible ?
It's not really a major problem but I am curious about it ?
Any thoughts anyone ?

[#2]

when you say power off do you mean you completely shut it down and then have to push the power button to make it start a boot cycle or do you select the option of restart from the operating system ?
the reason I ask is if its the latter then the problem would be with ram preloaded for Linux OS start and that could be causing the problem, something that sounds far more likely than your SSD being a problem at least, the symptoms you describe are of a temporary rather than persistent problem (when there is enough time for the ram to clear it doesn't have the same problems).

[#3]

I completely shut it down.
When I rebooted it I got the error so assumed that by completely shutting it down would clear the error  but it was still there.
Whenever possible I try do do any windows stuff first such as photography software that will only run on windows and then reboot into Linux for any Internet stuff.

[#4]

the only thing that would make sense would be if your bios was configured to not clear ram on boot so some junk was still in there given the symptoms your describing

[#5]

Linux makes no assumptions about the state of your RAM on booting,
i.e. it doesn't presume it will zeroed.
I know nothing about Windows (thankfully). 

[#6]

the bios sequence is supposed to perform a memory test and clear the ram when you do a cold start.....in some motherboards this option to clear the ram can be changed or indeed is defaulted to not happen as a single os system it's preferable to have that switch on sequence preloaded for a faster start up time ...
claiming Linux doesn't when you don't use any other os doesn't mean your correct unless you have added a custom command line as the first boot cycle entry of "clear ram"!

[#7]

Complete tosh...  The contents of RAM is not maintained during a power cycle.

[#8]

on some motherboards this might be true, on the majority they don't actually remove power from the board when you turn the system off, thus the ram retains its last known data settings, something that's normally advantageous for installing patches etc ....

[#9]

Complete nonsense    Even if the motherboard is powered the contents of the RAM prior to shutdown is of no use and will be overwritten as the OS reloads.
Fortunately, I suspect gleneagles has seen evidence of your technical 'knowledge' before and knows better than to believe you

[#10]

That would depend on the way the OS is setup, if you haven't told it that it's part of a duel boot system then why would it behave as if it is?
Clearly he has one system that is able to operate normally regardless of the OS selected and another that is not, the options as to why doing a  full shutdown from mint and an hour later booting into windows without problems yet doing a full shutdown from mint and then booting into windows for problem parts are limited to one specific option, with 3 possible reasons, that being ram, retaining info that isn't ignored during start up, reasons for this are limited to bios settings or an incorrectly installed os.
He has said that he has two systems that are both duel boot and the problem only happens on one so the reasons for why it's happening on the one and not the other are limited to ram remaining powered for a short time after shutdown causing boot cycle corruptions,
That's not negotiable as the cause of the problem as described, which means he probably needs to plug in a device that sucks juice from the system to get the latent power left after use faster

[#11]

Quote from: nanotm

on some motherboards this might be true, on the majority they don't actually remove power from the board when you turn the system off, thus the ram retains its last known data settings, something that's normally advantageous for installing patches etc ....

The only power from a standard ATX PSU is 5vSBY (from a separate PSU module to the main power) which, depending on M/B links, enables the power on switch circuit and power-up via RTC/PCI/USB/keyboard/mouse....  DRAM refresh is inoperative and decay times far less than power off to "power button press" response times.
Regarding patches, how would MS Windows determine if the "in RAM patch" memory has data from a reboot, shutdown/restart or total power down/restart? Probably by duplicating the patch on the disk and comparing, thereby rendering any "in RAM patch" totally redundant
gleneagles:
Do you have a spare HDD to receive a copy of the SSD and run the machine with that to eliminate the SSD as the problem source?
Phil

[#12]

According to https://en.wikipedia.org/wiki/Memory_refresh#Refresh_interval a RAM refresh interval of a few seconds is achievable...

[#13]

that's talking about charge state refresh and its in miliseconds not seconds based on ddr2 ....
@peegee
if the problem were the ssd it would be present every time he tried to boot into windows not only when doing so immediately after a shutdown..... ssd's don't loose data bits or we would all be crying every time we started up in the morning with a blank c drive .....

[#14]

That's an interesting approach - the only specified hardware difference is the presence/absence of a SSD, yet you eliminate that from consideration
Obviously, the storage element of SSDs consists of SRAM but there is a lot more electronics that does not "store" it's state on power-down and will need initialising on power-up, otherwise you get random unwanted events; even the possibility of a cache table being valid with the data being corrupt.
Whilst there are circuits to hold reset signals during power stabilisation, there is also a decay delay on this circuit which leaves the possibility of the SSD being a problem, whatever the probability may be.
Phil

[#15]

@PeeGee ,
I Do have a spare hdd but to be honest cannot be bothered switching things about having learnt the hard way that if things are working ok then leave well alone.
This odd behaviour is only a minor issue but I posted it just in case someone else using ssd had come across the same issue, however the post does seem to have resulted in one or two possibilities as to why this is happening.